Richard I.

Gillings
MATHEMATICS
IN THE TIME OF
THE PHARAOHS
 DOVER SCIENCE BOOKS
THEORIES OF THE WORLD FROM AN IQUFIY TO THE COPERNICAN REVOLUTION,
Michael J. Crowe. (26173-5) $6.95
A SHORT HISTORY OF TECHNOLOGY: FROM THE EARLIEST TIMES TO A.D. 1900,
T.K. Derry and Trevor 1. Williams. (27472-1)
FADS AND FALLACIES IN THE NAME OF SCIENCE. Martin Gardner. (20394-8)
$7.95
UNDERSTANDING ExisTEINS' THEORIES OF RELATIVITY: MAN'S NEw PERSPECTIVE ON
THE Cosmos, Stan Gibilisco. (26659-1) $8.95
DE MAGNETE, William Gilbert. (26761-X) $11.95
GOSSAMER ODYSSEY. THE TRIUMPH OF HUMAN-POWERED FuGm Morton
Grosser. (26645-1) $9.95
ON THE SENSATIONS OF TONE, Hermann Helmholtz. (60753-4)
THE STRANGE STORY OF THE QUANTUM, Banesh Hoffmann. (20518-5) $7.95
AL.cHEMY. E.J. Holmyard. (26298-7) $8.95
THE SCIENCE OF SOAP FILMS AND SOAP BUBBLES, Cyril Isenberg. (269604)

MENLO PARK REMINISCEINCES, VOLUME ONE, Francis Jehl. (26357.6) $13.95

FROM SUNDIALS TO ATOMIC CLOCKS: THEIR THEORY AND CONSTRUCTION, James
Jespersen and Jane Fitz-Randolph. (24265-)Q $8.95
ANCIENTS AND MODERNS, Richard Foster Jones. (24414-8) $7.95
EXPERIMENTS IN TOPOLOGY, Stephen Barr. (25933-1) $6.95
THE GREAT PHYSICISTS FROM GAL ILEO TO EINSTEIN, George Gamow. (25767-3)
$9.95
AN ILLUSTRATED GUIDE TO LINEAR PROGRAMMING, Saul 1. Gass. (26258-8)
THE SCIENCE OF MEASuREMENT, Herbert Arthur Klein. (25839-0) $16.95
EXCURSIONS IN NUMBER THEORY, C. Stanley Ogilvy and John T. Anderson.
(25778-9) $5.95
THE ENJOYMENT OF MATHEMATI S, Hans Rademacher and Otto Toeplitz.
(26242-1) $7.95
DISCOVERING THE NATURAL LAws, Milton Rothman. (261780 $6.95
RELATIVITY IN IILusTRATIONs, Jacob T. Schwartz. (25965-X) $7.95
THE HISTORICAL RooTs OF ELEMENTARY MATHEMATICS, Lucas N.H. Bunt et al.
(25563-8) $8.95
THE Ex. cr SCIENCES IN ANnQumr, 0. Neugebauer. (22332-9) $7.95
ENTERTAINING SCIENCE EXPERIMENTS wrim EVERYDAY OBJECTS, Martin Gardner.
(24201-3) $9.95
GREAT EXPERIMENTS IN PHYSICS, Morris H. Shamos. (25346-5) $10.95
ALL ABOUT LIGHTNING, Martin A. Uman. (25237-X) $6.95
THE COMPLEAT STRATEGYST: BEING A PRIMER ON THE THEORY OF GAMES OF
STRATEGY, J. D. Williams. (25101-2) $7.95
FAMous PROBLEMS OF GEOMETRY AND How To SOLVE THEM, Benjamin Bold.
(24297-8) $4.95
SIDELIGHTS ON RELATIVm; Albert Einstein. (24511-X) $3.95
(continued on back flap)
 Mathematics
in the Time
of the Pharaohs
The scribe Ka-Irw-Khufu of the Fourth Dynasty, excavated at Giza.
The statue is now owned by the Cairo Museum. Copyright Roger
\Vood, London.
 Mathematics
in the Time
of the Pharaohs
RICHARD J. GILLINGS

Dover Publications, Inc.

NEW YORK
 Copyright a 1972 by The Massachusetts Institute of Technology.
Appendix 15 copyright o 1981 by Academic Press, Inc.
All rights reserved under Pan American and International Copy-
right Conventions.
Published in Canada by General Publishing Company, Ltd., 30
1.esmill Road, Don Mills. Toronto, Ontario.

This Dover edition, first published in 1982, is an unabridged,

slightly corrected republication of the work first published in 1972. It
is reprinted by special arrangement with The MIT Press, 28 Carleton
Street, Cambridge, MA 02142, publisher of the original edition.
Appendix 15 originally appeared in Historia Mathematica 8 (1981),
pp. 456-459, and is presented here by permission of Academic Press,
Inc.

Manufactured in the United States of America

Dover Publications, Inc.
31 East 2nd Street
Mineola, N.Y. 11501

Library of Congress Cataloging In Publication Data

Gillings. Richard J.
Mathematics in the time of the pharaohs.
Reprint. Originally published: Cambridge, Mass.: MIT Press,
1972.
Bibliography: p.
Includes index.
1. Mathematics, Egyptian. I. Title.
[QA27.E3G52 1982] 510' .932 81-19445
ISBN 0-486-24315-X (pbk.) AACR2
Dedicated to 0. Neugebauer and A. Sachs,
in friendship and gratitude
I wish I knew as much as I thought I knew 10 years ago
0. Neugebaucr, 1953
 PREFACE xi
1

Introduction I

2
Hieroglyphic and Hieratic Writing and Numbers 4
3
The Four Arithmetic Operations 11
ADDITION AND SUBTRACTION 11

MULTIPLICATION 16
DIVISION 19
FRACTIONS 20
4
The Two-Thirds Table for Fractions 24
PROBLEMS 61 AND 61B OF THE RHIND MATHEMATICAL PAPYRUS 29
TWO-THIRDS OF AN EVEN FRACTION 32
AN EXTENSION OF RMP 61B AS THE SCRIBE MAY HAVE DONE IT 33
EXAMPLES FROM THE RHIN'D MATHEMATICAL PAPYRUS OF THE
TWO-THIRDS TABLE 34
5
The G Rule in Egyptian Arithmetic 39
FURTHER EXTENSIONS OF THE G. RULE 42
6
The Recto of the Rhind Mathematical Papyrus 45
THE DIVISION OF 2 BY THE ODD NUMBERS 3 TO 101 45
CONCERNING PRIMES 52
FURTHER COMPARISON OF THE SCRIBE'S AND THE COMPUTER'S
DECOMPOSITIONS 53

7
The Recto Continued 71
EVEN NUMBERS IN THE RECTO: 2 = 13 71
MULTIPLES OF DIVISORS IN THE RECTO 74
TWO DIVIDED BY THIRTY-FIVE: THE SCRIBE DISCLOSES HIS METHOD 77

Problems in Completion and the Red Auxiliaries 81

USE OF THE RED AUXILIARIES OR REFERENCE NUMBERS 81
AN INTERESTING OSTRAC'.ON A6
viii Contents

9
The Egyptian Mathematical Leather Roll 89
THE FIRST GROUP 95
THE SECOND GROUP 95
THE THIRD GROUP 96
THE FOURTH GROUP 98
THE NUMBER SEVEN 99
LINE 10 OF THE FOURTH GROUP 102
THE FIFTH GROUP 102
I0
Unit-Fraction Tables 104
UNIT.FRACFION TABLES OF THF RHIND MATHEMATICAL PAPYRUS I06
PROBLEMS 7 TO 20 OF THE RHIND MATHEMATICAL PAPYRUS 109

U
Problems of Equitable Distribution and Accurate Measurement 120
DIVISION OF THE NUMBERS I TO 9 BY 10 120

CUTTING UP OF LOAVES 123

SALARY DISTRIBUTION FOR THE PERSONNEL OF THE TEMPLE OF
I -IAH IN 124

12
Pesu Problems 128
EXCHANGE OF LOAVES OF DIFFER _NT P SUS 132

11
Areas and Vnlumes 137
THE AREA OF A RFCTANGI.E 137

THE AREA OF A TRIANGLE 138

THE AREA OF A CIRCLE 139

THE VOLUME OF A CVI_INDRlCAI_ GRANARY I4fi

THE DETAILS OF KAHUN IV, 3 151

14
Equations of the First and Second Degree 154
THE FIRST GROUP 154
SIMILAR PROBLEMS FROM OTHER PAPYRI 156
THE SECOND AND THIRD GROUPS 158
EQUATIONS OF THE SECOND DEGREE 161
 Contents ix

KAHUN LV, 4 162

SUGGESTED RESTORATION OF MISSING LINES OF KAHUN LV, 4, AND

MODERNIZATION OF OTHERS I63

15
Geometric and Arithmetic Progressions 166
GEOMETRIC PROGRESSIONS- PROBLEM 79 OF THE RHIND MATHEMATICAL
PAPYRUS 166

ARITHMETIC PROGRESSIONS: PROBLEM 40 OF THE RHIND

MATHEMATICAL PAPYRUS 170

PRORTFM 64 OF THE Rf-ITND \fATHFMfATIrAl_ PAPYRUS 173

KAHUN IV, 3 176

16
"Think of a Number" Problems 181
PROBLEM 28 OF THE RHIND MATHEMATICAL PAPYRUS 181

PROBLEM 29 OF THE RHIND MATHEMATICAL PAPYRUS 183

17
Pyramids and Truncated Pyramids 185
THE SEKED OF A PYRAMID 185
THE VOLUME OF A TRUNCATED PYRAMID 187
is
The Area of a Semicylinder and the Area of a Hemisphere 194

19
Fractions of a Hekat 202
20
Egyptian Weights and Measures 207
21
Squares and Square Roots 214
22
The Reisner Papyri: The Superficial Cubit and Scales of
Notation 218
APPENDIX 1
The Nature of Proof 232
APPENDIX 2
The Egyptian Calendar 235
x Contents

APPENDIX 3
Great Pyramid Mysticism 237
APPENDIX 4
Regarding Morris Kline's Views in Mathematics, A
Cultural Approach 240
APPENDIX 5
The Pythagorean Theorem in Ancient Egypt 242
APPENDIX 6
The Contents of the Rhind Mathematical Papyrus 243
APPFNDiX 7
The Contents of the Moscow Mathematical Papyrus 246
APPENDIX 8
A Papyritic Memo Pad 248
APPENDIX 9
Horus-Eye Fractions in Terms of Hinu: Problems 80, 81 of the
Rhind Mathematical Papyrus 249
APPENDIX 10
The Egyptian Equivalent of the Least Common Denominator 251
APPENDIX I I
A Table of Two-Term Equalities for Egyptian Unit Fractions 254
APPENDIX 12
Tables of Hieratic Integers and Fractions, Showing Variations 255
APPENDIX 1
Chronology 260
APPENDIX 14
A Map of Egypt 266
APPENDIX 15
The Egyptian Mathematical Leather Roll 267
BIBLIOGRAPHY 269
INDEX 279
 PREFACE

In 1811, Thomas Cooper, professor of chemistry at Dickinson College,

Carlisle, Pennsylvania, and a Britisher by birth, said at his opening
lecture, "The history of an art or science is a proper introduction to the
study of it; as giving a clear and concise view of the manner in which
improvements have been effected; as furnishing due caution against
future errors by exhibiting the mistakes of superior minds of olden
times, and as rendering merited honor to those who have benefitted
mankind by their discoveries."
I have often quoted these words in my own lectures on ancient
Egyptian mathematics, but wherever and whenever I have differed
from recognized authorities in this field of study, like Eisenlohr,
Gillain, Griffith, Sethe, Peet, Struve, Vogel, Chace, Neugebauer, Van
der Waerden, and Bruins, I have done so only after due caution and
consideration, yet nevertheless with firm and proper conviction. For it
was only after carefully examining their translations, their opinions,
their solutions and comments, their criticisms, and especially their
statements wherein I judged they may have erred, that I have been
enabled, I hope, to achieve a truer overall picture of the mathematics
of the ancient Egyptians. In other words, I have been privileged to
stand on the shoulders of savants and scholars, and the experience has
been profoundly rewarding.

R.J. Gillings

University of New South Wales,

and Turramurra, Australia, 1971
 Mathematics
in the Time
of the Pharaohs
 1 INTRODUCTION

One of the oddest of all the phenomena which come to the attention of
students of the history of mathematics is that logarithms were invented
by Napier more than a decade before Descartes first conceived the idea
of using indices in algebra. This oddity becomes even more striking
when we observe that mathematical textbooks today introduce the
subject of logarithms by a preliminary study of the index laws of algebra,
which is pedagogically perhaps the very best way to do it. Chrono-
logically, therefore, the expected order of invention seems to have been
reversed; things are the wrong way round!
A second oddity of the history of mathematics was brought to light
when the Babylonian clay tablet Plimpton 322 (museum number,
Columbia University, New York) was translated by Neugebauer and
Sachs in 1945. The translation established beyond any doubt that the
Pythagorean theorem was well known to Babylonian mathematicians
more than a thousand years before Pythagoras was born. The history
books tell us that the Greek mathematician sacrificed an ox to cele-
brate the discovery of the theorem named after him. Here then is an
unrewarded anticipation, for doubtless the name of the famous theorem
will remain as a true mumpsimus-" the Pythagorean theorem"-for
all time.
Now there is a third oddity in the history of mathematics, which,
however, we can clearly understand and explain; it is, indeed, one of
the raisons d'etre for this book. It is the circumstance that the mathe-
matics, astronomy, and science of the two most ancient of our recorded
civilizations, the Egyptian and the Babylonian, have only recently
been the subjects of historical research. And the very simple reason for
this is that for nearly 3,000 years no one knew what the many extant
hieroglyphic and cuneiform writings of these two civilizations meant,
nor indeed whether they were writing at all ! It was not until Cham-
pollion's Dictionnaire Egyptien appeared in 1842 that the Egyptian
hieroglyphs were at last deciphered, and not until the latter part of the
nineteenth century that the cuneiform writings (beginning with
Grotefend in 1802), with their many languages, were deciphered by
2 Chapter One

Assyriologists, and the secrets so well hidden for centuries-indeed for

millennia-at last unfolded.
As a result of these pioneering researches, scholars are now at work
in universities, institutes, and museums, transcribing and translating
inscriptions from temples and tombs, writings on clay tablets and
stelae, hieratic and demotic writings on papyri and ostraca, which had
lain unread for years as interesting objects and relics of past civiliza-
tions, ancient records whose meaning and significance were only to be
guessed or wondered at. Today, scientific and historical journals
throughout the world are receiving valuable and informative articles
on Egypt and Babylonia to such an extent that they can scarcely cope
with the material. In many cases they are months or even years behind
their regular publication schedules. The historians' output is setting
editors a task that is steadily becoming more formidable.
Of course, some research still proceeds on the histories of the Hindus,
the Persians, the Phoenicians, the Hebrews, the Greeks, the Romans,
and the Arabic nations, all of whom come much later than the Egyp-
tian culture with which we shall be concerned. The mathematics,
astronomy, and science of these other peoples are already well au-
thenticated; deciphered records and scholarly commentary fill many
shelves. From the seventeenth century onward, many volumes came
from the pens of mathematicians and historians: Moritz Cantor (4
volumes), Johannes Tropfke (7 volumes), Florian Cajori (2 volumes),
David Smith (3 volumes), etc. These commentaries and histories took
as their starting point the early Greeks, say Thales, about 600 B.C. But
knowledge of the earlier civilizations of Egypt and Babylon was not
available to these writers, and did not become generally known until
the beginning of the present century.
Well may we express our admiration of the wonderful architecture
of the Egyptian temples of Karnak and Luxor, at the grandeur and the
immensity of the Pyramids and at the construction of their magnificent
monuments. Well may we wonder at the government and the econo-
mics of a country extending nearly a thousand miles from north to
south through which ran the longest river of the then-known world.
And well may we marvel at the Egyptians' design of extensive irriga-
tion canals, at their erection of great storage granaries, at the organiza-
 Introduction 3

tion of their armies, the building of seagoing ships, the levying and
collection of taxes, and at all the thought and effort concomitant with
the proper organization of a civilization that existed successfully,
virtually unchanged, for centuries longer than that of any other nation
in recorded history.
What we today call science and mathematics must have played an
important role in the achievement of all this. I am reminded of a piece
of wisdom attributed to Arnold Buffum Chace, the principal author of
The Rhind Mathematical Papyrus:

I venture to suggest that if one were to ask for that single attribute of
the human intellect which would most clearly indicate the degree of
civilization of a race, the answer would be, the power of close reason-
ing, and that this power could best be determined in a general way by
the mathematical skill which members of the race displayed. Judged
by this standard the Egyptians of the nineteenth century before Christ
had a high degree of civilization.*
If we accept this thought as one containing a solid measure of truth,
then it will surely come as a great surprise to the readers of this history
to find that whatever great heights the ancient Egyptians may have
achieved scientifically, their mathematics was based on two very ele-
mentary concepts. The first was their complete knowledge of the twice-
times table, and the second, their ability to find two-thirds of any number,
whether integral or fractional. Upon these two very simple foundations
the whole structure of Egyptian mathematics was erected, as we shall
see in the following pages.
* A. B. Chace; L. Bull; H. P. Manning; and R. C. Archibald, The Rhind
Mathematical Papyrus, Vol. 1, Mathematical Association of America,
Oberlin, Ohio, 1927, Preface.
 2 HIEROGLYPHIC AND HIERATIC
WRITING AND NUMBERS

No Egyptian scribe could ever have claimed to be the first man to pick
up a mallet and chisel, and to have said to himself, "Now I am going
to invent hieroglyphics." He could never have set about carving on a
block of stone various figures that would have a special meaning or
would convey a message to those who might see it. Neither could it
have happened that some intelligent scribe could have been the very
first to think of slicing up some Nile River papyrus reeds and, by placing
some strips crosswise over others and pressing them flat, invented
" paper"; then, bruising the end of a smaller reed, and having dipped
it into a pot of " ink," made the dramatic announcement, "Now I am
going to write in hieratic characters!"
Neither of these things could have happened like that. The inven-
tion of hieroglyphics, which must have come first, took many many
years, perhaps centuries. And hieratic writing, the first cursive form of
hieroglyphics, developed much later, as a quicker and more convenient
way of recording an agreement, conveying a message, or making a
calculation with numbers than by the detailed drawing of pictographic
hieroglyphs. No one is able to say exactly when writing as we under-
stand it actually began. But with the Egyptians, as with other ancient
civilizations, the method used to represent numbers must have been at
least easier than writing their phonetically equivalent words. What
could be simpler than the scheme shown in Figure 2.1 ?
Other nations like the Romans, Babylonians, and Chinese thought
of something similar for numbers up to 9. The ability to write numbers
depended on simple counting, which could have been learned without
knowing every separate word for the numbers being written. Most
nations adopted a decimal system, no doubt because of the anatomical
circumstance that Homo sapiens has ten fingers. Others worked in
groups of 5, groups of 20 (the Mayas), or groups of 12 (the Romans) ;
and the Babylonians had a sexagesimal system, which worked partly
in groups of 10, and partly in groups of 60. The Egyptian symbols for
higher powers of ten are shown in Figure 2.2.
 Hieroglyphic and Hieratic Writing and Numbers 5

1 6 III

2 Il 7 II III
3 III 8 rill

4 (Ill 9 1III11

5 I It
FiouRE 2.1
The earliest Egyptian symbols for the numbers 1 to 9.

to r1

100 9
Lotus
1,000

1.0000 D
100,000 Tadpole or bird

1,000,000 "\,L, net ntaer us

FIGURE 2.2
Egyptian symbols for large numbers.

When writing a number in hieroglyphics, sometimes a large number

of characters would be required, and often a smaller number would
require more characters than a bigger one. Thus the number 1967 in
IIInnn99999
hieroglyphs is t ' n n n o a 9 9, requiring 23 characters, whereas the
number 20,000 is Ad, requiring only 2. It should be remembered that
the Egyptians carved or "printed" their hieroglyphs and wrote their
cursive hieratic (and later their demotic) forms from right to left, just as
the Hebrews did, and as the Arabs do today. In contradistinction, how-
ever, the Babylonians wrote their cuneiform characters on clay tablets,
6 Chapter Two

from left to right. Often the hieroglyphic and hieratic writing is written
vertically downwards, but still also from right to left. Where the align-
ment has not been carefully regarded by the scribe, a mathematical
papyrus is difficult to follow. In hieroglyphic and hieratic writing, the
various birds, reptiles, snakes, and other animals, the scribes, seated
or erect, and the human faces, all face the direction from which the
writing is coming, when drawn, as most of them are, in profile. It is
however an accepted convention that in translation into English or
other European language the direction is reversed for convenience,
but it is necessary and important, when comparing papyri with the
translations, to note whether or not this convention has been observed.
Lack of careful attention to this can be misleading, and even quite
wrong, as for example in a recent publication of the LIFE history of
mathematics.* There, the two hieroglyphs & and A, are stated as
meaning subtract and add, respectively, and the reader is referred to the
accompanying illustration where these signs occur. Now the truth is
exactly the opposite, and the sense of the mathematical problem,t if
examined from the reproduced illustration of the original, is com-
pletely changed.
In this present book, if hieroglyphs or hieratic writings are shown or
are being discussed as if in situ, then they will be written in the Egyp-
tian fashion from right to left, with hieroglyphs constructed, for ex-
ample, of animals in profile, facing toward the right. But if we are
treating a problem or discussing some mathematical question of
Egyptian techniques and methods, we will in translation, for our own
convenience, write from left to right.
Now practically everything that comes to our attention will origi-
nally have been written in the hieratic script; and, with Egyptian
scribes as with present-day handwriters, no two people write the same
hand. Some arc uniform and some irregular, some are angular and
some slope backwards, and as a result the reader needs to acquaint

* David Bergamini and the Editors of LIFE, LIFE Science Library: Mathematics,
p. 64. © 1965 by TIME Inc. TIME-LIFE International, Netherlands N.V.
t Problem 28 of The Rhind Mathematical Papyrus. See A. B. Chace; L.
Bull; H. P. Manning; and R. C. Archibald, The Rhind Mathematical Papyrus,
Vol. II, Pl. 51.
 Hieroglyphic and Hieratic Writing and Numbers 7

himself with the idiosyncrasies of different writers. So standard prac-

tice among Egyptologists is first of all to transliterate the "cursive"
hieratic into "printed" hieroglyphics, and then to translate the hiero-
glyphics into a modern language. The transliteration is from right to
left, and the translation is from left to right. When a restoration is made,
in the case of an illegible sign or where there is a lacuna in a poorly
preserved papyrus, the custom is to enclose the translation given in
square brackets [...], and this is done even when the restoration is
pretty obvious or even certain. The student of Egyptian and Babylon-
ian mathematics is seldom a competent Egyptologist or Assyriologist
as well as being a mathematician, so for translations he is dependent
on the skilled specialist in hieratic, demotic, or cuneiform writing.
Among such specialists are to be found those who are most competent
in some particular branch of the language, as for example, Middle
Kingdom hieratic or the Ptolemaic demotic (as on the Rosetta stone),
or the cuneiform script (Sumerian, Babylonian, or Akkadian). But
there is one powerful factor that is used to advantage by the historian
of mathematics. It is that the number systems can all be read and
translated by him, even if he cannot understand the words with which
the numbers are associated. Indeed he can go a little further. He can
with a little industry learn some of the standard symbols commonly
associated with mathematical situations-that is, recurrent phrases
and specific expressions-that will help in assessing an interpretation
provided by the translator, who may not himself be competent in the
mathematician's own field. And this is a most important factor with
mathematical papyri, because there is one criticism that certainly can-
not be leveled at their authors. They were never guilty of circumlocu-
tion. The authors of the mathematical papyri were never verbose;
indeed, we could often wish that they were sometimes more discursive,
that they sometimes went into greater detail. In Figures 2.3-2.5 are
given some phrases with their transliterations and translations, that
occur in the mathematical papyri. While these examples are by no
means exhaustive, they may assist a student who is unacquainted with
the hieratic script. They read from right to left, and appear therefore
just as they would on a papyrus.
Two or three examples of each are given to exhibit apparent
8 Chapter Two

yn.L 151. =f1

0
Iln opp ao(I
111 40 w

1z a I a a a
ff &taatuacr'd 2da"e 8

Znij,; tl P, A' 31Lq ti P, 1i Z.ul 1) &

le not 111 11 $to n oI l IIa 1"ng (aa if
7 10 tub 2 49n 9 20 tug a ifgn 1 40 tub 2 $1n,
ttlnlhplg IT sold 2 is t+ttched tll*1hply 2f uwtil 2 ii tttAea AWttyl' 41msltl t is tuclel

FIGURE 2.3
Examples of Hieratic script, with hieroglyphic transcriptions, phonetic
English transliterations, and English translations. The hieratic characters
and hieroglyphs are written from right to left, and orientable hieroglyphs
(here only "Syn ") face toward the right, as in the originals.

differences in cursive handwriting, where ligatures between adjacent

signs may sometimes alter considerably their individual appearance
or character.

FIGURE 2.4
Further examples of hieratic script, transcriptions, transliterations, and
translations. Note that in the examples of vertical writing ("working out")
the orientable figures still face toward the right.
 Hieroglyphic and Hieratic Writing and Numbers 9

*I 69-71411 ts ffi-.9111 U rp -fill I

.M.J DI- 49 son 119 a M. I1q- ak
it vsx Pt It 9! a 1u It vi n tt
Essafu of Pvoq Eacaatk Of %=f Fmsgu of Mof
1 0
a of 4
0 00
I*sef ort lhteef ov.

ss as
Ftal Ft" Fina

*14 Mw %

i iw S tai
Tetaa t Totat6 Totact 4

AT3 4160
nk I * ti g m a t'vi
16 ditay as it eos ha dotsF ss it oars Di as it occurs

ev q fqSU
C. 'M
t
won .4 a.

.kuy oat Woekag out

10 Chapter Two

JL1L 1
j
10 It,
\
14
U. ?t ksW k.*k vi UL ?t ), -, w k t-1. ri
tetki th.u tit 246 AAAAU ttt$ dW

A31A..3L0
"ON
O Gta '1

trp m 3 ltc to 3
to tit otddtb 118114

0-1113
I q;?
:113 eats 13
f rb h3 w m k3wto
To k acdded to tt To be adduct bit

f -r k M Mg fr kM- zay
,gou have corvafy joand tt yoU UvIt toMaly wit
FIGURE 2.5
Further examples of hieratic script, showing the idiosyncratic hand of
different scribes.
 3 THE FOUR ARITHMETIC OPERATIONS

ADDITION AND SUBTRACTION

Historians of Egyptian mathematics have seldom committed them-
selves regarding the methods of addition and subtraction employed by
the Egyptian scribes; they mostly take for granted the scribes' ability
to add or subtract fairly large numbers.* Now we of the twentieth
century can count to a million and well past a million, yet we have
difficulties in addition and subtraction, and we use desk calculators
and other aids to computation to overcome them. And we have only
to open a modern textbook on the teaching of arithmetic to be sur-
prised at the many different methods for subtraction now being
taught. Mankind has always had trouble with these two operations.
That is why the abacus was invented centuries ago, and is still being
used today in many Asian countries.
In the mathematical papyri that have come down to us, there are
many problems proposed and solved which require all four funda-
mental operations for their solution, and we can quite well deduce the
scribes' methods for multiplication and division. But for addition and
subtraction methods there are hardly any clues. It would seem that
these operations were performed and checked elsewhere by the scribes,
and the answers inscribed on the papyri afterwards. A scribal error in
this part of their arithmetic is such a rarity that one can be excused for
drawing the conclusion that they had tables for additions (and conse-
quently for subtractions), from which they merely read off the answers.
If such tables ever existed, however, no copies have come down to us,
so that the idea remains purely a conjecture insofar as integers are
concerned. For fractions, of course, many varied tables must have

* For example, see 0. Neugebauer, "For ordinary additions and subtrac-

tions nothing needs to be said" (The Exact Sciences in Antiquity, Harper
Torchbooks, Harper, New York, 1962, p. 73). Peet has dismissed the subject
by remarking that "people who could count beyond a million had no
difficulty about the addition and subtraction of whole numbers" (" Mathe-
matics in Ancient Egypt," Bulletin of the John Rylands Library, Vol. 15, No. 2
(Manchester, 1931), p. 412).
12 Chapter Three

- 1111 An
m nn 53
Iii non 77

till rhn
37
nnnn 4i
41 ANL III AA n" 83

m.# iiII Ao 99 25
it jisp n, nnnn
w firm 376
X1.11 "
l-F
'ii on 111 635
FIGURE 3.1
Examples of additions. Left, hieratic; center, hieroglyphic; right, Hindu-
Arabic.

existed. The EMLR* is itself very good evidence for the existence of
such tables for addition and subtraction of fractions. Today most
nations write from left to right, and our numbers are so written also;
but the values of the digits in our "Hindu-Arabic" decimal system
increase in place value from right to left. So if we have to perform an
addition or a subtraction, we begin with the units column on the right,
and work toward the left through the tens, the hundreds, the thousands,
and so on. In these calculations, including multiplication, we reverse
our direction of writing.
Conversely, as we have seen, the Egyptians wrote both their words
and numbers from right to left. Of necessity, however, the Egyptian
arithmeticians, like ourselves, had to start adding in the opposite
direction to that in which they were accustomed to write, so the
place value of the Egyptians' digits increases from left to right, and the
Egyptian system therefore runs widdershins to ours.
Figure 3.1 gives some simple examples of addition comparing
Hindu-Arabic, hieroglyphics, and hieratic. In Hindu-Arabic addition,
the number combinations 3 + 4 = 7, 6 + 7 = 13 and 6 + 9 = 15,
are learned by heart, "look-and-say," so that the addition (4 + 3) is

* Egyptian Mathematical Leather Roll. British Museum, London.

 The Four Arithmetic Operations 13

not done by starting with 4, and then counting, 5, 6, 7, and then stop-
ping. This counting method offers itself naturally to one performing
additions in hieroglyphics. In hieratic, however, counting does not so
lend itself, and we ask ourselves:
(a) Did they have an additions table?
(b) Did they add by simple counting?
(c) Did they learn number combinations?
If tables such as that shown in Figure 3.2 were prepared, they would
be equally useful in subtraction, because I am sure the Egyptians did
not say to themselves, "From 12 take away 7, answer 5," but rather,
"Seven, how many more to make 12? It needs 5," that is, the table
would supply the answer.
It is easy to say that multiplication by 10 was simply performed by
changing each I to t,1, each U to 9, each 0 to 1, each I to ,46;p, and

"2t M1= N~(4

1912 A I K M"I "+

1 *A

2911 2810 27 9 26 8 257 24i 235

3912 3811 3710 36 9 358 347
49 13 4.6 12 4 7 11 4610 4 99
5914. 5613 6 71Z 5611
69 is 68 14. 8713
T916 7615
8917
FIGURE 3.2
An addition table in hieratic script that could have been constructed by
the scribes, and its Hindu-Arabic translation.
14 Chapter Three

1012 1 2 8 01
q Yft N =oss 2 56 02
40:11 1 1 2 04
70691 itst 19 6 07

7 7
94 49
343 343
1042 2401
70861 16807
70691 T.tal 19607
FIGURE 3.3
Two additions from Problem 79 of RMP. Left, the hieratic; center, the
transliteration; right, in translation.

so on. But this would be true only of hieroglyphics, and in ordin-

ary everyday business, where these calculations would be needed,
hieratic-and later, the quicker demotic-was used by the clerk or
scribe, and in such writing this simple transfer of signs does not
apply.*
Again, multiplication by continual doubling, performed by a simple
duplication of each sign of a number, seems simple enough, even
allowing for "carrying," but is feasible only in hieroglyphs, which, as
we have seen, were seldom used. We are forced to the conclusion that
reference tables were used for the four fundamental operations (which
we indicate by +, - , x , and -) ; and it is probable that portions of
these tables were memorized much as we do today. When we add a
column of figures, 9 and 5 immediately suggest 4, 8 and 5 suggest 3,
7 and 5 suggest 2, 6 and 5 suggest 1, and in subtraction these particular
pairs suggest the same numbers, so that a kind of elementary theory of
numbers begins to arise. There are very many examples of the four
operations with unit fractions to be found in the papyri, but very few
showing actual addition and subtraction with integers. In Figure 3.3

* See Table 4.6 on hieratic multiplication and division by 10.

 The Four Arithmetic Operations 15

are displayed two additions from RMP* Problem 79, the controversial
problem thought by some to be the prototype of the Mother Goose
rhyme beginning, "As I was going to St. Ives, / I met a man with
seven wives...." In the first addition, by chance the units digits are
more or less in a vertical line, and (1 + 2 + 4) = 7 simply enough.
There are no tens digits. Then for the hundreds there are (800 + 600
+ 200), also more or less in line, which must be written as 6 hundreds
with one thousand to be carried. What the scribe's thought process for
this step was is the point we are doubtful about. Then for the thousands
he had (1,000 + 2,000 + 5,000 + 1,000) = 9,000 with nothing to
carry, and of course the one ten thousand was merely written down in
the total, the whole of which was pushed toward the left, because the
symbol for " total," _.a', being written first, required more space. His
multiplier 7, written by the scribe as (1 + 2 + 4), was not recorded in
the last line.
In the second addition, all digits are out of alignment, units, tens,
hundreds, etc., but the alternative symbol for "total," L, has not
pushed the answer out toward the left. In reading this addition greater
care is necessary. The scribe had (7 + 9 + 3 + 1 + 7), which, how-
ever he arrived at it, is 27, so he put down 7 and carried the 2 tens.
Notice the working is from left to right. For the tens he had, (20 +
40 + 40) = 100, so that there are no tens in the answer. The Egyp-
tians had no sign for zero, nor did they even leave a space to indicate
"no tens." For the hundreds, he had to add (100 + 300 + 400 +
800), which he finds is 1,000 and 600, (however he did it), so he put
down 600 and carried the 1,000. Then he had (1,000 + 2,000 +
6,000) = 9,000 with nothing to carry, which goes into the total with
the 10,000.
What we have not enough evidence to decide is whether in adding
(100 + 300 + 400 + 800) he counted (on his fingers so to speak), one
hundred; two, three, four hundreds; five, six, seven, eight hundreds;
then eight and eight hundreds makes sixteen hundreds: or whether he
thought of units, adding (1 + 3 + 4 + 8) as 16, and then calling
them hundreds. Or did he merely read off the answer from handy
prepared tables?
* Rhind Mathematical Papyrus. British Museum.
16 Chapter Three

MULTIPLICATION
It is not uncommon in histories of mathematics to read that Egyptian
multiplication was clumsy and awkward, and that this clumsiness and
awkwardness was due to the Egyptians' very poor arithmetical nota-
tion.
In Ahmes' treatment of multiplication, he seems to have relied on
repeated additions. Jourdain'
It is remarkable that the Egyptians, who attained so much skill in their
arithmetic manipulations, were unable to devise a fresh notation and
less cumbersome methods. Newmant
The limitations ofthis notation made necessary the use ofspecial tables.
With such a cumbrous system of fractional notation, calculation was a
lengthy process, frequently involving the use of very small fractions.
Sloley+
Such a calculus with fractions gave to Egyptian mathematics an
elaborate and ponderous character, and effectively impeded the
further growth of science. Struik§

If Egyptian multiplication was so clumsy and difficult, how did it

come about that these same techniques were still used in Coptic times,
in Greek times, and even up to the Byzantine period, a thousand or
more years later? No nation, over a period of more than a millennium,
was able to improve on the Egyptian notation and methods. The fact
is that, despite their notation, the scribes were adepts at solving arith-
metic problems and were in fact quite skillful in devising ingenious
methods of attack on algebraic and geometric problems as well, so
that their successors remained content with what came down to t'iem.
How far have we progressed in multiplication since the times of the
ancient Egyptians, or even since Greek and Roman times? What are
our grounds for being so critical of Egyptian multiplication, in which
* Philip E. B. Jourdain, in The World of Mathematics, James R. Newman,
ed., Vol. 1, p. 12, Simon and Schuster, New York, 1956.
t The World of Mathematics, Vol. 1, p. 172.
+ R. W. Sloley, in The Legacy of Egypt, S. R. K. Glanville, ed., pp. 168f.,
Oxford University Press, London, 1942 (reprinted 1963).
§ Dirk J. Struik. A Concise History of Mathematics, pp. 19f., Dover, New York,
1948.
 The Four Arithmetic Operations 17

it was only necessary to use the twice-times tables? In English-speaking

countries, at least, as late as the sixteenth century, it was not part of
the school curriculum to learn any multiplication tables at all. Samuel
Pepys, the famous diarist, educated at St. Paul's School and at Mag-
dalen College, Cambridge University, " an able man of business,"* was
secretary to the British Admiralty, and in that position must surely
have needed to know how to calculate. But note this entry:

July 4. 1662.... Up by 5 o'clock.... Comes M. Cooper of whom I

intend to learn mathematiques, and do, being with him to-day. After
an houres being with him at arithmetique, my first attempt being to
learn the multiplication table.t
If a graduate of Cambridge University was just beginning in his
thirtieth year to learn the multiplication tables, what are we to suppose
the average schoolchild knew of them? In 1542 the Welshman Robert
Recorde published The Grounde ofArtes. Teachyng the Worke and Practice
ofArithmetike,+ in which he shows how to multiply two numbers be-
tween 5 and 10.

MULTIPLY 8 BY 7

First set your digits one over the other. 8

7

Then from the uppermost downwards, and from the 8X

nethermost upwards, draw straight lines, so that they 7
make a St. Andrew's cross.
Then look how many each of them lacketh of 10, and
write that against each of them at the end of the line,
and that is called the difference.
8
7 X 2
3

I multiply the two differences, saying, "two times three

make six," that must I ever set down under the differ-
ences.
8
7 X 2
3
A

* Charles J. Finger, Pepys' Diary, p. 5.

t Ibid, p. 10.
+ See The Mathematical Gazette, London, Vol. XIV, No. 195 (July 1928),
pp. 196f.
18 Chapter Three

Take from the other digit, (not from his own), as the 8 2
lines of the cross warn me, and that that is left, must I 7 3
write under the digits. If I take 2 from 7, or 3 from 8, 5 6
(which I will, for all is lyke), and there remaineth 5,
and then there appeareth the multiplication of 8 times
7 to be 56. A chylde can do it.
Compare this technique of the sixteenth century A.D. with that of an
apprentice scribe of the iiyksos period of the Middle Kingdom, more
than 3,000 years earlier.
MULTIPLY 8 BY 7
\1 8
\2 16
\4 32

Totals 7 56

When the Egyptian scribe needed to multiply two numbers, he would

first decide which would be the multiplicand, then he would repeat-
edly multiply this by 2, adding up the intermediate multipliers until
they summed to the original multiplier. For example, to multiply 13
and 7, assume the multiplicand to be 13, doubling thus:
1 13
2 26
4 52

Here he would stop doubling, for he would note that a further step
would give him a multiplier of 8 which is bigger than 7. In this case he
would note that 1 + 2 + 4 = 7. So he put check marks alongside
these multipliers to indicate this.
\1 13
\2 26
\4 52

Totals 7 91

Adding together those numbers in the right-hand column opposite

check marks, the scribe would thus obtain the final answer. Had he
 The Four Arithmetic Operations 19

chosen 7 as the multiplicand, and 13 as the multiplier, his sum would

appear as follows :
1 7
2 14
4 28
8 56

Again he would cease doubling at 8, for a further doubling would give

16 which is past 13; then he would note that 1 + 4 + 8 = 13, and so
he would put check marks alongside these multipliers.
\1 7
2 14
\4 28
\8 56

Totals 13 91

Since 2 is not checked, he took care not to add in the 14 of the right-
hand column, where he has 7 + 28 + 56 = 91. The scribe's mental
arithmetic had to be pretty accurate, especially for large multipliers,
but of course he could keep a check of his totals on a papyritic scribbling
pad. These additions were made easier for the scribe by virtue of a
special property of the series
1, 2, 4, 8, 16, 32, ... ;

for any integer can be uniquely expressed as the sum of some of its
terms. Thus, for example, 19 = 1 + 2 + 16; 31 = 1 + 2 + 4 +
8 + 16; and 52 = 4 + 16 + 32. We do not know whether or not the
scribes were explicitly aware of this but they certainly used it, just as
do the designers of a modern electronic computer, and this is surely a
somewhat sobering thought.

DIVISION
For the Egyptian scribe, the process of division was closely allied to
his method of multiplication. Suppose that he wished to divide 184 by
8. The scribe did not say to himself, "I will divide 8 into 184." He
said, "By what must I multiply 8 to get 184?" Thus he had,
20 Chapter Three

1 8
2 16
4 32
8 64
16 128

At this stage he stops multiplying by 2, for his next doubling would give
256, which is well past 184. Now he must do some mental arithmetic
or use his memo pad to locate which numbers of the right-hand
column will add up to 184. Finding that 8 + 16 + 32 + 128 = 184,
he would place a check mark beside each of these numbers:
1 8/
2 16/
4 32/
8 64
16 128/
Totals 23 184

Then he must add the multipliers corresponding to the checked

numbers,
1+2+4+16=23,
which is his quotient.

FRACTIONS
When the Egyptian scribe needed to compute with fractions he was
confronted with many difficulties arising from the restrictions of his
notation. His method of writing numbers did not allow him to write
such simple fractions as 3/a or %, because all fractions had to have
unity for their numerators (with one exception*). This was because a
fraction was denoted by placing the hieroglyph o ("r," an open
mouth) over any integer to indicate its reciprocal. Thus the number
12, written in hieroglyphs as Ill, became the fraction 1/12 when written
as Il°(). In the hieratic or handwritten form, in which the scribe used a
* The fraction %. There is some evidence that a special hieroglyph for
3/4 existed.
 The Four Arithmetic Operations 21

reed brush and ink, the open mouth became merely a dot, and 1/12
would look like ItA. The dot being so much smaller than the "mouth,"
it was placed over the first digit of the number (here it is 10), and so in
reading numbers in hieratic papyri care must be taken not to think
that, say, nA is (1/10 + 2) instead of 1/12. Such a mistake is even more
likely with hundreds or thousands.
Thus all hieroglyphic and hieratic fractions are unit fractions (stamm-
bruchen), and have unit numerators in translation. The fraction 3/4 was
written by the Egyptians as 1/2 + 1/4, 6/7 was written as, 1/2 + 1/4 +
1/14 + 1/28, etc., for all fractions of the form p/q. To a modern arith-
metician this seems unnecessarily complicated, but we shall see that
the Egyptian scribes devised means and rules to meet the difficulties
of the method as they arose. The exception to the unit-numerator
usage-the fraction %-was denoted by a special sign: 4? in hiero-
glyphics, and Y in hieratic. There is no doubt that the Egyptians knew
that % was the reciprocal of 11/2, as the hieroglyph suggests,* for there
are many instances, particularly in the RMP, where this relation is
specifically shown. Thus :

RMP 33
Since 3 of 42 = 28,
then 28 of 42 = 1 2.

RMP 20
of 24=12x24,
= 36.

The numerator 1 of each fraction is here omitted, and 3 is %. In

multiplying and dividing fractions, the scribes used the same setting
out as they did for integers, but they needed to use various techniques
for the different problems that arose. In RMP 2, the fraction 5 is
explicitly multiplied by 10 (more often multiplications by 10 were
written down at once; see Table 4.3) :

* In earlier times, the two vertical strokes were often drawn the same
length, but this may have been merely lack of care.
22 Chapter Three

Do it thus
1. 1
1.2
1.3
1

\2
4
\ 5
$
3
15
10 30
1.4 \8 \1 $ 5 15

1.5 Totals 10 1 9 5 15 T5-

1.6 2

In line 2 the product 2 x 3 is seen to have been recognized by the

scribe as equal to the division 2 = 5, for the answer ($ 15) is that given
in the RMP Recto* Table, which lists such divisors of 2 (see Chapter
6). Lines 2 and 3 give the double of $ as 3, and 2 x F5- as (10 30),
which, again from the Recto Table, is the result of the division 2 = 15.
Lines 3 and 4 give the double of 3 as (1 5), the double of TO as 5, and
the double of TO- as 15. Check marks were put on lines 2 and 4,t and
(5 15) + (1 5 5 15) read off as the answer. The scribe would have
recognized this quantity to be equal to 2, by some papyritic jottings
on his scribbling pad, or merely by referring to his unit fraction tables
of addition.+
The constant necessity to double fractions in all multiplication gave
rise to the construction of the RMP Recto table, illustrated in Table
6.1, where unit fraction equivalents of 2 divided by the odd numbers
are recorded for the scribe's easy reference. Many of the simpler
equalities were no doubt committed to memory, just as were some of
the additions of unit fractions, as shown in the EMLR.
In RMP 9, 2 14 is multiplied by 1 2 4.
1. 1 \1 \2 14
1.2 \2 \4 28
1.3 \4 \$ T6-

1.4 Totals 1 2 4 2 4$ 14 28 56
1.5 1

* The Recto of the RMP is the first portion, dealing with the division of 2
by the odd numbers 3 to 101 (see Chapter 6). The remainder of the RMP
is called the Verso.
t Only on one column by the scribe. They are on both columns here for
ease of reading.
$ See Chapter 8.
 The Four Arithmetic Operations 23

Line 2 shows the halving of fractions by merely doubling the denomi-

nator numbers, and line 3 shows the same. Line 4 shows the totals of
both columns, and in line 5, the scribe wrote at once the answer 1.
This final addition of six unit fractions was quite possibly done men-
tally, for T4_ T8_ 56 = $ was a commonplace equality to the scribes,*
and the steps
2 4 ($ $)
2 (4 4)
(2 2)
1

would have been quite easy for a competent scribe, even though in
writing the details it appears rather long. In more difficult multipli-
cation involving fractions, as in 7 2 4 $ x 12 3 (RMP 70), which
the scribe showed to be equal to 99 2 4, he had to refer to his two-
thirds table for integers and fractions, to his rule given in RMP 61B,
to the Recto Table, and to his 2-term unit fraction tables, and he did
it accurately in six lines. It is instructive to calculate 77/s x 12%, as
we would do it today, and then compare the modern working with
that of the ancient scribe. It can be quite an enlightening comparison.

* For example, EMLR 1. 12 is 7 14 28 = 4, and simple doubling gives

82856=$.
 4 THE TWO-THIRDS TABLE
FOR FRACTIONS

The one remarkable exception to the fractions with unit numer-

ators was %, which was written with a special sign Y in hieratic
and 4? in hieroglyphic. The scribes used this fraction so freely as an
operator in their multiplication and divisions that one is led to believe
with Peet* that they must have used prepared tables-much of which
they probably knew by heart, as they did their twice-times table. This
two-thirds table was so much a part of a scribe's stock-in-trade that,
were he required to find one-third of a number, he would first find
two-thirds of it and then halve his answer, instead of simply dividing
by 3. This technique was so ingrained that we find it actually being
used for such simple operations as finding one-third of 3 and one-third
of 1! (See RMP 25 and 67.)
In this chapter we examine how the scribes could have produced
such a two-thirds table. Except for the interesting rule given in RMP
61B and a short table giving two-thirds of seventeen simple fractions
in RMP 61, no clear explanations of the Egyptians' methods of ob-
taining two-thirds of any given number, integral or fractional, has come
down to us in any of the mathematical papyri.
A careful check of the problems in the RMP shows that there are 24
examples of the scribe writing 3 of integers in one simple operation, of
which the following are illustrations:
of 27 is 18, (RMP Recto)
of 365 is 243 3, (RMP 66)
of 5432 is 3621 3. (RMP 33)
There are 14 examples where the scribe wrote down immediately 3 of
fractional numbers, such as :
3 of 3 is I8, (RMP 67)
9 of 13 23 is 8 346 138, (RMP 30)
9 of 7 2 4$ is 5 4. (RMP 70)
* T. E. Peet, "Mathematics in Ancient Egypt," Bulletin of the John Rylands
Library, Vol. 15, No. 2 (Manchester, 1931), p. 415.
 The Two-Thirds Table for Fractions 25

There are 18 examples where, in order to obtain one-third of a num-

ber, the scribe first found 3 of it, and then halved his answer. Some
examples are
3 of 8:
of 3 8 is 5 3,
hence, 3 of 8 is 2 . (RMP 43)
3 of 315:
3 of 315 is 210,
hence, 3 of 315 is 105. (RMP 67)
3 of 320:
3 of 320 is 213 3,
hence, 3 of 320 is 106 3. (RMP 38)

There are a further 17 examples of the scribe writing one-third of a

number, omitting the customary intermediate step of first writing its
two-thirds, so that we presume he did the necessary halving on an odd
piece of papyrus or that he did it mentally by reading directly from a
previously prepared two-thirds table. Thus,
3 of 4 32 is 12 96, (RMP 37)
3 of 4 53 106 212 is

12 159 31-8 636. (RMP 36)

It would be easy to say that
3 of 4 32=3x4 3x32
= 12 96,
and that the multiplication 3 x 4 and 3 x 32 were done in the usual
way. But this would be contrary to standard scribal procedure, in
which 3 was found first and then halved to give 3; thus,
3 of 4 32 = fi 48,
so 3 of 4 T2_ = 12 96.

And for the second sum, it would also be easy to suppose that 4, 93,
106, and 212 were each multiplied by 3, but the same objections still
apply. We should have
26 Chapter Four

5 of 4 53 106 212 = 9 (106 318) 159 318,

so 9 of 4 53 106 212 = 12 (212 636) 318 636.
(RMP 61B)
But this is not the scribe's answer. The scribe apparently knew that
212 636 = 159, for he wrote his answer as
12 159 318 636.

He could have gone a step further had he chosen, for

159 318 = 106,
and 318 636 = 212,

so that he could have given either of the simpler answers*

12 106 636 or 12 159 212.

However, in RMP 36 these latter refinements are of no moment to the

scribe, because these four unit fractions are only a portion of a group
of eighteen unit fractions that later have to be added up to give 1. We
are here concerned only with how two-thirds and one-third of various
numbers could have been found by the scribe, and how he might have
prepared his two-thirds table.
Now it would have been quite a simple matter for the scribe to
calculate one and one-half times the integers 1 to 10 and set them down
like this :
1 times 1=1 2
1 2 3
1 3 4 2
1 4 6
1 5 7 2
1 6 9
1 7 10 2
1 8 12
1 9 13 2
1 10 15

* See Chapters 5, 8.
 The Two-Thirds Table for Fractions 27

These numbers form two simple arithmetic progressions, which can be

extended by continual addition of 1 2. Again, this table may be re-
written in a slightly different form, in accordance with the prescription

if a x x = b, then b x x= a,
which was well known to the scribes, for there are at least 80 examples
of it in the RMP Recto alone.* Since the reciprocal of 1 2 was known
to be 3, he could have rewritten the table as:

of 1 2=1
3 2
4 2 3
6 4
7 2 5
9 6
10 2 7
12 8
13 2 9
15 10

Now the beginnings of a 3 table is emerging, and it may be extended

by inserting between each pair of consecutive terms of the progression
1 2, 3, 4 2.... two equally spaced numbers; for example, between
1 2 and 3, we insert 2 and 2 2, and between 3 and 4 2, we insert 3 2
and 4, and so on. A similar operation on the progression 1, 2, 3, .. .
would give the corresponding series 1, 1 3, 1 1, 2, 2 3, 2 1, 3, ... , so
that we can finally list the equalities shown in Table 4.1. This two-
thirds table would in general suffice for all the scribes' needs, and ex-
tended up to 100, it would enable them to find % of any number,
integral or fractional, because multiplication and division by 10 was a
commonplace operation with them.t Thus in the RMP alone, there
are 23 and 20 examples of the immediate multiplication and division,

* Examples are RMP 34: Since 4 x 1 2 4 = 7, then 7 x 1 2 4 = 4, and

RMP 32: Since 1 3 4 x 144 = 228, then 228 x 1 3 4 = 144.
t See Table 4.6, Multiplication and Division by 10 in the RMP.
28 Chapter Four

TABLE 4.1
The two-thirds table.
of 2 = 3 (RMP 61) of 8 = 53 (RMP 43)
I (RMP 67) 82 5
12 1 9 6
2 13 92 63
22 1 10 6
3 2 (RMP 25) 3 10 2 7
32 23 (RMP 69) 11 73
4 2 11 2 7
4 2 3 12 8
5 33 (RMP 46) 12 2 83
52 3 13 8
6 4 (RMP 39) 13 2 9
62 43 14 93
7 4 142 9
72 5 (RMP 70) 15 10.

respectively, by 10, of numbers like 53, 710, 79 108 324, 45 $, 365,

and the like.
RMP 40 requires I of 60, which if not read directly from the ex-
tended table, could be found as follows:
of 60 = (I of 6) x 10
= 4 x 10
=40.
RMP 66 requires the scribe to evaluate 3 of 365, the number of days
in the Egyptian calendar year:
of 365 = (3 of 36 ) x 10
=24 3 x 10
=240+3 5
= 243 9.

Although a two-thirds table for unit fractions, such as Table 4.1,

has not been preserved for us from the time of the pharaohs, such
tables did exist in later times, and some have been preserved. Indeed,
quite extensive tables were used as late as the sixth century A.D. The
Akhmfm Papyrus U. Baillet, Paris, 1892), in Greek, gives I of the num-
 The Two-Thirds Table for Fractions 29

bers 2, 3, ... , 10; 20, 30,..., 100; 200, 300,.. ., 1,000; and 2,000,
3,000,..., 10,000.
A similar table, in Coptic and Greek, and ornamented in red, green,
and yellow, occurs in Crum's Catalogue (London, 1905); this table may
date from as late as 1,000 A.D.

PROBLEMS 61 AND 61B OF THE RHIND MATHEMATICAL PAPYRUS

The two problems numbered 61 and 61B by Chace are not really
problems but tables of fractions, principally 3, 3, and 2, of other
fractions. They appear to have been written down as references for
the problems that follow them. They are not set down in any real
order, although the denominators do get larger as the table proceeds.
There is a certain similarity in this regard to the equalities of the
EMLR. Some five of the lines appear to be lost, and we can only
hazard a guess as to what these lost lines were; but if we rearrange in
slightly different order the 17 entries which we do have, we can come
close to what these equalities were.
The problem listed as 61B states very clearly the Egyptian rule for
finding two-thirds of any odd unit fraction. Chace's translation is:
The making of % of a fraction uneven. If it is said to thee, What is
% of 1/5? make thou times of it 2, times 6 of it; % of it this is. Behold
does one according to the like for fraction every uneven which may
occur.
As we shall see (line 4 of Table 4.2), the scribe knew that this rule also
applies to every "even" fraction; but he did not use it very much for
even fractions, because he had a much simpler rule for them (line 5).
By this latter rule, one merely adds one-half of the number to itself,
a technique that is used quite often elsewhere in the RMP. It must
have seemed too obvious for the scribe to state this rule in the same
detail as for the uneven" fractions; indeed, as we have observed, the
hieroglyph for I is Sr, which suggests that one and one-half times of a
fraction was something quite well known.*
There are one or two corrections and erasures in the table of
* There is an interesting line in Problem 19 of the MMP, Moscow Mathe-
matical Papyrus. "Calculate thou with 1 2 until you find 1. Result 3."
30 Chapter Four

TABLE 4.2
Lines taken from Problem 61 of the RMP.
line 1 of = 3 9
2 9 T8-
3 9 18
4 12 36
5 2 $
6 2 8
7 8 2 12
8 12 2 24
9 $ IS 34°
10 4 3 20
11 7 14 42°
12 2 7 14
13 11 22 66°
14 11 33-
15 2 11 12-
16 4 I1 44
17 5 10 30.

' The scribe could have checked from the Recto, 2 divided by the odd numbers, for
3 of I is the same as 2 = 9, which is there proved to be 6 18. See Chapter 6.
b Again in the Recto, 2 = 27 is given as 18 54. This line is repeated in the papyrus.
° As previously, the Recto gives 2 _ 21 = 14 421
e From the Recto, 2 = 33 = 22 66.

Problem 61, and line 9 is duplicated; the sequence as it now stands

(Table 4.2) contains no scribal errors.
From the equalities given in Table 4.2, we now make a table of
two-thirds of every unit fraction from 2 to 12, in which we have to
include four lines not given by the scribe (see Table 4.3).
The table for one-third of every fraction from 2 to 12 (Table 4.4)
contains only three of the scribe's entries. Indeed, the scribe did not
need an extensive one-third table, because each entry could be found
by merely doubling the corresponding entries in the two-thirds table.
Five equalities of the one-half table (Table 4.5) occur in RMP 61.
The scribe was also aware that every equality in these tables may be
stated in reverse; for example, 3 of 3 and 3 of 3 are the same (fi 18).
He also was aware that any one equality would produce another by
 The Two-Thirds Table for Fractions 31

TABLE 4.3
Two-thirds of unit fractions.
3 of 2 = 4 12 = 3
$ 3 6 i8
4 $ 24 6
5 10 30
6 12 36 $
7 14 42
16 48 T2-

$ T8- 54
10 20 60 15

11 22 66
12 24 72 18.

TABLE 4.4
One-third of unit fractions.
3 of 2 = $ 24 = 6
3 3 12 36 $
3 4 16 48 T2-
To- 60 15
3 6 24 T2- T8-
3 7 28 84 21
3 $ 32 96 24
3 9 36 108 27
3 10 45 120 30
3 Ti 44 132 33
3 12 48 144 36.

TABLE 4.5
One-half of unit fractions.
2 of 2 = 4
2 3 6
2 4 $
2 10
2 6 12
2 7 T4-
T6-
2 ?l T-8
2 10 20
2 11 H
2 12 24.
32 Chapter Four

multiplying the appropriate number on each side by 2,3,4..... Lines

6, 7, and 8 of Table 4.2 show this very clearly:
1.6 $ of 2=$,
1.7 of 2=12.
1.8 12 of 2=24.
Each line is obtained from the preceding on multiplying the first and
last numbers by 2. Again in lines 3 and 4, the same is observed:

1.3 9 of $=fi T8-

1.4 1 of 6 =12 36
Also,

1.15 of 11 =22.
2
1.16 of 11 =44.
4
That each of the , $, and 2 tables could be extended as far as he
desired was clear to the scribe, and in fact he was aware that further
tables for 4, 5, 6,. . . could as easily be drawn up, because many of the
entries would already have been included in the tables that he already
had.

TWO-THIRDS OF AN EVEN FRACTION

The following examples are from the RMP, where the scribe used the
rule for finding I of an even unit fraction, by adding to the number
its half:
3 of 2= $ Problems 16, 61, 69, 70.
8,32,70.
fi $ 18, 32, 42, 61.
12 70.
12 18 19, 32.
9 18 27 42.
24 36 20.
56 84 33.
114 771 32.
228 T42 32.
776 1164 33.
 The Two-Thirds Table for Fractions 33

AN EXTENSION OF RMP 61B AS THE SCRIBE MAY HAVE DONE IT

Since two-thirds of any odd (or even) fraction is equal to 2 times it plus 6 times
it, why not restate the rule as follows? The scribes may well have done
this. One-third of any odd (or even) fraction is equal to 4 times it plus 12
times it. Then what do we get?

of T= 4 12
S $ 24
$ 3 12 36
S 4 16 48
20 60
$ fi 24 T2_

But from simple first principles, we also have

Then we have the 2-term unit-fraction equalities:

4 12 = 3
$ 24 = fi

12 36
16 48 = 12
20 60 = 15
24 72 = 18

This is the table of equalities developed from the generator (1, 3) that
we will meet again when we discuss the G rule in Chapter 5.
By applying the scribe's rule of RMP 61B to the smaller odd num-
bers, we have
34 Chapter Four

I of 3= 6 18
3 5 10 30
3 7 14 42-
3 9 18 54
3 11 22 66

Now again from first principles, 3 of 3 can be expressed as 2 = 9, 3 of

5 as 2 = 15, and so on, so that we can rewrite the preceding table as
2-9 = 6 18
2=15=10 TO-

2 21 = T4 42
2=27-18 54
2=33=22 66

These are exactly the entries to be found in the Recto of the RMP, and
there are just 16 of them, those where the divisors are multiples of 3.
It is thus possible that the scribe checked his answers to these particular
divisions by means of this extension of RMP 61B.

EXAMPLES FROM THE RHIND MATHEMATICAL PAPYRUS OF THE

TWO-THIRDS TABLE
Considering the complexities of finding % of some numbers, which
the scribe apparently did in his head, it is surprising to see the working-
out of one-third of 3 and of 1, found in all detail in Problems 25, 67.
PROBLEM 25
3of3 = 1.
1 3
2
3 1.

PROBLEM 67
3 of 1 = S.
1 1
 The Two-Thirds Table for Fractions 35

PROBLEM 32
9of1 3 4 = 2 36.
1 1 5 4
9(618)6
(6 6) 18 (rearranging)
18
18

In Problem 32 the scribe does not show all the intermediate steps given
above; his work appears simply as follows:

1 154
1 18
2 36.
Without any explanatory steps, the scribe writes at once two-thirds of
numbers, both integral and fractional, on at least 38 occasions in the
RMP. Some of these appear quite difficult, as the following examples
will show.
PROBLEM 42
1 83618
3 5361827.
PROBLEM 33
1 16 56 679 77-6
3 10 3 84 1358 4074 1164.

PROBLEM 32
1 1 6 12 114 228
3 3 5 18 171 342.

The comparative ease and facility with which two-thirds of a com-

plicated fraction was achieved is astounding to us-more so, if we
should attempt to check the scribes' accuracy by our modern methods.
Scribal errors were rare!
RMP 61B gives the Egyptian rule for finding 3 of any odd fraction,
which is to take twice the fraction plus six times the fraction. The
36 Chapter Four

TABLE 4.6
Examples from the RMP of multiplication and division by 10. As can be
seen from the hieratic, all answers were given with no working.

2=15 15+f0 = i $ "lA >(

2+23 23 x10 = 230 '491 7R
2+47 47 x10 - 470
2+53 53 x10 = S30 r'I rA 00

2+61 6L x10 - 610 lu AJV

2+71 71 x10 = 710 14 Ay
2+79 79 x10 = 790 4L'I 96,p

2+ S9 89x10 = 890 IL°1

Pdlaw 21 1s +10 : I I IA >1

" 22 30+10 = 3 41
IN

29 10+10 = 1 A
" 30 135+10 = i S to x3o
35 3 3+10 : 3 114 1
320 LO = 32
35

39

41
4x10: 4.0
64 X 10 = 640
-
iU1.
U y1

-'
P 41 960 410 = 96 A.,,* "2 AL

M 42 79 0M 324 x 1o-1SO O iT ff it u9,).. IL.iI tWj'

43 10 3 X 10 = L06 3 IA
4+ 10 X 10 : 100 A
 The Two-Thirds Table for Fractions 37

TABLE 4.6 (continued

Probtam43 4559+1o = 45290

44 too x 10 = 1,000
44 1500+10 = 150
44 T5xto - 750
44 150 + 10 = 15

» 46 25 X 10 - 250
49 10o0x LO - 10000
» 49 10000x 10 = too 000
M 49 10000+10 1000
. sa 5 + 10 sa 2
66 166+10 = 361
69 37 X1O 35 >uI "1.15
» 70 loo x 1o - 1000
70 1M 64X 10 = 2 432
7 0 5 X 10 = 2
« 70 2432x10 = 72416
« 72 35+10 = 31
M 73 15x10= 150
« 76 22x10: 25
M 32 14x10-122
38 Chapter Four

scribes could have found out this rule from simpler 2-term unit frac-
tions, thus,
since

= 2
then
or of n = 2n 6n.

As we have seen, this 3 rule applies to all n, whether odd or even; but
the scribes see no good reason for applying it to even numbers, because
3 of 2n is obviously equal to 3 of n, and this by simple multiplication is
3n.
 5 THE G RULE IN EGYPTIAN ARITHMETIC

In attempting to establish that the Egyptian arithmeticians must have

been aware of what I have called the G rule, I will have to exhibit
many examples of 2-term equalities. By far the greatest number of
instances where the scribes appear to have used the G rule in one form
or another occur in the Recto and the 87 problems of the RMP, in the
MMP, and the KP.* However, it will be convenient to choose from
the equalities of the EMLR for purposes of illustration and demon-
stration, even though the EMLR itself will not be considered until
Chapter 9. The easy references and the small numerical magnitudes
lend themselves to such convenient analysis that I cannot ignore the
opportunity provided by the EMLR.
The G rule is, to my knowledge, nowhere explicitly stated in the
extant Egyptian papyri. However, I hope to show, from the mathe-
matical evidence at our disposal, that the scribes probably knew and
often used this rule. First, look at these 10 lines from the EMLR:
1.11 S 18= fi
1. 13 12 24 = $
1.19 24 48= 16
1.20 18 36 = 12
1.21 21 T2_ = 14

1.22 45 90 = 30
1.23 30 60 = 20
1.24 15 To- = 10
1.25 48 96 = 32
1. 26 96 192 = 64.
An intelligent scribe would certainly notice a certain simple relation
existing between the three terms of each of these equalities. The ex-
pression of this relation is the G rule. In modern mathematical terms
we may state it as follows:
G rule: If one unit fraction is double another then their sum is a
* Kahun Papyrus. British Museum, London.
40 Chapter Five

different unit fraction if and only if the larger denominator is divisible

by 3. The quotient of the division is the unit fraction of the sum.
But if such a rule were ever expressed by an Egyptian scribe, it would
have been much terser, probably something like this:
For adding 2 fractions, if one number is twice the other, divide it
by 3.
Line 11 of the EMLR illustrates the G rule:

$ i8=6.
This could be written down immediately, because 18 is twice 9, and
then 18 - 3 gives 6, which is the unit fraction sum. The other nine
lines also obey the rule in the same way.
The G rule would have been of further usefulness to the scribes
because, if the larger fraction does not give an integer when divided
by 3, then a single-term unit-fraction sum is not possible. Take for
example the addition of the two unit fractions, 5 and 10:
3 10= ?
Although 10 is twice 5, 10 _ 3 = 3 3 which is not an integer, so that
however 3 + 10 may be otherwise expressed, it certainly cannot be
written as a single unit fraction.
The G rule could have been extended by the scribes who observed,
for instance, the equality in line 3 of the EMLR. This is

4 i2=3.
One notices in this case that one of the paired unit fractions is 3 times
the other (4 is 3 times 12, or, as the Egyptians would have phrased it,
12 is 3 times 4) ; adding 1 to this multiplier (by analogy with the G
rule) and dividing into 12, we obtain the sum 3. Indeed the scribes
were able to go further. Let us look at lines 1 and 2, the first two
entries in the EMLR. They are
10 40=$,
3 20=4.
 The G Rule in Egyptian Arithmetic 41

It is a fair assumption that the scribe responsible for the EMLR put
them together because they are similar equalities. In each case, the
larger number is 4 times the other, and by adding I as before, one
obtains 5, which divided in turn into 40 and 20, respectively, gives the
sums 8 and 4. Thus the scribes could have arrived by induction at a
generalization that we can state in modem terms as :
Extension of the G Rule : If one of two unit fractions is K times the other,
then their sum is found by dividing (K + 1) into the larger number,
providing the answer is an integer.
Further, if the answer is not an integer, then a unit fraction sum is
impossible, and thus the rule becomes doubly useful.
I am sure that, over the many centuries in which the scribes added
and subtracted unit fractions, they often observed and made use of
this rule, perhaps not exactly in the form that we have expressed it.
But the principle must have been known, even though no explicit
statement of it is given in any of the extant Egyptian papyri. One thing
is certain. Anyone working today in Egyptian mathematics will find
the rule immeasurably useful in checking computations. All of the
2-term equalities shown in Table 5.1 appear to have been written
down mentally by the scribe of the RMP; of course, they may have
been checked from prepared tables.
TABLE 5.1
Instances of the scribe's probable use of the extended G rule in the RMP.
Location Equality
Recto, 2 + 17, 37, 43, 59, 73, 79, 83, 89, 95, 101. 3 6 = 2.
Verso, Problems 4, 5, 16, 38, 56, 66, 69.
Recto, 2 + 19, 23, 95. 6 12 = 4.
Recto, 2 + 59. Verso, Problems 17, 18, 67. 9 18 = 6.
Verso, Problems 30, 35. 15 30 = 10.
Recto, 2 + 59, 95. Verso, Problem 33. 4 12 = 3.
Recto, 2 + 29, 37, 41. $ 24 = 6.
Recto, 2 + 31, 67, 73, 83, 89. 5 20 = 4.
Recto, 2 + 61, 71. 10 40 = $.
Recto, 2 + 43. 7 42 = 6.
Verso, Problem 33. 14 84 = 12.
Recto, 2 + 47, 53, 79. Verso, Problems 4, 5, 56. 10 l5 = 6.
42 Chapter Five

I will now establish the standard equality, 7 14 TA_ = 4 as an exer-

cise in the application of the G rule. First multiply the left-hand side
by, 2, 3, ... .

1. 1 7 14 28
1. 2 multiply by 2 14 28 56
1.3 3 21 T2_ 9_4

1.4 4 28 56 112
1.5 5 35 70 140.

We choose the third line because it is the only one with numbers
divisible by 3. Then we have

21 42 84 = (21 42) 84
(14 84) [42 = 3 = 14.]
= 12. [84 = 7 = 12.]
Then, dividing both sides of 21 42 84 = 12 by 3 gives
7 1428=4.
The repeated multiplication illustrated here is an often-fruitful
technique for casting many-term fractions into a form amenable to
computation with the G rule. Here, although 7 divides 14, giving 2, yet
3 does not divide 14, and again, although 14 divides 28, giving 2, yet 3
does not divide 28; but multiplication by 3 gives the sum 21 42 84,
upon which the G rule may operate.

FURTHER EXTENSIONS OF THE G RULE

There are many examples in the arithmetical calculations of the
scribes where the subtractive concept with unit fractions predominates
over the additive concept. It was necessary then for them only to look
at the tables they already had for the addition of unit fractions (in,
e.g., the EMLR), from a slightly different point of view, and perhaps
even rewrite them, to produce a table for the difference of unit frac-
tions. One must remember of course that they had no specific signs
for plus and minus; mere juxtaposition of numbers meant "add," and
 The G Rule in Egyptian Arithmetic 43

if subtraction was meant, then the words in hieroglyphs or hieratic

had to be written. Although no such tables for the difference of unit
fractions have come down to us, it is instructive to examine how they
may have been derived from the addition tables, and to see what they
would have looked like. We introduce plus and minus signs in these
equalities for clarity, and, anticipating Chapter 10, we take first the
table whose generator is (1, 2)-that is, the table formed of consecu-
tive integral multiples of 1 and 2-and omit for simplicity the bars
over the numbers, for they are all unit fractions.
GENERATOR (1, 2)
3+ 6=2
6+12=4
9+18=6

We see that each entry in this table is a G-rule equality. And now we
rewrite the table as follows :
GENERATOR (1, 3)
2- 6=3
4-12=6
6 18=9

Now we can see that, with one simple alteration, the G rule applies to
the subtraction of unit fractions. Thus, for the first equality, one says
2 into 6 is 3, (3 - 1) is 2, and 2 into 6 is 3, so that the difference of the
unit fractions 1/2 and 1/6 is 1/3. This reasoning holds for every equality
of the table.
That this extension applies to all the succeeding tables can be seen
from the following:

GENERATOR (1, 3)
4 + 12 = 3
8 + 24 = 6
12+36=9
. ..,
44 Chapter Five

GENERATOR (1, 4)
3-12= 4
6-24= 8
9 - 36 = 12

GENERATOR (1, 4)
5+20= 4
10+40= 8
15 + 60 = 12

GENERATOR (1, 5)
4-20= 5
8-40= 10
12 - 60 = 15

Thus, by an examination of simple equalities, we are led to:

Further extension of the G rule: If one of two unit fractions is K times
the other, then their difference is found by dividing (K - 1) into the
larger number, providing the answer is an integer.
And again it can be said that, whether or not the ancient Egyptians
used this rule or not, anyone working today in Egyptian mathematics
will find it immeasurably useful in checking computations.
 6 THE RECTO OF THE RHIND
MATHEMATICAL PAPYRUS

THE DIVISION OF 2 BY THE ODD NUMBERS 3 TO 101

The Recto of the RMP occupies almost the first third of the roll, whose
overall length is 18 feet and whose height is 13 inches. It was copied
by the scribe A`h-mose during the period of the Hyksos or Shepherd
Kings (about 1650 B.c.) from writings made about 200 years earlier.
The papyrus is written in hieratic characters, and reads from right to
left. It contains some 87 mathematical problems. These arc preceded
by a table of the division of 2 by the odd numbers 3 to 101, the answers
being expressed as the sums of unit fractions (Figure 6.1).
For his introduction the scribe writes (Chace's translation) :*
Accurate reckoning. The entrance into the knowledge of all existing
things and all obscure secrets. This book was copied in the year 33,
in the 4th month of the inundation season, under the majesty of the
king of Upper and Lower Egypt, `A-user-Re`,t endowed with life, in
likeness to writings of old made in the time of the king of Upper and
Lower Egypt, Ne-ma-`et-REc.+ It is the scribe A`h-mose who copies
this writing.
This Recto Table (Table 6.1) is the most extensive and complete
of all the arithmetical tables to be found in the Egyptian papyri that
have come down to us. It must have been one of the most useful of all
the scribes' reference tables, and would have been regarded, we
might say, in about the same light as a modern mathematician would
regard a set of logarithm tables.
That this comparison is not an overstatement is evidenced by the
fact that many variations and extensions of the table have been found
on much later papyri, ostraca, and wooden and other tablets, dating
* RMP. A. B. Chace; L. Bull; H. P. Manning; and R. C. Archibald, The
Rhind Mathematical Papyrus, Mathematical Association of America, Oberlin,
Ohio, 1927 (Vol. 1), 1929 (Vol. 2). See Vol. 1, p. 49.
t Probably `Aweserre` Apopi, the sixth Hyksos ruler. See Gardiner, Egypt
of the Pharaohs, Oxford University Press, London, 1964, p. 158 and p. 443.
$ Nema`re` Ammenemes III of the Twelfth Dynasty, approximate dates
1842-1797, according to Gardiner, Egypt of the Pharaohs. Chace gives 1849-
1801, following Breasted, History of Egypt, New York, 1911 edition.
 The Recto of the RMP 47

from up to as many as 2,000 years after it was first written. We find, for
example, unit fraction equalities for the numbers 2, 3,..., 101, each
divided by 2, 3,..., 101, as late as the sixth century A.D. However
more sophisticated and advanced the mathematics of the Greeks,
Romans, Arabs, and Byzantines may have been, not one of these
nations over this long period of time had been able to devise a more
efficient technique for dealing with the simple common fraction p f q.
Thus the ancient Egyptian scribe, being required to divide 9 loaves
equally among 10 men (RMP 6), worked it out that each man would
receive 9 5 30 of a loaf. Again, in RMP 66, 3,200 ro* of fat are issued
for a year, and it is calculated by the scribe that this is equivalent
to using 8 10 2190 ro per day. These unit fractions may appear
clumsy, yet 2,200 years later a Greek papyrust shows us-in a quite
comprehensive table-that one-seventeenth of a silver talent (equal
to 6,000 copper drachmas) is equal to 352 2 17 34 51 drachmas.
Thus we see that the Greeks in their own arithmetical notations
retained the ancient Egyptian unit fractions. Indeed, the papyrus
material upon which the Greek table was inscribed also came from
Egypt, as probably did the brush and ink as well.
And now in the twentieth century A.D., nearly 4,000 years after the
Egyptians first devised their system for fractions, modern mathe-
maticians have tried to determine what principles and processes the
ancient Egyptian scribes used in preparing the Recto table. How
was it possible for them, with only a knowledge of the twice-times
table and an ability to find two-thirds of any integral or fractional
number, to calculate unit fractional equivalents of %, %, %,. ..,
2/ioi without a single arithmetical error? And how did it come about
that, of all the many thousands of possible answers to these decom-
positions, those recorded by the scribe of the RMP were in almost
every case the simplest and best possible, by his own prescribed
standards ?
Some of the mathematicians of the late nineteenth and the twentieth
centuries who have discussed these Egyptian unit fractions are
* A unit of measure. See Chapter 20.
t Michigan Papyrus 146. University of Michigan, Ann Arbor.
s The Greeks used the letters of their alphabet: a = 1, P = 2, y = 3, ... .
48 Chapter Six

Eisenlohr (1877), Favarro (1879), Sylvester (1880), Collignon (1881),

Shack-Schackenburg (1882), Tannery (1884), Mansion (1888),
Bobynin (1890), Loria (1892), Hultsch (1895), Simon (1907), Griffith
(1911), Vetter (1922), Vasconcellos (1923), Peet (1923), Neugebauer
(1926), Gillain (1927), Chace (1927), Vogel (1929), Van der Waerden
(1937), Hogben (1945), Struik (1948), Becker (1951), Bruins (1952),
and Vogel (1958). Here are some comments of these writers.

... the very beautiful ancient Egyptian method of expressing all

fractions under the form of a sum of the reciprocals of continually
increasing integers. SYLVESTER, 1882
Les d&ompositions sont toujours, a un point de vue ou d un autre,
plus simple que toute autre decomposition possible. (The decom-
positions are always, from one point of view or another, simpler than
any other decompositions.) MANSION, 1888
Attempts to explain it [i.e., the method of the Egyptians] ... have
hitherto not succeeded. HULTSCH, 1895
The men who designed the pyramids must have had insight into
scientific principles, hardly credited to the Egyptians from their
written documents alone. GRIFFITH, 1911

Of the discussions which I have seen, the clearest is that by Loria. But
no formula or rule has been discovered that will give all the results
of the table, and Loria expressly says that he does not attempt to
indicate how the old Egyptians obtained them. cHACE, 1927
The Recto is a monument to the lack of scientific attitude of mind.
PEET, 1931

They went to extraordinary pains to split up fractions like 2/43 into a

sum of unit fractions ... a procedure as useless as it was ambiguous.
The Greeks and Alexandrians continued this extraordinary perform-
ance. HOGBEN, 1945

All available texts point to an Egyptian mathematics of rather

primitive standards. STRUIK, 1948
Das prinzip der Berechnung scheint kein einheitliches zu sein. (The
principle of calculation does not seem to be uniform.) BECKER, 1951
 The Recto of the RMP 49

One would hardly have expected such diverse and contradictory

opinions from among such competent critics.
The usual statement made by students of the Recto decompositions
is that the scribe was seeking the "simplest" value available; but just
what constituted a "simplest" equality is seldom made quite clear.
Opinions are quite varied on the precepts, standards, or tests by which
the scribe was guided in his choice of values from the hundreds avail-
able to him. Some previous investigators have attempted to give the
scribe's possible precepts. I here present the five precepts which
I believe were the scribe's primary guide. The fifth precept has not
been suggested, to my knowledge, by any previous writer.
CANON FOR THE RECTO OF THE RMP
PRECEPT I
Of the possible equalities, those with the smaller numbers are pre-
ferred, but none as large as 1,000.*
PRECEPT 2
An equality of only 2 terms is preferred to one of 3 terms, and one of
3 terms to one of 4 terms, but an equality of more than 4 terms is
never used.
PRECEPT 3
The unit fractions are always set down in descending order of magni-
tude, that is, the smaller numbers come first, but never the same
fraction twice.
PRECEPT 4
The smallness of the first number is the main consideration, but the
scribe will accept a slightly larger first number, if it will greatly reduce
the last number.
PRECEPT 5
Even numbers are preferred to odd numbers,t even though they might
be larger, and even though the numbers of terms might thereby be
increased.
* The largest number in the Recto is 890. Of the 128 numbers of the table
only 11 exceed 500.
t There are 104 even and only 24 odd numbers used in the table.
50 Chapter Six

TABLE 6.1
The RMP Recto Table: Two divided by 3, 5, 7, ... , 101. Note that all
unit fractions are here written without overbars.
Divisor Unit Fractions Divisor Unit Fractions
3 3 53 30 318 795
5 3 15 55 30 330
7 4 28 57 38 114
9 6 18 59 36 236 531
11 6 66 61 40 244 488 610
13 8 52 104 63 42 126
15 10 30 65 39 195
17 12 51 68 67 40 335 536
19 12 76 114 69 46 138
21 14 42 71 40 568 710
23 12 276 73 60 219 292 365
25 15 75 75 50 150
27 18 54 77 44 308
29 24 58 174 232 79 60 237 316 790
31 20 124 155 81 54 162
33 22 66 83 60 332 415 498
35 30 42 85 51 255
37 24 111 296 87 58 174
39 26 78 89 60 356 534 890
41 24 246 328 91 70 130
43 42 86 129 301 93 62 186
45 30 90 95 60 380 570
47 30 141 470 97 56 679 776
49 28 196 99 66 198
51 34 102 101 101 202 303 606

In 1967 an electronic computer was programmed to calculate all

the possible unit-fraction expressions of each of the divisions of 2 by
the odd numbers 3, 5, 7, ..., 101, in order to compare the decom-
positions given by the scribe of the RMP with the thousands of possible
forms.* Such a comparison between the calculations of an ancient

* Professor C. L. Hamblin of the School of Philosophy, University of New

South Wales, kindly programmed the KDF-9 computer of Sydney Univer-
sity to do this, and supervised the production of the results. The time taken
by the computer was 5 hours.
 The Recto of the RMP 51

Egyptian scribe and the 22,295 values produced by a twentieth-cen-

tury computer, separated by a time of nearly 4,000 years, will un-
doubtedly be of great interest to historians of mathematics.
To illustrate the application of the five precepts of the canon to the
divisions of the Recto Table, let us take the division of 2 by 45. The
computer lists 1,967 possible decompositions of 2 = 45 into sums of
not more than four unit fractions, of which 7 have 2 terms, 134 have
3 terms, and 1,826 have 4 terms. Since there are seven 2-term decom-
positions, it would seem that it would be unnecessary to seek among
the 1,826 4-term values for the "simplest" value, and possibly also the
3-term values may not be needed. Indeed, none of these 3- or 4-term
values may even have come to the scribe's attention. The seven 2-term
equalities given by the computer are:

A 24 360
B 25 225
C 27 135
D 30 90
E 35 63
F 36 60
G 45 45.

Following the canon we at once rule out G, since the fraction 45 is

repeated. Next, E, C, and B are eliminated in that order, because
35 > 27 > 25, and also because they contain only odd numbers
Precepts I and 5'. Then there remain A, D, and F. Of these, the
first to go is A, for although 24 is less than 30 and 36, the 3-digit number
360 is so much greater than 90 and 60, that Precept 4 forbids the
choice. The choice then lies between D and F, and here Precept 4 is
of little help, because 60 is not sufficiently less than 90 to outweigh
the excess of 36 over 30. In decisions like this, the scribe's judgments
appear to have been flexible; further, we note that other considera-
tions favor D and not F, because

2 = 4 5 = 30 90,
52 Chapter Six

is easily derived by multiplication of the earlier values

2=15=10 30,
2= 9= fi 18

by 3 and 5, respectively.
There may have been perhaps a further reason for the scribe's
choice of D, which would apply to all divisors which are multiples of 3.
As we have seen (RMP 61B, p. 35), the scribe's rule for finding two-
thirds of any odd fraction can be briefly stated as " twice it, and six
times it." Then, since

2-45 3of15
= 30 90,

the scribe would enter this decomposition into the Recto Table. We
can conclude that, in this division as elsewhere, the computer did not
find a decomposition superior to that given by the ancient scribe.

CONCERNING PRIMES
Of the 50 divisors of 2 used in the RMP Recto Table, 25 are prime
numbers, and these are the divisors which must have given the scribe
most food for thought. An analysis of the computer output shows that
2-term decompositions are rare for primes, and that there are very
few 3-term values; so that the scribe's search for the simplest unit-
fraction equalities must have presented considerable difficulties. Thus
there are no 2-term or 3-term decompositions for the prime divisors 61,
79, 83, 89, and 101; and there are no 2-term decompositions for the
prime divisors 47, 53, 59, 67, 71, 73, and 97. There remain 13 prime
divisors for which, according to the computer, only one 2-term decom-
position is possible; all except four of these are unsuitable by the
precepts given on p. 49.
For 2 divided by 5, 7, 11, and 23, the scribe accepts the only 2-term
equalities available, but for 2 divided by 13, 17, 19, 29, 31, 37, 41, and
43 the solitary 2-term equalities are not acceptable, and he has to go
further afield. Table 6.2 shows this. The asterisks show the group to
 The Recto of the RMP 53

TABLE 6.2
Prime divisors between 3 and 101. Asterisks show the
group out of which the scribe chose his "simplest"
decomposition for each divisor.
Number of Number of Number of
Prime 2-term 3-term 4-term
divisor values values values
3 1 4 48
5 1* 8 260
7 1* 13 306
11 1* 16 367
13 1 12* 423
17 1 11* 467
19 1 16* 256
23 1* 18 368
29 1 8 203*
31 1 8* 155
37 1 6* 90
41 1 7* 179
43 1 6 117*
47 0 2* 54
53 0 1* 23
59 0 1* 19
61 0 0 7*
67 0 2* 12
71 0 2* 23
73 0 1 9*
79 0 0 3*
83 0 0 3*
89 0 0 6*
97 0 1* 7
101 0 0 1

which the scribe needed to go to find what he regarded as the simplest

or most suitable decomposition. For 2 = 3, the scribe merely put
even though he certainly knew he could have written 2 G.
FURTHER COMPARISON OF THE SCRIBE'S AND THE
COMPUTER'S DECOMPOSITIONS
2 = 5 The scribe's decomposition is 5 15, and from the 269 expres-
sions computed by KDF-9 we find that this is the solitary 2-term value
54 Chapter Six

possible. Although both numbers are odd, they are so small that they
do not induce a preference for any 3-term equality which may con-
tain only even numbers. The eight 3-term decompositions recorded
by KDF-9 are

A 3 16 240
B 8 18 90
C 8 20 9-0-

D 8 24 40
E 4 7 140
F 4 8 40
G 4 10 20
H 4 12 15.

F and G are the only ones having all numbers even, but he preferred
3 15 because Precepts I and 2, taken together, outweigh Precept 5.

2 = 7 The scribe wrote 4 28; the computer shows that of the 320
possible decompositions this is the only one consisting of only two
terms, and they are both even numbers. The scribe has therefore
chosen the simplest and best-possible value by his standards.

2 = 9 For this division, the computer lists 516 possible decomposi-

tions, of which the only ones with two terms are 5 45 and fi 18.
Since the latter has even numbers and 18 is less than 45, Precepts 1, 2,
and 5 decide that this is the simplest value. Furthermore, 6 18 is
derivable from
2-3=2 fi,

on multiplying through by 3.
Indeed, all the subsequent divisors that are multiples of 3 are
derivable by the successive multiplication through by 3, 5, 7,..., 33,
so that they are all 2-term values that conform to the canon. However,
we note that KDF-9 found other 2-term decompositions for these
multiples of 3. We therefore list these other decompositions (Table
6.3), and compare them with those written down by the scribe. Even
 The Recto of the RMP 55

if every equality in Table 6.3 had been known to the scribe, it would
still appear that nowhere has he made an obviously bad choice. In one
or two instances we might have some slight doubts, but even for these
Precept 4 vindicates the scribe's choice, the smaller first number
deciding for him. Thus,

for he chose he rejected

2 _ 15 10 30 12 20
2 _ 45 30 90 36 90-
2 = 63 42 126 56 72
2 = 75 50 150 60 100
2 -99 66 198 90 110

2 = 11 Of the 384 decompositions given by the computer, only one

2-term expression occurs, fi 66, which the scribe recorded. This is
clearly the simplest possible; both numbers are even.

2 = 13 Again there was only one 2-term decomposition, 7 91,

among the 436 listed by the computer. But the scribe would not accept
it, both numbers being odd (Precept 5). Now there are twelve 3-term
expressions available, but only four of them have three even numbers.
These are

A $ 36 936
B $ 40 260
C $ 52 104
D 10 20 260.

Of these he chose C, by virtue of Precept 1, and this is clearly the best

choice, noting the 104.

2=15 See 2-=9.

2 = 17 As was the case for 11 and 13, only one 2-term decomposition
is possible, namely 9 153, out of the 479 given by the computer.
However, the scribe rejected it. Both numbers are odd, and perhaps
we can find a 3-term equality with numbers less than 153. The eleven
56 Chapter Six

TABLE 6.3
Two-term decompositions of 2 divided by multiples of 3. Decompositions
that were chosen by the scribe are noted by an asterisk.
Total number of Number of Computer
Divisor computer values 2-term values values
15 1158 4 $ 120
tS 45
10 30*
12 20

21 1190 4 I1 231
12 84
14 42*
15 35

27 733 3 14 378
15 135
18 54*

33 1016 4 17 561
18 198
21 77
22 66*

39 894 4 20 780
21 273
24 104
26 78*

45 1967 6 24 360
25 225
27 1 55

30 90*
35 63
36 60

51 595 3 27 459
30 170
34 102

57 645 3 30 570
33 209
38 114*

63 1607 6 33 693
35 315
36 252
 The Recto of the RMP 57

TABLE 6.3 (continued)

Total number of Number of Computer
Divisor computer values 2-term values values
63 1067 6 T2- -12-6

45 105
56 72

69 500 3 36 828
39 299
46 138'

75 884 6 39 975
40 600
42 350
45 225
50 150'
60 100

81 339 2 45 405
54 162'

87 102 2 48 464
58 174'

93 58 2 5I 527
62 186'

99 710 5 34 594
55 495
63 131
66 198'
90 110

3-term values are

A 9 204 6-1-2-

B 9 254 442
C 10 65 442
D 10 68 340
E TO- T5- 1 00

F 10 90 153
G T2- T4- 204
H T2- T6- 153
I I2 51 T8
J 13 26 442
K 17 18 306.
58 Chapter Six

Of these, D and G are the only ones composed wholly of even numbers,
and if Precepts 4 and 5 are considered, one would expect the choice
to fall on G. But it does not! Precept 1 on smaller numbers must have
prevailed, for the scribe selected I, even though one of the numbers is
odd. We cannot know whether the scribe was aware of all these pos-
sibilities, but whether he was or not, he certainly found the only 3-term
decomposition consisting of two-digit numbers. Even if he had looked
among the 467 4-term decompositions, he would have found only 3
consisting of two-digit numbers, but all with numbers much greater
than 12 51 68; so that, however he did it, we can only compli-
ment him on an amazingly successful search.
2 = 19 The only 2-term decomposition here is 10 190, and the
scribe must have thought hard before he rejected it. Of the sixteen
3-term values available, only five consist wholly of even numbers.
These are
A 10 240 912
B 12 48 912
C 12 CO 1 00

D 12 76 114
E 16 24 912.
Precept I must have prevailed, for the scribe chose D, as having the
smallest numbers, for although 10 is less than 12, 114 is much less than
190 (see Precept 4).
2=21 See2-9.
2 _ 23 Another prime number divisor, and consequently only one
2-term decomposition is possible (12 276), which the scribe at once
accepted. All of the eighteen 3-term decompositions have last terms
much greater than 276, including the seven that contain only even
numbers, so that whether he looked further afield or not, he chose here
the simplest expression.
2 = 25 The computer records 619 values, of which
A 13 325
B 15 75,
 The Recto of the RMP 59

are the only 2-term expressions, from which the scribe naturally chose
B, primarily no doubt because it is at once obtainable from 2 = 5 =
3 15, upon multiplication by 5. If the scribe had sought among the
28 possible 3-term decompositions for even numbers, he would have
found only 7, all with much greater numbers,
A 14 140 700
B 14 200 280
C 16 80 210
D 20 36 450
E 20 40 200
F 20 50 100
G 24 30 OR
However, if he had been aware of these 3-term expressions, the de-
composition F must certainly have tempted him, but 15 and 75 are
less than 20 and 100.
2-27 See 2=9.
2 = 29 Of the 212 decompositions given by the computer, we find
1 containing 2 terms, 8 containing 3 terms, and 203 containing 4 terms.
The solitary 2-term decomposition is 15 435, where both terms are
odd and 435 is fairly large, so the scribe looked further afield. The 3-
term values are
A 16 232 464
B 16 240 435
C 18 87 522
D 18 90 435
E 20 58 580
F 20 60 435
G 24 40 435
H 29 30 870.

But he would have none of these either. All have numbers as large as
435 or larger, and all except A and E contain odd numbers. Perhaps
he could do better among the 4-term equalities! Whether he knew of
all these possibilities we cannot of course tell, but we do know that by
60 Chapter Six

the standards of the canon he found the very best available. For of the
203 decompositions listed by KDF-9, only three contain numbers less
than 300. They are
A 24 58 1 44 232
B 29 42 1 44 203
C 29 58 87 174.
By all the precepts of the canon he must choose A, and this is indeed
the equality recorded in the Recto. We cannot but admire the skill
with which the scribe found the equality with the smallest even num-
bers from the 212 available, and wonder just how he did it.
2 _ 31 Again a prime divisor with only one 2-term value, 16 496,*
and although both terms are even numbers, 496 is much too large and
perhaps the scribe can do better. Of the 163 other possibilities, 8 only
have 3 terms, and from these he was able to locate 20 124 155,
which has easily the smallest numbers, although he had to accept the
odd number 155. Those decompositions with all even numbers are
A 18 144 496
B 20 80 496
C 24 48 496,
which we would expect him to have rejected, for he had already
rejected the simpler 16 496 because of the large 496.
2 = 35 This division by 35 is the only division in the whole of the
Recto in which the scribe gives us any inkling of his method.t For
2 = 35, the computer lists 1,458 possible unit-fraction decomposi-
tions, of which only four contain two terms. These are
A 18 630
B 20 140
C 21 1 55

D 30 42.

* In a table of fractions dating from Greek times, published by Crum,

Coptic Ostraca, London, 1902, p. 46, later discussed by Sethe, Von Zahlen
and Zahlworten bei den alien Agyptern, Triibner, Strassburg, 1916, occurs
2 - 31 = 31 62 93 186. Compare 2 = 101, RMP Recto.
t The method will be discussed in detail in Chapter 7.
 The Recto of the RMP 61

Both on the score of smaller numbers and of even numbers, his choice
should have fallen on D, and this is indeed the equality given in the
Recto. There would have been no need to look at any of the other 1,454
values, nor to have considered deriving B from (2 = 7) x 5 =
(4 28) x 5 = (20 140), or C from (2 = 5) x 7 = (3 15) x 7 =
(21 105), although, of course, he may have checked on these.
Clearly, of the 1,458 possibilities, the scribe chose the simplest.

2 = 37 For this prime divisor only 97 answers are possible. The 2-

term decomposition is 19 703, which is clearly unsuitable, because
both numbers are odd, and 703 is far too large. Of the six 3-term
answers, there is only one having all numbers even, 20 370 740,
but these numbers are even larger than before, so it was also rejected.
From among the remaining five, the scribe looked for the equality
with the smallest numbers, and so selected 24 1 296, being forced
11

to accept one odd number; but this is consistent with his usual pro-
cedure.

2-39 See 2=9.

2 - 41 In this division by a prime number the scribe really shows
his capabilities in handling unit fractions. The sole 2-term decom-
position is 21 861, which he immediately rejects. There remain 186
other decompositions, of which only 7 consist of 3 terms. Now four of
these have 861 as their highest term, one has 902 as its highest term,
and another has 984, all large numbers, and in addition, other terms
are odd. The sole remaining 3-term expression is 24 246 328, by
far the simplest of the lot, for all the 4-term decompositions listed by
KDF-9 contain very much larger numbers. From the 187 decom-
positions available, the scribe certainly located that one which by his
standards clearly is the best offering, and he is to be complimented on
his success, whether accidental or not.

2 = 43 The scribe's effort here is truly amazing. From the 124

decompositions listed by the computer, those having two or three
terms are
62 Chapter Six

A 22 946
B 23 506 946
C 24 264 946
D T4- 4
344 516
E 26 143 946
F 30 86 645
G 33 66 946.
Now B, E, F, and G contain odd numbers, and so are not acceptable,
and although A, C, and D have only even numbers, they are far too
large for the scribe's purposes, even though A contains only two terms.
To seek smaller numbers, the scribe needed to search among the 117
4-term equalities, and he would have had to accept odd numbers.
Looking through this group, we find that 83 of them have their fourth
term greater than 900, three of them have their fourth term greater
than 800, one of them greater than 700, eighteen of them greater than
600, and eleven of them greater than 500. This is a total of 116, so that
one only remains. This sole remaining 4-term equality has its fourth
term 301, and although this is an odd number, we find that 42 86
129 301 has the smallest numbers of any of the possible values. The
smallest high number contained in any of the other 123 possible
answers is 516, whether they be 2-term, 3-term, or 4-term equalities,
and one can only remain lost in hopeless admiration of the ancient
Egyptian scribe, who could, with the meager arithmetical tools at his
disposal, so unerringly locate this value among the 124 available.
2 45 See 2 = 9 and pages 51-52.
2 = 47 Here no 2-term decompositions are possible, and only two
3-term equalities occur in the computer list, which totals 56 entries.
The 3-term decompositions are
A 28 188 658
B 30 141 470,
from which the scribe selected B because of the much smaller 470,
despite 141 being odd. However, he may have had some doubts here,
for there is little to choose between them. Precept 4 no doubt decided
for him.
 The Recto of the RMP 63

2 = 49 Of the 371 possible decompositions, the sole 2-term one is

28 196, with both numbers even and not too large. This is the value
given by the scribe, and it could have been obtained from 2 = 7 =
4 28 multiplied by 7. There are thirty 2-term values possible, all
with very high numbers, but probably none of them even came to his
attention. His choice here was clearly the best available.
2-51 See2-9.
2 - 53 The computer located only 24 possible values for 2 = 53,
none at all consisting of two terms, and only one, 30 31-8 795, of
three terms, and this is the value the scribe chose for the Recto. There
was certainly not much to choose from in this division, for of the
twenty-three 4-term equalities, all contain very high odd numbers
except 42 106 318 742, which he rejected by virtue of Precept 4, but
which he might have accepted by virtue of Precept 5. If this equality
did come to his attention, he must have thought very deeply before
deciding.
2 = 55 There are 1,128 decompositions of 2 = 55, of which 1,052
have 4 terms, 73 have 3 terms, and only 3 have 2 terms. The numbers
in this last group are so much smaller than all the others that the scribe
would not have fared any better if he had looked elsewhere, and so
his choice was reduced to
A 30 330
B 33 165
C 40 88.
Choice B would have been rejected because of the odd numbers,
which are inevitable, as it is derived from 2 -4- 5 = 9 l5 multiplied
by 11. However, A comes just as easily from 2 = 11 = fi 86 multiplied
by 5, and it contains only even numbers. If C did come to the scribe's
attention, Precept 4 may have swayed him in favor of A, for 30 being
less than 40 is his main consideration, even though 330 is much greater
than 88. We can imagine some queries in the scribe's mind, but the
easy derivation from 2 = 11 no doubt persuaded him to choose A.
2 =57 See2--9.
64 Chapter Six

2 _ 59 Here the scribe and the computer at once concur. Of the 20

possibilities, 19 contain 4 terms, not one of which is composed wholly
of even numbers. Indeed they average two odd numbers each, and
they are all large numbers. The sole 3-term decomposition is 36 F36-
53 1, which, although it has one odd number in it, is still the simplest
expression available, no other equality having its first term less than
36 nor its last term less than 531. No 2-term equality is possible for
2 - 59. This is the decomposition the scribe records in the Recto.
2 = 61 Agreement between the scribe and the KDF-9 is even more
obvious here. There are only 7 possibilities, all of which contain 4
terms, and only 2 of these consist wholly of even numbers. They are
A 40 244 488 610
B 48 1 22 366 976.

Since in A both the first and last terms are less than in B, this is the
value chosen by the scribe. It is clearly the best available.
2=63 See 2=9.
2 _ 65 There are 865 decompositions listed by the computer, but
only 3 of these contain 2 terms. They are
A 35 455
B 39 1 55

C 45 117.

All numbers are odd, so none has preference on that score. But B is
obtainable directly from 2 _ 5 = 3 15 on multiplication by 13, and
this probably weighed with the scribe when comparing C, considering
Precept 4 of the canon. Had he looked for 3- or 4-term values he would
have fared no better; the best offering is 2 - 65 = 60 130 156.
2 = 67 No 2-term decomposition exists out of a total of twenty-one,
and there are only two 3-term expressions, which are
A 40 335 536
B 42 201 938.

Both contain one odd number; but in A both the first and last terms
 The Recto of the RMP 65

are smaller, and so the scribe chooses A. Had he looked at the 4-term
expressions, he would have found only one without odd numbers,
42 268 804 938, of which the numbers are far too large to interest
him. He thus chose the best possible decomposition.
2-69 See 2=9.
2 - 71 Of 25 possible equalities, all except two have 4 terms, and
these are
A 40 568 710
B 42 426 497.

The scribe at once chose A, 40 being less than 42, and although 497
is less than 710, it is an odd number. Had the last term of B been even,
he would no doubt have selected it by virtue of Precepts 4 and 5.
2 = 73 The scribe had a poor selection to choose from here, for of
the 10 possible decompositions, all except 44 292 803 have 4 terms
with large numbers, 876 occurring twice, 803 five times, and 703 once.
In addition they all contain odd numbers. The scribe chose the one
remaining expression, 60 219 292 365, which is the one with the
smallest numbers. By his standards he certainly found the best
decomposition available.
2=75 See 2=9.
2 - 77 Of the 741 decompositions generated by the computer, only
3 have two terms :
A 42 462
B 44 308
C 63 99.

Choice C can be rejected because of its odd numbers, even though 99

is less than 308. Choice A is derivable from 2 = 11 = 6 66 on mul-
tiplying by 7, and B from 2 = 7 = 4 28 on multiplying by 11, and
so in conformity with Precepts 2, 4, and 5, we, as did the scribe,
select B, clearly the best choice available, without seeking further
among the 738 other 3- and 4-term possibilities.
66 Chapter Six

2 = 79 The prime divisor 79 has the least number of decomposition

of all divisions except 101, namely, 3:
A 60 237 31-6 790
B 60 158 790 948
C 79 99 711 869.

By Precept 1 the scribe should have selected A because of the smaller

numbers, even though one of them is odd. This the scribe did; it is the
best available choice.
2=81 See2=9.
2 _ 83 Like 79, the divisor 83 permits of only 3 decompositions:
A 56 332 581 664
B 60 332 415 498
C 60 249 415 996.

All contain odd numbers, and the only test applicable here is Precept 1
on the smallness of numbers, and unerringly the scribe found and
recorded choice B in the Recto Table.
2 = 85 Of 290 possible values, 255 have 4 terms, 32 have 3 terms,
and 3 have 2 terms. These last are
A 45 765
B 51 235
C 55 187.

All six numbers are odd, but A has a very large second term, so the
scribe's choice lay between B and C, and Precept 4 points at once to B.
Furthermore, B is easily derivable from 2 - 5 = 3 15 on multiplying
by 17, so that the scribe had the simplest value in the Recto Table.
2=87 See2-9.
2 = 89 The computer records only 6 possible values:
A 54 594 801 979
B 55 495 801 979
C 60 356 534 890
 The Recto of the RMP 67

D 63 231 801 979

E 63 267 623 801
F 66 1 88 801 979.
All have four terms, and all numbers are high. A, B, D, E, and F have
respectively 2, 4, 4, 4, and 2 odd numbers, while the remaining de-
composition C is the only one consisting wholly of even numbers, and
this is indeed the expression that the scribe recorded in the Recto.
The denominator 890 in the last term of this decomposition is the
largest number occurring anywhere in the Recto Table.
2 _ 91 Since 91 = 7 x 13, and thus is not a prime, undoubtedly
the scribe looked at the values derivable from his earlier divisions by
7 and 13. He would have found from 2 = 7, (4 28) x 13 = (Y2- 364),
and from 2 = 13, (7 91) x 7 = (49 637). Now if this was all he did,
he would quite naturally have selected 52 364 according to the canon,
and that would have been the end of the matter. But he looked further
afield for something better, and he found it. The computer records 216
possibilities, of which 185 have 4 terms, 28 have 3 terms, and 3 have
2 terms. These last are
A 49 637
B 52 364
C 70 130.
We already know how he could have found A and B, but how did he
find the only other 2-term equality C, in which 130 is less than half
364? By Precepts I and 4, he has certainly found the simplest value
in C, however he did it,* for a careful examination of the 214 remain-
ing 3- and 4-term values shows that in every case, the last term is more
than double 130, and in most cases is 3, 4, 5 and even 6 times it.
Whether the scribe even bothered to look at any 3-term or 4-term
values we shall never know; what we do know is that if he had, he
would have found nothing as good as the equality recorded in the
Recto Table.

* We can show by algebra that 2 = ab (where a and b are both odd, and
a + b = 20) equals 10a + T. Then 2 - 19 is 10 190, 2 = 51 is 30 170,
2 = 75 is50 150,2 = 91 is 70 130,and2 -- 99is90 110.
68 Chapter Six

2=93 See2-. 9.
2 = 95 There are 148 values recorded by the computer, of which
116 have 4 terms, 29 have 3 terms, and 3 have 2 terms. These last are
A 50 950
B 57 285
C 60 OR
Now by the canon, C should have been his immediate choice here; for
950 in A is far too large a number, and 57 and 285 in B are both odd.
But this is not the value the scribe gives. So far, we have been unable
seriously to challenge the scribe's choice of values in the Recto Table;
but here, perhaps, he faltered. "Even Homer nodded" on occasions.
The equality the scribe records is
60 380 570,
a 3-term value which is in fact equivalent to
60 228,
for 380 570 = 228, which he should have known.* What he must
have done here was to note that 95 = 5 x 19, then looked for 2 = 5
multiplied by 19, and also 2 _ 19 multiplied by 5; the same technique
as we judge him to have adopted for 2 = 91. From 2 = 5 = (3 15)
he has, on multiplying by nineteen, (57 285), both odd numbers. From
2 = 19 = (12 76 114), he has, on multiplying by five, (60 380 570),
all even numbers, and as this plan worked for 35 and 55, why not for 95 ?
So that must have been what he did ! We cannot of course be sure that
he did not search among the 3-term decompositions having even
numbers; but if he had, he could have found, as KDF-9 shows us,
A 56 532 760
B 76 152 760
C 76 160 608
D 76 190 380
E 80 190 304,

From his tables. See the EMLR 1.18 or the G rule. Since 10 15 = 6,
20 30 = 12, and 380 570 = 228, multiplying by 2 and 19.
 The Recto of the RMP 69

and no doubt D and E would have tempted him, by virtue of Precept

1, while Precept 4 may have made him hesitate. We can scarcely say
that the scribe made a bad choice for 2 = 95; we can only say that
he might have expressed his answer more concisely, and that it is a
great pity he did not check with his tables.
2 = 97 The only possible values for 97 are
A 56 679 776
B 60 679 7'76 840
C 63 504 679 776
D 64 448 679 776
E 70 280 679 776
F 72 252 '
6 79 776
G 84 168 679 776
H 88 154 679 776.

None is composed wholly of even numbers, and 776 is the least last
number in all cases. Then A is clearly his best choice, for 56 is the
smallest first term (Precept 4), and a 3-term decomposition is prefer-
able to a 4-term one (Precept 2). The scribe therefore wrote 56 679
776 in the Recto Table.
2=99 See2-9.
2 = 101 There is only one possible decomposition for 2 divided by
the prime number 101. It is 101 202 303 606, which KDF-9 and
the scribe both gave. It is derivable from the simple decomposition
2 9 fi = I that was very well known to the scribe. If this is rewritten
as I 2 3 6 = 2, it is possible to produce a whole new Recto Table,
consisting entirely of 4-term expressions,
2- 3= 5 fi 18
2= 5= 5 10 15 30
2= 7 = 7 14 21 42
2 = 9 = 9 18 27 Y4-

of which the scribe was well aware, but which he did not want; too
many terms, too many odd numbers!
70 Chapter Six

The conclusion to which we are led regarding the Recto Table as

a whole cannot be better expressed than in the words of P. Mansion
(quoted on p. 48), Summing up, all the decomposition of the Recto Table,
from one point of view or another, are the very simplest of all the decomposition
possible.
E. M. Bruins* thinks that the only values given by the scribe which
could have been improved upon are those for 2 divided by 13, 53, 61,
71, and 89, and he notes that 2 = 95 is reducible to a 2-term equality.
He further notes that all except four of the 25 prime divisors follow
the framework
2_p=N kp mp np,

wherep is a prime and N, k, m, and n are integers. But in fact, the whole
25 are so expressible, with mp, np omitted. Indeed, the whole fifty
equalities follows this framework, if mp, np may be omitted, with
slight variations for 2 - 35 and 2= 91. 1 cannot agree that the
scribal values for 2 divided by 13, 61, 71, and 89 could have been
improved upon by the canon I have postulated.
* E. M. Bruins, "Ancient Egyptian Arithmetic," Ken. Nederland Akademie
van Wetenschappen, Ser. A, Vol. 55, No. 2 (Amsterdam, 1952), pp. 81-91.
 7 THE RECTO CONTINUED

EVEN NUMBERS IN THE RECTO: 2 = 13

The observation that the scribe preferred even numbers to odd num-
bers in his choice of equalities for the Recto of the RMP arose from
pondering on the scribe's preference for $ 52 104 over 7 91 in the
division of 2 by 13. For Precept 1 (page 49) calls for the smallest
numbers, and Precept 2 says that a 2-term value is preferred to a 3-
term value, so that at first glance one would have expected the scribe
to have selected 7 91.
Clearly, the main purpose of the Recto Table is to facilitate divisions
involving fractions. Thus the unit fraction values of 3, 4, 5, ...., 12
divided by 13 can be most expeditiously found from the even decom-
positions of 2 = 13, as I will now show. Thus we can find quite neatly
the divisions
1=13 13 from:
2 = 13 $ 52 104 The Recto
3=13 8 13 52 104 3=1+2
4=13 4 26 52 4=2x2
5=13 4 13 26 52 5=1+4
6=13 4 $ 13 104 6=2+4
7=13 2 26 7=1+6
8=13 2 T3_ F6 8=2x4.
We need not go past 8 at this stage, because from 9 - 13 onwards the
fraction 3 can be introduced, and so the technique changes slightly.
Now we note that if the value 2 - 13 = 3 91 had been chosen, the
previous table would have appeared as
1=13 13 from:
2 - 13 7 91
3 = 13 7 13 91 3= 1+ 2
4 = 13 (4 28) (70 130) 4 = 2 x 2.
At this stage the scribe would have had to stop, for although 2 _ 7 =
4 28 had already been found, the equality 2 = 91 = 70 130 had not
72 Chapter Seven

yet been determined. He would have been in the same predicament

had he put 4 = 13 = (1 + 3) = 13; in either case he would have
been forced back to a tedious conventional division in order to
evaluate 4 - 13:
1 13
6 2
\4 \3 4
\26
\52 \ 4

Totals 4 26 52 4.

We conclude, therefore, that a preference for even numbers might

well have been a very important factor in the preparation of the Recto
Table as a whole. That there was a preference can be seen from an
examination of Table 7.1.
We also observe this preponderance of even numbers in the follow-
ing decomposition:
For the scribe prefers
2_9 fi 18 to 5 45-
2 = 33 22 66 21 77
2 = 35 30 42 21 105
2 = 55 30 330 33 T.
2 _ 13 $ 52 104 to 7 91
2 = 41 24 246 328 21 861
2 = 71 40 568 710 42 426 497
2 = 95 60 380 570 57 285.

2 = 29
2 = 61
24
40
58
244
-
-174
488
232 to 29
610 61
58
122
87
183
174
366
2 = 89 60 356 534 890 89 178 267 534
2 = 79 60 237 316 790 T9_ -15-8 237 474.

In this group of decompositions preferred by the scribe, only one unit

fraction is odd. In the complete table we find that the even unit
 The Recto Continued 73

TABLE 7.1
Comparison of the frequency of appearance of even and odd unit fractions
in the Recto Table of the RMP.
Number of
equalities Number of Number of Number of
in the 2-term 3-term 4-term
Recto Table equalities equalities equalities
50 29 13 8
Number containing
only even numbers 33 25 5 3
Number containing
odd numbers 17 4 8 5
105 even, 50 even, 31 even, 24 even,
Total of numbers 24 odd, 8 odd, 8 odd, 8 odd,
used = 129 = 58 = 39 = 32

fractions exceed the odd by 5 to 1. We may therefore fairly include

this fifth precept in the scribes' canon for the RMP Recto Table.
In the light of the foregoing, consider the following statement by
F. Ll. Griffith :

Egyptian fractions were all primary, having numerators 1 except 2/3

Then to express 9/13, they were obliged to reduce it thus.
9113 = 2/13 + 2/13 + 2/13 + 2/13 + 1/13,
= 4 x 2/13 + 1/13,
= 4(1/8 + 1/52 + 1/104) + 1/13, (Recto, 2 = 13).
= 1/2 + 1/13 + 1/26 + 1/13,
= 1/2 + (1/13 + 1/13) + 1/26,
= 1 /2 + (1/8 + 1/52 + 1 / 104) + 1/26, (Recto, 2 = 13).
9/13 = 1 /2 + 1 /8 + 1/26 + 1/52 + 1/104.*
Griffith continued, "This operation was performed in the head, only
the result being written down, and to facilitate it, tables were drawn
up of 2 divided by the odd numbers." t It is necessary to correct an
impression that might arise regarding the apparent complexity of
Egyptian fractions, if what Griffith offers as a decomposition of 9/ 3
* From Griffith's article "Egypt," Encyclopaedia Britannica, 11th edition,
1910-1911.
t Ibid.
74 Chapter Seven

is in fact how the scribes would have done it. There can be no question
as to Griffith's erudition as an Egyptologist, but his familiarity with
classical Egyptian arithmetic must here be questioned. Any com-
petent scribe would have found a much simpler value for a/i3 in terms
of unit fractions. Furthermore, he would have found it much more
expeditiously than Griffith suggests, by the following simple division
of9 by 13:
1 13
8

\39
3
\39
$

Totals 5 39 9.
Then, a/13 = 39. Or again, the scribe may have divided a little
differently :
1 13
\2 \6 2
4 3
2 fi
\39
Totals 2 6 39 9.
Then, 9/13 = 2 6 39. These two simple reductions to two-term and
three-term equalities serve to show that, despite the restriction to unit
numerators imposed by their notation, the Egyptians developed a
powerful technique for such reductions, not wholly dependent upon
the table of the RMP Recto, as Griffith implies. Furthermore, they
have a greater range of answers according to their choice of multipliers,
so that they could obtain the 4-term equality 9/13 = 2 $ 16 208, and
even a 5-term decomposition 9/13 = 2 13 14 35 65, but with smaller
denominators than those found by Griffith. With the Egyptian's well-
known preference for the fraction 3 whenever possible, I suggest the
scribe's value for 9/13 would have been 3 39.

MULTIPLES OF DIVISORS IN THE RECTO

The first entry in the Recto is 2 = 3 = , which could have been
written, as the scribe well knew, as
 The Recto Continued 75

2-3=2 6.

Now there are 17 divisions of 2 by the divisors which are multiples of

3, namely, 3, 9, 15,... 99; these are all obtainable from the first
equality on multiplying through by 3, 5, 7,..., 33, giving the entries
that are in the RMP:
2 divided by resulting equality 2 divided by resulting equality
3 6 57 38 114
9 6 18 63 42 126
15 To- 3-0 69 46 138
21 14 42 75 50 150
27 18 34 81 54 162
33 22 66 87 58 174
39 26 78 93 62 186
45 30 90 99 66 198.
51 34 102
After one or two of these multiplication, it would become evident
that these three columns of numbers form three series, so that it would
be easier to keep adding 6 to the left-hand sides and then 4 and 12 to
the two right-hand columns. Having his answers already prepared,
they are proved to be correct in exactly the same way, as illustrated,
for example, by 2 = 51 from the Recto:
1 51
9 34
\34 \1 2
\102 2

Totals 34 102 2.
This same sequence for this particular group of 17 divisions is also
obtainable using the rule given in RMP 61B (page 29), and it is
quite possible that the scribe obtained his answers this way. Thus
2- 3=3of1=2x1+6x1= 2 6
2- 9=Nof3=2x3+6x3= fi 18
2=15= of 5 = 2 x 5 + 6 x 5 = 10 30
2=21 x 7 + 6 x 7 = 14 42
76 Chapter Seven

There are 7 divisions of 2 by multiples of 5 (omitting those like 15,

which is also a multiple of 3), and from 2 - 5 = 3 15 we obtain
similarly by continuous multiplication the equalities
2_ 5= 3 15
2 = 25 = 15 75 (x 5)

2 = 35 = 21 105 (x 7) (but Recto has 30 42)

2 _ 55 = 33 165 (x 11) (but Recto has 30 330)
2 = 65 = 39 195 (x 13)
2=85=51 235 (x 17)
2 = 95 = 57 285 (x 19) (but Recto has 60 380 570).
We have already noted A`h-mose's poor choice in 2 - 95. In a similar
manner we find for multiples of 7,
2_ 7= 4 28
2=49=28 196 (x 7)
2=77=44 308 (x 11)
2 _ 91 = 52 364 (x 13) (but Recto has 70 130).
The multiples of 11, namely 33, 55, 77, and 99, have already been
considered as multiples of 3, 5, and 7; of the multiples of 13, namely
39, 65, and 91, there remains only 91 so far unconsidered. Then,
2 = 13 = 8 52 104, or 7 91
2 91 = 56 364 728 (x 7), or 49 637 (x 7).
But the Recto has 2 = 91 = 70 130.
We need go no further, because higher multiples of 17, 19, 23, 29,
etc., have already been considered or are greater than 101. It will be
noticed that for 2 = 91 the scribe entered a decomposition superior
to either one found by multiplication of earlier expressions of 2 = 13.
There is a further method by which the equalities for divisors which
are multiples of 3 could have been found, that utilizes the tables of
2-term unit fractions:
2-3= 3 3
= 3 (6 6)
= (3 6) 6
_ 6;
 The Recto Continued 77

2-9= 9
(18 18)

_ (S 18) 18
F=mi 6 18;

2 _ 15 = 15 15
= 15 (30 30)
= (15 30) 30
10 30;

2 _ 57 = 57 57
= 57 (114 114)
= (7 7114) 114
38 114.

This technique can be repeated up to 2 = 99.

TWO DIVIDED BY THIRTY-FIVE:

THE SCRIBE DISCLOSES HIS METHOD
In every one of the fifty divisions of the Recto, the scribe wrote his
equality on the top line in contrasting colors, the fractions constituting
the answers being in red, and all the rest in black. In 2 = 35 alone the
scribe added onefurther explanatory line, again in contrasting red and black,
which gives us the clue to his method, at least for the divisor 35. (What
we have here is, in fact, an example of the scribe's use of the so-called
red auxiliaries. In order to understand the division 2 = 35 we must
anticipate the discussion of Chapter 8, where the uses of the red
auxiliaries are treated in detail.)
The first line of Figure 7.1 shows the two fractions 30 and 42 in red
(reading from right to left). In the second line, there is first the number
6 in red, underneath the black divisor 35; this is followed by the
numbers 7 and 5 in black, placed underneath the red fractions 30 and
42, respectively. It is this second line, which never occurs in any other
division of the Recto, that enables us to understand and explain how the
scribe arrived at the answer of 30 42 for the division of 2 by 35. We
rewrite 2 = 35 in modern form from left to right, using boldface type
for those numbers that the scribe drew in red:
 78 Chapter Seven

35 30 of 35 = 1 6 42 of 35 = 6
6 7 5.

By this the scribe means that since

(30 of 35) + (42 of 35) = (1 + 6) + +

= 2,
then

30+42=2=35.
The red 6 placed underneath the divisor 35 is the number he has
chosen as the most convenient multiplier of 35, giving 210 which we
may regard as the Egyptian counterpart of our modern "common
multiple," which may or may not be at the same time the "least com-
mon multiple" so familiar to us. Then somewhere or other, perhaps
on a sort of memorandum papyrus pad, the scribe multiplied by 2 this
red 6, giving 12, which, in turn, he had to partition into two, three, or
perhaps more parts that will each divide 210 without remainder. In this

4I v'
yV . 0
IR T 4 I.
its
MA I0 $ 1 1 AAA

111
0
AM

FIGURE 7.1
The division 2 - 35 from the RMP. The red auxiliaries of the original
are shown here enclosed in boxes. The hieroglyphic transliteration is shown
below the hieratic version. Read from right to left.
 The Recto Continued 79

case, he found that he could do this with two parts, i.e., he found 12 =
7 + 5, with 210 = 7 = 30 and 210 = 5 = 42, and these are the
numbers, expressed as unit fractions, that he wrote in red above the
7 and 5 in the papyrus. Then in his two lines of proof he shows that
30 of 35 is indeed 1 6, and that 42 of 35 is 3 6. Of course the number
12 could have been partitioned differently, as, for example, 12 =
10 + 2, and the resulting equality would have been 21 + 105 = 2 -=
35. No doubt the scribe tried this and rejected it, preferring the former
equality. No other partitioning of 12 will give two numbers which will
divide 210 without remainder, but there could be other partitionings
into three or four parts, such as 12 = 7 + 3 + 2, leading to 30 + 70
+ 105 = 2 - 35, which the scribe would certainly have rejected,
had it come to his attention.
It is tempting to conclude that the scribe's technique as shown in
RMP 2 _ 35 was his standard method in all fifty divisions of the
Recto. But this was not the case, as a close examination of the various
divisions discloses; it appears that wherever possible he used it,
yet as he progressed he treated each division on its merits, finding real
difficulties with the prime divisors 29, 43, 61, 73, 83, and 89. The
somewhat controversial division of 2 by 13 illustrates one of his differ-
ent attacks in the divisions of the Recto. This is the calculation of
2 - 13 as it occurs in the Recto Table.

13 8of13=1 2$ 52of13=4 104of13=8

1 13
6
4 3 4
\$ \1 $
\ 52-
\104
Totals 8 32 104 1 2 4($ $)
= 2.

Had the scribe alternatively commenced with 1, then 3, 6, 12, ..., he

would have found the more cumbersome equality
80 Chapter Seven

12 26 52 78 = 2 = 13,

which the reader may care to deduce in detail for himself No doubt
he tried both, and preferred the first equality, while the equality,
7 91=2=13,
did not come to his attention; we can state this with some certainty,
because he could not take 7 of 13 unless he did the separate division of
6 = 7, and this results in a quite extensive calculation, which he would
not attempt.
 8 PROBLEMS IN COMPLETION AND
THE RED AUXILIARIES

USE OF THE RED AUXILIARIES OR REFERENCE NUMBERS

Problems 21, 22, and 23 of the RMP are called problems in completion,
since the scribe writes "complete 15 to 1," when he means "sub-
tract 9 15 from 1," which could also be expressed as, "What must
be added to 115 to make I ?" In these three problems, he shows
how to use a reference number,* which is the Egyptian counterpart of our
least common denominator, for handling various fractions. But the
Egyptians did not worry about the reference number being the least
which could be chosen. They usually chose the highest number of the
fractions before them, but not always. And in the calculations that
followed, they wrote what would correspond to our numerators in
red instead of black. For this reason, these numbers are sometimes
referred to as red auxiliaries.
The use of the red auxiliaries was so common in Egyptian com-
putation that no scribe's palette was made without at least one extra
depression for red pigment (Figure 8.1, bottom). Indeed, the two-
depression palette came to be regarded as a symbol of the scribe's
office, and the hieroglyph for "scribe" was a stylized palette and
writing stick (Figures 8.1, top and 8.2).
In the workings that follow, the red auxiliaries are set in boldface
type. I have added solutions to these three problems, using unit-
fraction equalities from tables that the scribe might have used if he
had not had another specific pedagogic purpose in view.
PROBLEM 21
Complete 3 15 to 1 [Answer: 5 15].
Take 15 as a reference number, and use the red auxiliaries: I of 15 is
10, 15 of 15 is 1, and 1 part of 15 is 15. Then,
3 15 + (some other fractions) = 1
10 1 ( 4 )=15.
* Called Hilfszahlen by Vogel. See his Vorgriechische Mathematik, Vol.
1, Vorgeschichte and Agypten, Schroedel, Hanover, 1958, p. 40.
82 Chapter Eight

FIGURE 8.1
Two scribes' wooden palettes. Left, palette inscribed with the name of
King Tuthmosis IV (1425-1417 s.c.) of the Eighteenth Dynasty. The
inscriptions along the ,ides are funerary invocations for the high official
Meryre, followed by the scribe's name (T unen). Length 13 inches. Right,
two-depression palette with a number of writing instruments. The inscrip-
tion is the name of King Antosis (1570-1546) of the Eighteenth Dynasty.
Length 11'/, inches. Courtesy British Museum.
 Problems in Completion and the Red Auxiliaries 83

FtouRE 8.2
A wooden panel from the tomb of Hesy-Ra, showing a scribe with the
insignia of his office over his shoulder and writing materials in his hand.
Courtesy Cairo Museum.
84 Chapter Eight

We have to find what fraction or fractions of 15 will give 4. Divide 4

by 15.
1 15
10 1 2
5 \3
\15 \1
Totals 5 15 4.

Therefore 3 15 is needed to complete 3 15 to 1. (Note the doubling

of 10 to obtain 5.)

PROOF
3 5 15 15 makes 1
10 3 1 1 makes 15.

Alternatively, the scribe might have said,

1= 5
_ 6 8 gen * (1, 1)
_ 9 (10 15) gen. (2, 3)
I 15 (fi 10)

therefore,

1=3 15 (fi 10),

or fi 10 is needed to complete 3 15 to 1.

PROBLEM 22
Complete 9 30 to 1 [Answer: 3 10].

Take 30 as a reference number, and use the red auxiliaries : 9 of 30 is

20, 30 of 30 is 1, and 1 part of 30 is 30. Then,

3 30 + (some other fractions) = 1

20 1 ( 9 ) = 30.

* See p. 104 for the definition and use of the generator.

 Problems in Completion and the Red Auxiliaries 85

We have to find what fraction or fractions of 30 will give 9. Divide 9

by 30.

1 30
\10 3
5 \6
Totals 5 TO 9.

Therefore 5 10 is needed to complete 3 30 to 1. (Again 10 is doubled

to obtain 5.)
PROOF
1 5 10 30 makes 1
20 6 3 1 makes 30.

Alternatively, the scribe might have noted from Problem 21 that

3 30 needs 30 more than 15 to make 1. Then fi To- To- must be
needed. But

fi 10 30 = (fi 30) 10
= 5 10. gen. (1, 5)

Or, he may have noted from the Recto, 2 = 91,'1 5 10 30 = 1.

PROBLEM 23
Complete 4 $ 10 30 45 to I [Answer: 9 30].

Take 45 as a reference number and use red auxiliaries. Then,

4 $ 10 30 45 + (some other fractions) =
(11 4 5 2 8 4 2 1 2 1) + (some other fractions) = 30
23248 68 =30.
We have to find what fractions or fractions of 45 will give 6 $. Divide
6 A by 45.*

* The scribe does not show this division in Problem 23. In Problem 21 the
scribe forgot to change to red ink for the auxiliaries, and in Problem 22
he used no red ink at all. This is most unusual. RMP 79 is the only other
instance where the statement of the problem is not in red.
86 Chapter Eight

1 45
$ 15
5
10 4
30 2 4
\40 \1 $
Totals 9 40 6 $.
Therefore 9 40 is needed to complete 4 $ 10 30 45 to
PROOF
4 $ 9 40 30 40 45 and 3 make I
11 4 528 5 42 12 18 1 and 15 make 45.
Alternatively,

_ (4 12) ($ 24) (9 18), gen. (1, 3) and (1, 2)

=(4$9) (1224) 18,
_ (4 $ 9) $ (30 45), gen. (1, 2) and (2, 3)
_ (4 $ 9) (10 40) (30 45), gen. (1, 4)
=(4$ 103045) (940),
so (9 40) is needed.

AN INTERESTING OSTRACON
The generally accepted meaning of the word "ostracon" is an in-
scribed fragment of pottery with Egyptian, Coptic, or Greek inscrip-
tions. Ostraca are most commonly found in Upper Egypt, and date
roughly from 600 B.C. to A.D. 400. The name comes from the Greek
oorpaicov, potsherd. Broken pieces of pottery had many uses in the
ancient world, one of which was as a substitute for expensive papyrus
if only brief notes needed to be made and if the clay had not previously
been decorated or inscribed.
In the tomb of Sen-mut, an architect for Queen Hatshepsut (1520-
1480 B.c.), several ostraca were found. Sen-mut designed the temple
at Deir el-Bahri, thought by many to be the finest in ancient Egypt,*
* E. K. Milliken, Cradles ofWesternCivilisation, Harrap, London, 1955, p. 92.
 Problems in Completion and the Red Auxiliaries 87

for the Queen. Sen-mut's tomb is at Thebes, where the New York
Metropolitan Museum of Art conducted an expedition, obtaining an
arithmetic computation on an ostracon that was subsequently trans-
lated.* This computation consists of three double lines in red and
black, showing the answer to the divisions of 2 and 4 by 7, expressed
in unit fractions. Now 2 divided by 7 is one of the fifty divisions of the
RMP Recto, where the standard answer is recorded as
2-7=4 28.

The interesting thing about this ostracon is that the decomposition

calculated on it is the unexpected 3-term one
2=7=fi 14 11,

and the author's technique in deriving it throws some light upon the
Egyptian method of using a reference number together with red
auxiliaries when adding fractions. The computation, with boldface
type indicating the red numbers, is
1.1 1 7
1.2 3
1.3 2 fi 14 11
1.4 32 12 1
1.5 4 2 14
1.6 101 1 2.
Whoever inscribed the ostracon was doing just what the scribe of the
RMP did in problems 28, 32, 36 and several others. The red 3 beneath
the 7 means "Take 3 as a multiplier of 7, to give the reference number
21." He then multiplied the 2 (of line 3) by his multiplier to give 6
which he then partitioned as 31/2, 11/2i and 1, each of which divides
the reference number 21 in integers, and wrote them in red (line 4).
These are the red auxiliaries.
The scribe of the ostracon then referred these auxiliaries to 21,
finding that 31/2 is 1/e of 21, 11/2 is 1/14 of 21, and 1 is 1/21 of 21, so that

* See W. C. Hayes, "Ostracon No. 153," Ostraca and Name Stones from the
Tomb of Sen-mut at Thebes, Publication 15, Metropolitan Museum of Art,
New York, 1942.
88 Chapter Eight

he wrote 6, 14, 21 in black in their proper places (line 3). In this terse
manner the scribe obtained his answer to the division,
2-7=fi 14 T.

Still with the same red multiplier 3 and the same reference number 21,
we note the 4 (of line 5) was multiplied by 3 giving 12 which was
partitioned as 101/2 and 11/2, each of which divides the reference
number 21 in integers 2 and 14, which he wrote as 2 and 14 in their
proper places (line 5) in black. It was thus he obtained his answer to
the division,
4-7=2 14.

We have no way of telling how the scribe came to choose 3 as his

multiplier* and, consequently, 21 as his reference number. Nor do
we know how he decided upon his particular partitions of 6 and 12.
This is part of his art, learned no doubt through past experience and
constant practice. The method of the ostracon is a powerful one and,
needing to write down only one line for each, the scribe could have
extended the calculation to obtain,
3-7=4 7 28
5=7=2 7 T4-
6 =7= 7 21,

which the reader is invited to verify as the scribe would have done it.

* See O. Neugebauer, The Exact Sciences in Antiquity, Harper Torchbooks,

Harper, New York, 1962, pp. 92-94, for another discussion of Ostracon 153.
 9 THE EGYPTIAN MATHEMATICAL
LEATHER ROLL

The Egyptian Mathematical Leather Roll (EMLR) was purchased

together with the RMP by the Scotsman A. H. Rhind (1833-1863)
at Luxor, Egypt, in 1858. This young lawyer came to Egypt's mild
climate for reasons of health; he became interested in Egyptology, and
went to Thebes in 1855, where he specialized in tombs. The EMLR
and the RMP were discovered in some ruins of the Rameseum at
Thebes, and later were acquired by Rhind. The papyri came to the
British Museum in 1864, and have remained there ever since. The
EMLR is roughly 10 inches by 17 inches; because of its very brittle
condition, it remained unrolled for more than 60 years.
Dr. Alexander Scott and H. R. Hall finally succeeded in unrolling
it in 1927,* and it was found to contain a collection, in duplicate, of
26 sums done in unit fractions. The lucky circumstance of duplication
helped to make possible the restoration of the right-hand column,
parts of which had been damaged. Various opinions as to its signifi-
cance have been advanced. S. R. K. Glanvillet regarded it as a
"handy table for popular use," probably, he thought, the work of a
junior official, "not of a schoolboy for the writing is far too good."
That it might have been an answer to an examination paper seemed
to him not to be the case, because of the duplicate copy. However, he
felt sure it must have been copied from a textbook as a practical guide
or table for future work. He goes on,
Its real mathematical interest lies in discovering what would have
been the use of such a table to a person armed with it, and further,
what was its relation, if any, to the Rhind Papyrus with which it was
discovered? Such an inquiry must follow a more detailed discussion
of the text itself.
From the scientific point of view, it can hardly be denied that the
dissemination of the knowledge of the chemical treatment of the
* A. Scott and H. R. Hall, "Egyptian Leather Roll of the 17th Century
B.C.," British Museum Quarterly, Vol. 2 (1927), pp. 56f.
t S. R. K. Glanville, "The Mathematical Leather Roll in the British
Museum," Journal of Egyptian Archaeology, Vol. 23 (1927), pp. 237-239.
90 Chapter Nine

leather, is of greater value than the publication of the contents

inscribed on it.*

In like vein, Scott and Hall express the opinion that

The roll has not justified the hope that it might prove to contain
material of importance. It is simply a series of sums in additions of
fractions, repeated twice over, apparently a scholar's exercise.t
A much more optimistic opinion as to the value of the contents of the
EMLR was held by Vogel,+ who wrote
Ich mochte den Inhalt als hochst bedeutsam ansehen, trotzdem er
lediglich in 26 Stammbruchsummen besteht. (I consider the content
as most important, although it consists only of 26 unit fractions.)
And Neugebauer similarly regards the EMLR as of much greater
significance; he wrote§
Ich mochte, im folgenden zu zeigen versuchen, dass ein so pessimis-
tisches Urteil vielleicht doch nicht ganz am Platze ist. (I will try to
show in what follows, that such a pessimistic judgment would perhaps
not be appropriate.)
When after sixty years the contents of the EMLR became known, the
disappointment of Glanville and Scott and Hall is quite understand-
able. They, naturally enough, expected that any hieratic writing
important enough to warrant inscription on costly leather, instead of
the much more common and cheaper papyrus, would excite Egyp-
tologists far beyond what previous historical disclosures had done. It
might have spoken, for example, of the construction of the pyramids,
of the temple of Karnak, of Abu Simbel; it might have contained the
words of some famous pharaoh, perhaps Ramses, Akhenaton, or
* Ibid.
f Scott and Hall, "Egyptian Leather Roll."
$ K. Vogel, " Erweitert die Lederolle unserer Kenntniss agyptischer Mathe-
matik?" Archiu f r Geschichte der Mathematik, Vol. 2 (1929), pp. 386-407.
§ 0. Neugebauer, "Zur agyptischen Bruchrechnung," Zeitschrift fur
Agyptische Sprache, Leipzig, Vol. 64 (1929), pp. 44f. See also B. L. Van der
Waerden, "Die Entstehungsgeschichte der agyptischen Bruchrechnung,"
Quellen and Studien zur Geschichk der Mathematik, Part B, Study IV, Berlin,
1937-1938, pp. 359-382.
 Egyptian Mathematical Leather Roll 91

Cheops; or recorded something of the diplomatic situation with the

Babylonians or Akkadians, comparable perhaps with the Tell el
`Amarna tablets; indeed it might have shed light on any one of the
many questions with which Egyptologists had been occupied for over
a century. But it did none of these things.
A reconstruction and translation of the EMLR is given in Figure
9.1; a photograph of the roll itself is given in Figure 9.2.
However the table came to be written, let us imagine with Glanville
that it was the work of a junior official, and that a chiefscribe, perhaps
the teacher, was required to show it to the head of the school for his
approval. A kind of supervisor's inspection of work, so to speak. We
can imagine the examiner's comments.
... His writing has improved, though the tails on his hundreds are
rather long. And he has not arranged the equalities at all systemati-
cally. Observe some very simple ones mixed up with much harder
ones. Look at number 8 for instance! I'm glad to see number 12 there,
one of the most useful of all. And I detect a major error in line 17.
Surely that should read 26 39 78 = 13. It stares one in the face, from
2 9 fi = 1. However, only one mistake is not bad. Of course the last
eight lines are pretty repetitious, and should have included numbers
11 and 13 among them. On the whole, a good piece of work, and a
very useful table for the addition of fractions....
These comments are of course the comments we ourselves would make,
but more importantly from the modern point of view the EMLR table
throws great light upon the mechanical arithmetic of the RMP, the
MMP,* the KP,t RP,+ BP,§ AMP, 11 MichP.,# and many of the later
Greek and Byzantine tables of fractions, as well as offering justification
* Moscow Mathematical Papyrus (originally, Golenischev Papyrus).
Moscow Museum of Fine Arts. Inventory No. 4576.
t Kahun Papyrus. British Museum. Found by W. M. F. Petrie (grandson
of the famous Australian explorer, Matthew Flinders) at Kahun, Egypt,
in 1889. The papyrus contains six mathematical fragments, not all ofwhich
have been penetrated. See p. 176.
+ Reisner Papyri. Museum of Fine Arts, Boston. Acquisition No. 38.2062.
§ Berlin Papyrus. Staatliche Museen, Berlin. No. 6619.
u Akhmim Papyrus. Cairo Museum. No. 10,758 in the Catalogue of Papyri.
# Michigan Papyri. University of Michigan, Ann Arbor. Vol. III of the
University of Michigan Collection.
92 Chapter Nine

fill nA ui't A ft a JUL A

/1 fat x
fit

f k w_ JUL 1 fit 1 Mx
fn /A iL f= A
fn A -Ait f*I ff1r
fit (I ru _ =. fill 7 Pa 4 fm >
4 /lL tv . /b Afih1t fI Y
fµ =pMA-t' fm
ft A

/W =A A
J zn
1$

fitiii -.

FIGURE 9.1
Left, reconstruction of the EMLR; right, Glanville's translation. Courtesy
S. R. K. Glanville.
 Egyptian Mathematical Leather Roll 93

1 10 40 s
2 s 20 4
3 4 12 3
4 10 to S
S 6 6 3
6 6 6 6 2
r 3 3
8 25 is 7s be s
9 60 30 is0 400 t6
10 as so lie 1s
It 9 is 6
u 7 14 2s 4
13 it 24
14 K 21 42 7
1s 1s 2T 54 9
16 22. 33 66 11

17 2S 49 196 a
is so 45 90 15
19 24 46 16
to is 36 12
21 21 42 14
22 45 90 30
23 30 60 20
24 i5 30 10
25 4$ 96 32
26 96 192 64
 Egyptian Mathematical Leather Roll 95

for presuming the existence of standard fraction tables, the G rule, and
the two thirds table.
We first of all rewrite the table of the EMLR to restore the order of
difficulty of the equalities, and to systematize the groups into which
they most naturally fall.

THE FIRST GROUP

1.7
1.5 6 6=3
1.4 10 10 = 6
1.6 6 6 = 2.
In this first group, the four equalities are so simple* that one might
well ask why the scribe wrote them, for the mere understanding of
what is meant by a fraction makes their existence self-evident. But
they are fundamental, and indeed, just a cursory glance at lines 5 and
6 gives us at once the important relation,
A 3 6=2,
from which so many other equalities are derived. Then, considering
lines 7 and 5 together, giving 3 6 6 we find using equation A that
B 2 6=3,
an equality of great importance to the scribes.

THE SECOND GROUP

The members of the second group offer themselves at once by their
similarity, and we note that in each, the second term is double the
first:
1.11 $ 18 = 6
1.13 12 24= 6
1.24 15 30 = 10
1.20 18 36 = 12
1.21 21 42 = 14
1. 19 24 48 = 16
* Two fractions that are "alike" are never given by the scribes as data or
answers to problems. They are met only in tables and, even then, very
rarely.
96 Chapter Nine

1.23 30 60 = 20
1.22 45 90 = 30
1.25 48 96=32
1.26 96 192 = 64.

Now, even though these ten equalities are scrambled (so to speak) in
the EMLR, the scribe could not have failed to notice that the answer
is in every case one-third of the second term. Here is the beginning of
the G rule of Chapter 5. Of more importance to him, however, was
the recognition of the equality A

3 6=2.
We can note here a further extension of the second group, building
on lines 3 and 2 thus :

1.3 4 12=3
1.2 3 20 = 4
G 30=5
7 42 = fi
8 56=7

This extended table could have been discovered by the scribes, using
nothing more than the natural number series for the two outside
columns and the continuous addition of the even numbers 8, 10, 12,
... for the center column.
We shall come back to this table in Chapter 10. For the present,
I remark that we have here the faint beginnings of a theory of numbers,
which as far as we know remained unwritten for 2,000 years.

THE THIRD GROUP

The third group consists of

1.3 4 12=3
1.2 5 30=4
1.1 10 40=$.
 Egyptian Mathematical Leather Roll 97

A little deeper thought was required to establish these. But the scribe
had more than one method.* First, he can put for line 3
6 fi =9 (from line 5)
6 (T2- 12-) = 9 (line 5 x 2)
(6 12) 12 = 9

4 12 = 3. (equation A x 2)

Second, he could have used the method of the red auxiliaries described
in the preceding chapter. Thus if he wished to add 4 + 12, he could
have chosen a reference number, say 12, and then reasoned as follows :
applied to 12, 4 is 3, and 12 is 1, so that added together, 4 and l2 is
3 and 1 or 4, which applied to 12 is 3; therefore,
4 i2=9.
Third, he could have referred to the previously established equality B,
and then, dividing through by 2, he would have I 12 = 9.
Of course we cannot know whether the scribe found line 3 by one
of these or some other method. But having established it, the successive
multiplication by the odd numbers produces the sequence of equalities
12 36 = 9
20 60 = 15
28 84 = 21

Now, by multiplying through by 2 (i.e., halving the denominators),

he has
6 18=2= 9
10 30 = 2 = 15
14 42=2-21

* K. Vogel, Vorgriechische Mathematik, Part 1, Vorgeschichte and J&pten,

Schroedel, Hanover, 1958. On p. 40 Vogel suggests that from 3/,j + 1/4 = 1,
division by 3 gives 1/4 + 1/1s = 1/3. This may have been the scribe's thought
process, but he had no way of writing 3/*, 4/a, or 6/Q!
98 Chapter Nine

But these are exactly the expressions given in the RMP Recto Table
of 2 divided by the odd numbers. Then we are presented with the
possibility of deriving line 3 of the EMLR from (2 = 3) of the Recto,
or, vice versa, deriving (2 _ 3) of the Recto from line 3 of the EMLR,
which lends some credence, perhaps, to Glanville's thought that the
EMLR and the RMP were in some way related.
Line 2, and consequently line 1, can be established in much the
same way as line 3:
5 5 5 5 5= 1, (definition of
one-fifth)

20 20 20 20 20 = 4,
(20 20) (20 20) 20 = 4,
(10 10) 20 = 4, (line 4, x 2)
5 20=4,
which is line 2. Doubling each number of this produces line 1. Or, as
with the previous groups, the reference number method could be used
if the scribe chose.

THE FOURTH GROUP

There are seven 3-term equalities in the EMLR, excluding line 6, of
which the five consecutive lines 14-18 are all derivable from the
original prototype
C 2 3 fi=T,
easily established from equality A and 2 2 = 1.
1.12 7 14 28 = 4
1.14 14 21 42 = 7
1.15 18 27 54= 9
1.16 22 33 66 = 11
1.17 26 39 78 = 13
1.18 30 45 90 = 15
1. 10 25 50 150 = 15.
 Egyptian Mathematical Leather Roll 99

Lines 14-18 were inscribed in their correct sequence in the EMLR,

being derived from equation C, on multiplying the denominators by
the odd numbers 7, 9, 11, 13, and 15, respectively. It is this particular
sequence that convinces me that the error of line 17 is indeed here
properly corrected from

28 49 196 = 13,

as on the EMLR, to

26 39 78 = 13.

It is possible that the scribe deduced the three-term equalities of the

fourth group from the entries which we have recorded in the first and
second groups. We illustrate how he could have done this for line 18.
From the first group, line 4, he has 10 10 = 3, which when multiplied
through by 3 gives 30 30 = 15. Then from the second group, line 22,
he has 45 90 = 30, giving 30 45 90 = T5_ on substitution, which is
line 18 of the fourth group. In a similar manner, lines 14-17 may be
as easily derived. And, of course, he may have used the method of the
red auxiliaries.

THE NUMBER SEVEN

Line 12 of the fourth group,

7 14 28 = 4,

is perhaps the most interesting of all the 26 entries of the EMLR. To

the ancients of all nations, the number 7 seems always to have held
some ineffable fascination. We can immediately think of the 7 wonders
of the world, the 7 hills upon which Rome was built, the 7 deadly sins,
the 7 against Thebes, the 7 champions of Christendom, the 7 muses,
devils, sisters, and veils, 7 years' bad luck, and the 7th heaven. The
number 7 was constantly before the Egyptian scribe, for in his mul-
tiplication, based upon the method of continued doubling (and
halving in division), the sequence
1+2+4=7
100 Chapter Nine

appeared frequently. Furthermore, the Egyptian table of length was

4 digits (or fingers) = 1 palm,

7 palms = 1 royal cubit,

hence, 28 digits = 1 royal cubit.

At once from this the scribe has
1 digit = 28 cubit,
2 digits = 14 cubit,
4 digits = 7 cubit,
7 digits = 4 cubit,
so that, since
4+2+1=7
in digits, then

7 14 28 = 4

in cubits.

Establishing this important relationship between unit fractions might

very well have been a useful and regular teaching point in the scribal
schools, and perhaps its derivation provided a sort of academic pastime
for the scribes themselves. Each of the following five derivations illus-
trates a different method of arriving at the relationship.

1. The scribe in RMP 34 shows that

4x (1 2 4)=7;
he then writes

7of(1 2 4)=4,
whence 7 14 28 = 4.
2. Since from ordinary multiplication with integers,

1+2+4=7,
 Egyptian Mathematical Leather Roll 101

dividing throughout by 7 (using the Recto Table, where, 2 - 7 =

4 28, and hence, 4 = 7 = 2 14) gives
7 (4 28) (2 14) = 1;

rearranging, this is
(2 4) (728) = 1.
14
Noting that 2 4 4 = 1, subtract 2 4 from each side, whence
7 14 28 = 4.
3. Using a reference number (here obviously 28), the reasoning is
7 is 4,
14 is 2,
28 is 1,

then adding, 7 14 28 is 7, which applied to 28 is 4.

4. The EMLR third group slightly extended is
4 12=3
3 20=4
9 30=5
7 42 = fi.
Doubling the last equality gives
14 84= 12,
and adding these two gives
7 14 (42 84) = 6 12;

hence 7 14 28 = 4, for both 42 84 = 28 and 12 = 4 are members

of the second group, in which the second term is always double of the
first.
5. The equality can be arrived at by the multiplication of 4 by 7,
\1
\2
\ 4

\4 \1
Totals 7 1 2 4.
102 Chapter Nine

But this means that 1 2 4 = 4 x 7; so divide by 7 to get 7 14 28

=4.
We can be pretty sure that these five examples by no means exhaust
the ways in which this important equality could have been arrived at
by the scribes.

LINE 10 OF THE FOURTH GROUP

The last equality of the fourth group, line 10 of the EMLR, is an ex-
pression quite unlike any of the other 25. It would be most interesting
to know just how it was established by the scribe. It is quite easily
proved to be true by using the red auxiliaries. Thus, applied to the
reference number 150,
25 is 6,
50 is 3,
150 is 1;
then adding, 25 50 150 is 10, which is 15 of 150, so that it is proved
that 25 50 150 = 15. It would not be easy to imagine the circum-
stances in which this equality would arise. None of its terms is prime,
and the common factor 5 reduces it to
3 10 30=$,
which the scribe could have established in several different ways, e.g.,
2 fi = I (equation B)
1 $=1 (adding unity)
3 TO 30 = 3 ; (dividing by 5)
Or perhaps he said : since 6 3 1 = 10 in integers, divide by 30 to
obtain 3 10 30 = 3.

THE FIFTH GROUP

1.8 25 15* 75 200 = $
1. 9 50 30# 150 400 = 16.

* This is one of the rare occasions where a set of fractions belonging to the
same group is not exactly arranged in descending order of magnitude: 15
should precede 25, and 30 should precede 50.
 Egyptian Mathematical Leather Roll 103

Using the red auxiliaries, and (as is most usual) taking 200, the
greatest denominator for reference number from the left-hand side of
the equality of line 8, we would have
25 is 8,
15 is 13 9,
75 is 2 3,
200 is 1;
adding, 25 15 T5- 200 is 25, which is 8 of 200. Therefore line 8 follows.
Simple doubling gives line 9.
The above, however, is only a proof, by a standard scribal technique,
that the equality is true. It does not show how the four terms were
chosen in the first place, nor how the scribe knew their sum to be A.
Perhaps we shall never find out, but we can still hazard a plausible
guess as to how he arrived at it. One way, perhaps a most unlikely one,
would be for the scribe to have noted that
24 40 8 3=75,
in integers; then, on dividing through by 600, he would have had line
8. But we can establish it more plausibly by selecting fi 30 = 5 from
the extended second group (p. 96). Multiply this through by 5, so
that
25 = 30 150,

then add 15 75 to both sides and regroup, giving

25 T5- T5- = (15 30) (75 150) (from
= 10 50, line 24)
then add 200 to obtain
25 15 75 200 = 10 (50 200) (from line
= 10 40 2 x 10)
_ $. (fromline 1)
We can conclude that Vogel, Neugebauer, and Van der Waerden
were indeed farseeing when they considered that the EMLR would
add considerably to our knowledge and understanding of the Egyptian
techniques for handling unit fractions despite Glanville's and Scott
and Hall's disenchanted reflections when it was first unrolled.
* Here we see a possible explanation of why 25 preceded 15.
 10 UNIT-FRACTION TABLES

In this chapter, most numbers are denominators of unit fractions.

Hence the overbars will be omitted unless there is some chance of
confusion. The exception to the unit-fraction rule is of course a/a,
which is written 3. In order to identify groups of unit fractions such as
1 2 3 5
2 4 or 6 10
3 6 9 15,
where each fraction is composed of terms that are integral multiples
of the terms of the first fraction, we shall refer to each group by its
initial pair, calling this pair the generator of the group; i.e., the left-
hand group shown here has the generator (1, 2) and the other, (3, 5).
Note that, as elsewhere in this book, no plus sign is included in the
unit fractions, mere juxtaposition implying addition, just as, say, 41/z
means 4 + 1/2 in modern notation. However, of necessity we use the
modern sign = for "equals," although the Egyptians had no such
sign. In the EMLR there is on each line a symbol j, which means
" this is." This notation is the nearest counterpart to a modern equals
sign in Egyptian arithmetic.
The scribes' preoccupation with operations involving unit fractions
rather than integers is quite marked in Egyptian mathematical papyri.
Of the 87 problems in the RMP only a mere six do not involve frac-
tions, while the Recto of the RMP, which constitutes nearly one-third
of the 18-foot-long papyrus, deals entirely with odd fractional divisions
of 2, as we have seen in Chapters 6 and 7. Because the Egyptians per-
formed their multiplications and divisions by doubling and halving,
it was necessary to be able to double fractions as well as integers. For
fractions with even denominators this was easy; but with odd numbers
as denominators this doubling could sometimes be quite difficult, so
that a previously prepared table would have been a great aid to
computation. Indeed, a portion of the RMP Recto Table is included
in the KP. Other papyri, such as the MMP, also show this obsession
with unit fractions. Of the 25 problems of the MMP only three do not
involve calculations with fractions.
 Unit-Fraction Tables 105

The Egyptians' concern for the accurate dealing with fractions

almost certainly originated from practical problems, such as the
division of food, supplies, and other things, either equally or in some
specific ratio, among families, troops, members of working gangs,
large crews, etc., in a country which had no metallic currency or
money, and in which payments were made in kind. Only for large
payments might weights of precious metals like silver or gold be used
(see Chapter 20). If a fractional number like our % needed to be
expressed or evaluated, as would arise for example in the division of
3 loaves among 5 men, the Egyptians had no other way of writing it
than as 9 5 F5- or some other combination of unit fractions, or by
saying "three divided by five." Today, if a new concept arises, mathe-
maticians devise at once a new notation for it, but the Egyptians, never
thinking to improve or alter their notation for fractions, developed
instead special techniques for dealing with the notation they already
had. To divide 3 loaves equally among 5 men, each man would be
given three separate portions, a 1/3, a 1/, and a 1/15.One advantage of
this division was that not only was justice done, but justice also appear-
ed to have been done. In a modern distribution, three of the five men
would get % of a loaf in one large piece, while the other two men
would get two smaller pieces, % and 1/5 of a loaf, which division might
be regarded as an injustice by an ignorant workman.*
The scribe would never write 3 5 3 for % except in tables used
for calculating. No two similar fractions ever occur in the scribe's
answer to a problem or a calculation, even though they may have
occurred several times repeated in the mechanical operations that the
problem required. Just exactly what was the reason at the back of the
Egyptian mind for this is still not clear. It is, however, certain that
the concepts of" one-half," "one-third," or "one-quarter," and so on,
must have come before the invention of the hieroglyphs that denote
* Problem 6 of the RMP requires the division of 9 loaves among 10 men.
While the modern answer is that each man gets a/10 of a loaf, this division
requires that the last man must get (1/10 1/10 1/1o 1/ 0 1/ 0 1/10 1/ 0 1/10 1//o)
of a loaf; already sliced so to speak! The Egyptians would have none of such
a solution. The answer given in the RMP is that every man gets exactly the
same number of pieces and exactly the same-sized pieces, namely, 2/3i 1/s,
and 1/30 each, and justice is again obviously done.
106 Chapter Ten

them. Thus also the concepts preceded the conventional notation of

writing the fractions with their numbers. And so likewise would the
concepts of " the two parts" (in three) and "the three parts" (in four)
have come quite early, just as today we often speak of something being
" three parts full," when we mean three-quarters full. In modern
German, there is a word anderthalb, meaning literally another or
second half, so that anderthalb pfund means 11/Z pounds.
If one-fifth, let us say, was cut off from one end of a loaf of bread,
the Egyptian did not seem to think of what was left as being'/5 of the
original loaf, but rather as the loaf reduced; if a fifth of this reduced
loaf was cut off, the two "fifths" thus removed would of course be
unequal. Perhaps this circumstance may help us to understand why
the scribes so carefully avoided writing 5 5 for %/5 and wrote instead
15. The ancient Egyptians might even have claimed with some
justification that their method of division was superior!
Divisions of loaves and other commodities could often be further
complicated by the provisos that overseers were to get three times as
much as the boatmen and doorkeepers, who in turn were to get double
portions. How will the supervisor cut up the loaves ifeach man expects
to get the same number of pieces as other workers of his class? There
were many such problems of partitioning, some far more complicated
than this, and thus inevitably, when the system of writing numbers and
fractions had been evolved, one of the first requirements was to prepare
useful tables of unit-fractional equivalents for computation. The
EMLR is one such table, though perhaps a rather sophisticated one.
In the RMP there are, besides the Recto Table, five other shorter
tables of equivalent fractions. Other similar tables are to be found in
other papyri, on ostraca, on leather, even on wood.

UNIT-FRACTION TABLES OF THE RHIND MATHEMATICAL PAPYRUS

The Recto of the RMP, as we have already seen, contains decomposi-
tions of the division of 2 by the odd numbers 3 to 101 (see Chapter 7).
In each division, the scribe first gave his answer, and then proved it
correct by multiplication of the fractions involved in each answer,
which necessitated the addition of several fractions to total 2. In only
one of the 50 divisions does the scribe show how he "knows" that the
 Unit Fraction Tables 107

fractions concerned do in fact total 2 (see pp. 77-79), but they all
do, and nowhere has he made an error. A close examination of the
Recto leads one to the conclusion that the scribe must have referred
to some set of tables giving the sums of 2-term and 3-term unit frac-
tions. Either that, or he worked each one out separately on a papyritic
memo pad, or he was superbly competent at mental arithmetic. The
first conclusion is the most likely, for it is already well established that
the Egyptians made constant use of various types of tables. The set of
tables that the scribe who was calculating the divisions of the Recto
had for reference must have contained the equalities given in Tables
10.1 and 10.2. Whether the simple relations 2 2 = I and 4 4 = 2
were included or not is a matter of conjecture. They occur frequently
in the problems (17 times in the Recto and more than 40 times in the
Verso) ; they might very well have been considered to be too simple
to bother recording in a standard table. The same would apply to the
relation 3 3 = 1, which appears in at least 14 of the Verso problems.
In this list, each equality is easily located either in the Recto or the
Verso of the RMP as shown, but I am sure that Tables 10.1 and 10.2
are by no means complete. That these tables are not as complete as
the scribes' may have been can be illustrated by the last entry for the
gen. (4, 5, 20). It is possible that the scribe first noted that (10 40) = 8
from gen. (1, 4) and then put (8 8) = 4 from gen. (1, 1), but we are
unable to decide precisely which equality was used. Not only were all
the equalities in these tables used by the scribe as a commonplace
procedure not requiring any proof or justification : he sometimes went
much further; for example, in Problem 65, he wrote without
comment 100 = 13 = 7 3 39, as if he did the division mentally,
which may well have been the case. But to say the same of the two
following equalities, particularly the last, would be to strain our
credulity too far!
2 5 10 10 10 = 1, (Problem 35)
2 6 12 14 21 21 42 63 84 126 126 168 252
336 504 1008 = 1. (Problem 70)
If the case for the use of tables has not been completely established up
to this point, then surely this last equality would clinch it. The reader
108 Chapter Ten

TABLE 10.1
Two-term equalities appearing in the RMP.
Generator Equality Occurrence
(1, 1) 6 6=3 Recto, 2 i 11, 35, 41, 53, 55.
Verso, Problem 40.
8 8=4 Recto, 2 i 13, 67, 71, 97.
10 10 = 5 Recto, 2 i 89.
14 14 = 7 Verso, Problem 69.
(1,2) 3 6=2 Recto, 2 i 17, 37, 47, 73, 79, 83, 89, 95.
Verso, Problem 69.
6 12 = 4 Recto, 2 i 19, 23.
21 42 = 14 Verso, Problem 69.
81 162 = 54 Verso, Problem 42.
(1,3) 4 12 = 3 Recto, 2 i 95.
8 24 = 6 Recto, 2 i 29, 37, 41.
(1, 4) 5 20 = 4 Recto, 2 + 31, 67, 73,83, 89.
10 40 = 8 Recto, 2 i 71.
(1, 6) 7 42 = 6 Recto, 2 i 43. Verso, Problem 69.
(1, 7) 8 56 = 7 Verso, Problems 7, 7B, 11.
16 112 = 14 Verso, Problems 7, 7B, 11.
(2, 3) 10 15 = 6 Recto, 2 i 47, 53, 79.
20 30- 12 Verso, Problem 76.

TABLE 10.2
Three-term equalities appearing in the RMP.
Generator Equality Occurrence
(1, 1, 1) 6 6 6= 2 Recto, 2 i 29.
(1, 2, 4) 7 14 28 = 4 Recto, 2 i 97. Verso, Problems 38, 69.
14 28 56 = 8 Verso, Problem 24.
(1, 2, 6) 5 10 30 = 3 Recto, 2 -= 91. Verso, Problems 1,3,6,30.
(1, 3, 3) 5 15 15 = 3 Verso, Problems 2, 30.
(1, 3, 5) 23 69 115 = 15 Verso, Problem 30.
46 138 230 = 30 Verso, Problem 30.
(2,3, 6) 2 3 6= I Recto, 2 i 43,101. Verso, Problems 66, 69.
6 9 18 = 3 Verso, Problems 17, 42, 67.
18 27 54 = 9 Verso, Problem 42.
36 54 108 = 18 Verso, Problem 42.
(3,4,6) 9 12 18 = 4 Recto, 2 _ 59.
(3, 10, 15) 3 10 15 = 2 Verso, Problems 4, 5, 35, 56.
(4, 5, 20) 8 10 40 = 4 Recto, 2 i 61.
 Unit-Fraction Tables 109

is invited to check this equality by modern arithmetic, or by using the

scribes' methods.

PROBLEMS 7 TO 20 OF THE RHIND MATHEMATICAL PAPYRUS

These 15 problems (including Problem 7B) form in reality a table of
3- and 6-term unit-fraction equalities, similar to the Recto Table and
the tables of the EMLR. The two numbers 1 2 4 and 1 3 3 are used
as multipliers for a succession of multiplicands, so chosen as to result
in certain equalities likely to be of use in the later problems of the
papyrus. Chace regards them as examples of simple multiplications
of fractional expressions, and Neugebauer as completion problems for
2 - 7 and 2 = 9 of the Recto. The following multiplications were
carried out by the scribe:
1 2 4 times 7 (Pr. 11)
14 (Pr. 12)
28 (Pr. 14)
14 (Pr. 9)
4 T8 (Prs. 7, 7B, 10)
16 112 (Pr. 13)
32 224, (Pr. 15)
1 times Q (Pr. 16)
IV
(Pr. 17)
4 (Pr. 8)
(Pr. 18)
12 (Pr. 19)
1f 24. (Pr. 20)

As a result of these 15 multiplications, the following equalities are

established in this group of problems:

7 14 28 = 4 (Prs. 11, 10)

14 28 56 = 8 (Prs. 12, 9)
28 56 112 = 16 (Pr. 14)
112 224 448 = 64 (Pr. 13)
224 448 896 = 128, (Pr. 15)
110 Chapter Ten

2 9 6=I (Pr. 16)

4 6 12 = 2 (Pr. 8)
6 9 18 = 3 (Prs. 18, 17)
12 18 36 = 6 (Pr. 19)
24 36 72 = 12. (Pr. 20)
h en, from
1=2 2
=2 4 4
=2 4 8 8,
and since 14 28 56 = 8 from the foregoing list,
2 4 $ 14 28 56 = 1 (Pr. 9)
x2 4 8 16 28 56 112 = 2 (Prs.10, 7, 7B)
x4 16 32 64 112 224 448 = 8 (Pr. 13)
x2 32 64 128 224 448 896 = 16. (Pr. 15)

It is my view that this group of problems either was included in the

RMP to establish a set of 3-term and 6-term equalities for inclusion
in Egyptian standard tables, or was taken from a set of such tables.
Working only with the same methods, techniques, and notations
available to an Egyptian scribe, we now attempt to reproduce some
of these unit-fraction tables ab initio. We must of course eschew any
modern refinements that could lead us to obvious simplifications. It
may be a little irksome, but we have to try to think how the scribe
would have thought, to imagine we are writing in hieratic, and to be
logical only to the extent that we could expect the scribe to have been
logical. And so our very first table of fractions will come from the
original meaning of the word "fraction," a part, where two halves,
three thirds, four quarters, and so on, make one whole. Then we have
our first and most elementary table (Table 10.3). Each of the unit-
fraction equalities of Table 10.3 will produce other equalities by the
use of generators, i.e., by the successive multiplication by integers.
Table 10.4a thus appears by so treating the first equality of Table 10.3.
To Table 10.4a could be added its only odd-number member,3 3 = 3,
and that only because of the uniqueness of 3. Similarly from the other
equalities of the basic Table 10.3, Tables 10.4b-d arise.
 Unit-Fraction Tables 111

TABLE 10.3
The most elementary table, from the definition of a fraction. Overbars not
shown.
2 2=1
3 3 3= 1
4 4 4 4= 1

TABLE 10.4
The table produced b applying generators to the equalities of Table 10.3.
The equality 3 3 = f can be considered to be included in a. Overbars
omitted from this table.
Generator Generator Generators Generators
(1, 1) (1, 1, 1) (1, 1, 1, 1) (1, 1, 1, 1, 1)
2 2=1 3 3 3=1 4 4 4 4= 1 5 5 5 5 5= 1
4 4=2 6 66=2 8 8 8 8=2 10 10 10 10 10=2
6 6=3 9 99=3 12 12 12 12=3 15 15 15 15 15 = 3

(a) (b) (c) (d)

If Table 10.4 appears to be elementary, it is worth noting that the

scribe of the EMLR thought some of the entries worthy of being
recorded in that work:

line 4 of the EMLR, 10 10 = 5;

line 5 of the EMLR, 6 6 = 3;
line 6 of the EMLR, 6 6 6 = 2;
line 7 of the EMLR, 3 3 = 3.
This suggests that lists similar to those in Table 10.4 were at some
time constructed by the scribes, and were in fact used by them, even
though no such tables have come down to us in toto. Their most prob-
able use would have been in the scribal schools.
We now endeavor to establish further equalities derivable by the
scribes from the Tables 10.3 and 10.4.
We begin with Table 10.4a. Adding 6 to each side of line 3 gives

3 6=6 6 6.
112 Chapter Ten

The right-hand side of this equals 2 by the generator (1, 1, 1) (line

of Table 10.4b), or

3 6=2.
This equality could have been derived otherwise by the scribes. For
example, from generator (1, 1) (Table 10.4a) they could have written
3 3 = 3 and 6 6 = 3; then by addition,

(3 6) (3 6) = 1,

or, by applying line 1 of generator (1, 1), 3 6 = 2.

Now the equality 3 6 = 2 belongs to a new group, for since the
left-hand side can be written as (3 x 1 3 x 2), its generator is (1, 2) ;
successive multiplication of this generator by 2, 3, 4, ... , produces
the table
3 6 2
6 12 = 4
9 18=6

The next table to look for is, naturally, that whose generator is
(1, 3), and then, if successful, (1, 4), (1, 5), and so on.
Adding 12 to each side of line 2 of the just-generated table gives

4 12 = 6 12 12
=6 6 (from gen. (1, 1))
3. (from gen. (1, 1) )

Again, this equality may be otherwise derived. Thus, from generators

(1, 2) and (1, 1) we have

3 6 =2
(6 6) (12 12) = 2
6 6 12 12 = 3 6
6 12 12 = 3
4 12 = 3.
 Unit-Fraction Tables 113

Successive multiplication of this last equality produces the table of

the generator (1, 3) :

4 12=3
8 24=6
12 36 = 9

We proceed :

gen. (1, 1) 5 = 10 10
gen. (1, 1) = (20 20) (20 20)
Add 20 to each side 5 20 = (20 20 20 20 20)
gen. (1, 1, 1, 1, 1) gives 5 20 = 4.

Then the table based on generator (1, 4) follows:

5 20 = 4
10 40 = 8
15 60 = 12

Again,

gen. (1, 2) 15 30 = 10
gen. (1, 1) 15 (60 60) = 10
(15 60) 60 = 10
gen. (1, 4) 12 60 = 10
Divide both sides by 2 6 30 = 5.

We have then the table of generator (1, 5) :

6 30 = 5
12 60 = 10
18 90 = 15

The scribes no doubt also noted that, besides multiplying the pre-
ceding equalities to produce new equalities, they could also add or
114 Chapter Ten

TALE 10.5
Ordered table of generators derived solely from the simpler generators of
Table 10.4. The initial equality associated with each generator is given
opposite the generator itself
gen.(1,l) 2 2= 1
(1,2) 3 6=2
(1,3) 4 12 = 3
(1,4) 5 20 = 4
(1,5) 6 30 = 5

even subtract them to obtain new equalities, provided that those so

operated on belonged to the same generator group.
At this stage, the scribes might well have looked over what they so
far had achieved and then set down in order the equalities established,
making a table of the basic generators (Table 10.5). We may assume
that it was not difficult for the scribes to see that they had lighted upon
a simple property of sequences of fractions. I suggest that such a dis-
covery by the scribes, and the embryonic theory of numbers that it
implies, might help to explain why so many of the problems of the
mathematical papyri are freely interspersed with illustrations worked
out in fractions when integers would have sufficed. By a simple induc-
tion the scribes could have concluded that the series of natural num-
bers in the table of generators (Table 10.5) could be extended as far
as they pleased. The first and third columns of numbers consist simply
of the natural number series, and the middle column progresses in a
uniform way, for their differences in order are the even numbers 4, 6,
8,. .. . More obvious perhaps would be the observation that in each
equality, the product of the two outside numbers equals the middle
one. And further, each element of the table of generators produces
another unending table upon the successive multiplication by 2, 3, 4,
... , so that a comprehensive set of tables could be constructed, con-
taining all possible 2-term unit-fraction equivalents. The EMLR is
probably one small selection from such a set of tables.
However powerful the 2-term equalities may have been as an aid
to computation, there was no need to stop at 2, and indeed the scribes
did not, as the EMLR itself illustrates. Three-term equalities may be
 Unit-Fraction Tables 115

derived, sometimes very easily, from the 2-term equality tables already
established, and this we now examine. The first and most obvious
3-term equality to establish has the generator (2, 3, 6), because it
arises from the first of the 2-term equalities:

gen. (1, 1) 2 2=1

gen. (1, 2) 2=3 6
Substitution 2 3 6 = 1.
For the Egyptian scribe, this equality was almost a standard form;
it often occurs in calculations. In like manner the scribe could write

gen. (1, 2) 3 6=2

gen. (1, 3) 6=8 24
Substitution 3 8 24 = 2.

This is an entirely new 3-term equality. The field would begin to

widen at this stage, because there happen to be other substitutions
available for 6, which the scribe can make; thus

gen. (1, 2) 3 6=2

gen. (1, 2) 6=9 18
Substitution 3 9 18 = 2.

The three equalities deduced so far turn out to be prime equalities;

i.e., they cannot be simplified by dividing their terms through by some
common factor. But, as previously, each may be multiplied in suc-
cession by the integers to produce further endless sets of tables. Looking
at the last two of these 3-term equalities, we observe, as no doubt the
Egyptian scribes did, that since 3 8 24 = 2 and 3 9 18 = 2, then
surely

8 24 = 9 18.

Such equalities, composed of members both of which are themselves

multi-termed, may have been another source of thought for the ancient
Egyptian student of numbers. In the Recto of the RMP, where the
116 Chapter Ten

scribe divided 2 by 17, we find 3 12 replaced by the more convenient

4 6 without any explanation. And in dividing 2 by 23 he wrote 1 4
for 2 4 6, again without explanation. The reason for the substitution
would appear to be because the 2-term equivalent is shorter, and also
perhaps because it contains his favored . So equalities composed of
multi-term members may have been another facet in his vista of unit-
fraction equalities.
Let us now suppose that, in endeavoring to establish a certain 3-
term equality, it became necessary to refer to the 2-term tables for the
single-unit-fraction sum 28 70. Now this pair of fractions should be
found in the table whose generator is (2, 5). But we do not have such
a table, and in fact at this stage we do not know whether such a sum
has a single-unit-fraction equivalent. Therefore we must consider
such pairs before proceeding with the tables for 3-term equalities. We
must determine if they in fact do exist, and, if they do, find out how
to develop them in a properly ordered sequence.
Although the scribes may already have noted how the table of
generators could be extended merely by extending certain series, there
remains the question of the relations between the three individual
members of each equality. We look again at the table of generators
(Table 10.5) that has already been established. There are many
properties which an observant student of these equalities might dis-
cern, but the relevant ones are those which apply in exactly the same
way to all five of them, as well as to all the succeeding equalities. One
such property may be seen by taking the last one listed as an example :
if the elements of the generator of (1, 5) be added, then 6 results; and
if the elements be then multiplied by this 6, one obtains the left-hand
side of the equality, namely, 6 30. Clearly this simple property is
possessed by all the other generators and their corresponding equalities
of Table 10.5. The right-hand side of each equality is equal to the
second element of the corresponding generator, which in this case is
the same as the product of the two elements. Reasonably enough, the
scribe might have wondered if such simple relations would apply to
other generators when the first element is other than unity. The follow-
ing are a likely set of generators he would set down to test this con-
jecture.
 Unit-Fraction Tables 117

(2, 1) (3, 1) (4, 1) (5, 1)

(2,2) (3,2) (4,2) (5,2)
(2, 3) (3, 3) (4, 3) (5, 3)
(2, 4) (3, 4) (4, 4) (5, 4)
(2, 5) (3, 5) (4, 5) (5, 5)

In this array there are some duplications. For example, (2, 2) is the
same generator as (1, 1), (2, 4) is the same as (1, 2), and (3, 1) is the
same as (1, 3), and so on. To eliminate duplications, we may exclude
those pairs of generators having a common factor. Then the generators
to be examined are those in the amended array:

(2, 3) (3, 4) (4, 5) (5, 6)

(2, 5) (3, 5) (4, 7) (5, 7)

(2, 7) (3, 7) (4, 9) (5, 8)

(2, 9) (3, 8) (4, 11) (5, 9)

(2, 11) (3, 10) (4,13) (5, 11)

Applying the rules deduced above to the first pair of generators in

this array, the scribe would have: the sum of the elements is 5, and
multiplying the elements 2 and 3 by this 5 gives 10 15, which should
be the left-hand side of the resulting equality. For the right-hand side
he should have either the second member 3, or the product 2 x 3 = 6.
Now, according to the development that I am postulating here, the
scribe would have known at once that the sum cannot be 3, for he
would remember generator (1, 1), line 3 (Table 10.4a), i.e., in which
6 6 = 3 and both 10 and 15 are smaller fractions than fi. The scribe
then had to test the truth of the putative equality

10 15 = 6. gen. (2, 3)

If the equality is not true, he need have gone no further. If it is true,

he would still have had to try several more in order to justify a gen-
eralization. The proof follows on the next page.
118 Chapter Ten

From gen. (1, 5) 10 = 12 60

Add 15 to each side 10 15 = 12 (15 60)
From gen. (1, 4) = 12 12
From gen. (1, 1) 10 15 = 6.

A proof such as this would have established the truth of this equality
for the scribe. He would next try other equalities. For example, he
might have taken the generator (3, 4). Following the same method-
ology as before, the relevant numbers are 3 + 4 = 7 and 3 x 4 = 12,
so that the resulting equality which had to be tested is 21 28 = 12
(gen. (3, 4)) :
From gen. (1, 7) 8 56 = 7
Add 42 to each side 8 42 56 = 7 42
From gen. (1, 6) = 6
From gen. (1, 3) =8 24
Then, 8 42 56 = 8 24
Subtract 8 from each side 42 56 = 24
Divide by 2 21 28 = 12.

These two equalities having been established, the scribe would have
treated a third in the same way, from the generator (4, 5), say; since
here 4 + 5 = 9 and 4 x 5 = 20, the equality to be tested is 36 45 =
20 (gen. (4, 5)) :
From gen. (1, 9) 10 90 = 9
Add 72 to each side 10 72 90 = 9 72
From gets. (1, 8) = 8
From gen. (1, 4) = 10 40
Then 10 72 90 = 10 40
Subtract 10 from each side 72 90 = 40
Divide by 2 36 45 = 20.
There is a sameness about each of the three preceding proofs, which
suggests that, given any two numbers at all as elements of a generator,
the same approach will establish the relevant 2-term equality. How-
ever, by choosing the numbers 5 and 7 for elements, the scribe would
have found that the method does not always work, in this case because
5 and 7 are both odd numbers. So he would have had to look for an-
 Unit-Fraction Tables 119

other method, which he easily found. For example, the scribe could
have established 60 84 = 35 (gen. (5, 7)) :
From gen. (1, 6) 60 = 70 420
Add 84 to each side 60 84 = 70 (84 420)
From gen. (1, 5) = 70 70
From gen. (1, 1) 60 84 = 35.
Trying the same procedure with the generator (3, 5) gives
From gen. (1, 4) 24 = 30 120
Add 40 to each side 24 40 = 30 (40 120)
From gen. (1, 3) = 30 30
From gen. (1, 1) 24 40 = 15.
If the scribe had tried as generator (3, 7), his technique would have
had to be slightly different (he would have had to start with the larger
of the summed unit fractions). Thus, to establish 30 70 = 21 (gen.
(3, 7)) :
From gen. (1, 5) 70 = 84 420
Add 30 to each side 30 70 = 84 (30 420)
From gen. (1, 14) = 84 28
From gen. (1, 3) 30 70 = 21.
Again, the scribe interested in what we have called his "theory of
numbers" or, more properly, his "theory of fractions" would natu-
rally enough stop at about this stage and take stock, in an orderly
fashion, of what he had so far established. A collection of 2-term
equalities obtainable by means of the methods just discussed is shown
in Table All .1. The established 2-term equalities, together with
their generators, are in roman type; an overall view of such a collection
would have at once revealed to the scribe many ordered sequences,
what today we would call series. Indeed, such is the regularity of the
various columns that it would be a very unimaginative scribe indeed
who would not be tempted to extend his table upwards, as shown by
the italic entries (this is downward, so to speak, in the magnitudes of
the denominators). By so doing, the scribe would see that he has really
rewritten his table so that entries in corresponding positions above
and below the diagonal are the same; thus the table reads the same
either horizontally or vertically (see Appendix 11).
 11 PROBLEMS OF EQUITABLE DISTRIBUTION
AND ACCURATE MEASUREMENT

DIVISION OF THE NUMBERS I TO 9 BY 10

TABLE 11.1
Quotients of 1, 2, ... , 9 divided by 10, as listed in the RMP.
Number Quotient
1 10
2 5
3 To
4 3 15
5 2
6 2 10
7 3 To
8 1 10 30
9 3 5 30

Table 11.1 is a translation of the one made by the scribe in preparation

for the first six problems of the RMP. In these problems, which im-
mediately follow the original table, 1, 2, 6, 7, 8, and 9 loaves are to be
divided equally among 10 men. In each case the answers are given in
the form of those unit fractions shown in the table. In each problem
the scribe first states his answer from the table, and then by multiplica-
tion he proves that it is correct. We are here more interested in how
the table was prepared rather than in its application to the division
of loaves among workmen, and we now attempt to reconstruct the
scribe's original derivation of this table. The following calculations
show how the table could have been derived by straightforward
division.
1 10
1 10 10 I

\10 \1 5 \2
10 1. 5 2.
 Equitable Distribution and Accurate Measurement 121

1 10 1 10
\10 \1 \6
\5 \2 \30 \ 5

5 10 3. 3 30 7.

1 10 1 10
6 \6
\\15
1
\35 \10
\ \
1

\30 5

515 4. 3 1030 8.

1 10 1 10

\5 6

5.
\5 \2
\30
10
1
1 530 9.
2 \5
\10 \1
2 10 6.

But the table could have been otherwise constructed. The alternate
technique of the succeeding paragraphs may appear to be much
longer, but this is only because of the explanatory matter. Actually
it is a good deal shorter than the straightforward division just shown.
It was possible to write at once the division of 1, 2, and 5 by 10:
Number Quotient
1 10
2 5
5

From these entries, the scribe could immediately obtain 3 = 10 from

(1 + 2) = 10, so that he had his third entry 5 10. In like manner,
since 6 = 10 is (1 + 5) = 10, the sixth entry is at once 2 10. The
scribe could also have found 6 = 10 by doubling 3 = 10, but most
probably rejected it as less simple. Performing the doubling, however,
gives the unit-fraction decomposition
122 Chapter Eleven

2 x (3 = 10) _ (2 _ 5) + (2 - 10)
_ (3 15) 5,
where the 3 15 comes from the Recto Table (Table 6.1). Thus the
scribe would have obtained the useful equality 3 5 15 = 2 TO. The
fourth entry can come from 1 + 3 or from 2 x 2, but in each case the
scribe would have had 3 5, which was not acceptable, the two frac-
tions being the same (Precept 3, p. 49). But since two-fifths may be
expressed as 2 - 5, he had only to look in the Recto Table to find the
quotient 3 15. He now had
Number Quotient
1 T
2 5
3 5 10
4 3 15
5 2
6 2 10
7
8
9
The scribe had available several alternate ways of obtaining the
quotients for 7, 8, and 9. Thus, he could have found 7 - 10 by con-
sidering
(1 + 6) _ 10 = 10 (2 10) =2 5.
(2+5) = 10= 5 2 =2 5.
(3 + 4) = 10 = (5 1-0)(9 15) = 3 5 10 15.

But he would have none of these, because now he could include his
most important (and his largest) fraction 3, which was not possible
for the preceding dividends. In the same way for 8 - 10, he had
(1+7)=10=10 ? _ ?
(2+6)=10= 5 (2 10) =2 5 TO.
(3 + 5) = 10 = (3 10) 2 =2 5 TO.
(4 + 4) = 10 = (3 1-5)(9 15) = (3 3) (15 15)
_3 10 30.
* From 2 = 15 = 10 30, RMP Recto Table.
 Equitable Distribution and Accurate Measurement 123

This last value is the one the scribe chose, most probably because of
the presence of the 3.
The table was now complete except for 7 and 9.
Number Quotient
1 10
2 5
3 5 10
4 15
5 2
6 2 To
7
8 3 10 30
9

These last two now could have been easily found from 9 - 10 =
(8 + 1) - 10 (adding 10 to 3 TO- 50- to give 3 5 30), and from 7
10 = (8 - 1) _ 10 (subtracting 10 from 3 10 30 to give 3 30).

CUTTING UP OF LOAVES
The scribe now had the complete list of divisors and quotients as
shown in Table 11.1. As I have stated, no working is shown in the
RMP for the nine equalities of Table 11.1. To show their uses the
scribe chose six practical problems: the division of 1, 2, 6, 7, 8, and 9
loaves among 10 men. The scribe read off the answers from the table;
he proved that his choices were correct by performing the appropriate
multiplication by 10, as in the following translation of Problem 3:
Divide 6 loaves among 10 men.
Make thou the multiplication 2 10 times 10.
The doing as it occurs.
1 2 10
\2 \1 5
4 2 3 15 [2 _ 5 = 3 15, Recto Table]
\8 \4 3 TO- TO [2 - 15 = 10 30, Recto Table]
Total 10 6 the same, this is. [5 10 30 = 3, Table 10.2]
124 Chapter Eleven

The calculations included by the scribe for divisions of 1, 2, 7, 8, and

9 loaves are similar to this. These calculations are reproduced here:

1 10 1 5
\2 5 \2 \ 3 15
4 15 4 3 To- TO
\8 \ I 10 30. \8 \1 5 5 15.
10 1 10 2

1 1 30 1 110 30
\2 \115 \2 \110
4 25 1030 4 35
\8 \5 2 10. \8 \6 9 15.
10 7 10 8

1 9 530
\2 \ 1 '9 10 30
4 3 2 10
\8 \7 5.
10 9

SALARY DISTRIBUTION FOR THE PERSONNEL OF THE

TEMPLE OF ILLAHUN
Borchardt has given the translation of the salary distributions for the
priests of the Illahun temple during the Middle Kingdom* (Table
11.2). The calculations are for payments in loaves of bread and jugs
of beer for the various personnel of the temple. For distribution there
were 70 loaves, 35 jugs of Sd' beer, and 115 2 jugs of Hpnw beer. The
unit of distribution was one forty-second part of each of these quan-
tities, calculated and written by the scribe as 1 3 loaves, 16 Sd' beer,
and 2 3 TO Hpnw beer. This shows an error by the clerk in the por-
tions of Hpnw beer; it should be 2 2 4, so that he records 60 of a jug
too much. This was a pity, because it made his calculations more
* L. Borchardt, "Salary Distribution for the Personnel of the Temple of
Illahun," ZeitschriiftfiirAgyptischeSprache, Vol.40 (Leipzig, 1902-1903), pp.
113-117.
 Equitable Distribution and Accurate Measurement 125

TABLE 11.2
Salary distributions of the personnel of the Temple of Illahun, based upon
Borchardt's translation (Z. Agypt. Spr., Vol. 40, pp. 113-117).
Number of Loaves of jugs of jugs of Corrected
Portions Bread S4' Beer Hpnw Beer Hpnw Beer
Personnel 42 11 16 2 3 TO- 224
The temple director 10 16 83 27 3 27 2
Head lay priest 3 5 22 8 5 10 84
Head reader 6 10 5 16 2 i0 16 2
Scribe of the temple 13 2 6 IS 13 3 3 45 33
Usual reader 4 63 33 11 15 11
Wtw priest 2 33 1 5 2 30 52
Imi ist c priest 2 33 1 5 230 52
Ibh priests (3) 6 10 5 16 2 TO- 16 2
Royal priests (2) 4 63 33 11 15 11

Md3w 1 13 3 23 10 224
Thur guardians (4) 13 2 6 18 1 3 3 45 33
Night watchmen (2) 19 2 18 1 2 3 90 1 16
Temple worker 3 2 18 436 3 4 180 34
Another worker* 3 2 18 4 36 3 4 180 34
Totals (clerk) 42§ 70§ 35§ 115 2#
Totals, without an-
other worker 41 69 3 9 34 3 18 115 6 S 114 3 4
Totals, including an-
other worker 42 70 35 115 2 115 2

Omitted by the clerk. Or perhaps there should have been two temple workers.
§ Clerical errors, but only if there was in fact only one temple worker.
# If there were two workers, as seems most likely, then this should be 116 3.Ob-
viously the clerk did not add up all the fractions. He knew what they ought to total,
and so he just wrote the numbers down without checking.

difficult. These he performed without further error-except that,

when checking his totals, in column 4 he put the total as 115 2 jugs
of Hpnw beer as he knew it should be, instead of 116 5, which it intact
adds up to. The difference is 2 3, which of course resulted from his
original error of 60 jug too much multiplied by 42, which is 2 5. The
clerk cheated a little here, and we cannot tell how he made the distri-
bution, as his records claim that he did; for either he had 2 18 of a
loaf, 6 5 jugs of Sd' beer, and 6 18 jugs of Hpnw beer left over, or-
126 Chapter Eleven

if the correction I have made (really an omission corrected)* is the

right one-he found that he was 2 5 short of Hpnw beer. Thus,
because of his initial error, the last temple worker may have received
only 6 18 jugs of the last kind of beer instead of 14 180 jugs.
No explanation is given for the number of portions allotted to each
person. One is surprised to note that the head reader was paid twice as
much as the head lay priest, and the usual reader, three times as much as
the scribe of the temple!
This clerical record interests us not only because it sheds light upon
the relative importance of the various temple personnel; it also causes
us to ask, were such meticulous fractional divisions actually carried
out in bread and beer? And if they were, how were they done? How,
for instance, would the distributor measure out a 45, a 90, or a 180 part
of a jug of beer? We have already seen how loaves of bread were
divided equally among certain numbers of men in the RMP problems
1 to 6 (pp. 105 and 120-124). Then we can check how the fractional
parts of the loaves could in actual fact be distributed according to the
clerk's calculations recorded in Table 10.2:
Fractional portions Number
4
2
2
fi 2
1

18 4.

This is a total of 15 fractional portions, and we see that these could be

exactly cut from 5 loaves, with no surplus. For the sum of the fractions
just given is 5 (2 1 + 9 + 1 + 9 + S + fi 18). Whether the scribe
knew this or not, we do not know.
Figure 11.1 shows how such a cutting up of loaves could have been
done. Indeed, it would appear that the whole of the calculations of
Table 11.2 would need to have been completed before any attempt at

* I have added the line, "Another worker," or temple worker, so that the
total of portions is 42 exactly, just as it should be.
 Equitable Distribution and Accurate Measurement 127

distribution was made; if the order of payment was as indicated in

Table 11.2, the scribe of the temple would get a portion of the fourth
loaf, and a portion of the fifth loaf. How he received 5- of a jug of beer
I do not know! But we can hazard a guess regarding the Sd' beer by
supposing a dipstick with 36 division marks, or perhaps equally-
spaced marks on the containers themselves. But even if this sounds
plausible, it strains our credulity too far to suppose that with the
clerk's fractions for the Hpnw beer he would so divide the stick or the
container into 180 equal parts and make the division in the same way.
Of course, had he not made his original error of division, then he would
only have needed a dipstick with 12 divisions, to measure out 3, 2, 4,
and fi, and this would have been relatively easy.

ti
I 2 I 2 I

I 3 IJ

I
3

I
3

I 3

FIGURE 11.1
Cutting up five loaves.
 12 PESU PROBLEMS

A peso is a measure of the strengths of beer or bread, after either of

them is made. It is not a measure of the quality of the barley, wheat,
wedyet flour, emmer, besha, spelt-date, dattel, or grain that may have
been used to make the beer or bread, although of course all these com-
modities could vary in quality and strength.* The pesu of the beer or
bread made was determined by the Egyptians thus : If one hekatt of
grain were used to produce only one loaf or one des-jug of beer,+ then
the pesu of both the bread and the beer was said to be one; if one hekat
produced two loaves or two des-jugs of beer, then their pesu was said
to be two; if one hekat produced three loaves or three des-jugs of beer,
then their pesu was three; and so on, so that the higher the pesu, the
weaker the beer or bread, and possibly the smaller the loaf. The
relation between the amount of grain used and the pesu of the beer
or bread produced was thus:

_ number of loaves or jugs

pesu
number of hekats of grain
Generally speaking, when the Egyptians made beer and bread, they
used more grain for their beer than for their bread, or we could say
that the same quantity of grain would produce more loaves than des-
jugs of beer, which was therefore relatively stronger. In the twenty
"pesu-problems" of the RMP and MMP (ten in each), the values of
the pesus of the beers lie between 1 and 4, while for the loaves of bread
the pesus vary from 5 to as much as 45.
The first of the ten pesu problems of the RMP is number 69:
3 2 hekats of meal are made into 80 loaves. Find the amount of meal
in each loaf and the pesu.

* Authorities use various terms to describe pesu; for example, cooking

ratio, baking number, or baking value.
t Chace gives I hekat = 292.24 cubic inches and Vogel I hekat = 4.75
liters; roughly 1/9 bushel.
$ A des-jug of beer was approximately 7/e of a pint.
 Pesu Problems 129

Since there are 320 ro in each hekat, the scribe first found the number
of ro in 3 2 hekats.
\1 320/
\2 640/
\2 160/
Totals 3 2 1120 ro.
He then divided 1120 ro by 80. He wrote,
Make thou the operation on 80, for the finding of 1120. The doing as
it occurs.
1 80
\10 \800
2 160
4 \320
Totals 14 1120.

The answer is 14 ro in each loaf, or 32 hekat 4 ro.

To find the pesu of each loaf, he only had to divide 80 by 3 . He
wrote : Make thou the operation on 3 2 for the finding of 80.
1 3 Q

\20
10 35
70 //
//
\2
\
\21
7

\7 /
Totals 22 ' 7 21 80.
The pesu is 22 3 7 T.

Problems 70 and 71 are like Problem 69, but the remaining seven
(Problems 72 to 78) deal with exchange of loaves and beer. Problem
73 is:
100 loaves of pesu 10 are to be exchanged for loaves of pesu 15. How
many of these will there be?
130 Chapter Twelve

The scribe wrote : "Reckon the amount of wedyet flour in these 100
loaves. It is 100 divided by 10, namely, 10 hekats. The number of
loaves of pesu 15 from 10 hekats is 15 times 10, namely, 150. This is the
number of loaves for the exchange."
The other six problems on the exchange of loaves and beer of
different pesus appear at first glance to be similar to Problem 73, but
a closer examination reveals some new scribal techniques and arith-
metical procedures. We look then at Problems 74 and 76.
PROBLEM 74
1,000 loaves of pesu 5 are to be exchanged, a half for loaves of pesu 10,
and a half for loaves of pesu 20. How many of each will there be?
PROBLEM 76
1,000 loaves of pesu 10 are to be exchanged for a number of loaves of
pesu 20 and the same number of loaves of pesu 30. How many of each
kind will there be?
The scribe's solution for Problem 74 reads as follows: "1,000 loaves
of pesu 5 require 200 hekats, and if these are halved, a half of 200
hekats is 100 hekats. Multiply 100 by 10; it makes 1,000, the number
of loaves of pesu 10. Multiply 100 by 20; it makes 2,000, the number
of loaves of pesu 20. The answer is 1,000 and 2,000 loaves."
The solution for Problem 76 is:
For loaves of pesu 20, the first kind, 20 hekat produces 1 loaf. For
loaves of pesu 30, the second kind, 30 hekat produces I loaf. Then
20 30 = 12 hekat produces 2 loaves, one of each kind. Then 1 hekat
will make 24, or 12 loaves of each kind. The quantity of wedyet flour
in the 1,000 loaves of pesu 10 is 100 hekats. Multiply 100 by 12; the
result is 1,200, which is the number of loaves of each kind for the
exchange.
In order to acquaint ourselves with the scribe's processes in this
problem, we set it down in modern terms using what is usually called
the unitary method.
First, 1,000 loaves of pesu 10 required 100 hekats of flour. Now 1
hekat of flour produces 20 loaves of pesu 20, and I hekat of flour
produces 30 loaves of pesu 30.
 Pesu Problems 131

Then 3 hekats of flour produce 60 loaves of pesu 20, and 2 hekats of

flour produce 60 loaves of pesu 30.
Therefore 5 heka.s of flour produce 60 loaves of each kind, so that 100
hekats of flour produce 60 times 20 loaves of each kind, and the result
is 1,200 of each kind of loaf.
A further modern approach now becomes clear to us following the
foregoing solution, using the principle of the harmonic mean.
The harmonic mean of the two pesus 20 and 30 of the exchange
loaves is twice their product divided by their sum, so that the har-
monic average of the pesus of the two kinds of loaves considered
together is (2 x 20 x 30) _ (20 + 30) = 24. Now the pesu of the
original 1,000 loaves was 10, so that the total number of loaves to be
received in exchange is greater in the ratio of 24 is to 10, namely,

1,000 x lU = 2,400 loaves.

Then there will be a half of 2,400 or 1,200 loaves of each kind, pesu
20 and pesu 30, received in exchange for 1,000 loaves of pesu 10.
How similar Problems 74 and 76 appear on casual reading, the
first asking for half for loaves of one pesu and a half for loaves of
another, while the second problem asks for equal numbers of loaves
of the two pesus the scribe mentions. But it was a trap for the unwary.
It is interesting to note how well A`h-mose chose his two numbers 20
and 30 for the pesos. The choice is reminiscent of the modern problem
which asks: If a man drives from one town to another at an average
speed of 20 miles per hour, and returns at an average speed of 30 miles
per hour, what is his average speed for the double journey ? This is
like A`h-mos4Ps Problem 76, for the answer to both problems is the
harmonic mean of 20 and 30, which is 24 and not 25 (the arith-
metic mean). Problem 74 is one in arithmetic means.
The problems of the MMP* dealing with pesus are much the same
as those of the RMP, but the scribe was not as careful in his copying
* W. W. Struve. "Mathematischer Papyrus des Museums der Schonen
Kunste in Moskau," Quellen and Studien zur Geschichte der Mathematik, Part
A, Vol. 1 (Berlin, 1930), p. 98.
132 Chapter Twelve

as was A`h-most. In addition there are some minor errors of arith-

metic, so that the meanings of some of the problems are not entirely
clear.' MMP 21 is one such problem, and the reader is invited to
decide for himself whether or not it resembles Problem 76 of the RMP
and whether the suggestion of a harmonic mean can be found there.

1. 1 Method of calculating the mixing of sacrificial bread.

1.2 If one names 20 measured as $ of a hekat and 40 measured as F6_
of a hekat,
1. 3 compute 8 of 20. Result 2 2.
1.4 Compute 16 of 40. Result 2 2.
1.5 The total of both these halves is 5.
1. 6 Compute the sum of both halves. Result 60.
1. 7 Divide thou 5 by 60.
1.8 Result 12. Lo! the mixture is 12. You have correctly found it.

In line 7 the scribe divided 5 by 60 where we would have expected 60

divided by 5 giving the pesu of the sacrificial bread as 12. If one hekat
of grain produced 12 loaves of bread then each of these loaves would
have a pesu of 12. But the scribe has expressed this differently by
saying that each loaf contained one twelfth of a hekat of grain, which
is correct. This method of expressing pesu appears to be consistent
with line 2 where the fractions $ and 16 are written for what we would
call pesus 8 and 16. If then following the scribe's thoughts we think
of fractions only, we come quite naturally to the observation that the
answer 12 is the harmonic mean of the two fractions $ and 16, being
equal to twice their product divided by their sum.

EXCHANGE OF LOAVES OF DIFFERENT PESUS

The three problems numbered 72, 73, and 75 of the RMP are all
phrased similarly by the scribe, and they treat the same topic.t
RMP 72
100 loaves of pesu 10 are exchanged for loaves of pesu 45. How many
of these are there?
See Appendix 7.
t Problems 5 and 8 of the MMP are done in the same way.
 Pesu Problems 133

RMP 73
100 loaves of pesu 10 are exchanged for loaves of pesu 15. How many
of these are there?
RMP 75
155 loaves of pesu 20 are exchanged for loaves of pesu 30. How many
of these are there?
Since the greater the pesu, the greater the number of loaves from
the same quantity of meal, these three problems are very easily solved
by simple proportion as follows :
RMP 72
No. of loaves = 100 x 45/10
= 10 x 45
= 450.
RMP 73
No. of loaves = 100 x 15/1,
= 10 x 15
= 150.
RMP 75
No. of loaves = 155 x 30/2o
=724x30
= 232 2.
This is exactly how the scribe did solve them, that is, all except Prob-
lem 72, which for some reason was done in an entirely different
manner. Chace remarks that "He arrives at the result in a round-
about way," which is very much of an understatement, and although
quite true, it adds nothing to our understanding of the scribe's thought
processes in arriving at the correct answer. What was the reasoning
behind this round-about solution? Was it perhaps a more advanced
technique? Was he attempting to introduce some new concept into
mathematical methods?
To attempt to answer these questions, we set down the steps in the
argument exactly as the scribe gave them (Chace's translation from
Vol. 2), line for line.
134 Chapter Twelve

RMP 72
1. 100 loaves of pesu 10 exchanged for loaves of pesu 45. How many
of these loaves are there?
2. Find the excess of 45 over 10. It is 35. Divide this 35 by 10. You
get 3 2.
3. Multiply this 3 2 by 100. Result 350. Add 100 to this 350. You
get 450.
4. Say then that the exchange is 100 loaves of pesu 10
5. for 450 loaves of pesu 45.

In order to examine logically the steps of the scribe's reasoning, we

restate the preceding solution in modern symbolic terms.

1. If x loaves of pesu p are exchanged for y loaves of pesu q, find y

if x, p, and q are known.
2. Find the excess of q over p. It is (q - p). Divide this (q - p) by p.

You get (L;- p)

3. M ultiply this (q P) by x. Result (q p)x. Add x to this. You

P P
get(q p)x+x.
4. Say then that the exchange is x loaves of pesu p
5. for (q P p)x + x loaves of pesu q. Then,
y= (gPP)x+x

= (q - 1)x + x

which is the scribe's formula or method for RMP 73 and 75.

Now how did the scribe come to think of all this? The only data
which he had a priori was the relation,
 Pesu Problems 135

number of loaves
number of hekats of meal =
pecu
Following immediately from this, the scribe can write,
x y
p q
whence, y = x x q/p, which is just what he did for RMP 73 and 75.
But for RMP 72, with the same data, to achieve the same steps in
his argument as shown in our symbolic transcription of his solution,
we are forced to proceed as follows :
Given
x y
p q
then
y q
(T he modern concept of alternando)
x p
and (line 2)
Y
-x x = q p- p (The modern concept of dividendo) ;

hence (line 3)

y - x=(q pp x,

so that (line 3)

therefore,

exactly as before.
136 Chapter Twelve

However one looks at this "round-about" method of solution, it is

entirely logical and indeed elegant, whether or not the scribe arrived
at it by some algebraic or symbolic thought processes, or by some
other means. Whatever the true story behind it is, we can only be
amazed at such an achievement in 1850 B.c., and suggest that here
perhaps, as the scribe wrote it, we are looking at the very earliest
example of rhetorical algebra to come to the attention of the historian
of mathematics. We cannot avoid introducing the concept of alter-
nando and dividendo in some form or other, because of the scribe's
direction in line 2, "Find the excess of q over p."
 13 AREAS AND VOLUMES

THE AREA OF A RECTANGLE

From the extant papyri, etc., it is clear that the scribes found the areas
of rectangles by multiplying length and breadth as we do today. In
Problem 49 of the R MP, the area of a rectangle of length 10 khet (1,000
cubits) and breadth 1 khet* (100 cubits) is found to be 1,000 x 100 =
100,000 square cubits. The area was given by the scribe as 1,000 cubit
strips, which are rectangles, usually of land, 1 khet by 1 cubit (Figure
13.1).
Again in Problem 6 of the MMP, calculation of the area of a rect-
angle is used in a problem on simultaneous equations. The following
text accompanied the rectangle shown on the right in Figure 13.1:

1. 1 Method of calculating a rectangle.

1. 2 If it is said to thee, a rectangle of 12 in the area (is] 2 4 of the
length
1. 3 for the breadth. Calculate 2 4 until you get 1. Result 1 3.
1. 4 Reckon with these 12, 1 3 times. Result 16.
1. 5 Calculate thou its angle [square root]. Result 4 for the length.
1.6 2 4 is 3 for the breadth.

FIGURE 13.1
D A,,

Rectangles from the RMP and the MMP. Left, the 1-khet by 10-khet
rectangle of Problem 49 of the RMP. Note that the scribe has made an
error in copying, showing the breadth as 2 khet. The working accompany-
ing the figure was done for a breadth of 1 khet. Right, a rectangle of area
12 and breadth 2 4 of the length, from Problem 6 of the MMP. The scribe
wrote his correct answers for breadth and width on the figure as shown.

* The scribe made a careless copying error, putting 2 khet for I khet, and
he repeated it in his figure. But there are no errors in his calculations with
1 khet.
138 Chapter Thirteen

Rewriting this in modern form, we would have:

If
A = lb = 12 and b=(24)1, (from 1. 2)
to solve for 1, first substitute for b: 1 x (2 4)1 = 12.
Then
1x1=12=24 (1.3)
=12x1 (1.3)
= 16. (1.4)

Therefore
I = 4 for the length, (1. 5)
and

2 4 of 4 is the breadth 3. (1.6)

THE AREA OF A TRIANGLE

For the area of a triangle ancient Egyptians used the equivalent of the
formula A = 1/2bh. In RMP 51 the scribe shows how to find the area
of a triangle of land of side* 10 khet and of baset 4 khet. The scribe
took the half of 4, then multiplied 10 by 2 obtaining the area as 20
setats of land. Then in MMP 4 the same problem was stated as finding
the area of a triangle of height* 10 and baset 4. As stated by the scribe
the method was to calculate with a half of 4 and then to reckon with
10 twice, giving an area of 20. No units such as khets or setats were
mentioned.
There have been differences of scholarly opinion among philologists

eld

FIGURE 13.2
Triangles accompanying Problem 51 of the RMP (left) and Problem 4
of the MMP (right).
* The word meret (or meryct) translated as side or height.
t The word Leper (or tepro) translated as base or mouth.
 Areas and Volumes 139

regarding the precise meaning of meret and of Leper. If maet meant the
side rather than the height, the area would be in error unless the
triangle were right-angled. Scribal sketches of triangles in the papyri
suggest that a right-angle may have been intended, but some observers
have thought that isosceles triangles with acute vertical angles were
meant, and if matt was side the area would again be in error by a
small amount. However, these differences of opinion are academic;
and modern-day historians agree that perpendicular height is meant
by the scribe. Thus Peet writes* that he believes "that the Egyptians
had found the correct formula, half the base multiplied by the vertical
height for the scalene triangle."
D. J. Struik writes,t "The area of a triangle was found as half the
product of base and altitude," while Carl B. Boyer has,+ "Problem 51
(of Ahmes) shows that the area of an isosceles triangle was found by
taking half of what we would call the base and multiplying this by the
altitude."

THE AREA OF A CIRCLE

In Problem 50 of the RMP, the scribe showed how to find the
area of a circle. To explain his method, he assumed a circle of diameter
9 khet as a matter of arithmetical convenience and not because it is a
really practical problem. A khet was 100 royal cubits (approximately
57 yards) ; a square khet was called a setat (about two-thirds of an acre).
The circle the scribe refers to would have an area of over 40 acres and
a circumference of nearly a mile.

FIGURE 13.3
Circles drawn by the scribe of the RMP. Left, from Problem 41; right, from
Problem 50.
* T. E. Peet, "Mathematics in Ancient Egypt," Bulletin of the John Rylands
Library, Vol. 15, No. 2 (Manchester, 1931), p. 430.
t D.J. Struik, A Concise History of Mathematics, Dover, New York, 1948, p. 22.
$ Carl B. Boyer, A History of Mathematics, Wiley, New York, 1968, p. 18.
140 Chapter Thirteen

A circle with hieratic inscriptions (Figure 13.3) was included by

the scribe in his statement of Problem 50. A translation of Problem
50.is:
Take away S of the diameter, namely, 1.
The remainder is 8.
Multiply 8 by 8.
It makes 64.
Therefore it contains 64 setat of land.
Do it thus.

The remainder is 8.
1 8
2 16
4 32
\8 \64
The area is 64 setat.
The scribe's method of finding the area of a circle can thus be restated :
Subtract from the diameter its one-ninth part, and square the remainder. This
is its area. We ask ourselves how close this is to the true value, and how
did the scribe arrive at his formula? If we use the modern value for ir,
a circle of diameter 9 khet would have an area of 63.6174 setat, so
that the Egyptian value is in error by less than 0.6 of one percent.

lit . +
FIGURE 13.4
Problem 48 of the RMP as the scribe wrote it, including his geometrical
illustration.
 Areas and Volumes 141

Problem 48 of the RMP is unique among the 87 problems of the

papyrus in that the usual statement of what A`h-mose proposed to do
was replaced by a geometrical illustration (Figure 13.4) from which,
clearly, the reader is expected to deduce and to understand the nature
of the problem. A. B. Chace in his translation of the RMP has
written :
PROBLEM 48
Compare the area of a circle and of its circumscribing square.
The circle of The square of
diameter 9 side 9
1 8 setat \1 9 setat
2 16 setat 2 18 setat
4 32 setat 4 36 setat
\8 64 setat \8 72 setat
Total 81 setat
We note that Problem 50 and the first part of Problem 48 appear
to be the same, yet their expressions are by no means identical. Indeed,
even the numbers as written in the hieratic are not the same. In Prob-
lem 48, the 9 is written , the special sign for 9 khet, also used for 9
hekats of grain, instead of the normal hieratic 9, written &(,, as in
Problem 50 and also in the geometrical diagrams of these problems,
where in both it means 9 khet. In Problem 50 the scribe is thinking
and writing ordinary numbers simply as units, whereas in Problem
48 he is thinking of setats or square khets as his units, and this is the
important circumstance that gives us the clue as to the true purpose
of Problem 48. We must conclude that Chace was in error when in
translating he interpreted the scribe's working as meaning that the
figure is supposed to represent " a circle and its circumscribing square."
Chace himself must have had some doubts about the circle, even
though he may have thought that it was good enough for a freehand
drawing with a scribal reed pen, or fine brush, for he had only to
glance at the excellent freehand circle of diameter 9 khet of Problem
50, drawn with the same hand with just two confident sweeps of the
pen, to see how good a calligrapher A`h-most really was. Another
circle of diameter 9 cubits in Problem 41 was equally well drawn
by A`h-mose (Figure 13.3), and there are other circles in Problems 42
142 Chapter Thirteen

and 43 which support us in the conclusion that the figure within the
square of Problem 48 is an octagon with straight sides drawn within
a previously drawn square of side 9 khet, and not a circle with a square
circumscribing it as Chace supposed. Let us look more closely at this
octagon of Figure 13.4 now that we agree that it really is one. Was
A`h-mose intending to inscribe a regular octagon by joining eight
points on the four sides of the square? The answer is no! We have no
evidence that the ancient Egyptians knew the geometrical construc-
tion for determining these points, such that by cutting off the four
corners of a square the resulting figure would be an eight-sided figure
with all the sides of equal length. And even if A`h-mose did know of
such a construction, he would also know that it would produce a
regular octagon of area most certainly greater than that of a circle of
diameter 9 units, for it would be an escribed octagon like the square
itself. Of course, it would be a much closer approximation to the area
of the circle than the square would be, but he would also surely see
how he could find a much closer approximation to the circled area,
and with a much simpler and more obvious construction. All he would
need to do would be to join the adjacent points of trisection of the sides.
That this is in fact what he did, or was aiming to do, is suggested by
his careful choice of a square of side 9 units to allow of easy trisection.
Such an octagon would have each pair of opposite sides equal, and
in the papyrus except for the top right-hand corner this certainly
appears to be the case.
Vogel (1958) came to the same conclusion when he compared our
modern formula for the area of a circle, F = Ir(d/2)2, with the equiv-
alent of the Egyptian formula, F = (8d/9)2,* from which one derives
an Egyptian value for jr of 25% , which is approximately 3.1605.
Vogel then remarks, "Just how this remarkably close approximation
was found, we do not know, but we can offer a suggestion on examining
the diagram of RMP 48." He then refers to the diagram of Figure
13.5, which he says ". . . seems to represent a figure whose area
approaches the area of a circle inscribed in the square." t
* K. Vogel, Vorgriechische Mathematik, Vol. 1.
t Ibid., "scheint einen Kreis in Annaherung darzustellen, der einem
Quadrat einbeschrieben 1st."
 Areas and Volumes 143

FIGURE 13.5
Vogel's diagram of the inscribed octagon of Problem 48 of the RMP.

Then, he says, the area of the octagon is equal to the original square
less the two small squares made up by the four cut-off corners. "Then
the area of the octagon is (81 - 4 x 9/Z) = 63, which would corre-
spond to a square whose side is 163, which is approximately 64 = 8.
Thus probably the area of a circle formula (8d/9)2 might have
originated."*

ALTERNATE METHOD OF OBTAINING THE FORMULA

By drawing a diagram as that shown in Figure 13.6 on a piece of
papyrus, the scribe would conclude that the octagon was pretty
closely equal in area to the inscribed circle because some portions of
the circle are outside the octagon and some portions of the octagon
are outside the circle, and mere observations by the naked eye suggest
these are roughly equal. He then would sketch a square of sides repre-
senting 9 khet, trisect the sides, join the adjacent points of division,
and then by drawing all the lines necessary to actually see or visualize
each of the square khets or setats he can count these squares in any
way he pleases to find the number of them in the octagon (see Figure
13.6).
Now the Egyptian scribes found the areas of squares and rectangles
with ease. Then if the two top shaded corners of 4 2 setats (or square
khets) each, which add to 9 setats, were to replace the top row of 9
setats and if, similarly, the two bottom shaded corners of 9 setats were
to replace the left-hand column of 9 setats, then the figure remaining
* Ibid.
A t

m
LI

%.00IN
_ 00 w MEN
...
\R 0-0I/
I..I;
 Areas and Volumes 145

would be a square instead of an octagon, the area of which the scribe

can easily calculate.
The scribe could now properly conclude that the area of the circle
inscribed in a square of side 9 khets is very closely equal to the area
of a square of side 8 khets.
Therefore, in Problem 48 the scribe finds the total number of
setats in this square of eight rows of 8 setats, which he gives as the area
of a circle ofdiameter 9 khets. Of course he certainly knows his method
is not exact, because he has, so to speak, cut off one setat twice-the
one in the top left-hand corner. But his method allows him to find a
square nearly equal to a circle, so that we can, "en caprice," as it were,
credit A`h-mose with being the first authentic circle-squarer in record-
ed history! Problem 48 shows us that in counting the number of setats
in his new square he notes that there are eight rows, each containing
8 setats, so that he multiplies 8 setats by 8, giving him 64 setats. Each
step of his calculation gives the product in the special setat hieratic
number signs. He does the same in his second multiplication, where
he finds the number of setats in the original square to be 81 setats.
Nowhere in this problem does the scribe give the direction, "Take
away thou one-ninth of the diameter," as he does in the four other
Problems 41, 42, 43, and 50, for this is where he is showing how he
discovered his now classical rule. And now we have to meet the critics
of ancient Egyptian mathematics, e.g., Sloley,* " Mathematical
knowledge in ancient Egypt was essentially practical in character,
and must have developed as occasion arose in dealing with problems
encountered in daily life." Then Peet, ". . . the Egyptians evolved
no better means of stating a formula than that of giving three or four
examples of its use, and this is hardly a tribute to the scientific nature
of their mathematics."t Such statements as these are legion in histo-
ries of mathematics, and they bring to our attention a consideration
of the question, What is the nature of proof?
Nowadays one tends to consider that an argument to be rigorous
* R. W. Sloley, in The Legacy of Egypt, S. R. K. Glanville, ed., Oxford
University Press, London, 1942, p. 173.
t T. E. Peet, "Mathematics in Ancient Egypt," Bulletin of the John Rylands
Library, Vol. 15, No. 2 (Manchester, 1931), p. 439.
146 Chapter Thirteen

must be symbolic. This is not true. A nonsymbolic argument can be

quite rigorous, even when given for a particular value of the variable*
(in this case it is 9). The only requirements are that the particular
value be typical and that the generalization to any value be immediate.
Both of these requirements are satisfied in the argument we have
attributed to the scribe in deducing his method, for the value of the
diameter, 9, is typical, and the generalization to any value like 12 or
15 is immediate, as it is also to a number like 20, even though it takes
a little longer. The rigor of the argument is implicit in the deduction.
Peet devotes five pages to Egyptian geometry but only eight lines
to the circle, even though there are five problems of the RMP deal-
ing with its area. He writes:
The best achievement of the Egyptians in two-dimensional geometry
is undoubtedly their close approximation to the area of the circle.
They squared eight-ninths of its diameter, giving 256/81r2, where r
is the radius. Comparing this with our own 7rr2 we get for the Egyptian
value of it 256/81, or 3 13/81, a very close approximation to 3 1/7,
which we find good enough for practical purposes. We have no idea
how this result was obtained. The expression of the area as a square
suggests a graphic solution.t
Peet's last sentence appears certainly to have been prophetic if the
analysis based on Figure 13.6 is accepted.

THE VOLUME OF A CYLINDRICAL GRANARY

The method for finding the volume of a cylinder was, just as it is
today, to first find the area of the circular base and then multiply by
the height. In Problem 41 of the RMP, for a granary of diameter 9
cubits and height 10 cubits, the scribe subtracted from 9 its 0 part,
then multiplied 8 by 8, giving 64, then multiplied by 10, to get 640
cubic cubits. This he expresses in a more convenient unit for grain by
multiplying by 1 2, for there are 1 2 khan in a cubic cubit, so the vol-
ume is found to be 960 khar. This last step is analogous to a modern
calculation in which the volume would be found in cubic feet, then

* I am indebted to Mr. H. Lindgren of Canberra, A.C.T., for suggesting

this aspect of Problem 48.
t "Mathematics in Ancient Egypt," p. 434.
 Areas and Volumes 147

afterwards be converted to conventional bushels, or cubic feet to

gallons of liquid. This is a first, easy problem, to show the method.
Problem 42, which we now discuss, is much more difficult and requires
some complicated arithmetic. In this problem we are required to find
the contents in khar of a cylindrical granary of diameter 10 cubits and
height 10 cubits.
The scribe of the RMP, by the method which he revealed in Prob-
lem 41, found the contents to be 1185 6 54 khar, and the arithmetic,
which involves many operations in unit fractions, will well repay a
careful examination. But first we look at a modern solution, using the
Egyptian equivalent of zr, 25%1.
V= Irr2h
= 256/81 x 52 x 10
=791/81 x 10
= 7901% cubic cubits.
To express this in khar, add to 7901% its half, which is 395%, which
gives a total of 1185%7 khar. Now we express %7 in unit fractions, in
the Egyptian tradition:
%7 = %7 + %7 = 9 18 54 = 6 54 khar.
Here is a complete statement of the scribe's solution of Problem 42:
Take away 9 of 10 namely 1 9.
The remainder is 8 9 6 18.
Multiply 8 3 fi 18 by 8 9 fi 18.
It makes 79 108 T24 square cubits.
Method of working out:
1.1 1 8 5 fi 18
1.2 2 17 5 9
1.3 4 35 2 18
1.4 \8 \71 9
1.5 3 \5 3 6 T8- F7
1.6 9 2 9 6 12 36 54
1.7
1.8 \18 \ 1 9
3
12
9
24
27
72
108
T08
T24

1.9 Total 79 108 324.

148 Chapter Thirteen

1. 10 1 79 108 324
1.11 10 790 18 27 54 81
1.12 2 395 36 54 108 162
1.13 Total 1185 fi 54 [khar].
It is left to the reader to examine and explain how the scribe:
found $ of 10 to be 1 9;
subtracted 1 9 from 10 to give 8 3 9 18;
squared the number 8 3 6 18 to get 79 108 324;
multiplied this by 10 to get 790 18 27 54 F1_; and
multiplied this by 1 2 to find the contents, 1185 G 54 khar.
In Problem 43, the third problem on the contents of a cylindrical
granary in the RMP, the scribe proposed to show how the contents
may be found directly in khar without having first to calculate the
volume in cubic cubits. The scribe did this by means of a very neat
and interesting transformation that, correctly interpreted, throws
light on the mental processes of the Egyptian scribes. But the copyist
A`h-mose made two errors in his copying of the earlier work from
which the RMP was derived. Because of these errors, it was some years
before the problem was properly understood upon the work of Schack-
Schackenburg, who correctly interpreted a similar problem in the
KP.* In that problem, KP IV, 3, the scribe of the KP obtains the
contents of a cylindrical granary directly in khar without first finding
the volume in cubic cubits, by adding to the diameter its one-third
part, squaring this number, then multiplying by two-thirds of the
height. In copying a similar calculation in Problem 43 of the RMP,
A`h-mose gave the wrong dimensions for the silo; then he added an
extra line of directions, which almost certainly was a carry-over from
the two previous problems. The extra line, "take away S of the diam-
eter," should have been replaced by, "add to the diameter its i," and
not been an addition to it. With these corrections made, Problem 43
should read :
A cylindrical granary of diameter 8 and height 6. What is the amount
of grain that goes into it?
* F. Griffith, The Petrie Papyri : Hieratic Papyri from Kahun and Gurob. Univer-
sity College, London, 1897, pp. 15-18.
 Areas and Volumes 149

Method of working out:

\8
5 5
\2
Totals 1 3 10 3.

1 10
\10 \106
\3 \7 $
Totals 10 113 3 $.

1 113 9
2 227 2 18
Totals \4 \455 $ khar, the answer.*
The scribe does not bother to show that 4 is two-thirds of 6 cubits.
The second and third lines of doubling show a neat piece of mental
reckoning:
1 113 9 $
2 226 1 5(6 18) (Recto 2 - 9)
or, 227 2 18 (G rule)
\4 \454 1 $
or, 455 9.

By taking 7r equal to 3'/T, we find that the scribe's answer exceeds

the correct value by only 1.8 percent. It is easy enough for us to deter-
mine how the scribe's transformation can be arrived at; but it is quite
another matter to determine how the ancient Egyptians themselves
arrived at the transformation with the much cruder methods and
arithmetical tools at their disposal. Despite these handicaps their
formula, checked against their standard rule, is found to be correct to
within 0.56 percent, also in excess. Let us attempt to see how this may
* We can find modern counterparts to this simplification of a formula in
almost any technical manual. For example, in Shapes and Sections (a manual
of Lysaght's, a subsidiary of B. H. P., the Australian firm, Melbourne,
1937, p. 438), we find the dimensions of a cylindrical tank given in feet,
the contents being found directly in imperial gallons rather than in cubic
feet, by means of the formula V (in gallons) = 5 d2h.
150 Chapter Thirteen

have been done, using only those operations that were available to the
scribes. The rules are as follows.
STANDARD RULE

Subtract from the diameter its one-ninth.

Multiply this number by itself.
Multiply by the height.
Multiply this number by 1 2 (or add to it its half).
NEW FORM OF RULE

Add to the diameter its one-third.

Multiply this number by itself.
Multiply by two-thirds of the height.
For our own convenience we put d and h for diameter and height, and
we thus have to transform
(d-9d) x (d-9d) x h x (1 2)

into
(d + 3d) x (d + 3d) x 3 x h.
Now,

(d - 9d) x (d - 9d) x h x (1 2)
1.1 =('3+fi+18)dx('3+9+18)dxhx(1 2)
1.2 = ('3+6+ 18) x 1 2 x (3+6+ 18) x (1 2)
xdxdxh+(12)
1.3 =(1+4+12) x (1+4+ 12) x d x d x h x ( )
1.4 = (1 +3)d x (1 +3)d x 5h
1. 5 = (d + 3d) x (d + 3d) x 3h.
This is the sequence of operations that could have been made by
the scribe (but not, of course, in this form), although it is quite possible
that some other sequence might be found to be equally tenable. Line
I of the sequence, (1 - 9) = 3 fi 18, is shown in line 1 of Problem 42
of the RMP (p. 147). In line 2 of the sequence, multiplication by an
extra 1 2 is balanced by division by 1 2, which in line 3 becomes a
 Areas and Volumes 151

multiplication by 3, the well-recognized reciprocal. And again in line

3, multiplying 3 fi 18 by 1 2 is

\'
1
\
\1

2
\3
(9
fi
12

3) (fi
18
36

12) (18 36),

or 1 4 12, (G rule)
and is line 4, 1 3. (G rule)

However the scribe may have devised the method of Problem 43,
whether it was along the lines I have described or any other, there
can be no doubt that, for his era, he was a mathematician of no mean
ability.
In Problems 41, 42, and 43 of the RMP, the answers given here in
khar are expressed in terms of hekats of grain (where 20 hekats equal
I khar) ; but as the numbers are likely to be large for a granary,
instead of multiplying by 20 the scribe divides by 20 and calls the
answer hundreds of quadruple hekats. This is not of importance re-
garding the volume of cylinders, but it will come up again when tables
of weights and measures are discussed.

THE DETAILS OF KAHUN Iv, 3

As in other mathematical workings contained in the Kahun Papyrus,
no explanation of what he proposes to do is given by the scribe. All
that one reads is the drawing shown in Figure 13.7, accompanied by
the following calculation :

Totals 1 3 16

\1 \ 16
\10 \160
\5 \80
Totals 16 256
152 Chapter Thirteen

\1 \ 256
2 512
\4 \1024
\3 \ 85 5
Totals 59 1365 9.
The scribe was here finding the contents of a cylindrical granary of
diameter 12 (cubits) and height 8 (cubits). Now if he had followed the
standard procedure illustrated by the scribe of the RMP in Problem
50, he would have taken away S of the diameter 12, squared this, and
then multiplied by the height 8. This would have given him the
volume 910 18 in cubic cubits. If he required the answer in khar,
he would add a half of 910 fi 18 to itself, giving 1365 9 khar, since
there are 11// khar in a cubic cubit. This is the answer the scribe has
written within his freehand-drawn circle to represent the cylindrical
granary.
But here the scribe of the KP has used a different technique for the
volume of a cylindrical granary, by means of which the contents is
determined directly in khar, and it gives evidence of considerable
ingenuity on the part of the ancient Egyptian who devised it. In
modern terms, it amounts to establishing the "new rule" given on
p. 150. The interpretation of the scribe's method puzzled the original
translator, F. L. Griffith, who wrote,
It would seem as though the problem had been to find the contents
of a circular granary, of which the height and the diameter were 12
and 8 cubits respectively; but if so the method adopted and the result
are quite wrong, whether we look for the answer in cubits cubed, in
khar, (= 2/3 cubits cubed), or in quadruple heqat.*
12

FtoURE 13.7
The circle drawn in KP IV, 3, with a translation of the accompanying
hieratic.
* The Petrie Papyri.
 Areas and Volumes 153

With so little indication given by the scribe, it is not so surprising that

the numbers 12 and 8 written outside the circle were erroneously
considered by the translator to refer to the height and diameter,
rather than the opposite. Thus the scribe shows the working according
to the new rule given on p. 150: 1 3 times 12, giving 16. Then squaring
16 to obtain 256, which the scribe then multiplied by 5 3, giving the
answer, 1365 3 khar. Had he shown the working, or given a hint
that the multiplying factor 5 3 was I of the height 8, he would have
helped his readers considerably. But finding two-thirds of any num-
bers at all was mere routine to the Egyptian mathematician, and he
no doubt read it off from his two-thirds tables, and thought no more
about it. R. C. Archibald, in his bibliography to Chace et al., The
Rhind Mathematical Papyrus, wrote that Schack-Schackenburg (1899)
was the first to explain this problem.
 14 EQUATIONS OF THE FIRST AND
SECOND DEGREE

PROBLEMS 24-34 OF THE RMP: SOLUTION OF

FIRST-DEGREE EQUATIONS
As we have seen, many of those who have studied Egyptian mathe-
matical papyri have expressed the opinion that Egyptian mathematics
was wholly practical and indeed proceeded by trial-and-error. In
order to refute this charge of lack of scientific attitude of mind and the
charge that Egyptian mathematicians were concerned only with the
practical arithmetic of everyday life, I now exhibit Problems 24-34
of the RMP. These eleven problems deal with the methods of solving
equations in one unknown of the first degree. Based upon the order
of difficulty and method of solution, these problems fall into three
groups.

THE FIRST GROUP

Pr. 24: A quantity and its 7 added becomes 19. What is the quantity?
Pr.25: 2 16. ?
Pr.26: 15. ?
Pr.27: 21. ?

Each of these problems is solved by the method known as false

position, and each deals with abstract numbers unrelated to loaves of
bread, hekats of grain, or the areas of fields. It is as if the scribe is
showing with four similar problems, but different numbers, a general
method of solution for this type of question. The number "falsely
assumed" in each case is the simplest that could be chosen, namely,
7, 2, 4, and 5, respectively.

PROBLEM 24
Assume the false answer 7. Then 1 7 of 7 is 8. Then as many times as
8 must be multiplied to give 19, just so many times must 7 be multiplied
to give the correct number. Thus, divide 8 into 19.
 Equations of the First and Second Degree 155

1 8
\2 \16
2 4
\4 \2
\$ \1
Totals 2 4 $ 19.
Now multiply 2 4 $ by 7.
\1 \2 4 $
\2 \4 2 4
\4 \9 2.

Totals 7 15 (2 2) (4 4)$
7 16 2 $.
The answer, then, is 16 2 $.
PROBLEM 25
Assume the false answer 2. Then 1 2 of 2 is 3. Then as many times as
3 must be multiplied to give 16, just so many times must 2 be multiplied
to give the correct number. Then divide 3 into 16.
\1 3
2 6
\4 \12
2
\$ \1
Totals 5 5 16.
Now multiply 5 $ by 2.
1 5 $
\2 \10
The answer is 10
PROBLEM 26
Assume the false answer 4. Then 1 4 of 4 is 5. Then as many times as
5 must be multiplied to give 1 5, j ust so many times must 4 be multiplied
to give the correct number. Then divide 5 into 15.
* Note the method for finding one-third of a number. The scribe first found
two-thirds of it, and then halved it, even for the simple case of 1/3 of 31
156 Chapter Fourteen

\1 \5
\2 \10
Totals 3 15.

Now multiply 3 by 4.
1 3
2 6
\4 \12
Totals 4 12.

The answer is 12.

PROBLEM 27
Assume the false answer 5. Then 1 5 of 5 is 6. Then as many times as
6 must be multiplied to give 21, just so many times must 5 be multiplied
to give the correct number. Then divide 6 into 21.
\1 \6
\2 \12
\2 \3
Totals 3 2 21.

Now multiply 3 2 by 5.

\1 \3 2
2 7
\4 \14
Totals 5 17 2.

The answer is 17 2.

SIMILAR PROBLEMS FROM OTHER PAPYRI

The following problem (KP LV, 3) is one of a number of mathemat-
ical fragments found at Kahun in 1889. It was restored and translated
by F. L. Griffith.*
* F. L. Griffith, ed., The Petrie Papyri : Hieratic Papyri from Kahun and Gurob
(2 vols.), London, 1897.
 Equations of the First and Second Degree 157

A half and a quarter are taken away, and 5 remains.

What number says this?
Make thou the remainder after 4 has been taken from 1.
Result 4.
The remainder is 4, if the number was 1.
Then the remainder is 4 x 4 = 1, if the number was 4 x 1 = 4.
And the remainder is 5 x 1 = 5, if the number was 5 x 4 = 20.
Therefore, the number that said it was 20.
Though this is a very simple piece of arithmetic, the scribe's
reasoning allows us to consider it an example of false assumption with
the number assumed being one.
Problem 19 of the Moscow Mathematical Papyrus is as follows. It
is not solved by the method of false assumption.
Method of calculating a heap.
1 2 times together with 4,
it has become 10.
What is this heap?
Compute thou the excess of these 10 over these 4,
it is 6.
Calculate thou with 1 2 until you find 1.
Result 9.
Reckon thou 3 of these 6.
Result 4.
Lo ! It is 4. You have correctly found it.'
If we express the scribe's reasoning in modern notation, using x to
represent the heap, we have
1 2x+4= 10
12x= 10 - 4
= 6;
x=6-1Q
=6xI
= 4.

* W. W. Struve, the translator, has "Du hast richtig gefunden," the MMP
scribe's equivalent of the RMP, "Do it thus" or "The doing as it occurs,"
or even perhaps Euclid's "Quod Brat faciendum," Q.E.F.
158 Chapter Fourteen

In view of the abundant evidence that the Egyptian scribes knew as

a commonplace statement that the reciprocal of 1 2 was 3, it is quite
surprising to find the scribe of the MMP writing, "Calculate thou
with 1 2 until you find 1."
For each of the Problems 24-27, the scribe supplies a proof that the
answer he gives is correct. For Problem 24 it was:

Prove that 1 7 of 16 2 $ is 19.

\1
7
\ 16
2
2
(4
$
28) 14 56
(2 - 7 = 4 28, Recto RMP)
2 4 (14 28 56)

\7 \=2 4 A. (Pr. 12)

Totals 1 7 18 2 4(8 8)
1 7 19.

THE SECOND AND THIRD GROUPS

SECOND GROUP
Two problems constitute the second group. They are:

PROBLEM 28
A quantity and its I are added together, and from the sum a third of
the sum is subtracted, and 10 remains. What is the quantity?

PROBLEM 29
A quantity and its 3 are added together, 9 of this is added, then 3 of
this sum is taken, and the result is 10. What is the quantity?
Both of these problems are discussed in Chapter 16, along with
other "think of a number" problems. These problems are not "aha"
(quantity) problems.

THIRD GROUP
The third group consists of Problems 30-34:
PROBLEM 30
If the scribe says, "What is the quantity of which (S 10) will make
10," let him hear!
 Equations of the First and Second Degree 159

PROBLEM 31
A quantity, its 5, its 2, and its 7 added becomes 33. What is the
quantity?
PROBLEM 32
A quantity, its 3, and its 4 added becomes 2. What is the quantity?
PROBLEM 33
A quantity, its 3, its 2, and its 7 added becomes 37. What is the
quantity?
PROBLEM 34
A quantity, its 2, and its 4 added becomes 10. What is the quantity?
The scribe solved these problems by a different method, that of
division. Again, we could suppose that these problems might appear
in a modern algebra book, were it not for the choice of the numbers
33, 2, and 37, which lead to enormous unit fractions. So awkward is
the mechanical work involved that one is likely to lose sight of the
method adopted.
The solutions given by the scribe are:

Pr. 31. 14 4 56 97 194 388 679 776.

(In modern notation,142%7)
Pr. 32. 1 fi 12 114 228. (1 s/19)
Pr. 33. 16 56 679 776. (16%7)
Pr.34. 5 2 7 Ti. (5%)

A glance at the answers-all of which are quite correct-convinces

us that these problems could have had no practical applications. They
were meant to illustrate one method for the solution of simple equa-
tions of this type, and although they did this, the simplicity of the
method has been masked by the complexities of the unit fractions that
arise in the process and by the unexpected operations to which the
scribe was forced to resort. Thus in one part of the proof of Problem 33,
the scribe found that he had to add up 16 fractions, of which the last
half-dozen are 1164 1358 1552 4074 4753 5432. This is a formid-
able task in anyone's language or notation. And this the scribe has
160 Chapter Fourteen

apparently done in his head. In this same problem, the scribe needed
to find 3 of the unit fraction 679, which he wrote at once as 1358 4074.
This is a remarkable example of the rule given later in the RMP
(Problem 61B) referred to in Chapter 4. In Problems 30 to 34 of the
RMP the scribe was floundering in a maze of huge fractions. He had
chosen his numbers badly, but having started, he continued manfully,
and let us remember, successfully, to correct solutions. If the author
of one of the earliest mathematical textbooks ever written faltered
anywhere, then it was here that he did so. Had he chosen the numbers
97, 19, 194, and 14, in that order, his purpose would have been equally
well served, and the mechanical work much simplified. To show how
the scribe solved these four problems, we select Problem 34 as the
least complicated of them.
PROBLEM 34
A quantity, its 2 and its 4 added becomes 10. What is the quantity?
Multiply 1 2 4 so as to get 10 (or, divide 10 by 1 2 4).
\ 1 \1 2 4
2 3 2

\
4
7 \ 7

(Since 4x 124=7,7x 1 24=4)

4

4 28 2 (2x7=2-7
\2 14 \ 1. = 4 28)
Totals 5 2 7 14 9 2 (4 4)
9 (2 2)
5 2 7 14 10.
The answer is 5 2 7 T. The proof follows:
Find 1 2 4 of 5 2 7 14.
\1 \5 2 7 14
\2 \2 2 4 14 28
\4 \1 4 $ 28 56.

Totals 1 2 4 9 (2 8) 7 14 14 28 28 56.
The scribe refers the last 6 fractions to 56, using red auxiliaries, here
rendered in boldface:
 Equations of the First and Second Degree 161

7 14 14 28 28 56
8 4 4 2 2 1 total 21;
4
14 7 total 21; therefore they are equal.
Then 9 (2 $) 4 $ = 10.
EQUATIONS OF THE SECOND DEGREE
Two problems in the Berlin Papyrus (Figure 14.1), restored and
translated by Schack-Schackenburg,' appear to deal clearly with the
solution of simultaneous equations, one being of the second degree.
The papyrus is mutilated, so that the restorations, although quite
reasonable and plausible, perhaps still remain open to some slight
reinterpretation. In essence, Schack-Schackenburg concludes that
the scribe proposed to solve the following two sets of equations,
expressed in modern algebraic notation:
xs + ys = 100
(1)
4x - 3y = 0

x2 + y2 = 400
(2)
4x - 3y = 0.

The translator's rendering of the first problem (Equations 1) is as

follows :
If it is said to thee ... the area of a square of 100 [square cubits) is
equal to that of two smaller squares.
The side of one is 2 4 the side of the other. Let me know the sides of
the two unknown squares.
Always take a square of side 1. Then the side of the other is 2 4.
Multiply this with 2 4. It gives 2 16, the area of the small square.
Then together, these two squares have an area of 1 2 16.
Take the square root of 1 2 16. It is 1 4.
Take the square root of this 100 cubits. It is 10.
Divide this 10 by this 1 4. It gives 8, the side of one square.
The remainder is very much damaged, but what does remain leads
Schack-Schackenburg to restore as
Take 2 4 of these 8. It gives 6, the side of the other square.
* H. Schack-Schackenburg, Der Berliner Papyrus 6619, Zeitschrif? flit,
Agyptische Sprache, Vol. 38 (1900), pp. 135-140 and Vol. 40 (1902), p. 65f.
162 Chapter Fourteen

FiouRE 14.1
Berlin Papyrus 6619.

In the papyrus, the squaring of 2 4 is not shown by the scribe, and

neither are the extractions of the square roots of 1 2 16 and 100. If
these were not done mentally, they were no doubt read off or checked
from a table of squares (see Chapter 21). Neither is the working
shown for 2 4 of 8 = 6, and we may justly conclude that the scribe
was more concerned to show how his method of false position or false
assumption was applied to the solution of equations than to teach
the arithmetic of multiplication of fractions and the squaring and
square roots of simple fractions.
In the second problem (Equations 2), the amount of restoration is
considerable, but as the problem is closely allied to the preceding one
the difficulties are lessened. The ratio of the sides of the two smaller
squares is given as 2 is to 1 2, and so these are the sides the scribe
assumed instead of, as in the first problem, "Always take a square of
side 1." Then he gave the sum of the areas of these two squares as
4 + 2 4 = 6 4, the square root of which is given as 2 2 (no working).
The square root of 400 being 20, the scribe divided this 20 by 2 2
giving 8, which on multiplication by 2 and 1 2 gave 16 and 12 as the
sides of the two smaller squares.
KAHUN LV, 4
I will now provide a suggested restoration of Problem LV, 4, of the
Kahun papyrus, of which the translator F. L. Griffith wrote,
I do not know how to complete this problem, nor what is meant by
hayt.... The word for square root is new and interesting.*
0 The Petrie Papyri.
 Equations of the First and Second Degree 163

My analysis suggests this is a problem of simultaneous equations, one

of the second degree, of the general form*
xy = A...... (i)
x=ky...... (ii).
1.31 (Title) ....................................
1.32 ..........................................
1.33 ..... of the hinu ...........................
1. 34 Make thou that 40, 3 times.
1. 35 The result thereof is 120. Make thou,
1. 36 10 of 120. The result thereof is 12.
1.37 Make thou that 2 4 to find 1.
1. 38 The result thereof is 1 3 times. Make thou,
1. 39 that 12, 1 9 times, the result thereof is 16.
1. 40 Make thou a corner (square root) as 4. Make thou
1. 41 2 4 of 4. The result thereof is 3.
1.42 The result thereof is hayt (?) 10 of 4 cubits: 3.
Note first ofall the following grain measures, as used, e.g., in RMP 41,
42:
a. 10 hinus = 1 hekat.
b. 20 hekats = 1 khar.
c. 1 2 khar = 1 cubic cubit.
Hence,
d. 300 hinus = 1 cubic cubit.
e. 200 hinus = I khar.
f. 30 hekats = 1 cubic cubit.

SUGGESTED RESTORATION OF MISSING LINES OF KAHUN LV, 4,

AND MODERNIZATION OF OTHERS
1. 31 (Rectangular Granary). The side is 2 4 of the front. 40 baskets
each of 90 hinu,t
1. 32 are poured in, to a depth of 1 cubit. Find the side and the front.

$ After simplifying the various units, A = 12 and k = 2 4.

t A 90-hinu basket would hold slightly more than (1 8), a bushel of grain,
a reasonable load for a laborer to carry.
164 Chapter Fourteen

FIGURE 14.2
Rectangular granary of KP LV, 4.
1. 33 Make thou 30 to find 90 "of the hinu." The result is 3.
1.34 Make thou that 40, 3 times.
1.35 The result is 120. Make thou,
1.36 TO- of 120. The result is 12.
1.37 Divide 1 by 2 4.
1. 38 The result is 1 9.
1. 39 Multiply 12 by 1 S. The result is 16.
1. 40 Find the square root of 16. It is 4, the front.
1.41 2 4 of 4 is 3, the side.
1.42 The result thereof is (hayt(?) 10 of) 4 cubits: 3 cubits.
In modem notation we can write:
The volume of grain is f x s x d = 90 x 40 hinu, (1. 31)

or f x s x 1= 90 x 40
30 x 10 cubic cubits
(f,a.1.32,
33)
=3x 10 (1.33)

120
(1 . 34 , 35)
10
Then, fxs=12....... (i), (1.36)
and s = 4 f,...... (ii). (1.31)

Substituting for s in (i), fx4f=12, (1.37)

or f2 = 12 x 4 (1.38)

so that f2= 16. (1.39)

Take the square root, f = 4 cubits; (1.40)
hence, s=4f ((i) or 1.31)
 Equations of the First and Second Degree 165

=4x4; (1.41)
or, s = 3 cubits. (1.41)
Then the front of the granary is 4 cubits, and the side 3 cubits (1. 42).
As a check we note that
90 x 40 = 3600 hinu
3600
= cubic cubits
= 12 cubic cubits
=4x3x1,
and since volume = f x s x d, the depth of grain in the granary is
I cubit (Figure 14.2).
The approximate modern equivalents of Egyptian dry measure are:
1 hinu = 4/g pint.
1 hekat = 1 gallon = 1/8 bushel.
1 khar = 21/, bushels.
I cubic cubit = 3% bushels.
According to Gardiner's dictionary, the hayt was the principal mul-
tiple of the cubit and was equal to 100 cubits. A partial explanation
for "hayt 10 of" (line 42) that I can suggest is that, in the restored line
32, the depth of the grain might have been expressed as 100 of a
hayt, which would have been 1 cubit. But this still would not wholly
explain the " 10 of." There is no suggestion of a hayt in any other part
of the problem; and just why the scribe introduced it (if indeed he
did) in line 32, 1 am unable to say.
 15 GEOMETRIC AND ARITHMETIC PROGRESSIONS

GEOMETRIC PROGRESSIONS: PROBLEM 79 OF THE

RHIND MATHEMATICAL PAPYRUS
Because of the Egyptians' method of performing all multiplications by
continued doubling, it was natural enough that they should be inter-
ested in numbers arranged serially, and especially the series 1, 2, 4, 8,
16,. . ., which so often confronted them when they multiplied integers.
This series is what we today call a geometrical progression (G.P.).
This particular G.P. whose first term is 1 and whose common mul-
tiplier (or common ratio) is 2 has a special property, which the Egyp-
tians must have been aware of and which is today made use of in the

TABLE 15.1
Multipliers written as sums of entries of the geometrical progression 1, 2, 4,
8,16,....
Resulting
Series Terms to be Added Multiplier
l
2
1 2 3
4
1 4 5
2 4 6
1 2 4 7
8
1 8 9
2 8 10
1 2 8 11

4 8 12
1 4 8 13
2 4 8 14
1 2 4 8 15
16
16 17
2 16 18
1 2 16 19
4 16 20
 Geometric and Arithmetic Progressions 167

design of modern electronic computers. This property is that every

integer can be uniquely expressed as the sum of certain terms of the
series. Thus any integral multiplier, when partitioned in this form,
can be used in Egyptian multiplication. This property is more clearly
seen from Table 15.1.
The importance of this property to the Egyptian scribes lies in the
uniqueness of the partitioning of any multiplier. For example, the
multiplier 26 can be expressed as the sum of terms of this series in one
way only, namely, 2 + 8 + 16. It is thus readily understandable that
the Egyptians' attention would quite naturally be directed to the sum
of certain terms of this and other series, and that those properties of
progressions that could be used in subsequent calculations would
interest them deeply. Let us therefore look at one such property of
G.P.'s, to which our attention is drawn by RMP 76. We look at the
series of multipliers that the scribe used in every calculation, viz.,
2, 4, 8, 16, 32, 64.... :
The sum of the first 2 terms is 6 = 2 x (1 + 1st. term) = 6,
14 = 2 x (1 + 1st. 2 terms) = 14,
30 = 2 x (1 + 1st. 3 terms) = 30,
,,
62 = 2 x (1 + 1st. 4 terms) = 62,
,,

31 126 = 2 x (1 + 1st. 5 terms) = 126,

,,

Then by inductive reasoning the scribe could have concluded that,

in any G.P. whose common multiplier (or common ratio) is the same
as the first term (in this case 2), the sum of any number of its terms is
equal to the common ratio times one more than the sum of the preced-
ing terms. Of course this generalization, on the evidence of only one
such series, would not be sufficient for any scribe, and so it would be
natural to test if the property were true of other series in which, say,
the first term is 3 and the common ratio 3, or the first term 4 and the
common ratio 4, and soon. Then for the G.P. 3,9,27,81,243,729.... :
The sum of the first 2 terms is 12 = 3 x (1 + 1st. term) = 12,
2 39 = 3 x (1 + lst. 2 terms) = 39,
4 120 = 3 x (1 + 1st. 3 terms) = 120,
5 363 = 3 x (1 + 1st. 4 terms) = 363,
6 1,092 = 3 x (1 + 1st. 5 terms) = 1,092,
168 Chapter Fifteen

Thus he would find that the property holds for 3 as well as for 2, and
very likely this would have been sufficient grounds for an Egyptian
mathematician to conclude that it was true for 4, 5, and all other
integers, perhaps even fractions.
Let us assume, however, that he wished to further test it for a G.P.
whose first term is 7 and whose common ratio is 7. The number 7 often
presents itself in Egyptian multiplication because, by regular doub-
ling, the first three multipliers are always 1, 2, 4, which add to 7.
Then we will suppose that he tried the G.P. 7, 49, 343, 2,401,
16,807, ... :

The sum of the first 2 terms is 56 = 7 x (1 + 1st. term)

= 7 x 8 = 56,
3 399 = 7 x (1 + 1st. 2 terms)
= 7 x 57 = 399,
4 2,800 = 7 x (1 + 1st. 3 terms)
= 7 x 400 = 2,800,
5 19,607 = 7 x (1 + 1st. 4 terms)
= 7 x 2,801 = 19,607,....

Thus the procedure holds also for the number 7. Now we refer to
Problem 79 of the RMP, which deals with exactly this situation. The
scribe, as was usual, gave very little explanation of what he planned
to do, and even the few words that he did give are not exactly clear
even to the most competent translators of hieratic and hieroglyphic
scripts. Nevertheless, whatever the words do signify, the meaning of
the sequence of numbers is perfectly clear to us. Problem 79 is quite

TABLE 15.2
Sums of terms of various G.P.'s; such a table may have been known to
the scribes.
G.P. whose first term
and common ratio is 2 3 4 5 6 7

Sum of 2 terms is 6 12 20 30 42 56
3 14 39 84 155 258 399
to 4 30 120 340 780 1,554 2,800
91 5 62 363 1,364 3,905 9,330 19,607
 Geometric and Arithmetic Progressions 169

short. The following is Chace's translation:

COL. 1 A HOUSE INVENTORY
1 2,801
2 5,602
4 11,204

Total 19,607.
COL. 2
Houses 7
Cats 49
Mice 343
Spelt 2,401
Hekat 16,807
Total 19,607.
Clearly column 1 is equivalent to the last step in the preceding argu-
ment, 7 x 2,801, but one would have expected to see some reference
to where the number 2,801 originated. Either that or a similar multi-
plication of 400 by 7 ought to be shown, giving 2,800, especially if the
scribe was meaning to explain and introduce this shorter method of
obtaining his answer. But I suggest this was not the scribe's intention.
There seems little doubt that all this detail was earlier known to the
Egyptian scribes, and that in Problem 79 use was being made of past
experience. It is quite possible that 2,801 had merely to be read off
from a table (such as Table 15.2) prepared long before, as (2,800 + 1).
There is a minor scribal error in column 2 of the RMP problem, where
the scribe wrote 2,301 for spelt, instead of 2,401, yet there is no mistake
in the total. This suggests the scribe of the RMP either repeated a
previous error or made one himself. Whichever it was, we conclude
that the detail of Problem 79 had already been completed before it
was set down in the form in which we read it in the RMP.
There have been some fanciful ideas suggested about this problem.
One is that here we have the origin of the Mother Goose rhyme,
As I was going to St. Ives,
I met a man with seven wives,
Each wife had seven cats, .. .
All the available evidence for this is here before us, and one is entitled
170 Chapter Fifteen

to draw whatever conclusions one wishes. It is indeed tempting to be

able to say to a child, "Here is a nursery rhyme that is nearly 4,000
years old ! " But is it really? We shall never truly know.
Returning to more mundane things, we can state Problem 79 of
the RMP succinctly as follows:
Find the sum of 5 terms of the G.P. whose first term is 7 and whose
common ratio is 7.

ARITHMETIC PROGRESSIONS: PROBLEM 40 OF THE

RHIND MATHEMATICAL PAPYRUS
In Problem 40 of the RMP, the scribe speaks of an arithmetic progres-
sion (A.P.) of 5 terms, in which the sum of the three largest terms is
seven times the sum of the two smallest terms. In modern algebra it
is usual, though not essential, to think of an A.P. as starting with the
smallest term and ending with the largest. The Egyptians reversed
this order. For the present discussion we will adopt the modern con-
vention for convenience, and we look first at the very simplest A.P. of
all, namely,
1, 2, 3, 4, 5.

We observe that 3 + 4 + 5 = 12, which is 4 times (1 + 2). If we

look for other series that have similar properties, we can find
1, 4, 7, 10, 13,

in which 7 + 10 + 13 = 30, which is 6 times (1 + 4), and

1, 14, 27, 40, 53,

where 27 + 40 + 53 = 120, which is 8 times (1 + 14). These are

easily found by simple mental arithmetic, but if one were to look for
series other than the 4 times, 6 times, and 8 times, say 3 times or 7
times, as the scribe has in RMP 40, then it is a little more difficult,
though not so much more, requiring easy fractions such as
1, 1 2, 2, 2 2, 3,

in which, 2 2 2 3 = 7 2, which is 3 times (1 + 1 2), and the

 Geometric and Arithmetic Progressions 171

common difference is 2. Then it is reasonable to suppose that after a

little trial and error, the scribe would light on the series
1, 6 2, 12, 17 2, 23,

where 12 17 2 23 = 52 2, which is 7 times (1 + 6 2), and the

common difference is 5 2. Indeed, this is the series that the scribe used
in Problem 40 of the RMP, just as in Problem 79 he used a G.P.
whose common ratio is 7, where there are again 5 terms, but the first
term is 7, not 1. This is how he framed his problem in Problem 40:
100 loaves for 5 men. 7 of the 3 men above, to the 2 men below. What
is the difference of the shares? The doing as it occurs. The share differ-
ence being 5 2.*
1 \23
1 \17 2
1 \12
1 \6 2
1 \1
Total 60.

\1 60
\9 40
[Total 100]. (because there
are 100, not 60 loaves)
Make thou the multiplication, I
23 38 3
17 2 29
12 20
6 2 10 fi
1 1 3

Totals 60 100.

The scribe has found each man's share, and not stated the new com-
* This is not the share difference being sought, but the common difference
of the A.P. the scribe suggested that one start with.
172 Chapter Fifteen

mon difference. But this is very easily found by subtraction: 1 3 from

1036is96.
One would have expected a proof here, following the usual scribal
procedure, that the 3 men above did receive 7 times the 2 men below;
thus,
38 3 29 6 20 is 87 2 (men above),
and
10 16 1 3 is 12 2 (men below).
Then,
\1 \12 2
\2 \25
\4 \50
Totals 7 87 2.

But this proof is not given in the papyrus.

We restate the problem in less archaic language.
Divide 100 loaves among 5 men, so that the shares of the 3 highest
are together 7 times the shares of the 2 lowest. What is the difference
of their shares?
There was no need to specify that the shares are in A.P., for that is
inherent in the phrase, "difference of their shares." In his solution,
the scribe at once assumed a common difference of 5 2, because he
knew that the A.P.
23, 17 2, 12, 6 2, 1

exactly fulfills the conditions of the problem, except for the 100 loaves.
So then he added up the 5 terms giving 60, which, he noted, will be-
come 100 if multiplied by 1 1. Therefore he said, "multiply each of
the five terms by 1 s," giving
38 5, 29 6, 20, 10 1

which, as a check, he showed to add up to 100. This is his answer.

The problem asked for the difference of the shares, and presumably
the scribe supposed that whoever wishes to know will subtract any
two adjacent terms and find it to be 9 6. But he was more interested
 Geometric and Arithmetic Progressions 173

in how the 100 loaves were divided among the five men, and clearly
this is what he really meant to ask in the problem. For otherwise, if it
were only the common difference he sought, then he could have found
this at once by multiplying the original 5 2 by 1 3, giving 9 $, without
bothering to multiply out the five terms of the series by 1 .

PROBLEM 64 OF THE RHIND MATHEMATICAL PAPYRUS

Problem 64 of the RMP is:
Example of distributing differences. If it is said to thee, divide 10*
hekats of barley among 10 men, so that the difference of each man
and his neighbor in hekats of barley, 8 this is, what is each man's
share?
The average share [or " regular" share in equal distribution] is 1 hekat.
Take away 1 from 10. Remainder is 9.
Half of the difference is found, namely, 16 hekat.
Make up to times 9 [i.e., multiply by 9], there becomes, 2 16 hekat.
Add this on to the share average.
Take away the 8 hekat for each man until you come to the last.
The doing as it occurs.
1 2-F6, 1 4 8 16, 1 4 16, 1 8 16, 1 16, 2 4 8 16,
2 4 16, 2 8 16, 2 16, 4 8 16, total 10 hekats.
And now I restate the problem in less obscure language:
The sum of 10 terms of an A.P. is 10 and the common difference is $.
What are the terms of this series?
Find the average value of the terms, 10 = 10 = 1.
The number of differences is one less than the number of terms, = 10
- 1 = 9.
Find half of the common difference, 2 = 16.
Multiply 9 by 16 and you get 2 16.
Add this to the average value of the terms, = 1 2 16.
This is the highest term.
* The numbers and fractions in italic type are Horus-eye fractions used
solely for measuring grain.
174 Chapter Fifteen

Now subtract the common difference 8 nine times, until you reach the
lowest term.
Then the series is
12 6, 1 4$ 16, 1 4 16, 1$ 16, 1 16, 2 48 16,
2 4 16, 2$ 16, 2 16, 4$ 16.
The total is 10 hekats or, as we have it, simply 10.
Alternatively and equally simply, the scribe could have directed
us to find the lowest term first and then to add the common difference,
8, nine times until we found the highest term; thus,
Subtract thisfrom the average value of the terms, = 4 $ 16.
This is the lowest term.
Now add the common difference, $, nine times until you reach the
highest term.
Then the series is
4$ 16, 2 16, 2$ 16, 2416, 24816, 1 16,
1 8 6, 1 4 16, 1 4$ 16, 1 2 16.

The total is 10 hekats or, as we have it, simply 10.

The scribe preferred the first alternative, because in any series of
numbers, it was his custom to write the larger numbers first and the
smaller ones following, in descending order of magnitude, just as he
did for his fractions. We now follow the scribe's directions word for
word, but we substitute for the numbers he used those letters com-
monly used in a modern algebraic treatment of arithmetic progres-
sions, thus,
a = 1st term (and lowest).
I = last term (and highest)
d = the common difference.
n = the number of terms.
S = the sum of n terms.
The scribe directs,
Find the average value of the n terms, = Sl n.
 Geometric and Arithmetic Progressions 175

The number of differences is one less than the number of terms, _

(n - 1).
Find half of the common difference, = d/2.
Multiply this (n - 1) by d/2 and you get (n - 1) x d/2.
Then either, or,
Add this to the average value, Subtract this from the average value,
d d
n+(n-1)2 n - (n - 1)
This is the highest term 1. This is the lowest term a.
Then, n+(n-1)2=1, Then,n-(n-1)2=a,
or,n=1-(n-1)2 or,n=a+(n-1)2
=2[21-(n-1)d]; =2 [2a+(n-1)d];
hence, S = 2 [21 - (n - 1)d]. hence, S = 2 [2a + (n - 1)d].
How naturally these formulas were deduced ! Logically and simply
from the scribe's directions ! All we have done is to adapt a few ele-
mentary algebraic transformations, and we find we have not one, but
two formulas for the sum of n terms of an A.P. The first is perhaps the
less familiar. Indeed, of the 23 algebra texts on my own study shelves,
only one* includes it; but they all have the second in exactly the same
form as here.
The conclusion is inescapable : hidden in a hieratic script of the
Middle Kingdom, expressed in words, without the assistance of any
form of algebraic notation at all, is the equivalent of a perfectly well-
known modern formula for the sum to n terms of an arithmetic
progression. In view of this, it is difficult to accept the recent pro-
nouncement of Professor Morris Kline:
The mathematics of the Egyptians (and Babylonians) is the scrawling
of children just learning how to write, as opposed to great literature.t
Scrawling indeed !
* W. E. Paterson, School Algebra, Clarendon Press, Oxford, 3rd ed. (1916),
p. 385.
t Mathematics, a Cultural Approach, p. 14.
176 Chapter Fifteen

KAHUN rv, 3
Tucked away in odd corners of the Kahun Papyri are six relatively
small items of mathematical import, which were reproduced, trans-
lated, and discussed by Griffith. They were judged to be written about
1800 B.c., and thus would be contemporary with both the RMP and
the MMP.
Only three of these so-called mathematical fragments have been
clearly explained. KP IV, 2 is a portion of the Recto of the RMP: the
odd numbers 3 to 21 divided into 2. KP IV, 3 (columns 13 and 14)
deals with the volume of a cylindrical granary, and was first satis-
factorily explained by Schack-Schackenburg in 1899. KP LV, 3 solves
the equation 2x - 4x = 5. The remaining three items, KP IV, 3
(columns 11, 12), KP XLV, 1, and KP LV, 4 have not yet been pene-
trated. Of the first-mentioned, Griffith says, "I must confess I do not
see the connexion between the two operations," meaning the columns
11, 12 of KP IV, 3, the second of which appears to be some kind of a
series. KP XLV, 1 is a column of seven quite large numbers in de-
creasing order of magnitude, "not," writes Griffith, "in any fixed
proportion, yet it seems probable that they formed part of a consider-
able mathematical calculation." KP LV, 4 is a vague problem, which
Griffith could not complete, and which Schack-Schackenburg in 1900
thought dealt with a quadratic equation. A new word for square root
occurred in this problem.
It is KP IV, 3 (columns 11 and 12) that I wish to discuss here and
to offer an explanation of its mathematical meaning. My rendering
and translation of this problem is shown in Figure 15.1.
At first, in 1966,* I thought the numbers were a solution of, or
partly a solution of, a problem that we could express in modern
nomenclature as:
In a series of 17 numbers in A.P., whose common difference is double
the lowest term, the sum of the 12 largest terms is 110. What is this
series ?
This indeed does explain both columns of KP IV, 3, though perhaps
a little laboriously.
* R. J. Gillings, "Mathematical Fragment from the Kahun Papyrus,"
Australian Journal of Science, Vol. 29, No. 5 (November 1966), p. 126.
 Geometric and Arithmetic Progressions 177

Col. it

Ce1. 1 COLa

\1 3 t2 `11o

i
2. 3 6 13 s It

4 t3 `iz3
'8 3a ` lit u
Total 3!1 ` it 6 12
`1012
`9Y6t2
\8 ii
\7 16 U
\712
`6712
FIGURE 15.1
Top, KP IV, 3; bottom, the translation.

Later, following a closer examination of RMP 64, I developed a

modification of this* that appeared to me to be more plausible. It
read:
The sum of 12 terms of an A.P. is 110, and the common difference is
2 3. What is this series?
* R. J. Gillings, "Mathematical Fragment from the Kahun Papyrus,
IV, 3," Australian Mathematics Teacher, Vol. 23 No. 3 (November 1967),
p. 61.
178 Chapter Fifteen

This explanation appeared to receive the approbation of those scholars

interested in ancient Egyptian mathematics. But it seems to me that,
as we have so few problems on progressions in the extant papyri, those
that we do have should be subjected to the very closest scrutiny. So
we take a third and closer look at KP IV, 3. Let us first of all clear the
numbers of column 12 (the right side of Figure 15.1) of fractions. This
we can do by multiplication throughout by 12. Then we obtain the
series
33, 31, 29, 27, 25, 23, 21, 19, 17, 15,
which if continued would give
13, 11, 9, 7, 5, 3, 1,
where it needs must stop. This is the scribe's familiar series of odd
numbers, and we look at it to see if it contains any propositions that
would have enabled the scribe to frame problems for solution. The
scribe would be at this stage in exactly the same position as a modern
textbook writer searching for exercises. He could have observed then
that
A The sum of 4 terms is 120,
B 6 168,
C 10 240,
D 12 264,
E 16 288.
Of course he could have made similar statements for all the possible
sums of the series, but each of the five we have chosen has the factor
12, which property he could have utilized for the framing of problems.
Thus if he chose D, he could ask,
If the sum of 12 terms of an A.P. whose common difference is 2 is 264,
what is this series?
But if this appeared to him to be perhaps a little elementary, he could
have made it somewhat more difficult by dividing the common differ-
ence and the sum by the same number, which can be 2, 3, 4, 6, or 12,
whichever he chooses, for 12 is an abundant number.* His problem
then reads, dividing by 12,
' An abundant number is one with many divisors, whose sum exceeds the
number itself.
 Geometric and Arithmetic Progressions 179

If the sum of 12 terms of an A.P. whose common difference is $ is

22, what is this series?
But perhaps even this may not be quite challenging enough for his
brighter students, so he further multiplies the common difference
and the sum by 5, and he has
If the sum of 12 terms of an A.P. whose common difference is 2 3 is
110, what is this series?
And now he has a problem of the right order of difficulty. Of course,
it might well have been that the scribe of KP IV, 3 had himself
been set this problem for solution, and that what we read 3,700 years
later is his own solution to it; for no explanation of the various steps
is written on the papyrus, just as in modern times we find on the
papers of most examination candidates. They do not have the time !
I must remark that if the scribe was indeed looking for an interesting
problem, he missed a very neat form of it:
If the sum of 10 terms of an A.P. whose common difference is 2 3
is 100, what is this series?
Indeed, we may justifiably ask, why was this not in fact the original
problem being solved? It seems to adequately fit the numbers given
in column 12. The only objection to it is column 11 (left halfof Figure
15.1), in which a check is made by the scribe. There, the total 3 9 12
is the thirteenth term in the series being considered, that is, it is too
far forward to be a check on the scribe's work in column 12. This
column 11 is our stumbling block. Had the scribe added all the prod-
ucts, he would have had
\1
\2
\\ 9
3
12
6
\4 \1 3
\8 \3 3

Totals 6 4.

This finds the tenth term 6 4, which would have fitted so beautifully
with the interpretation we are considering that we would be almost
180 Chapter Fifteen

certain that we had here the true answer. But then, how would we
explain the 110 at the head of column 12?
We conclude, therefore, that KP IV, 3 is a scribal solution to a
problem which we would express in modern terms as
The sum of 12 terms of an arithmetical progression of common
difference (9 fi) is 110. What is this series?
I now apply the directions of Problem 64 of the RMP to KP IV, 3:
Find the average value of the terms, 110 = 12 = 9 .
One less than the number of terms, 12 - 1 = 11.
Find half the common difference, 16 = 3 T.
Make up to 11 times (i.e., multiply by 11), there becomes 4 2 12.
Add this to the average value of the terms, 9 9 4 2 12 = 13 3 T.
This is the highest term.
Subtract the common difference, 11 times until you reach the
lowest term.
The doing as it occurs.
13 12, 121(;12, 1212, 11fi12, 10312, 996 12,
8 1 12, 7 3 fi 12, 7 12, 6 fi 12, total 100.
However, the scribe of KP IV, 3 made only 9 of the 11 subtractions.
Why did he not complete the subtractions? How can we ever know?
But we may surmise that he was checking the progression totals, and
when he reached 100, he thought he had finished at 110. Or he may
just have got tired of the interminable subtractions. But note that his
check multiplication (column 11 in Figure 15.1) totals to the thir-
teenth term. Including a check mark at 2, a further addition gives
3 3 12 + 3 fi = 4 2 12, the twelfth term; and a further check mark
at 4 gives 8 + 4 + 1= 3 9 1 3 9 12 = 5 9 12, the eleventh term.
These were the two terms he omitted. Had he added all four lines, he
would have had 8 + 4 + 2 + 1 = 6 fi 12 (the tenth term), which he
already had included. All these additions the scribe could quite easily
have checked mentally, and no doubt he did this.
I am sure we have found here the true explanation of KP IV, 3,
columns 11 and 12, and we find it closely allied to Problem 64 of the
RMP, where we have already found hidden the formulas for the sum
of n terms of an arithmetic progression (p. 175).
 16 "THINK OF A NUMBER" PROBLEMS

Problems 24-27 of the RMP have been discussed in Chapter 14. The
two following problems, 28 and 29, were included by Chace, with
other aha or "quantity" problems. However, I consider these two to be
the very earliest examples of think of a number problems on record. As
long ago as the third century A.D., Diophantus of Alexandria in his
Arithmetica proposed a class of problems called "find a number" prob-
lems, which, however, were mostly concerned with indeterminate
equations. In Charles Hutton's* translation of Montucla's (1725-1799)
edition of Ozanam (1640-1717), we find the first problem of Chapter
10 to be:
To Tell the Number Thought of by a Person. Desire the person, who
has thought of a number, to triple it, and to take the exact half of that
triple, if it be even, or the greater half if it be odd. Then desire him
to triple that half, and ask him how many times it contains 9; for the
number thought of will contain the double of that number of nines,
and one more if it be odd.
Vera Sanfordt records that Kobel (1514) directed a person to think
of a number, add a half of it, add half the sum, divide by nine and tell
the result. The result is one fourth of the original number.
For proofs, Hutton and Kbbel chose some particular number and
then show by arithmetic, not algebra, that the number thought of has
been found, and this is exactly what the scribe of the RMP did in his
problems 28 and 29.;

PROBLEM 28 OF THE RHIND MATHEMATICAL PAPYRUS

Chace's translation of Problem 28 (see Figure 16.1) is:

* Charles Hutton, Recreations in Mathematics and Natural Philosophy, London,

1840.
t Vera Sanford, A Short History of Mathematics, Harrap, London (1930), p.
225.
t R. J. Gillings, "Think of a Number: Problems 28, 29 of the RMP,"
Mathematics Teacher, Vol. LIV, No. 2 (Washington, D.C., February, 1961),
pp. 97-100.
182 Chapter Sixteen

Two-thirds is to be added. One-third is to be subtracted.

There remains 10.
Make 10 of this, there becomes 1. The remainder is 9.
3 of it namely 6 is to be added. The total is 15.
9 of this is 5. Lo! 5 is that which goes out, and the remainder is 10.
The doing as it occurs!
Let us state this in modern terms, adding a few clarifying details:
Think of a number, and add to it its 9. From this sum take away its
9, and say what your answer is. Suppose the answer was 10.
Then take away 10 of this 10, giving 9. Then this was the number first
thought of.
Proof. If the number were 9, its 3 is 6, which added makes 15. Then S
of 15 is 5, which on subtraction leaves 10. That is how you do it!
Just as Hutton and Kobel showed how the number originally thought

.A VAX

`/!7i1 ti3
Q
f
*46
:
rL1-
Or
- >n 1s

s is to 1, aaa.a s is tote =Wad" 1o etv.i s

Mat ii of es lUsse wes+wes I tssssia4v is,
14i!ssss'sisbh.j"N%t Beds 1gNsi.a
tai s is Vhst Vast sst Ths eeasl04s is to so i sat u.a s !
the 4hy an it aaoss a3 so ma 7,1
FIGURE 16.1
Left, Problem 28 of the RMP, and right, Problem 29 of the RMP, as the
scribe wrote them, with translations. Here the lines have been opened up
to accommodate the rectangles, which enclose material drawn in red by
the scribe.
 "Think of a Number" Problems 183

of is found by the exhibition of a specific case, so the scribe of Problem

28 explained how he did it by means of a simple example, in this case
using the number 9. Because of the Egyptians' predilection for the 3
fraction and the ease with which they could find 3 of any number, it
was only to be expected that these numbers would be used in framing
a "think of a number" problem to illustrate the "entrance into the
knowledge of all existing things and all obscure secrets."* The scribe
could not have chosen a number simpler than 9 to show how his
"magic" worked. But to convince his audience, he would have fol-
lowed with another example. Let us say the number thought of was
54. Then I of 54 is 36, which being added is 90. One third of this is
30, which being subtracted leaves 60, which was the answer told to
the scribe. All he needed to do now was to subtract from 60 its tenth
part, 6, and he at once knew the number thought of was 54.
I must add that neither Peet nor Chace gave this interpretation in
their translation and commentaries on the RMP. Peet, for example,
suggested that the last line, "the doing as it occurs"t or "do it thus"+
really belongs to the next problem, and Chace said, "the solution does
not seem to be complete." They both regarded it as of the same class as
the four preceding problems of the RMP. In no other problem of the
RMP do the words "do it thus" occur at the end instead of the begin-
ning of a problem, as both Peet and Chace have pointed out. This I
submit gives us the clue to the real intention of the scribe. Having
disclosed his "obscure secret," he concluded with-in modern phra-
seology-" And that is how you do it."

PROBLEM 29 OF THE RHIND MATHEMATICAL PAPYRUS

Problems 28 and 29 stand side by side in the papyrus, and Problem
29 (Figure 16.1) is clearly of exactly the same type as 28. For this
reason, the explanatory detail of 29 is the same as that of 28, and so
the scribe did not bother to repeat it. If we restore this detail,§
following 28 as closely as possible, we have

* From the scribe's introduction to the RMP.

t Chace's translation in Vol. 2 of The Rhind Mathematical Papyrus.
+ Chace's translation in Vol. 1 of The Rhind Mathematical Papyrus.
§ As Peet (1923) and Chace (1927) also did, but not in quite the same words.
184 Chapter Sixteen

Think of a number, and add to it its 3. To this sum add its 3. Find 3
of this result, and say what your answer is. Suppose the answer was 10.
Then add 4 and 10 to this 10, giving 13 2. Then this was the number
first thought of.
Proof. If the number were 13 2, its 3 is 9, which added makes 22 2.
Then 3 of 22 2 is 7 2, which added makes 30. Then 3 of this 30 is 10.
That is how you do it!
In the papyrus the scribe started with his student's answer which he
said was 10, adding to it its 4 and 10.
1 \10
4 \2 2
\10 \1
Total, 13 2. (This is the number thought of.)
Next he proved that he correctly divined it.
\13 2
\9
Total, \22
\7
Total, 30
9 20
\9 \10. (This is the answer he was given.)
The scribe did not put any check marks in either of his calculations for
these two problems. We note again* his formality in finding one-third
of a number. For 3 of 22 2, he writes at once, 7 2, yet for 3 of 30, he
first finds 9 of 30 to be 20, then halves the 20 for 10, which is 3 of 30.
This then is the second of the " think of a number" problems of the
RMP. No other problems of this type occur in any other mathematical
papyrus known to me.

* See Chapter 4 on the Egyptian I Tables.

 17 PYRAMIDS AND TRUNCATED PYRAMIDS

THE SEKED OF A PYRAMID

We ask ourselves what the extant papyri and ancient records tell us
about the ancient Egyptians' knowledge of the geometry and mathe-
matical properties of that most famous geometrical figure of antiquity,
the right pyramid. The answer is that they tell us very little. All we
know with any certainty is that the scribes knew how to calculate:
1. The seked (slope of the sides) of a pyramid.
2. The volume of a truncated pyramid (or a frustum).
3. The volume of a pyramid.
We know of (3) only because we are certain of (2) from MMP 14.
The only other references to calculations on pyramids are Problems
56, 57, 58, 59, and 60 of the RMP. All of these are very simple and
deal with the sekeds of various right pyramids. The following is a
translation of the text of Problem 56, which is accompanied by the
drawing shown in Figure 17.1:
Example of reckoning a pyramid.
Height 250, base 360 cubits.*
What is its seked?
Find 2 of 360, 180.
Divide 180 by 250, 2 5 50 cubit.
Now a cubit is 7 palms.
Then multiply 7 by 2 3 50.
1 7
\3 2
\1
\50 \ 3
10
15
25

Totals 2 3 50 5 25 palms. This is its seked.

* Converting to feet, we find that this problem speaks of a pyramid of
height 429 feet and base 618 feet. The actual measurements of the Giza
pyramids are : Cheops-481 feet (height), 756 feet (base) ; Chephren-471
and 708 feet; and Mycerinus-218 and 356 feet.
186 Chapter Seventeen

i ioui 17.1
The scribe's drawing of a pyramid accompanying Problem 56 of the RMP.

This means that the slope of the triangular faces of this pyramid is
5 25 palms horizontally for every rise of one cubit in height.
The workers building a pyramid needed to preserve their directions
very carefully in order to obtain the same seked for each subsequent
block of stone, and this may be one reason why the orientation of the
pyramids was so accurately north-south and east-west.
We note that the scribe did not show the division of 180 by 250,
which is simple enough. We note also that in 7 x (2 5 50) no check
marks were made to show which fractions are to be added,* but again
this is no serious omission. What is not so obvious is how he found 50
of 7 to be 10 25 by finding TO of 1 5 15. No doubt on a papyritic
memo pad, or perhaps even mentally, he had
5 1 9 15
50 10 30 150
10 25. (G rulet)

Nor is it obvious how the sum (3 2 1 9 15 10 25) equaled 5 25

palms. Again, if we could have glanced at his memo pad, we would
no doubt have seen something like this:
(3 1) 2 5 (o- 15) 25
4 2 (5 fi) 25 (from his tables)
4 (2 2) 25 (from memory or tables)
(4 1) 25
5 25 palms.
We may consider the sekeds given in the pyramid problems of the
* We have included them here.
t Or perhaps from 6 30 = 5 on multiplication by 5.
 Pyramids and Truncated Pyramids 187

RMP and MMP to be the cotangents of the angle of slope of the faces
of pyramids. Then we may compute these angles and compare them
with those actually employed in the Giza pyramids. We obtain :
RMP 56 54° 14'
RMP 57 53° 8'
RMP 58 53° 8'
RMP 59 53° 8'
RMP 60 750 58'
MMP 14 80° 34'
Cheops 51 ° 52'
Chephren 52° 20'
Mycerinus 50° 47'.

The remaining problems on pyramids are similar to Problem 56;

they read as follows :
RMP 57
The seked of a pyramid is 5 palms 1 finger,* and the base is 140 cubits.
What is its height? [93 3].
RMP 58
The height of a pyramid is 93 9 cubits, and the base is 140 cubits.
What is its seked? [5 palms 1 finger].
RMP 59
The height of a pyramid is 8 cubits, and the base is 12 cubits. What is
its seked? [5 palms 1 finger].
RMP 60
A pillar (pyramid?) is 30 cubits high, and its base is 15 cubits. What
is its seked ? [1/4].

THE VOLUME OF A TRUNCATED PYRAMID

The only other problem dealing with pyramids is Problem 14 of the
MMP; this problem establishes beyond any doubt that the Egyptians
had a standard method for finding the volume of a truncated pyramid.
This would represent the very acme of Egyptian mathematical
* Four fingers (or digits) equal one palm.
188 Chapter Seventeen

achievement, except for MMP 10, which some think with justifica-
tion establishes a formula for finding the area of the curved surface of
a hemisphere. The following translation of Problem 14 of the MMP is
due primarily to Struve,* though the responsibility for its further
translation from German into English is mine (see the scribe's ac-
companying illustration in Figure 17.2).

Method of calculating a truncated pyramid.

If it is said to thee, a truncated pyramid of 6 client in height,

Of 4 ellen of the base, by 2 of the top,

Reckon thou with this 4, squaring. Result 16.

Double thou this 4. Result 8.
Reckon thou with this 2, squaring. Result 4.

Add together this 16, with this 8, and with this 4. Result 28.
Calculate thou 9 of 6. Result 2.
Calculate thou with 28 twice. Result 56.

Lo! It is 56! Thou has found rightly.

FIouRz 17.2
Left, the scribe's illustration for Problem 14 of the MMP; right, a truncated
pyramid with symbols replacing particular values.

* W. W. Struve, "Mathematisch Papyrus des Museums in Moskau,"

Quellen end Studien zur Geschichk der Mathematik, Series A, Vol. 1 (Berlin,
1930), p. 135.
t Cubits.
 Pyramids and Truncated Pyramids 189

Following the scribe's directions, we may write

volume = (4 x 4+ 2 x 4+ 2 x 2)
= (16 + 8 + 4) x 9 x 6
= 28 x 2
= 56 cubic cubits.
We now replace the particular value of the base 4 by a, the top 2
by b, and the height 6 by h. Then we have
V= (ax a + b x a + b x b) x 3 x h

= 3 (a2 + ab + b2).

This is the standard formula for the frustum of a pyramid (see Figure
17.2). But how did the scribes arrive at this?
It has been generally accepted that the Egyptians knew of a method
for the volume of a square pyramid, and that it was probably the
correct one, V = 1/3ha2 ; but this is nowhere specifically attested, to my
knowledge. Problem 14 of the MMP is strong evidence that the
Egyptians knew this formula or some equivalent; but it still has not
been easy to establish how, even with this powerful tool, they were
able to deduce (in a most compact and far from obvious way) the
formula for the frustum, which, in the words of Gunn and Peet, "has
not been improved on in 4,000 years." This is the question to which we now
address ourselves.
It would of course have been a simple operation to construct a
hollow pyramid and a hollow rectangular box of the same base and
height, to determine that the pyramid had a capacity exactly one-
third of the box by simply pouring sand or water. That the Egyptians
understood the volume of a rectangular solid to be I x b x d is well
attested,' so that the volume of an equivalent pyramid would be
expressed as one-third of the area of the base times the height, or one-
third of the height times the base. In like manner, the two figures
could have been made solid of sun-dried Nile River clay and then
weighed in the usual Egyptian way. Not so simple is the method of
dissection, in which a pyramid is cut up and the parts reformed into a
0 See, e.g., the Reisner Papyri, and Problems 44, 45, and 46 of the RMP.
190 Chapter Seventeen

rectangular solid whose volume can be easily calculated. None of the

dissections I have seen are simple or convincing, but I suggest the
following would be within a scribe's capabilities. A right pyramid is
constructed of clay or wood, whose perpendicular height is exactly
half the side of the square base. This pyramid is then cut into four
equal oblique pyramids by two planes passing through the vertex and
the midpoints of opposite base lines (Figure 17.3). Then three of these
four oblique pyramids fit together to form a cube whose sides are half
the base of the pyramid. Therefore in volumes the cube is 3% the
pyramid, or the pyramid is 4/3 the cube. Then the volume of the pyra-
mid is found to be V = 1/3haa.
Another method is to make six congruent juel* pyramids, that is to
say, right pyramids whose heights are half of the sides of the square
bases. These would have their triangular sides sloping at 45°, so that
an Egyptian would declare their sekeds to be seven palms. These can
be put together with their vertices coincident and with their six bases

riovxx 17.3
Dissection of a pyramid whose height is half its base.

* I do not know the origin of this word.

 Pyramids and Truncated Pyramids 191

FIGURE 17.4
Six fuel pyramids fitted together to form a cube.

forming the six faces of a cube (Figure 17.4). An inspection of the

dissection immediately yields the correct formula for the volumes of the
component pyramids.
Early attempts to ascertain how the Egyptians could have estab-
lished the equivalent of the formula for finding the volume of a trun-
cated pyramid were made by Gunn and Peet (with Engelbach),* by
Kurt Vogel,t by P. Luckey,-' by W. R. Thomas,§ and more recently
by Van der Waerden.11 In 1964 the dissectionist H. Lindgren of
Canberra, Australia, communicated to me a method by which three
truncated juel pyramids could be dissected into three rectangular
prisms, (a x a), (a x b), (b x b), each of thickness h, which would
establish the formula for this particular case. The volume of the trun-
cated pyramid is the volume of the whole pyramid minus the small
pyramid cut from the top. Here k = h + 1, and with the dimensions
of MMP 14, h = 1. Then, from Figure 17.5,
volume of frustum = Sa2k - 3b21
= 5[a2(h + 1) - b21]
= 3[a2h + a21 - b21]
= 3h[a2 + (a2 - b2)].
In MMP 14, a = 2b, so that the scribe would need to cut from the
* Journal of Egyptian Archaeology, Vol. 14 (1927).
t Journal of Egyptian Archaeology, Vol. 17 (1930).
+ Zeitschrift fur Mathematik and Naturwissenschaften, Vol. 41 (1930).
§ Journal of Egyptian Archaeology, Vol. 18 (1931).
fl Science Awakening (English translation by Arnold Dresden), NoordhofF,
Groningen, 1954, pp. 34, 35.
 192 Chapter Seventeen

% k

b
b

bs b bz b

FIGURE 17.5
Volume of a truncated pyramid by dissection. The top edge is half the base.

square of side a a square whose side is b, where a = 2b. Then what

remains is the rectangle ab and the square b2, so that (a2 - b2) _
(ab + b2). Then,
volume of frustum = 3h(a2 + ab + b2).
To check, in the scribal manner, on the validity of this method we
take a second case (Figure 17.6), where the base of the pyramid is three
times the base of the smaller top pyramid, so that h = 21, and a = 3b.
Then, from Figure 17.6,
volume of frustum = 3a2k - 5b21
= 5[a2(h + 1) - b21]
= 3[a2h + a21 - b211
= 3h[a2 + a22 b22]-
= 3h[a2 + 2(a2 - b2)].
 Pyramids and Truncated Pyramids 193

CL

ab b

ab b
ti

bs
MW b bV b

FIGURE 17.6
Second example of the volume of a truncated pyramid by dissection. Here
the top edge of the frustum is one-third the base, i.e., the height of the
frustum is two-thirds that of the complete pyramid.

To evaluate 2(a2 - b2), which is half the difference of the top and
bottom squares of the truncated pyramid, the scribe would have
needed to cut from the square of side a a square of side b, where a =
3b. There remain the two rectangles, each ab in area, and the two
squares, b2 in area. Thus we have a2 - bs = 2ab + 2b2, and, as before,
volume of frustum = Sh[as + ab + ba].
In like manner, the scribe could have continued this cutting of
smaller pyramids from the top of the larger pyramid to give a = 3b,
a = 4b, a = 5b, as far as he wished, or even introduced fractions, so
that, for example, a = 1 2b, or b = 3a, and thus concluded that his
method held in all cases.*

* See Appendix 1.
 18 THE AREA OF A SEMICYLINDER AND THE
AREA OF A HEMISPHERE

This chapter is based on my article of the same title* in which the

views and opinions of Struve, Peet, Neugebauer, and Van der Waer-
den were considered in attempting to arrive at the conclusion that
MMP 10 deals with the area of the curved surface of a hemisphere
(and consequently of a sphere) rather than the area of the curved
surface of a semicylinder (and consequently of a cylinder).
Struve, the original translator, certainly thought that the scribe's
calculations referred to a hemisphere, but Peet thought it more likely
a semicylinder was meant; Neugebauer and Van der Waerden in-
clined toward this latter view. Since my article was published, Van
der Waerden has said that he now agrees with Struve and with the
arguments I adduced in his favor,t but Neugebauer still feels that the
text is not paleographically certain enough to be definitive,+ so that
a further and closer examination of the original papyrus is desirable.
Struve died in 1965 and Peet in 1934.
After trying for more than three years I at last received from Mos-
cow a clear photograph of Problem 10 of the MMP from the original
papyrus in the Museum of Fine Arts, a copy of which I sent to T. G. H.
James, Assistant Keeper of the Department of Egyptian Antiquities
of the British Museum, London. A recognized authority on Middle
Kingdom hieratic, Mr. James had earlier agreed to lend his services,
and in November 1970 he wrote me as follows:
The difficulties reside in the first six lines and are specifically the
meaning of nbt, the meaning of the words at the end of line 2 and the
beginning of line 3, the meaning of `d and the restoration of the word
in line 6. I much prefer to take nbt as "basket" and consequently to
consider the problem as dealing with the area of a hemisphere, but
the difficulties pointed out by Peet for lines 2 and 3 remain. In addition
I am not sure that the word tp-r means "mouth" or diameter. But I
* In the Australian Journal of Science, Vol. 30, No. 4 (October, 1967), pp.
113-116.
t Letter from Van der Waerden, November 11, 1967.
+ Letter from Neugebauer, December, 8, 1967.
 Area of Semicylinder and Hemisphere 195

feel in spite of all this that your general interpretation must be along
the right lines.
Struve's translation (1930) of Problem 10 of the MMP was published
in German;* the following translation into English is mine.
1 Method of calculating a basket.
2 If it is said to thee, a basket with an opening (mundung),
3 of 4 2 in its containing, oh!
4 Let me know its surface.
5 Calculate thou 9 of 9, because the basket
6 is the half of an egg. There results 1.
7 Calculate thou the remainder as 8.
8 Calculate thou 9 of 8.
9 There results 1 18.
10 Calculate thou the remainder of these 8 left,
11 after taking away these 1 ti 18. There results 7
12 Reckon thou with 7 5, 4 2 times.
13 There results 32. Lo! This is its area.
14 You have done it correctly.
In 1931, Peet reviewed very carefully and in great detail the whole of
Struve's translation and commentary of the Moscow Papyrus.t There
he wrote :
If I could believe with Struve that No. 10 involved an approximate
determination of the curved area of a hemisphere, this judgment+
would have to be revised. But I do not! It would be very flattering to
the Egyptians, and very important for the history of mathematics, if
we could place this brilliant piece of work to their credit.
He then follows in his review with his own rendering of Struve's
translation, which is substantially the same as the one I have given,
except that his third line reads "of 4 2 in preservation," instead of
"of 4 2 in its containing." Any real significance in this difference is
* W. W. Struve, "The Moscow (Golenishchev) Mathematical Papyrus,"
Quellen and Studien zur Geschichte der Mathcmatik, Ser. A, Vol. 1 (Berlin, 1930).
t T. E. Peet, A Problem in Egyptian Geometry, Journal of Egyptian Archae-
ology, No. 17 (1931), pp. 100-106.
+ In 1923, eight years earlier, Peet had examined photographs of the MMP,
and he then wrote "It contained nothing, apart from the problem on the
truncated pyramid (No. 14), which would greatly modify our conception
of Egyptian mathematics."
196 Chapter Eighteen

not immediately apparent. Peet gives his own translations of MMP

10, offering two alternative renderings. And this he finds necessary
because of the difficulties of interpreting the scribe's handwriting, in
which he says the forms of his signs are "criminally inconsequent";
furthermore, "in some problems he was dealing with a faulty original,
or with an original which he did not understand."
Peet's first alternative assumes that the scribe was merely finding
the area of a semicircle whose diameter is 9. Now if this were the
problem (as in Problem 50 of the RMP), the answer would have been
very simply found by subtracting from 9 its one-ninth part, giving 8,
squaring this, giving 64, and then halving for a semicircle: 32, as the
scribe had. Peet himself thought this interpretation had grave dis-
advantages but would not dismiss it "out of hand" as a possibility.
However, he says it is "unthinkable" that the scribe should find the
area of a semicircle by taking % of % of the diameter (length) and
multiplying by the radius (breadth) ; and so he dismisses the semicircle,
in preference for his second alternative of the semicylinder. I have not
included Peet's version of MMP 10 assuming a semicircle, as being less
plausible than the semicylinder. Peet's second alternative introduces
a number that he supposes the scribe inadvertently omitted (4 2 of
line 2) and makes some further changes from his first alternative,
indicated by angular brackets, where he used "circle" for "cylinder."
1 Example of working out a <semicylinder>.
2 If they say to you, A <semicylinder of 4 2* in diameter>
3 by 4 2 in <height>, pray,
4 let me know its area. You are to
5 take a ninth of 9, since a <semicylinder>
6 is half of a <cylinder>; result 1.
7 Take the remainder, namely 8.
8 You are to take a ninth of 8;
9 result 3 6 18. You are to take
10 the remainder of the 8 after (subtracting)
11 the 3 fi 18; result 7 9.
12 You are to take 7 9 4 2 times;
13 result 32. See, this is its area.
14 You will find it correct.
0 In Peet's first (alternative) translation 4 2 was replaced by 9.
 Area of Semicylinder and Hemisphere 197

The replacement of "a basket" by "semicylinder" (lines 1, 2, 3)

leads to the inclusion of an extra 4 2 (line 2), which Peet explains by
supposing that in copying the scribe thought that writing 4 2 twice
in succession was unnecessary. It also means that Peet needs to intro-
duce two new terms not previously in evidence: "diameter" (line 2)
and "height" (line 3). Then the translation can be paraphrased, "the
container is a basket or vessel of cylindrical shape, cut in half, of
diameter 41/2 and height 41/2 "* (see Figure 18.1) . Now find % of 2d
(lines 5, 6, and 7). Then again, % of this = % x % x 2d (lines 8, 9,
10 and 11). Multiply this by h, = '2%ld x h = area (lines 12, 13).
Since this may be written
A = 1/2 x 25%1dh,

and because 25%, is indeed the Egyptian counterpart of the modern

ir, we have for the area of the curved surface of the semicylinder
A = 1//irdh.
Then if Peet's second alternative is right, we must come inevitably to
the conclusion that MMP 10 establishes that the ancient Egyptians
knew that the circumference of a semicircle was 1/znrd, and therefore
that the circumference of a circle was C = 7rd. This would represent
a considerable mathematical sophistication for those times and, if

FIGURE 18.1
The cylindrical container whose surface area is calculated in Problem 10
of the MMP, according to Peet.

* We are not told whether the units are cubits or palms.

198 Chapter Eighteen

true, would antedate the Greek Dinostratus by more than 1,400 years.
To my knowledge, intimation of this formula is nowhere else attested
in the Egyptian mathematical papyri. It is therefore surprising that
Peet should write,

The conception of the area of a curved surface does not necessarily

argue a very high level of mathematical thought so long as that area
is one which, like that of the cylinder, can be directly translated into
a plane area by rolling the object along the ground.*
For should this rolling lead to the evaluation of the area of a plane
rectangular figure as 21rrd, then it would indeed show a quite high level
of mathematical thought. Peet very modestly concludes by saying,
I do not know how many mathematicians I shall convince that this
problem deals not with a hemisphere, but with a semicircle or a
semi-cylinder.
I now attempt to establish the case for the hemisphere. So far, no
one has thought to examine the arithmetical detail of MMP 10, which
the scribe probably worked out more fully on his memo pad. Reverting
then to Struve's earlier translation (p. 195), we note that the scribe
first doubled the diameter 4 2, getting 9 (line 5), which simplified the
next step of finding 9 of 9, which is I (lines 5, 6), and then subtracted
(9 - 1 ) = 8 (line 7). In lines 8 and 9, he found $ of8 to be 3 fi 18,
but did not show the working. Following normal technique, he would
then have divided 8 by 9. Thus, what the scribe did up to line 7 was
to find % of twice the diameter, and in lines 8 and 9 he found 1/6 of
this. The division of 8 by 9 would have proceeded as:
1 9
\6
3
\1 2

\18
$
\ 1

Totals 3 fi 18 8. Answer 3 S 18.

* T. E. Peet, A Problem in Egyptian Geometry, Journal of Egyptian Ar-

chaeology, Vol. 17 (1931), p. 100.
 Area of Semicylinder and Hemisphere 199

In the next three lines (10, 11, 12), the scribe subtracts this answer
from 8, giving him 7 9, which is his way of finding % of 8 to be 7 S,
because he cannot write the fraction % in his notation; but he does
know that it is the same as I - J. The subtraction 8 - (16 18) =
7 9 is not shown by the scribe, but this he may well have done mentally
or with a few brief jottings, thus:

8=7
=7 (fi fi)
=7 (9 18) ti
=7 9 (3 18). The remainder is 7

Finally, in lines 13 and 14 he states that this 7 9 multiplied by the

diameter 4 2 is 32, the answer to his problem. This multiplication
again is not shown, but would have been similar to:
\1 \4 2
\2 9
\4 \18
3
1 2

Totals 7 9 32. Answer 32.

Clearly the scribe was concerned with method rather than calculational
technique. Let us now condense these arithmetical operations into a
simpler form by using modern algebraic notation, using d for the
diameter and r for the radius of the hemisphere (see Figure 18.2).

(Line 5). Double 4 2. Double the diameter = 2d.

(Lines 6, 7). Find % of this. % x 2d, = 2 x 11/9 x d.
(Lines 8, 9, 10, 11). Find % of this. 2 x 11/11 x 11/9 x d.
(Line 13). Multiply by d. 2 x 84/81 x d2 = 2 x 11!/81 x (2r)2, or
A = 2 x 25%jr2
A = 2rrr2, where 7r = 2s%1.
This is indeed the modern formula for the curved surface of a hemi-
sphere. If this interpretation of MMP 10 is the correct one, then the
200 Chapter Eighteen

scribe who derived the formula anticipated Archimedes by 1,500

years !
Let us, however, be perfectly clear about both the semicylinder and
the hemisphere. In neither case has any proof that either Acsllnder =
1/2irdh or A11emIsphe e = 2irr2 been established by the Egyptian scribe
that is at all comparable with the clarity of the demonstrations of the
Greeks Dinostratus and Archimedes. All we can say is that, in the
specific case at hand, the mechanical operations performed are con-
sistent with those operations which would be made by someone apply-
ing these formulas even though the order and notation might be
different. Whether the scribes stumbled upon a lucky close approxi-
mation or whether their methods were the results of considered
estimations over centuries of practical applications, we cannot of
course tell. I find it difficult to accept Peet's suggestion that the area
of the curved surface of a semicylinder was meant, but the hemisphere
lends itself to plausible speculation. The conventional shapes of
Egyptian baskets that are found in murals and other Egyptian art
are as shown in Figure 18.3; these were apparently made of rushes or
reeds, papyrus, leather, skins, or perhaps even wood. And they must
have been made in fairly large numbers, and the art of the basket-
maker or weaver must have been one of some consequence in the
Egyptian economic world. When one is weaving baskets which are

Fioua 18.2
The area of a hemisphere for the second interpretation of Problem 10 of the
MMP.

FIGURE 18.3
Conventional forms of ancient Egyptian baskets, taken from murals, etc.
 Area of Semicylinder and Hemisphere 201

roughly hemispherical one requires a quantity of material for the

circular plane lid that is about half that required for the basket itself.
Since the calculation of the area of a circle was a commonplace opera-
tion to the scribes (Problem 50 of the RMP), over a period of years it
could have come to be equally commonplace that the curved area of
the hemispherical basket was double that of the circular lid.
 19 FRACTIONS OF A HEKAT

Problem 35 of the RMP reads as follows:

I have gone three times into the hekat-measure, my 9 has been added
to me, (and I return), having filled the hekat measure. What is it,
that says this?

Call 1 out of 3 3 [i.e., divide 1 by 3 3].

Totals 5 Ti 1.

PROOF

\1 \5 To
\2 \2
\
10
\3 10

Totals 3 3 1.

The scribe's answer is that, if a certain container or scoop filled 3 3

times was required to exactly fill a measure of 1 hekat capacity, then
that container must have held 5 10 hekat. In the RMP, the scribe
repeated the above in Horus-eye fractions and ro, where 5 TO hekat
is 96 ro; 96 multiplied by 3 3 equals 320 ro or 1 hekat. He checked
that 5 10 hekat is 96 ro first.

1 320
\ 10 32
5 64

Total 96.
 Fractions of a Hekat 203

In the division of 1 by 3 3, one step is omitted; we should have

\1 \3 3
\2 \6
Totals 3 10.
(If axx=b
Then \10
\3
\ 3 then, b x x = a)
(3=2 x 10.)
Totals 5 10 1.

The full proof would be somewhat as follows :

\1 3 10
2 (3 15) (10 TO) (Recto 2 = 5)
3 (10 15) 10
_ (3 ) 10 (Tables)
\2 = 2 10. (Tables)
again, 3 TO' No- 15 (Rule 61B, RMP)
= 10 (15 30)
= To- TO (Tables)
3
\3 = TO.

Totals 3 3 5 10 2 10 10
5 (To' To) 10
3) 10
2 (3 15) 10 ( R ecto 2 = 5)
2 (3 fi ) ( -
To

3 3 1.

Many of the detailed steps shown here may well have been done
mentally by the scribe or checked on his memo pad.
Problem 36 of the RMP is the same as Problem 35, except that the
container-scoop goes 3 3 5 times into the hekat-measure. A newer
204 Chapter Nineteen

technique is evident in Problem 36, in that instead of dividing 1 by

3 5 5 the scribe divides 30 by 106. Thus,
1 106
53
\26 2
\106 \1
\53 \2
\212
Totals 4 53 106 212 30.

The container therefore holds (4 53 106 212) of a hekat, and the

proof by multiplying by 3 5 5 is quite a mammoth calculation,
involving 46 fractional numbers, the largest being 1060. This multi-
plication is not any more difficult than others, it is only longer.
Repetition with Horus-eye fractions does not occur as in Problem 35.
What most interests us here is the very powerful method the scribes
had at their disposal for what might otherwise be a quite difficult
division with unit fractions. Expressed in algebraic terms, it is
equivalent to
a - b=ax - bx.
The container in Problem 37 of the RMP has to be filled 3 2 18 times
to completely fill the one-hekat measure, and therefore 3 2 T8- has to
be divided into 1. Using the technique of Problem 36, this could be
done easily by dividing 18 by 64:

Totals 4 32 18. Answer 4 32 hekat.

But the scribe did not do this. He reverted to his ordinary method and
thus had a long calculation of unit fractions on his hands. His proof
was equally long, but he stuck manfully to his task and then repeated
 Fractions of a Hekat 205

it with ro, just as he did in Problem 35-the whole entirely without

error, which was no mean achievement.
In Problem 38 of the RMP, 3 7 times the container is required to
fill the hekat-measure, so that the scribe had to divide 1 by 3 7. Had
he used his technique of Problem 35, he would have divided 7 by 22
as follows :

1 22
14
7
\fi \3 1
\11 \2
\22 1

\66
Totals Ti 22 66 7. Answer 6 11 22 66 hekat.
But again he did not use this elegant technique, which clearly he was
aware of. But he did something which is closely akin to it, in dividing
1 by 3 7. He wrote
1 3 7
22 7

in which he was saying that

ifa x x = b
then zxb=a,
and thus, because he knew that 7 x 22 = 3 7, then
22x37=7.
Then he wrote

\\ 1

22 \3 7

\\
7
Ti 4 28
66 2 14

Totals 11 22 66 1.
206 Chapter Nineteen

Here the third line comes from doubling the second, so that 2 x 22 =
11, and 2 x 7 = 4 28, from the Recto 2 _ 7. Again the fourth comes
from doubling the third, so that 2 x 11 = 6 66, from the Recto
2 _ 11, and 2 x 4 28 = 2 14. The addition of the right-hand side
column to give 1 is

2 7 4 14 28 = 4 (7 14 28)

= 1.
The value 7 14 28 = 4 is known from tables or from Problem 10
of the RMP and was an equality very well known to the scribes. The
proof that follows is also not simple, and when the whole is repeated in
Horus-eye fractions and ro, the solution to Problem 38 appears as a
rather long and involved calculation, when in fact it could have been
a simple one if the scribe had taken care to use his more efficient
methods.
 20 EGYPTIAN WEIGHTS AND MEASURES

CUBrr :
A cubit was originally the length of a forearm, from the elbow to the
tip of the middle finger. Of course, individuals' limbs varied in length;
and two standard cubits came into common use early, the Royal Cubit
and the Short Cubit. The former was the cubit usually meant in measur-
ing in everyday life and was 20.6 inches (more accurately 20.59),*
while the short cubit is reckoned to be 17.72 inches, hence the "cubit
and an hand breadth" :
Behold... there was a man ... with a line of flax in his hand, and a
measuring reed ... of six cubits long by the cubit and an hand
breadth; so he measured the breadth of the building, one reed, and
the height, one reed.
Then measured he the porch of the gate, eight cubits; and the posts
thereof, two cubits.
The foundations of the side chambers were a full reed of six great
cubits.... And these are the measures of the altar after the cubits:
the cubit is a cubit and an hand breadth. And the altar shall be twelve
cubits long, twelve broad, square in the four squares thereof. EZEKIEL
40:3, 5, 9; 41:8; 43:13, 16.
In later times the term "cubit" was still used, the Greek cubit being
18.22 inches and the Roman 17.47 inches.t Ezekiel is contemporary
with the Babylonian King Nebuchadnezzar and the Egyptian Phar-
aoh Apries (Hophra of the Bible), who reigned from 589 to 570 B.c.+
PALM:
The palm, or handbreadth, was one-seventh of a cubit, and thus 2.94
* Sir Alan Gardiner, Egyptian Grammar, Oxford University Press, London,
3rd edition, 1957, p. 199, has royal cubit = 0.523 meter. See also I. E. S.
Edwards, The Pyramids of Egypt, Pelican, London, 1952, p. 208; R. W. Sloley
in The Legacy of Egypt, S. R. K. Glanville, editor, Oxford University Press,
London, 1942, p. 176. Both give 1 cubit = 20.62 inches. I adopt Gardin-
er's value of 20.59 inches.
t Webster's New International Dictionary, 2nd edition, 1934.
+ Gardiner, Egypt of the Pharaohs, Oxford University Press, London, 1961,
p. 360; S. R. K. Glanville, editor, The Legacy of Egypt, p. 233.
208 Chapter Twenty

inches if taken from a royal cubit and 2.53 inches from a short cubit.
To the nearest tenth of an inch, the royal cubit was 2.9 inches longer
than the short cubit. Some authorities state that for the short cubit six
instead of seven palms was the equality used. This is plausible, for then
the short-cubit palm would be 2.95 inches, very close to the royal-
cubit palm of 2.94 inches.
FINGER:
A finger, sometimes called afingerbreadth or a digit, was one-quarter of
a palm or handbreadth. Thus 28 fingers equaled a cubit. It was nearly
3/4 inch, or 0.735 inch from the royal cubit and slightly less from the
short cubit.
HAYT:
The chief multiple of the cubit was the hayt (rod or cord) of 100 cubits.
REMEN:
A double-rerun was the length of the diagonal of a square whose side
was one cubit. Using the royal cubit, which was most commonly the
case, a double-remen was therefore 29.1325 inches( x 20.6), and
consequently the remen was 14.566 inches. It is thought that the double-
remen was used in measuring land, because it enabled areas to be
halved or doubled without altering their shapes.
Doubling of numbers was standard technique in Egyptian arith-
metic; so in measuring land areas the relations between the double-
remen, the cubit, and the remen enabled areas (whether squares,
rectangles, triangles, or other shapes) to be doubled and halved merely

Pools-g,Wtw. 29-lf isdta

Cow 20.6 is&dd;

3taeu ! it"
FIGURE 20.1
The double-remen, the cubit, and the remen.
 Weights and Measures 209

TABLE 20.1
Table of length and fractions of a cubit.
4 fingers = I palm
7 palms = 1 cubit
Fractions of a Cubit
Palms Cubits
1 7
2 4 18-

3 4 7 28
4 2 T4-

5 2 7 14
6 2 4 14 28

From the RMP Recto, 2 = 7.

by changing the units of measurement while preserving the propor-

tions of the figures. Thus the relative lengths in Figure 20.1 are such
that a square on the side of the double-remen is double the area of a
square on the cubit, while a square on the side of the remen is half a
square on the cubit.
From a table such as Table 20.1 the scribe could have deduced from
(1 + 6) or (2 + 5) or (3 + 4) palms equaling one cubit, the equality
2 4 7 14 28 = 1, which is what he gave in RMP 38. Further, since
2 4 4 = 1, he had at once 7 14 28 = 4, an equality he frequently
used, as for example in RMP Problems 9, 10, 11, 12, 14, 24, and
elsewhere.
ARUBA:
A unit of area not commonly met is the arura, equal to the area of a
square whose side was 100 royal cubits, and thus 10,000 square cubits.
The linear length of 100 royal cubits was called a schoenia in the time
of the Ptolemies, and this would be the length of the side of the square
of one arura. See hayt and khet. The measures quoted here are from
land measures in Greek papyri of the first century A.D.
KHET, SETAT:
The common unit for linear measures of land at the time of the RMP
was the khet, of 100 cubits. A setat was a square khet or 10,000 square
cubits. What was called a cubit-strip was a long rectangle of 100 cubits
210 Chapter Twenty

by I cubit, or one khet by one cubit, 100 of which cubit-strips would

make one setat.
HEKAT:
A hekat was a half-peck dry measure for barley, wheat, corn, and grain
generally. It was thus 1/9 bushel, or 4 quarts, or 8 pints dry measure.
Chace gives it as 292.24 cubic inches,* while half a peck in British
measure was 277.36 cubic inches, so that a hekat was slightly more than
half a peck. Eisenlohr, Sethe, and Struve call it the schefel, Gunn uses
gallon, and Peet bushel. For stating the contents of larger grain vessels,
the units might be double-hekats or more commonly quadruple-hekats,
which would therefore be 1/4 and 1/2 of a modern bushel, respectively.
For storage granaries an even larger unit was needed, and use was
made then of "100 quadruple hekat" units. One cubic cubit contains
30 hekats of grain.
HINU :
The hinu was a smaller unit for grain, being one-tenth of a hekat.
KHAR:
This was two-thirds of a cubic cubit, or 20 hekats of grain. Or we can
say 1 2 khar make a cubic cubit.
RO:
The smallest named unit for grain was the ro, which was 1/320 part of
the hekat. It would be between a dessertspoon and a tablespoon full
of grain. The only fractions of a hekat used were 2, 4, $, 16, 32, 64, and
these were written in a special way, quite unlike ordinary fractions.
They were called Horus-eye fractions, and were used solely for grain
(see Figure 20.2). Then 64 of a hekat contained 5 ro, and for any
fraction of a hekat less than 64, ro and sometimes fractions of a ro had
to be used. Horus was the son of Osiris, who was treacherously slain
by his brother Seth. In revenge, Horus sought out his uncle and slew
him, but in the fight lost an eye, the broken parts of which were later
restored by the god Thoth. Isis was the mother of Horus, and the wife
and sister of Osiris.
* A. B. Chace; L. Bull; H. P. Manning; and R. C. Archibald, The ltlund
Mathematical Papyrus, Vol. 1, Mathematical Association of America, Ober-
lin, Ohio, 1927, p. 31.
 Weights and Measures 211

32 j

o ro i
Q
11

2 ro Si

3 ro °i

FIGURE 20.2
o 4 ro
Horus-eye fractions.
212 Chapter Twenty

DEBEN,SHATY:
In Problem 62 of the RMP, the value in shaty of one deben (weight) of
gold is given as 12. Then one deben of silver is 6 shaty and one deben
of lead is 3 shaty. According to Chace, the shaty was a seal (sic), the
word representing a unit of value. It was not a coin. The deben was
a weight of about 91 grams and consequently about 3.2 ounces
avoirdupois. However Sloley* gives a deben as "the weight (1,470
grains) of the anklet of the same name, of which the tenth part was
the quedet, the weight of the finger ring." Then at 7,000 grains per
1 lb. avoirdupois this would give a deben of 3.36 ounces avoirdupois.
SEKED :
The seked of a right pyramid is the inclination of any one of the four
triangular faces to the horizontal plane of its base, and is measured
as so many horizontal units per one vertical unit rise. It is thus a mea-
sure equivalent to our modern cotangent of the angle of slope. In
general, the seked of a pyramid is a kind of fraction, given as so many
palms horizontally for each cubit vertically, where 7 palms equal one
cubit. The Egyptian word "seked" is thus related to our modern
word "gradient."
PESU :
A pesu is a unit measuring the strength of beer, bread, or cakes,
according to the amount of grain used. If one hekat of grain were used
to make 10 loaves of bread, then their pesu was said to be 10; if one
hekat made 15 loaves, then their pesu was 15. In the same way, if I
hekat of grain was used to make 5 des-jugs of beer, then the beer was
said to have a pesu of 5; if it made only 3 des-jugs, their pesu was 3.
Thus the less the pesu, the stronger the beer (or bread), and the higher
the pesu, the weaker the beer (or bread). The formula

number of loaves or des jugs

peso =
number of hekats

expresses the relation. Authorities use the following terms to explain

the meaning of pesu :

* The Legacy of Egypt, p. 176.

 Weights and Measures 213

Chace -Cooking ratio.

Vogel -Backungzahl or Baking number.
Struve -Backverhaltnis or Baking proportion.
Van der Waerden-Baking Value.
Peet -Cooking figure.
DES-JUG:
Struve says that one des-jug of beer is approximately half a liter, so
that it would therefore be about 7/8 of a pint.
 21 SQUARES AND SQUARE ROOTS

The squares of numbers, both integral and fractional, are quite often
stated and calculated in the mathematical papyri, but square roots
are far less common, and although stated, they are not calculated.
There would have been no need for the Egyptians to devise a means of
finding the square roots of perfect squares. These could have been
read off from a table of the squares of integers. Such a table would
have been very easy to construct, and indeed very probably was drawn
up by the scribes. Similar tables involving the simpler fractions could
equally well have been made by them, using ordinary Egyptian
multiplication; and although no such tables have been preserved, if
they were in fact made, they would have looked like Table 21.1.
These can be read forwards and backwards equally well, and they
would have been sufficient for all ordinary requirements. These tables
could also be used to obtain good approximations to numbers not
specifically listed. Let us suppose for instance that the square root of
40 is required. From the tables we read that the square root of 39 16
is 6 4, while the square root of 40 $ is 6 3. Then the square root of 40
lies somewhere between 6 4 and 6 3, and closer probably to 6 3, so
that this may be a sufficiently close approximation for the purpose at
hand. But if it is not, we can proceed as follows with standard multi-
plication, but starting with the lesser value, 39 16 being below 40, so
that we can add smaller fractions to 6 4, rather than subtract them
from 6 3. The value 6 4 is less than the exact answer by 0.0746 to four
decimal places in our notation. For the square of 6 4 the working
would be:
1 6 4
\2 12
\4 25
3 $
1 T6'

Totals 6 4 39 16.

Now the table tells us that 6 3 x 6 3 is too great for the square root
of 40, and therefore, since 4 12 = 3 (by application of the G rule or
 Squares and Square Roots 215

TABLE 21.1
Perfect squares as they might have been tabulated by the scribes.
No. Square No. Square No. Square No. Square

1 1 2 4 3 9 4 T6-
2 4 12 24 13 19 14 1216
3 9 27 64 23 539 24 516
4 16 32 124 33 11 34 io716
5 25 42 204 43 189 44 1816
6 36 52 304 53 2839 54 27216
7 49 62 424 63 409 64 3916
8 64 72 564 73 5339 74 52216
9 81 82 724 83 6939 84 6816
10 100 9 2 90 4 9 3 87 9 9 4 85 2 16

No. Square No. Square No. Square No. Square

24 216 3 30 5 25 6 36
1 24 316 13 233 15 1 3 1525 16 1 336
224 7216 23 79 25 43102530 26 4336
324 1416 33 1339 35 10325 36 1036
424 22216 43 2130 45 17351525 46 17336
524 3316 53 329 55 2725 56 26336
624 45216 63 4439 65 3831525 66 3836
724 6016 73 5833 75 513103025 76 51336
824 76216 83 753 85 67525 86 66336
924 9516 93 9339 95 84351525 96 8436

No. Square No. Square No. Square

7 49 $ 84 0 81
17 1 42849 i$ 1 484 10 1 618 81
27 421449 2$ 4264 29 43381
37 924142849 30 92464 30 9391-
47 17749 4$ 1784 43 16361881
57 26472849 5$ 26464 59 26381
67 3727 1449 6$ 37284 69 37381
77 5149 7$ 50 249 79 5021881
87 6642849 8$ 6664 80 653381
97 832 1449 9$ 83 484 93 83 81
216 Chapter Twenty-One

gen. (1, 3) ), the addition of 12 to 6 4 giving 6 4 12 as a closer approxi-

mation would be too great. Then the choice is to be made from the
smaller unit fractions, 13, 14, 15, 16, F7-,.. ..Here a certain judgment
is required, in which past experience would help. It is also of some
practical importance to keep the multiplication as simple as possible.
Suppose we choose 16. The multiplication is:
1 6 4 16
\2 12 2 $
\4 25 4
3 $ 32
\4 1 2 16 64
4 32 128
\ 66 4 $ 64 256

Totals 6 4 16 39 448 $ 16 64 64 256

39 2 4 16 32 256.

Then 6 4 16 is a closer approximation and may have been sufficient

for the scribe's purposes, although it is still too small by 0.0121.
In KP LV, 4, lines 39, 40, Griffith translates, "make thou a corner
(square root o f 16) as 4," where the pertinent hieroglyphs are:09.3
A corner thou make.*
In the Berlin Papyrus 6619, Schack-Schackenburg translates as,
"Nimm die Quadratwurzel daraus (1 2 16), das giebt 1 4" (Take
the square root of 1 2 16, it is 1 4), the line he displays asst
a!6 W G %_IX
UL_ CDT 40
UN-%C7

In the Moscow Papyrus, Problem 6, Struve's translation reads,

"Berechne du seinen Winkel (Quadratwurzel 16). Es entsteht 4."
(Calculate thou its angle (square root 16). Result 4.)
Peet notes in his Mathematics in Ancient Egypt that in the Berlin
Papyrus 6619 the square root of 6 4 is also correctly given by the
scribe. See the table of squares (Table 21.1) and Chapter 14.
* Read the hieroglyphs from right to left.
t Read left to right.
 Squares and Square Roots 217

R. C. Archibald* notes that in a Greek papyrus dating from the

second century (P11529, Berlin Museum), in the first of five problems
dealing with areas of fields in arurae, the square root of 164 is found
in the form 12 5 15 26 32, which is a good approximation. If 26 were
an error for 24 the approximation would indeed be remarkable.

* The Rhind Mathematical Papyrus, Vol. 1, p. 176.

 22 THE REISNER PAPYRI: THE SUPERFICIAL
CUBIT AND SCALES OF NOTATION

The Reisner Papyri were found by Dr. George Reisner in 1904 at

Nag`ed Deir in Upper Egypt, during excavations being carried out
for Harvard University and the Boston Museum of Fine Arts. The four
badly worm-eaten rolls were found lying on a wooden coffin. Dr. H.
Ibscher, the director of the papyrus collection of the Akademie der
Wissenschaften (Berlin), unrolled and restored RP 1, but the remain-
ing rolls remained in Berlin until the end of World War II. These papy-
ri are for the most part the official registers of dockyard workshops,
Nagced Deir being a necropolis for the town of This. Originally the
whole papyrus was about 111/2 feet long and about a foot high. The
editor and principal translator,* Dr. W. K. Simpson, concludes that
the papyri date from the reign of Sesostris I of the Twelfth Dynasty,
from which I conclude their approximate date to be 1880 B.c. In the
first two volumes on the Reisner Papyri,fi Simpson makes 17 divisions
as shown in Table 22.1.
We are concerned only with the sections marked G, H, and I. Section
G is an analysis of the volume calculations of the rectangular ex-
cavations planned for the temple to be built and the determination
of the number of enlistees, or workmen, required to make them.
Division by 10 suggests that 10 cubic cubits is the unit of volume ex-
pected of each enlistee per day. Section H is an analysis of the calcu-
lations for the volumes of blocks of stone from the storehouse; while
Section I, which is the "least clear of the three,"-+'- probably deals with
the volumes of rectangular walls or floors, as I 7 suggests, length 52
cubits, width 3 cubits, and thickness (or depth) 4 cubit, perhaps a wall
or pathway. No workings for the clerk's calculations are shown, but

* Earlier Dr. Alan H. Gardiner and Paul C. Smither (d. 1943 at age 29)
made some translations.
t Papyrus Reisner I (1963), II (1965), III (1969), W. K. Simpson, Boston
Museum of Fine Arts.
$ Simpson.
 The Reisner Papyri 219

TABLE 22.1
W. K. Simpson's seventeen divisions of the Reisner Papyri.
Sections A, B. Rosters of enlistees.
Section C. List of foremen and laborers.
Section D. List of enlistees going upstream.
Section E. Names of foremen and their crews.
Section F. Lists of foremen, clerks, and crews.
Section G. Calculations of temple excavations and enlistees needed.
Section H. Calculations of volumes of stone blocks from the storehouse.
Section I. Calculations, floor plans, walls, trenches, corridors.
Section J. Totals of enlistees from the above sections.
Section K. Divisions by 10 for enlistees from the above sections.
Section L. Account of cargo, hides, cattle, fish, oil, pigeons, fowls.
Section M. Various totals.
Section N. List of officials, mostly women.
Section 0. Like a balance sheet, 6 long columns of numbers.
Section P. A second roster of names, men.
Section Q. Contains many large numbers, such as 40,566, 39,548.

certainly they must have been done on some sort of papyritic writing
pad, from which the answers were transferred to the RP lists. What
would we not give to see that piece of papyrus ! I would hazard the
guess that as each block of stone was brought to the site, an official
measurer would call out the dimensions for the clerk to write down.
While waiting for the next to arrive, he would do the multiplication,
which would be entered on the appropriate line. Multiplications and
divisions would also have been done for the heading "units," where
there was more than one unit of the same size, and the heading "en-
listees," for the results of the divisions by 10.
The measuring reed used, whatever its overall length, must have
been marked in cubits, palms, and fingers, as well as in halves, thirds,
and quarters of cubits. It was probably 6 cubits long and may have
looked like the one shown in Figure 22.1.
Table 22.4 is one that could have been constructed by the scribe in
order to ease the work of dividing by 10. This table may be constructed
by ordinary division, or from the table preceding Problems 1-6 of the
RMP. The entries of Table 22.4 should be compared with the scribal
work collected in Tables 22.2 (correct calculations) and 22.3 (scribal
 N
O

EGYPTIAN ROYAL CUBIT

THE CUBIT OF KING AMENHOTEP I 1559-1539 B.C. 19th DYNASTY,
ORIGRUL IN THE LOUVRE. PtRIS

Am, TT
-fL : v J
Q n fc a 4; 3c OP- *
,(Lr A-
-ft d7 .I'Li. . .. .°n He Io a I1MpNM ° °.. ,°. .Y. °. 4. a
_r. _.... _V .-1. .r ..6... .L V- N6.- _ I°n ,
I I
lawsy-NOWO
pr01) 1.
n.r i - o yln.
M
IIn

viouRE 22.1
A possible division of an Egyptian measuring-rod. Courtesy M. J. Puttock B.Sc., National Standards Laboratory,
C.S.I.R.O., Sydney, Australia.
 The Reisner Papyri 221

errors and approximations). Thus, in line G10 (Table 22.3B) and line
17 (Table 22.2), the clerk has written the approximation 39 = 10 = 4
in his calculations for the Eastern Chapel. But he need not have made
this approximation, for a glance at Table 22.4 gives immediately
39-10=(30+9)=10
=3 2 5 i5.
TABLE 22.2
Calculations by the scribe of the Reisner Papyri which are quite correct.
Line 1. b. d. Units Volume Enlistees Detail
G5 3 2 2 1 12 15
G 6, H 32 8 5 4 1 10 1 Eastern Chapel.
G 14 8 3 5 1 8 2 4 20
G15 6 4 2 1 48 42420
G 16 4 2 2 1 16 1 2 10
G 17, H 33 4 4 2 2 64 6 4 10 20 Footings.
G 18, H 34 3 3 2 2 36 3 2 10 Footings.
H 31 15 5 4 1 18 2 4 Great Chamber.
H7 34 12 2 2 424$
H8 22 12 2 1 124$
H9 42 12 1 2 13 2
H II 3c Ip 1 1 1 3c lp
H 17 4 lc 3p 1 4 22c 6p
H25 12 24 1 2 24
H26 22 24 1 2 324
H27 32 12 12 2 !524
H 30 12 5 4 1 15 Great Chamber.
I2 12 5 2 1 30 Great Chamber.
I3 15 5 2 1 37 2 August Chamber.
I4 8 5 2 1 20 Eastern Chapel.
15 18 11 1 132
16 32 4 4 1 32 Western.
I 7, G 10 52 3 4 1 39 Eastern.
18 24 5p 2 1 8c 4p
I9 26 2 5p 1 11 Ic 3p Carrying srft.
I 10 20 5 5p I 71c 3p Carrying srft.
1 12 27 7 2 1 378 Loosening brick clay.
I 13 8 7 2 1 112 Water from a field.
1 14 12 12 2 2 9 For tower.
115 22 12 122 11 4 For tower.
120 8 6 1 1 48
222 Chapter Twenty-Two

TABLE 22.3
(A, B, C, D, E) Calculations which appear to be errors, major and minor,
with restorations.
A. Calculations by the Clerk Correct Except for Simple or Obvious
Scribal Errors.
Line 1. b. d. Units Volume Enlistees Detail
G8 35 11 2 1 1922 192for194
H35 8 5for9 4 1 18
116 32 22 12 2 25 4 for 26 4 For the tower.
117 4 22 12 2 36 for 30 For the tower.
I 18 10 52 4 for I I 55 Brick clay.

B. Errors, Misreadings, or Approximations.

Line 1. b. d. Units Volume Enlistees
G7 3 22 4 1 2 2 for 1 2 4 8 5 20 correct for 2 2.
G9 13 11 1 2 1 2142 21 2for21 45.
G 10 52 3 4 1 39 4 for 3(2 3 15).
G 11 32 4 2 1 85 for 64 8 2 correct for 85.
G12 32 2 3 1 43 To-

G13 102 82 3 1 27 for 29 2 4 2 2 5 correct for 27.

H 20 3c 3p Ic 3p I I 5c 5c for 4c 6p (4 28)p.

C. Minor Errors.
Line 1. b. d. Units Volume
H10 4c lp 1 2 2 1 (3c 2f) for (3c 3f).
H 13 2c 3p 2c 3p 3 1 (3c 6p 1 3f) for (3c 6p 2f)(14 42f).
H 14 2c 2f 1 2 Ic lp If 1 (3c 4p 12f) for (3c 4p 2f)(2 28f).
H 15 Ic 5p 1 2 5p 2 (3c 5p) for (3c 4p 2f)(2 4 14 28f).
H 18 3c 2p lc 2p 6p I (3c 3p 2 3f) for (3c 4p I Sf)(15f).
H 22 1 lc 3p I I (2c 2p 2f) for (2c 2p 2 3f).
H 24 3c 5p is 2p 6p 1 (4c 2f) for (4c 2 3f) (5 15) approx.
H 28 4c 4p lc 5p lc 2p 2 (20c lp 1 2f) for (20c lp 4f) approx.

D. Major Errors.
Line 1. b. d. Units Volume
H 16 2c 3p Ic 4p 5p 2f I (2c 5p 2 2f) for (2c 6p 3 2f) approx.
H 19 3c 5p 2f Ic 3p I I (4c 2p 3f) for (5c 2p 3 4f) (7 28).
H 21 is 5p Ic 3p 1 1 (2c 4p I ... f) for (2c 3p 2 14f).
H 23 4 1 c 6p 6p 1 (4c If) for (6c 2p 2 4f)28.
 The Reisner Papyri 223

E. Possible Restorations.
Line 1. b. d. Units Volume Restoration
H2 2c 5p 6p [ ] 2 4c 1 3f d = 6p if (probably too great).
H3 24 6p [] 1 1c2p2...f d=5p(givesv=1c2p2214f).
H424 6p [ ] [ ] 1 ... 2p 2 ... f d = 5p (units = 1, as above).
H5 24 6p [ ] 1 lc 3p 1 2f d = 5p 2f (probably too great)
H6 44 [] [] 2 624 b=6p
TABLE 22.4
Table for dividing by 10 for enlistees.
The Number Divided by 10 Alternatively

2 20
30
4 40
To
2 5
3 3 10
4 3 15 (4 10 20)
5 2
6 2 10 ( fi 10)
7 23 (' 30)
8 2 3 10 (2 4 20)
9 23 15 (3615)
10 1

20 2
30 3

My interpretation of Simpson's translation of Reisner Papyri I and

II leads me to conclude that the chief overseer of the dockyard ap-
pointed a skilled scribe to instruct the tally clerk in multiplication and
division for the records of the workshop. The scribe's teaching for
integers and the simple fractions of a cubit like 3, 2, 3, 4 was well done,
because of the 65 entries made by the clerk, there were 11 that con-
tained only integral values of cubits, and for these the clerk's arith-
metic is 100 percent accurate. As an example, we take line G 18,
footings (Table 22.2) :
length breadth depth units volume enlistees
3c 3c 2c 2 36c 3 2 10
224 Chapter Twenty-Two

The calculations not shown by the clerk must have been,

\l c 3c
\2 c 6 cc
Totals 3 c 9 cc
lc 9 cc
Totals \2 c 18 ccc
1 18 ccc
Totals \2 36 ccc volume.
36 ccc = 10: \ 1 10
\2 20
2 5
\10 1

Totals 3 2 10 36 enlistees. (See Table 22.4)

Of course, some of the simpler steps may have been done mentally,
and almost certainly, Egyptian equivalents of our abbreviations c,
cc, and ccc were not included in the calculations, although they must
have been mentally noted.
In those lines where fractions of cubits were included, only 3 errors
occurred, so that in the 31 of these calculations, the clerk was 90 per-
cent accurate. Here is an example from H 27 (Table 22.2) :
length breadth depth units volume
32c l2c 12c 2 1524
The calculations not shown by the clerk may have been,
\l c 3 2c
\2 c 1 2 4 cc
Totals 1 2c 5 4 cc.

\1 c 5 ccc
\2 c 2 2 $ ccc
Totals 1 2c 7 2 4 8 ccc.

1 7 2 4 $ ccc
Totals \2 15 2 4 ccc volume.
 The Reisner Papyri 225

TABLE 22.5
Table of fingers, palms, and cubits.
4 fingers = 1 palm
7 palms = 1 cubit
lp = 7 c
2p = 4 28 c
3p = 4 7 28 c
4p = 2 14 c
5p = 2 7 14 c
6p = 2 4 14 28 c

There are 11 entries that include measures in cubits and palms but
no fractions. Now here the clerk began to find a little difficulty, because
the palms must be expressed as fractions of a cubit, and so a table of
fractions of a cubit would need to be prepared and handy for reference.
The table would have been something like Table 22.5. See also Table
20.1.
Then let us look at line H 24, Table 22.3C, which falls in this cate-
gory:
length breadth depth units volume
3c 5p lc 2p 6p 1 4c 2f
The first multiplication here is (3 2 7 14) x (1 4 28), and we set
it down as it must have been done by the clerk.
3 2 7 14
1 2 4 14 28
4 $ 28 56-
\F8 14 28 56 1 66 392

Totals 1 4 28 c 4 2 7 9 196 392 cc.

Now we have not shown how the clerk added the thirteen fractions
(2 7 14) (2 4 9 28 56) (14 28 56 1 66 392), and there remains
further the multiplication by (2 4 14 28) for the depth 6 palms,
which will produce more than 20 fractions to be added ! Clearly, a
shorter method had to be found by the scribe and then explained to
the clerk, a good calculator but not necessarily a mathematician.
Perhaps the scribe might have thought to show how each dimension
226 Chapter Twenty-Two

could be expressed in palms, and thus the multiplication for line H 24

(Table 22.3C),26 x 9 x 6, would give the volume in cubic palms,
but the product would need to be divided by 343 in order to obtain
cubic cubits. Thus,

\1 26
2 52
4 104
\8 208
Totals 9 234.
1 234
\2 468
\4 936
Totals 6 1404 cubic palms.

343
686
1372
171 2
85 2 4
42 2 4 $
21 4 $ T6-
10 2 8 16 32

Totals 4 T6- 5-2 1404 $ 32. ($ 32 in excess)

This is inaccurate, and furthermore, the scribe still had to find some-
thing simpler for the clerk. What the scribe found I think we can locate
in line H 11 (Table 22.2) :

length breadth depth units volume

3c l p 1 1 1 3c l p
The volume, here found mentally, clearly means 3 cubic cubits and
one seventh of a cubic cubit, although it is written as 3 cubits 1 palm.
One seventh of a cubic cubit would be a flat rectangular prism, I cubit
 The Reisner Papyri 227

by 1 cubit by 1 palm, and from the analogy of the modern "superficial

foot" in measuring timber, I will call this the Egyptian "superficial
cubit." This I think is quite justified, since in every one of the RP
calculations, the volume is stated simply in cubits, palms, and fingers,
without any suggestion of square cubits or palms, nor of cubic cubits
or palms. And this simplified method of stating the volume or cubic
contents allows of a much easier way of doing the required multiplica-
tions, which I am sure the scribe invented and explained to the clerk,
although I have no direct evidence to prove it, working only with
what is to be deduced from the multiplications before us in the Reisner
Papyri.
We revert then to H 24 (Table 22.3C), where the ordinary stand-
ard Egyptian techniques left us in a maze of complicated multiplica-
tions of almost unmanageable fractions. This is what I think the
scribe devised for this kind of multiplication. I repeat line H 24
here for convenience:
length breadth depth units volume
3c 5p lc 2p 6p 1 4c 2f.
Then,
c p c p
3 5

3 5
\ 2 6 10

Totals 1 2 4 5 3.

4 5 3

\ 6 24 30 18

Totals 6 4 0 4 4.

The answer is 4 c 0 p and (4 4) ; amounts which the clerk wrote (in

error I think) as 2 f, though it is a rough approximation. What the
scribe has invented is what we call in modern textbooks scales of nota-
tion. In this case the additions were made in the scale of 7, simply
because there are 7 palms in a cubit. Thus in the first addition, since
10 = 7 + 3, put down 3 and carry 1; then 1 + 6 + 5 = 12, which
228 Chapter Twenty-Two

is 7 + 5, so put down 5 and carry 1; then I + 3 = 4. In the second

addition, 18 = 2 x 7 + 4, put down 4 and carry 2; 2 + 30 = 32 =
4 x 7+ 4, put down 4 and carry 4;4+24=28=4x 7+0,
put down 0 and carry 4. Of course zero, which had not yet been in-
vented, was not written down by the scribe or clerk; in the papyri, a
blank space indicates zero.
This technique works splendidly with every calculation of the RP,
but it was new to the clerk, who got into many difficulties, particularly
with H 14 (Table 22.3C), H 16, and H 19 (Table 22.3D), where the
measurer was so meticulous as to include fingers in his linear measure-
ments. But we should not be too critical of him, for he was seldom very
far from the correct answer. It will help us to understand the scribe's
thought processes if the multiplication is rewritten to include the units
c and p for cubits and palms, so that cc and pp mean square cubits
and square palms, and ccc and ppp mean cubic cubits and cubic
palms. On this basis, cp means an area one cubit by one palm, and
ccp means one square cubit by one palm, and thus a superficial cubit.
Again, we have 7 cp = I cc, 7 pp = 1 cp, 7 ppp = 1 cpp, and soon,
because we are working in the scale of 7. Then here is the working of
H 24, repeated with these refinements.
3c 5p
\1 c 3cc 5cp
2p 6cp 10pp
Totals 4 cc 5 cp 3 pp
6p 24 ccp 30 cpp 18 ppp
Totals 4 ccc 0 ccp 4 cpp 4 ppp.
Thus the volume is 4 cubic cubits, 0 superficial cubits, plus the small
volumes represented by 4 cpp and 4 ppp, which the clerk recorded
as 2 f, presumably as an approximation. If, by analogy with ccp for a
superficial cubit (written simply as p by the clerk), ccf was written by
him simply as f, then we should expect for (4 cpp + 4 ppp), 2 ccf +
(2 14 28 196) ccf,* which latter he may have regarded as negligible,
* 4 cpp = (2 14) ccp = (2 4 28) ccf. And 4 ppp = (2 14) cpp = (14 98)
ccp = (4 28 28 -19-6) ccf = (4 14 -19-6) ccf. See table 22.5.
 The Reisncr Papyri 229

and thus wrote 2 f as an abbreviation meaning 2 ccf. The foregoing

calculation and all the other calculations provide strong evidence that
this technique, with perhaps some hieratic signs to identify ccc, ccp,
cpf, ppf, etc., was indeed used by the clerk on his memo pad, and all
the answers confirm the conclusion that, in stating the volume or
cubical contents, c meant cubic cubits, and p meant a superficial
cubit, that is, an area of one square cubit by one palm in thickness.
I am equally well convinced that f or fb for fingerbreadths in the
volume column meant a quarter of a superficial cubit, or an area of
one square cubit by one fingerbreadth, since a fingerbreadth is a
quarter of a palm. The evidence available is not so strong for this
latter, because none of the lines involving fingerbreadths is entirely
free of an error ofsome kind. Indeed where fractions, cubits, and palms
are involved, the clerk is in error (even though very slightly) in 44
percent of the cases; but where fractions, cubits, palms, and finger-
breadths are concerned, he is wrong in every case. But despite this,
what we may call the circumstantial evidence clearly points to a
cubical content measure of one-quarter of a superficial cubit, which
is itself one-seventh of a cubic cubit, and the scribal designation of
this unit is f. Very little thought is necessary to see that we cannot call
it a superficial palm, meaning one square palm by a fingerbreadth, for
this volume would be one forty-ninth of the volume the clerk indicates
by f. We will just have to remember that f means a quarter of a super-
ficial cubit, as the clerk must have had to do. Thus,
1 cubic cubit = 7 superficial cubits,
or1c=7p.
1 superficial cubit = 4 quarters of superficial cubits,
or1p=4f.
The analogy with the more familiar Egyptian table of length must
have been a very important factor in devising and adopting this
method of calculating volumes, at least in the dockyard workshops
of Nag`ed Deir.
I conclude this chapter by attempting to show how the clerk cal-
culated H 14 of Table 22.3C, in which the measurer included cubits,
palms, fingerbreadths, and fractions; the lot! We note also that when
230 Chapter Twenty-Two

adding, besides working in the scale of 7 for the conversion of p to c,

we must work in the scale of 4 for the conversion of f to p. It could
well be that it was in these conversions that the clerk became confused.
All that is shown for H 14 is:
length breadth depth units volume
2c 2f 12 lc lp if 1 3c 4p 1 2f
The working would have been :
c p f c p f
2 2
Totals \1 2 3cc 3cf
3cc 3cf
\1 c 3 ccc 3 ccf
1p 3 ccp 3 cpf
1f 3ccf 3cfT
Totals 1c lp 1f 3 ccc 4 ccp 2 ccf 3 cpf 3 cf..
The volume is therefore (3 c 4 p 2 f) plus some smaller fractions
represented by (3 cpf 3 cff). The clerk had somehow made an error,
where he had 1 2 f for the finger column, but we can perhaps find out
where the 2 came from. Thus;

3 cpf 3 cff = 3 cpf (2 4) cpf (3 = 4)

_ (4 7 28) ccf (14 28) ccf (3 = 7, tables)
_ (4 7 14 28 28) ccf (rearranging)
_ (4 7 14 14) ccf ( gen. (1, 1))
7) ccf (gen. (1, 1))
_ (4 4 28) ccf (Recto, 2 = 7)
= r1 28 ccf. ( gen. (1, 1) )

The correct answer is therefore 3 c 4 p 2 2 28 f, for which the

clerkhad3c 4p 1 2f.
If making such calculations as these was standard procedure in the
building of a temple, and if the distribution of salary or rations to the
overseers and workers was made in a manner similar to that shown in
 The Reisner Papyri 231

Chapter 11 for the temple of Illahun, then what must have been the
colossal mass of arithmetical calculation demanded by the building
of the Great Pyramid or any of the other immense structures and
monuments of ancient Egypt?
 APPENDIX 1

THE NATURE OF PROOF

Those historians who have adversely criticized the mathematics of

the ancient Egyptians confine themselves generally to the lack of
formal proof in the Egyptian methods and to the apparent absence
of what they call "the scientific attitude of mind" in the Egyptians'
treatment of mathematical problems. Thus these commentators have
written,
The fact that the Egyptians had evolved no better means of stating a
formula than that of giving three or four examples of its use is hardly
a tribute to the scientific nature of their mathematics.
The table [of the RMP Recto] is in itself a monument to the lack of
the scientific attitude of mind in the Egyptians.
That they did not reach the conception of scientific mathematics and
its dependence on cogent a priori demonstration is merely another
instance of the vast debt which the world owes to the Greeks. PEET*
There are no theorems (in the RMP) properly so called; everything
is stated in the form of problems, not in general terms, but in distinct
numbers. JouanAu t
All available texts point to an Egyptian mathematics of rather
primitive standards. srautx$
Egyptian geometry is not a science in the Greek sense of the word, but
merely applied arithmetic.
The Greeks may also have taken from the Egyptians the rules for the
determination of areas and volumes. But for the Greeks, such rules
did not constitute mathematics; they merely led them to ask; how
does one prove this? VAN DER WAERDEN§
Perhaps a more discerning assessment of the situation was expressed
by J. R. Newman when he wrote,
A sound appraisal of Egyptian mathematics depends upon a much
broader and deeper understanding of human culture than either
Egyptologists or historians of science are wont to recognise.4l
* "Mathematics in Ancient Egypt," Bulletin of the John Rylands Library,
Vol. 15, No. 2 (Manchester, 1931), pp. 439, 440, 441.
t "The Nature of Mathematics," in The World of Mathematics, James R.
Newman, editor, Vol. 1, Simon and Schuster, New York, 1956, p. 12.
$ A Concise History of Mathematics, Dover, New York, 1948, p. 23.
§ Science Awakening, Arnold Dresden, translator, Noordhoff, Groningen,
1954, pp. 31, 36.
1 The World of Mathematics, Vol. 1, p. 178.
 The Nature of Proof 233

Let us see if we can develop and amplify Newman's statement some-

what, and perhaps dispel some of the more depressing opinions ofsome
of the less charitable commentators.
It is true that the Egyptians did not show exactly how they estab-
lished their rules or formulas, nor how they arrived at their methods
in dealing with specific values of the variable. But they nearly always
proved that the numerical solution to the problem at hand was indeed
correct for the particular value or values they had chosen. For them,
this constituted both method and proof, so that many of their solu-
tions concluded with sentences like the following:
The producing of the same. (RMP 4)
The manner of the reckoning of it. (RMP 41)
The correct procedure for this [type of] problem. (MMP 9)
Manner of working out. (RMP 43, 44, 46)
Behold ! Does one according to the like for every uneven fraction which
may occur. (RMP 61B)
Thus findest thou the area. (RMP 55)
These are the correct and proper proceedings. (MMP 6)
The doing as it occurs. [or] That is how you do it. (RMP 28 and 23
other problems)
Shalt do thou according to the like in relation to what is said to thee,
all like example this. (RMP 66)
Twentieth-century students of the history and philosophy of science,
in considering the contributions of the ancient Egyptians, incline to
the modern attitude that an argument or logical proof must be
symbolic if it is to be regarded as rigorous, and that one or two specific
examples using selected numbers cannot claim to be scientifically
sound. But this is not true! A nonsymbolic argument or proof can be
quite rigorous when given for a particular value of the variable; the
conditions for rigor are that the particular value of the variable should
be typical, and that a further generalization to any value should be
immediate. In any of the topics mentioned in this book where the scribes'
treatment follows such lines, both these requirements are satisfied, so
234 Appendix One

that the arguments adduced by the scribes are already rigorous; the
concluding proofs are really not necessary, only confirmatory. The
rigor is implicit in the method.
We have to accept the circumstance that the Egyptians did not
think and reason as the Greeks did. If they found some exact method
(however they may have discovered it), they did not ask themselves
why it worked. They did not seek to establish its universal truth by an
a priori symbolic argument that would show clearly and logically
their thought processes. What they did was to explain and define in
an ordered sequence the steps necessary in the proper procedure, and
at the conclusion they added a verification or proof that the steps
outlined did indeed lead to a correct solution of the problem. This was
science as they knew it, and it is not proper or fitting that we of the
twentieth century should compare too critically their methods with
those of the Greeks or any other nation of later emergence, who, as it
were, stood on their shoulders. We tend to forget that they were a
people who had no plus, minus, multiplication, or division signs, no
equals or square-root signs, no zero and no decimal point, no coinage,
no indices, and no means of writing even the common fraction p/g;
in fact, nothing even approaching a mathematical notation, nothing
beyond a very complete knowledge of a twice-times table, and the
ability to find two-thirds of any number, whether integral or frac-
tional. With these restrictions they reached a relatively high level of
mathematical sophistication.
 APPENDIX 2
THE EGYPTIAN CALENDAR

The Egyptian year consisted of 12 months of 30 days each, or 36

decades of 10 days each, with 5 "year-end" (epagomenal) days that
were dedicated to the gods Osiris, Horus, Seth, Isis, and Nephthys;
the year-end days were their gods' birthdays. The 12 months were
divided into 3 seasons of 4 months each, which were the inundation or
sowing period, the coming forth or growing period, and the summer or
harvest period. This civil year of 365 days was retained because there
was no break in its continuity, and was still used in later Hellenistic
times; indeed, it was used in the Middle Ages by Copernicus and other
astronomers. "This calendar," writes Neugebauer, "is indeed the
only intelligent calendar which ever existed in human history."* It
is simpler even than the "perpetual calendar," t which, though recom-
mended for worldwide use by astronomers, seems condemned to
remain forever in some official pigeon-holes in all countries. Now,
the true length of the year (the solar year) is 3651/4 days,$ as the Egyp-
tians knew. This meant that their civil year slipped steadily backwards
through the solar year, 1 day every 4 years, 30 days or 1 month every
120 years, and thus 12 months or 1 year every 1,440 years,§ and the
seasons therefore were very slowly changing, though not specially
noticeably in one man's lifetime. What did it matter? There were few
worries in the ordinary affairs of daily life.
The most important event in Egyptian life was the annual flooding
of the Nile River, the inundation period, which coincided pretty
closely with the heliacal rising (just before dawn) of Sirius, the Dog
Star, the brightest star in either hemisphere. Thus the solar year
became also the Sirius year, by which sowing and agriculture gener-
ally was controlled. The first brief appearance of Sirius in the east-
ern sky was an important event in the Egyptian year. The next
morning Sirius would appear some minutes earlier, and so on, so that
before long, Sirius would no longer herald the dawn, and some other
bright star would serve this purpose. This measuring of the days by
* 0. Neugebauer, The Exact Sciences in Antiquity, Harper, New York, 1962,
p. 81.
t "ANZAAS Committee on Calendar Reform," Australian Journal of
Science, December, 1943; R. J. Gillings, "Perpetual World Calendar,"
Australian Mathematics Teacher, Vol. 1, No. 1, (April, 1945), p. 24.
$ Less 11 minutes.
§ Approximately of course. The cycle would then start again.
236 Appendix Two

the heliacal rising of stars gave rise to the system of decans, in which
each chosen star would serve its duty of noting the last hour of night
for 10 days (or nights), so that there would be 36 decans distributed
through the mornings of the year. Of course, not all decans would be
visible through any given night. At the time of the inundation, when
Sirius rises heliacally, 12 decans rise during the night, and thus the
"hours" of the summer night were determined. In winter there would
be more decans visible; thus the length of hours varied slightly, both
for the seasons and for nighttime and daytime. We see here the origin
of the division of the day into 24 hours that is now universally adop-
ted. These two calendars (of 365 and 3651/! days) existing side by side
from, it is thought, the time of the first pharaoh of Upper and Lower
Egypt, was " the most scientific organisation of calendars which has
yet been used by man."*
* J. W. S. Sewell, "The Calendars and Chronology," in The Legacy of
Egypt, S. R. K. Glanville, editor, Oxford University Press, London, 1963,
p. 7.
 APPENDIX 3
GREAT PYRAMID MYSTICISM

Perhaps the most famous and best known of all the architectural
constructions of the ancient Egyptians are the pyramids, the Great
Sphinx, and the Temple of Karnak. And of the 80 or so pyramids,
there is no doubt that the Great Pyramid of Khufu (in Greek, Cheops),
which was built during the Fourth Dynasty (c. 2644 B.c.), is the one
which has most stirred the thoughts and fired the imaginations of
all interested in ancient Egypt. Authors, novelists, journalists, and
writers of fiction found during the nineteenth century a new topic, a
new idea to develop, and the less that was known and clearly under-
stood about the subject, the more freely could they give rein to im-
agination and invention. These writers were forerunners of the
American, Edgar Rice Burroughs, who created the fictional character
Tarzan of the Apes, and set him up in central Africa, a country Bur-
roughs had never visited and knew nothing about. Burroughs let his
imagination run riot, and his novels (translated into fifty-six langu-
ages) achieved sales that were exceeded only by the Bible and
Euclid's Elements.
Many writers have propounded theories on the origins, the mathe-
matical properties, and the pseudo-astronomical marvels of Cheops,
and further, made extravagant prophecies about the Great Pyramid.
A resurgence of this cult occurred when Carter discovered the tomb
of Tutankhamen in 1923, so that these fictions were presented all over
again to the general reading public. Some of them still live on!
It may therefore come as a surprise to those readers with whom any
memories remain of those "wonderful disclosures" that most of the
miraculous stories written by these writers have no foundation in
scientific fact at all; that the remarkable mathematical properties
attributed to the Great Pyramid measurements are nowhere attested
by scholarly Egyptological studies. It is only because they then were,
and perhaps still are, so widely distributed and accepted that any
reference to them at all is made in this book. And if some long-cher-
ished illusions are thus destroyed, it is simply because they truly de-
serve to be destroyed, as being entirely contrary to fact, to history,
and to truth.
Among the extraordinary things claimed by these writers, one can
read that Piazzi Smyth asserted that half the distance round the square
base of the Great Pyramid divided by its height was exactly equal
238 Appendix Three

to Tr, the ratio of the circumference to the diameter of a circle; and

that 1/3., part of the base equals one five-millionth part of the earth's
axis of rotation, whatever that might mean!*
Even such a sober-minded person as H. W. Turnbull writes, "Their
land surveyors were known as rope stretchers, because they used ropes
with knots or marks at equal intervals to measure their plots of land.
By this simple means, they were able to construct right angles, for they
knew that three ropes of lengths three, four, and five units respectively,
could be formed into a right-angled triangle."t
It is, however, nowhere attested that the ancient Egyptians knew
even the very simplest case of Pythagoras's theorem! But Turnbull
goes further: "As Professor D'Arcy Thompson has suggested, the very
shape of the Great Pyramid indicates a considerable familiarity with
that [sic] of the regular pentagon. A certain obscure passage in Hero-
dotus, can, by the slightest literal emendation, be made to yield ex-
cellent sense. It would imply that the area of each triangular face of
the Pyramid, is equal to the square of the vertical height. If this is so,
the ratios of height, slope, and base, can be expressed in terms of the
golden section, or of the ratio of a circle to the side of the inscribed
decagon."
I am unable to understand exactly what Turnbull means by this
last sentence. But whatever it means, with further slight emendations,
the dimensions of the Eiffel Tower or Boulder Dam could be made to
produce equally vague and pretentious expressions of a mathematical
connotation. I am also unable to locate the reference attributed to
D'Arcy Thompson. I would say it is certainly not in his well-known
"Growth and Form."+ Nor can I locate the "certain obscure passage

* Piazzi Smyth was Astronomer Royal of Scotland in the late nineteenth

century. His discussion of the Great Pyramid may be found in his Our
Inheritance in the Great Pyramid, London, 1877.
t H. W. Turnbull, The Great Mathematicians, 4th edition, Methuen,
London, 1951, pp. 2f.
Unless it be the brief reference on p. 931 of Vol. 2. There, however,
Thompson makes no such suggestion: "The sectio aurea or `golden mean'
of unity, (V5- - 1)/2 = 0.61803... , is a number beloved of the circle
squarer, and of all of those who seek to find, and then to penetrate, the
secrets of the Great Pyramid. It is deep set in the regular pentagon and
 Great Pyramid Mysticism 239

in Herodotus," to find out what this slight "literal emendation"

might be.
Anyone wishing to look further into this pyramid mysticism should
refer to Leonard Cottrell's The Mountains of Pharaoh,* whose Chapter
11 is titled, "The Great Pyramidiot," meaning Smyth but which
refers also to John Taylor, Rev. John Davidson, and Edgar Stewart.
For a readily available text, The Pyramids of Egypt, by I. E. S. Edwardst
of the British Museum, can be consulted. That doyenofbibliographers,
Raymond Clare Archibald, records the following in his bibliography
to The Rhind Mathematical Papyrus, which are "enthusiastic in support
of the pyramid mysticism of Taylor and Piazzi Smyth" :

M. Eyth
Der Kampf um die Cheopspyramide, Heidelberg, 1902.
0. Nairtz
"Die Cheopspyramide, ein viertausendjahriges Rathsel," Prometheus, Vol.
17 (1906).
H. Neikes
Der golden Schnitt and die Geheimnisse der Cheops-Pyramids, Cologne, 1907.
J. and M. Edgar
The Great Pyramid Passages and Chambers, Glasgow, 1910.
K. Kleppisch
Die Cheopspyramide, ein Denkmal mathematischer Erkenninis, Munich and Berlin,
1921.
F. Noetling
Die kosmischen Zahlen da Cheopspyramide der mathematische Schlfissel zu den
Einheits-Gesetzen im Aufbau des Weltalls, Stuttgart, 1921.
D. Davidson and H. Aldersmith
T e Great Pyramid, its Divine Message, London, 1924.

dodecahedron, the triumphs of Pythagorean or Euclidean geometry. It is

a number which becomes by the addition of unity its own reciprocal-its
properties never end." On Growth and Form, Cambridge University Press,
London, 1951.
* Pan Books, London, 1963.
t Penguin Books, London, 1952.
 APPENDIX 4
REGARDING MORRIS KLINE'S VIEWS IN
MATHEMATICS, A CULTURAL APPROACH

Morris Kline, in his book on the history of mathematics,* has pro-

duced a scholarly volume of seven hundred pages. He is professor of
mathematics at New York University. In his opening chapters he
speaks of the mathematics of the early civilizations of Egypt, Babylon,
India, and China, the last two of which he dismisses very briefly, but
then he devotes less than three pages to Egypt and Babylonia. Three
pages, to cover an expanse of three thousand years!
Of course, Professor Kline might think that from a "cultural"
point of view, these civilizations merited no more than three pages in
seven hundred. But concerning Egypt, he mentions briefly the Egyp-
tian equivalent of 7r, the areas of fields, the volumes of structures, the
quantity of material needed to erect pyramids, the grain required to
make beer of a certain alcoholic content, Herodotus's remarks about
the flooding of the Nile, geometry, and the Egyptian calendar. He
then praises their achievements compared with contemporary civili-
zations, noting that they reached relatively high levels in religion, art,
architecture, metallurgy, chemistry, and astronomy, but considers
that their "contributions to mathematics were almost insignificant."
This pretty sweeping summing up of the Egyptian culture is followed
by the astounding statement that, compared with the Greeks, "The
mathematics of the Egyptians and Babylonians, is the scrawling of
children just learning how to write, as opposed to great literature"!
He claims that, "they barely recognized mathematics as a distinct
subject" (he apparently has not heard of the Rhind Mathematical
Papyrus) and that, "over a period of 4,000 years hardly any progress
was made in the subject "t
Such aggravated and condemnatory statements by an author of the
stature of Professor Kline can only imply that he has not fully informed
himself of the extent and nature of either Egyptian or Babylonian
mathematics. And this is most surprising, for he knew at least of
Neugebauer and Eves, because he lists their books+ in his work.
* Morris Kline, Mathematics, A Cultural Approach, Addison-Wesley, Reading,
Mass., 1962.
t All quotes cited here, Ibid., p. 14.
+ O. Neugebauer, The Exact Sciences in Antiquity, Princeton University Press,
Princeton, 1952; H. Eves, An Introduction to the History of Mathematics,
Rhinehart, New York, 1953.
 Regarding Morris Kline's Views 241

The Exact Sciences in Antiquity alone would have directed his atten-
tio;. to the Rhind Mathematical Papyrus of Chace et al. in two large
volumes, published conveniently in Ohio more than a quarter of a
century earlier; the Egyptian mathematics discussed therein most
certainly would not have come under the heading of "the scrawlings
of a child just learning how to write." If then Professor Kline were not
familiar with, or was unaware of, the writings of Eisenlohr, Peet,
Chace, Struve, Griffith, Schack-Schackenburg, Van der Waerden,
Vogel, and others on Egyptian mathematics, then he would have been
wiser to omit any reference to this ancient civilization in his "cultural
approach," and to have devoted his opening chapters to the early
Greeks, about whose work he was certainly very well informed.
 APPENDIX 5
THE PYTHAGOREAN THEOREM
IN ANCIENT EGYPT

There is no document to prove that the Egyptian knew even a par-

ticular case of the Pythagorean theorem. R. C. ARCHIBALD*
In 90% of all the books, one finds the statement that the Egyptians
knew the right triangle of sides 3, 4 and 5, and that they used it for
laying out right angles. How much value has this statement? None!
B. L. VAN DER WAERDENt
There is no indication that the Egyptians had any notion even of the
Pythagorean theorem, despite some unfounded stories about "har-
pedonaptai," who supposedly constructed right triangles with the aid
of a string with 3 + 4 + 5 = 12 knots [sic]. DIRK J. STRUIK+
The historian Cantor had conjectured that the Egyptians knew that a
(3, 4, 5)-triangle is right-angled, and that they used this knowledge
to construct right angles. A. SEIDENBERG§
There seems to be no evidence that they knew that triangle (3, 4, 5)
is right-angled; indeed, according to the latest authority (T. Eric Peet,
The Rhind Mathematical Papyrus, 1923), nothing in Egyptian mathe-
matics suggests that the Egyptians were acquainted with this or
any special cases of the Pythagorean theorem. T. L. HEATH1I

* "Outline of the History of Mathematics," American Mathematical Monthly,

Vol. 56, No. 1 (January, 1949), p. 16.
t Science Awakening, Noordhoff, Groningen, 1954, p. 6.
Concise History of Mathematics, Dover, New York, 1948, p. 23.
§ Scripta Mathematica, Vol. 24, No. 2 (New York, June, 1959), p. 10, foot-
note.
q The Thirteen Books of Euclid's Elements, Vol. 1, Cambridge University
Press, London, 1962, p. 352.
 APPENDIX 6
THE CONTENTS OF THE RHIND
MATHEMATICAL PAPYRUS

The Recto contains the result of the division of 2 by the 50 odd num-
bers from 3 to 101. This is followed by a table of the division of the
numbers 1 to 9 by 10, expressed in unit fractions.
PROBLEMSI-6
The division of 1, 2, 6, 7, 8, 9, loaves among 10 men.
PROBLEMS 7-20
The multiplication of (1 4) and (1 3 3) by various multipliers
containing unit fractions.
PROBLEMS21-23
Examples in subtraction. I (I 30), 1 - (4 $ 10 30
45).
PROBLEMS 24-27
Solution of equations in one unknown of the first degree, resolved by
the method of false assumption.
PROBLEMS 28-29
"Think of a Number" problems.
PROBLEMS 30-34
More difficult equations of one unknown of the first degree, resolved
by the method of division.
PROBLEMS 35-38
Even more difficult equations in one unknown of the first degree,
resolved by the method of false assumption but set in terms of hekats
of grain, containers, and hekat measures.
PROBLEMS 39-40
Arithmetic progressions.
PROBLEMS 41-46
Volumes or contents of rectangular and cylindrical granaries.
PROBLEM 47
A table of the fractions of a hekat in Horus-eye fractions.
PROBLEMS 48-55
Areas of triangles, rectangles, trapezia, and circles.
244 Appendix Six

PROBLEMS56-60
Sekeds, altitudes, and bases of pyramids.
PROBLEMS 61-61B
Tables of and rule for finding two-thirds of odd and even unit
fractions.
PROBLEM 62
A rather vague problem in proportion, concerning precious metals
by weight.
PROBLEM 63
The proportional division of loaves among men.
PROBLEM 64
An arithmetic progression, and S,, = (n/2)[21 - (n - 1)d].
PROBLEM 65
The proportional division of loaves among men.
PROBLEM 66
Division of fat. The amount issued per day.
PROBLEM 67
The proportion of cattle due as tribute.
PROBLEM 68
The proportional division of grain between gangs of men.
PROBLEMS 69-78
The pesus of bread and beer. Exchanges. Inverse proportion. The
concept of a harmonic mean.
PROBLEM 79
A geometric progression whose common ratio is 7.
PROBLEMS 80--81
Tables of Horus-eye fractions of grain in terms of hinu.
PROBLEMS 82-84
Unclear problems, dealing with the amounts of food for various
domestic animals, as geese, other birds, and oxen.
 Contents of the RMP 245

PROBLEM 85
Enigmatic writing. Upside down on the papyrus.
PROBLEMS 86-87
Memoranda of certain accounts and incidents, not altogether clear,
parts of which are missing.
 APPENDIX 7
THE CONTENTS OF THE MOSCOW
MATHEMATICAL PAPYRUS

The Moscow Mathematical Papyrus was purchased in 1893 by V. S.

Golenishchev from Abd-el-Rasoul, one of the brothers who found the
king's mummy at Deir el-Bahri. The papyrus was originally named
after its owner, but in 1912 it passed to the Museum of Fine Arts,
"avec toute ma collection, a Moscou, contre une rente viagire, que le Gouverne-
ment Russe s'itait engage de me payer, ma vie durant."* After the Russian
revolution of 1917, this life annuity was not paid. Golenishchev died in
1947. The papyrus is over 5 meters long and 8 centimeters high,
containing 25 problems, many of which are not clearly readable.
Turajeff and Tsinserling wrote on MMP 14 in 1917, but in 1930
W. W. Struve published a complete translation and commentary in
Quellen and Studien, Section A, Quellen, Berlin. The scribe of the MMP
in the view of Egyptologists was a very bad writer.
No. Detail.
1 Damaged and unreadable.
2 Damaged and unreadable.
3 A cedar mast. 3 5 of 30 = 16. Unclear.
4 Area of a triangle. 2 of 4 x 10 = 20.
5 Pesus of loaves and bread. Same as No. 8. See Chapter 12.
6 Rectangle, area = 12, b = 2 41. Find I and b.
7 Triangle, area = 20, h = 2 2b. Find h and b.
8 Pesus of loaves and bread. See Chapter 12.
9 Pesus of loaves and bread. See Chapter 12.
10 Area of curved surface of a hemisphere (or cylinder). See Chapter
18.
11 Loaves and basket. Unclear.
12 Pesu of beer. Unclear.
13 Pesus of loaves and beer. Same as No. 9. See Chapter 12.
14 Volume of a truncated pyramid. V = (h/3) (a2 + ab + b2). See
Chapter 18.
15 Pesu of beer. See Chapter 12.
16 Pesu of beer. Similar to No. 15.
17 Triangle, area = 20, b = (3 15)h. Find h and b.
* "With my whole collection in Moscow, in return for a life annuity which
the Russian government promised to pay me as long as I live."
 Contents of the MMP 247

18 Measuring cloth in cubits and palms. Unclear.

19 Solve the equation, 12x + 4 = 10. Clear. See Chapter 14.
20 Pesu of 1,000 loaves. Horus-eye fractions. See Chapter 12.
21 Mixing of sacrificial bread. See Chapter 12.
22 Pesus of loaves and beer. Exchange. See Chapter 12.
23 Computing the work of a cobbler. Unclear. Peet says very diffi-
cult.
24 Exchange of loaves and beer. See Chapter 12.
25 Solve the equation, 2x + x = 9. Elementary and clear.
On analysis of these problems we find,
[2 problems] Nos. 1 and 2 are not readable.
[11] Five problems (8, 9, 13, 22, 24) on the pesus of loaves and beer
are not perfectly clear. Three problems (5, 20, 21) deal with the pesu
of loaves only. They are difficult to understand. Three problems (12,
15, 16) deal with beer and its pesu only. They are clear and simple.
[6] Three treat the area of a triangle. No. 4 merely finds the area
of a right triangle, while Nos. 7 and 17 are equivalent to the solution
of two simultaneous equations, one of the second degree. Two prob-
lems (19, 25) concern the solution of equations of the first degree,
which are very simple, and No. 6 is on simultaneous equations, one
of the second degree.
[4] Problems 3, 11, 18, and 23 are miscellaneous problems, none of
which is entirely clear.
[2] No. 14 on the volume of a truncated pyramid is a most important
problem in the history of Egyptian mathematics. It has no counter-
part in any other mathematical papyrus. No. 10 deals, I consider,
with the area of the surface of a hemisphere, as Struve thought, and
if this is so, it becomes the outstanding Egyptian achievement in the
field of mathematics.
 APPENDIX 8
A PAPYRITIC MEMO PAD

I have often made references to the probable use of papyritic memo

pads by the scribes when we have been confronted with what appears
to be a piece of mental arithmetic that seems to be rather too long
or too difficult to be done in the head. If such memo pads were some-
times used, we can envisage what one might have looked like with
the following example, taken from Problem 35 of the RMP.
The scribe writes at once,
I of(5 10) = 5.
How might he have done this on his memo pad?
Either,

I of (5 10) = (10 30) 15 [RMP 61B.]

= 10 (15 30) [Rearranging.]
= 10 10 [EMLR 24, or G rule.]
EEE 5. [gen. (1, 1)..]

Or,
I of (5 10) _ 3 of (1 2 of 5) [TO = 2 of 5.]
_ (I of 1 2) of 3 [Rearranging.]
1 of 5 [1 = reciprocal of 1 2.]
3.

Or,
3 of (5 10) = I of (10 10) 10 [gen. (1, 1).]
= 3 of 10 10 10 [Removing brackets.]
= 10 10 [ 2 parts out of 3.]
= 5. [gen. (1, 1).]
Or,
3 of (3 10) = 1 9 of (10 20) [Mutt. and divide by 2.]
= 4 x (30 60) [Mutt. and divide by 3.]
=4x 20 [EMLR 23, or G rule.]
= 5.
 APPENDIX 9
HORUS-EYE FRACTIONS IN TERMS OF HINU:
PROBLEMS 80,81 OF THE RHIND
MATHEMATICAL PAPYRUS

PROBLEM 80
This is not really a problem but a reference table, which the scribe
states is for the use of the "functionaries" of the granary, in conjunc-
tion with their vessels for measuring out the grain. The table is based
on the equivalence 10 hinu = 1 hekat:
Hekat Hinu
T 10
5
4 2 Q
$ 1 4
16 $
32 4 16
64 $ 32

PROBLEM 81
This is a more extensive table, the first portion of which is an exact
repetition of Problem 80. Then follow 28 entries, giving various frac-
tions of a hekat of grain, expressed in terms of hinu and in Horus-eyc
fractions and ro (see Table A9. 1). There are some duplications, as the
first column of line numbers (following Chace) in Table A9.1 shows.
I have, of course, altered the scribe's sequence in this table.
250 Appendix Nine

TABLE A9.1
Horus-eye fractions in terms of hinu.
Line Equivalent in Horus-eye Fractions Equivalent
(Chace) Fraction of a Hekat 2 48 16 32 64 ro. in Hinu

a3, d3. T2- 33 65

c3. 2 5
a6, d4. 3 4 16 64 1 3 3 3
a7, c4. 4 4 22
a8, bl. 5 $ 16 4 2
a9. 6 $ 32 33 1

d5. $ $ 14
W. 10 16 32 2 1

b5. 15 16 13
d6. 16 T6-

b3. 30 T21 64 1 2
cl. 30 32 3 3
el. 32 32 4 16
M. 40 64 3 4
c2. 60 64 3 6
e2. 94- 64 $ 32

c5. 24 2 4 72
a4,dl. 2$ 2 $ 64
2 16 2 16- 52$
2 32 2 32 5416
264 2 64 5$§2-
a5, d2. 4$ 4$ 324
4 16 4 T6' 3$
4 32 4 32 2 2 4 16
464 4 64 22$32
24$ 2 4$ 824
2416 2 4 16 8$
2432 2 4 32 72416
2 464 2 4 64 7 2$32
 APPENDIX 10
THE EGYPTIAN EQUIVALENT OF THE
LEAST COMMON DENOMINATOR

TABLE A10.1
Least common denominators in the Rhind Mathematical Papyrus.
RMP No. Fractions to be Added Reference No. The Sum

21. r3 5 15 15 15 1

22. 3 5 10 To- 30 1

23. 4 8 9 10 30 40 45 45
32. 12 18 24 36 48 114 228
912 4
342 456 684 9121
33. 56 84 112 392 679 776 1164
5432 21
1358 4074 4753 1358 1552 5432}
34. 7 14 14 28 28 56 56 4 8
36. 12 20 30 53 53 106 166 159
1060 4
212 265 318 318 530 626 795 10601
37. 16 32 84 T2- 5T7-6 576 8
37. 12 T6- T2- 36 36 96 288 288 288 4
38. 11 11 22 22 33 86 66 66 9

In the group of problems RMP 7 to 20, which show how to multiply

certain fractional numbers, there are a further 8 examples of the
addition of fractions by the use of red auxiliaries, but the reference
numbers are confined to the numbers 18 and 28. But these are a special
group. In all of the 10 other examples of the RMP where this method
is used, the scribe chose the largest denominator as a reference number,
and in four of these (34, 37, 37, 38), it happens to be what we call the
least common denominator, but this is just by chance (Table A10.1).
The same applies to numbers 21 and 22, but for some reason the
auxiliary numbers are in black like all other numbers, not in red;
this is probably an accident-the scribe may temporarily have run
out of red ink. Nowhere does the scribe show how he did all the lesser
divisions, for example, the 16 divisions in RMP 36, where e.g., 1060
159 = 6 9, or 1060 = 12 = 88 9. Now, however adept the scribe may
have been at mental arithmetic, he surely must have made use of
memo pads, some kind of papyrus scribbling blocks, to perform these
divisions, and to have access to 3 tables, for these two divisions would
have to be done as follows.
252 Appendix Ten

1060 = 159:
1 159
\2 \318
\4 \636
\9 \106
Totals 6 1 1060.

1060 = 12:
1 12
2 24
4 48
\8 \ 96
\16 \192
32 384
\64 \768
8
\3 \ 4
Totals 88 3 1060.

It is difficult to imagine such divisions as these being done in the head.

To sum up, we find that there are in the RMP about 130 occasions
where the scribe needed to add up unit fractions, and quite often a
large number of fractions, as in RMP 70, where there are 20 of them,*
yet he used his technique of a reference number and the red auxiliaries
only 18 times, i.e., in less than 15 percent of the cases where they
would be applicable. It seems pretty clear that in Problems 7 to 20,
where he was teaching the multiplication of fractions, he included
this method 8 times as instruction to the student in very simple cases,
as for example (RMP 8), he used the reference number 18 to prove
that 4 12 = 2. In Problems 21 to 23, where he showed how to
subtract fractions, the additions required are a little harder (see
table) and he used it 3 times; while in Problems 30 to 34, where he
showed how to divide by fractions, he again used it 3 times, one of
which was unnecessary. Finally, in Problems 35 to 38 he had occasion
to use it 4 times, and possibly a couple of these were unnecessary from
0 The last three are 336, 504, and 1008.
 Egyptian Equivalent of the Least Common Denominator 253

the point of view of difficulty. In this last group of problems the scribe
was dealing more particularly with the divisions of the hekat for
grain, in which Horus-eye fractions were used. I mean by "unneces-
sary " that in Problem 34, the addition of the fractions
7 14 14 28 28 Y6_ = 4 $

was purely mental for the scribe, because he had just established that
7 14 28 = 4 (Problem 11)
and 14 28 56 = $, (Problem 12)
so that the sum followed at once by inspection. But he was teaching a
certain technique, so he used his reference number method for the
benefit of his readers.
TABLE A11.1
Two-term equalities for Egyptian unit fractions.
(1, 1) 2 2=1 (2, 1) 6 3=2 (3,1) 12 4=3 (4,1) 20 5=4 (5,1) 30 6=5
(1,2) 3 6=2 (2,2) 8 8=4 (3,2) 15 10 = 6 (4,2) 24 12 = 8 (5,2) 35 14=10
(1, 3) 4 12 = 3 (2, 3) 10 15 = 6 (3, 3) 18 18 = 9 (4, 3) 28 21 = 12 (5, 3) 40 24 = 15 tz+

(1,4) 5 20 = 4 (2,4) 12 24 = 8 (3,4) 21 28 = 12 (4,4) 32 32 = 16 (5,4) 45 36- 20 to

r
(1 , 5) 6 30 = 5 (2,5) 14 35 = 10 (3, 5) 24 40 = 15 (4, 5) 36 45 = 20 (5, 5) 50 50 = 25
(1,6) 7 42 = 6 (2,6) 16 48 = 12 (3,6) 27 54 = 18 (4,6) 40 60 = 24 (5,6) 55 66 = 30 C O
(1,7) 8 56 = 7 (2,7) 18 63 = 14 (3,7) 30 70 = 21 (4,7) 44 77 = 28 (5,7) 60 84 = 35
(1, 8) 9 72 = 8 (2,8) 20 80 = 16 (3, 8) 33 88 = 24 (4,8) 48 96 = 32 (5,8) 65 104 = 40
(1, 9) 10 90 = 9 (2, 9) 22 99 = 18 (3, 9) 36 108 = 27 (4, 9) 52 117 = 36 (5, 9) 70 126 = 45
(1,10) 11 110 = 10 (2, 10) 24 120 = 20 (3, 10) 39 130 = 30 (4,10) 56 140 = 40 (5,10) 75 150 = 50 rte
(6,1) 42 7=6 (7,1) 56 8=7 (8,1) 72 9=8 (9,1) 90 10 = 9 (10,1) 110 11 = 10
(6,2) 48 16 = 12 (7,2) 63 18 = 14 (8,2) 80 20 16 (9,2) 99 22 = 18 (10,2) 120 24 20
(6, 3) 54 27 = 18 (7, 3) 70 30 = 21 (8, 3) 88 33 = 24 (9, 3) 108 36 = 27 (10, 3) 130 39 = 30
(6, 4) 60 40 = 24 (7,4) 77 44 = 28 (8,4) 96 48 32 (9,4) 117 52 = 36 (10,4) 140 56 = 40
(6, 5) 66 55- 30 (7, 5) 84 60 = 35 (8, 5) 104 65 = 40 (9, 5) 126 70 = 45 (10, 5) 150 75 = 50 nP
(6, 6) 72 72 = 36 (7, 6) 91 78 = 42 (8, 6) 112 84 = 48 (9, 6) 135 90 = 54 (10, 6) 160 96 = 60
(6, 7) 78 91 = 42 (7, 7) 98 98 = 49 (8, 7) 120 105 = 56 (9, 7) 144 112 = 63 (10, 7) 170 119 = 70
(6,8) 84 112 = 48 (7,8) 105 120 = 56 (8,8) 128 128 = 64 (9,8) 153 136 = 72 (10,8) 180 144 = 80
(6,9) 90 135 = 54 (7,9) 112 144 = 63 (8,9) 136 153 = 72 (9,9) 162 162 = 81 (10,9) 190 171 = 90
(6,10) 96 160 = 60 (7,10) 119 170 = 70 (8,10) 144 180 = 80 (9, 10) 171 190 = 90 (10,10) 200 200 = 100
0
PO
 APPENDIX 12
TABLES OF HIERATIC INTEGERS AND
FRACTIONS, SHOWING VARIATIONS

TABLE A12.1
Integers as written in various papyri.
Reisner MMP KP EMLR RMP Berlin

I t, 1 41 1
II 2 L) 9 it Ll "u 11

III 3 Ill lq ILI '" (LI 4

IIII 4 K`) "` Ill( 'i - »
'1 7
'
5 "1 "1

Ifl 6 :i 00 we

III
n
9-
to
IL A
A
nn 20
A A A
nn
n nn
30
40
50
-
`7
-
2I l
nnn 60
nnn TO

n nn $0 -%:
nnnn
nnn 90
256 Appendix Twelve

TABLE A 12.1 (continued)

Rcisner MMP KP EMLR RMP Berlin

9 I00

-09 200

909 300
900
400
99 500

999 600
9999
9 09 700
9909
9999 800

99 9 900

f 1000

ff 2000

Ire 3000

f : :: 4000 A
 Hieratic Integers and Fractions 257

TABLE A12.2
Hieratic integers, showing variations.

I I%A 6i pa 91 1

2 4ya N1(32 11"A 62 11 92

3 "1W:l Mw33 ul n 63 91 AL 93 u1=
4 w" -. _ 16 SM =A 64 AL p y1 94 V UL
5 "1 "1 35 '4-A ''l -A 65
6 2"36 w 66
7 37 ' 271 6T
a SS

9 69 S 99
10 1 1440 . 70

u t4)11141 W 200 ) 9
12 L4 'l 111 42 1+. 11 300 .
13 WA 43 *I 400
14 -A " .... -'1404,500 y
p1=
15 pIA 45 a 600
s6

17
is
V 46
2A+7
A 48 78
a

zq o
=
700 5, J
goo
19 WA 49 79 W-1 401000
20 i1 A 'A So 511 so .. y1 z4 2ooo
`1
1
258 Appendix Twelve

TABLE A12.2 (continued)

81 VA 51 1-1 91 JAIL SM

22 it'll 52 all 82 4 Ad 000

23 QUA 53 1111 83 ,d.& 5000 A4

24 --AA ji " * 1.0 ci-.1 84 1dl.iW 6000

25 'I l1 ss 85 u`aI, 41 im too

28 to 56 so
L4o 9000 4
2T 'A 57 2-1 8T AZ4 f " i0,= .
1 1

88 all 58 88 W AL 100,000 /;O yo

29 4`11 59 SUI 4121 99 OWL socav
30 T -T 60 .:, su 90 y= ' iaa as
 Hieratic Integers and Fractions 259

TABLE A 12.3
Hieratic fractions, showing variations.

if 4111
TABLE A13.1 Chronology.
o w
o u
V t0 py %v
C U C A
OpO O v U U O OC. A CC
.
c ?i ; u 0. r-
0. 0. m0 0 a, £ U'y
< 0. < 0 is 0 U to a. at <v -eR15 ta03u
Archaic
4241 Calendar.
6 6 ? ? Metal cast.
Early Dynastic
1 8 14 3110 3100 3188 3100 3188 3200 3400 250 250 Mcncs-Upper & Lower Egypt.
D.C. +150
II 9 23 2883 2847 2780 150 400 Capital-Memphis.
Zoser-Step pyramid-Sakkara.
Old Kingdom
III 9 32 2664 2700 2815 2670 2815 2778 2980 80 480 Copper tools.
Hieroglyphic-Hieratic writing.
IV 8 40 2614 2620 2600 2690 2680 2900 140 620 Cheops-Great Pyramid--Giza.
Cephrcn Pyramid-Sphinx-Giza.
Mycerinus Pyramid-Giza.
V 8 48 2501 2480 2560 2565 2750 140 760 Neferikara Pyramid-Abu Sir.
VI 11 59 2341 2340 2420 2420 2423 2625 160 920 Ncrenra Pyramid-Sakkara.
Pepi II Pyramid-Sakkara.
World's longest reigning king.
90+ yrs.
First Intermediate Period
VII ? ? ? ? 2294 2270 2294 2263 2475
VIII 18 77 2174 65 985
IX 14 91 2154 2445 73 1058
Inkrrtgnum
X 18 109 2100 2239 68 1126
2233

Middle Kingdom
XI 8 117 2134 2134 2133 2134 2160 143 1269 Capital-Thebes.
Reisner Papyri-Hieratic-Boston.
XII 8 125 1991 1991 2132 2000 1990 1891 2000 205 1474 Akhmim tablets-Hieratic-Cairo.
Senusret I Pyramid-Lisht.
Second Intermediate Period
XIII 1785 1786 1777 1788 Rhind Mathematical Papyrus-
Hieratic-British Museum.
Moscow Mathematical Papyrus-
Hieratic-Moscow.
Hyksos (Shepherd Kings) Period
XIV 1785 1700 1700 Kahun Papyri-Hieratic-London.
Hyksos
XV 1678 Domination.
XVI Rhind Mathematical Papyrus
132 257 1680 211 1685
1647 copied-A°h-most.
Egyptian Mathematical Leather
Roll-British Museum.
XV i I 1. 1600 1600 1570 Hyksos expelled.
TABLE A13.1 (continued)

1.

X .O V 'V x2 K ro
X'C
rs 2-
a a o
Q <a < 2 za 0 W (5 to C1 dvo d
New Kingdom
XVIII 16 273 1570 1575 1573 1567 1573 1580 1580 267 1952 0. Hatshepsut. Bronze appears.
Tutmosis.
A khenaten-Neferti ti.
Tutankhamen.
Ramscs If.
The Exodus.
XIX 9 282 1304 1308 1320 1314 1320 1350 124 2076 Berlin Papyri-Hieratic-Berlin.
Anastasy Papyrus-Hieratic-
Leyden.
XX 10 292 1192 1184 1200 1194 1250 1200 100 2176 Temple of Karnak-Abu Simbel.
Late Dynastic Period
XXI 7 299 1075 1087 1090 1085 1085 1090 142 2318 Tanites.
Religious writing still Hieroglyphic.
Other, Hieratic.
XXII 8 307 940 945 950 945 128 2446 Libyans.
XXIII 4 311 761 817 745 745 87 2535
XXIV 52 313 725 730 718 718 5 25381 Overlapping period.
XXV 16 319 736 725 712 712 751 712 61 25995 Ethiopian conquest. Demotic
writing.
XXV1 9 328 664 664 663 664 656 663 140 2739 Saite period.
Assyrian domination.
Persian Conquest
XXVII 8 336 525 525 525 525 525 525 120 2859 Persian conquest.
Cambyses-Xerxes-Darius.
Thales-Pythagoras.
Late Dynastic Period
XXVIII 1 337 404 404 404 404 5 2864 Herodotus visits Egypt (450).
XXIX 5 342 398 399 398 398 399 20 2884
XXX 3 345 378 380 378 380 38 2922
XXXI 3 348 341 343 332 332 332 10 2932 Aristotle.
Three Persian Kings.
Greek Period
332 332 332 332 332 332 9 2941 Capital Alexandria-Coptic
Alexander
writing.
the
Archimedes.
Ptolemies
4 62 23 23 23 93 234 Cleopatra-Last of the Ptolemies
(30).

Roman Perio d
30 30 30 30 30 30 30 30 3264 Julius Caesar-Mark Antony.
Brass appears.
Province 0 0- 0- 0- 0- 0- 0- 3294 Birth of Christ.
Revillout Papyrus-Demotic-
Roman
of London (140).
Period
24 64 64 64 64 64 .D. 658 Thompson Tablet-Greek-
Rome A.D. 400 3694 London (350).
Michigan Mathematical Papyrus
500 3794 -Greek-Ann Arbor (350).
TABLE A 13. 1. (continued)

o L
o T
y C
y OO
Y V
V.4 E
E cud
a as 0. .
a. tS V u M CO d'o <s k 03u
A <c a d °
Byzantine Empire
Justinian (538).
Byzantine 600 3894 Crum Ostracon-Coptic--London
Control (550).
690 Hegira (622)-Mohammed-
Flight from Mecca.
700 3994 Alexandria captured by the Arabs
(642).
800 4094 Baillet-Akhmim Papyrus-Greek
--Cairo (750).
Islam 900 4194 Crum.-Parchmcnt-Palimpsest.
Demotic & Greek-
1000 4294 London (circa 950--1000).
1100 4394
1200 4494 Saladin-Sultan of Egypt (1137-
1193).
1300 4594 Mamclukes conquer Egypt 1250.
1400 4694 End of Byzantine Empire (1453).
Mamc- Constantinople captured by
lukes Ottoman Turks.
1500 4794 Mamclukes defeated (1517).
I
 1600 4894 Robert Recordc-Arithmetic.
Galileo.
Napier-Logarithms. Descartes---
I ndices.
Turkish Pepys (1662) learns multiplication
Domina- tables (Age 30).
tion 1700 4994 Newton-Principia (1687).
Calculus. Leibnitz.
Napoleon's Egyptian expedition
(1789).
1800 5094 Ro setta stone found-Nile delta
(Mahomet (1799).
Ali) Young (1814).--Bankcs -Philae
Obelisk (1819).
C:hampollion-",)c ticns I'affaire."
(1822).
1900 5194 R M P found -Thebes (1859).-
Suez Canal (1869).
World Flinders Petrie.
War I Tutankhamen's tomb-Carter
(1922).
World Chace- RMP (1927).
War 11 2000 5294 K. Fuad (1936). K. Farouk. Col.
Nasser.
0-
362 62
Pharaohs English Kings 0
0)
is
 APPENDIX 14
A MAP OF EGYPT
 APPENDIX 15

THE EGYPTIAN MATHEMATICAL LEATHER ROLL-LINE 8.

HOW DID THE SCRIBE DO IT?

The Egyptian Mathematical Leather Roll (EMLR) contains two

identical tables whose 26 equalities are the sums of various unit
fractions; in modern notation they would appear as:

1 + 1 = 1
Line 3.
4 12 3

However, the ancient Egyptians did not use unity as the numerator of
any fraction. They merely placed a mark or a short line above the
number, so that 'A was written as 4 , and thus line 3 appeared as

4 12 3

Just how they discovered such equalities we do not know for certain. It
is possible that they may have used what I will refer to as the "G-
Rule," illustrated by the following example: 4 divided into 12 equals
3, so that, increasing this quotienty unity gives 4, which divided into
12 gives 3, hence the equality T 12 = 3 results.
Scribes never used signs to denote plus, minus, or equals; indeed,
they often omitted the bar mark above a number to indicate it was a
fraction, whenever there was no doubt, as was the case in the EMLR.

Line 1 10 40 8
Line 2 5 20 4
Line 3 4 12 3
Line 4 10 10 5
Lines 5, 6 3 6 2

Of the 26 equalities in the EMLR, 16 are binomials and 8 are

trinomials, so that there remain only two equalities with four terms,
which are lines 8 and 9

25 15 75 200 8
50 30 150 400 16

Unexpectedly, the fraction 25 precedes 15, contrary to the normal

procedure, and so this oddity might give some hint as to how the scribe
proceeded on this occasion. I contributed an article to Historia
268 Appendix Fifteen

Mathematica [ 1979, Vol. 6, No. 41 in the hope that perhaps a reader

might conjecture what the scribe's method of deriving this unusual
four-term, unit-fraction equality might have been. But so far, no such
suggestion has been received.
One must remember that twice-times was the only multiplication
table that the ancient Egyptians used. I conclude this note with two
possible methods the scribe might have devised in producing Number
8 of the 26 equalities in the EMLR. The first is, in my view, the better
method, but the second shows how he may have worked, so that 25
comes before 15. The question which remains unanswered is, "Why
did he not then reverse them?"

Starting from EMLR Line Number I

EMLR
lines
1 10 40 =8
24,2 15 30 50 200 =8
23 15 30 75 150 200 =8
- 15 30 150 75 200 =8
G
Rule 15 25 75 200 =8
1 10 40 =8
4, 2 20 20 50 200 =8
2 25 100 20 50 200 =8
- 25 20 50 100 200 =8
24 25 20 75 150 100 200 =8
- 25 20 100 150 75 200 =8
G
Rule 25 20 60 75 200 =8
3 25 15 75 200 =8
 BIBLIOGRAPHY

Akhmim Papyrus (AP)

The Cairo Museum, Egypt, Catalogue No. 10758.
Archibald, R. C.
Bibliography of Egyptian Mathematics, Mathematical Association of America,
Oberlin, Ohio, 1927.
Archibald, R. C.
Chace Memorial Address, The Review Club of Pembroke College, Providence,
Rhode Island, 1932.
Archibald, R. C.
"Outline of the History of Mathematics," The American Mathematical
Monthly, Vol. 56 (Menasha, Wisconsin, 1949).
Becker and Hofmann
Geschichte der Mathematik, Athenaum-Verlag, Bonn, 1951.
Bergamini, David
LIFE Science Library: Mathematics, TIME-LIFE International, Netherlands
N.V., 1965.
Berlin Papyrus (BP)
Staatliche Museen zu Berlin, Catalogue No. 6619.
Borchardt, Ludwig
Gegen die Zahknmystik an der grossen Pyramide bei Gise, Berlin, 1922.
Borchardt, Ludwig
"Salary Distribution for the Personnel of the Temple of I Ilahun," Zeitschrjft
fur Agyptische Sprache, Vol. 40 (Leipzig, 1902-1903), pp. 113-117.
Boyer, Carl B.
A History of Mathematics, John Wiley, New York, 1968.
Breasted, J. H.
A History of Egypt, Hodder and Stoughton, London, 1946.
Broken Hill Proprietary Company
Shapes and Sections Handbook, Melbourne, 1937.
Brugsch, H.
"Uber Bau and Masse des Temples von Edfu," Zeitschrift fur Agyptische
Sprache, Vol. 8 (Leipzig, 1870).
Bruins, E. M.
"Ancient Egyptian Arithmetic," Kon. Nederland Akademie van Wetenschappen,
Ser. A, Vol. 55, No. 2 (Amsterdam, 1952).
Bruins, E. M.
"Over de Benandering van or/4 in de Aegyptische Meetkunde," Kon.
Nederland Akademie van Wetenschappen, Vol. 48 (Amsterdam, 1945).
270 Bibliography

Budge, Sir E. A. W.
The Rosetta Stone, Trustees of the British Museum, London, 1913. Reprinted
with revision, 1957.
Cantor, M.
Vorlesungen uber Geschichte der Mathematik, Vol. 1, Leipzig, 1880.
Chace, A. B.; Bull, L.; Manning, H. P.; and Archibald, R. C.
The Rhind Mathematical Papyrus, Mathematical Association of America, Vol.
I (1927), Vol. 2 (1929), Oberlin, Ohio.
Chronique D'Egypte
Bulletin Periodique de la Fondation Egyptologique Reine Elisabeth, publil avec le
Contours du ... la Fondation Universitaire de Belgique, Vol. 41, No. 81, Brussels,
1966.
Cottrell, L.
Life Under the Pharaohs, Pan Books, London, 1964.
Cottrell, L.
The Lost Pharaohs, Evans, London, 1949.
Cottrell, L.
The Mountains of Pharaoh, Pan Books, London, 1964.
Cottrell, L.
The Penguin Book of Lost Worlds, (2 volumes) Penguin, London, 1966.
Crum, W. E.
Catalogue of the Coptic Manuscripts in the British Museum, London, 1905.
Crum, W. E.
Coptic Ostraca from the Collections of the Egypt Exploration Fund, the Cairo Museum
and Others, London, 1902.
Davidson, D., and Aldersmith, H.
The Great Pyramid, its Divine Message, London, 1924. Reprinted 1937.
Desroches-Noblecourt, Christiane
Egyptian Wall-Paintings From Tombs and Temples, Fontana Unesco Art Books,
Collins, Uncsco, Milan, 1962.
Driver, G. R.
Semitic Writing from Pictograph to Alphabet (British Academy Schweich Lec-
tures, 1944), Oxford University Press, London, 1948.
Edgar, J. and M.
The Great Pyramid Passages and Chambers (2 volumes), Glasgow, 1910. Re-
printed 1923.
Edwards, I. E. S.
The Pyramids of Egypt, Pelican Books, London, 1947.
Egyptian Mathematical Leather Roll (EMLR)
See S. R. K. Glanville, "The mathematical leather roll in the British
Museum," Journal of Egyptian Archaeology, Vol. 13 (1927), pp. 232-238.
 Bibliography 271

Eisenlohr, A.
Ein Mathemaiisches Handbueh der Alien Agypter, (Papyrus Rhind des British
Museum), ubersetzt and erkldrt, Leipzig, 1877.
Encyclopaedia Britannica
" Egypt-Mathematics," by F. L. Griffith, 11th edition, Vol. 9 (1910), pp.
46-47.
Engelbach, R.
"Mechanical and Technical Processes," in The Legacy of Egypt, S. R. K.
Glanville, editor, Oxford University Press, London, 1942. Reprinted 1963.
Erman, Adolf
The Ancient Egyptians, A Sourcebook of their Writings, Introduction by W. K.
Simpson, Harper Torchbooks, Harper, New York, 1966. First published
1923.
Eves, Howard
An Introduction to the History of Mathematics, Holt, Rinehart and Winston, New
York, 1964. First published 1953.
Finger, Charles J.
Pepys' Diary, Haldeman-Julius Company, Girard, Kansas, 1922.
Gardiner, Sir Alan
Egyptian Grammar, being an Introduction to the Study of Hieroglyphics, Oxford
University Press, London, 1957. First published 1927.
Gardiner, Sir Alan
Egypt of the Pharaohs, Oxford University Press, London, 1964. First published
1961.
Gillain, O.
La Science Egyptienne: l'Arithmftique au Moyen Empire, Paris and Brussels,
1927.
Gillings, R. J.
"An Interesting Ostracon," The Australian Mathematics Teacher, Vol. 19,
No. 1 (Sydney, 1963), pp. 15-19.
Gillings, R.J.
"Mathematical Fragment from the Kahun Papyrus," The Australian
Journal of Science, VoI.29, No. 5 (November, 1966), pp. 126-130.
Gillings, R. J.
"Mathematical Fragment from the Kahun Papyrus IV, 3, The Australian
Mathematics Teacher, Vol. 23, No. 3 (Sydney, 1967), pp. 61-64.
Gillings, R.J.
"Perpetual World Calendar," The Australian Mathematics Teacher, Vol. 1,
No. 1 (Sydney, 1945), p. 24.
Gillings, R. J.
"Problems 1-6 of the Rhind Mathematical Papyrus," The Mathematics
Teacher, Vol. 55, No. I (Washington D.C., 1962), pp. 61-69.
272 Bibliography

Gillings, R. J.
"Sum of n Terms of an Arithmetical Progression in Ancient Egypt," The
Australian Journal of Science, Vol. 31, No. I (Sydney, 1968), pp. 47-50.
Gillings, R. J.
'` The Addition of Egyptian Unit Fractions," Journal of Egyptian Archaeology,
Vol. 51, (London, 1965), pp. 95-106.
Gillings, R.J.
"The Area of the Curved Surface of a Hemisphere in Ancient Egypt," The
Australian Journal of Science, Vol. 30, No. 4 (Sydney, 1967), pp. 113-116.
Gillings, R.J.
"The Egyptian Mathematical Leather Roll (B.M. 10250)," The Australian
Journal of Science, Vol. 24, No. 8 (Sydney, 1962), pp. 339-344.
Gillings, R. J.
"The Egyptian 2/3 Table for Fractions," The Australian Journal of Science,
Vol. 22, No. 6 (Sydney, 1959), pp. 242-250.
Gillings, R. J.
"The Remarkable Mental Arithmetic of the Egyptian Scribes," The
Mathematics Teacher (Washington, D.C., 1966). Part I: Vol. 59, No. 4, pp.
372-381. Part II: Vol. 59, No. 5, pp. 476-484.
Gillings, R. J.
"The Volume of a Cylindrical Granary in Ancient Egypt," The Australian
Mathematics Teacher, Vol. 22, No. 1 (Sydney, 1966), pp. 1-4.
Gillings, R. J.
"The Volume of a Truncated Pyramid in Ancient Egyptian Papyri," The
.llathematics Teacher, Vol. 57, No. 8 (Washington D.C., 1964), 552-555.
Gillings, R. J.
"`Think of Number' Problems 28, 29 of the Rhind Mathematical Papyrus
(B.M. 10057-8)," The Mathematics Teacher, Vol. 54, No. 2 (Washington
D.C., 1961), pp. 97-100.
Gillings, R. J. and Rigg, W. J. A.
"The Area of a Circle in Ancient Egypt," The Australian Journal of Science,
Vol. 32, No. 5 (Sydney, 1969), pp. 197-200.
Glanville, S. R. K. (editor)
The Legacy of Egypt, Oxford University Press, London, 1963. First printed
1942.
Glanville, S. R. K.
"The Mathematical leather roll in the British Museum," Journal of Egyptian
Archaeology, Vol. 13 (London, 1927), pp. 232-238.
Griffith, F. L.
"Egypt-Mathematics," Encyclopaedia Britannica, 11th edition, Vol. 9 (1910),
pp. 46-47.
 Bibliography 273

Griffith, F. L.
The Petrie Papyri: Hieratic Papyri from Kahun and Gurob (2 volumes), F. L.
Griffith, editor, University College, London, 1897.
Hayes, W. C.
"Ostracon No. 153," Ostraca and Name Stones from the Tomb of Sen-mut at
Thebes, Publication 15, The Metropolitan Museum of Art, New York, 1942.
Heath, Sir Thomas L.
The Thirteen Books of Euclid's Elements (2nd edition, 3 volumes), Cambridge
University Press, London, 1926.
Herodotus
History, Translated by George and Sir Henry Rawlinson, Everymans,
London, 1912.
Hogben, Lancelot
Mathematics for the Million, George Allen and Unwin, London, 1945. First
published 1936.
Hutton, Charles
A Philosophical and Mathematical Dictionary (2 volumes), F. C. and J. Riving-
ton, London, 1815.
Hutton, Charles
Recreations in Mathematics and Natural Philosophy. Translated from Montucla's
Edition of Ozanam, revised by Edward Riddle, Thomas Tegg, London,
1840.
Iverson, Erik
The Myth of Egypt and its Hieroglyphs in European Tradition, G. E. C. Gadd,
Copenhagen, 1961.
Jaffe, Bernard
Men of Science in America, Simon and Schuster, New York, 1944 (p. 93).
James, T. G. H.
Egyptian Sculptures, Fontana Uneseo Art Books, Collins, Milan, 1966.
Jourdain, Philip E. B.
"The Nature of Mathematics," in The World of Mathematics, Vol. 1, James
R. Newman, editor, Simon and Schuster, New York, 1956.
Kahun Papyrus (KP)
The British Museum, London. Listed under: Kahun Papyri IV.2, IV.3,
XLV.1, LV.3, and LV.4.
Kleppisch, K.
Die Cheopspyramide, ein Denkmal mathematischer Erkenntnis, Munich and Berlin,
1921.
Kline, Morris
Mathematics, A Cultural Approach, Addison-Wesley, Reading, Mass., 1962.
274 Bibliography

Life Science Library

Mathematics, by David Bergamini and the Editors of Life. Time-Life
International, the Netherlands, 1965.
Mansion, P.
"Sur une table du Papyrus Rhind," Annales de la Socilti Scienhfque de
Brv, elles, Vol. 12 (Louvain, 1888), p. 46.
Michigan Papyrus (Mich.P)
University of Michigan Collection, Vol. III, Papyri Nos. 146, 621. See also
L. C. Karpinski in his, Vol. 5 (Smithsonian Institute, Washington, D.C.,
1923), pp. 20-25.
Milliken, E. K.
Cradles of Western Civilization, Harrap, London, 1955.
Moscow Mathematical Papyrus (MMP)
Moscow Museum of Fine Arts. No. 4576 of the Inventory.
Nairtz, O.
"Die Cheopspyramide, ein Viertausendjahriges Rathsel," Prometheus, Vol.
17 (1906).
Neikes, H.
Der goldene Schnitt and die "Geheimnisse der Cheopspyramids," Cologne, 1907.
Neugebauer, O.
The Exact Sciences in Antiquity, Harper Torchbooks, Harper, New York,
1962. First published 1951.
Neugebauer, O.
Vorgriechische Afathematik (Vorlesungen), Springer, Berlin, 1934.
Neugebauer, O.
"Zur agyptischen Bruchrechnung," ZeitschnfifwdgyptischeSprache, Vol.64
(Leipzig, 1929), p. 48.
Neugebauer, O. and Parker, R. A.
Egyptian Astronomical Texts 1, Brown University Press and Lund Humphries,
London, 1960.
Newman, James R.
"The Rhind Papyrus," in T e World of Mathematics, Vol. 1, James R. New-
man, editor, Simon and Schuster, New York, 1956.
Paterson, W. E.
School Algebra, 3rd edition, Clarendon Press, Oxford, 1916.
Peet, T. Eric
"A Problem in Egyptian Geometry," Journal of Egyptian Archaeology, Vol.
17 (London, 1931), pp. 100-106.
Peet, T. Eric
"Mathematics in Ancient Egypt," Bulletin of the John Rylands Library, Vol.
15, No. 2 (Manchester, 1931).
 Bibliography 275

Peet, T. Eric
Review of Struve's Translation of the Moscow Mathematical Papyrus,
Journal of Egyptian Archaeology, Vol. 17 (London, 1931).
Peet, T. Eric
The Rhind Mathematical Papyrus, British Museum 10057 and 10058, London,
1923.
Petrie, W. M. F.
Illahun, Kahun and Gurob, London, 1891.
Petrie, W. M. F.
The Pyramids and Temples of Gizeh, London, 1883.
Recorde, Robert
The Grounde of Artes. Teachyng the Worke and Practice of Arithmetike, 1542. See
the Mathematical Gazette, Vol. 14, No. 195 (G. Bell and Sons, London, July,
1928), p. 196.
Reisner Papyri (RP)
The Museum of Fine Arts, Boston, Mass. Museum No. 38.2062.
Sanford, Vera
A Short History of Mathematics, Harrap, London, 1930.
Sarton, George
The Study of the History of Mathematics, Dover, New York, 1957. First pub-
lished 1936.
Schack-Schackenburg, H.
"Das Kleinere Fragment des Berliner Papyrus 6619," Zeitschrift fur Agyp-
tische Sprache, Vol. 40 (1902), p. 65.
Schack-Schackenburg, H.
"Der Berliner Papyrus 6619," Zeitschrift fur Agyptische Sprache, Vol. 38
(1900), p. 135.
Scott, A. and Hall, H. R.
"Laboratory Notes: Egyptian Leather Roll of the Seventeenth Century
B.C.," British Museum Quarterly, Vol. 2 (1927), p. 56.
Sethe, Kurt H.
"Von Zahlen and Zahlworten bei den Alten Agyptern," Schriften der wis-
senschaftlichen Gesellschaft in Strassburg, Part 25 (Trubner, Strassburg, 1916).
Sewell, J. W. S.
"The Calendars and Chronology," in The Legacy of Egypt, S. R. K. Glan-
ville, editor, Oxford University Press, London, 1963.
Simpson, W. K.
The Papyrus Reisner I (1963), II (1965), III (1969). The Museum of Fine
Arts, Boston, Mass.
Sloley, R. W.
" Science," in The Legacy of Egypt, S. R. K. Glanville, editor, Oxford Univer-
sity Press, London, 1963. First published 1942.
276 Bibliography

Smyth, Piazzi
Our Inheritance in the Great Pyramid, Daldy, Isbister, London, 1877. First ed.
1864; 2nd ed. 1874; 3rd ed. 1877; 4th ed. 1880; 5th ed., abridged, 1890.
Struik, Dirk J.
A Concise History of Mathematics, Dover, New York, 1948.
Struve, W. W.
"Mathematischer Papyrus des Staatlichen Museums der Schdnen Kunste
in Moskau," Quellen and Studien zur Geschichte der Mathematik, Part A, Vol. 1
(Berlin, 1930).
Thompson, D'Arcy W.
On Growth and Form, Vol. 1 (1951), Vol. 2 (1952), Cambridge University
Press, London.
Thompson, H.
"A Byzantine table of fractions," Ancient Egypt (London, 1914), p. 52.
Tsinserling, D. P.
"Geometry in Ancient Egypt," Bulletin de l'Acadlmie des Sciences de l'Union
des Republiques Soviltiques Socialistes, Vol. 19 (Leningrad, 1925), p. 541.
Turaev, B. A.
"The volume of the truncated pyramid in Egyptian mathematics,"
Ancient Egypt (London, 1917), p. 100.
Turnbull, H. W.
The Great Mathematicians, Methuen, London, 1951. First published 1929.
Van der Waerden, B. L.
"Die Entstehungsgeschichte der agyptischen Bruchrechnung," Quellen and
Studien zur Geschichte der Mathematik, Part B, Studien IV (Berlin, 1937-1938),
p. 359.
Van der Waerden, B. L.
Science Awakening, English translation by Arnold Dresden, Noordhoff,
Groningen, 1954.
Vogel, Kurt
"Eweitert die Lederrolle unsere Kenntniss agyptischer Mathematik?" in
Archiv fur Geschichte der Mathemalik, der Naturwissenuhaften and der Technik,
Julius Schuster, Berlin, 1929, p. 386.
Vogel, Kurt
"The Truncated Pyramid in Egyptian Mathematics," Journal of Egyptian
Archaeology, Vol. 16 (London, 1930), p. 242.
Vogel, Kurt
Vorgriechische Mathematik, Vol. 1, Vorgeschichte and Agypten, Schroedel,
Hannover, 1958.
Wheeler, N. F.
"Pyramids and their Purpose," Antiquity, 9 (Gloucester, England, 1935).
 Bibliography 277

Wilson, John A.
Signs and Wonders upon Pharaoh : A History of American Egyptology, The Univer.
sity of Chicago Press, Chicago, 1964.
Yeldham, F. A.
The Teaching of Arithmetic Through 400 Years, Harrap, London, 1936.
 INDEX

Abacus. II Assyriologists, 2
Abd-el-Rasoul, 246
Abundant numbers, 178n Babylon, 2 240
Abu Simbel, 90 Babylonian
Addition clay tablet (Plimpton 322), 1
examples of in hieratic. 10, 12, 14, 15 cuneiform script, 7
hieroglyph for, 6 direction of writing, 5 6
how the scribe did it, 13, 15 Babylonians. 91, 240
possible tables for. See Tables Baskets
scribes' ability in, 11 conventional shapes for Egyptian,
Aha (quantity) problems, 158,181 20
A'h-most (Ahmes), the scribe, 45 76 half-an-egg, 195
131-132,139,141-142.145,148 hieratic word nbt for, 194
falters with some fractions, 160 of grain, 90-hinu. 163
Akhenaton, Pharaoh, 90 Becker, 0., 48
Akhmim Papyrus (AP or AMP), 28, 91 Beer
Akkadian cuneiform script, 7 des-jugs of. 212
Akkadians, 91 division of. See Fractions
Aldersmith, H., 239 strength of. See Pesu
Alternando, 135-136 two types distinguished, 124, 125
Amenhotep t. King, 220 Bergamini, David, 6n
Amosis, King, 82 Berlin Papyrus 6619 (BP), 91, 161n,
Anklet, weight of, 212 216
Apries, Pharaoh, 207 hieroglyphic sign for square root,
Arabic, direction of writing, 5 216
Arabs, 2 ,47 Besha, 128
Archibald, R. C., 3n, 6n, 153, 217, 239 Bible. 237
Archimedes, 200 Birds, and direction in which hiero-
Areas glyphs face, 6
of circles, 139-140, 243 Bobynin, V. V., 48
of triangles and rectangles, 137, 243, Borchardt, L., 124-125
246-247 Boyer, Carl B., 139
of hemispheres. See Hemisphere Bruins, E. M., 48,.7-0-
Arithmetic progressions Bull, L. 3n, fin
examples in RMP, 243-244 Burroughs. Edgar Rice, 237
properties of, 170 Bushel, 147, 210
terms in descending order of Byzantine
magnitude, 170, 174 tables of fractions, 91
with common difference 5%.171 times, 16,47
Arura
areas of fields in, 217 Calendar, Egyptian
definition of, 209 only intelligent one in human
280 Index

Calendar, Egyptian (continued) Day, origin of division into 24 hours,

history, 235, 240 236
solar year of 36511{ days, 235 Deben, definition of, 212
Canon for decompositions in the Decan. 236
RMP, 49. See also Precepts; Recto Decimal system, 4 12
Table of the RMP Degree of civilisation, 3
Carter, Howard, 237 Deir el-Bahri, temple of, 86 246
Chace, Arnold B., 3 48 109, 128n, Demotic writings, 2,5
133, 141-142, 153, 168, 181, 183, Descartes, Reni, I
210.212-213,249 Des-jug
Champollion, Jean Francois, I of beer, 128, 212
Check marks, 18, 19, 20.22n quantity contained in, 213
Cheops (Khufu), pyramid of, 91, 231, See also Beer
237, 238 Dictionnaire Egyptienne, I
Chronology, 260-265 Digit. See Finger
Circles, area of. See Areas Dinostratus, 198, 200
Collignon, E., 48 Diophantus of Alexandria, 181
Coming forth (growing) period, Direction of ancient Egyptian writ-
235 ing. 6
Computers Dividendo, 135-186
54 Division
Use of series in, 19 by the clerk of the Reisner Paypri,
Copernicus, 235 219, 22fi
Coptic of loaves among workmen, 120
inscriptions, 86 of loaves in various proportions,
times, 16 171-173
Cottrell, Leonard, 239 of loaves, justice appears to be done,
Crum, W. E. 105
catalogue of, 29 of particular numbers: 184 by 8 19-
table of fractions, On 201060 by 12 and 159, 252
Cubit of nine loaves among ten men, 41
cubic, 163-165 105n
Greek, 201 a powerful technique with fractions,
Roman, 201 244
royal, 207-208 .220 by ten, 27,219.223
short, 202208 used to solve equations, 159
strip, 137. 209 "Do it thus" occurring at the end of a
superficial, analogy with superficial problem, 183
foot, 227-229 Dot indicating hieratic fractions,
Cuneiform writings, 1, 5 21
Cylinder, area of curved surface of, Doubling of Mo to obtain %, 84 85
194 Drachmas and fractions, 42
Dresden, Arnold, 191n, 232n
Dattel, 128
Davidson, D., 239 Edgar, J. and M., 239
Davidson, John, 239 Edwards, I. E. S., 207n, 239
 Index 281

Egyptian Mathematical Leather Roll of different pesus, 129, 130, 132-133;

(EMLR), 12, 22 23n, 39, 89, 106, using modern algebraic techniques
109 for. 134, 136
author of, 89 Eyth, M., 239
dimensions of, 89 Ezekiel, 207
early opinions of, 9fl
entries with like terms, I l l False assumption (false position),
error in line 17 of, 99 method of, 154, 157, 162, 243
grouping of equalities in, 95 Favarro, A., 48
photograph of, 94 "Find," the hieratic sign for, 9
reconstruction of, 92 Finger, Charles J., 17n
three- and four-term equalities, 114 Finger (fingerbreadth)
translation of, 93 definition of, 208
unrolling of, 89 inclusion of, in measurements at the
Egyptian standard tables, 95 110 dockyard workshops, 228-229
Egyptian symbols for numbers, 5 used in calculation of sekeds, 187
Egyptologists, 90-91.246 Flinders, Matthew, 91n
Egypt, map of, 266 Foundations of Egyptian mathe-
Eisenlohr. A., 48 210 matics. 3
Emmer, 128 Fractions
Ellen (cubits), 188 Egyptians unable to write the com-
Englebach, R., 191 mon fraction p/q, 234 and
Enigmatic writing in the RMP, 246 fractional divisions of bread and
Enlistees (workmen), 218-219, 224 beer, 126-127
Epagomenal ("yearend") days, 235 hieroglyph for, 24
Equals, the hieratic sign nearest to it. Horus-eye, 173n, 210-211
See "This is"; Hieratic characters meaning of, to scribes, 110
Equations, solution of multiplication of, 224
in MMP, 241 no two alike together, 95n, 105-107.
in RMP, 243 See also Egyptian Mathematical
Errors Leather Roll, grouping of equalities
in addition and subtraction, II possible hieroglyph for :Y4. 29n
infrequent in the Recto of the problems involving, genesis of, 105
RMP,1 scribe flounders in a maze of, 160
in multiplications involving finger- series of, in descending order of
breadths, 229 magnitude. 102n
found in the Reisner Papyri, 222 sign for, in hieratic, 21
Even number fractions, 71-73. See with very large denominators, 159
also Odd and even numbers in the
RMP Recto Gallons, 147, 210
Eves, Howard, 240 Gardiner, Alan
Exact Sciences in Antiquity, l In, 88n, Egyptian dictionary of, 165
235n. 240n, 241 Egyptian Grammar, 20TH
"Examples of proof." the hieratic on the Reisner papyri, 218n
form of this phrase, 9 Generators
Exchange of loaves definition of, 43,104
282 Index

Generators (continued) Head lay priest of the temple,

examples of, 108, 111-113 126
for pairs of fractions, 33k 4344 Head reader of the temple, 126
table of, 116 Heap, an unknown quantity, 157
Geometric progression Hekat (heqat)
unique properties of, 167 definition of, 210
with 7 the common ratio, 168, 244 measure of grain, 128n, 151, 163,
Gillain, 0., 48 165.2 253
Gillings. R. J., 176n, 177n, 181n Hemisphere, area of surface of, 188,
Glanville, S. R. K., 16n, 89-91.95, 194, 195. 196. 198, 200, 246
103. 145n, 241n Herodotus, 238_240
Golden section (sectio aurea), 238 Hesy-Ra. panel from tomb of, 83
Golenischev, V. S., 91n, 245. See also Hieratic
Moscow Mathematical Papyrus characters, 4-10
Gradient. See Seked integers from various papyri, 255-
Granaries, 2. 146-148 256
rectangular, 163-164 signs for khets and hekats, 141
volumes of, 243 sign "this is," meaning equals, 104,
Great Sphinx, 237 123
Greek variations in writing fractions, 259
papyrus, 217 variations in writing integers, 257-
tables of fractions, 91 258
times, 16 Hieroglyphs
Greeks, 2 47.232, 234, 240 decipherment of, I
Griffith, F. Ll., 48 73-74. 148n, 152, direction in which pictographs face,
156n, 162, 176, 216 fi
Grotefend, Georg F., I for numbers, 4 5
Grounde of Aries, 17 for two-thirds, 21
G rule invention of, 4
beginnings of, 9593 written from right to left, 5
examples of, 186, 215 Hilfszahlen, 81n
an exercise in, 42 Hinu, 163-165
extension of, 41 definition of, 210
further extension of, 44 Hogben, Lancelot, 48
not explicitly stated in papyri, 39 Hophra. See Apries, Pharaoh
probable use of, 41 Horus, Egyptian God, 210,235
Gunn, B. G., 189, 191, 210 Horus-eye fractions, 173n, 202. 204
206
Hall, H. R., 89, 90, 103 for grain, 210, 244, UL 249, 253
Hamblin, C. L., 5IIn hieroglyphic and hieratic signs for,
Handbreadth. See Palm 211
Harmonic mean, the concept of, 131, in terms of hinu, 250
244 use of, in the RMP, 243
Hatshepsut. Queen, 8fi Hultsch, F., 45
Haves, W. C.. 87n Hutton, Charles, 181-182
Hayt, 162-165, 208 Hyksos period, 18, 45
 Index 283

Ibscher, H., 218 table of common measures, 251

Illahun, temple of, 124, 231 Life Science Library: Mathematics, fi
Illegible signs, how indicated, 7 Lindgren, Harry, 146n, 191
Index laws of algebra, I Loaves of Egyptian bread, 212, 244
Introduction to the RMP by Logarithms. I
A'h-mose, 183 Loria, G., 4$
Inundation (sowing) period, 235 Lotus, 5
Inverse proportion, 244 Luckey, P., 191
Irrigation canals, 2 Luxor, temple of, 2 89
Isis, Egyptian Goddess, 210, 235
Isosceles triangle, area of, 139 Manning, H. P., 3n, 6n
Mansion, P., 48 70
James, T. G. H., 194 Mathematical papyri, 6
Jourdain, P. E. B., 16, 232 Measuring reed, or rod, 219-220
Justice appearing to be done in the Mental arithmetic of scribes
division of loaves, 105 evidence of, 19, 248.251-252
further illustrations of, 159-160, 186,
Kaltun Papyrus (KP), 39 91 199, 224. 226
portion of the RMP Recto included impossibility of, in Problem 71 of
in KP tv 2, 104, 176 the R1?P, 107
Problem KP iv 3 (contents of a Meret (side or height of a triangle),
granary), 148, 151 138n, 139
Problem KP t.v 4 (rectangular Meryre, the official, 82
granary). 162-165, 216 Michigan Papyrus (Mich. P.), 91
Problem KP LV 3 (mathematical Middle Kingdom. 124, 175
fragment), 156 Middle Kingdom hieratic, 7 18
six mathematical items, not all Milliken, E. K., $&n
penetrated yet, 176 Montucla, 181
Karnak, temple of, 2, 90, 237 Moscow Mathematical Papyrus
KDF-9 computer, SOn (MMP), 39 91, 104, 246-247
Khar Problem 6 reference to square root,
contents of granaries in, 147-148, 216
152 Problem 10, 194
definition of, 146, 151, 210 Problem 19, 157-158
values in, 163, 165 Problem 21, 131-132
Khet Mother Goose rhyme and Problem 72
measure of length, 137-139 of the RMP, 15,169
definition of, 209 "Mouth.- hieroglyphic word tp-r for,
hieratic signs for, 141 194
Kleppisch, K., 239 Multiplication
Kline, Morris, 175, 240 by continued doubling. 166
Kiibel, Jacob, 181-182 examples of. 17-19, 204 219, 221-
223
Largest number in the Recto of the form of, devised for the clerk of the
RMP, 49n, 61 dockyard workshops. 227 2311
Least common denominators, 78 81 of fractions. 22-23, 225
284 Index

Multiplication (continued) Papyri

hieratic signs for. 8 10 Greek, 209
of one-fifth by ten, 22 hieratic, 2
since the times of ancient Egypt, 16 Papyritic memo pad, 19, 22 107, 186,
in the Sixteenth and Seventeenth 203.219.248.251
Centuries, 17 Papyrus reed, 4
by ten, 13-14, 21, 23, 27n Paterson, W. E., 175n
Multi-term fractional equalities. 115 Peet, T. Eric, 1 In, 24 48 139, 145-
146,183. 189, 191, 194-198, 200
Nag'ed Deir 210, 213, 216, 232
dockyard workshops of, 218, 229 Pepys, Samuel, 17
Reisner papyri found at, 218 "Perpetual World Calendar," 235
Nairtz, 0., 239 Persians, 2
Napier, John, I Pesu
Nebuchadnezzar, King, 247 definition of, 128, 212
Neikes, H., 239 of bread and beer, 128. 240
Nephthys, Egyptian God, 235 exchange of loaves using modern
Neugebauer, 0., 1. 1In, 88. 90 103, algebra. 134-136
109. 194, 235, 240 loaves of different pesus exchanged,
Newman, J. R., 16, 232-233 129-133
Nile River meaning of, 128n
annual flooding of, 235 various modern terms for, 213
clay from, 189 Petrie Papyri, 148n, 152n, 156n,
papyrus reeds from. 2, 4 I62n
Noetling, F., 239 Petrie, William Matthew Flinders.
Number systems, readable, 7 91n
Numbers, very large, 219 Phrases, mathematical, in hieratic, 7
Pi (:), the Egyptian equivalent stint
Octagon for, 142, 197, 199, 238 240
inscribed in a square, 143 Plimpton 322 (Babylonian clay
regular, 142 tablet), I
Odd and even numbers in the Recto Potsherds, 86
of the RMP, 49. Precepts for the RMP Recto fractions,
One-third 49
of any number, 25-26 Prime equalities, 115
of mixed numbers, 35 Problems
of one and three, 34, 155n in completion. 81
of 30, finding two-thirds first, 184 RMP 28 on addition and subtrac-
Osiris, Egyptian God, 210, 235 tion. fin
Ostracon No. 153 from the tomb of Profiles in hieroglyphs, 6
Sen-mut. 8588 Proof, on the nature of, 145-146.
Ozanam, 181 232
Ptolemaic demotic writing, 7
Palette as scribe's insignia, 81, 82 Ptolemies, 209
Palm, 185, 207 Puttock, M. J., 22Q
Paper, 4 Pyramidiot, The Great, 239
 Index 285

Pyramids Red auxiliaries, 78 81, 85 87, 97 99

dimensions of the Giza group of, 102-103, 160-161, 251-252
185n Reference numbers for the red
dissection of, 189, 190 auxiliaries, 97 101, 103, 251-252
juel type of, 190 "Regular share," meaning of the
orientation of, 186 hieratic phrase, 173
Problems RMP56-6 0 and MMP 14 Reisner Papyri, 91, 218
on, 185-189 the seventeen divisions of, 219
right, 185 "Remainder," hieratic sign for, 8
scribes' drawings of, 186, 188 Remen
standard formula for a frustum, 189, definition of, 208
192, 193, 246 and double-remen, 208
truncated form of, 185, 187, 188, 193, use in doubling areas, 208
247 Reptiles, directions in which hiero-
volume of, 185, 189 glyphs face, 6
Pythagorean theorem Rhind, A. H., 89
in ancient Egypt, 242 Rhind Mathematical Papyrus
known to the Babylonians, I (RMP), 3 3n, 21, 24, 39 773 90 154
suggestions of, in equations, 161 details of the contents of, 243-245
unknown to Egyptians, 288. dimensions of, 45
mechanical arithmetic of, 91
Quadruple hekat, 151, 210 scribe's introduction to, 45
Quedet (weight of finger ring), 212 Rigor, conditions for, of a nonsym-
bolic proof, 231
Rameseum, 89 Romans, 42
Ramses. Pharaoh, 20 Rosetta Stone, 2
Rarity of scribal errors, 11. See also Ro
Errors definition of, 210
Reciprocal of is two-thirds, 21, 27 hieroglyphs and hieratic signs for,
157, 158 211
Recorde, Robert, 17 unit of measure, 47, 129, 202L 205
Rectangle, area of, 137 Royal cubit, 139
Recto Table of the RMP, 22, 34, 109,
243 Sachs, A., I
A'h-mosk falters on 2 = 95 68 Sacrificial bread, 242
canon or precepts for the values of, Saint Andrew's cross in multiplica-
44 tion, 17
derivation of some of the values of, Sanford, Vera, 181
98 Scales of notation
lack of scientific attitude claimed, scale of 4, 2311
232 scale of 7 227-228
method for 2 = 35 not standard, 79 Schack-Schackenburg. H., 48, 148,
most extensive of all extant tables, 153, 161, 176, 216
45 Scheffel, 210
simplest and best values, 42 Schoenia, 209
a whole new one possible, 69 Science, as the Egyptians knew it, 234
286 Index

Scott, Alexander, 89 90, 103 second degree, 162

Scribes Sirius (the Dog Star), heliacal rising
accuracy in multiplication, 222-223 of, 235-236
directions in which hieroglyphs Six-sevenths, as written in Egyptian
face, fi fractions, 21
handwriting of, 4 196 Sloley, R. W., 16, 145, 207n, 210
hieroglyphs for, 81 Smither, Paul C., 218n
never verbose, 7, 168 Smyth, Piazzi, 237
preoccupations with fractions, 104 Snakes, directions in which hiero-
of the temple. 126 glyphs face, 6
Seked Spelt-dates, 128
calculation of, 185 Square root
and cotangent of angle, 187, 212 approximations to, 215
definition of, related to gradient, hieratic word and signs for, 162-163,
212 176
of Giza pyramids, 187 hieroglyphic signs for, 216
of juel pyramids, 190-191 of particular numbers: li lSa, 161-
and slope of the sides of pyramids, 162; 164 in approximation. 217;
185, 187, 244 61+. 162
Semicircle possible tables for, 214-215
area of, 196, 198 simple cases of, 138, 143, 161-162,
circumference of, 197 164
Semicylinder, area of curved surface Squaring numbers, 214
of, 196-198, 200 the fraction %, 1%, 161-162;
Sen-mut, tomb of, 8681 the mixed number 83 U, M,48, 147
Series the numbers 2 and 4 188
arising from ordered Squaring the circle: perhaps
equalities, 119 the first of the circle squarers, 145
a special property of, for addition, Stammbruchen, 21
19 Stammbruchsummen, 90
used in multiplication, 166 Standard equality in fractions (the
Sesostris 1, Pharaoh, 218 equality 1L M 1/, a 1/4 of the
Setat (square khet), 38-39. 140-141, EAILR), 100-102,206,209,20
143 Standard fraction tables, 91, 110
definition of, 209 Stewart, Edgar, 239
Seth, Egyptian God, 210, 233 Struik, D. J., 16, 139, 232
Sethe, K. H., 210 Struve, W. W. 131n, 157n, 185, 194-
Seven, the number, 99 168-169. See 195,198.210.21,-,216,246-94 7
also Scales of notation. Subtraction
Sewell, J. W. S., 236n Egyptian method for, 11
Sexagesimal system, 4 examples of, in RINIP, 243
Shapes and Sections, 149n the hieroglyphic sign for, 6 10
Shaty, definition of, 212 Sumerian cuneiform script, 7
Simon, \I., 48 Summer (harvest) period, 23.5
Simpson, W. K., 218, 223 Sum to n terms of an arithmetic
Simultaneous equations, one of the progression, 175
 Index 287

Superficial cubit, and superficial foot, Three-term unit fractional equalities,

227-229 108
Sylvester, J. J., 48 "Total," hieratic sign for, 9 15
Transliteration, hieratic to hiero-
Tables glyphics, 7
of addition and subtraction in Triangle, area of, 138
hieratic, 11-14 Tsinserling. D. P., 246
elementary forms of, for fractions, Turajeff, B. A., 248
III Turnbull. H. W., 238
of fractions, 11, 15, 22, 32 Tutankhamen, Pharaoh, 237
of fractions of a cubit, 209, 225 Tuthmosis iv, Pharaoh, 82
from geometric progressions, 168 Twice-times table, 3. 234
in the RMP and elsewhere, 106 Two divided by
of squares, 162, 214-217 forty-five, 51, 52
of two-thirds of numbers, 153 multiples of 3 alternative methods,
for use of functionaries of the 77
granary, 242 thirteen, 71, 79-80
Tadpole, 5 thirty-five, scribal method disclosed,
Tannery, P., 48 77-7R
Tarzan, 237 Two-thirds, 20n, 21
Taxes, 3 Two-thirds of any number
Taylor, John, 239 examples and technique, 24
Tell el 'Amarna, clay tablets from, 91 in later times, 28-29
Temples, 2, 231, 237, 24fi rule for even fractions, 38
Teper (base or mouth of a triangle), rule for odd fractions, 29
138n, 139, 194 scribes' ability to find, 3 224, 244
Thaies, 2 Two-thirds of the fraction 1A69, 160
Thebes, Z 82 Two-thirds tables
"The doing as it occurs," the hieratic for any number, routine. 153
form of this phrase, 9 for integers and fractions, 25
Theory of numbers, faint beginnings possible method of constructing, 24
of, 96 114, 119 some justification for, 95.251
"Therefore," the hieratic form of this
word, 2 Unitary method, 130
"Think of a number" problems, 158, Unit fractions
181-183, 243 Egyptian, 21-23
This (Egyptian town), 218 used by Greeks two millenia later,
"This is," hieratic sign for 47
occurring in each of RMP Problems Usual reader of the temple, 126
1-6 and 87 times altogether, 123
occurs 80 times in EMLR, 92.104 Van der Waerden. B. L., 48 90n, 103,
Thomas, W. R., 191 191, 194, 213, 232
Thompson, D'Arcy W., 23$ Verso of the RMP, 22n
Thoth, Egyptian God, 210 Vertical writing, 6
Three-quarters as written in Vogel, Kurt, 48 81 n, 90 97n, 103,
Egyptian fractions, 21 128n, 142, 191, 213
288 Index

Volume
of a cylindrical granary, methods for,
146
of a rectangular prism, 189
rule for finding directly in khar, 148,
150
standard Egyptian rule for finding,
150

Wedyet flour, 128,130

"Working out," the hieratic form of
this phrase, 2
Writing
how it started, 4
instruments used for, 82-83
vertically downwards, 6

"You have correctly found it"

the hieratic signs for this phrase, 10
as used in the MMP. 157

Zero, indicated by a blank space, 22$

A CATALOG OF SELECTED
DOVER BOOKS
IN ALL FIELDS OF INTEREST
 A CATALOG OF SELECTED DOVER
BOOKS IN ALL FIELDS OF INTEREST
CONCERNING THE SPIRITUAL IN ART, Wassily Kandinsky. Pioneering work
by father of abstract art. Thoughts on color theory, nature of art. Analysis of earlier
masters. 12 illustrations. 80pp. of text. 5% x 8Y,. 23411-8 Pa. $4.95

ANIMALS: 1,419 Copyright-Free Illustrations of Mammals, Birds, Fish, Insects, etc.,

Jim Harter (ed.). Clear wood engravings present, in extremely lifelike poses, over
1,000 species of animals. One of the most extensive pictorial sourcebooks of its kind.
Captions. Index. 284pp. 9 x 12. 23766-4 Pa. $ 14.95

CELTIC ART: The Methods of Construction, George Bain. Simple geometric tech-
niques for making Celtic interlacements, spirals, Kells-type initials, animals, humans,
etc. Over 500 illustrations. 160pp. 9 x 12. (USO) 22923-8 Pa. $9.95

AN ATLAS OF ANATOMY FOR ARTISTS, Fritz Schider. Most thorough refer-

ence work on art anatomy in the world. Hundreds of illustrations, including selec-
tions from works by Vesalius, Leonardo, Goya, Ingres, Michelangelo, others. 593
illustrations. 192pp. Iii x 10%. 20241.0 Pa. $9.95

CELTIC HAND STROKE-BY-STROKE (Irish Half-Uncial from "The Book of

Kells"): An Arthur Baker Calligraphy Manual, Arthur Baker. Complete guide to cre-
ating each letter of the alphabet in distinctive Celtic manner. Covers hand position,
strokes, pens, inks, paper. more. Illustrated. 48pp. 81A x 11. 24336-2 Pa. $3.95

EASY ORIGAMI,John Montroll. Charming collection of 32 projects (hat, cup, pel-

ican, piano, swan, many more) specially designed for the novice origami hobbyist.
Clearly illustrated easy-to-follow instructions insure that even beginning paper-
crafters will achieve successful results. -48pp. 84 x 11. 27298-2 Pa. $3.50

THE COMPLETE BOOK OF BIRDHOUSE CONSTRUCTION FOR WOOD-

WORKERS, Scott D. Campbell. Detailed instructions, illustrations, tables. Also data
on bird habitat and instinct patterns. Bibliography. 3 tables. 63 illustrations in 15 fig-
ures. 48pp. 5'% x 8%. 24407-5 Pa. $2.50

BLOOMINGDALE'S ILLUSTRATED 1886 CATALOG: Fashions, Dry Goods

and Housewares, Bloomingdale Brothers. Famed merchants' extremely rare catalog
depicting about 1,700 products: clothing, housewares, firearms, dry goods, jewelry,
more. Invaluable for dating, identifying vintage items. Also, copyright-free graphics
for artists, designers. Co-published with Henry Ford Museum & Greenfield Village.
160pp. 8% x 11. 25780-0 Pa. $10.95

HISTORIC COSTUME IN PICTURES, Braun & Schneider. Over 1,450 costumed

figures in clearly detailed engravings-from dawn of civilization to end of 19th cen-
tury. Captions. Many folk costumes. 256pp. 8% x 11 %. 23150-X Pa. $12.95
 CATALOG OF DOVER BOOKS

STICKLEY CRAFTSMAN FURNITURE CATALOGS, Gustav Stickley and L &

J. G. Stickley. Beautiful, functional furniture in two authentic catalogs from 1910. 594
illustrations, including 27 photos, show settles, rockers, armchairs, reclining chairs,
bookcases, desks, tables. 183pp. 6Y x 9'.G. 23838-5 Pa. $11.95
AMERICAN LOCOMOTIVES IN HISTORIC PHOTOGRAPHS: 1858 to 1949,
Ron Ziel (ed.). A rare collection of 126 meticulously detailed official photographs,
called "builder portraits," of American locomotives that majestically chronicle the
rise of steam locomotive power in America. Introduction. Detailed captions. xi +
I29pp. 9 x 12. 27393-8 Pa. $ 13.95
AMERICA'S LIGHTHOUSES: An Illustrated History, Francis Ross Holland, Jr.
Delightfully written, profusely illustrated fact-filled survey of over 200 American light-
houses since 1716. History, anecdotes, technological advances, more. 240pp. 8 x 10%.
25576-X Pa. $12.95
TOWARDS A NEW ARCHITECTURE, Le Corbusier. Pioneering manifesto by
founder of "International School." Technical and aesthetic theories, views of indus-
try, economics, relation of form to function, "mass-production split" and much more.
Profusely illustrated. 320pp. 6% x 9'4. (USO) 25023-7 Pa. $9.95
HOW THE OTHER HALF LIVES,Jacob Riis. Famous journalistic record, expos-
ing poverty and degradation of New York slums around 1900, by major social
reformer. 100 striking and influential photographs. 233pp. 10 x 7A.
22012-5 Pa. S 11.95
FRUIT KEY AND TWIG KEY TO TREES AND SHRUBS, William M. Harlow.
One of the handiest and most widely used identification aids. Fruit key covers 120
deciduous and evergreen species; twig key 160 deciduous species. Easily used. Over
300 photographs. 126pp. 5% x 8`n. 20511-8 Pa. $3.95
COMMON BIRD SONGS, Dr. Donald J. Borror. Songs of 60 most common U.S.
birds: robins, sparrows, cardinals, bluejays, finches, more-arranged in order of
increasing complexity. Up to 9 variations of songs of each species.
Cassette and manual 99911-4 $8.95
ORCHIDS AS HOUSE PLANTS, Rebecca Tyson Northen. Grow cattleyas and
many other kinds of orchids-in a window, in a case, or under artificial light. 63 illus-
trations. 148pp. 5% x 814. 23261-1 Pa. $5.95
MONSTER MAZES, Dave Phillips. Masterful mazes at four levels of difficulty.
Avoid deadly perils and evil creatures to find magical treasures. Solutions for all 32
exciting illustrated puzzles. 48pp. 8'4 x If. 26005-4 Pa. $2.95
MOZART'S DON GIOVANNI (DOVER OPERA LIBRETTO SERIES),
Wolfgang Amadeus Mozart. Introduced and translated by Ellen H. Bleiler. Standard
Italian libretto, with complete English translation. Convenient and thoroughly
portable-an ideal companion for reading along with a recording or the performance
itself. Introduction. List of characters. Plot summary. 121pp. 514 x 8%.
24944-1 Pa. $3.95
TECHNICAL MANUAL AND DICTIONARY OF CLASSICAL BALLET, Gail
Grant. Defines, explains, comments on steps, movements, poses and concepts. 15-
page pictorial section. Basic book for student, viewer. 127pp. 5%% x 81h.
21843-0 Pa. $4.95
 CATALOG OF DOVER BOOKS

THE CLARINET AND CLARINET PLAYING, David Pino. Lively, comprehen-

sive work features suggestions about technique, musicianship, and musical interpre-
tation, as well as guidelines for teaching, making your own reeds, and preparing for
public performance. Includes an intriguing look at clarinet history. "A godsend,"
The Clarinet, Journal of the International Clarinet Society. Appendixes. 7 illus.
320pp. 5% x 814. 40270-3 Pa. $9.95
HOLLYWOOD GLAMOR PORTRAITS,John Kobal (ed.). 145 photos from 1926-
49. Harlow, Gable, Bogart, Bacall; 94 stars in all. Full background on photographers,
technical aspects. 160pp. 876 x 11 %. 23352-9 Pa. $12.95
THE ANNOTATED CASEY AT THE BAT: A Collection of Ballads about the
Mighty Casey/Third, Revised Edition, Martin Gardner (ed.). Amusing sequels and
parodies of one of America's best-loved poems: Casey's Revenge, Why Casey
Whiffed, Casey's Sister at the Bat, others. 256pp. 5% x 81/1. 28598-7 Pa. $8.95
THE RAVEN AND OTHER FAVORITE POEMS, Edgar Allan Poe. Over 40 of
the author's most memorable poems: "The Bells," "Ulalume," "Israfel," "To Helen,"
"The Conqueror Worm," "Eldorado," "Annabel Lee," many more. Alphabetic lists of
titles and first lines. 64pp. 57fr, x 8%. 26685-0 Pa. $1.00
PERSONAL MEMOIRS OF U. S. GRANT, Ulysses Simpson Grant. Intelligent,
deeply moving firsthand account of Civil War campaigns, considered by many the
finest military memoirs ever written. Includes letters, historic photographs, maps and
more. 528pp. 61A x 91l. 28587-1 Pa. $12.95
ANCIENT EGYPTIAN MATERIALS AND INDUSTRIES, A. Lucas and J.
Harris. Fascinating, comprehensive, thoroughly documented text describes this
ancient civilization's vast resources and the processes that incorporated them in daily
life, including the use of animal products, building materials, cosmetics, perfumes
and incense, fibers, glazed ware, glass and its manufacture, materials used in the
mummification process, and much more. 544pp. 6'A x 9'/.. (USO)
40446-3 Pa. $16.95
RUSSIAN STORIES/PYCCKNE PACCKA3bl: A Dual-Language Book, edited by
Gleb Struve. Twelve tales by such masters as Chekhov, Tolstoy, Dostoevsky, Pushkin,
others. Excellent word-for-word English translations on facing pages, plus teaching
and study aids, Russian/English vocabulary, biographical/critical introductions,
more. 416pp. 53k x 8'6. 26244-8 Pa. $9.95
PHILADELPHIA THEN AND NOW: 60 Sites Photographed in the Past and
Present, Kenneth Finkel and Susan Oyama. Rare photographs of City Hall, Logan
Square, Independence Hall, Betsy Ross House, other landmarks juxtaposed with
contemporary views. Captures changing face of historic city. Introduction. Captions.
128pp. 8'd x 11. 25790-8 Pa. $9.95
AIA ARCHITECTURAL GUIDE TO NASSAU AND SUFFOLK COUNTIES,
LONG ISLAND, The American Institute of Architects, Long Island Chapter, and
the Society for the Preservation of Long Island Antiquities. Comprehensive, well-
researched and generously illustrated volume brings to life over three centuries of
Long Island's great architectural heritage. More than 240 photographs with authori-
tative, extensively detailed captions. 176pp. 8% x 11. 26946-9 Pa. $14.95
NORTH AMERICAN INDIAN LIFE: Customs and Traditions of 23 Tribes, Elsie
Clews Parsons (ed.). 27 fictionalized essays by noted anthropologists examine reli-
gion, customs, government, additional facets of life among the Winnebago, Crow,
Zuni. Eskimo, other tribes. 480pp. 64 x 9'4. 2737-6 Pa. $10.95
 CATALOG OF DOVER BOOKS

FRANK LLOYD WRIGHT'S DANA HOUSE, Donald Hoffmann. Pictorial essay

of residential masterpiece with over 160 interior and exterior photos, plans, eleva-
tions, sketches and studies. 128pp. 9'/, x 1034. 29120-0 Pa. $12.95

THE MALE AND FEMALE FIGURE IN MOTION: 60 Classic Photographic

Sequences, Eadweard Muybridge. 60 true-action photographs of men and women
walking, running, climbing, bending, turning, etc., reproduced from rare 19th-centu-
ry masterpiece. vi + 121pp. 9 x 12. 24745-7 Pa. $10.95

1001 QUESTIONS ANSWERED ABOUT THE SEASHORE, N. J. Berrill and

Jacquelyn Berrill. Queries answered about dolphins, sea snails, sponges, starfish, fish-
es, shore birds, many others. Covers appearance, breeding, growth, feeding, much
more. 305pp. 5'!, x 814. 23366-9 Pa. $9.95
ATTRACTING BIRDS TO YOUR YARD, William J. Weber. Easy-to-follow guide
offers advice on how to attract the greatest diversity of birds: birdhouses, feeders,
water and waterers, much more. 96pp. 5'/« x 8/4. 28927-3 Pa. $2.50

MEDICINAL AND OTHER USES OF NORTH AMERICAN PLANTS: A

Historical Survey with Special Reference to the Eastern Indian Tribes, Charlotte
Erichsen-Brown. Chronological historical citations document 500 years of usage of
plants, trees, shrubs native to eastern Canada, northeastern U.S. Also complete iden-
tifying information. 343 illustrations. 544pp. 6'/, x 9'b. 25951-X Pa. $12.95

STORYBOOK MAZES, Dave Phillips. 23 stories and mazes on two-page spreads:

Wizard of Oz, Treasure Island, Robin Hood, etc. Solutions. 64pp. 8'b x 11.
23628-5 Pa. $2.95
AMERICAN NEGRO SONGS: 230 Folk Songs and Spirituals, Religious and
Secular, John W. Work. This authoritative study traces the African influences of songs
sung and played by black Americans at work, in church, and as entertainment. The
author discusses the lyric significance of such songs as "Swing Low, Sweet Chariot,"
"John Henry," and others and offers the words and music for 230 songs.
Bibliography. Index of Song Titles. 272pp. 61/1 x 9'/,. 40271-1 Pa. $9.95
MOVIE-STAR PORTRAITS OF THE FORTIES, John Kobal (ed.). 163 glamor,
studio photos of 106 stars of the 1940s: Rita Hayworth, Ava Gardner, Marion
Brando, Clark Gable, many more. 176pp. 8% x 11 %. 23546-7 Pa. S 14.95

BENCHLEY LOST AND FOUND, Robert Benchley. Finest humor from early 30s,
about pet peeves, child psychologists, post office and others. Mostly unavailable else-
where. 73 illustrations by Peter Amo and others. 183pp. 5% x 89.22410-4 Pa. $6.95
YEKL and THE IMPORTED BRIDEGROOM AND OTHER STORIES OF
YIDDISH NEW YORK, Abraham Cahan. Film Hester Street based on YekI (1896).
Novel, other stories among first aboutJewish immigrants on N.Y.'s East Side. 240pp.
51 x 8'G. 22427-9 Pa. $6.95
SELECTED POEMS, Walt Whitman. Generous sampling from Leaves of Grass.
Twenty-four poems include "I Hear America Singing," "Song of the Open Road," "I
Sing the Body Electric," "When Lilacs Last in the Dooryard Bloom'd," "0 Captain!
My Captain!"-all reprinted from an authoritative edition. Lists of titles and first lines.
I28pp. 5i x 814. 26878-0 Pa. $1.00
 CATALOG OF DOVER BOOKS

THE BEST TALES OF HOFFMANN, E. T. A. Hoffmann. 10 of Hoffmann's most

important stories: "Nutcracker and the King of Mice," "The Golden Flowerpot," etc.
458pp. 5% x 8Y. 21793-0 Pa. $9.95
FROM FETISH TO GOD IN ANCIENT EGYPT, E. A. Wallis Budge. Rich
detailed survey of Egyptian conception of "God" and gods, magic, cult of animals,
Osiris, more. Also, superb English translations of hymns and legends. 240 illustra-
tions. 545pp. 5% x 816. 25803-3 Pa. $13.95
FRENCH STORIES/CONTES FRANcAIS: A Dual-Language Book, Wallace
Fowlie. Ten stories by French masters, Voltaire to Camus: "Micromegas" by Voltaire;
"The Atheist's Mass" by Balzac; "Minuet" by de Maupassant; "The Guest" by
Camus, six more. Excellent English translations on facing pages. Also French-English
vocabulary list, exercises, more. 352pp. 51 x 8%. 26443-2 Pa. $9.95
CHICAGO AT THE TURN OF THE CENTURY IN PHOTOGRAPHS: 122
Historic Views from the Collections of the Chicago Historical Society, Lang A.
Viskochil. Rare large-format prints offer detailed views of City Hall, State Street, the
Loop, Hull House, Union Station, many other landmarks, circa 1904-1913.
Introduction. Captions. Maps. 144pp. 9% x 12%. 24656-6 Pa. $12.95
OLD BROOKLYN IN EARLY PHOTOGRAPHS, 1865-1929, William Lee
Younger. Luna Park, Gravesend race track, construction of Grand Army Plaza, mov-
ing of Hotel Brighton, etc. 157 previously unpublished photographs. 165pp. Bib x 11'.
23587-4 Pa. $13.95
THE MYTHS OF THE NORTH AMERICAN INDIANS, Lewis Spence. Rich
anthology of the myths and legends of the Algonquins, Iroquois, Pawnees and Sioux,
prefaced by an extensive historical and ethnological commentary. 36 illustrations.
480pp. 51 x M. 25967-6 Pa. $10.95
AN ENCYCLOPEDIA OF BATTLES: Accounts of Over 1,560 Battles from 1479
sc to the Present, David Eggenberger. Essential details of every major battle in
recorded history from the first battle of Megiddo in 1479 a.c. to Grenada in 1984. List
of Battle Maps. New Appendix covering the years 1967-1984. Index. 99 illustrations.
544pp. 6% x 9'4. 24913-1 Pa. $16.95
SAILING ALONE AROUND THE WORLD, Captain Joshua Slocum. First man
to sail around the world, alone, in small boat. One of great feats of seamanship told
in delightful manner. 67 illustrations. 294pp. 51 x 8'%. 20326-3 Pa. $6.95
ANARCHISM AND OTHER ESSAYS, Emma Goldman. Powerful, penetrating,
prophetic essays on direct action, role of minorities, prison reform, puritan
hypocrisy, violence, etc. 271pp. 51 x 8%. 22484-8 Pa. $795
MYTHS OF THE HINDUS AND BUDDHISTS, Ananda K. Coomaraswamy and
Sister Nivedita. Great stories of the epics; deeds of Krishna, Shiva, taken from
puranas, Vedas, folk tales; etc. 32 illustrations. 400pp. 5% x 8%. 21759-0 Pa. $12.95
THE TRAUMA OF BIRTH, Otto Rank. Rank's controversial thesis that anxiety
neurosis is caused by profound psychological trauma which occurs at birth. 256pp.
5% x 8Y. 27974-X Pa. $7.95

A THEOLOGICO-POLITICAL TREATISE, Benedict Spinoza. Also contains

unfinished Political Treatise. Great classic on religious liberty, theory of government
on common consent. R. Elwes translation. Total of 421pp. 5% x 8%. 20249-6 Pa. $9.95
 CATALOG OF DOVER BOOKS

MY BONDAGE AND MY FREEDOM, Frederick Douglass. Born a slave,

Douglass became outspoken force in antislavery movement. The best of Douglass'
autobiographies. Graphic description of slave life. 464pp. 5% x 8Y,. 22457-0 Pa. 58.95
FOLLOWING THE EQUATOR: A Journey Around the World, Mark Twain.
Fascinating humorous account of 1897 voyage to Hawaii, Australia, India, New
Zealand, etc. Ironic, bemused reports on peoples, customs, climate, flora and fauna,
politics, much more. 197 illustrations. 720pp. 5% x 26113-1 Pa. $15.95
THE PEOPLE CALLED SHAKERS, Edward D. Andrews. Definitive study of
Shakers: origins, beliefs, practices, dances, social organization, furniture and crafts,
etc. 33 illustrations. 351pp. 5% x 8'h. 21081-2 Pa. $8.95
THE MYTHS OF GREECE AND ROME, H. A. Guerber. A classic of mythology,
generously illustrated, long prized for its simple, graphic, accurate retelling of the
principal myths of Greece and Rome, and for its commentary on their origins and
significance. With 64 illustrations by Michelangelo, Raphael, Titian, Rubens,
Canova, Bernini and others. 480pp. 5% x 8'b. 27584-1 Pa. $9.95
PSYCHOLOGY OF MUSIC, Carl E. Seashore. Classic work discusses music as a
medium from psychological viewpoint. Clear treatment of physical acoustics, audi-
tory apparatus, sound perception, development of musical skills, nature of musical
feeling, host of other topics. 88 figures. 408pp. 5'% x 8Y,. 21851-1 Pa. $11.95
THE PHILOSOPHY OF HISTORY, Georg W. Hegel. Great classic of Western
thought develops concept that history is not chance but rational process, the evolu-
tion of freedom. 457pp. 5% x 8Y,. 20112-0 Pa. $9.95
THE BOOK OF TEA, Kakuzo Okakura. Minor classic of the Orient: entertaining,
charming explanation, interpretation of traditional Japanese culture in terms of tea
ceremony. 94pp. 5% x 8'b. 20070-1 Pa. $3.95
LIFE IN ANCIENT EGYPT, Adolf Erman. Fullest, most thorough, detailed older
account with much not in more recent books, domestic life, religion, magic, medi-
cine, commerce, much more. Many illustrations reproduce tomb paintings, carvings,
hieroglyphs, etc. 597pp. 514 x 8'1.. 22632-8 Pa. $12.95
SUNDIALS, Their Theory and Construction, Albert Waugh. Far and away the best,
most thorough coverage of ideas, mathematics concerned, types, construction,
adjusting anywhere. Simple, nontechnical treatment allows even children to build
several of these dials. Over 100 illustrations. 230pp. 5% x 81l. 22947-5 Pa. $8.95
THEORETICAL HYDRODYNAMICS, L. M. Milne-Thomson. Classic exposition
of the mathematical theory of fluid motion, applicable to both hydrodynamics and
aerodynamics. Over 600 exercises. 768pp. 6'/w x 9'/4. 68970-0 Pa. $20.95
SONGS OF EXPERIENCE: Facsimile Reproduction with 26 Plates in Full Color,
William Blake. 26 full-color plates from a rare 1826 edition. Includes "TheTyger,"
"London," "Holy Thursday," and other poems. Printed text of poems. 48pp. 5K x 7.
24636-1 Pa. $4.95
OLD-TIME VIGNETTES IN FULL COLOR, Carol Belanger Grafton (ed.). Over
390 charming, often sentimental illustrations, selected from archives of Victorian
graphics-pretty women posing, children playing, food, flowers, kittens and puppies,
smiling cherubs, birds and butterflies, much more. All copyright-free. 48pp. 914 x 12' .
27269-9 Pa. S7.95
 CATALOG OFDOVER BOOKS

PERSPECTIVE FOR ARTISTS, Rex Vicat Cole. Depth, perspective of sky and sea,
shadows, much more, not usually covered. 391 diagrams, 81 reproductions of draw-
ings and paintings. 279pp. 51 x 84. 22487-2 Pa. $7.95

DRAWING THE LIVING FIGURE, Joseph Sheppard. Innovative approach to

artistic anatomy focuses on specifics of surface anatomy, rather than muscles and
bones. Over 170 drawings of live models in front, back and side views, and in wide-
ly varying poses. Accompanying diagrams. 177 illustrations. Introduction. Index.
144pp. 81 x 11 4. 26723-7 Pa. $8.95
GOTHIC AND OLD ENGLISH ALPHABETS: 100 Complete Fonts, Dan X. Solo.
Add power, elegance to posters, signs, other graphics with 100 stunning copyright-
free alphabets: Blackstone, Dolbey, Germania, 97 more-including many lower-case,
numerals, punctuation marks. 104pp. 816 x 11. 24695-7 Pa. $8.95
HOW TO DO BEADWORK, Mary White. Fundamental book on craft from simple
projects to five-bead chains and woven works. 106 illustrations. 142pp. 5-% x 8.
20697-1 Pa. $5.95
THE BOOK OF WOOD CARVING, Charles Marshall Sayers. Finest book for
beginners discusses fundamentals and offers 34 designs. "Absolutely first rate ... well
thought out and well executed."-E. J. Tangerman. 118pp. 7% x 1016.
23654-4 Pa. $7.95
ILLUSTRATED CATALOG OF CIVIL WAR MILITARY GOODS: Union Army
Weapons, Insignia, Uniform Accessories, and Other Equipment, Schuyler, Hartley,
and Graham. Rare, profusely illustrated 1846 catalog includes Union Army uniform
and dress regulations, arms and ammunition, coats, insignia, flags, swords, rifles, etc.
226 illustrations. 160pp. 9 x 12. 24939-5 Pa. $10.95
WOMEN'S FASHIONS OF THE EARLY 1900s: An Unabridged Republication of
"New York Fashions, 1909," National Cloak & Suit Co. Rare catalog of mail-order
fashions documents women's and children's clothing styles shortly after the turn of
the century. Captions offer full descriptions, prices. Invaluable resource for fashion,
costume historians. Approximately 725 illustrations. 128pp. 8Y x 11%.
27276-1 Pa. $11.95

THE 1912 AND 1915 GUSTAV STICKLEY FURNITURE CATALOGS, Gustav

Stickley. With over 200 detailed illustrations and descriptions, these two catalogs are
essential reading and reference materials and identification guides for Stickley furni-
ture. Captions cite materials, dimensions and prices. 112pp. 6'l x 9'%.
26676-1 Pa. $9.95

EARLY AMERICAN LOCOMOTIVES, John H. White, Jr. Finest locomotive

engravings from early 19th century: historical (1804-74), main-line (after 1870), spe-
cial, foreign, etc. 147 plates. 142pp. 11% x 8Y,. 22772-3 Pa. $10.95
THE TALL SHIPS OF TODAY IN PHOTOGRAPHS, Frank O. Braynard.
Lavishly illustrated tribute to nearly 100 majestic contemporary sailing vessels:
Amerigo Vespucci, Clearwater, Constitution, Eagle, Mayflower, Sea Cloud, Victory,
many more. Authoritative captions provide statistics, background on each ship. 190
black-and-white photographs and illustrations. Introduction. 128pp. 84 x 11X4.
27163-3 Pa. $14.95
 CATALOG OF DOVER BOOKS

PIANO TUNING, J. Cree Fischer. Clearest, best book for beginner, amateur.
Simple repairs, raising dropped notes, tuning by easy method of flattened fifths. No
previous skills needed. 4 illustrations. 2Olpp. 5% x 8'/. 23267-0 Pa. $6.95
HINTS TO SINGERS, Lillian Nordica. Selecting the right teacher, developing con-
fidence, overcoming stage fright, and many other important skills receive thoughtful
discussion in this indispensible guide, written by a world-famous diva of four
decades' experience. 96pp. 5'/6 x 8'/:. 40094-8 Pa. $4.95
THE COMPLETE NONSENSE OF EDWARD LEAR, Edward Lear. All nonsense
limericks, zany alphabets, Owl and Pussycat, songs, nonsense botany, etc., illustrated
by Lear. Total of 320pp. 5`% x 8'/!. (USO) 20167-8 Pa. $7.95
VICTORIAN PARLOUR POETRY: An Annotated Anthology, Michael R. Turner.
117 gems by Longfellow, Tennyson, Browning, many lesser-known poets. "The
Village Blacksmith," "Curfew Must Not Ring Tonight," "Only a Baby Small," dozens
more, often difficult to find elsewhere. Index of poets, titles, first lines. xxiii + 325pp.
5% x 8!,. 27044-0 Pa. $8.95
DUBLINERS, James Joyce. Fifteen stories offer vivid, tightly focused observations
of the lives of Dublin's poorer classes. At least one, "The Dead," is considered a mas-
terpiece. Reprinted complete and unabridged from standard edition. 160pp. 5'/in x 8%.
26870-5 Pa. $1.00

GREAT WEIRD TALES: 14 Stories by Lovecraft, Blackwood, Machen and Others,

S. T. Joshi (ed.). 14 spellbinding tales, including "The Sin Eater," by Fiona McLeod,
"The Eye Above the Mantel," by Frank Belknap Long, as well as renowned works
by R. H. Barlow, Lord Dunsany, Arthur Machen, W. C. Morrow and eight other
masters of the genre. 256pp. 53, x 8'/. (USO) 40436-6 Pa. $8.95

THE BOOK OF THE SACRED MAGIC OF ABRAMELIN THE MACE, trans-

lated by S. MacGregor Mathers. Medieval manuscript of ceremonial magic. Basic
document in Aleister Crowley, Golden Dawn groups. 268pp. 5% x 8`/,.
23211-5 Pa. $9.95

NEW RUSSIAN-ENGLISH AND ENGLISH-RUSSIAN DICTIONARY, M. A.

O'Brien. This is a remarkably handy Russian dictionary, containing a surprising
amount of information, including over 70,000 entries. 366pp. 4% x 6%.
20208-9 Pa. $10.95

HISTORIC HOMES OF THE AMERICAN PRESIDENTS, Second, Revised

Edition, Irvin Haas. A traveler's guide to American Presidential homes, most open
to the public, depicting and describing homes occupied by every American President
from George Washington to George Bush. With visiting hours, admission charges,
travel routes. 175 photographs. Index. 160pp. 8% x 11. 26751-2 Pa. $11.95

NEW YORK IN THE FORTIES, Andreas Feininger. 162 brilliant photographs by

the well-known photographer, formerly with Life magazine. Commuters, shoppers,
Times Square at night, much else from city at its peak. Captions by John von Hartz.
181pp. 9Y, x 10l.. 23585-8 Pa. $13.95
INDIAN SIGN LANGUAGE, William Tomkins. Over 525 signs developed by
Sioux and other tribes. Written instructions and diagrams. Also 290 pictographs.
I 1 I pp. 61.G x 94. 22029-X Pa. $3.95
 CATALOG OFDOVER BOOKS

PHOTOGRAPHIC SKETCHBOOK OF THE CIVIL WAR, Alexander Gardner.

100 photos taken on field during the Civil War. Famous shots of Manassas Harper's
Ferry, Lincoln, Richmond, slave pens, etc. 244pp. 10% x 8%. 22731-6 Pa. $10.95

FIVE ACRES AND INDEPENDENCE, Maurice G. Kains. Great back-to-the-land

classic explains basics of self-sufficient farming. The one book to get. 95 illustrations.
397pp. 5 x 816. 20974-1 Pa. $7.95

SONGS OF EASTERN BIRDS, Dr. DonaldJ. Borror. Songs and calls of 60 species
most common to eastern U.S.: warblers, woodpeckers, flycatchers, thrushes, larks,
many more in high-quality recording. Cassette and manual 99912-2 $9.95

A MODERN HERBAL, Margaret Grieve. Much the fullest, most exact, most useful
compilation of herbal material. Gigantic alphabetical encyclopedia, from aconite to
zedoary, gives botanical information, medical properties, folklore, economic uses,
much else. Indispensable to serious reader. 161 illustrations. 888pp. 6'h x 914. 2-vol.
set. (USO) Vol. 1: 22798-7 Pa. $9.95
Vol. 11: 22799-5 Pa. $9.95

HIDDEN TREASURE MAZE BOOK, Dave Phillips. Solve 34 challenging mazes

accompanied by heroic tales of adventure. Evil dragons, people-eating plants, blood-
thirsty giants, many more dangerous adversaries lurk at every twist and turn. 34
mazes, stories, solutions. 48pp. 8'% x 11. 24566-7 Pa. $2.95

LETTERS OF W. A. MOZART, Wolfgang A. Mozart. Remarkable letters show

bawdy mvit, humor, imagination, musical insights, contemporary musical world;
includes some letters from Leopold Mozart. 276pp. 5% x 811.. 22859-2 Pa. S795

BASIC PRINCIPLES OF CLASSICAL BALLET, Agrippina Vaganova. Great

Russian theoretician, teacher explains methods for teaching classical ballet. 118 illus-
trations. 175pp. 3% x 81h. 22036-2 Pa. $5.95

THE JUMPING FROG, Mark Twain. Revenge edition. The original story of The
Celebrated Jumping Frog of Calaveras County, a hapless French translation, and
Twain's hilarious "retranslation" from the French. 12 illustrations. 66pp. 5% x 8%.
22686-7 Pa. $3.95

BEST REMEMBERED POEMS, Martin Gardner (ed.). The 126 poems in this
superb collection of 19th- and 20th-century British and American verse range from
Shelley's "To a Skylark" to the impassioned "Renascence" of Edna St. Vincent Millay
and to Edward Lear's whimsical "The Owl and the Pussycat." 224pp. 5% x 8'f,.
27165-X Pa. $5.95

COMPLETE SONNETS, William Shakespeare. Over 150 exquisite poems deal

with love, friendship, the tyranny of time, beauty's evanescence, death and other
themes in language of remarkable power, precision and beauty. Glossary of archaic
terms. 80pp. 5K,, x 8%. 26686-9 Pa. $1.00

BODIES IN A BOOKSHOP, R. T. Campbell. Challenging mystery of blackmail

and murder with ingenious plot and superbly drawn characters. In the best tradition
of British suspense fiction. 192pp. 5% x 8'/%. 24720-1 Pa. $6.95
 CATALOG OFDOVER BOOKS

THE WIT AND HUMOR OF OSCAR WILDE, Alvin Redman (ed.). More than
1,000 ripostes, paradoxes, wisecracks: Work is the curse of the drinking classes; I can
resist everything except temptation; etc. 258pp. 5% x 8' . 20602-5 Pa. $6.95
SHAKESPEARE LEXICON AND QUOTATION DICTIONARY, Alexander
Schmidt. Full definitions, locations, shades of meaning in every word in plays and
poems. More than 50,000 exact quotations. 1,485pp. 6' x 9'.1. 2-vol. set.
Vol. 1: 22726-X Pa. S 17.95
Vol. 2: 22727-8 Pa. S1795
SELECTED POEMS, Emily Dickinson. Over 100 best-known, best-loved poems by
one of America's foremost poets, reprinted from authoritative early editions. No
comparable edition at this price. Index of first lines. (i4pp. 5'. x 8'H.
26466-1 Pa. $ 1.00
THE INSIDIOUS DR FU-MANCHU, Sax Rohmer. The first of the popular mys-
tery series introduces a pair of English detectives to their archnemesis, the diabolical
Dr. Fu-Manchu. Flavorful atmosphere, fast-paced action, and colorful characters
enliven this classic of the genre. 208pp. 51;6 x 8%. 29898-1 Pa. $2.00
THE MALLEUS MALEFICARUM OF KRAMER AND SPRENGER, translated
by Montague Summers. Full text of most important witchhunter's "bible," used by
both Catholics and Protestants. 278pp. 6% x 10. 22802-9 Pa. S12.95
SPANISH STORIES/CUENTOS A Dual-Language Book, Angel
Flores (ed.). Unique format offers 13 great stories in Spanish by Cervantes, Borges,
others. Faithful English translations on facing pages. 352pp. 51% x 8YL
25399-6 Pa. $8.95
GARDEN CITY, LONG ISLAND, IN EARLY PHOTOGRAPHS, 1869-1919,
Mildred H. Smith. Handsome treasury of 118 vintage pictures, accompanied by care-
fully researched captions, document the Garden City Hotel fire (1899), the Vander-
bilt Cup Race (1908), the first airmail flight departing from the Nassau Boulevard
Aerodrome (1911), and much more. 96pp. 87/A x 11'/o. 40669-5 Pa. $12.95
OLD QUEENS, N.Y., IN EARLY PHOTOGRAPHS, Vincent F. Seyfried and
William Asadorian. Over 160 rare photographs of Maspeth, Jamaica, Jackson
Heights, and other areas. Vintage views of DeWitt Clinton mansion, 1939 World's
Fair and more. Captions. 192pp. 8A x 11. 26358-4 Pa. $12.95
CAPTURED BY THE INDIANS: 15 Firsthand Accounts, 1750-1870, Frederick
Drimmer. Astounding true historical accounts of grisly torture, bloody conflicts,
relentless pursuits, miraculous escapes and more, by people who lived to tell the tale.
384pp. 5% x 8`. 24901-8 Pa. $8.95
THE WORLD'S GREAT SPEECHES (Fourth Enlarged Edition), Lewis Copeland,
Lawrence W. Lamm, and Stephen J. McKenna. Nearly 300 speeches provide public
speakers with a wealth of updated quotes and inspiration-from Pericles' funeral ora-
tion and William Jennings Bryan's "Cross of Gold Speech" to Malcolm X's powerful
words on the Black Revolution and Earl of Spenser's tribute to his sister, Diana,
Princess of Wales. 944pp. 516 x 8%. 40903-1 Pa. $ 15.95
THE BOOK OF THE SWORD, Sir Richard F. Burton. Great Victorian
scholar/adventurer's eloquent, erudite history of the "queen of weapons"-from pre-
history to early Roman Empire. Evolution and development of early swords, varia-
tions (sabre, broadsword, cutlass, scimitar, etc.), much more. 336pp. 6'i x 9%.
25433-8 Pa. $9.95
(continued from front flap)

THE GEOMETRY OF ART AND LIFE, Matila Ghyka. (23542-4) $6.95

MATHEMATICS IN THE TIME OF THE PHARAOHS, Richard J. Gillings. (24315-X)
$8.95
ESSENTIALS FOR SCIENTIFIC AND TECHNICAL WRITERS, Hardy Hoover. (24060-6)
$7.95
PHYSICS AND PHILOSOPHY, Sir James Jeans. (24117-3) $7.95
SCIENCE AND MUSIC, Sir James Jeans. (61964-8) $7.95
PRACncAL STATISTICS SIMPLY EXPLAINED, Russell Langley. (22729-4) $9.95
VISUAL II3.UsIoNS: THEIR CAUSES. CHARACTERISTICS AND APPLICATIONS, M.
Luckiesh. (21530-X) $6.95
PRELUDE TO MATHEMATICS, W.W. Sawyer. (24401-6) $5.95
GREAT IDEAS OF MODERN MATHEMATICS: THEIR NATURE AND USE, Jagjit Singh.
(20587-8) $8.95
WONDERS OF THE SKY, Frederick Schaaf. (24402.4) $8.95
How TO CALCULATE QUICKLY, Henry Sticker. (20295-X) $3.95
1001 QuEsnoNs ANSWERED ABOUT EARTHQUAKES, AYAtA.NCHFS. FLOODS AND
OTHER NATURAL DISASTERS, Barbara Tufty. (23646-3) $7.95
BASIC MACHINES AND How THEY WORK. U.S. Navy. (21709-4) $7.95
BASIC OPTICS AND OPTICAL INSTRUMENTS, U.S. Navy. (22291-8) $16.95
LADY LUCK: THE THEORY OF PROBABILITY, Warren Weaver. (24342-7) $8.95
CRYSTALS AND LIGHT: AN INTRODUCTION To OPTICAL. CRYSTALLOGRAPHY.
Elizabeth A. Wood. (23431-2) $6.95
TEMPERATURFs VERY Low AND VERY HIGH, Mark W. Zemansky. (24072-X)
$5.95
How To SOLVE MATHEMATICAL PROBLEMS. Wayne A. Wickelgren. (28433-6)
$7.95
LANDMARK EXPERIMENTS IN TwENnETH CENTURY PHYSICS, George L. Trigg.
(28526-X) $8.95
INTRODUCTION TO LOGIC AND TO THE METHODOLOGY OF DEDUCTIVE SCIENCES,
Alfred Tarski. (28462-X) $8.95

Paperbound unless otherwise indicated. Prices subject to change with-

out notice. Available at your book dealer or write for free catalogues to
Dept. 23, Dover Publications, Inc., 31 East 2nd Street, Mineola, N.Y.
11501. Please indicate field of interest. Each year Dover publishes over
200 books on fine art, music, crafts and needlework, antiques, lan-
guages, literature, children's books, chess, cookery, nature, anthropol-
ogy, science, mathematics, and other areas.
Manufactured in the U.S.A.
 Richard J. Gillings I
MATHEMATICS IN THE
TIME OF THE PHARAOHS
In the past, scholars have tended to dismiss the mathematics of the ancient Egyp-
tians as "child's play." compared with the achievements of the Greeks and other
later civilizations. Nevertheless, in a society that achieved the marvelous accuracy
of construction revealed in the Pyramids, extensive systems of irrigation canals, the
erection of large granaries, levying and collecting of taxes, and other evidences of a
well-organized and highly developed culture, mathematics must have played a
major role.

In this remarkably erudite work, the first book-length study of ancient Egyptian
mathematics, Prof. Gillings examines the development of Egyptian mathematics-
from its origins in commercial and practical computations to such accomplishments
as the solution of problems in direct and inverse proportion; the solution of linear
equations of the first degree; determining the sum of arithmetical and geometrical
progressions, and the use of rudimentary trigonometric functions in describing the
slopes of pyramids.

Drawing on all the extant sources-the Egyptian Mathematical Leather Roll, the
Reisner Papyri, the Moscow Mathematical Papyrus, and, most extensively, the
Rhind Mathematical Papyrus, a training manual for scribes-the author shows that
although the mathematical operations of the ancient Egyptians were limited in
number, they were adaptable to a great many applications. Professor Gillings is
also at pains to debunk such myths as the numerical mysticism that arose in con-
nection with the construction of the great Pyramids, and the oft-repeated assertion
that the Egyptians were conversant with the Pythagorean Theorem.

Enhanced with photographs of age-old papyri and other artifacts, as well as the
author's own calligraphic renderings of hieroglyphic and hieratic words and
numerals, this carefully researched and well-presented study will fascinate Egyp-
tologists, mathematicians, engineers, archaeologists- any student or admirer of the
remarkable civilization that flourished on the shores of the Nile so many centuries
ago.

Unabridged republication of the edition published by MIT Press, Cambridge,

Mass.. 1972. Some text figures. 14 appendices. xiii + 286pp. 53/s x 8"s. Paper-
bound.

Free Dover Mathematics and Science Catalog (59065-8)

available upon request.

ISBN 0-486-24315-X

$9.95 IN USA 9 780486"243153