'e-

tl
l1-]]I

/\
I rlirl; r:r:,. *:: I I :
.
__. r::,:. r... 'r.i;/
.]':i ltrl,', .:

1',
L\.\
i
\l
.,::'" '.,\ j,, " " '

"ttl'''- '
;::''..;r.:1--f:i I .l
o,
"' ,t r.r-

.: '< .

.)<

;r-'
':

{> t.i t\
.f{

ll
L(

tl,
.Y Silt,,.

W
i
J
I
.r,.l
J
1,,
\t

*{ffi 3,

.\
*

.0-,,

7#
,

- .::,: *<r-t.:.::::. . rr.:.:ti::.i.. ,tri

1

I 1{w
I
;l.i. l,,i.tr: ;4f1,:, i
:':'
'
{, ;--
...tr': ir .1 . .' i ,'
./._.-a
?:
FfFE'5
illriillr
';;,;:,.;,.; {'tt}.:;;r,,,
, .r.',:. ......,
o
tLU
J
J
CI

U
z
o
o
i
e
o
a

6
o
.E
q

F.

z
The Musician ll9l$by Louis Marcoussis

f he Polish painter Louis Casimir Ladislas Marcoussis (1883-1941) and motion. (Some Cubists would present multiple sides of a
I was one of the group o{ artists in Paris who were exploring new three-dimensional object simultaneously, mixing "points of view"
possibilities of spatial representation in the second decade of this in a single "snapshot, " to suggest another dimension-time.)
century. The movement eventually came to be called "Cubism," Other aspects of the painting, though, suggest restraint and
and its practitioners included some of the most illustrious names discipline: the narrow range of colors used, the way much of the
o{ modem art. Some Cubists, Marcoussis among them, saw painting seems to radiate from a single point, the sense of
themselves as obeying a severe discipling one inspired directlyby architecture in the placement of the gray slabs.
mathematical laws. Others, like Pablo Picasso and Georges Many artists have been profoundly affected by the new theories
Braque, saw a greater role for the individual imagination and in physics that have bubbled up into the public consciousness
denied being "Cubists" at all. Cubists, they felt, simply adhered throulhout the 20th century. And many of them have been
to a strict set of compositional rules. fascinated by music and frequently used musicians as their
Labels aside, it's apparent that Marcoussis wasn't content to subjects. No doubt you've come across discussions o{ the more
produce an objectively "realistic" image of a musician playing his traditional physics and mathematics of music-columns of
instrument. After all, that's what cameras are for. As a self- vibrating ar, the relationship of string length to pitcl! and so on.
conscious painter, Marcoussis was able to use fractured planes, But what about the music of physicists? For a lighthearted peek
skewed angles, and jumbled shapes to create a sense of rhythm into that relatively unexplored area, see page 54.
SEPTEMBER/OCTOBER
UA TU 1 990 VOLUME
.1,
NUMBER 1

F EATU R ES

0 fating Humble Pi
[elusion ol' Inaud?
byAN Zaydel

12 A Mathematical Taxonomy
fhltille in fie lam oladuersity
by A P Vesyolov

20 Physics in the Banya

Att ittttihtion lo fie lafihouse
by l.l. Mazin

Cover art by Leonid Tishkov

2l Huid Statics Revisited
Maybe you saw it on the news a while
Boy-oh[uoyancy!
back a perfectly sane-looking man fiv-
ing headfirst off a bridge, elastic straps by Alexander Buzdin and Sergey Krotov
attached to his ankles and nothing but
water to look at on the way down. From
that height, at that speed, it looked
about as inviting as a sheet of concrete.
DEPARTMENTS
He was wired for sound (he carried a tiny
video camera on his back as well), and 3 Puhlisher's PaUe 44 Loo[inu Baclr
the sounds he emitted during his flight
l0 At tle Blarftb0al'd lules Verne's cryptogram
were remarkable in their variety and
intensity. When his fall was success- Taking a flying leap 4$ 0etting Io lfiow...
fu1ly broken by the ankle straps, his 10 Bl'ainteasel's The elementary particles
relief was apparent in his shouts and
23 llow [o You tigure?
5 2 0mltlutn Smiles
laughter. Physics limericks The
This man was, in [act, anAustralian 24 Malhemalical Sut'lrises music of physicists
television reporter, and he was investi- An old fact and some 50 lla[[enin0s
gating a new sport that has sprouted new ones
downunder. Newas a sport/ anyway. The USA Mathematical
12 l(aleidosroIe Talent Search ... Bulletin
It's actually based on a ceremonial prac- Board
tice that can be seen on the islands of Botanical geometry
Vanuatu, which is the subject of "Tak- 58 $olulions
34 ln I[e lal
ing a Flying Leap" onpage 10. (And the
gendeman in the long coat on our cover/
Through a glass bilghtly 04 [[eckmate!
calmly making calculations-just who 40 C onlest Fantasy chess
is he?ll A pigeonhole for every
pigeon ... Click, click,
click

OUAlllTUltl/[OllJTElllIS
Circle No. 1 on Beaders Service Card
 about assumptions, think about alter-
PUBLISHER'S PAGE natives-in short, to think. It,s dedi-
cated to the proposition that thinking
is enjoyable.
Can you read most of what is printed
in Qu antum, and do you understand
A Uood qtle$liolt what you are reading? You probably
can't just say "yes" or "flo" to that
question. Is some of it easy? Does
some of it require careful study while
worth a thousand routine answers you fill in details left out of the article?
-itb Is some material simply too hard?
We aim Quantum mainly at young
people who would answer "yes" to al).
of these questions. These young people
T QUANTUM WE'RE CON- things tl:rough. Too often we've taught would be among our nation/s best and
tinually questioning what we do. and tested for facts and information. brightest. But we need to know for
We're a rather new magazine, so As my friend has observed, there a1- sure that we haven't aimed either too
that's to be expected, but I hope most seems to be a "destructive collu- high or too low. Again, what do you
we keep this self-critical attitude as sion between students and teachers- think? We don't have a Letters to the
we grow older. If you don't mind, I'd a collusion in which students agree to Editor column, but we do read and
like to draw you into our questioning accept bad teaching provided they are think about every letter we get.
mode... given bad examinations." What do So write a letter and tell us if Quan-
Do you think you're intelligent? you think? tum is interesting/ challenging, and
Do you knowphysics or math pretty We publish Quantum for you. It is mostly comprehensible to you. If it
well? Have you thought much about meant to be challenging, as well as isn't, tell us exactly what you think
what it means to "know" something? entertaining, and should demonstrate we could do to make it more nearly
I have a good friend who has carefully what it means to think about a prob- what you'd like.
probed for genuine understanding of lem, think about a solution, think Aldridge
physics amongyoungpeople who have -BillG.
studied it. His observations suggest
that true knowledge and understand- Here is a book filled
ing are frequently elusive. with physics demonstrations
Knowing a fact or how to solve a that are amazingly simple,
particular class of problem isn't enough. often playful, and always
How do you know the "fact" ?. Why do instructive. Each of the
you believe it? And, as for the prob- .l
75 demonstrations uses
lem, do you merely know a procedure inexpensive, everyday items-
that can unquestionably (and unques- rubber balls, a plastic ruler,
tioningly) be foliowed to ar:iye at a Styrofoam cups, string, etc.-
solution? Or, instead, can you iden- and each is very clearly
tify the relevant laws, or principles, described. lntended for science
and definitions and use them in a teachers, from middle school
correct and efficient approach to the to college level, this is also a
problem? great book for students who
Can you understand the difference want to experiment (and learn)
between tacts, data, or observations on their own.
and inferences? How is a theory or Paper: $14.95 ISBN 0-691-02395-6
model created, and what makes one Shipping: $2.75 lor 1st book;
better than another? 50 cents each additional book.
The poing of coursg is this. It's not VISA, Mastercard, and American
Express accepted by mail or phone.
what you profess to know that's im-
portant. How do you know some- qft-trrje_qt
thing? Why do you believe it? How
would you find o:ut? That'swhat's ORDER FROM YOUR BOOKSELLER OR FROM

really important. I regret that we/ as Princeton University Press

teachers, haven't emphasized the abil- ORDER DEPT., 3.175 PRINCETON PIKE, . LAWRENCEVILLE, Nl 08648
ity to ask good questions and think ORDERS: 800-PRIISBN (7 77 47 26)

Circle No. 2 on Beaders Service Card

OUAlt'lIUttl/PUBI.ISllIR'S PAOI
 Be a laclor in lhe

OUANTUM
Bqttatiolt!

Have you written an article that you

OUANTUM
THE STUDENT MAGAZINE OF MATH AND SCIENCE
A publication of the National Science Teachers Association (NSTA)
thinkbelongsin Quantum? Do you
& Qwrntum Bureau of the USSR Academy of Sciences
have an unusual topic that students
in conjunction with
would find fun and challenging? Do
the American Association of Physics Teachers (MPT)
you know of anyone who wouldmake
& the National Council of Teachers of Mathematics (NCTM)
a gr eat Quant um author? Write to us
and we'll send you the editorial guide-
lines for prospective Quantum con- Publisher
tributors. Scientists and teachers in Bill G. Aldridge, Executive Director, NSTA
any country are invited to submit ma-
USSR editor in chief Managing editor
terial, but it must be written in collo-
Yuri Ossipyan Timothy Weber
quial English and at a ievel appropri-
Vice President, USSR Academy of Sciences Production editor
ate for Qunntum' spredominantly high
Elisabeth Tobia
school readership. US editor in chief for physics
Int e mational c ons ult ant
Sheldon Lee Glashow
Edward Lozansky
Send your inquiries to: Nobel Laureate, Harvard University
Advertising manager
US editor in chief for mathematics Paul Kuntzler
Managing Editor William P. Thurston Director of NSTA publications
Quantum Fields Medalist, Princeton University Phyllis Marcuccio
l7 42 Connecticut Avenue NW
Washington, DC 20009-l 17 I
US advisory board
Bernard Y. Khoury, Executive Officer, AAPT
James D. Gates, Executive Director, NCTM
A IIATIOlIAt COMPETITIOI{ FOB HIGH SCHOOT Lida K. Barrett, Dean, College of Ars and Sciences, Mississippi State University, MS
SIUDE]ITS A]ID TEACHEBS George Berzsenyi, Professor of Mathematics, Rose-Hulman Institute of Technology, IN
Arthur Eisenkraft, Science Department Chair, Fox Lane High School, NY
Judy Franz, Professor of Physics, West Virginia University, WV

super[uesT
SuperQuest challenges students at any high school in
Donald F. Holcomb. Professor of Physics. Comell University, NY
Margaret J. Kenney, Associate Professor of Mathematics, Boston College , MA
Lany D. Kirkpatrick, Professor of Physics, Montana State University, MT
the country to design an origi nal science project for i n- Robert Resnick, Professor of Physics, Rensselaer Polytechnic Institute, NY
vestigation on some of the fastest computers available. Mark E. Saul, Computer Consultant/Coordinator, Bronxville School, NY
There is virtual ly no I i mit to the range th'ese projects can
Barbara I. Stott, Mathematics Teacher, Riverdale High School, LA
[ake, from the design of traffic light cycles t0 the
clocking ot a black hole i n space. Super0uest is hosted
by the Cornell Theory Center in lthaca, New York. USSR advisory board
'Entering Sergey Krotov, Chairman, Quantum Bureau
teams consist of one teacher-coach and three
t0 f0ur students lrom grades 7-12 \ou must register
Victor Borovishki, Deputy Editor in Cbjef , Kvant magazine
.1
your i ntent to apply no later than March , 1.991 ; ful I ap- Alexander Buzdin, Professor of Physics, Moscow State University
plications with project descriptions are due April 12, Alexey Sosinsky, Professor of Mathematics, Moscow Electronic Machine Design Institute
1 991 . Winning teams come t0 Cornell (expenses paid)
.1991
in July for the SuperQuest Summer lnstitute for
three weeks of.training 0n the IBM ES/3090 600J Quarttunt (ISSN 1048-8820) contains authorized English-language translations tlon.r
supercomputers. Stipends for winners are $3,000 lor Kvcut. a phl,sics and mathematics ma-sazine published br the Acaden.rl of Sciences of
teachers and $1,000 for students. ln the fall, each the USSR and the Academl,of Pedagogical Science-r oithe USSR. as rre1l as ori_cinal
winning team will receive lBlM workslation configura
tions and continued access to the supercomputers.
material in Enslish. Copr right O 1990 \atronal Science Teachers .\ssociation. Sub-
scriptronpnceforoner.earrtburi\\Lles): indiriduai.SlE:student.S1.1;institr-rtion/
The potential of supercomputing is as great as your
imagination and curiosity. lf you can think of a compu
Library. S28: tbreign. 516. BuLk :ubscription: tbr students: 20-.19 copies, S12 each; 50 +
tational problem that needs t0 be solved, the IBM ES/ copies. S10 each. Corespor.rdence about
3090 600J supercomputers can help you solve it! edttonal This project was supported, in part,
subscriptions. adr erti,sing. and
bv the
FOR A1I APPTIGATIOT BOOI(EI A1{D
Oucrrt-
matters shoulcl be addressed ro -
MORE II{FORMAIIOlI 01{ SUPEBOUEST CAtt
rum, 1t 42Connecticut ;,;".'.-Nw, N"tj9-lql 9-gj gry-9
' Opinions expressed ol th ldation
are those-F9y authors
1-607-255-4859 WaShingtOn, DC 20009. and not necessarily those of the Foundation

Circle No.3 on Readers Service Card

srPrr]lllBrRl0cr0BrR 1gg0
 DURACELL
NSTA
ffiffiffiffitueffiffiffiffiffi
ffiffiffiffiffiwffiwffiffiru

EntertheNlhffi
rghip Gompetition
TO ENTER: o Obtain officialentry lorm and rules by writing to the address below.
o Design and build a battery-powered device.
o Submit an essay, wiring diagram, photos of the project
and a completed entry
form by January 25, lgg1.

1990 second-place winner Albert Huntington at the Duracett booth in Ailanta.

a Your student in grades g-12 could win e Allentrants receive a Duracell Designer
a $10,000 College Scholarship! Athletic Bag.
ta Over $30,000 in prizes are awarded ta Teachers win too!
each year to 41 winners: Place: One Personal Computer
-First
-First Place: One $10,000 Award -Second Place: Five Laptop Computers
-Second Place: Five $3,000 Awards Place: Ten $50 Gift Certificates for NSTA
Place: Ten $500 Awards
-Third
-Third Publicalions
-Fourth Place: 25 $100 Cash Awards e For more information and entry kits
e The top six winners, their parents and write to:
teachers will be flown to Houston for an DuracelI/NSTA Scholarship Competition
Awards Ceremony at the NSTA National 1742 Connecticut Ave., NW
Conventlon. Washington, DC 20009
Circle No. 4 on Beaders Service Card
 o
c

=
!a
o
lo
c

inventedby the French naturalist G.L.L.

Buffon ,1707-1788). Later Buffon's
origural experiment was repeated time
and again to confirm or refute some
conclusions of probability theory (or,
to be more precise, their applicability).
This resuLted rn an amusing incident,
Delusion or lraud? ir'huch I'm about to relate. But first
describe the experiment.
I'11

Buflon's B)tperitnettfl
To do Button's expenment you need
Tossing a needle and counting crossings to a needle and a honzontal surface ruled

calculate n-and the suspicious results of a by a grid of parailel equidistant lines.

The distance fi between th: lines and
certain Lazzarini the length -lof the needle 1frg. I ) must
satisfy the reiation I < Jr. Toss the
needle above the surface, each time
by A.N. Zaydel giving it a little flip so that it falls
freely from a height of about 5G{0 cm
S THERE AN EXPERIMENTAL by the diameter. There is also, how- and lands at random angles relative to
way to determine the value of n? Of ever, a completely different way. We the lines. After each toss write down
mr.lse there is: take a thread, measure can get an approximate value of n by whether the needle does or does not
the crcumference of a circle with a using a needle . . . and a touch of intersect one of the parallel lines {fig.
knovm diameter, and dir,rde that amount probability theory. This approach was 1a and fig. 1b, respectively) and calcu-

stPIttllBtR/0[I0BtR 1gg0
 Iate the frquency of such "crossings,,- proximately equal to this probabiliry equation (1), we must be able to pre-
that is, the ratio of the number of we can approximate n by using the ciseh measure h and I (or rather, their
throws m resulting in intersection to equation ratio). This can be done only by actu-
the total number of throws n. you,ll ally measuring both values. Errors in
soon see that as the number of tosses measurement will surely affect the
in our experiment increases, the fre-
2ln
It=-.-
htn (1) accutacy of n. Let's estimate the
quency scarcely changes. Not only magnitude of the error. Suppose the
that, if we perform many trials con- If I = h, we can simply say that n is needle is 50 mm long, which we,ll
sisting of many tosses/ the frequency approximately equal to twice the total take to be equal to the distance be-
of crossings is approximately the same number of tosses divided by the num- tween the lines. Using ordinary meas-
for every trial. ber of crossings. uring devices-for instance, a slide
This property of frequency "stabili- gage or vemigl salipers-we can meas-
zation" (along with the unpredictabil- What fie sttrul'ilneltt$ sfiowed ure both lengths to an accuracy of 0;1
ity of the result of each individual The accuracy of the approximation mm (0.2 %).1
throw) is a characteristic feature of all in equation (t) depends on the number Making use of more sophisticated
experiments in probability theory. A of tosses n. At first giance it seems instrumentation, we can measure all
certain number p (0 s p s 1) is assigned that by increasing the number of tosses the lengths to an accuracy of 0.01
to each outcome of such an experi- we can obtain the value of n to any mm. At that point we've pretty much
ment and is called its probability. This desired precision. Try to cany out reached the practical limit-it's nearly
number simply expresses the likeii- such an experiment and you'll see that impossible to reduce the error to
hood of this outcome in the experi- it's quite easy to obtain the value 3.1 0.001 mm because a variation in the
ment. The probability is the value for n. But the next decimal, 4, is much needle's or the surface's temperature
around which the frequency of the harder to get. In the 19th century, of only I or 2 degrees results in a
outcome oscillates in an experiment when probability theory was often variation of about 0.001 mm in the
of sufficiently long duration. So the regarded as a semiempirical science, measured distances. Deformation of
approximate value of a probability can such experiments were of great value the needle caused by its collision with
be obtained empirically by calculat- andwere scrupulously
ing the coresponding frequency. (The staged by many scien-
Number Experimental
precise mathematicai formulation of tists. A table from Name Year of tosses value of p
how the experimental {requency tends B.V. Gnedenko's text-
to a certain limiting value, or proba- book on probability
bility, is grven by the law of large num-
bers, proved in its simplest form by
(which is well known Wolf 1850 5000 3.1596
and widely used in the Smith 1855 3204 3.1553
facob Bemoulli 300 years ago.) On the Soviet Union)is given Fox 1894 1120 3.1419
other hand, probability theory makes here by way of illus- Lazzarini '1901 3408 gj4i1g2g
it possible in most cases to calculate a tration. (The names
probability theoretically by examin- of the scientists in- True value of ru
ing experimental conditions.
Later we'll show how Buffon's prob-
volved in the needle
throwing are listed in
to the seventh decimal place: 3.i415927
lem-that is, howto find theprobabfity the first column.)
p that the needle will intersect a line- Compare the experimental results the surface, wear atits tips, deforma-
is solved. The result is really quite with the true value of n. The values in tion of the surface itself-all these
amazing: p = lzlnl\lh)! Since the the first two lines differ from n by factors make it quite unreasonabie to
frequency of intersections m/n is ap- 0.01-0.02. The value obtained by Fox try to achieve a level of experimental
is greater than n by only 0.0003. error of 0.001 mm. Even without a
This is an amazingresult. But more detailed evaluation, we can still
the value obtained by Lazzaini say with assurance that the experi-
is only 0.0000002 over the true ment can't determine the value of n to
value. This is a miraclel (Orso it afl accuracy better than 0.2-0.02o/o,
seems.)
IFor a small number of tosses
(approximately 0.14.2% of the total
Tle ilteuihtility ol mrot'
number of tosses) the distance from the
What astonishes us and, to be needle's tip to the line will be less than
frank, makes us skepticai of the 0.1 mm. In this case the naked eye can,t
result of the last experiment? discem whether there is an intersection.
Several things. First, the accu- This may also contribute to the resulting
error, although this contribution is
racy of the measurements. To smaller than that caused by other
obtain the precise value of nfrom sources of experimental error.
Figure 1

OUAIT,IIU]Il/IEAIURE
This means that with conventional nearly 12,000 times. The experiments 3.16, and the third should read3.l4.
measuring devices the best we can listed in the table (except for Fox's)
eq)ect to obtain is n = 3.141 + 0.006. If involve only 3,000 to 5,000 throws. And so-delusion ol' lnaud?
the experiment is conducted with the According to equation (21, the result-
But how can we explain the result
utmost rigor, we might hope to get fi = rng u is somewhathigher then-about
obtained by Lazzarin? We can hardly
3.1416 + 0.0006. Wanting to derive 0.025-0.030, which is in good agree-
suspect him of deliberate fraud. When
the value of n in this experiment to ment with the results of Wolf and Lazzarini was throwing his needle,
eight decimal places is like trying to Smith. About the same level of error the law of large numbers was already
weigh a match on a railroad scale. In occurs when the needle's length is
'
well known-our calculation could
both cases, the instruments are too measured to an accuracy of 0.1 mm.
have been done by any mathemati-
crude. So there's no need to stdve for greater
cian of the time. Most probably the
In any measurement we should aco)racy in these measurements: it's
scientific community never took his
evaluate the accuracy first and then useless to measure the length of the
result seriously, despite the fact that it
reduce the result of the calculation to needle to an accuracy of 0.001% if the
has been republished a number of times.
the corresponding number of decimal limitation imposed by the number of Before he began his experiment,Laz-
places. If in measuring a value A we tosses results in an overall error of
zarini should have known what he
get A = 2.474329, and the accuracy of 0.24.3Y".
could expect from it. And even after
the measunng device is 1%, we should Choosing a number of tosses be-
obtaining a number as a result of cer-
write A =2.47, dropping all the subse- tween 3,000 and 5,000 makes good
tain arithmetic exercises, he should
quent digits, which merely reflect the sense. One toss might take about 5
have refrained from publishing such a
acuJracy of our computing device. seconds (try to toss a needle faster). An
fantastic result until he was able to
(Even ordinary calculators now dis- experiment consisting of 10,000 tosses
reproduce it in an independent series
play 6 to B decimal places.) By writing would take approximately 14 hours-
of tests. Personally, I would guess that
all the decimal places obtained, we two fuii working days. If we want to
he was too eager to outdo all his prede-
overrate the accuracy of the experi- obtain results 10 times as precise, it'll cessors. Overweening ambition some-
ment and thereby report a misleading take 100 times as long-200 days times deludes researchers, leading them
result. (according tothe"lf ntlz law"), which
to findwhat they want to find.
There is, however, no use in trying is too long for such an experiment. In
Another possible explanation is that
to increase the accuracy of our mea- order to obtain theLazzar\ni result,
while throwing the needLe Lazzarini
surements. It's quite sufficient to whose accuracy was 0.0000002, we
calculated n after eachthrowand ended
measure the needle's length to 0.1 would have to throw the needle for the experiment after 3,408 throws,
mm. That's because the accuracy of about 4 million years! (And this result
having obtained the value given above
determining n is already limited by a could have been obtained only if the
by pure chance. Of coursg even after
completely di{ferent, "probabilistic" length measurements were made to
10,000 throws it's practically impos-
circumstance-the impossibility of an absolutely unreal level of accu-
sible to obtain a given value to an
performing enough tosses to ensure racy.) So, starting his experiment in
aco;rtacy of 2 . l0r even once-the
that the approximating equation (1) 1901 and throwing the needle until
probability is about lOr-but it could
holds with a relative error of less than now,Lazzariniwould have been as far
happen. If that was indeed the case,
0.01! There's away of estimating the from his published result as he was on
Lazzairi deceived himselJ rather than
rate at which the differencelmf n-p\ the first day.
others. Unfortunately, this happens
decreases as the number of tosses in- There's one more source of experi-
in science from time to time.
creases. It turns out that to increase mental error in Buffon's experiment.
Although the resuit of Lazzarini's
the accuracy of the approximate equal- For the probability of intersections to
experiment didn't confirm the con-
ity mln=pby afactor of ll, we have to equal the theoretical valre oI p :211
clusions of probability theory (taken
increase the number of tosses by a nh, we must ensure that all the needle's
seriously, it would have contradicted
factor of N 2. In other words, the eruor positions with respect to the lines
them), it did serve to generate this
is inversely propofiionalto the squarc upon the surface are absolutely equiva-
cautionary tale. Once again we are
root of the number of tosses. A de- lent-that is, that none of them has an reminded to be cautious with experi-
tailed examination of Buffon's experi- intrinsic tendency to occur more of-
mental results and the statistical analy-
ment shows that for h 1, the acnur:acy ten than the others. (A more precise
= sis of these results.
o that can reliably be expected for n formulation of this condition is given
alter n tosses is given by the formula below. ) In a real experiment it's very
Soluing Bullon's
E_
0=Vr' rzr
unlikely that this could be achieved.
So there are ample grounds for
concluding that the number of deci-
Uotleln
It remains for us to explain how the
probability that the needle will inter-
If we demand that cx be less than mal places given in the table is too sect a line in Buffon's experiment is
0.02, we should be prepared, accord- high. To be credible, the first two calculated. We'll denote the distance
ing to equation (2), to toss the needle entries in the last column should read from the needle's center to thenearest

stPTtllllBtR/0cT0BtR 1990
line by a l0 < a shlzl and the angle In particular, the arca of the curvi-
between the needle and the line by q (0 linear trapezoid formecl by the sinu-
sq slElzl. Then, as figure 1a shows, soidal curve is equal to
the needle intersects a line i{ and only Back issues ol
if J,=
n.1 I
.J lsirro,lq
I
r o2
aS
7sinQ. tlr r ( 0) ]

2L'-(:O\2)- -(:()\ l
|

Each possible outcome in our experi-

You may order copies of
ment is described by a point in the I
plane having coordinates (ag) and lyrrg a
the |anuary (premier) and May
L
inside the rectangle bounded by the issues of Quantum.
coordinate axes and the straight lines Since the area S of the rectanglc
a = hl2 and q = nlL(h9.2). The points ecluals rhf 4, we get the probability Single copies: $S
of the rectangle lying below the curve
S, 2-19 copies: $4lea
a = (ll}lsing represent the intersec- 2t 2049 copies: $3/ea
tions, while the points above the curve 'Snh
tt=
50 or more: $2lea
correspond to the outcomes when the as we predicted a few pages
needle doesn't cross a line.
The problem can now be reformu-
o Send your order to:

lated in the following way: A point is

chosen at random within the rec-
tangle {0 < 9. nl2; 0 < a.hl2}. What Quantum
is the probability that it lies below the 1742 Connecticut Ave. NW
sinusoidal cuwe a: (ll2)sin<pl Washington, DC 20009
The term " at rarrdorn" here replaces
the requirement that all the positions
of the fallen needle be equivalent.
There are several ways of precisely
defining this term. The simplest is to
require that the probability that a point
will fall into a sqwre {g, s g s g2i a, < a
. arllgz-9r= az- at= d)with a given
side doesn't depend on the position of
AFS
INTERNAT!ONAL/INTERCULTURAL PROGRAMS
the square inside the rectangle (at- IN 70 COUNTRIES AROUND THE WORLD
though it obviously depends on the
length d of its side). For instance, the
probabilities of finding the point in-
side the squares K, and Kr(fig.2l are
equal. From dris we can easily deduce
that the probability of landing in any
figure inside our rectangle is propor-
tional to its area and consequently
equals the ratio of this area to the area
of the entire rectangle.

wu
CHALLENGE YOURSELF!
High School student exchanges . Teachers programs
Hosting opportunites for families

Call Us!
FOR AN EDUCATIONAL ADVENTURE THAT WILL LAST A LIFETIME
1-800-AFS-lNFO.
lnternational lntercultural Programs
313 East 43rd Street New York NY 10017

OUAilTUlll/TtATURt
 AT THE BLACKBOARD

Ta[inU a IIyiltU leap

Applying Hooke's law on a South Seas r,s/e

by A.A. Dozorov

ATIVES ON ONE OF THE IS- F= -kx. (1) ation the term stress, o = F/S, is often
lands of Vanuatu (formerly the used. ili a rod of cross section S is
New Hebrides) celebrate their The factor k is the rrgrdity coefficient compressed b,v a force F, then o is the
festivals in a rather btzarueway. of the stretchedbody (ior example, a average prcssure on its end.)
A young man mounts a special tower/ string or spring). The absolute value of the relative
attaches vines to his anldes, and then, The greater the cross section of the stretching e for srnall deformatlons of
accompanied by the music of a ritual string, the greater the force that must elastic bodles is proportional to the
dance, throws himsel{ off. The vines, be applied to achieve the same stretch- absolute value of the stress o:
tied at the other end to the top of the ing of the string. In other words, the
tower/ break his fall in such a way as to rigidity coefficient is a function of the r= EI^o. (2)
dampen the shock, and the young string's cross section. In such a situ-
man lands safely. This relationship reflects another for-
The height of the towers used on mulation o{ Hooke's law. The coeffi-
the island varies from 15 to 30 meters. cient E is called Young's moduius.
You'dthink the load on the jumper's Using equation {2) and recalling that
legs would increase as the tower grows t = xf l, we can wnte the expression for
taller and the vines grow longer. As a the force in another way:
r*J
result, there should be a maximum _ .sE
f - _t (3)
altitude for safe jumping. This suaight- I
forward and obvious conclusion,
however, is absolutely wrong. The Comparing equations (1) and (3),
right answer is provided by Hooke's we get the following relationship be-
law. tween Young's modulus and the rigid-
Let the length of a fteely suspended ity coefficient: k = SEll. Young's
string be equal to 7. When a force is modulus is dependent only on the
applied to the string, it stretches to the materialfromwhich the string is made,
lengthl + x. The valuexis called the while the coefficient k depends also
absolute deformation of the string, on the shape of the stretched sample.
and the vahrc xf 1= e is called the According to the definition, Young's
relative def ormation. This deforma- modulus is numerically equal to the
tion is said to be "elastic" i{, after the force stretching a string with a unit
force is removed, the string retums to c,ross section to double its length (the
its original length. Generally, for small dimensionality of Young's modulus
elastic deformations (x << 1) the value in SI units is N/m'z).
of the absolute deformation is propor- Let's return to Vanuatu now and
tional to the force applied. The direc- try to estimate the maximum stretch-
tion of the elastic force tending to ing of the vines. This amounts to
retum the string to its initial state and solving the following problem.
the direction of deformation are oppo- Amassm= 72kgis suspended on
site. This situation is described by an elasticweigfutJess string and fuopped
Hooke's law: from the point where the other end is

10 stPTIllllBtR/0CI0BtR lgg0
attachecl. Find the ntttxintttm f orce The overload is the difference between critical height beyond which the human
sftetching the strtng nnd the tnaxj- this force and the force of gravity mg. body could not withstand the over-
munt ttcceleratiotl of tlte mass whtle Therefore, the maximum overload load (that is, the additional g-forces). If
its fall is beitgbroken. Yotng's nndnlus acceleration is the elastic properties of the vine are
for tlte srring.rs E: 10' N lm2, and its close to those of rubber (that is, E = IO7
crossSectionS=9ctn2. N/m'), the vine's cross section S =
a=g
The work cxpended in strctching 9 crr,z, and the jumper's mass is 72 kg,
the string b1'an arnount Ax is erlual ro (This obviously occurs at the lowest we get a = 59. The human body can
point.) withstand such overloads.l O
AA = F.Ax. So neither the maximum stretch-
ing force nor the maximum overload rHere's something else to think
The force F is propor-tronal to x (fig. 1). acceleration depends on the string,s about: is the overload more or less i"f we
The work expcnded over thc interval length. take the vine's weight into account?
Ax is numerically equal to thc arca of And on the same taclg who will
As for jumping from a tower/ we
the trapezoi d AIICD.
experience more g's, a little person or a
can now conclude that there is no big person?-Ed.
If the string stretchcs from length 7
to length 1+ 5 thc work expended (and
consequently the potential energy and lsaac Nffion Forces:
acquired by thc string) is determined Could lsaac Newton realy 6elteve
by adding up al1the components of the that a thrown object wouid continue
work. That is, the string's potential at a constant ve locity in a strarght lrne?
energy W is defined by the area of the This manual is designed to help
triangle OCD: teachers introduce the sometimes
daunting subject of Newtonian
w =lr, =i.r.,.. (4) mechanics to students rn the mrddle
grades. The 27 teacher -created
The potential energy of the mass is activities-including marble races, a
transfomed into the krnetic cnergy of tractor-pull using toy cars, fettucini
its motion and then into the energy of carpentry, film container cannons,
the string's deforn-ration. Since thc and others-wrll make teachers and
total height from which the rnass falls students look forward to class. All
is cqual to 1 + r, then the actiivites use readily avariable
materials to give students visual, aural,
,iiqt1-.r,= jrr' and tactile evidence to combat their misconceptions. Background for
teachers and a master materrals list follow the activities and make this manual
So the stretching oi the string is ec1ual useful for inservice workhops. (grades 6-jO) #pB_3g, lglg,
1g2 pp. $16.50
to
We live in a universe that is
,.
_,r,.,* \ ;,,-s, * lAr,.,1 constantly in motion. Do your
students understand why objects
move and how? Topics such as
and the maximum stretching force
projectile motion, work, energy,
(since k = SE l1) is given by
machines, torque, and center of
gravity are introduced with the first-
F = mg,-t +2mgES. time student of mechanics in mind.
The informal, hands-on activities
use readily-available materials and a
variety of instructiona I techniques
to combat common misaoncep-
tions and make the sometimes-
intimidating subject of mechanics
i m med iate ly accessi ble-and
fun-for teachers and students. Show your students the Evidence of Energy
that is all around them and they will see just how exciting learning can be.
(grades 6-10) #PB-80, 1990, 200 pp. $16.50
All orders of $25 or less must be prepaid. Orders over $25 must include a postage
and handling fee of $2. No credits or refunds for returns. Send order to: publication
Sales, NSTA, 1742 Connecticut Ave. NW, Washington, DC 20009-IL7l, (Z0Z) 3/zg_
Figure 1 5800. Quantity discounts are available.

OUA|llIUI'll/AT T1lt BI.ACI(BOARO lt

 "1'
1

, !u'
.* .'
'"..*.;/
, I
"?.,"*

g-
.;!
I 1 --'-\
i /"':
n:lr:
-.1 ii
. ::., j L.
4

I ta

lntr
D
Ratr

L]
D
-I
Fort

N-
Add

C*

ji

"a

Ca

r_

a:
?

1;.. U.&oq*{
 Flexihle in lhe lace ol aduersily
ldentifying and classifying a certain breed of fanciful "creatLtre"

by A.P. Vesyolov

Figure 1
HE HERO OF OUR STORY IS
the "flexie"-an imaginary flex-
ible organism living in or.r three-
dimensional world and capable
of marvelous transformations. You
canthink of aflexie as aflat amoebalike
creature of finite size and zero thick-
ness having a single mobiie closed line
as its boundary. In the vicinity of each
inner (that is, nonboundary) point a
flexie resembles a tiny piece of plang members of the flexie family. August Ferdinand Moebius { 1 790-1868),
possibly a bent one. Its life consists The simplest flexie (fig. 1) looks a leading geometer of his time, was the
not only of continuous deformations like a genuine amoeba, although it director of a German observatory and de-
(thrt ir, bending stretching, and shrink- can take the form of a trianglg a sqrurg voted much of his time to astronomy. But
ing without tearing and gluing) but {or many years he also contemplated the
or even a star. A special term, "dislg"
properties of objects sirnilar to our {lexies, in
sometimes of more dramatic transfor- has been coined to designate such particular those of the Moebius strip shown
mations of the "cut-and-hea1" t1pe. objects. in figure 4. Only after many yean did Moebius
Here's how that happens. First, a The flexie shown in figure 2 is venture to submit the results o{ his work to
self-inflicted wound appears on the called (more or less understandably) a the French Academy in Paris, but the subject
of his "Memorandum on Single-sided Sur-
flexie's surface-a cut that goes along "handle," although it resembles a faces" was so unusual his manuscript gath-
aL arc that starts and ends on the punctured inner tube more than the ered dust on the Academy's shelves until the
boundary. Then the flexie experi- broken handle of a cup. Its shape, author finally decided to publish it as a sepa-
ences convulsions-stretching shrink- however, can change beyond recogni- rate book at his own expense. Around the
ing or twisting in any possible way- tion (fig.3). same time the German astronomer I.B. List-
until, finally, the torn edges rejoin ing (1808-1882) independently obtained and
The next flexie (fig. 4) merits a bit
published results similar to those of Moebius.
(each point joining precisely the point more discussion. We'lI call it
from which it had been torn awayl, "amoebius" after its discoverer, A.F. It's easy to construct a model of the
and the wound heals again. This cut- Moebius (though it's usually known amoebius-in fact,you should do that
and-heal act doesn't alter the inner as a "Moebius strip"). before we go any further. Take a strip
structure of the flexie, but it may

*ffi
thoroughly change its position in space
(resulting in entanglement or disen-
\\
tanglement of the flexie). If you've
ever tried to untangle a fishing line
-'+ u
i::{ili;j*1
ad .l;l'
you'll certainly appreciate the extent
to which your task would be simpli- I w]""

o
(n
o
oo
fied if the fishing line had the same
properties. We'll come back to the
transformations of flexies a little later,
but first let's get to know some of the
Figure 3 .Lffi-'r
* #i,,t.,.,,',

-o
l
o
OUAIIIIlJllll/ItATUBI l3
 amoebius under natural conditions.
First of all, owing to its cut-and-heai *-.. \ *
ability it can acquire the shape of a
strip twisted by any odd number of
half turns (fis. 5). (Notice that a strip
twisted by a whole number of turns
isn't a flexie at all since it has two
closed curves for its boundaries. ) in
fact, after cutting itself apart the amoebius Figure 7
can untwist itself by any whole num-
ber of tums and then heal itself along most symmetric states.
Figure 4 the same edge again. (Check this with hr exactly the same way the handle
of paper (approximately 20 cm by 4 a strip of paper.) Simiiarly, the amoebius generates a family of two-sided flexies
cm), draw a blue line down the middle can tie itself into any knot and disen- obtained by sequentially fusrng a number
along its entfue length on one side and tangle itself again (fig. 5). of handles to one another (fig. 9). For
a red line on the other, and then glue lack of a better name, we'llcall this
the ends of the strip together after the "bilater alfamily." The bilateral
grving one of them a twist of 180" (fig.
ffi family likewise consists of an inJinite
5). Where the ends meeg the blue line ':
number of species described by the
runs into the red ong so that now both Y number of handles making up a par-
lines turn out to be on one and the !i
ticular flexie. Since they are two-
same side of the surface. This means sided, disks naturally belong to this
*E:r'
the amoebius has only one side! You @, family as well. (In this case the num-
can now run your finger along the Figure 6 ber of handles N= 0.)
entire surface of the amoebius with- I just can't decide which of these
out going over the edge, which You No temporary self-inllicted wounds like many shapes is the most beautiful, so
these would havebeennecessary i{ the amoeblus thebilateral family is represented twice,
couldn't have done with the original
lived in four-dimensional space. In addition in figures 10a and 10b. Take the
strip of paper that had two sides (with to providing a convenient escape from a
blue and red lines). completely sealed room (an achievement double handle (N: 2), for examPle,
Let's perform an act of pure barba- well known and widely exploited in science and try to see for yourself that you
rism: take a pair of scissors and cut the fiction), the fourth dimension enables one to actually have two different forms of
freely undo knots and disentangle flexies. In the same flexie. It's quite a challenge,
amoebius along the red-blue line. How
the course of evolution the cut-and-heal but it will bring true pleasure to a
do you like the result? Be honest--did ability of our flexies arose precisely because
genuine geometrician. Maybe you'lI
of the "limited dimensionality" o{ our space.
discover even more elegant forms of
We can identify an entire (infinite) these same flexies.
family of fiexies, which we'll call the By now you're probably thinking
"amoebius [atr,i7y." After the amoebius that, various and rich as the class of
itself, the next representative of this our flexies is, their external appear-

ffiffi
Fi(;ure 5
family is obtained by "fusing" two
amoebii together along a section of
their boundaries (fig. 7). The third
representative is obtained by fusing
ance is just as complex. So the follow-
ing resulg which is the comerstone of
"flexiology, " rr,ay catch you by sur-
prise:
you expect such an outcome (unless three amoebii, and so on. The shapes The class of all flexies consists of two
you had aLready heard about it)? A1- of these flexies can be quite diverse, so infinite f amilies-the amoebius f amily
though the amoebius didn't split into they're not easy to recogrize. Figure 8 (fis. S) and the bilateral family (fis. 10).
two pieces, the cut stiil had a disas- shows the entire family in one of its
trous effect: the boundary of the new k twists
surface consists of two closed curves. Figure B
(Check to make sure.) So, according
to our definition, it's no longer a flexie!
Not only thag it has acquired a second
side! The next cut won't produce
such a striking result, although I doubt
you/ll be able to predict what will
happen.
Let's quit our brutai behavior and
take a look at what goes on with an N-/ N-2 ,ry -.i N-k

14 srPTtllllErR/0cI0BtR 1g$0
 k hol er

tw
& i.-
,.,'
a
w
ffi
ffi
'.w

N- 2

'i.\ ]
I
ii/t,tt I

2k t_y.lt.tr.
Ei^,,.^
rrgutu 10
Figure 9
Quantum,May l99O, "The Geome-
This is a qpicalclassification theo- try of Population Genetics" by I.M.
rem. It states that any representative Yaglom.)
of a certain class (in our case/ any The time has come, however, to
flexie) can be classified as belonging to prove our main result. We'lI make use
one species in a certain list (in our of the intuitively obvious fact that an
case, the amoebius andbilateral spe- appropriate number of cuts can trans-
cies). form any flexie into a disk. Two cuts
"No wayl" you may exclaim after are necessary for a handle, whereas
thinking about it for a while. "To one cut is sufficient for an amoebius.
begin with, you forgot about the flexie The boundary of the disk obviously
we get by fusing an amoebius with a includes the traces of the cuts (fig. 13).
handle" (fig. 11). But haVe I reaily To restore the flexie we should per-
overlooked it? Let's analyzethe situ- form the inverse operation and"heal"
ation more closely by taking, for the them. Here's how to do it. For each o{
sake of convenience, the handle in the the cuts, take a strip of paper and glue
one end to one edge of the cut. Then
Figure 12 stretch the strip over the disk in an arc
The amazing transformation of a and glue the other end to the other
handle fusad with an amoebius into edge. Notice that the strip obtained in
a tiple amoebius. Fitst, one of the this way can be one of two types. We'li
bases of the amoebius is dragged
call them type 1 (fig. l4al or type 2 (fig.
along the edge of the handla (a) itntiJ
it takes a position between the two 14b), depending on the relative orien-
neigfiboring b ases of the handle (b), tation of the cut's edges on the bound-
Then the two right bases of the ary of the disk.
handle are dragged one after the The resultingflexie looks iike the
Figure 1'1
other aJongthe edge of the twisted
one in figure 15.
amoebius strip (see figwe b).
form presented at the extreme right in Any flexie can be deformed into
Moving along the edge of the
figure 3 (transferred now into figure amoebius, the two strips of the such a shape in two stages. Stage 1 is
12). The amoebius "tums the handle handle arc twisted, which results in shown in figure 15. You take a type 1
inside out, " transforming it into two two new twisted strips, one of them strip and drag the ends of all other
amoebii so that our flexie becomes embracingthe other. The strip that
strips from under the first one/ sepa-
is embraced canbe rcIeasedby
nothing other than a tripie amoebius, rating it from the rest. For type 2 strips
dragging both its bases along the
shown in figure 8 (marked "N = 3"). edge of the embracing strip. This the situation is different. We can't
This is really an amazingresult, since resuhs in a triple amoebius!
we could have achieved it by replacing (Here we have fuagged the end of
the handle with a double amoebius- a strip along the flexie's edge. Any
an absolutely different flexie! Now flexie can be subiected to such an
operation: the section along the
it's easy to understand that the fusion
flexie's boundary is exceptionally
of k amoebii (k 2 0) and t handies (12 0) elastic, whercas thebase of the strip
yields a (k + 2/)-fold amoebius. So becomes rigid; the sUip moving
combining handles with amoebii gets along its edge stretches the flexie's
elastic edge behind it and
us nothing new. (Notice that from a
compresses it infuont. To an outside
biologist's point of view two-sided- obsewer it looks as if thebase of the
ness is a recessive trait and amoebius- sttip simply slides alongthe flexie's Figure 13
ness is dominant. See, for example, edge.)

OUAllJTUtlll/IIATURt t5
 A is, be transformed
lif,a ^\
*s -"res. into one another
j
according to the %.,
(1 fu.*
fla A natural laws gov-
i\
,:.\ :l
erning the behav-
B
iorofflexies)? The
answer is no, but I
won't give you the
B
' proofhere. Check
'. .,.t*
, r{. ,: i the books men-
A
i"-) tioned at the end of _,",w;.-
the anicle if you're
' Figure 15
cunous.
Figure 14 Notice that our
proof provides an
remove all the other ends from under effective method of determining to
a type 2 strip (otherwise the boundary which of the species a given flexie be-
of the whole flexie would consist of at longs. In particular, it makes it pos-
least two curves). One end of a strip sible to judge when two given shapes
must remain. The remaining strip is are actually different forms of one and
necessarily aiso of type 2. It's logical the same flexie. ..fl
to call this pair of strips a handle (recall Problem. Determine the species of flexie
I
figure 3). depicted in figure 18. Also, show that the
In stage 2 a similar operation moves shapes given in iigure 10 are in fact different
forms of the same bilateral flexie.
everything from under each handle
(fig. 17) and the flexie acquires the
shape of a disk with separate "pure"
To avoid any possible misunder-
standing, before I finish I'd like to
ffi
handles and strips of type 1 or type 2 emphasize that the subject here has ally a topological classification of two-
attached to it. (No doubt you recog- not really been biology but rather to- dimensional manifolds with a con-
nized the latter as amoebiuslike.) If pology-an area of mathematics deal- nectedboundary. It yields, in turn, a
there is at least one amoebius, it will ing with the properties of bodies that topological classiJication of such con-
tum all the handles into double amoebii retain their shape after being arbitrar' nected mani{oids as the sphere or torus
(see figure 12), yielding a representa- ily shrunk and stretched. So I'11give (the surface of a doughnut), since cut-
tive of the amoebius family (fig. 8). you a short dictionary for translating ting a small hole in such a manifold
Otherwisg the flexie consists only of our main ideas from the language of tums it into a flexie.
handles and is therefore bilateral, which biology used here into the language of A more detailed treatment of these
completes our proof. mathematics used elsewhere. "Flex- problems can be found in an excep-
But is this really the end of the ies" are compact connected two-di- tionally interesting book by V.G.
proof? We've established that arry mensional manifolds with a connected Boltyansky and V.A. Efremovich,
flexie belongs to one of the species in b oundaty. " BTlateral" and "amoebi- Topology in Pictues (Moscow, Nauka
the list. Can it belong to several spe- uslike" flexies correspond to oriented Publishers, 1980, in Russian). I also
cies at once? In other words, can the and nonoilented two-dimensional recommend you take a look at Experi-
two species actually be the same {that manifolds with a connected bound- ments in Topology by StePhen Barr
aryl respectively. "Amoebius," as I (New York, Thomas Y. Croweii
mentioned earlier, isthe Moebius strip. Company,I954l. o
The terms "disk" and "handle" are
borrowed from topology and need no
translation. So what we have is actu-

t , ',' '.lr**rl:lt*- ,i;'ifl{1''lilitk ffi

\
'r
l*u*d @
#
....*,,,
I:1
Figure 17 Figure 1B

SEPII]llBIR/OCTOBIR 18gO
 romforyou.

problems like engrneenng, involved in the most exciting

o coreer rn orf ificicrlintell on eorth *the
You'll
complex It won't be
requireiereotive The rewor
You'll
Avioiion offers rncny diverse fields ond the
which require o strong scientific thon 25,000
eduQs;$q1n,.,,1,,
,
PIeose send me more informolion
An Erndli$tl?iddle educotion ;
Imbry-Riddte Aeronouf iaol Univer.$ity.. Or coll''
ioll{r6e ]-800-222-ERAU,' 'i MZ
Embry.fiiddlloi,,pr icr,ns,include
Nome
Address
Ciiir Stq:te-..r+;ip
Phone{rr )
 The Sky's Nof the Limit!
The year 1992 has been declared the krternational Space Year (ISg by the United Nations. Scientists from many countries
will meet atvarious conferences, seminars, and s5rmposiums to discuss the future of intemational cooperation in space. We
hope there will be many new agreements on joint projects, including perhaps one about a joint Marc mission. All these projects
will need many nev/ researchers. Many of them will be arnong those who are presenfly going to higlr school. For this reason
workwithyouthhasbeenanimporbantpartofthelsY. Oneoftheprojectsunderdevelopmentisthe 1992krternationalSpace
Ollrmpiad in Washington, DC.

Summer study in the USSR and US Three-stage competition

To prepare for this oll,rnpiad, several American and Soviet Sixty students from the US and 60 from the USSRwill be
organizations, including the magazines Ku'ont and QuarLhtm, selected, and we expect that students from other countries
the US International Space Year Association, the Soviet will also be interested in participating. The selection process
Aerospace Society "Union," the National Science Teachers As- will be based on the results of a three-stage competition. The
sociation, and the International Educational Network, have questions for the Iirst round are printed below. The second
decided to organize an International Summer Institute in the round will also be by correspondence and will include two
summer of 1991 in the United States and the Soviet Union. math and two physics problems related to space. A total of
The program will feature advanced classes in mathematics, 3OO students will be invited to participate in the third round,
physics, biologr, and other space-related subjects; lectures which will be given at local universities or schools in the
by prominent scientists; trips to major scientific laboratories; presence of the organizers' representatives.
sports and recreation; and marry cultural activities.
Threeweek program
The winners will participate in either the American or the
Soviet part of the program, which will each last three weeks.
The American session will take place July 1-21, 199I, while
the Soviet sessionwill take place August 1-2 1, 199 1. Each
session will feature two weeks of study and one week of travel
in the host country. The winners of the competition, depend-
ing on their total score, will receive scholarship prizes and
a\ilards that will cover all or part of the program costs.

To enter the competition, please fill out the form and mail
it, along with your answers to the questions printed below,
postrnarked no later than December 31, 199O, to:

Dr. Edward Lozansky, President

lnternational Educational Network
300'1 Veazey Terrace, NW
NofuL latreate Sfedon Ctaslanu oJ Hannrd Unircrsfu insfucts
Washington, DC 20008
partbipants in a preuiaus InternationaL Edilcafional Nedpork
(Telephone: 202 362-7 855)
surTtfi'er camp.

Yes, lam interested in the 1991 lnternational Summer lnstitute!

Last name First name
Home address
City _ State Zio Birthdate Sex-
Home phone (_) Parent's office phone (-)-
School name
School address
Phone (

Name of math or science teacher who can recommend you (print first and last name)

Please answer the following questions:

1. When was the ftust manned space ship launched! Who piloted this ship!
2. Who was the first man on the moon!
3. Name all American and Soviet women who have been in space.
4. Write a shofi essay explaining why you would like to patticipate in this progtam.
5. Could you vwite this essay with a ball point pen whiJe orbiting the Earth! Explain.
Teachers are encouraged to copy this page and distribute it to potential participants.

l8 SIPTEltllBER/OCIOBER IggO
 BRAINTEASERS

Jusl [ol' IhB lun ol it

Problems offered for your enjoyment
by A. Savin, G. Galpeiln, M. Lobak, Y. Kurlyandchik,
and S. Sefibekov

811 815
How can a polygonal Winnie-the-Pooh and Piglet went to visit
Iine BDEFG be each other. They started at the same time
drawn in a triangle and walked along the same road. But since
ABC so that the five Winnie-the-Pooh was absorbed in compos-
triangles obtained ing a new "ht)m" and Piglet was trying to
have the same area?

4
812
The product of a billion natural numbers is
equal to a billion. What's the greatest value
the sum of these numbers can have?

813
A glass flask of an ir-
regular shape con-
tains a certain
amount of liquid. Is
it possible to tell
(without any meas-
uring devices or
other containers)
whether the flask is
more or less than o
6
half full? N
6
z
-
(!
B
E
814 Lrl

c
Is it
possible to add i
four digits to the count up all the birds overhead, they didn't
right of the number notice one another when they met. A min-
9999 so that the ute after the meeting Winnie-the-Pooh was
eight-digit number at Piglet's, and four minutes after the meet-
obtained becomes ing Piglet was at Winnie-the-Pooh's. How
the square of an inte- long had each of them walked?
ger? SOLUTIONS ON PAGE 61

OUAlllTUIIl/BRAINItAStRS
 Nl il,xIJlf,Y"';nl:fl,t
I hnds." wrote an eleventh-cen-
I ,- Russian ciuorucier. "I saw
wooden bathhouses, and the people

An inuilation Io the heat them until they are red-hot, and

they undress, and they are naked, and
they pour kvassr over themselves, and
they take green twigs, and they whip

halhhotl$E themselves, and they beat themselves

until they arebarely alivewhen they
leave the place, and they pour cold
water over themselves, and in this
lmagine you're the proprietor of a Russian way they come back to life."
bathhouse-how do you keep the air so You don't have to be a time-trav-
eler, however/ to see Russians enthu-
pleasantly dry? siastically "torture" themselves in the
bathhouses. Why do they do it? Well,
it's thought that the body benefits
by l.l. Mazin when the skin is exposed to high
temperatures (in this case, hot air)for
a short period of time. In humid air,
though, the high temperature can't be
withstood even briefly. So good Rus-
sian bathhouses feature not only heat
but also a dry atmosphere. It's not an
easy thing to build a good banya, and
it's not obvious how you are to " pre-
a
pare the steam" (ot, to speak in more
t
scientific language, provide the opti-
mum microclimate). Our predeces-
sors were able to perform this task
i back in the llth century, but they
I
hardly would have been able to ex-
I plain why things are done this way,
and not that way, in the banya.
Nowadays any teenager has enough
knowledge of physics to answer the
questions that arise in an ordinary
bathhouse. And this is exactly what
I'm now inviting you to do. Let's
analyze a session in the bathhouse
step by step, formulating relevant
physical problems and then trying to
answer them.
And so-to thebanyal To orient
ourselves, let's use the picture on the
left.
We've entered the steam room of
the bathhouse. Wow, is it hot in here!
Let's stay down below where it's a
little bit cooler, at least for the time
being.
-
c Now that we/ve gotten used to the
o
E
O heag we can go up on one of the tiered
o
6
L
1A slishtly fermented drink made
!
! from bread and raisins.

20 stPTttllBtR/0cT0BtR 1gg0
benches. Hot? No, it's not so hot Question 7. Why is the stale afu in since it's not directly related to the
today. Some days you can't walk the steam room forced out by the bathhouse, and concentrate on water
barefoot on the wooden floor let alone steam! in its liquid and gaseous states.
sit on the benches. But if the bench Now thatyou've mastered themles The process by whidr a liquid charrges
has iron n4ils in it, you'll do well to of the banya, let's try to answer all the into a vapor is called evaporation;
steer clear of it even i{ it's not too hot questions. conversely, the process by which a
in the steam room-the head of the The first question is so easy all of liquid forms from its vapor is known
nail will still give your skin a nice you probably answered it right away. as condensation. During evaporation
little bum. So let's skip to the second one. heat is absorbed, while condensation
Question l. Why is it cooler down What happens when you step or sit of the same mass of vapor releases the
below in the steam room than it is on on ahot bench? Yourbody tempera- same amount of heat. The atmos-
the tiered benches! ture doesn't exceed 40"C whereas in a phere always contains a certain amount
Question 2. Why is it possible to good steam room the temperature of of water vapor. For instance, in a
sit onheatedwoodbut not on anfuon the air, and consequently that of the living room there is approximately
obiect at the same temperature! benches, varies from 80" to 120"C. A 10 g of vapor per cubic meter. The
Gradually we're getting used to the process of heat transfer from the hot density of water vapor present in a
steam room. The air doesn't seem so body (the bench) to the coid one (you) unit volume of air is cal1ed the abso-
hot now. Still, we feel that it's rather begins when such contact is made. lute humidity.
humid-there are wet spots here and What's the rate of this process? It Let's bring a saucer of water into
there on the benches and on the floor. depends on the thermal conductivity the room. The water evaporates and
We can fix that. We need to "add the of the hot body. The higher its ther- changes into vapor. This causes the
sfssrn//-lphich is to say, we should mal conductivity, the faster heat is absolute humidity in the room to in-
throw boiling water in small portions transferred from its hotter areas to its crease/ although not significantly-
onto the glowing stones in the stove. cooler areas. Along with other met- the volume of the room usually amormts
Immediately a hot wave rushes up- als, iron's thermal conductivity is sig- to several dozen cubic meters. If the
ward from the stove. It's getting hot- nificantly higher than that of wood saucer contains i0 g of water, the
ter up on the benches, and the heat lby a factor of approximately 300). humidity increases byno more than
dries up the wet spots there and down When you touch a hot wooden bench 1 g/m'. What happens i{ the same sau-
below. you cool anareaadjacent to the area of cer is placed in a sealed flask with a
Question 3. Why does it get drier contact/ drawing heat from just a small volume of only 1 1(10r m3)? The
when water is thrown on the hot volume of the bench. The situation is amount of water in the saucer de-
stones in the stove! quite different if there is a naii in the creases until the vapor in the flask
Question 4. Why must the watet area of contact-the heat is pulled becomes saturated. Thishappens when
be thrown in small pofiions! Why from the whole lengh of the nail and the number of molecules leaving the
can't we iust throw a whole tub of quicldy gathers at the area of contact. waterper unit of time and the number
water into the stove! Also, iron's speci{ic thermal capacity of molecules entering the water be-
Question 5. Why must we use (that is, the thermal capacity calcu- come equal. From this time on, the
boilingwateil lated for a unit volume) is about 40 absolute humidity of the air in the
. . . Over an hour has passed. Many times that of wood; so under similar flask doesn't change. (It's assumed, of
people have been to the steam room. cooling conditiong you get much more course/ that the temperature of the
The air has lost its freshness. It's too heat from an iron object than from a flask is kept constant.)
humid. Leaves from the twigs are wooden one of the same volume.
scattered here and there. The time Now I think you're capable of for- Prrmm Hg
has come to "elean" the steam room. mulating the answer to question 2.
Here's how it's done. The room is (You'll notice I didn't mention either
vacated for about 10 minutes. Lr that the thermal capacity or the thermal
time we have to sweep the floor, hose conductivity of the human body. Try
it down, open the steam room door, to analyze the role of these parameters
and throw several tubs of coldwater in the process on your own.)
on the floor in front of the door. Then In order to answer the remaining
we start "adding the steam." The new questions, let's recall several concepts
steam forces out the stale air. Now of molecular physics, in particuiar
everything is ready again, and visitors those related to water vapor.
can retum and begin their enjoyment As we all know, water exists i.n
anew. three different states: solid (ice), liq-
Question 6. What's the pwpose of uid (what we usually mean when we
the puddle of cold watu at the en- say "water"l, and gas (vapor or steam).
ftance to the steam rcom! We'll leave ice out of our discussion, Figure 1

QUAlllTt|ttl/ftATURI 21
 So for every constant temperature the absolute humidity obviously in- Now that we've answered ques-
there is a maximum absolute humid- creases). Why does this happen? When tions 3, 4, and 5, question 5 can be
ity equal to the density of the satu- a small amount of water is thrown dealt with easily enough. Near the
rated water vapor at that temperature. vigorously into the stove, it tums into puddle at the entrance/ the tempera-
fhe higlrer the temperatwe, the greater tiny droplets. Landing on stones heated ture is below the dew point, so the
the density of saturated vapor. to hundreds of degrees, the droplets cooler "waste" vapor leaving the steam
Here's another relationship that immediately evaporate, and the tem- room quicldy condenses, or "precipi-
comes into play in the bathhouse: the perature of the steam produced is close tates," on the puddle. In well-de-
lower the absolute humidity relative to that the stones. The steam bursts
of signed steam rooms the entrance and
to the maximum humidity attainable ' out of the stove, and the overall tem- the stove are located at opposite ends
at a given temperature, the more in- perature in the steam room increases. so that the steam created in the stove
tense the evaporation. The ratio of the The higher the temperature in the passes through the entire stearn room/
absolute humidity to the density of room, the greater the density of satu- cools along the way, and precipitates
the saturated vapor for a given tem- rated vapor. So despite an increase in at the exit.
perature is calied the relative humid- the absolute humidity, the relative Finally, the last question: why
ity, expressed as a percentage. Since humidity should decrease. And, in does the fresh steam force the stale air
the pressure exerted by the vapor is fact, that's what happens. out of the steam room? When we
proportional to its density, relative Now it's easy understand why the cleaned the steam room/ we threw
humidity can be defined in another water should be thrown in small por- boiling water (no less than 10 kilo-
way-namely, as the ratio (expressed tions. A large amount of water would grams) into the stove. The tempera-
as a percentage) of the partial pressure plop on the stones in the form of a ture of the steam created by this water
of the water vapor to the pressure of huge "drop." Such a"dtop" car.'t is about300oC; at this temperaturg 10
the saturated vapor at a given tem- evaporate as quickly as a small one. It kilograms of steam occupy a volume
perature. starts to boif which creates steam at a of about 25 m3 at a pressure of 1 bar.
Both the pressure and the density temperature of 100"C or a bit higher. We can deduce that in a shortperiod of
of the saturated vapor increase as the This is just what we're trying to avoidl timewehave generated enough steam
temperature increases. The graph in The secret of the banya lies in the to fill about one third of the room. The
figure 1 demonstrates this depend- rapidity of the process. For the same steam is hot, so it rises and forces the
ence. reason we mustn't use cold water. stale air down and out the door.
An increase in the absolute humid- After all, the stones have a rather low Well, we're finished with the phys-
ity at a given air temperature causes thermal conductivity-even a small ics of that venerable Russian institu-
an increase in the relative humidity. droplet being heated to 100"C cools tion, the banya. With this introduc'
The same result can be achieved in the portion of the stone it lands on, tion, I hope you'llmake a point of
another way-by decreasing the air and this lowers the temperature of the visiting one of ourbathhouses ifyou
temperature while keeping the abso- steam produced. ever get the chance! O
lute humidity at a fixed level. The
relative humidity will again increase.
At a certain temperature it reaches
the 100% level-the vapor becomes
saturated, which results in condensa-
our
tion and the creation of fog and dew.
The temperature at which this hap-
WARE!
IIIII
pens is called the dew point. qO !
c
But what happens if the absolute 1. Fill a beaker with o
o
6
i
humidity and the temperature in- water, one test tube 0 o 6 lr
at a time. Record the fitt !
o
6
O
crease? [r this case, the relative humidity height of the water a
depends on whichever increases more after each addition.
;A''E o
O
c
l
o .o6 ol
"]r
o op
quickly: the density of the vapor in c E
.E I
o 6 o
2. Graph your results. 4. Try plotting other contain-
the air or the pressure of the saturated
ers on the same
!
O
o,9a
E a
sl
'I
vapor. o 6
o'E
Nowwe cangetback to ourques-
graph. ;-i = o
o:., lr
I UJ
(L o
tions. o
6 a o
We've made it clear that the evapo-
CE
o o o. c I
o..c
e) o F.
ration rate depends on the relative o O o. I
o q o '=
(not absolute) humidity. If the steam E
c E o
l lr
c o
room gets drier after boiling water is l f
tr E- a
o lr
tossed in the stove/ this means that
trtr .[
the relative humidity decreases (whereas
Circle No.6 on Headers Service Card -IIII
E
:I
22 srPrtt'llBtH/0cr0BrR 1gg0
 HOW DO YOU FIGURE?

Challeltue$ in phy$ics and lnath

tlllAII
M11
Bisector estimated. The lengths of
AB
two sides of a triangle are 10 and 15. Figure 2
\
Prove that the bisector of the angle
between them is no greater thanL2.
Can any them get to the fourth vertex
after several jumps? (Y. Ionin)
\3 -\
(N. Vasilyev)
Ml5 Figure 3
Ml2
Counting pafus of lntegers. Prove
Counting the sides of a polyhedron's
that faces. Provethat arty convex polyhe- neath. Find the bell's mass if its radius
any nonnegative integer n can be rep- dron has two faces with the same is R and the density of water is p.
resented in the form n :flx + yll + 3x + number of sides. (A. Gruntal)
yll 2 with nonnegative integers x and
y, and that such a representation is Pl4
Hot walls or cold walls! The tem-
unique.
Physirs perature of the walls of a vessel con-
tammg a gas is 7. The temperature of
the gas is 7,. When is the gas pressure
P11 on the walls greater-when the ves-
rUheel balancing. Automobile wheels sel's temperature is lower than that of
have to be accuratelybalanced in or- the gas (T . f ) or vice versa (? > 7,)?
der to position the wheel's center of (V. Myakishev)
mass exactly on the rotation axis.
What's the purpose of this operation?
(S. Semenchinsky)
Pl5
Unknovvn resistances. Figrre 4 shows
part of an electric circuit consisting of
Figure 1
Pl2 unknown resistances. Is it possible to
Ml3
A quadrangle in ap arallelogram. A
Breaking a sfting. A weight is sus-
pended from an elastic string. An
find the value of one of the resistances
usrng an ammeter/ voltmeter, battery,
quadrangle is inscribed in a parallelo- increasing force (whose initial value and connecting wires without break-
gram whose area is twice that of the is zero) is applied to the weight until ing any contact? (A. Zilberman) O
quadrangle, as shown in figure 1. Prove the string breaks under a force F,.
that atleast one of the quadrangle's What's the minimum force that must
diagonals is paraliel to one of the be applied to break the string if the
parallelogram's sides. (E. Sallinen) force could reach a constant value
instantaneously? (G. Baronov)
Ml4
Leapfrog. Three frogs are playing-
what else?-leapfrog. When hogA
P13
Thebell. Water ls poured through an
jumps overfrog B, it lands at the same orifice into a hemispheric bell lying
distance from B as it was before the on a table, tightiy pressed against its
jump (an{ naturally, on the same line surface (fig. 3). When the water level
AB-see figure2l. Initially the frogs reaches the orifice the bell lifts up and
are located at three vertices of a square. the water begins to flow from under- SOLUTIONS ON PAGE 58

OUAIIITUltl/llOlll DO YOl TIGURI? 23

 MATH EMATI CAL SU RPR SES I

An old lact and $olne ltBW olte$

Another excursion into the region of shape-numbers, number-
shapes-mysterious numbers that may put a spell on you!
by John Conway

DD UP THE FIRST n odd numbers, starting from 1. The cubes are just the sums of the first few of these:
What do you get?
1 : 1, 1 +7 =8, I +7 +19 =27,....
t:1
1+3 : 4 Boring! Why should anyone be interested in these myste-
1+3+5 : 9 rious numbers?
1+3+5+7 = 15 Let's try again.
1+3+5+7+9 = 25.
trom at{ftmelh
Do you recognize these numbers? You probably do. The sum of several numbers is just the number of them
They're the square numbers multiplied by their average. So, for instance, the sum 1 + 3
+ 5 + 7 + 9 is just 5 times 5 (the middle number). The
1: 1. 1, 4:2.2, 9 =3.3, l5:4.4, 25:5.5,.... averages for the sums

This is our oid fact. Can we explain it? Generalize it? Can 1, 1+3, 1+3+5, l+3+5+7,...
we get the cube numbers
are indeed
1= 1. 1. 1, 8=2.2.2, 27 =3.3.3, 64=4.4.4,...
1,2,3,4,...,
in a similarway?
Well, there are lots of ways to explain this fact, and they since they can be found as half the sum of the first and last
lead to lots of different new facts. Let's try. terms.
This does give us a nice way to get the cubes. Rather
than always taking the sum of the first n odd numbers, we
Fom aluetna
take the sum of the next n, starting from where we left off.
What we have to prove is that the differences between
So, instead of
adiacent square numbers 0, t, 4, 9, 16, 25,... are just the odd
numbers l, 3, 5,7,9, 11,... . Butthis is easy-the typical rl
1
dilference is
1+ 1
J a)
1+3+ 5 - 3',
(n + l)2 - n2 : ln' * 2n + ll - ln2) = 2n + |
:
we have
by some easy algebra,
You can see in the same way that the differences 1 1.,
between adjacent values of any polynomial are the values
of a polynomial function of the next lower degree. So, for
3+ 5 )r

instance, adjacent cubes differ by the numbers of the form

7+9+ 11 30,

3n2 + 3n + l-nameiy, the mysterious numbers

That's a bit nicer, but let's try yet another idea.

1,7 , 19,37 ,51, ... .

24 stPItllllBtR/0cr0BtR 1gg0
 Ft'om UeomeFy It seems there are lots of ways to generalize our basic
Iact, andthis is nicg because it means that no matter how
fust look at this pattern:
long you live, you'll always see some new ones. Here,s a
striking one that was discovered only recently.

[l,Ioessnen's maUit
First, the squares. We write down the first few whole
It shows very clearly how a typicai square number can be numbers, circle every second one, and then add up the
broken up into consecutive odd numbers. Is there away to others:
see our mysterious numbers 1, 7 , 19 , 37 ,6 I as patterns of

6 @ 6,@ 6,ca b@ b;
dots? The way that seven pennies naturally arrarLge
themselves into a neat figure

Of course, we get the squares.

Now, instead, circle every third number in the top row,
get the next row as before, circle every second number in ig
might suggest something . . . Yesl The mysterious num- and add the uncircled numbers in this row to get the last
bers row:
t7 t9 o/

::ii::.
tt tt
t -t'

. ... i,
. . at
a- a

Surprise! There are the cubesl Can you show that this
continues forever?
are what combinatorialists call the "hex,, numbers. This seems toworkforallpowers. Forinstance, if we
(Beware-they use the term "hexagonal numbers,, for circle every fifth number in the first row, every fourth
something else!) number in the second, and so on/ we at least get the first
Is it really true that the sum of the first few hex numbers three fifth powers correctly:
is a cube number? Yesl To see this, just draw in some lines

13 e frt @ 6t5 23
7 B \@,,.,.
L2 3 4
qL qql sl
| 4 do-l 26 49 r80i
rGt Y - 3t @) -
131
ztr
and those pattems of dots become cubical shells.
o" @-@
My daughter had some hexagonal blocks like the ones Can you verify that this, too, will continue forever?
pictured below, which she used to stack into hexagonai In general, you can circle any selection of numbers in
pyramids. How many blocks did she need to build a the top row, thus dividing the remaining numbers into
pyramid n layers high? blocks; then rn subsequent rows you circle the last number

O 1990 Geomety Supercomputer Project

Thesepictwes werc rendercd by Toby orloff at the ceometry Supercomputer Prciect and printed on a color laser imager
dweloped at 3M.

O U A III I U lt| / tll A T II E llll A T I C A I. SURPRI$ I S

Ln each block, sum the uncircled numbers to get the next Answel's
rou., and so on.
Of course, my daughter used exactly n3 blocks to make
We've seen that i{ in the top row you circle the numbers
her pyramid of n layers.
Yes, those cubes do continue forever! It's quite easy to
).+1,2+2,3+3,4+4,... proye this by working out the general form of the pattem:

you get the squares-namely, the numbers

3n- t E'p,T.'1u.,
t l, 2.2, 3 .3,
. 4 . 4, ...,
@
@
@ @@
while circling It's a bit harder to prove that the kth power rule works
for allk. We'lI publish the best proof we receive.
1+ I + l,2+2+2,3+3+3,4+4+4,... From the triangular numbers
gives the cubes
l, l+2, l+2+3, l+2+3+4, l+2+3+4+5,...
1. 1. 1, 2.2.2, 3.3.3, 4.4.4,.... the Moessner magic leads (of course!) to the factorials:
What happens when you circle the trrangular numbers
r, 1.2, 1.2.3, 1.2.3.4, 1.2.3.4.5,.... o
l, l+2, l+2+3, l+2+3+4, l+2+3+4+5,...?

ShirtWear!
Broadcast your pride in physics
with aL\\Vo cotton T-shirt from AAPT! Get "Molecules Moving, HUG a Physicist" is perfect for
the student who wants to stand out. Available in neon yellow and black. For your teacher, "Another
Awesome Physics Teacher" has nary letters on white or gray, or white letters on a teal T-shirt.
Only $13.75. And if you're an AAPT member and order more than one shirt, your ?-\Vo discount
means your T-shirts will only cost $11-! To order, send your check or money order to

Aneican Assrciation of Physics Teacherc (AAPT)

5112 Beruyn Road, College Park, MD 20740
301134s-4200 Fil( 301134s-18s7

20 STPTI]IlBER/OCIOBIR IggO
 FLUID STATICS REVISITED

Boy-oh-[uoyaltcy!

We hope you won't get dunked by these probtems from Soviet

university entrance exams

by Alexander Buzdin and Sergey Krotov

IRST, WE'LL REFRESH YOUR sure. If a liquid (and when we say The pressure in a liquid is govemed
memory by reviewing the basic Iiquid we'll also mean gas) is at rest- by Pascal's law: pressure applied to a
iaws of fluid statics. that is, in static equilibrium-viscos- liquid is transmitted without change
A Iiquid or gas/ when it moves ity doesn't appear because the liquid to every part of the liquid. If only the
as a unit, constitutes
a mechanical friction emerges only when layers of force of gravity is applied to a liquid,
system in which different parts inter- Iiquid move relatively to each other or the pressure p increases with depth ft
act with each other only through pres- a solid body in contact with the liquid. according to the law p = pgh, where p is
the density of the liquid. Therefore,
different forces of pressure are applied
to various parts of a body immersed in
the liquid so that there is upward (lift)
force, which is called the buoyancy
force. This phenomenon is described
by Archimedes's law: a body totaliy
immersed in a fluid is pushed upwards
by a buoyancyforce equal to theweight
of the fluid displaced by the body-
that is, the weight of the fluid con-
tained in the volume of the body.
We should notice that Archimedes,s
law can't be applied in the situation
when a body is tightly pressed to the
walis or the bottom of a container. For
example, a submarine lying on silt is
taken to be pressing against the ground
and not being pushed upward.
Now let's turn to some specific
problems.

Problem l. One pan of a balance

holds a glass of watu and a stand
from which a weight is suspended
(fiS. 1). What will happen to the
E
o balance's equilibrium if we lengthen
! the string so as to immerce the weight
t
in thewateil

OUANTUltll/IIATURE 27
 The most common answer/ and the or iJ the stand were on the other pan of
wrong one, is that the equilibrium the balance?
will be disrupted. Some students say
that the buoyapcy force acts on the Problem 2. A glass containing a
weiglrt in accordance with Archimedes's smailballfloats in a vessel of water
law and decreases the tension of the (fis. 2). How does the level of the
string so that the pressure of the stand watu change if the ball, which is
on the balance pan is decreased as made of either steel or wood, is uans-
well. Others say that because the .fered fuom the glass to the vessell
weiglrthas been immersed in thewater, The force of pressure on the bottom
the water level will increase so that of thevessel equals the weight of the
the pressure on the bottom of the glass watert the glass, and the balI. If we put
will increase as well and the left pan the vessel on a balance, which for
simplicity's sake we'll consider weiglrt-
Figure 2
will drop down.
To get the right answer we need to less, it will indicate the weight of the
notice that the contents of the Pan total contents. It's important that its
don't depend on the position of the reading doesn't depend on whether Problem 3. A steel ball floats in
weight, inside or outside the water; the ball is inside the glass or in the mercury. Watu is added so as to
consequently, the equilibrium of the vessel of water. On the other side, the cover theballwithwater (fig. 3). How
balance will be preserved. Now let's balance must indicate the force that does the depth of immersion of the
find the errors in the arguments the acts on the bottom of the vessel, and ball in the mercury change!
students offered. initially the force was determined only You're tempted to apply Archimedes's
We'll take into account the fact by the level of the water in the vessel. law right awayt atenttyou? But the
that when the weight is lowered into If we transfer the wooden ball, it difficulty is that different parts of the
the water the tension of the string is will float on the surface of the water, ball are in different liquids so that we
decreased owing to the buoyancy force andthe force acting onthebottom of can't consider the ball as a whole
acting on the weight. So the force of the vessel will be determined by the while applying this law.
the pressure of the stand on the pan level of the water. Since the force Let's choose a small area of the
also decreases. But according to doesn't change, the water level must ball's surface inside the mercury and
Newton's third law, the force acting also remain the same. find the force of pressure acting on it.
on the water and the bottom of the The result will be different if the It's not hard to see that it equals
vessel will be increased by the amount ballis made of steel. Such a ball drops
of the buoyancy force. Therefore, the to the bottom of the vessel, and the f =brshr+ P2gh2)aS,
pressure of the glass on the pan in- total force of pressure on the bottom
creases. We see that the decrease in comprises the force of pressure of the where p, is the density of water/ p2 is
the pressure due to the stand will be water and the force of pressure of the the density of mercury, and AS is the
compensated by the increase in the ball. Therefore, the level of the water area. Let's cast this equation in the
pressure of the glass on the pan. From in this case will decrease. form
this we 8et the correct answer: the A similar problem involves a glass
equilibrium will be preserved. in which a piece of ice floats with (1) a f =lorslt ,+hr)+lpz-p,)shrlls =f ,*f,.
Can you figure out what wiii haP- piece of cor( (2) a small lead pellet, or
pen with balance i{ you put your finger (3) a bubble of air embedded in it. How Next, we'll sum the forces of Pres-
in the glass of water without touching does the water levei change in the sure that act on all areas of the ball's
the walls and the bottom of the glass, glass after the ice melts? surface that touch both the water and

Figure 3

28 stPIt]IlBtR/0tI0BtR l0g0
 the mercury, obtainiirg the two forces we need to know where the force is F,=p N.
F, andFr. Thefirst force, applied. Let's imagine a certain region
in the liquid. In the state of equiiib- Because the plank is in a state of
F:r=Prg(vr+vrl, rium the force of gravity that acts on equilibrium, the sum of aII forces act-
the region is balanced by the buoy- ing on it is equal to zero. Let,s write all
is the buoyancy force that acts on the ancy force. The angular momentum these forces by using the projections
ball when it's immersed only in water. of the force of gravity with respect to on directions tangent and perpendicu-
The second force, the center of mass of the region is lar to the plank:
taken to be equal to zero. Conse-
F, = (p, _p)sv, quently/ the sum of angular moments 2xpo Sg sincr-gl Sg sino + pN = 0,
of the forces of pressure is also equal to 2xpo Sg coso-g/ Sg coss + N: 0.
is the buoyancy force if the ball is zero. If we replace the liquid with a
immersed down to the level occupied rigid body of the same shape, we can From the equations given above we
by the mercury in liquid with a den- convince ourselves that theforces acting injer that
sity of pz - p' The resultant buoyancy on it from the surrounding medium
force equals don't change. So we can inJer that the p: rg0(.
forces of pressure are equivalent to a
f :Fr+Fr=p,gV, +pzgVz. {orce that acts vertically through the On the other hand, from figure 4 we
center of gravity of the displaced liq- inJer that
We see that the force is resolved uid. It should be noticed that we,ve
into two terms-the first for the wa- found only the line of action of the )
ter, the second for the mercury----cor- buoyancy force, but we can't say any-
(t-2rf
responding to the parts of the ball thing about the point at which it,s The quantity x can be found from
immersed in either liquid. We may applied. the requirement that the sum of the
say there is a principle of independ- Therefore, in our case the buoy- angular moments of all the forces act-
ence of the buoyancy forces so that ancy force has an upward direction ing on the plank be equal to zero. It,s
either liquid makes a contribution to and passes through the center of the convenient to consider the angular
the resultant force, even though we immersed part of the plank (the center moments of the forces with respect to
might think that mercury pushes the of mass of the displaced water). Let the point at which the plank touches
ball out while the water presses it the area of the section of the plank be the stone (point O)because the angu-
against the mercury. S, the length of the immersed part of lar moments of the friction force, and
Thus, the water appears to help the the plank 2x, andthe angle that the the reaction of the prop, are equal to
mercury hold the ball up so that it plank forms with the horizon (the zero at this point.
emerges a little from the mercury, and surface of the water) o. Then The buoyancy force passes through
the depth of the ball's immersion in the center of mass of the immersed
the mercury decreases. Fo = 2xpoSg part of the plank so that its arm with
and respect to point O equals (1- x)coscr.
Problem 4.Athinplankof lengthl The arm for the force of *a\nty equals
is propped up at its higher end by a Mg=pS1g. Yzlcoso^ The equation for the total
stone that emerges above the water's angular momentum reads
surf ace to a heightH (fi7. q. What is Since we're interested only in the
the minimal coefficient of ftictionbe- minimal value of the friction coeffi- I
p/S;lcosa- (1-x)coSu = 0.
pn2xS
tween the stone and the plank that is cient p, we can assume that
needed for the plank to remain at rest!
(We'Lllet the densities of water and
woodbe poand p, rcspectively.)
Four forces act on the plank (l) the
force of gravity Mg, l2l the force of
reaction of the prop N, (3 ) the force of
friction F,, and (4) the buoyancy force
Fo. The first force is applied to the
center of the plank, the second and
third at the point where the plank
touches the stone. The force N is
directed along the normal to the plank,
the force F, along the plank.
IJntjl nowwe had to know only the
magnitude of the buoyancy force. Now Figure 4

OUAlllTUll/l/fIATURI
OI with the depth of immersion h accord-
AP ing to the linear law p : pgh. The resul-
x2-lx+t' =0. tant force of pressure is directed hori-
4Qo
zontally, and the problem is to find its
From this *" gLt magnitude. To solve ig students usu-
ally introduce the formula
_(
..-ll, lp
'- F:p
I

t']%,J LAV S,l

'['-
We've dropped the second root be- with S = 7H (the area of the wall in
cause it doesn't satisfy the constraint contact with water) andp", being the
2x < l. Finally, we obtain average pressure, which is equal to the up23
=PBH tHE = 208.3 N.m.
pressure at the middle dePth. The
l-t = answer is the right one, but we need to You'llnotice that in this solution
know why this formula was chosen we haven't allowed for atmospheric
r'(r -r)-a' for the average pressure. pressure. Can it change the answer?
Let's consider a rectangular Prism One more question is whether the
The problem of the angular mo- made of amaterial of density p and force of pressure on the wall changes if
mentum of the buoyancy force is of height l that has an isosceles right the aquarium wall is made of rubber.
primary importance for studying the triangle with a side H as its base. Place
equilibrium of floating bodies. In fact, the prism on a horizontal surface (fig. Problem 6. What happens to the
an important concept used in shiP- 6). It's easy to convince ourselves that depth of immersion of a balJ floating
'building the force of pressure of the prism on in a glass of water if the gJass begins to
is the "metacenter," which
is where the line of action of the buoy- the surface equals the pressure of the mov e with accelerution upwar d!
ancy force of a ship in a slanting water on the side surface of the aquar- Let's consider a system comprising
position intersects the plane of the ium because of the fact that pressures the water and the ball floating in it,
ship's symmetry (fig. 5). The meta- are ecluaily distributeci on the contact and let's suppose that it moves uP-
center (point M) isn't allowed to de- surface. But the force of pressure for ward with an acceleration a. The
scend below the ship's center of grav- the prism is its weight; so we have acceleration is caused by the interplay
ity (point O); if it did, the angular of the force N,, which is the pressure
psHH psh. exerted by the bottom of the glass, and
momentum of the buoyancy force !- ='-_:-= .
'21 lH = 1,250N.
2 the force of gravity of the system (M +
couldn't retum the ship to its upright
position. Consequently, the mean pressure of mlg where M is the mass of the water,
water is to be taken as the pressurq at m is themass of theball, and
Problem 5. Anaquailumof rectan- the middle depth.
gularcross section is fiJled with water The second question of the prob- N"-(M+ ml1=lM+m)a.
(whose density p = 10s kglms) to the lem is more difficult because both the
heightH= 0.5 m. Findtheforce acting pressure and the arms of the corre- When the system was at rest, the
on the aquarium wa|L (whose length I sponding forces depend on the depth. pressue Af, at the bottom of the glass
= 1 m) and the angular momentum of Sometimes, drawing an analogy from is determined by the equation
the pressure on the wall with rcspect the preceding result, students propose
to its lowest edge. to use the average force of pressrue and No- (M + m)g = 0.
In this case the pressure changes the average arm, equal to HfZ, for
finding the angular momentum of the Comparing the forces N" and No, we
forces of pressure. But this is quite get
Nu
wrong. To get the right answer we'll -a*8
use a different analogy, one based on
Nog
the prism mentioned above. The cor-
responding angular momentum for Let's show that the ,r"rr*" ala1
the prism M" equals the product of the
^t
point of the liquidhas increasedby the
force of gravity on the prism and the same ratio. Imagine a water cylinder
arm taken with respect to the straight of section AS, one of its bases is at the
line AA'(fig. 6). Since the center of surface of the water and the other is at
mass for a homogeneous triangle is a depth h. We can write the following
the point of intersection of its medi- equations of motion for the cylinder
ans, the line of action of the force of in the upward direction:
gravity is at a distance of ( 1 /3 )H from
the edge AA'. Thercfore, we have Ph AS - pAShg = PLSha,

30 stPrttllBtR/0cI0BtR lgg0
 where Pnis the preisure of water at a Let's write the equations of motion Exercises
depthT: and p is the density of water. for this region by using projections of 1. A mercury manometer (fig. g) consists of
We see that the {orce and acceleration in the verti- two tubes with cross sections S, and S, such that
Sr/S, = 2. Find the change in measured pressure if
cal and horizontal directions:
' Pn:p@+ alh.
the level of mercury in the first tube increases by
Ah = 10mm.
Ncoso-Amg=0, 2. A frmnel of mass 1tZ, which has the shape o{
That is, it has increased by a factor of Nsincr:Ama. a truncated cone with a base of radius R, stands
(g + a)/g compared to the static case. on the table. The edges of the funnel are tightly
Consequently, the buoyancy force lras pressed against the surface of the table. How
From this we infer that
increased by the same ratio. Now we much water must be poured into the funnel if, at
the moment it breaks awayfrom the table, the
write the equation for vertical motion tna. =!. water level in the funnel is equal to I?
d
of the ball: 3. A cylindrical weight suspended fror4 a
so that the angle of the slope doesn,t springbalance is lowered into a vessel ofwater
a+p until the waterlevel is changed by Al = g cm
PVrg-;a-mg = ma. depend on the choice of point A. {fu.
9). The reading of the spring balance is changed
Let's choose a plane inside the water by AF = 0.5 newon. Determine the vessel,s cross
From this *" irier that the volume V that is parallel to the surface of the section.
oJthe immersedpart of the ball dudng water and at a distance ft from itin the 4. Where does a gas bum better, on the ground
the accelerated motion of the glass vertical direction. Let,s show that the floor or the top floor of
a fourteen-story building?
doesn't depend on the system,s accel- 5. A wooden ball floats in a glass that is filled
pressure of the water at ali its points is
eration and equals V m I p. Conse- with waterup to the brim and closed on top. How
given by the formula does the pressure o{ the ball on the cover change
quently, the depth oI^=the part of the if the glass is moved upward with an acceleration
ball immersed in water doesn,t change. Pt':PSh or a:
o
Problem 7. An aquarium in the To this end let's imagine a slant cylin-
shape of a cube with edgeL is half der of slant height h and base AS. SOLUTIONS ON PAGE 61
filled with water. Find the shape of Since it doesn't move in the vertical
the surface of the water in the aquar_ ciirection, the sum
of ail the vertical
ium andthepresswe atpointMif the projections of forces acting
on the
aquarium moves in the horizontal cyiinder is equal to zero:
dfuecttonwith anaccelerution a(a.
(fis.7).
d
pgTr AS cosu - Pn AS coscr : 0.
We'll show that the surface of the
water is an area in a plane that forms Here the first term is the force of
an angle a with the horizon. gravity for the cylinder, the second the
Imagine a small region of the liquid vertical projection of the force of pres-
-
of mass m close to a point A on the sure on the lower base. From this we
surface of the water. The resultant of get
forces of pressure from all other parts
of the water is normal to the surface at
Pn: PSh'
that point. Let it be equal to N and
form an angle cr with the verticai. The Therefoie, the surfaces of constant
fixed area of the surface, which we can pressure are planes paralle1 to the free
consider flat because it,s small, then surface of the water.
forms the same angie with horizon.
(Can you e4piain why?) - To find the pressure at point Mr
let's notice that the middle point tr
remains at rest because of the incom-
pressibility of water. So we have

P-= pse*r,so)= o!@*s1

We'll leave a question for you to

answer: have we employed the condi-
tion a < g? Also, find the forces of
pressure of the water on the walls and
bottom of the aquariumwhenitmoves
with acceleration.

Figure 7 Figure 9

OUAIIITIII|l/TEATURE 31
 , r';,$ -rEr
e3,r
l
.)
f*
J.-

t
|\
ba)
+-r
o
su
'+iJ
'J \\, \"'..
CU
ct) \
O\'r
s
et
E \r\
'\'r
(\'t\

o
q) o\\.
\-r\ ,-r--_J*
,\\
'_"

t,
o \
\
j--,...
a, H\

\.- ,\
i..-.-

HQ
qJ
z -ot
q5 s)
\ -- -.\_+
', '\j ..
\'..\ \
\
I --os *._
'\ti__.
-..
#
:s
tu
q) ..', ,l
^\ -:--- - /
0u CDf, *i - - '< e-'^/ .l'
ct)
oa) ".._-er
\
o' SN ""

\'t
sF- t\'-H -tn

I
L*-

giE+
C=T
:=r :i EE s :f;j t ; ;
iE E:;EE= E * E
C5 F-
E E $ [E!i6 EEgillqEEaiEi [i Zi ,
-II
CE'
1-
1-
C=T a-
I
F-
U
U
CE' ?212i1E?!iiEEE:iiEisttzi*;;
trEl
-
-
=I:, E*i,E?=E E sir;j EiE:jE iE it:
=;E =ii'==r-zzlil.i=?',9E Etz=-a
= =
 lrv
^ouE^l ^a6ras ^q

EEHil;$E FHEE'
I
E*aeCt;eeg sE$S H

i
\ EtEgfEC*e+
I33EEz=RE siH-
SSB) $
O

I Et}fE*E; !EE; 3 LO
O

EEE;+tre:= :8,s1
I L
f
o)
H LL

!l;tgEIiE*: "SEi!
E#EEEEse
=

=P$:*
EEEr E3-rE
-ssEt EsH{-E
rF isI f l.
flii:e E*tls ssn nl $
o

srs'Hu l?l$ss
l
f,;F#E tES ig ir
o)
=."8*:E
AH:! "; T {iS E$
Er
\ ''",
il
fl;;gE E+H5$g sEE U
+
li
HEEEE
Ei$ f,fl cq
+
r I '..., li 'EE:$E F-
Q
I
t-a)
FS

!i
c)
''ll tL
ts--)
\o)
riI

,
I ,,1
.--.----*d--

{-
rl
ll iIiEiE{Igiigii$-gii
*..--_*_
\l
-. -J I

EtggttigEi:gggcEEEi N
o
ff FNE{s E$Esi i uEEs= 8hE E4 l

EH Hrs:* EE*5E- $
o)
gf;EiE*{gE ig ir
gf tESst :sEE= ;E $tEEEgg Ei i
HE FNs$iu !i$l:i n* nrf;+sEf;
AS iE
;EE SEf$E: xfcse= *nEEEEEEEH
cg Ess$Ee lteEgt !S gEgtEEegFif r

EEB
eefgrr Eg3igj EE C)
f
O)
ir
=gEjEHsEggig
 IN THE LAB

Thl'otluh a Ulass hriUhlly

Theres more to green glass than meets the eye

by B. Fabrikant

HAT COLOR IS GREEN GLASS? in alcohol, you can get a chlorophyll solution and perform
This question may bother you. You'll probably the following experiment.
answer that green glass is called green because Put a giass on a sheet of white paper and slowly pour in
it's . . . But don't be in a hurry to offer conde- the chlorophyll solution. The bottom of the glass will
scending explanations. A simple experiment will show appear green at fusq then, as the layer of solution increases
you that the question of the color of green glass is not as in thickness, it will take on a deep red color.
simple as it
seems. Let's get back to the green glass. We can muddy the
If you have a piece of green glass, break it carefully so as problem of its color even more if, after the filament, we
to get several pieces (not too smail). Then iook through iook at the end of a red-hot poker. With only *ree pieces
one of them at the incandescent filament o{ a clear light of glass it will already appear ruby-red. So here we have our
bulb. As expected, the filament appears green (fig. 1). Place second unexpected result: the visible color of glass de-
this piece of glass on another one and look at the filament pends not only on its thickness but also on the properties
again. of the object we're observing. Three pieces of green glass
You probably won't see any change in the color of the layered together look discolored when we look at the
filament-it'll appear green, as before. But i{ you lay a third filament but red when we're looking at the red-hot poker.
piece of glass on the first two and look at the filament We can perform another experiment that has a practical
through al1 three, you'll see that it's discolored and whit- outcome. When taken out of the fireplace, the poker cools
ish. The filament will appear reddish when seen through very quickly. Try observing the poker as it cools down. As
four pieces, and ruby through five pieces. we noticed earlier, the end of the poker appears red through
This result is totally three pieces of green glass. After cooling a bit, though, it
r.rnexpected and rnstruc- appears red through only two pieces. If you wait a little,
tive. It turns out that you'll see the poker as red through a single piece of glass.
the color of glass depends Our experiment shows that the higher the temperature of
on its thiclcress, and glass a red-hot objecg the more layers of glass needed for its color
to change. So we can estimate the temperature of a red-hot
: .ti::"; ', not so thick changes its object by the thickness of the glass needed to change its
".,.,.
, ..;,.,;,.1,1ir1.;,lr:;!i:7,\. ; ,. COIOf tO fed When it iS color.
k- lo cluite thick' Not every This experiment with the poker helps explain the
. a:-':' : i.
.'"" ,:,,,,,:.t,;..#,*;i *,,r"'Y,
i kind of green glass has desrgn of a simple yet ingenious insrmment for determin-
;ht;-proi"r,v, f,ut most ing the temperature of red-hot objects-the optical py-
rometer lhg.2). It consists of a wedge of green glass whose
thickness gradually increases from one end to the other.
teresting that this prop- The wedge can be moved in a metal holder with an
erty is characteristic of opening for observing red-hot objects. A temperature scale
the most common dye runs along the edge of the glass wedge.

Fisure 1
Alteration of thevisible
i,l,tflTk"lt?i".'^i#.
color of phvll
The opening is aimed at the object under consideration
and the wedge is moved inside the holderuntil the color o(
eives the leaves o{ the object seen through the opening changes. Then a o
on lnconaescent ruament rrom c
green to rcd for iiff"r^i plants their green color' reading is taken from the scale at the point opposite the :c
"o7
pi"""t of "luiUurt
green'glass. By putting some leaves opening, andin this way the temperature of the object is o
o
ao
o
34 $tPIt]I/lBtR/0Cr0BtR lg90
:
 / 'ia..
.', ..s

*-,, /

\ .,lq.
. ad,

t .,\A-t6" .li
 asceruained The and dark-bluebeams to pass but absorbs the green ones.
opticalpyrome- Before your anazedeyes the whole of nature is completely
teriswidelyused transformed, and under the usual dark-blue sky we see
todeterminethe dark-red vegetation. Do the troubles that landscape paint-
temperature of ers continually have to overcome lie in this specific
molten metals property of chlorophyll? No doubt the painter's palette
(for exainple, in doesn't include the green hues peculiar to brightly colored
open-hearth green plants."
. {urnaces). De- But let's leave painting for now and retum to the optical
spiteits simplic- pyrometer, making a few changes in the experiment with
ity, in experi- the wedge of green glass. We'll use an incandescent
enced hands it filament as the fuht source and put the optical pyrometer
Figure 2 provides a high between it and a prism (fig. a). Again two bands willbe
Optical pyometer for determining the degree of accu- thrown on the wal1, green and red; the relative brightness
temperatwe of red-hot obiects. tacy. of these bands will depend on the thickness of the wedge
So now you know about a clever application of the at the point where the light passes through it. If the beam
strange properties of green glass, but the mystery of the passes through the thin part of the wedge, the green band
green glass itself remains unsolved. is brighter than the red one. As the thickness of the wedge
increases, the brightness of the green band diminishes
and, after a certain point, the red band will be brighter.
An erueriment llleuJmlt didn'tper{orln altd a l00k a[
When the green band is brighter, the filament is seen as
landscaE [ailtilittU green, when the reverse is trug the filament is seen as red.
I'm sure many of you remember Newton's famous If the two bands are equally bright, the filament appears
experiment in which he split a beam of sunlight into a discolored.
rainbow-colored band (the visible spectrum) by means of a So the mystery of the green qlass seems to be solved. But
glass prism. The experiment showed that sunlight is a it remains to be explained why the ratio of the brightness
mixture of beams of different colors: red, orange, yellow, of the red and green bands is inverse as the thickness of the
greeq blue, dark blue (indigo), and violet. For some reason, glass increases. To get an answer, we need to look at an
Newton didn't perform a more sophisticated experiment important optical law discovered by a French scientist
by putting colored glass or a beaker of colored liquid in the about 200 years ago.
path of the sunbeam. At least, he never mentioned any Pierre Bouguer was the fust to focus on the problem of
such experiment. measuring the intensity of light and illumination. He
As it tums out, i{ red glass is used, the experiment gives devised the first instruments for measuring the intensity
us nothing new. Instead of a multicolored spectral band, o{ light, discovered that the intensity of the Sun's Iight is
we get only a narrow band corresponding to red beams. 300 times that of the Moon, and in his Optical Treatise
This result could have been predicted beforehand: the red formulated the important law describing how the inten-
glass is red precisely because it allows only red fuht to pass sity of light fiminishes in absorbing media.
through it and absorbs all the others. To understand the meaning o{ this law, which we'll call
An experiment with green glass or a beaker filled with the Bouguer law, we'll employ an analogy borrowed from
chlorophyll solution is much spolts-not particularly accurate but conveniently graphic.
more interesting. hstead of Imagine that we're watching a seven-kilometer race. It
one band, two bands remain- tums out the race is rather poorly organized. The partici-
green and dark red. This pants'lack of training becomes apparent right at the out-
means the green glass and seg and the observers soon discover the following interest-
the chlorophyll allow not only
green but also red beams to
pass through.
The famous Russian sci-
entist KA Temerezyaev made
the following very interest-
ing observation about chlo-
rophyll: "It's very easy to
convince yourself that chlo-
rophyll allows only redbeams
to pass-it's enough to look
Figure 3 at a sunny landscape through Figure 4
A green landscape seen a piece of special dark-blue Newton's experiment: the optical pyrometer allows only the
througb dark-blue glass. glass (fig. 3) that allows red green and red components of white light to pass tfuouglt.

36 SIPII]'llBIR /0cT0BrR 1gg0

ing law: only one third of the runners that begin a given Let's jot down the numbers of runners at each stage of
kilometer make it to the end of it. There were 2,187 the racg the greens above and the reds below:
participants at the start of the race; at the end of the first
kilometer, 729 remain; at the end of the second, 243; atthe 2,t87 729 2A3 81 27 9 3 I
end of the third, B1; at the end of the fourth, 27; atthe end stz 256 t28 & 32 t5 I 4
of the fifth, 9 i at the end of the sixth, 3. Only one runner
makes it across the finish line at the end of the seventh Lr the second line each number after the first is one half
kilometer and is naturally declared the winner. the number on its left; in the first line, as previously, one
Let's write out the numbers of.runners that ran the thfud. It tums out that this small difference in the divisors
different distances: 2,187, 729, 243, 81, 27, 9, 3, 1. It's easy is enough to compensate for the great numerical advan-
to see that these numbers form a geometric progression in tage of the green team and lead the red team to victory. The
which every number after the first is one thfud its predeces- race just needs to be long enough (at least 4 kilometers)."
sor (standing to the left of it). The behavior of green and red beams is analogous to the
Nowlet's getbackto optics. Take apiece of a colored performance of the green and red teams (fiS. S). The green
glass. Suppose it allows one third of the incident light to glass transmits dark-red beams better than it does green
pass through. Add another piece of glass similar to the ones, and according to the Bouguer law the difference in
first. It transmits one thfud of the light that passes through the aansmission of these beams increases rapidly with the
the first piece-that is, one ninth of the light that falls on thickness of the glass layer (the "long-distance effect" in
the first piece. Adding another piece, we get one twenty- the race).
seventh part, and so on. Obviously, the same result is The question arises: why does the glass appear green
obtained by doubling tripling and so on, the thickness of for a thin layer if it transmits red beams better than green?
the glass. The explanation lies in the spectral composition of light
We can conclude that if the thickness of the glass from the incandescent filament used in the experiment-
increases, the ratio of transmitted light decreases in a its green band is much brighter than its dark-red band. Lr
geometric progression. This is the law Bouguer discov- a thin layer (a short race, according to our analogy), the
ered. As we saw in the example of the race, numbers in a difference in the absorption of dark-red and green beams
geometric progression can decrease very quickly. isn't large enough to compensate for the initial advantage
of the brighter green beam, so that the green beam plays
A little lnol'e $pom the maior role and provides the dominant color.
Equipped with the Bouguer 1aw we can attack the Now we need only explain the part played by the
mystery of the green g1ass, but first let's revisit the seven- temperature of a red-hot object viewed through the glass.
kilometer race. Suppose the novices who ran so poorly It's widely known that the higher the temperature of a
were conceited enough to challenge experienced athletes heated object, the whiter the light radiated. For examplg
to arace. These well-trarned athletes accept the challenge at low voltages an incandescent filament gives off a
and suggest very generous conditions: 2,187 novices and reddish light; at normal voltage its light is much whiter.
only 512 experienced athletes will take part in the race. This effect is expiained by the fact that the brightness of
Whichever team has the most members who cross the the green and blue beams increases with temperature
finish line after 7 kilometers wins. faster than that of the red beams. So at high temperatures
Both teams arrle at the event dressed in colored T- the difference in brightness between the green and red
shirts: the novices wear green/ the experienced athletes bands of the spectrum is much greatert and it's difficult to
red. A-fter the first kilometer the supporters of the novices compensate for it with the glass. This is why at higher
are encouaged by the results: as in the previous competi- temperatures we need a thicker glass to change the color of
tion,729 novice runners remain in the race, while only 256 the heated object under observation.
experienced athletes remain. The numerical superiority
remains with the novices. The supporters of the experi-
enced athletes are disappointed by the fact that half the
runners on thei-r team have quit. But one of the fans, a{ter
doing some simple calculations on a race program/ states
firmly that if the race continues the way it's going, the j
athletes will win. i
After the second kilometer 243 greens and 128 reds are t..
f i .i _
::"
left; alter the third, 81 greens and 64 reds; after the fourth, I;:
I , .;..
27 greens and32 reds. Everybody looks respectftilly at the
person who made the calculations. The remaining three a ---
?, +, &-- F_
kilometers only worsen the defeat of the greens: 9 greens
and 16 reds remain after the fifth kilomete4 3 greens and Figure 5
8 reds after the sixth kilometer. Fina1ly, 1 green and 4 reds Bouguer's Law: an explanation of how inueasing the number
arive at the finish line at the end of the seventh kilometer. of pieces of green gJass aherc the oolor of a flament.

0uAilTllllil/tlt Tltt t_AB t7

 When liglrt reflected from such fabric is split by a prism,
Ancieltl Hussian h0ns and an ohset'uaton hy Lnonal.do da
there is oniy one band in its spectrum instead of the two
Uinci bands for double-colored fabrics. For example, light re-
Some ancient Russian icons are particularly striking flected once from yellow velvet produces a broad band in
because of the f eculLarway the clothing of the saints is the spectrum with the greatest intensity in the middle of
painted. Folds are painted with colors that are in sharp the yellow range. There are also green and blue beams in
contrast to the smooth parts of the garment. For example, the spectrum in addition to the yellow ones. This band in
we may see red folds on a green coat (fig. 6) or orange folds the spectrum becomes narrower for double reflections
on a blue garment. The experienced eye of the Russian because the blue beams disappear almost completely and
painters of antiquity noticed that some fabrics have the the green ones become substantially weaker. This is a
propefty of being double colored-that is, their folds take result of the law of geometric progressions discussed
on a color different from that of their flat surfaces. We can above. Consequently, the yellow-orange color becomes
say that the cause of this phenomenon is the same as the darker.
one we discovered in our experiments with the optical Leonardo da Vinci, a multitalented genius who was not
pyrometer. only a painter but a sculptor, architect, engineer, writer,
If we allow a beam reflected from a double-colored musician, and anatomist, noticed the peculiarity of folds
fabric to pass through a prism, two colored bands will in fabric and came up with the correct explanation for this
remain in the spectrum. For green double-colored fabric, phenomenon. In his Tt eatis e on P aintinghe wrote that
the scenari.owill be the same as for green glass: the green reflected colors are more beauti{ul than the original ones
and red bands remain, and all other beams are absorbed. and that this fact is particularly noticeable with folds in
In fact, double-colored green fabric refiects red beams golden fabrics when one surface is reflected in another and
better than green ones, but the greater intensity of the then reflected back, over and over again.
green beams is decisive when light is reflected from a Maybe I've given you a good reason to take another look
smooth surface. We conclude that green beams are at the work of the old masters. Perhaps they can pass their
predominant in light reflected only once from the green powers of observation on to us, and we can share in their
fabric. profound understanding of the properties of 1ight. Once
When there are folds in the fabric, however, the beams you've leamed to notice the subtleties of multiple reflec-
of light are reflected more than once. At the second tion and the transmutation of colors, no doubt the worid
reflection, the red beams are reflected more strongly than willneverlookquitethesame... O
the green beams, so the doubie refiection results in the
same ef{ect as that for thick green glass-the red beams
become more intense than the green ones, and the fabric
changes color. Further reflection simply increases this Why all the Russian names in
effect.
Most ordinary fabrics have properties that are just the
Quantum ?
opposite of those discussed above. hr their folds we merely
see a color that's darker than that on a smooth surface. The Because Quantum is the product of a joint
cause? Again, multiple reflection. venture with the Soviet rnagazurre Kv ant.
Much of what you read in Quantum is
translated from Russian. Our goal is to
.1,
]
r.l
:l f increase the content contributed from this
lf side of the globe to the fifty-percent level.
-;t
d t! t.
,ifl
.#";. Qu antum's Soviet and American staff s
work very hard to make the translations
readwell. How are we doing? If you don't
look at the names of the authors; canyou
tell which articles were ffanslated and which
were written originally in English?

We hope you enioy the glimpses of So-

viet life offered lrl Quantum. And we hope
you enjoy the wonderful articles from our
sister maqazirrc Kvant, written in the inter-
Figure 6 national language of math and science!
An oncient Russian icon: alterution of color in the fabilc
folds because of multiple reflection.

38 srPrrlllBrR/0cI0BtR 1gg0
|
7l; -TG E
oJ Tecth'i(ol [c ': l'/
l

earned its
academic reputa I [o ,

by degrees,
Florida Institute of Gchnology has everything you'd expect
from a university. Including a lot of degrees both in and
out of the classroom.
-
For example, we offer more than 121 degree programs, from
A.S. to Ph.D., specializing in Science, Engineering,
Business, Psychology and Aviation. Our modem campus is
located on Floridds famous Space Coast, in the heart of one
of America's fastest-growing business areas.
No*, add an annual average temperature of 7 5 degrees,
miles of clean, uncrowded beaches, and every water sport
you can think of, and you know why students prefer EIT
For more facts about EII, the Univenity with all those
degrees, call TOLL FREE 1-800-357-8324, IN FLORIDA
1-800-3494636.

Florida Institute of Technology

MELBOURNE
150 West University Blvd., Melbourne, FL 32901
Circle No.7 on Fleaders Service Card
 CONTEST

A piUeonhole lur Etlery ilUeolt

An attempt to provide private accommodations for that

special mathematical bird

by George Berzsenyi

I NMARTIN GARDNER'S AUGUST that the sum of the elements of S is at can Lrvitational Mathematics Exami-
I uso Mathematical Games column most 90 + 91 + ... + 99, or945, and so nation (AIME): Let S be a subset of {1,
I n Scientific American, the famous the subsets of S can be sorted accord- 2, ...,251such that for every two dis-
I *rth.-atrc, exposiror R.oss ingto the sum of their elements into joint subsets o{ S, the sum of the e1e-
Honsberger showed the following: If pigeonholes numbered l, 2, ..., 945. ments of one subset is different from
S is a subset oI ll, 2, ..., 99} and i{ S has For the pigmng he chose the nonempty that of the other subset. Find the
10 elements, then S must have two subsets of S, of which there are 2'o - I, maximum vaiue of the sum of the
disjoint subsets A md B such that the or 1,023. Thus, there must be a pi- elements of S.
sum of the elements of A is the same gmnhole with more than onepigeon- In view of the time limitations of
as the sum of the elements of B. Thus, that is, there must be at least two the AIME, it was felt that even this
forexample, if S = {3, 9,14,21,27 ,35, subsets of S whose elements have the problem was a bit too ambitious. Since
42, 59, 63, 7 6l,thenA = ll4, 63} and B same sums. Upon discarding com- I was chairing the AIME Subcommit-
= 135,4z]yield the same sum (77), and mon elements, the reduced sums will tee at that timg I replaced 25 with 15
so doA = |.3, 9, 14) and B = {25}. remain equal for the resulting disjoint and posed the revised problem on the
To prove his assertion, Honsberger subsets of S. 1986 AIME, savingAndy Liu's origi-
applied the pigeonhole principle, which As he was reading Honsberger's nal problem for this year's USA Mathe-
was the topic of a beautiful article by arguments/ Andy Liu, another Cana- matical Talent Search (see Happen-
Alexander Soifer and Edward Lozan- dian mathematician, became inter- ings).
sky in the premier issue of Quan- ested in the maximum size of S that Here I'd like to reopen Andy Liu's
tum.t More specifically, he observed will allow for the sum of the elements investigations in a more general set-
l
of each pair of its disjoint subsets to be ting searching for maximal subsets of
answer M 15 on page 5 9 for a
See
distinct. In otherwords, hewas dis- S, of {1, 2, ..., nl. As it turns ort, for n
restatement of this principle.
tressed by the over- :15, S15 :ll5'14' 13' 11' g)'withfive
crowdedmnditions of elements whose sum is 61; while for n
the pigeonholes and :
= 25, SN 125, 24,23, 21, 18, 12), with
wantedto ensurepri- six elements whose sum is 123. ln
vate accommodatiors both cases (as well as for other values
foreachof thepgmrs. of n investigated so far), S" is uniquely
To make the arith- determined by the Greedy Afuorithm,
metic more manage- according to which we always pick
ablg heswitchedfrom the largest avaliablenumbersfor S,
99 to25 and submit- that don't lead to contradictions.
ted the following I invite you to explore this problem
problem to the sub- more systematically, possibly by first
committee of the gathering some more data via clever
Mathematical Asso- computer programs. Here are some
ciation of America in
charge of the Ameri- CONTINUED ON PAGE 42

40 srPTtl'lBtR/0cIIBtR lgg0
 CONTEST

Cliclr, cliclr, cliclr r r r

"The causes of events are ever more interesting than the events
themselves."-Cicero, Ad Atticum , Book lX, Section 5

by Arthur Eisenkraft and Larry Kirkpatrick

O DOUBT MANY OF YOU ity v andone ball leave with velocity 2vwould conserve momentum (2mv)
have played delightedly with 2y? Momentumwouldbe conserved. but would have a kinetic energy of
Newton's collision toy (though Couldn't two bails enter with veloc- 2rl! Similarly, all other possible colli-
not one so large as in the pic- ity v and four balls leave with velocity sion scenarios that conserve momen-
ture!). In this toy five identical balls v/2? Although momentum conserva- tum do not conserve kinetic energy
are suspended by strings so that they tion allows this, these events never except for the one that we really ob-
1ie along a line. If you pull one ball occur. Nature is warning us that we serve. The collision toy has yielded
back and release it, the three middle don't know the whole story. There some important physics.
balls remain stationary and the last must be another restriction on the The collision toy leads us to won-
bali flies off the other end. If two balls motion of the balls that forbids these der what would happen if the balls
are raised on one sidg two fly off
ba1ls other events. didn't have the same mass. Take any
the other end. The pleasure we take in The second restriction is the con- two balls of diff erent mass (a basket-
the toy comes from the repetitive servation of kinetic energy (K=Vzmtr'l. ball and a table tennis ball would
motion of the colliding balls and the The sum of the kinetic energies before work well). Drop each one separately
click, click, clicking sound of the a collision must be equal to the sum of onto the ground and observe the heiglrt
collisions. the kinetic energies after the colli- each reaches. Now place the table
The physics of the toy is both inter- sion. When two balls of mass m elter tennis ball atop the basketball and
esting and informative. The toy pro- with velocity v, the momentum is release the basketball. Watch your
pels the inquisitive idler into an ex- 2mv andthe kinetic energy rs mvz. II eyesl The table tennis ball goes "sky
amination of the conservation laws in two balls leave with velocity v, the high." Here, the collision is between
nature. Conservation o{ momentum momentum and the kinetic energy the Earth, the basketbalt and the table
states thag in a system free of outside have both been conserved. We can see tennis ba[. Imagine three balls of un-
forces, the momentum (mass times that one ball leaving with a velocity equal mass on the Newton collision
velocity)before a collision must be
equal to the momentum after the
collision. In our collision toy, one ball
with velocity v collides with the hang-
ing ba1ls and a single ball leaves with
velocity y. Momentum is conserved.
Two ba1ls in, two balls out-momen-
tum is conserved. Three bails in,
three balls out-momentum is con-
served. The toy certainly obeys the o
law of conservation of momentum. a

If momentum conservation were -V6

the only law goveming the collisions,
o
then other results would be plausible.
Couldn't two balls enterwith veloc- !

OUANTUttl/COll,ITIST 4l
toy. The large mass bal1is pulledback In our discussion so far we've as- This interesting physics has some
and released. The collision occurs. sumed there's no energy loss in the interesting applications. When one
The big ball hits the middle ball, the system. We know that the kinetic wishes to hammer a small nail, an
middle ball hits the smal1ball, and the energy must continually decrease, since intermediate mass called a punch is
small ball flies off the other end. the Newton toy eventually stops. We placed between the massive hammer
Here's part A of our contest prob- also know that some of the kinetic and the less massive nail. In some
lem: Find the mass of the middle ball, energy becomes sound energy-that/s gravitational wave detectors/ the tiny
in terms of the masses of the big and the click, c1ick, click. Since collisions signal hits a succession of masses in
small balls, such that the velocity of with macroscopic objects aren't per- just the ratio discovered in part B of
the small ball will be greatest. fectly elastic, we need a means by this month's problem. Can you think
Part B: Extend your proof to find which we can quantify the loss of of other applications for maximizing
the relationship between the masses kinetic energy. Newton defined the collisions of unequal masses?
of the balls if three balls of intermedi- coefficient of restitution e of a colli- Those of you who are iust begin-
ate mass are involved in the collision. sion as the ratio of the final relative ve- ning your study of physics may at-
locities to the ratio of the tempt part A alone. When submitting
rnitial relative velocities your solution, indicate yourphysics
of two l-ralls. Newton then background so that we can reward our
discovered, expenmentally, younger readers for their excellent
that this number stays rela- attempts. Here's our address: Quan-
tively constant for bal1s of rum,1742 Connecticut Avenue NW,
a given material. Washington, DC 20009.
Finally, part C: Deter- We're still getting some interesting
mine the middle mass in answers to our first contest problem,
a thremass collision, given so we'll hold off selecting the best
a coefficient of restitution ones until the next issue. O
e for each collision.

CONTINUED FROMPAGE 40
Highly recom mended"
-Booklist.
questions to answer: Is it true that n,
n-l,n-2mustbe elements of S"for "ln the iconoclastic tradition ol
n > 3? Is the number of elements of S, Aldo Leopold, Rachel Carson, and
always maximizedwhen the sum of James Lovelock, Daniel Botkin
its elements is largest? Is there some has used a lifetime of research in
n forwhich there are several choices the ecological sciences as a basis
for S,? Send your findings to Quan- for reexamining the human-nature
tum, 1742 Connecticut Avenue NW, relationship. Discordant Harmo-
Washington, DC 20009. The best nies will be provocative to histori-
results will be aclcrowledged, and their ans and philosophers as well as
creators will receive a free one-year scientists. lt is a book to pack in
subscription to Quantum. our intellectual baggage as we
We've received some very impres- prepare lor the journey into the
sive answers to our first contest prob- 21st century."
lem. In fact, we're still receiving Frazier Nash, author of
-Roderick
answers as we go to press, so I'11wait Wilderness and the American Mind
until the November/December issue
to discuss thebest ones. O
Just published by Oxtord
DarurEl B-BorKrN

;
r$[[[[ lill ll[illl ill lts
NEW Ecorocv FOR THE TwENTY- FrRST CEnrunv

Circle No. 8 on Readers Service Card

42 stPTttllBtR/0cI0BtR lgg0
f,ttention High School Students !

TffiH,:.5
fol
A Cholleme
Arnedc on High Sfiod $udenrs

-I-
c Are you ihterested in math yierrce, engineeing or technologr?
o Would you lihe to chech your potential for succeeding in
- Do you haw $uder8 who are god problem-sohxrs and uxll
engineering xhool fufore you apply to college? rounded in the academics?

If so, take the - h Wu want to giue your students aary chance to grory and
master the academic concepts prexnted in the danircrrm?
National Engineering Aptitude Search! Do your students e4ioy oprytunitia to compete wiilt other
The NEAS is a guidance exam for fth tluou$ 12th gaders con-
- studenB who haw similar interes&?
sidering careers in engineering, mathematics, science or technolory. fso, rhenTEAlVt up with JETS!
The NMS covers mathematical undentanding, science reading
IEAMS is a urdque academic competition wittr a name that
aptitude, and problem-solving, It also includes an interest inventory
sugests both !!e team effort needed to win and the types of tats
ex[acunicular accomplishment scale and tiographical questionnaire.
taken-Tests of Engineering Aptitude, Mathematics, and-science
Students receive a comprehensive score report and intex?retation
Students work as a group in an opendlxussion, open-book
gide to asist with high schml guidance efforts. Transcripts are atrnosphue The group format slresses cmperative leaining and
available upon requesl inUoduces students to the'uEineering team" work environinent
Register today for the NEAlit's today's tool for tomonow's engineers, Start buildtng your team now for the state and national TEAI,IS
technologists and technicjansl competitions held eadr year in most states.

TD{ll{S otrers high school students a national challurge

and helps preparc then todayfortomorronrs worldl

$onsored by the funior Enginming Tectrnical Sociely (JEIS)

trffi
q&6
Commended by the Engineering Deans Council
Approved by the National Association of Secondary Schml principals
For registration inlormation write
For more information write
American College Testing (ACI)
JEISTE.AIIIS
NEAS Regrstation (82), P.0. Box 168, Iowa Crty, 1A52243
1420 King Sfeet, Suite 405, Alexandia, VA mt+2715
Circle No, 9 on Readers Service Card Circle No. 10 on Readers Service Card

dii;;*l9f
iGiEiai&l
:{
I
iL earr). LeAd. Serve.
Over 70 undergraduate The nation's leading A university which challenges
3.r
sii

and 60 graduate programs Catholic university in its students, faculty, and

of study. Our student-to- federally sponsored alumni to make a difference.
teacher ratio is 17 to 1 research in the scien- Our teachers are respected
and most classes have ces and engineering. and committed to educa-
fewer than 25 students. tion for more
Excellence in science and than a
engineering is our tradition. career.

+ flan
@
A M arietr. ist, C crth.ol ic
tLniuersitu & Ohirt's
largest prit:cLte
tntirersity . Four-year academic scholarships
Office of Admission o Undergnaduate research opporbunities
300 College Park
Da;Jton. Ohio,15,169-161
o Respected honors program
1

(513) 229-1.1 1 l o Residential campus

Circle No. 11 on Readers Service Card
 LOOKING BACK

Jules Uerns'$ ct'ypluuram

"Everybody realized that if the attempts to decipher the document
failed, the convict would stand no chance."

by G.A. Gurevich

ECRET WRITING AND THE real culprit is his only hope, but it's After a thorough analysis of the
mystery it conceals . . . puzzling written in some unknown code. Here's text, the judge comes to the conclu-
out the code that finally gives the the text to be deciphered: sion that the key to the code is a
clue to hidden ffeasures or saves number. He explains to the convict's
a human life . . there's quite an WFDPFWQQUMWYMPVXRNIE son Manuel how the document was
attraction in a mystery story that's not LEPQSEVEQFPJOPHFLVDV enciphered.
only well written but contains in- GUTOWLHPNHKXRHOBQYDT "Let's take a phrase, any phrase.
triguing coded messages-drink of "The DUKIWYJOWCVCHPQHHKLI This one, for example: '|udge |ar-
Gold Bug" by Edgar Allan Poe or "The WIHPYXHRWAXJAMVPTUMS riques is cute.' And now I take any
Adventure of the Dancing Men" by D QIBF SVVFVQNXUYMB GRP number at random to make a cryPto-
Sir Arthur Conan Doyle. IFPRHFYPGIDVMCXXPGFN gram. Let's assume that it's a three-
In his works |ules Veme also made LWHTFCOIFNJS NSIVMFGM digit number'. 423, for example. I
agooduse of mysterious documents DOTOGHLUYSLGWCINLRLU wrrte'423' underneath the words so
whose secret is disclosed at the very YSDXLQSZHWPGFMRRHYMJ that each letter corresponds to one of
end of the story- Captain Grant's Chil- FLLJJSHFBUNHQMQODPZR the digits, and I repeat this process
dten and A lourney to the Center of QCRFRWVGLB. until I reach the end of the sentence:
the Earth are good examples. Another ]S
JUDGE JARRIQUES CUTE
nanel, I angada, or Eight Hundr ed Miles )udge |arriques volunteers to un- 42342 342342342 34 2342
Dot+m the Amazrsn, isrl t xwell knovm. ravel the knot. "The first thing we
It's full of detailed geographic, histori- need," he declares, "is a system. Sys- Then we replace each letter in the
cal, and ethnographic descriptions as tem means logic, and logic means sentence with the one that follows it
well as pages devoted to the curious success." The judge doesn't have a in the alphabet by the number of
peculiarities of plant and animal life shadow of a doubt about success. He places indicated by the coresponding
in the Amazon basin. But the most decides to employ the method bril- number. Forexample, if thenumber
gripping chapters of the book are de- liantly described by Poe, which is '3'stands under the letter 'D,'YotJ
voted to the deciphering of a docu- based on a comparison of the fre- count off three letters and replace it
ment containing the confession of a quency with which different syrnbols, with the letter'G.' If the letter is at the
criminal who took part in a diamond in ciphertext, and letters, in ordinary end of the alphabet and there aren't
robbery 23 years before the action in plaintext, occur: "I a:-l,anged all the enough letters after it, we continue
the novel. letters of the alphabet in numerical counting from the beginning of the
The story has it that |oao da Cost4 order, starting with the most prepon- alphabet.
througfr a fatal c.oincidence, finds himsell derant, and replaced the letters in the "So let's complete our cryptogram
on trial, facing a charge of theft and document with new ones according to based on the key number 423-which,
murder. The crime was committed a the procedure described by our im- mind you, was chosen at random.
long time ago andhe's unable to pro- mortal analyst Edgar Allan Poe, and L:stead of our plaintext messagg we'll o
E

duce any evidence of his innocence. A then I tried to read the message . . . endup with the following coded one: Y
message with the confession of the but, alas, I failed!" NWGKGMETUMSX ] ULWEXXG .,, t
o
o
:
44 stPTt[llBtR/0cI0BtR 1gg0
 After the judge arriv6s at the con- Let's get back to the text of the component). Some words l"the,"
clusion that the cryptogram has a novel. |arriques's line of thinking was "which," "that," as well as suffixes
numerical key, his certitude gives way as follows: "I am sure that the name of like -tion, -ing, -able, and so on) occur
to the darkest pessimism. Calcula- loao da Costa is mentioned in the quite frequently in all kinds of text.
tions carried Out by iarriques show document. Had the lines of the mes- Naturally, in this case the scope of the
that arandom search for the key, by sage been separated into words, we search is substantially increased, but
going through all possible combina- could have picked out the pairs of the chances would still be good enouglr.
tions of numbers containing not more words that could stand for " daCosta"- It seems, though, that a more rational
than 10 digits, will take over 300 years! . that is, 'two letters-space-five letters' way is to analyze the frequency with
Eventually, he gets bogged down in combinations. Trying them one by which different letters occur in the
guesswork and turns into a gambler one, we could possibiy find the key to cryptogram. According to |arriclues,
who is trying to hit upon the right the cryptogram." however, this approach will be unac-
number. It's not quite clear why the absence ceptable if the key to the code is a
Meanwhile, the day of the execu- of spaces between the words seems an number: "Consequently, the mean-
tion is approaching. |oao da Costa is insurmountable barrier to the judge. ing of each letter is determined by the
gorng to the gallows. In facg it merely increases the range of underlying number chosen at random,
But all ends weil. Luckily, |oao's the exhaustive search. That's why and the same letters in the cryptogram
friendcomes to knowthat thename of Manuel, who has a better grasp of the never corespond to a particular letter
the man who had written the crypto- problerry disagrees with faniques: "What in the plaintext."
gram was Ortega. The judge places of it? If we assume the name of da
the letters o, & T, E, G, A over the last Costa is mentioned and take each
six letters of the message, determines letter in tum to be the first letter of his
the amounts of the shift, and obtains name, we'll eventually find it the
the key to the code: key."
That's it! The direct way to the
ORTEGA solution has been found. Not only
34325L that, the range of the search is not so
RVWGLB
wide. The text consists of Z3}letters,
)ules Veme is a great writer, and he which means that the number of pos-
easily leads the reader to believe that sible combinations doesn't exceed 223.
except for a happy coincidence, it's Eventually, having written the words
impossible to guess the number 343:251.1 "da Costa" over the IBFSWF frag-
Now it's time to tell you that |arri ment, we'd determine the following
ques could actually have deciphered sequence of figures: 5134325. It would Figure 1

the cryptogram without waiting for a be natural to assume that the last
lucky break. The most amazing thing number opens up the following nu- The special coding device shown in
about it is that the judge was on the merical pattem: figure 1 witt help us see that the judge
right track and had practically solved T-)AT-NqrflA was wrong. To simplify the matter
thepuzzle. He had the key right in his . . . s134325L34325L34 . . we'll assume that the key is a three-
pocket. ...TUMSDQTBFSWFVQN.. digit number. The letters of the alpha-
bet are written in order around the
So instead of the key 343251, we outside edges of four concentric disks.
rAn interesting fact is mentioned in have found its cyclic permutation Three of them canrotate, while the
the commentary appended to the novel: 513432, which in no way prevents us inner disk is stationary. The inner
"The author . . . received a letter from his from deciphering the text. (And, by disk can be considered the "plaintext"
friend Professor Maurice d'Ocagne
informing him that a student at a chancg it's the very combination that disk, the three rotating disks the "ci-
polytechnic school had managed to read opens the coding line of numbers.) phertext" disks.
the cryptoglam lflng at the core of Finally, let's consider the following Suppose the key number is 259.
langada. At the time, the novel was still problem. In the case descdbed above Tum the first ciphertext disk counter-
being published serially in a magazine.
we knew what kind of document it clockwise through two letters; the
So it was not too late to correct the
re$ettable inadvertence. Before the was and so were able to guess one of next disk, through five letters; and the
book appeared as a separate edition, fules the words, which gave us a clue to the third disk, through nine letters (fig. 2).
Veme had time to think up a more solution. But what do we do when the Now we're ready for coding. Find the
complex code----one that precluded content of the document is completely first letter of the plaintext on the in-
premature deciphering of the
ol:scure? nermost disk and replace iS with the
document. . . .

"One will undoubtedly not find as There are several possible paths to letter across from it on the ffust rotat-
intricate a cr)?togram in any other of his pursue. |ust as with our cryptogram/ ing disk; replace the second letter with
works." we can try to guess a word (or its the corresponding letter on the second

40 stPIt]llBtR/0cI0BtR 1900
 Table 1: The number of instances of each letter in sets selected from a cryptogram with a 6-digit key

Set No A B c D E F G H I J K L M N o P o R S T U V W X Y Z

1 0 0 0 2 0 4 0 1 3 4 0 3 4 2 1 1 1
'1
2 3 0 0 1 1 5 0

2 0 3 2 0 2 4 0 2 1 4 0 0 2 2 3 4 1 0 4 0 2 1 0 1 0 1

3 1 1 0 3 0 2 3 3 0 1 2 4 0 1 2 1 4 2 0 0 1 2 3 1 0 1

4 0 0 1 0 2 1 2 2 4 0 0 2 4 1 0 2 0 5 3 0 0 2 2 4 1 0

5 0 1 0 4 0 2 1 tE 1 0 0 4 0 0 0 2 4 3 0 0 0 3 5 1 0 0

6 1 0 2 0 1 3 3 1 1 1 0 0 0 3 2 3 3 0 0 2 5 3 1 0 3 0

The columns under letterc obtained fuom the letter E by shifting no more than 9 places are tinted yellow. In each line, rcd
indicates the number under the letter that actually denotes an E in the conesponding set; geen-the maximum number of
instances in the yeilow (that is, permissible) runge; bluo-the maximum values that fall outside the permissible range.

rotating disk, and replace the third magnitude of the shift? The trick is decoding. Now we can try to przzle
letter with the corresponding letter on that in the set of correctly shifted out the key number right away, com-
the outer disk. In other words, the letters, the frequency of each letter's paring the frequencies of the different
respective letter of the jth rotating occurence is approxrmately the same letters. Because the most frequent
disk replaces the (3k + rlth letter of the as in the language as a whole. That's letter in English is E, we can assume
text (where i -- l, 2,3 and k -- 0, l, 2, ...1. the gist of the matter! By comparing that its counterparts are the most fre-
The very process of coding suggests frequencies at different shifts, we'll quent in their respective code sets. For
that if the difference between the determine the most probable fust digrt. example, in the first set the prime
numbers of any two letters in the text Then similar analysis of the set con- suspect is Y (5 entries-see the first
is a multiple of 3, the same number- taining the 2nd, 5th, Bth, .. . letters of line of table 1). But Y is 20 places away
that is, the same disk-is used to enci- the cryptogram will give the second from E in the alphabet, whereas the
pher them. So |arriques was mistaken digit of the key, while the third set (the maximum shift is 9 places. So Y falls
in maintaining that the same letters 3rd,6rh,9d1 ... Ietters)will suggest the away. For the same reason we can
in the ciphertext never denote the third digit. Finally, there might be generally confine ourselves to the 10
same letters in the plaintext. several sufficiently probable keys at yellow columns in table 1 (from E to
Now, let's start deciphering. Sup- our disposal-we just have to choose N). The predominant "yeIlow let-
pose we know that the key is a three- the one that gives a coherent text. ters" in the first line are F, f, and M,
digit number. To determine its first All that remains is to clarify how to which means the fust digit of the key
digit we should analyze the 1st, 4th, approach the problem when the number should be l, 5, or 8. The second line
7th, ... letters of the cryptogram. If the of digits in the key is unknown. This gives us F and I as plausible letters (or
first digit of the key is 1, all these case, too, requires a good deal of se- i and 5 as digits); the third line sug-
letters should be replaced by those quential searching. First, we assume gests L (or 7)i and so on. Although
immediately preceding them in the that the key is a two-digit number; we've already seen that the actual key
alphabeq i it's 2, the letters should be thery a three-digit numbe4 and so on, number is 513432, to play fair we
shiftedback two places; andso on. But until the text has been deciphered. shouldcheckallthe3 . 2. I . 2. 1 . 3 =
how can we determine the actual Omitting the intermediate variants, 36 possibilities. It looks to be a little
let's make use of the method to deci- boring, and we still wouldn't be all
pher our cryptogram, whose key, as that sure we didn't miss the right digit
we already know, is a six-digit num- (in fact, we did miss it in the third
ber. Lr this casg the text of the crypto- line). But there's nothing to be sur-
gram is divided into six sets of letters prised at: our sets are toci small for us
according to the pattem described above to draw reliable conclusions about the
(first set: the letters 1,7, 13, ...,229i frequencies of individual letters. And
second set: the letters 2,8, 14, ...,80; yet, if. we take a group of the most
...; sixth set: the letters 6, 12, 18, ..., common letters right off the bat, sta-
2281, The first two sets have 39 letters tistical laws will inevitably take over.
in each; the resq 38 letters. To make the superiority in frequenry
Tobeginwith, we countup allthe significant, agroup of fourletterg E, I
instances of each letter in each set. A, O, will suffice. This timewe'llbe
The results are listed in table 1. And clever right from the start and restrict
Figure 2 that's all the in{ormation we need for ourselves to this group andits shifts:

IuAilIlJ1lll/t001ilil0 BAct( 4t
Table 2. The number of instances of the letters E, T, A, and O in each cryptogram set

Magnitude of Shif t

Set No. 0123456789

1 4 5 1 5 6 16 2 2 5 6

2 5 13 4 2 8 B 3 3 6 9

3 3 5 9 11 1 3 7 10 4 3

4 2 3 5 I 13 2 2 6 10 6

5 0 5 8 19 2 2 1 14 7 1

6 6 11 11 2 2 I 8 5 2 6

In each line red denotes the number

corresponding to the actual magnitude
of the shift for the set. It's always the
greatest number in the line, and it's
ShouldUgu
almost always unique. (The exception is
the sixth line, where there is another
tn
number as large, which is marked
gteen.)
They are!
F, V C, Q; ...; N C, }, X (that
U, B, R G,
is, the "yellow portion" of a giant
"table of letter-quadruplets"). The
fu@fr.@XEROX
number of letters of each group in each
set is easily calculated by adding up dTdEEEI
the corresponding four numbers in
table 1. This gives us tableL, by means --F-r
of which we can more or less defi- 'Trademark oi The Dow Chemi.a Comp6ny

nitely conclude that the first five key

digits are 5, l, 3,a, 3. Although the last
digit remains uncertain-it's a choice
between 1 and 2-full decoding of the
cryptogram now poses no problem.
I leave it to you to figure out what's
written in the ciphertext! O These and 400 other top corporations "grow their own" englneers, managers, and
corporate executives at GMl...
"Closely coupled" cooperatlve education - GMI's unique partnership lvith maior
corporations - provides extraordinary opportunities lor high ability students'
Learn and Earn during paid co-op work experiences. GMI students average $56,000 in
co-op earnings over the tive-year program (range $35,000 ' $75,000).

DEGREE PROGRAMS

Engineerlng Management
Electrical Accounting
lndustrial General Management
Manulacturing Systems lnformation Systems
Mechanical Marketing

For more intormation call:

GMI Engineering and Management lnstitute

'1700 West Third Avenue
Flint, Michigan 48504

1-800-955-4464 l-313-762-7865

Circle No. 12 on Readers Service Card

48 stPItllllBtR/0cT0BtR lgg0
 GETTING TO KNOW

The elemelthry parlicles

Fishing for the Higgs boson, stalking the top quark.

by Sheldon Lee Glashow

]ulass, spin, and fie antharlicle

Quantum mechanics and relativ-
ity tell us that every particle is charac-
terizedby a nonnegative mass and a
nonnegative integer or half-integer sprn,
and that it has an antipart:icle with the
same mass and spin but opposite elec- f Le{_rtetL {hi&iatu*
tric charge. Massive particles travel
slower than light and can be brought
*'13 i C
to rest, while massless particles (like
photons and gravitons) travel at light
speed relative to all watchers. Spin is
the measure of a particle's intrinsic
angular momentum. A massive par-
ticle of spin s may be found in any one
(B..rr*u.'%#'SW
(A_) Ure
W
of 2s + I different quantum spin states.
The antiparticle of the electroq called
the positron, was seen first in cosmic
rays in 1932. Antiprotons were first eL {.1 RU!.\ D+u,r{ ft\"'aRiL e L{{ it$*
produced and detected at the Berkeley n0 Li{ R rft+
Bevatron in 1955. Photons are their
own antiparticles. Particles annihi-
Iate their antiparticles on contact. All
earthly matter (and virtually a1l celes-
tial matter) consists of particles, not
antiparticles. Otherwise, we wouldn't
be here to tell the tale.
MWMW
YfrU0f\ 5lRi,'r\&&
L,tu {,il {. r{.
rn u sy\
$e'-A{
'{lnt}
( Hi^nrne;
Qud^rtlt
Fenmions altd hosolt$
Particles with half-integer spin (like
electrons with spin ll2l satisfy
Fermi-Dirac statistics, which means
no two of them may be in the same a

quantum state at the same time (the r{u tt"?fln ts*rtos^ fd\$ ,?0F Gu&ttK -
,a
Gu*.tlK n&Lr[ rtr l\0 :o
Pauli Exclusion Principle). Such par- :.
.6
ticles are calied fermions. Particles .. -
,a
with integer spin (like photons) satisfy z
a
Bose-Einstein statistics. Many of these

0l]Atr,lIU]ll/0rrTrilrG I0 t(il0l'll... 4g
bosons can (and, in a sense, like to) seen as particles in their own right. complete. Whether it really is or not,
congregate in the same quantum statg The word lepton comes from the Greek we shall see!
which is the principle underlying the leptos meaning "small" or "slight,"
operation of lasers. and was coined by L. Rosenfeld in Basic hosons
1948 to mean any fermion of small These particles mediate forces
mass, like the electron or neutrino. among the fundamental fermions.
Fundamenffil lermions-{ttat'16 altd
Today, leptons include any of six lcrown Electromagnetism results from the
hrult$ fermions lacking strong nuclear inter- exchange between charged particles
Our particle directory lists twelve actions. Three are electrically charged: of masslessphotons, the particles of
spin 1/2 particles: srx cluarksand six the electron, the muon (about 200 light. The strong nuclear force arises
leptons. Quarks (rhymrng with forks| times heavier), and the tau-lepton (about from the exchange between quarks of
were invented by M. Gell-Mann and 17 times heavier yet). Each one is massless gluons. The weak nuclear
G. Zweig n 1963. "Up," " chamed," associated with its own sort of neu- force is produced by the exchange of
and"top" quarks carry electric charges trino, making six leptons in all. Neu- massive W or Z bosons between any of
of.2l3, while "down," "strarrge," and trinos are very light, perhaps even the fundamental fermions. Gravity
"bottom" quarks carry charges of massless. Recent experiments sug- may be thought of as due to the ex-
-1/3. An individual quark can't be iso- gest that there are no more than three change of massless gavitons. Gluons,
lated from t]l'e hadron of which it neutrino species. This implies that like quarks, are "confined": they can't
forms apafi. Thus, quarks can't be our list of fundamental fermions is be seen as isolatedparticies. Charged
W's and nettra|Z'swere discovered at
the European Center for Particle Phys-
ics (CERN) in 1983. The last of the
basic bosons in our bestiary is the
Higgsboson, an elusive and still hypo-
thetical particle responsible for gener-
b65tc ating all particle masses. It should
BO50ns show up at the Superconducting Su-
per Collider, now a-building in Texas.

Hadrons
Inl962the Sovietphysicist L. Okun
?ARTILLQ, 6Pr^ ahD ma95 used the Greek word adros, meaning
FOR,CA
thick and bulky," in choosing a name
for any seemingly elementary particle
eLaCrRO- 1
that partakes in the strong nuclear
tA66t,.ec9ri t6LEaLA9, force, like the proton but not the elec-
tron. Today, a hadron is any particle
;\-\\. i., made up of quarks. Three quarks stick

weaK
?Or(LA
*frv)!k),, }\/-\' 1

v$659L0Et
together to form a baryon; a quark
binds to an antiquark to form a rnesoni
and three antiquarks form an antibar-
wtBogon z Bo90n yon. These are the only known ways
in which quarks combine to form
{ hadrons. Because they are made up of
I
an odd number of fermions, baryons
t^d35Le55
and antibaryons are themselves fermi-
GVV0T\6 ons. Mesons are bosons.

Illucleolts
This is a word that has been used
since 1941 to refer to neutrons or
protons. An atomic nucleus with
mass numberZcottains A nucleons,
Z of which are protons. Nuclei with
t^La9Le5' the same Z but differing A are known
GRAVITY
as isotopes. Nucleons are fermions.
They are the lightest baryons, consist-
ing exclusively of up and down quarks:

50 SIPTIlllBER/OCIOBIR IgOO
two ups and a down make a proton/ and muons were first seen in cosmic cists working at the Fermilab
while two downs and an up make a rays. Many other kinds of meson have proton-antiproton collider within two
neutron. About 99.98% by weight of been dismvered since. Yukawa's mesons years.
all ordinaryrnatter consists of nucle- became known as pi-mesons and even-
ons. The rest is electrons. tually pions. They aren't elemen-
as ltleuFinos
tary: like allmesons, they're each Neutrinos produced by a nuclear
Pions and lnttolt$ made up of one cluark and one an- reactor were first observed in 1953.
Hideki Yukawa suggested in the tiquark. Since then, physicists have observed
1930s that the nuclear force results neutrinos produced at particle accel-
{rom the exchange ofhypothetical ele- T[ehp qumlr erators/ by cosmic rays, by the nuclear
mentary particles between nucleons. Our theory demands that such a furnace of the sun, and by the last
He called his particles mesotrons (soon particle exists and weighs no more "nearby" supernova in 1987 (which
truncated to mesons) because they than 200 protons. Experimenters have was a mere 160,000light-years awayl.
had to be intermediate in mass be- not yet found it. They are con-fident it Some scientists believe that neutri-
tween electrons and nucleons. Par- must be heavier than 100 protons, nos have mass and that the mysteri-
ticles with such masses were observed otherwise, it would have shown up ous dark matter of the universe con-
in 1938, but they turned out to be already. This window is rapidly being sists of swarms of neutrinos left over
muons. Yukawa's particles were fi- closed: I predict that"top" (the Last of from the Big Bang. o
nally discovered in 1947. Both pions the Quarksl) will be found by physi-

r-\
/,
(
t
\*
r.AliiluuArJ

Factol' lne irtu IhE OUANTUM equaliolt!

Four colorful issues per school year

lntroductory rate if you subscribe beforeOctober3l , 1990:

il $9.95 (US) t $17.95 (Foreign)

Rates after 10/31/90: (checkone)

Bulk rates for students:
il $'18lndividual il $28 lnstitution/library 20-49 copies: $12
il $i4 Student il $14 Gift 50+ copies: $10
Forcign orderc, please add $8 for surface postage d handling

Name
Send to:
Address
Quantum
17 42 Connecticut Avenue NW
CitylStatelZip Washington, DC 20009

IuAt'tTlJlll/GtTTill.l0 T0 l(i|0u{... 5l
 Physics limel'iclts
Mysteries of the universe in rhyme,
the riddle of time in verse
by Robert Resnick

UTTING PHYSICS in limerick form has become

popular recently. It's really an old game, though. And {inished him off in mid-air.
Back in Albert Einstein's day, when there was a
writer named Gertrude Stein ("a rose is a rose is a I shouid tell you the sort of legitimate answers I had in
rose") and a modemist sculptor Louis Epstein, a fashion- mind for this one. Remembering that the electron and the
able limerick went positron circle about their common center of mass, form-
ing a short-lived positronium atom before mutual annihi-
There's a curious family named Stein- lation, the student might have used a couplet like
There's Gertrude, there's Ep, and there's Ein.
Gert's verses are bunk, Sheput him in a trance
Ep's statues are iunk, withher infamous dance
And nobody understands Ein.
A so*e*hat incorrect answer/ for which I'd give part
I work humor into my physics classes and often cite a credit, might be
relevant limerick. For example, in waming students what
could happen i{ they try to defy the laws of physics, I would She meant him no harm,
say But tumed on her charm,

There was a young woman named Bright A recent correspondentl who heard of my test offered this
Whose speed was much faster than light. answer:
She eloped one fine day
In a relative way She needed no tact,
And conceived on the previous night. For unlikes atttactl

Of course, today that could be considered a sociological So you see/ you really can find lots of meaningful solutions.
limerick. Another test item was (
Or, in another veiq
There once was ahardgamma ray
A mathematician named Haines, And a nucleus it forced to decay.
After infinite racking of brains, A resultant bambino
Now says he has found Was called the neutrino
Anewkindof sound
That travels much faster than planes.
The same correspondent met the challenge with this last
Back in 1958 I gave the students in a modern physics line:
class an examination in limerick form. They had to
complete the limerick I startedby adding the couplet or the You ask if it had mass? No weigh!
last line. One test item, for example, was
Now I'11give you some of the other test items and let
An electron quite debonair
Spied a positron up on the stair.
tB arbar a Lei, Phy s ic s T o d ay.

52 SEPIItljlBIR/OCIOBIR 1O$O
 you try your hand at fi[ing in the blanks. Remember, you An electron was spinning around
must make physical sensg be clever, and keep the gallop- And moving quite close to the gouno
ing rhythm (or at least try)12 Exotic its rapture
At the thought of K-capture
Saida slow little neutron ere fission,
"Don't speak of me with such derision.
I may have no charge, That exam spread far and wide since 1958. I do think it
Andbe not so large, had something to do with popularizing physics limericks.
Since then physics journals and others have held physics
limerick contests.3 Once you open it up to classical phys-
An atom that came from the tap ics, as well as modem/ and allow complete ones instead of
Had electrons all over her map, just test items, the sky's the limit. I must have filed
But in her interstices hundreds of them away-not all the greatest/ of course. "
Lurked a much worse disease A recent articlea rnThe Physics Teacher reprinted a
story of mine from the Rensselaer alumni rnagazinethat
told of my experiences with student limericks. Already
A meson descending in flight I'm getting mail with complete physics [mericks. One of
Was veering first left and then right. the best sets came from a high school student.s Here are a
So brisk was its action, few of his creations:
The Lorentz attraction
Is a quarka thing or a wave?
This question spurs many to rave.
An X ray shot out like a tear Please, don't you jeer,
Took off for a crystal quite bare. I'm being sincere-
It wasn't the plasticity, A quark can as either behave.
But that dam periodicity
Where is it? 'Tis really uncertain,
Like trying to peer through a curtain.
The mesons are nuclear glue- Heisenberg had no doubt
fust listen to what they can do: You can never find out
Both position and momentum for certain.

The pi ones, that is, not mu. And one for skeptics:

2There was a young man from San Fran They thought they discovered cold fusion,
Whose verses never would scan. And in general caused quite some confusion.
When asked why this thing Fleischman and Pons
Never went with a swing, May unfortunately be cons-
He said, "Itry to get as many words into the last line as I Their research might just be illusion.
possibly can."

So maybe we should have a contest

among students for the best complete
physics limericks. What do you think?
(For our foreign readers, this may push
your English as wellas yourphysics to
the limitl) Meanwhile, send in your
lines to complete the protolimericks I've
given.
Next time-palindromes. That's a lot
tougher! O
Robert Resnick is Prcfessor of Physics at
Rensselaer Polytechnic Institute, Troy, NY.
3SeeThe Physics Teacher, September
1986 and September 1987.
C a"Thete Was a Professor from Troy,"
-a
o The Physics Teacher,larruary 1990.
sPatrick T. Baker of Rockville,
C
Maryland.
-o
.a
Z
!

OUAIITUr'll Sr'llII-tS 53
 OUANTUM SMILES

The music ol fhysicisls

Short tales (not "from the Vienna woods")
that show the human side of some of the
giants of modern science

Albert Einstein (1879-1955) Iiked to play the violin, and he especially liked being
accompanied by the great pianist Artur Schnabel. One particular day, Einstein played
a wrong note-after all, he was a physicist, not a professional musician! He and Schna-
bel tried the passage again, and again Einstein missed the note. Schnabel lost his cool.
"Wrong, Albert, wrongl fust listen to me play it: one/ two/ three . . . Dear me, how
could it be you can't count? !"

Robert Bunsen (1811-1899) invented many things, including the carbon zir,c elec-
tric cell and the ice calorimeter, but he played only a minor role in developing the
ubiquitous burner that bears his name. Once he went to the loca1 music conservatory
for a concert and a bulb went on in his head. During the intermission he turned to the
person sitting next to him: "Tellme, are all the violins over there playing the same
thing?" His neighbor told him yes, they allplay the same notes. Bunsen shook his
head. "We11," he said, "that'spretty uneconomical. They ought to exchange them for
one big violin and have just one person play it!"

Max Planck (1858-1947), the instinctively conservative scientist who revolutionized

physics with his work on the quantum theory and relativit, was strongly drawn to mu-
sic in his youth. Infact, he seriously considered making it his career. Planck had such a
good ear that, as he used to tell his friends, no concert could be completely pleasurable for
him because he always noticed even the slightest mistakes the musicians made. Only
after many years did he-to his great joy-lose this "supersensitivity."
1-\<-',
t

The great nuclear physicist Emest Rutherford (1871-19371, on the other hand, didn't
haveaverygoodearformusic. Buthedidhaveaprettyloudvoice. Hisrepertoirecon-
sisted of just two things, which his lab assistants reliably used to ascertain his mood. If
Rutherford was walking down the hall bellowing "Onward, Christian soldiers" (recog-
-c
o nizable only by the words, not the tune), work was going well. But iJ he was carefully
I
O fitting words to the doleful strains of a ponderous dirge, his coworkers began mentally
preparing themselves: Watch out, Rutherford's in a rotten moodl
o
L
o O
c

54 StPItI'llBER/OCIOBtR 19gO
 Wanted! Womeru in
science arud math
Clare Boothe Luce Scholarships for Women
at Marymount University
can help you pay the way.
llethods of lUotion
An Introduction to Undergraduate study ln:
Mechanics, Book I Biology
Isaac Newton really believed
Computer Science
that moving objects continue Mathematics
at a constant speed in a Physical Science
straight line? Do your
students? This manual was
created to help teachers For eligibility and application information, call
introduce the sometimes (800) 548-7638 . (703) 284-1500
daunting subject of Newtonian
mechanics to students in the
middle grades. The 27
Marymount .JZ-
OUniversity
activities presented here use 2807 N. Glebe Road . Arli Ya. 222074299
readily available materials to
give students visual, aural, Circle No. 13 on Readers Service Card

and tactile evidence to combat

their misconceptions. And the

TABI|RAIIIRY 4lIllII
teacher-created and tested
modules are fun: Marble races,
a tractor-pull using toy cars,
EWA*,4
-
fettucini carpentry, and film
container cannons will make &BIOTOGIGAT MIGRII$GOPE
teachers and students look HAS ALL THE EXPENSIVE FEATURES FOUND
forward to class. Readings for ON MICROSCOPES COSTING TWICE AS MUCH
teachers, a guide for workshop o lnternational Standard DIN Objectives and Eyepieces
leaders, and a master o Built-in llluminator
materials list follow the A truly professional laboratory instrument with all of the
f ine features found on Japanese scopes costing twice as
activities, making this manual much, Ruggedly built for heavy duty classroom use.
useful for inservice Featuring coated fully-achromatic optics, the resolution,
centering and parafocal features all perform excellently.
workshops. (grades 6-10) Mechanical movements are smooth and positive, dual
coarse and f ine focus controls plus full security locks on eye-
#PB-39, 1989, 157 pp. $16.50 piece, stage clips and rack stop. One year limited warranty.

All orders of $25 or less must be

prepaid. Orders over $25 must include
a purchase order. All orders must
include a postage and handling fee of
$2. No credits or refunds for returns.
Send order to: Special Publications, "=

M
NSTA, 1742 ConnecticutAve. NW,
o#P# $
Washington, D.C. 20009.
lf fiiflg,?, fJff iv 259
Send us yout school bids, you will be pleased with ou prices and services.
Learning with Write for FREE 188 page catalog featuring over 7,000 products.
SERVING EDUCATORS SINCE 1942
NSTA
G ff E**n gr,{S,g!
r il}i[:r^F,n;
Tel. 1-609-573-6250
Fax. 1-609-573-6295

Circle No. 14 on Readers Service Card

 HAPPENINGS

The USA tUlalhemalical Talenl Seal'ch

Where else do you get an entire month to solve five problems?
by George Berzsenyi

aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
I N THE FALL OF 1989, VIA A Prove that an integer can be expressed as the arithmetic average of two perfect
I "nlrmn of the same name in Con- squares if and only if it is the sum of two perfect squares.
I sortium,I initiated the USA Mathe-
I matical Talent Search (USAMTS), For what values of n is it possible to partition the set {1, 2, ..., n} into five disioint
broadening the well-known Wiscon- subsets so that within each subset the sum of the elements is the same?
sin Mathematical Talent Search to aaaaaaaaaaaaaaaaoaaaaaaaaaaaaaoaaaaaaaaaaa
the national level. This endeavor was this year's International Mathemat- double our efforts to atfract even more
supported by Rose-Hulman Institute ics Olympiad team took part in it. of you to the USAMTS.
of Technology, the Consortium for Many of the participants commented To this end, I offer two problems
Mathematics and Its Applications on their preference for this type of from year 2, rognd 1, in the hope that
(COMAP), and the Eroron Education competition, which doesn't impose they will whet the appetites of thou-
Foundation. It was aimed at talented stringent time constraints but instead sands of Quantum's student readers.
high school students to attract them encourages careful exposition of mathe- For the complete set of five problems,
to the fields of science and engineer- matical thought. That takes time, see the Fall 1990 issue of Consortium
ing. Four sets of five problems were and the beautiful work submitted by or write to me at the USA Mathemati-
published in the quarterly Consor- most of the contestants is most re- calTalentSearc[ bxl}l,Rose-Hulman
tlum during the 1989-90 school year. warding. Unfortunately, year-round Lrstitute of Technology, Terre Hautg
The participants were given one month problem solving is not yet a national N 47803. O
to submit their solutions to each set of pastime in America/ so we must re-
problems, and their work was care-
fully evaluated by ateam of faculty Bulletin Board
members at Rose-Hulman. These
evaluations, aiong with complete sets $uWcomlutinU lon [igh $il001$fttdsilts
of solutions with insightful comnen- The Cornell Theory Center, one of four national supercomputing
tary,were sent to the students, who centers/ offers a summer program called "SuperQuest." Open to all of
could gather 5 points for each perfect the 23,000 higfr schools in the United States, SuperQuest is the onlypro-
solution and thus a total of 100 points gram to offer advanced supercomputing specifically forhigh schools.
over the year. On the basis of the Four teams, consisting of 3-4 students and one teacher-coach, are
outcome, several winners were de- selected to come to Comell for one month in the summer to leam about
clared in each of grades 9 through i 2 supercomputing research and its applications. The students take classes
and were awarded valuable book prizes in supercomputing techniques, meet with supercomputer researchers
by COMAP. such as Carl Sagan, and work with Comeil's technical staff to develop
During the first year of the USAMTS, their own programs.
nearly 300 students took advantage of Sponsored by IBM and the National Science Foundation,'Super-
this unique opportunity. The states o{ Quest's goal is to foster creativity in devising computational solutions to
New York, Illinois, and Texas pro- scienti{ic problems, and no arca of scientific endeavor is out of bounds.
vided the largest numbers of competi- For an application booldet and more inJormation on SuperQuest, write
tors, but most of the other states were to SuperQuest, P.O. Biox6745, Princeton, NJ 08541, or call 607 255-4859.
also well represented. It was particu-
larly gratifying that more than hal{ of lllaliolta! $cienm 0lymilad ru$ilts
the participants of the US Mathemat- Approximately 2,000 students representing 94 schools in 35 states
ics Ollnnpiad werc acrive in the US,AMTS gathered at Clarion Universrty of Pennsylvania in May to takepart in the
and that five of the six members of sixth arurual National Science Olympiad. The students competed in 32

50 SIPIE]IlBIR/OCTOBIR IggO
science events testing their knowl- Earth's topics range from the well-
arship Competition awards 41 stu-
edge of biology, earth science, chemis- dents in grades 9 through 12 over
publicized and widely di.scussed to
try, physics, computers/ and technoi- the obscure and unexplored issues of
$30,000 in scholarships and cashprizes
ogy. Winners received medals, tro- for battery-powered devices they have the environment. The series will also
phies, or scholarships for their efforts. designed and built. |udges evaluate include interviews with environmen-
This year Irmo, South Carolin4 bested devices on the basis of originality, tally active celebrities and local he-
competitors in both the high school creativity, and practicality, ar'd all roes, and pointers toward leading an
and middle/junior high school divi- entrants receive an award certificate environmentally positive lif estyle.
sions. and a special gift. Network Earth invites viewers to
The Science Olympiad, a nonprofit This year's top winners will also participate with its staff, environmental
organization headquartered in Roch- win an expense-paidtrip to Houston experts and organizations, and each
ester, Michigan, seeks to improve the for the awards ceremony. All entrants other via computer. Through Com-
quality of science education, increase must have a teacher/sponsor/ so ask puServe InJormation Service, viewers
student interest in sciencg and recog- your science teacher for rules and with access to a personal computer
nize outstanding achievement in sci- applications. Formore information, andmodemwillbe ableto logonto the
ence education. The teams advancing write to Duracell/NSTA Scholarship system/ read about current environ-
to the finals at Clarion University Competitiorl National Science Teach- mental activities, ask questions about
were the survivors of regional and ers Association, 1742 Connecticut what they have seen on the show,
state Science Olyrnpiad toumaments. Avenue NW, Washington, DC 20009. access material that could help them
Next year's National Science OIym- become more environmentally awarel
piad will be held at Penn Valley lmodttdltg tlleltllonk Eaflr and participate in live computer con-
Community College in Kansas City, The Turner Broadcasting System ferences.
Missouri. For more information, write recently premiered "Network Earth," Network Earth runs on Sundays at
to National Science Olympiad, 5955 an inlormative and entertaining weeldy 11:00 p.u. on TBS. For more informa-
Little Pine Lane, Rochester, MI 48604. pro$am that offers a provocative look tion, write to Network Earth, One
at current environmentai problems CNN Center, Box 105366, Atlanta,
Durauell scholanshhs and their solutions. Presented in a GA30348.
Eachyear the Dwacell/NSTA Schol- fast-paced magazine format, Network

ltllltath happeninU? ln the next /ssue of Quantum . . .

Summer study .. competltlons ... new
books . . ongo ng activlties .. clubs and What are the pospects for using light instead of electricity for information trans{er
assocrations .. free samples . . contests in electronic devices? Y.R. Nosov gives us a grrided tour of this area of solid-state
... whatever t is, if you think it's of rnterest
physics in "Lightning inaCrystd7."
Io Quantum readers. et us know about
f
itl He p us Happenings and the Bul e- What does it mean to "take the extreme casg" as we were told to do in the answer
t n Boaro wlth short news items. firsthand to Math Challenge 15? A.L. Rosenthal plays variations on that rich theme in
repofts, and announcements of upcom- "Math to the Max."
ing events
Have you ever wondered why the holes in Swiss cheese are round? In his article
Sergey Krotov reminds us not to be afraid of asking "childish" questions.
Whal's on you'lniltd?
Wr te to usl We want to kno|whatyou What do Euclid's greatest common divisor, Pythagorean triplets, and geneological
thlnk of Ouantum What do you l ke the trees have in common? A.A. Panov shows us what in his mathematically arboreal
most? Whatwould you l keto see more article.
of? And, yes what don't you I ke about
Auantum? We want to make it even bet- How did Native Americans throw a tomahawk so that it stuck instead of
ter. but we need your he p. bouncing? V.A. Daqrdov's love of )ames Fennimore Cooper's novels led him to
investigate, and he offers his results in "Tomahawk Throwing Made Easy."

Whal's ollr addrs$$? Plus...

Quantum
17 42 Conneclicut Avenue NW . A glitch while inventing the steam engine
Washington, DC 20009
. Physics for dummies

Be a lactol' in the .A discomforting incident in a railway tunnel

.New problems from the Toumament of Towns

. . . and our regularfeatures!

OUAlllIll]Il/llA PIlll IltlE

 SOLUTIONS

IUlAIh

Ml1
From a great variety of solutions to
into one and only one of the sets V(s),
it can be represented in the required
this problem we've chosen two of the
form. Also, s, x, and y = s 'x are
most instructive and elegant, in our
determined by n uniquely.
view.
By the way, our formula shows that
In a triangle ABC let CD = 1 be the
m pairs (x, y) can be enumerated by the
bisector of angle ACB, AC = b, BC = a.
numbers r?/ as seen in figure 4. (N.
We'lI prove that Figure 1
Vasilyev)
.,2ab
'a+b M13
To expose the main idea of the proof
For a = 10, b = 15, this givesT < (2. 10.
Iet's consider a triangle XYZ with a
1s)/(10+ l5)= 12.
fixed base YZ arrd vertex X moving
along a line 1(fig. 5). It's quite obvious
Solution 1. Draw a line through D
that the area of XYZ tsconstant if J is
parallel to BC and intersecting AC at E
parallel to YZ, otherwise, it varies
(fig. 1). Obviously, CD < CE + ED, and
monotonously as long as X doesn't
the angles marked in the figure are cross YZ. (Actually, the area is propor-
equal. Therefore, triangle CDE rs
tional to the distance fuomXtoYZ.l
isosceles, CE=ED =x, andbecauseof
Now let's denote the given quad-
the similarity of triangles ADE and rangles as in figure 5. If a diagonal of
ABC the inscribed quadrangle, say KM, is
CE ED CB parallel to a side of the parallelogram
EA EA Figure 2
CA IAB or CDl,we'redone. Otherwisewe
mark the point P on BC such that PM
or xl @ - xl : aI b, So x = ab I la * bl and is parallel to AB. Using our moving
1<2x=zablla +bl. M12 method it's easy to transform PIMN
let s set the sum s = x+ y 20. Thenthe
into the triangle ABC, which is just
olution 2. If we fix the vertices A set V(s) of values assumed by
S
half of ABCD, so its area remains
and C, the locus of B wili be the circle
unchanged. Thus the quadrangles
with center C and radius a (fig.z) . By (r+y)2 + 3x *.r *-
the well-known property of a bisector,
=
f .
KLMN and PIMN have the same area,
equal to half the area of ABCD. Sub-
which can be obtained easily from whenxvaries from 0 to s, consists of
tracting the triangle IMNfrom both
figure 1 and ecluation (1), BDIDA = aII the integers from (s2 + sll2to (s2 + s)/
of these quadrangles, we get two triangles,
BC ICA = alb. Tlr,l.:s ,4D IAB = blla + bh 2 + s, each of them assumed once.
INI( and INP (fig. 7), with the com-
therefore, D is the image ol B after a Now let's notice that the last number
v
dilation with center A ardscale factor of V(s)andthefirstnumberof V{s + 1)
blla + b). This dilation maps the locus are consecutive:
of B onto the locus of D. So the latter
is a circle of radius ab I la + b) passing s2+1 , ^, ,I -(s*l)2+s+1
through C (seefigure 2), andl= CD is 2'"'' \f
--
2
always less than its diameter, Zablla + Therefore, the sets V(s) cover all the
b). Also, we see immediately that for nonnegative integers n without over-
anylfrom the interval 0 < 1 < Zab f la + laps or gaps (fig. 3). Since any n gets
b)there exists a triangle with a, b, and ;-_@
J astwo sides and the bisector between ,t 2, ,3
v(s)tU 4 5, ,6 7 I q, -

s0 I 2 3
them. x0 0 I 0 I 2 0 l2 3
y0 l 0 2 t 0 3 2 t 0

Figure 3 Figure 4

58 $IPTIttlBER/OCIOBEB lggO
 LI
Figure 6
Figure 5 + 1 faces (including F), because alace of
mon base lNand equal areas. From a convex polyhedron can't have more
this it follows that KP (or the side BC) than one edge in common with any
is parallel to the diagonal IN. (V. other face.
Dubrovsky) On the other hand, all the faces can
have no more than m different num-
Ml4
The answer is no. To prove it, let's
bers of sides (even less, since I and}
arg in facg impossibte). So there are at
introduce a coordinate system on the least two faces having an equal num-
plane such that the initial positions of Figure 7 ber of sides.
the frogs get the coordinates (0,0), (1,0), (the red "subgrid" is qnnmetrical about This last conclusion is based on yet
and (0,1) (fig. 8). It's easy to see that any point of the whole grid). Simi- another useful principle, this time a
when a frog sitting at{x,yljumps over Iarly, the "blue" and "black" frogs very simple theorem known as the
a frog at la,bl, itlands at the point (2a - have to stay on blue and black points, pigeonhole (or Dirichlet) principle: i{
x,2b -y) (fig. 9). So the parities of a respectively. We leave to you to prove more than m pigeons are placed in m
frog's coordinates don't change aker a that each frog can get to any point of its pigeonholes, then at least one pigeon-
jump. At the start each frog had at color. More difficult questions are hole has more than one pigeon in it.
least one even coordinate. Thereforg these: (1) Can two frogs simultane- (See the Contest Problem in this issue
none of them can hit a point with two ously get to any two given points of for more pigeon talk.)
odd coordinates, in particular the point their respective colors? (ZlWhat are Although our solution seems very
( 1, I
)-that is, the fourth vertex of the the triplets of points accessible to three much like many other pigeonhole
square. frogs at the same time? solutions, it can't do without convex-
This solution can be explained in a The second question has a simple ity, and so it's essentially geometrical.
more visual way with the grid shown and beautiful answer, but we won't [r facg two adjacent faces of a noncon-
in frgure 10. The grid contains three of deprive you of the pleasure of finding vex polyhedron can have more than
the four vertices of the initial square at it onyour own. (N. Vasilyev) one common edge (like a and b in
which our frogs start, and it's sym- figure 1 1). In this case our reasoning
metrical about each of its points.
Thereforg the frogs can't leave it to hit
Ml5
One of the most useful principles for
fails. Nevertheless, the statement of
the problem remains valid for non-
the fourth vertex. soiving olympiad problems says: "Take convex polyhedrons/ too. The only
A1so, for a similar reason the "red the extreme case." Following this rec- condition is that they shouldn't have a
frog" (starting at the red vertex in ommendation, let's consider the face hole that goes all the way through,
figure 10) can get oniy to red points F of a given polyhedron with the great- like the one shown in figure 12. The
est number of sides; let this number be proof is based on Euler's famous for-
(0,1) o m. Eachside of the face Fbelongs to mula
another face. This gives us atLeast m

CI taaaar)aaaeat
(0, 0) q..- w ( 1 ,o)
aw
laaaaa
a
aa la 3aa at aStlo
\--
Figure 8 ---/ a a O,;II aoa
a
aa oo oi'f ac altaa
a a a o aaa
oa ao oa lt ataaa
'/"r-----
(2a-x,2b-y) aa a a ao,
(a,b ) taaaC aO aa alact
( x,v) aa a o
Figure 9 Figure 10 Figure 11

OlJAlllTUllll/SOI.UIIOlllS 5g
 areas of the tire may wear more rap- F
idly than others. The practical result? F-l
ntin 2
You may have to buy new tires sooner
thanyou would if the wheels hadbeen Pl3
The total pressure on the table is the
balanced.
total force of gravity acting on the bell
Pl2
TheElastic string can be considered a
and water:
A
F = Ms + .
spring. Let the spring's rigidity equal ]nn3ps
k. The string breaks when the full
force reaches the value But when the water lifts the bell and
starts to leak out, the bell's weight
F,:mg+F11 (1) itself no longer acts on the table. The
Figure 12 pressure on the table then equals the
where m is the mass of the weight. pressure of the water multiplled by
Let's see what happens when the ap- the area of the base. The water pres-
v-e+f=2 plied force reaches the value F imme- sure is the same at all points (pgR), so
diately and then doesn't change. F: pgRrR2. This implies that
(where v, e, md'f standfor the number At the initial moment (when the ,/.
of vertices, edges, and faces in a poly- force is equal to zerol, the string is rR3Pg = Mg +]nn3Pg .

hedron), but it's rather cumbersome stretched because of the weight of the
and not elegant enough to present hanging mass mg by the amount x/ which yields
here. And yet another problem: does definedbythe expressionkxo = mg-
our statement remain true if holes are that is, the coordinate of the string's 14 =lna3p
J
.

allowed? (G. Galperin) end at equilibrium is equal to xo: mgf

k (the origin is assumed to be the A Kv ant r eader proposed an inter-
position of the string's end when the esting approach. Assume that the bell
Physics weight isn't attached). After alorce F is placed in a cylindrical container
is applied, the coordinate of the string's whose radius andheiglrt are both equal
Pl1 end at equilibrium is equal to x' : mgf to R. We'll fill the container with
The wheel is subjected to the reaction k + F lk. But the system reaches the water and take the bell's mass to be
force N. If the wheel is balanced, this equilibrium state only after some negligible. We can easily see that
force is equal in absolute valueto lf 4 oscillations. The initial amplitude of since the inside and outside pressures
of the automobile's weight Mlgl (as- these oscillations depend on the ini- on the bell are equal at all points (this
suming the car is traveling along a tial deviation of the system from the follows from Pascal's law and from
smooth, horizontal road). 1
equiiibrium position-that is, A = x - the assumption that the bell is very
If the wheel is unbalanced, its cen- xo: Flk. The damping rate of these os- thin), the water's equilibrium won't
ter of rnass moves along a circle of cillations is quite low, so the full ex- be affected if the bell is removed- Neither
radius z, and the centrifugal accelera- tension of the string alterhalf of the does the pressure of the water on the
tion is ro2r (r is the distance from the period is table change. But this means the
wheel's center of mass to the wheel's pressure of the water we poured into
center, ro is dre angular velocity of the
mg+F mg ^F
the cylindrical container acts exactly
wheel). When the center of mass
rkkk like the belt so the bell's mass is equal
reaches the lowest point, lNl oceeds 7+ This means that a total force of to the mass of the poured water:
Mlgl by mrffr (mis the wheel's mass). r\
When the center of mass is at the -lF.' : mc +2F M = I Y.r,,ro"r- Yt.*i.pt.r. JP
upper point, lNl is less th an1/+ Mlglby I . r ,\
the same value. When the wheel slips is being applied to the string. =
The string snaps when this force [R,iR--;,TR'rp
on the road surface (a certain amount
of slipping always occurs), the result- exceeds F,-see equation (1); this de- = 1nR3o .

ing force of friction between the wheel fines the string's strength limit. So the J

and road may have a different value. minimum force Fn* under which the P14
So with an unbalanced wheel some string breaks can be determined from The gas temperature is defined by the
the condition Fr' = Fr-that is, average kinetic energy of its mole-
lThe exact {raction of the
cules
automobile's weight corresponding to mg+2F^n=1118+Ft,
each wheel depends on the position of
*r2
the automobile's center of mass. In 22=1kr -

general, it'snot equal to 1/4. which readily yields

00 stPIt]tlBtR/0cI0BtR 1gg0
where k is tire Boltzmann constant.
This means that the higher the gas
measures the voltage drop on this re-
sistance. Dividing the voltmeter read-
814
No, since every such number is less
temperature, the greater the averagc ingby that of the ammeter, we find the than 10,0002 but greater that9,9992 :
velocity of its molccules and, consc- value of the resistance. 99,980,001.
quently, the greater the average mo-
lccular momentuln. 815
If the ternpcrature of the wall is the Brainlea$8r$ Winnie-the-Pooh had walked for 3
same as that clf the gas, a molecule minutes and Piglet for 6 minutes.
colliding wrth the wall changes its 81 1 Suppose it took x minutes for Winnie-
morrlentuln from po to -p,,. The change Point D shouid be positioned so that the-Pooh and Piglet to walk from their
in momentum is 2p,,. Whcn 7> 7,, the segment cD is equal to ll5 of segment respective homes W and P to the meeting
gas gets heated, which means the gas AC (fig.15); then the areao{ triangle point M. Winnie-the-Pooh spent x
molecules move from the wal1s u,ith a DBCwlllbe 1/5 thatof ABC. Simi- minutes walking from 1// to M and'
greater vclocity than the.v hacl before Iarly, point E is positioned so that BE : 1 minute hom M to P; therefore,
the collision. So the resulting mo- AB I 4,point F so th at FD :
AD 13, and W M I MP = x. The same reasoning for
mentum is greater than the initial point Gso thatEG:AE12. Piglet gles PMIMW = xl4. Since
molnentuln (fig. 13a), ancl the change WMIMP . PMIMW = l, xzl4 = l,
in momenturn is pgeater than )po. 812 whichgivesusx=2.
II T < T t, the gas gets cooler, so after Answer: I,999,999,999. If there are In tackling this kind of problem,
the collision a molecule 's momen- two numbers a and b greater than 1 you'll find it helpful to begin by plot-
tum is less than beiore ifig. 13b). In among the given numbers, then, re- ting the motions in question. The
this case the changc jn mornunrurr is piacing one of them with ab andthe graphs can then prompt you how to
evidently less than rvhen T > 7,. Ac- other with 1, we'll retain the product work out an equation or simply render
cording to Nertron's second 1ar'v, the of all the numbers and increase their the problem as pure geometry, as in
change in momentum is proportronal sum because the inequality la - lllb - figwe 16.
to the avcrage force from the wa11 1) > 0 implies that ab + | > a+ b. Thus,
acting on the molecule; and according the sum will be greatest if one of the
to Nervton's third law, the average numbers is a billion and all the others Boy-oh-[uoyaltcy!
force exerted on thc molecule is equal are equal to 1.
to the average forcc acting on the wall; l. Lp = 3p_ gAh = 4kPa (p* is the
therefore, the pressure of the gas on density o{ mercury).
the wall is greater when 7 > T, than 2. m = pnRzh-M(p is the density of
when 7< 7,. water).
3. S:M/(pgAhl = 6.25 cm2.
Pl5
The devices should be connected
4. The intensity with which the gas
as bums is determined by the difference
shown in figure 1:i. Pomts O, A, and B in the pressure of the gas and air. The
have the same potentials. iThe resis- pressure of the gas in the pipes of a
tance of the ammeter is 1ow, and rt,e building is usually low, and its density
can ignore the corresponding voltage is lower than the density of air. The
drop.) Consequently, there is no cur- decrease in air pressure on the top
rent through the resistances connect- floor of a fourteen-story building is
ing point O with A arrd B. This means greater than the decrease in gas pres-
Figure 14
that the ammeter registers the current sure. As a result, the difference be-
passlng through the resistirnce between
points O and C, while the voltmeter
813
Mark the level of the liquid and turn
the flask upside down.

T,{T T,)T

x x WMPM
x+l 'x+4 WP'PW '

Figure 13 Figure 15 Figure 16

OUAIIIUrII/$OI.UTIOilS
tween the pressures incieases, and the
gas bums better on the top floor.
5. The pressure the ball exerts on
the cover must increase.

lhleidoscopE
1. In the notations for figure I 7 let
the circumcircles passing through A'
meet again at P. Then angle BPC :
350"-angle APB -angle APC :360"-
(180"-y) - ( 180'-B) =y+ B. Thus, cx +
arrfle BPC = cr + y + f3 = 180o, so the third Figure 19
circumcircle also passes through P.
The casein which Plies outsidetriangle
ABC ts treated similarly.
2. In figure 18, angle ATC + angle
ATC f 120'+ 60" = 1B0o; therefore, ?
lies on line CC,. The case in which 7
lies outside the triangle recluires obvi-
ous changes.
3. Let the rotation about A through
60o turn an arbitrarypointMtnto M'
(fig. 19). Then AM = MM', BM = CtM.
The segments CM, MM', and M'C,
form one line if and only if M = T and
the angles of triangle ABC don't ex- Figure 1B Figure 20
ceed 120". tr this case CM + AM + BM
= CM + MM' + M'C,., CCt= I = CT +
TT'+T'Cr: CT +AT + BT.IIA>12fr'
lhg.2}l,l= CCl= CT-TT'+T'Cr= g7 lndex of Advertisers
-AT + 87. We leave it to you to prove
that in the latter case the smallest
valueof AM + BM + CM is attainedfor Arnerican Association of Physics Teachers 25
M=4. Cornell Theory Center 4
4. The answer is the intersection of
the diagonals.
Duracell 5

5. Hint: each of the centers of the Edmund Scientific 55

Torricelli circles passing through, say, Embry-Riddle Aeronautics L7
vertex A is equidistant from the ends Florida Institute of Technology 39
of segment 7H.
General Motors Institute 48
Grinnell College Back cover
Iona College 2
Intemational Educational Network 18
JETS 4?
Ma4rmount University 55
National Council of Teachers of Mathematics 65
Cnl'reflion
NSTA Special Publications 11,55,63
[r equation (1) on page 18 of the Oxford University Press 42
May issue, the plus sign should be Princeton University Press 3
a raised dot. (The equation is cited
corectly on page 21.) TOPS Learning Systems 22
University ofDayton 43

02 srPTrltlBtR/0cT0BrR 1gg0
 Soor to New Heights with NSTA Publicotions
Flights of lmagination: Experimentation
An lntroduction to Aerodynamics and Measurement
Revised Edition W. J. Youden
Wayne Hoskings This text takes students
Go fly a kite, and watch student involvement in scrence through the processes of
soar! The clear instructions in these '18 projects transform measuring: from taking
trash bags, dowels, and tape into high{lying lessons in measurements and
general science, math, recognizing possi ble sources
and science process of error to learning which
skills. Let Flights of measurements are important
lmagination add a and choosing proper
s- A.llioL(1110H
1O AI"}ODRMI]i
springtime lift to your equipment. Experiments are
middle- through high- rncluded so students can convince themselves of the
school science teaching value of statistics and careful measurement.
(grades 5-19) (grades 9-college)
#PB-61,'1990, 56 pp. #PB-2,1985,98 pp. $6 00
$7.00

All orders of 595 or less must be prepaid. Orders over $95 must include a postage and handling fee
of $2. No credits or refunds for returns. Send order to: Publication Sales, NSTA,
1 7 42 Connecticut Ave. NW, Washin gton, DC 2OO09
-1 1 7 1, (902) 398-5800.
Quantity discounts are available.
_l

Science and
Math Events:

Connecting
and
Competing
\ -:- .'
*=r
.*t9
: -"i i i
.
j, ':-.'.- lli, . ,.:
'...,.

.jl.
l'-'tti,'r
l
Er
jr rl',
OUANTUM lnaltes a [BrleclUiltl
=.
When you are trying to build student
interest and enthusiasm in math and
science, few resources can match the Use the response cardin this issue to
excitement generated by science clubs
and competitions. But how do you get order Quantum for your chil{ $and-
your high-school students involved? And chil{ nephew niece, mother, Iather,
how do you keep them involved? With
plans for successful fairs, details on 25
friend ... Four colorful, challenging,
national and international contests, and entertaining issues for only $14.00!
commentary by 89 prize-winning
scientists, this new publication prepares
you and your students for
connecting and competing in the 1990s. Factor x into fhe Quantum equation,
#P847,1990, 196 pp. $7.00
All orders of $25 or less must be prepaid.
where x rs any potential Quantum
Orders over $95 must include a purchase order.
All orders must include a postage and handling reader you know!
fee of $2. No credits or refunds for returns.
Send orderto: Publications Sales, NST 1742
Connecticut Ave. N!fl, \Tashington, D,C. 20009.

OUAITIIUIIl 03
 CHECKMATE!

tanla$y chess
By introducing new parameters, you can add some
interesting twists to the age-old game

by Yevgeny Gik

MONGTI{EUNUSUAIPROB- It seems that white will achieve a can't take the 91 rook with the bishop
lems and brainteasers in the realm checkmate inone move, and in two or the rook because the white rook
ol"fantasy chess" (that is, chess different ways: 1. Qa4xb3+ or 1. Bd7- will appear on al, putting the black
with an extra rule or two), there g4+. But it's not so simple . . . king in check again with white yet to
are some rather rare but extremely 1. Qa4xb3 (the knight leaves the move. And opening with 1. Rg7xgl +
witty variations. Let's take a look at board because gB is occupied by the won't work because the bishop re-
three of these. black rook)-mate? The defense 1. ... turns to f8 and the white king is al-
Kdlxd2 is impossible because the ready put in check.
Cil'ce captued bishop retums to cl and the Finally, 1. c7-cBQ(N)goes nowhere
This highly original version of the black kingwouldthusbe put in check because of 1. ... RgSxcB, and 1. c7-c8B?
fantasy genre differs from ordinary with white yet to move. But black has is a dead endbecause of 1. ... Bg1xe3 !
way: after an
chess in the following another, more clever defense up its
enemy piece is captured,it isn't re- sleeve: 1. ... NaBxcT!, and the white Irellh board
moved from the board-it's retumed pawn that appears on c2 blocks the In this interesting version of fan-
to the position it originally occupied diagonal a4-dl. It seems white will tasy chess the board is broken into
at the start of the game. Rooks and nevertheless achieve its goal 2. c2-c3 sixteen squares of four spaces each.
knights return to the square of the mate, but after 2.... b4xc3l a white The rules are simple: a piece is power-
same color as the one on which they pawn again appears on c2. less in the square it sits in (it can't
were taken, and pawns go to the start- l.Bd7-g4 won't work either. That move or attack enemy pieces); it can
ing position in the row in which they move can be repulsed by 1. ... Bglxe3! become active only by moving into
were taken. If, however, the point of The white pawn is restored to e2 and another square.
relocation is already occupied, the there's no mate after one [iov; there's
captured piece must leave the board, none after two moves eitherbecause
as usual. the e2 pawn can't move anywhere.
The quiet move 1. c7-c8R! will lead
TN to success. This panm disappears from
-.-N i NAN S the board and 2. Qa4xb3+ becomes a
threat. Black answers with 1. ... Rg8xc8,
\\N
\B\.
I
WN.Nr
'r whereby the newly made white rook
.Ni
.ur..\l\
\t-a.s i appearc on h1. The captr.re 2. Qa4xbl
E*.N is parried by 2. ... Rc8xc5l, and again a
white pawn shows up on c2. But now
"*N s i
we can get a checkmateby 2.8d7-g4+,
bringing the hl rook unexpectedly
N*ri into play: the black bishop is pinned E. Wisserman, 7955
and the response Bglxe3 can't be made. Mate in 2 moves on a tr ellis bo ar d
N. Macleod,7978 Can't black be savedby 1. ... Bglxe3
Mate in 2 moves (Circe) 2.Qa4xb} Be3xc5l? In this case [r this position the black king can't
2. Rg7-g1+ will prove decisive: black move to h5 orh6 (these spaces are in

04 srPItttlBIR/0[T0BtR lgg0
the same square as the king), while h2xg3 (the pawn turns into a rook)
the pawn on b2 can transform itself 2. Qh4-h5+! g4xh5+ (the pawn tums
into another piece only by capturing a into a queen), and the black king is
piece on c.l (in the adjoining square)! checkmated. It's amusing that the
1. Bf5-d7! This threatens 2. Bd4- white pawns in this diagram tum into
f6+, against which there's no defense. all sorts of pieces and at a rather
If black responds with 1. ... Bb7-f3, great distance from the last rank,
2. Nd2-E4+! is decisive-the bishop .NS*_N*Sft$ where these kinds of things usually
on f3 finds itself in the same square as
the knight and isn't allowed to cap-
happen. e
ture it. In response to 1. ... Qhl-e4, Does your library
which pins the bishop on d4, white's N. Bal<ke, 7986
next move is 2. Nd2-f3+ (the queen Cooperative mate in 2 moves have Quantum ?
protects the knight from the bishop (Erankfurt chess) If not, talk to your librarian!
on b7-an impracticable idea in an or-
dinary two-move problem). As you know, in a cooperative prob- Quantum is a resource that be-
Other variants: 1. ... Rc7-c6 2.R12- lem biack moves first and helps white longs in every high school and
g2+!; 1.... Qhl -f3(c6l 2. h2-h4+ ; L. ... achieve a checkmate. 1. 0-0-0! f4xe5 college library.
Rgl-S4 2. Rf2-f5+. hr normal play the (now e5 is occupied by a white bishop). See page 5I for subscription in-
knight on 96 would be defenseless, 2. Nc5-e7+ d6xe7+l Mate is effec- formation.
but on a trellis board it's untouchable. tively achieved by the new knight on
e7. Share the
tl'anklurt chess Curiously enough, this problem
In this kind of fantasy chess the has a very attractive twin. If the rook OUANTUM
capturing piece turns into the cap- is taken off a8 and put on h8, castling experience!
tured piece (without changing color). again leads to a solution: 1. 0-0!

Comic relief for the serious student.

MATHEMATICS AND HUMOR
Edited by Aggie Azzolino, Linda Silvey, and Barnabas Hughes

Read this unusual collection of

limericks, riddles, jokes, and car-
toons that poke fun at the usually
serious subject of mathematics.

You'll find that mathematics can

be funny and that mathematicians
can laugh at themselves.

Share a good laugh with your

friends! Order your copy now.
58 pp., #266, $4.50.

Ndtional Council of Teachers of Mathematics

1906 Association Drive, Reston, VA2209l
Tel. (703) 620-9840; fax (703) 476-2970

IUIIITlJllll/ClltCI(II4ATI!
ffiffiffiffiffiffiffi,ffiffiffi

ffiffiffiffiffiffiffi

ffiffiffiffiffiffi ffiY

r%ff$$",*&**:fl $
ffiH,fH'HIHfl;h,{.

ffi{k}
t; $
E ;s : n
#ffiffi
'*+ J"i'a 4#i.
'* A;{ f. H
1'!-" .& # ;:;:
?,.*- '.3+ fi&;..l:;U.4
Wt*f

1989 NobelLaureate in chemistryThomasR. Cech, recognizedlorhis RNAresearchwhich mayprovide

a new lo0l l0r gene technology, with potential to creale a new delense against viral infections.

You may be surpnsed to leam that Thomas Election to the National Academy of For more information,
R. Cech, the biochemist who shared the Sciences is anhonor second only toreceiving please urrite ot call:
1989 Nobel Prize in chemistry, is an honors the Nobel Prize. Six of the top 10 member-
gmduate of Grinnell College. producing institutions, I 1 of the top 20, and Office of Admission
Robert Noyce, the co-inventor of the 15 ofthe top 25 come from that group of48 Gnnnell College
integrated circuit and the father ofthe Infor- small liberal-arts colleges. P.0. Box 805
mation Age, also graduated with honors The sciences do not exist in a vacuum in Grinnell, Iowa 501 12-0807
from Grinnell College. the larger world. Nor do they at Grinnell. (s15) 269-3600
In fact, Grinnell College is one of 48 The college's open curriculum encourages FAX-(515) 269-4800
small liberal-arts colleges that historically science students to take courses in other
have produced the greatest number of sci- areas.
entists inAmerica. Grinnell andthese other Students who wish to focus their study
small colleges compare favorably with ma- may engage in scientific research, usually in
jor research universities, showing a higher a one-to-one relationship, under the direc-
per-capita production of graduates with tion of a Grinnell College faculty member.
sciencedegrees. The small colleges comprise Undergraduate student researchers often be-
five of the top 10 and 13 of the top 20 come the authors of scientific papers with
baccalaureate institutions in the proportion their professors at Grinnell College.
of graduates eaming Ph.D.s.
Circle No. 15 on Readers Service Card