25.

Alphametics
Consider the following problem. This is a problem on simple addition where each letter represents a unique
digit and M is not equal to zero.

S E N D
+ M O R E
M O N E Y

This classic example was published in the July 1924 issue of Strand Magazine by Henry Dudeney. Here, one
has to find the digits that are represented by the letters to make this addition correct. Most important task in this
activity is proper analysis and reasoning. At first, this may appear to be a daunting task, but if you take it step by
step, you will find it rewarding and entertaining.

The sum of any two four digit numbers cannot yield a number greater than 19,998. [The maximum possibility is
9999 + 9999]. Therefore M = 1.

Since M = 1, MORE < 2000. SEND < 10,000. This implies that MONEY < 12,000. Thus, ‘O’ can be either 0 or
1. But 1 is already assigned to M. Therefore, O = 0.
Now, we have;
S E N D
+ 1 0 R E
1 0 N E Y

Now, MORE < 1,100. If SEND is less than 9,000, then MONEY < 10,100, which would imply that
N = 0. But this cannot be, since 0 is already assigned to O. Therefore, SEND > 9,000. So, S = 9.
Now we have;
9 E N D
+ 1 0 R E
1 0 N E Y

Consider the digits in hundreds place. If there is no carryover to the hundreds column, then E + 0 = N, which
implies that E = N. This is not permissible. Therefore, there must be a carryover to the hundred column. This
implies that E + 0 + 1 = N, or E + 1 = N. It further implies that N + R = E + 10, if there is no carry over to tens
place or N + R + 1 = E + 10, in case there is carry over to tens place.

First we shall presume that N + R = E + 10 holds. Substituting the value of N = E + 1 in this equation, we get:
(E + 1) + R = E + 10, which implies that R = 9. But 9 has already been assigned to S. This implies that our
assumption N + R = E + 10 is not correct. So, N + R + 1 = E + 10 should be correct
N + R + 1 = E + 10
(E + 1) + R + 1 = E + 10 (Since N =E +1). This implies that R = 8.

273
We now have;
9 E N D
+ 1 0 8 E
1 0 N E Y
From the remaining list of available digits, we find that D + E < 14. Since there is carry over to tens place,
D + E = Y + 10. Further Y can not be 1, since M =1. So, Y is either 2 or 3.
If Y = 3, then D +E = 13, implying that the digits D and E can take only 6 or 7. If D = 6 and E = 7, then from the
earlier equation E + 1 = N, we get N = 8 which is unacceptable since R = 8. If D = 7 and E = 6, then, from the
earlier equation, E + 1 = N we get N = 9, which is again unacceptable since S = 9. Thus Y cannot be 3.
Therefore, Y = 2.

We now have:
9 E N D
+ 1 0 8 E
1 0 N E 2

Thus, D + E = 12. The only way to get this sum is with 5 and 7.
If E = 7, we again get, N = E + 1 = 7 +1 = 8, which is not acceptable.
Therefore, D = 7 and E = 5. We can now again use the equation E + 1 = N to get N = 6.
Finally, we get the solution:
9 5 6 7
+ 1 0 8 5
1 0 6 5 2

This is one of the celebrated problems of the world of “Alphametics”, an interesting bi-product of mathematics
which deals with the numbers in disguise! It is also called as “cryptarithms” (This name is originated from the
word “cryptography”, a branch of science that deals with encoding and decoding the passwords and letters
written in secret languages). A cryptarithm is an arithmetical puzzle in which each digit of an arithmetical
operation (addition, subtraction, multiplication ...) is assigned a definite letter as in a code. The problem is to
decode and find the digits assigned to each of the letters.

While posing the cryptarithm problem one must keep the following rules in mind:
1. The code must be one-one. All occurrences of a given digit must be replaced by the same letter, and
different digits cannot be represented by the same letter.
2. Common arithmetical conventions must be observed. For example, the first digit of a number cannot be 0.
Thus, in the problem given above, we must have S  0 and M  0.
3. The problem and codes must be such that precisely one answer exists. Problems with multiple solutions are
generally not considered “nice”. It is difficult to achieve and we shall sometimes be forced to relax this
requirement.
4. Finally, it is also an attractive feature to select the code so that actual names or dictionary words appear in
the problem.

274
Example 1
T H R E E
+ T H R E E
+ T W O
+ T W O
+ O N E
E L E V E N

Here, each letter represents a unique number and also we know that “three + three + two + two + one = eleven”!
It is hard to construct such a beautiful Alphametics!

The answer is,

8 4 6 1 1
+ 8 4 6 1 1
+ 8 0 3
+ 8 0 3
+ 3 9 1
1 7 1 2 1 9

Example 2
A much more elementary example is the following problem:
O N
 3
G O

The problem may also be written as ON + ON + ON = GO.

Here is how we do the analysis:
 Since 3  ON is a 2-digit number, 3  ON  99, therefore ON  33 and O  3. Thus, O = 1, 2 or 3.
 Since 3N  O, we examine the multiplication table for 3. We deduce the following:
O=1N=7
O=2N=4
O=3N=1
 Each of these possibilities gives rise to a solution. The solutions are:
17  3 = 51, 24  3 = 72, 31  3 = 93,
It is clear that this list of solution is complete.

Here, we have three possible values to each of the letters in the given problem. But, generally, such problems
with multiple answers are not entertained.

Example 3
Decode the following problem.

275
 = THREE
We reason as follows:
 Since has 5 digits, it follows that TWO  316 (because lies between 316 and 317). Therefore,
T is 1, 2 or 3.
 Since the left-most digit of is the same as that of TWO, namely T, we must have T = 1.
(If T were 2, then would exceed 40000 and we would have T  4; and if T were 3, then
would exceed 90000 and we would have T = 9)
 We know that no perfect square ends with 2, 3, 7 or 8. So, the digit in the unit place of a square must be one
of the following: 0, 1, 4, 5, 6, 9. Since O and E are distinct digits, they can be neither 0 nor 5. Therefore, E
is 1, 4, 6 or 9.
 We now use the property of perfect squares. The digit in the tens place of the square of an odd number can
never be odd. (For example, consider the squares of odd numbers: = 15129 and = 18769,
).
 Since the last two digits in THREE are both E, it follows that E cannot be odd. Therefore, E is either 4 or 6.
 Since E is even, it follows that THREE is even. So TWO itself is even. Therefore THREE is divisible by 4.
We now use the test for divisibility by 4. A number is divisible by 4 iff the 2-digit number formed by its last
two digits is divisible by 4. This implies that the number EE is divisible by 4. Since EE is either 44 or 66,
and of these only 44 is divisible by 4, it follows that E = 4. This implies, in turn, that O is either 2 or 8.
 The partial reconstruction now has the following appearance:
= 1HR44.
Since the number on the right is less than 20000, it follows that 1WO  141 (since
 141.42). As O is 2 or 8, it follows that W < 4; therefore, W is either 2 or 3. (Since T = 1, the
possibility that W = 1 is ruled out).
 Since O = 2 or 8 and W = 2 or 3, it follows that TWO is one of the following: 128, 132, 138. (Only three
possibilities are left!). The finish is now rather quick. We have, = 16384, = 17424,
= 19044. Only the third possibility fits (it is the only one where both the last two digits are 4). So we
have our answer:
TWO = 138, THREE = 19044

Example 4
Solve this alphametic.
A B C D
 9
D C B A
Here what could A be? If A is greater than 1, then we would have ABCD  2000; but
9  2000 = 18,000, a 5-digit number, whereas the product DCBA is a 4-digit number. So, A cannot exceed 1.
Further, as A is the first digit, it can not be 0. So, A = 1.

What would D be? Arguing from the units end, 9  D is either A or an integer with units digit A. Since, A = 1,
it follows that 9  D is a number with units digit 1. This forces D to be 9 (because 9  9 = 81)
Now, we have;
1 B C 9
 9

276
9 C B 1
We observe that 1200  9 = 10,800 a 5-digit number. This implies that B  1. Since we already know that A =
1, B can not be again 1. So, B = 0.

Now, we have;
1 0 C 9
 9
9 C 0 1

Since (9  C) + 8 is a number with units digit 0, 9  C is a number with units digit 2; therefore
C = 8 and ABCD = 1089.

Now check: 1089  9 = 9801. Here, we have not only obtained an answer but also shown that the solution is
unique!

Exercise

Choose the correct answer and enter it in the bracket provided

1. Find the number which represents L E T, choosing from the alternatives.

L E M
+ M E L T
= 9 9 9 9

(1) 4 5 0 (2) 9 4 0 (3) 4 0 5 (4) 5 0 4 [ ]

2. Find the number which represents S E T, choosing from the alternatives.

E N T
+ T E N S
= 1 8 6 7

(1) 6 3 1 (2) 6 4 1 (3) 4 3 6 (4) 3 8 4 [ ]

3. Find the number which represents M E N, choosing from the alternatives.

M A N
+ N A M E
= 3 3 3 3

(1) 3 2 1 (2) 1 3 2 (3) 1 2 3 (4) 1 0 3 [ ]

4. In the following multiplication problem, find the numerical values of S, V and Z in that order.

S T
U V W
 1 2
4 W Z R S
S T 4 V W
S 8 R U 7 2

277
 (1) 2, 5, 9 (2) 2, 9, 5 (3) 5, 9, 2 (4) 9, 5, 2 [ ]

5. Given that A C T  A T = 1 1, find out which of the following numbers does not stand for CAT?
(1) 2 4 6 (2) 6 1 5 (3) 6 2 4 (4) 8 3 5 [ ]
6. Given the following subtraction problem, find out which of the following numbers does not stand for
C A R T?

C A R
 A R T
= 2 2 2

(1) 6 4 2 0 (2) 7 5 3 1 (3) 8 4 2 0 (4) 9 7 5 3 [ ]

7. Find the number which represents S 0 I L, choosing from the alternatives.

C O C A
+ C O L A
= O A S I S

(1) 3 2 5 1 (2) 7 2 1 6 (3) 2 1 9 0 (4) 6 8 2 9 [ ]

8. Find the number which represents R I C E, choosing from the alternatives.

C I R C L E
+ C I R C L E
+ C I R C L E
= S P H E R E

(1) 8 0 1 5 (2) 8 3 2 4 (3) 9 1 4 5 (4) 5 3 8 4 [ ]

9. Find the number which represents H E L P, choosing from the alternatives.

C I R C L E
+ C I R C L E
+ C I R C L E
+ C I R C L E
= S P H E R E

(1) 8 5 7 0 (2) 7 0 6 5 (3) 5 7 2 1 (4) 4 2 6 0 [ ]

10. Find the number which represents S A T, choosing from the alternatives.

S T A R
+ R A T S
= S T A R T
(1) 1 9 8 (2) 1 8 9 (3) 8 9 0 (4) 1 8 0 [ ]

278