You are on page 1of 28

Piin the

Sky
Issue 14, Fall 2010

In This Issue:
(-1) x (-1)=1 . . . but WHY?
Palindrome Dates
Primes from Fractions
Reasoning from the Specific to the General
The Mathematics of Climate Modelling
On the Cover
Predicting the evolution of the Earth’s climate system is one of the Pi in the Sky is aimed primarily at high school students and
most daunting mathematical modeling challenges that humanity teachers, with the main goal of providing a cultural context/
has ever undertaken. In his article on page 17, Adam Monahan landscape for mathematics. It has a natural extension to junior
discusses the interweaving of mathematics with modern climate high school students and undergraduates, and articles may also
science. put curriculum topics in a different perspective.

Editorial Board
John Bowman (University of Alberta) Submission Information
Tel: (780) 492-0532, E-mail: bowman@math.ualberta.ca For details on submitting articles for our next edition of Pi in the
Murray Bremner (University of Saskatchewan) Sky, please see:
Tel: (306) 966-6122, E-mail: bremner@math.usask.ca http://www.pims.math.ca/resources/publications/pi-sky
John Campbell (Archbishop MacDonald High, Edmonton)
Tel: (780) 441-6000, E-mail: jcdotcalm@shaw.ca
Florin Diacu (University of Victoria)
Tel: (250) 721-6330, E-mail: diacu@math.uvic.ca
Table of Contents
Sharon Friesen (Galileo Educational Network, Calgary)
Tel: (403) 220-8942, E-mail: sfriesen@ucalgary.ca Editorial
Gordon Hamilton (Masters Academy and College, Calgary) by Anthony Quas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Tel: (403) 242-7034, E-mail: gamesbygord@gmail.com
(-1) x (-1) = 1 . . . but WHY?
Klaus Hoechsmann (University of British Columbia)
by Marie Kim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Tel: (604) 822-3782, E-mail: hoek@pims.math.ca
Dragos Hrimiuc (University of Alberta) Palindrome Dates in Four-Digit Years
Tel: (780) 492-3532, E-mail: dhrimiuc@ualberta.ca by Aziz S. Inan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Michael Lamoureux (University of Calgary)
Tel: (403) 220-8214, E-mail: mikel@math.ucalgary.ca Primes from Fractions
David Leeming (University of Victoria) by Alex P. Lamoureux and Michael P. Lamoureux. . . . . . . . . . . 8
Tel: (250) 472-4928, E-mail: leemingd@uvic.ca Reasoning from the Specific to the General
Mark MacLean (University of British Columbia) by Bill Russell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Tel: (604) 822-5552, E-mail: maclean@math.ubc.ca
Patrick Maidorn (University of Regina) Book Review
Tel: (306) 585-4013, E-mail: Patrick.Maidorn@uregina.ca Pythagorean Crimes
Wendy Swonnell (Greater Victoria School District) Reviewed by Gord Hamilton . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Tel: (250) 477-9706, E-mail: wswonnell@shaw.ca The Mathematics of Climate Modelling
Managing Editor by Adam H. Monahan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Anthony Quas (University of Victoria) Pi in the Sky Math Challenges
Tel: (250) 721-7463, E-mail: aquas@uvic.ca Solutions to problems published in Issue 13. . . . . . . . . . . . . . . 22
New Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Contact Information
Pi in the Sky
PIMS University of Victoria Site Office, SSM Building Room A418b
PO Box 3060 STN CSC, 3800 Finnerty Road, Victoria, BC, V8W 3R4
T/(250) 472-4271 F/(250) 721-8958 E-mail: pims@math.uvic.ca

Pi in the Sky is a publication of the Pacific Institute for the Mathematical


Sciences (PIMS). PIMS is supported by the Natural Sciences and Engineering
Research Council of Canada, the Province of Alberta, the Province of British
Columbia, the Province of Saskatchewan, Simon Fraser University, the University
of Alberta, the University of British Columbia, the University of Calgary, the
University of Lethbridge, the University of Regina, the University of Victoria and
the University of Washington.
Editorial
Anthony Quas

MATHEMATICAL Modelling: When light


is reflected
WHAT, HOW AND WHY? in water the
In my work and outside it too I often hear about reflection
Mathematical Modelling. In this editorial I'll say seems to be
a bit about what it is; how to do it; and why it's stretched out
important. along the line
connecting the
What is Mathematical Modelling? viewer and
the source.
First let's try a quick answer: Mathematical
Why is this?
Modelling is the process of using mathematics
Why isn't it
to understand something `in the real world'. For
stretched in the
high school students the closest thing to this in
perpendicular
the curriculum might be the (often-dreaded) word
direction
problems. Here is an example taken from the
(horizontally in
internet:
this picture)?
Mr.S is planting flowers to give his girl
In the case of a fast-spreading epidemic involving
friend on Valentine's day, which is half
an unknown disease (e.g. the SARS epidemic that
a year away. Currently the flowers are 7
hit China in late 2002 / early 2003 and spread
inches tall and will be fully grown once
to other countries including Canada) what are
they reach one foot. If the flowers grow
the best policies to follow to minimize sickness
at a rate of half an inch per month, then
and death from the disease while avoiding over-
how large will they be on the day? Will the
reacting?
flowers be fully grown? or will he have to
find another gift? Finally one of the most important cases of all:
climate modelling. Here you're trying to predict
The aim here is to take something which is on
the weather patterns in 10 years; 20 years; 100
the surface not a mathematical question, translate
years. For more information on this see the article
it into a mathematical question, solve the
in this issue by Adam Monahan.
mathematical question and translate the answer
back into the framework of the original question.
How do you do Mathematical Modelling?
In this case we let x denote the height of the flowers Just as in the case of the word problem, a first
on Valentine's Day and say that x  =  7  +  t / 2 step is trying to decide which variables to study
where t is the number of months until Valentine's in the problem. There is an important difference
Day. Since we're told it's half a year away we have though. The word problem is typically constructed
t = 6 so that we can compute x = 10. Since we're to give you all of the information you need to
told the flowers will be 12 inches when they're solve the problem (and no more). Generally the
fully grown it sounds as though `Mr.S' will have variables are given to you in the statement of the
to think of a new present. problem itself. When you're doing more advanced
modelling you have to decide which variables to
For some more advanced examples, here are some
include (and often equally importantly which ones
other questions that might be approached using
to leave out of consideration).
mathematical modelling.
1
This last idea is at first sight quite surprising. where you know the answer? If it's not working
Surely the best model is the one that takes you may need to tweak the model to improve it.
everything into account? Actually that's not
usually considered to be the case. Rather a good Why does mathematical
model is usually considered to be the simplest one modelling matter?
that explains the behaviour you're trying to study. Mathematical modelling can be great or it can
This is an important idea that is sometimes called be useless. It all depends on the model chosen.
`Occam's Razor' (named after a 14th Century Good mathematical modelling should make some
monk who first expressed it). predictions that you can test with existing data
For example in the case of the reflected lights one (it's important that you haven't already used the
might try to take account of the positions of all data to build the model-otherwise you end up with
of the water molecules! This would probably be a a circular argument: you build the model to work
bad idea because people don't know exactly how for a certain set of data and then check that it
liquids work. Also because there are a number of works with that same data). It should then make
trillion trillion water molecules in the picture it some predictions that you can test in the future.
would be impossible to do any computations even You can use mathematical modelling to test ideas
if you did know where they all were. that would be impractical or unethical to do as
In practice what you do is this: pick out some an actual experiment. For instance in the SARS
variables that you think could be important (in modelling example you might want to predict what
the water case I used the `slopiness' of the water: would happen if we took no action and compare
how far away from being flat it is as a variable for it to the effect of (say) shutting down schools and
example) and try to write down some equations airports for two weeks. Doing an actual experiment
relating the observations to the situation you're would be politically impossible but doing it in the
modelling. This may involve making some guesses model doesn't impact people in the same way.
as to how things work. Test your model: does it In the best case, mathematical modelling tells you
give the results you expected?; can you apply it to how things work and informs decisions. It is an
make predictions in situations other than the one invaluable tool across science and economics.

www.xkcd.com

2
(-1) x (-1) = 1 . . . but WHY?
Marie Kim, Cheong Shim International Academy

Positive and negative numbers are often explained


Marie Kim is at Cheong Shim International Academy,
currently in 11th grade. She's from South Korea, and
with the model of `debt and profit.' Profit is
plans to go to the U.S. for college and major in biology positive and debt is negative; adding positive
or neurology. numbers results in profit while adding negative
numbers deducts profit - adding to debt. This
Negative numbers. Most students find the model however, failed to make people understand
concept hard to understand and to accept at first. the concept of multiplying a negative number by a
Mathematicians before Descartes refused to accept negative number since it is hard to find such cases
negative numbers, including the great Pascal in everyday life.
himself. Negative numbers actually are believed
How is the concept `negative number x negative
to have been found in the East the earliest, in
number' explained, then? There are quite a few
China. An ancient Chinese text, written in
illustrations and models that are available to help
B.C. 1000, titled \Ku Jang San Sul" - meaning
people understand and accept such a concept. Out
­\nine arithmetic formulas"- includes in part a
of those explanations, here are three easy ones.
computation of negative numbers.

1. Pattern
2 x 2 = 4 1 x 2 = 2 0 x 2 = 0 (-1) x 2 = (-2) (-2) x 2 = (-4)
2 x 1 = 2 1 x 1 = 1 0 x 1 = 0 (-1) x 1 = (-1) (-2) x 1 = (-2)
2 x 0 = 0 1 x 0 = 0 0 x 0 = 0 (-1) x 0 = 0 (-2) x 0 = 0
Take a look at the pattern above. In each row, you'll be able to find that constant decrease in the
number multiplied results in constant change in the product. The same for each column. Applying this
same pattern, the next row will be:
2 x (-1) = (-2) 1 x (-1) = (-1) 0 x (-1) = 0 (-1) x (-1) = 1 (-2) x (-1) = 2
2. Number line
This pattern is an initial indication why the statement \negative number multiplied by negative number
results in positive number" might be correct.

On the number
2. Number line line, `2 x 3' is thought as `moving by 2 three times in the
same direction as 2 - the positive direction.'
On the number line, `3 x 2' is thought of as `moving by 2 three times in the same direction as 2 - the
positive direction.'

3
direction of 2 - the negative direction.'

Similarly, `3 x (-2)' is thought of as `moving by 2 three times in the opposite direction of 2 - the negative
direction.'

Now, keeping those two models in mind, `(-2) x (-3)' can be thought as `moving by (-2) three times in the
opposite direction of (-2)'; so it will be the same as `moving by 2 three times in the positive direction.'
This results in `(-2) x (-3) = 6'; negative number times negative number equals positive number which
Now,
would look keeping
like the firstthose
numbertwoline. models in mind, `(-2) x (-3)' can be thought as
(-2) three times in the opposite direction of (-2)'; so it will be the same as `m
3. Proof
three times in the positive direction.' This results in `(-2) x (-3) = 6'; nega
times negative
This one number
may be the equals positive
most complicated out of the number
three. Usingwhich would look
the distribution likeit can
property, thebe first
proved that multiplying two negative numbers result in a positive number.
Let's put `a' and `b' as two real numbers.
x  =  ab  +  (-a)(b)  +  (-a)(-b)
Let's expand it this way first.
x  =  ab  +  (-a)  {(b)  +  (-b)}  (factoring `-a' out)
Proof
3. x  = ab  +  (-a)  (0)
x  = ab  +  0
x  = ab
This one may be the most complicated out of the three. Using the distribution proper
Do it again but this time in a different order.
proved that multiplying two negative numbers result in a positive number.
x  =  {a  +  (-a)}  b  +  (-a)(-b)  (factoring `b' out)
x  =  (0)b  +  (-a)(-b)
x  =  0  +  (-a)(-b)Let's put `a' and `b' as two real numbers.
x  =  (-a)(-b)
So `x = ab and x = (-a)(-b),' which leads us to `ab = (-a)(-b)'
x= ab + (-a)(b) + (-a)(-b)
Other explanations for `(-)  x  (-)  =  (+)' do exist and can be found easily, for example using complex
numbers. But, don't just go and look it up; take your time and think of your own way to explain
`(-)  x  (-)  =  (+).'

*2

4
Palindrome Dates in Four-Digit Years
Aziz S. Inan

Aziz Inan was born in Turkey, did his PhD at Stanford is, \Can some of these eight-digit full dates be
and now teaches Electrical Engineering at University palindrome numbers?" (A palindrome number
of Portland. He enjoys posing Recreational Math Puz- is a number that reads the same forwards or
zles. He has noted that even his name has a puzzling backwards [1-2].)
geometric property. Write out AZIZ INAN. Swap A’s
and I’s and rotate the consonants by 90 degrees. The The answer is yes, and these special dates are
two names switch places! called palindrome dates.

Introduction For example, in the DDMMYYYY date format,


the first palindrome date of this (21st) century
In most of the world's countries, a specific occurred on 10 February 2001 since this date,
calendar date in a four-digit year is expressed in expressed as the single date number 10022001, is
the format DD/MM/YYYY (or DD.MM.YYYY indeed a palindrome number. Note that palindrome
or DD-MM-YYYY) where the first two digits date 10022001 is also the first palindrome date of
(DD) are reserved for the day, the next two (MM) this century that occurred in the MMDDYYYY
for the month and the last four (YYYY) for the date format, but it corresponds to a different
year numbers. (The United States is one of the actual date, which is October 2, 2001.
few countries which use the MM/DD/YYYY date
format in which the places of the month and the Generally speaking, palindrome dates are very rare
day numbers are switched.) In general, if one and sometimes don't occur for centuries. If any, only
removes the separators between the day, the month a single palindrome date can exist in a given four-
and the year numbers, a full date number consists digit year Y1Y2Y3Y4. Also, among all four-digit
of a single eight-digit number sequence given as years, a specific date designated by both month
DDMMYYYY. For example, the birth date of and day numbers as D1D2M1M2 (or M1M2D1D2)
the famous American recreational mathematician can be a palindrome date only once represented
Martin Gardner is 21 October 1914 which can be by a date number D1D2M1M2M2M1D2D1 (or
expressed as a single date number as 21101914 M1M2D1D2D2D1M2M1).
in the DDMMYYYY format (or 10211914 in the
MMDDYYYY format). In the DDMMY1Y2Y3Y4 date format, since
each single palindrome date number must satisfy
Palindrome
D=Y4 D=Y3 M=Y2 M=Y1 Y1 Y2 Y3 Y4 N
Dates
0 1 to 9 0 1 to 9 1 to 9 0 1 to 9 0 81
Assuming each 0 1 to 9 1 1,2 1,2 1 1 to 9 0 18
date in all the 1,2 0 to 9 0 1 to 9 1 to 9 0 0 to 9 1,2 180
four-digit years 1,2 0 to 9 1 1,2 1,2 1 0 to 9 1,2 40
is assigned a 3 0 0 1,3 to 9 1,3 to 9 0 0 3 8
single eight- 3 1 0 1,3,5,7,8 1,3,5,7,8 0 1 3 5
digit date 3 0 1 1 1 1 0 3 1
n u m b e r 3 1 1 0,2 0,2 1 1 3 2
DDMMYYYY,
Table 1
a question that
comes to mind Palindrome date combinations in the DDMMY1Y2Y3Y4 date format. Note that the total number of
palindrome dates in each category adds up to 335.
5
one being 29 November 1192 (29111192). No
1001 to 2000) can occur only in the first four
other palindrome dates occurred between the
centuries of each millennium. On the other
Y4Y3Y2Y1Y1Y2Y3Y4, M=Y4 M=Y3 D=Y2 D=Y1 Y1 Y2 Y3 Y4 N
palindrome dates can 0 1 to 9 0 to 2 1 to 9 1 to 9 0 to 2 1 to 9 0 243
only
TABLEoccur2 in years 0 1,3,5,7,8 3 1 1 3 1,3,5,7,8 0 5
ending with a digit Y4 1 0 to 2 0 to 2 1 to 9 1 to 9 0 to 2 0 to 2 1 81
less than four (since day 1 0,2 3 1 1 3 02 1 2
Table cannot
number 2. Palindrome
exceed date combinations
Table 2.
in the MMDDY1Y2Y3Y4 date format. Note that the total
number
31). In ofaddition,
palindrome
thedates in each category adds up to 331.
Palindrome date combinations in the MMDDY1Y2Y3Y4 date format. Note that the total
hundreds digit Y2 of number of palindrome dates in each category adds up to 331.
the year number of the
palindrome date must *5
either be zero or one (since month number cannot with day number equal to Y2Y1. Table 2 provides
exceed 12). In the special case when the thousands all possible combinations of palindrome dates in
digit Y1 of the year satisfies Y1 > 2, Y2 must equal the MMDDY1Y2Y3Y4 date format categorized in
zero. In other words, palindrome dates in the terms of different values and ranges of digits Y4,
second and third millenniums (years 1001 to 3000) Y2, Y3 and Y1, where N is the total number of
can only occur in the first two centuries of each palindrome dates in each category.
millennium. On the other hand, palindrome dates
According to Table 2, there are a total of 331
that fall between fourth and tenth millenniums
palindrome dates in the MMDDYYYY date
(years 3001 to 10000) can only occur in the first
format involving all the four-digit years.
century of each millennium. Also, between years
1000 to 10000, palindrome dates in each century Note that even if some palindrome date numbers
all fall on the same month with month number represented by Y4Y3Y2Y1Y1Y2Y3Y4 are valid
Y2Y1. Table 1 provides all possible combinations date numbers in each date format, unless the
of palindrome dates in the DDMMY1Y2Y3Y4 date day and the month numbers are the same, they
format categorized in terms of different values correspond to different actual dates in each date
and ranges of digits Y4, Y2, Y3 and Y1, where N format (e.g., palindrome date number 10022001
is the total number of palindrome dates in each as mentioned earlier). There are also palindrome
category. Based on Table 1, in the DDMMYYYY date numbers which are only valid dates in one
date format, a total of 335 palindrome dates exist date format but not the other.
among all the four-digit years.
For example, palindrome date number 21022012
In the MMDDY1Y2Y3Y4 date format, palindrome is a valid date number in the DDMMYYYY
dates represented by Y4Y3Y2Y1Y1Y2Y3Y4 can date format and represents 21 February 2012.
only occur in years ending with digit Y4 equal However, 21022012 is not a valid date number in
to either zero or one since the month number the MMDDYYYY format since its month number
cannot exceed 12. In the case when Y4 = 1, the 21 exceeds 12.
tenth digit Y3 of the year number cannot exceed
two. In addition, the second digit Y2 of the year Palindrome dates in the
number of the palindrome date must be less than second millennium
four since the day number cannot exceed 31. In the DDMMYYYY date format, a total of
Furthermore, if Y1 > 1, then, Y2 < 3. That is, for 61 palindrome dates occurred in the second
the MMDDY1Y2Y3Y4 date format, palindrome millennium (years 1001 to 2000), 31 in the 11th
dates in the second millennium (years 1001 to century (all in the month of January) and 30 in
2000) can occur only in the first four centuries of the 12th century (all in November), with the last
each millennium. On the other hand, palindrome one being 29 November 1192 (29111192). No other
dates between the third and tenth millenniums palindrome dates occurred between the 13th and
(2001 to 10000) only occur in the first three 20th centuries, more than 800 years.
centuries of each millennium. Also, between
years 1000 and 10000, palindrome dates in each In the MMDDYYYY date format, a total of 43
century all fall on the same day of the month, palindrome dates occurred during the second

6
millennium, split as 12, 12, 12 and 7 among the all in the month of December, to occur during
11th, 12th, 13th and 14th centuries respectively. the 22nd century and no more after that until the
The 11th century palindrome dates all occurred end of the third millennium. In the MMDDYYYY
on the first day of the month, the 12th century format, 12 more palindrome dates (all being on the
ones all on the 11th day, 13th century ones all on 12th day of the month) are to occur in the 22nd
the 21st day of the month, and the 14th century century followed by 12 more (all on day 22 of the
ones all on the 31st day of the month. The month) in the 23rd, and no more afterwards until
last palindrome date of the second millennium year 3001. In addition, there is a second palindrome
occurred on August 31, 1380 (08311380) and no date to occur in this millennium, 12122121, which
other palindrome dates existed between the 15th is not only common to both date formats but also
and 20th centuries, over 600 years. represents the same actual day in each format,
which is 12 December 2121. In the DDMMYYYY
Also, interestingly enough, among all the palin- date format, starting with the fourth millennium
drome date numbers in the second millennium there will be 31 more palindrome dates (all in the
common to both date formats, only two corre- month of March) to occur in the 31st century, 30
spond to the same actual dates: 1 January 1010 (all in April) to occur in the 41st, 31 (all in May)
(01011010) and 11 November 1111 (11111111). to occur in the 51st, 30 (all in June) to occur in
the 61st, 31 (all in July) to occur in the 71st, 31
Palindrome dates in the 21st century (all in August) to occur in the 81st, and 30 (all
The 21st century has 29 palindrome dates in the in September) to occur in the 91st centuries, the
DDMMYYYY versus only 12 in the MMDDYYYY last one being 29 September 9092 (29099092). No
date formats. The first palindrome date in this other palindrome dates will exist in all the other
century in each date format was 10022001, in
2001. The second palindrome date of this century
Number MMDDYYYY DDMMYYYY
in the DDMMYYYY date format was 20022002
10022001 10022001
representing 20 February 2002. The third 1
October 2, 2001 10 February 2001
palindrome date in the DDMMYYYY date format,
01022010 20022002
which also happens to be the second palindrome 2
January 2, 2010 20 February 2002
date in this century in the MMDDYYYY format, 11022011 01022010
is 01022010. This date number represents 1 3
November 2, 2011 1 February 2010
February 2010 in the DDMMYYYY date format 02022020 11022011
4
versus January 2, 2010 in the MMDDYYYY February 2, 2020 11 February 2011
format. The first twelve palindrome dates of this 12022021 21022012
5
century in both date formats are provided in Table December 2, 2021 21 February 2012
3. Note that in this century, palindrome dates in 03022030 02022020
6
the DDMMYYYY date format all occur in the March 2, 2030 2 February 2020
month of February. On the other hand, in the 04022040 12022021
7
MMDDYYYY date format, palindrome dates of April 2, 2040 12 February 2021
this century all fall on the second day of the month. 05022050 22022022
8
The last palindrome date of this century will be 29 May 2, 2050 22 February 2022
06022060 03022030
February 2092 (29022092) in the DDMMYYYY 9
June 2, 2060 3 February 2030
date format versus September 2, 2090 (09022090)
07022070 13022031
in the MMDDYYYY format. Also, there is one 10
July 2, 2070 13 February 2031
common palindrome date to occur in this century,
08022080 23022032
02022020, which corresponds to the same actual 11
August 2,2080 23 February 2032
date in each format, that being 2 February 2020. 09022090 04022040
12
September 2, 2090 4 February 2040
Palindrome dates after the 21st century Table 3.
After this century, in the DDMMYYYY date The first twelve palindrome dates of the 21st century in
format there will be 31 more palindrome dates, each date format.
7
centuries that fall between the fourth and tenth date in four- digit years in the MMDDYYYY date
millenniums (year 3001 to 10000). format will be 09299290, which is September 29,
9290 and after this one, no more palindrome dates
The number of palindrome dates to occur in the will occur until the next one which is October 10,
MMDDYYYY date format starting with the 10101 (101010101) to occur in year 10101.
fourth millennium up to the tenth is 36 per each
millennium, distributed as 12, 12 and 12 between Lastly, between the fourth and tenth millenniums,
the first three centuries of each millennium. The there is one common palindrome date that exists
12 palindrome dates in each century all fall on the in each millennium that corresponds to the same
same day of the month, which equals the reverse actual date in each date format. For example,
of the first two digits of the year number. For common palindrome date number 05055050 to
example, there will be 36 palindrome dates in the occur during the first century of the 6th millennium
6th millennium and these 36 palindrome dates will represents 5 May 5050 in each date format.
be distributed as 12, 12 and 12 between the 51st,
52nd, and 53rd centuries respectively. The 12 in References:
the 51st century will all happen on the fifth day
of the month, the 12 in the 52nd on day 15 of the 1. M. Gardner, The Colossal Book of Mathematics,
month, and the 12 in the 53rd are all going to be on Chapter 3, W. W. Norton & Company, 2001.
the 25th day of the month. No further palindrome 2. J. N. Friend, Numbers: Fun and Facts, Chapter
dates will exist in the other centuries (54th to VIII, Charles Scribner's Sons, 1954.
60th) of the 6th millennium. The last palindrome

Primes from Fractions


Alex P. Lamoureux and Michael P. Lamoureux
Alex Lamoureux is a senior at Queen Elizabeth decimal gives you 1," he points out, \which is not
School in Calgary. He enjoys reading, fencing, and a prime. Next one is 16, also not a prime. And
video games – in addition to his school work. after that is 166, 1666, 16666, none of which are
primes. So what are you talking about?"
Michael Lamoureux is a professor of mathematics at
the University of Calgary, He does research in analy- \No, Dad, you have to round up the numbers. So
sis and its applications to geophysics and signal pro- you start with 1.666 . . . which rounds up to 2,
cessing, and teaches courses from calculus to gradu- which is a prime. Next is 16.666 . . ., which rounds
ate research seminars. up to 17, which is also a prime.
After that is is 166.666 . . ., which rounds to 167,
Coming home also prime, and after that is 1667, which I'm pretty
\Hey Dad," announced Alex one day, arriving sure is a prime too."
home from school. \There is something weird
\You're right Alex, all of those are primes. But
about the fraction one-sixth. See, if you write it
the next one, 16667, is that prime?"
out in decimal form,
1/6 = .16666 . . . \I don't know, but let's check. Of course two
doesn't divide it, since it is not even. Three does
and start moving the decimal place, you get
not divide it since the digits don't add to a multiple
primes!"
of 3. Five doesn't divide it since it doesn't end in 0
\That's nuts," said his father, Michael. \The first or 5. What about seven?"
8
\Yes, what about seven? With our calculator, or 3 = 11(binary)
even in our head, we can check that 16667/7 = 7 = 111(binary)
2381. So that one is not prime." 31 = 11111(binary)
127 = 1111111(binary)
\Still, Dad, this fraction one-sixth is doing pretty
good. See, by comparison, if you look at one-half, and even the huge prime number
in decimal it is 243112609 - 1 = 111 . . . (binary) with 43112609
1/2 = .5000 ..., ones, which was discovered in 2008.

so shifting decimals gives you 5 (a prime), then So perhaps our one-sixth prime generator can also
the numbers 50, 500, 5000, etc., none of which are produce big primes for us. How can we tell?
prime."
Testing for primes
\Same thing with one-third, in decimal
To find out experimentally what is going on, we
1/3 = .3333 . . . , can use a computer program that can look at a big
so shifting gives you 3 (a prime), then the numbers integer and decide whether it is prime. There are
33, 333, 3333, etc., also none of which are prime. lots of tools out there: MAPLETM, MathematicaTM,
The fractions one-fourth, and one-fifth also give GIMPS, among others. Since the second author is
only one prime each." a mathematician, he has access to lots of these
tools. So we just use any one of them.
\So the one-sixth is pretty special. Why is that,
Dad?" Mathematica is a useful mathematical tool, with
plenty of commands to do all sorts of mathematical
Why is that? calculations quickly and painlessly. It has a very
simple command, PrimeQ[n], which tests whether
Why indeed? This is a question for a number
an integer n is prime or not. Mathematica is also
theorist, and neither of us are such. But it still
smart enough to keep track of all the digits of
is interesting to think about. The pattern of the
a very long number. Another nice command,
numbers generated by one-sixth are pretty special.
FactorInteger[n] will compute the prime factors of
Except for the first number 2, the integers we get
n, if we are interested in those.
are of the form of a one, followed by many sixes,
and ending in seven. That is, they look like this: So, for instance, we type in PrimeQ[16667] and
1666 ... 6667. discover the number 16667 is not a prime. Type
in PrimeQ[166667] and discover 166667 is actually
So these are always odd numbers, which is good, a prime.
as they will not be divisible by 2. The digits never
add up to a multiple of 3 (since the 1+7 is not A little \FOR" loop can be used to test a bunch
a multiple of 3, while the sixes all are), so this of numbers:
number is not divisible by 3. And of course it is
For[n=0, n<101, If[PrimeQ[Round[(1/6)*10^n]],
not divisible by 5. So at least we have a few factors
Print[n]]; n++].
that \can't" happen.
We can make the code a bit nicer by typing out
In fact, this pattern of repeating digits might
both the prime and its index n:
remind us of Mersenne primes, those primes of
the form n = 2k - 1. In binary notation, where we For[n=0, n<101, If[PrimeQ[x = Round[(1/6)*10^n]],
use only the digits 0 and 1, these primes are in Print[x,n]]; n++].
the form
With this little piece of code, we can find the
n=111 . . . 111(binary),with exactly k repeats of 1
prime numbers hidden in the decimal expansion
Of course, not all numbers binary numbers of this of any fraction.
form are prime, but many are, including

9
Some fractions
We start with the fraction 1/6 = .16666 . . ., and see how many primes appear in the first 100 digits.
We test for number of digits n up to 100, and get the following twelve primes:
2, n=1
17, n=2
167, n=3
1667, n=4
166667, n=6
1666666667, n = 10
166666666667, n = 12
166666666666667, n = 15
166666666666666666666666666666667, n = 33
1666666666666666666666666666666666666666666666666666667, n = 55
16666666666666666666666666666666666666666666666666666667, n = 56
1666666666666666666666666666666666666666666666666666666666667, n = 61

In fact, if we are patient (and have a fast enough a ratio of two integers. Everyone knows about
computer), we can use Mathematica to find more the number pi, which is irrational. Its decimal
primes of this form. Up to one-thousand digits, expansion starts out as π = 3.1415926 . . . Here, in
we find primes of the form 1666 . . . 6667 for digit the first 100 digits, we see only four primes:
lengths of n = 154, 201, 462, 570, 841, and 848.
There may be more! 3, n=1
31, n=2
Alex pointed out that one-sixth seems really 314159 n=6
special. So let's use this numerical test to compare 314159265359 n = 12
what we get with other fractions.
Another prime that we learn about in calculus is
To start, let's try the fraction 1/7 = .142857 . ... the exponential number e = 2.718 ....
We get the following five primes, with digit length
less than 100: There are only two primes that appear in the first
100 digits of the expansion.
1429, n=4
1428571, n=7 3,  n=1
1428571429, n = 10 2718281828459045235360287471352662
1428571428571429, n = 16 4977572470936999595749669676277240
1428571428571428571428571, n = 25 76630353547594571,  n=85

For the fraction 1/9 = .1111 . . . we only get get Wow, that's not a lot of primes!
three primes in the first one hundred digits:
Conclusion
11, n=2
1111111111111111111,  n = 19 Well, one-sixth is pretty special. We get a dozen
11111111111111111111111, n = 23 primes in its first one hundred digits, which is a
lot more than our other test cases! We have no
So again, it looks like one-sixth is pretty special. idea why. Michael thinks Alex should experiment
with more fractions, and more digits, to see what
Irrational numbers is going on. Or maybe some of the readers of this
article might be inspired to look into why we get
Not all number are fractions, some of them are primes this way.
irrational - numbers that cannot be written as

10
Reasoning from the Specific
to the General
Bill Russell
Bill Russell teaches math at James Bowie High concept such as odd x odd = odd. The following is
School in Austin, Texas, where he's worked for the a more advanced example of this principle.
past 22 years.

Solving a math problem that involves numbers can Problem A (Take 1)


often open the door to discovering a more general A total of 240 meters of fencing is to be used to
mathematical concept. Much of mathematics is enclose a rectangular region and divide it into 15
about relationships, and once these relationships smaller rectangular regions (see Figure 1). Find
are recognized, the next logical step is to try to the values of x and y that will maximize the total
extend them into more profound revelations. enclosed area.

For example, while practicing multiplication Solution: From the picture, 240 = 4x + 6y, or
tables, an elementary student might observe that y = -  x + 40. The goal is to maximize the area.
the product of two odd numbers always appears
to be an odd number. However, since there are
infinitely many 2-number combinations of odd
(
A(x) = xy = x -   x = 40  ) (2)

numbers, it is not possible to check each pairing


and make sure that the product is odd. However, We observe that the function A(x) is quadratic
a first-year algebra student has the tools necessary and therefore its graph has a vertical line of
to prove this relationship is always true. symmetry midway between its zeros. Since
equation (2) above is in factored form, one can
Any odd number can be represented as 2n + 1, easily finds its zeros by setting each factor equal
where n is some integer. A second odd number, then, to zero and solving. This reveals that the function
can be represented as 2m + 1 for some (possibly has zeros at x = 0 and x = 60. Furthermore,
different) integer m. The product of these two since the leading coefficient of (2) is negative, the
integers, then, is (2m + 1)(2n + 1)=4mn + 2m + 2n +1 function will take on a maximum value at the
 =2(2mn + m + n)+1 vertex, which, as symmetry dictates, is located
This last expression represents one more than midway between the zeros at x = 30. This gives a
an even integer, and is therefore an odd integer,
thus proving that the product of 2 odd integers is corresponding width of y = -  (30) + 40 = 20 ft.
always odd. Thus, the maximum area enclosed is
20 m x 30 m = 600 m2, a correct but less than
This very simple example shows how a specific provocative outcome.
numerical problem such as 3 x 5 = 15 can be
extended to arrive at a more general mathematical Of much greater interest is to observe that in this
optimum situation, the total horizontal fencing
Figure 1
used is 4 x 30 = 120 m, and the total vertical
Specific example of the fencing used is 6 x 20 = 120 m! The question that
fencing problem
should arise is whether this is a coincidence |
an accident of the specific numbers chosen |
or whether this is a numerical example of some
greater law of rectangular regions. Moreover, if
11
true, it is necessary to solve the related general problem by replacing the numbers with v
mind, the original problem can now be restated.
ue,
true,it itis isnecessary
necessarytotosolve
solvethe relatedgeneral
therelated generalproblem
problembybyreplacing
replacingthethenumbers
numberswith withvariables.
variables.With Withthis
thisinin
mind,
mind,the the original
original
problem
problem can
can now
now bebe Problem
restated.
restated. A (Take 2): A total of T meters of fencing is to be used to enclose a rectangular
this is a universal truth, then how would one prove a general theorem that would encompass
into smaller rectangular regions using m horizontal pieces, each of length x and n vertica
prove this? length y (see Figure 2). many
Show thatsuch
when specific
the enclosed problems,
area we restated
isdivide
maximized the
roblem
ProblemA A(Take totalofofT Tmeters
(Take2):2):A Atotal meters ofoffencing
fencing is is
totobebeused
used totoenclose
enclose a arectangular
rectangular region
region and
and divideit it that the total am
nto
intosmaller
smaller rectangular
rectangular
It would be easy regions
regions
enough using
usingmfencing
to repeat theused
mhorizontal
horizontal equalseach
pieces,
pieces,
computations the
each total
ofof amount
original
length
lengthx andof nhorizontal
xproblem
and in more
nvertical
vertical fencing
pieces,
pieces, used.
general
each
eachofterms
of and tried
lengthy (see
ength yusing
(seeFigure
Figure
other2).2).Show
Showthat
numbers, thatwhen
butwhen
even the
theifenclosed
enclosed
similararea
to prove
areais ismaximized
results maximized thatit.the
that Although
thetotal our ofproof
totalamount
amount depended mostly
ofvertical
vertical
encing
fencingused usedequals
equalsthethetotal
totalamount
amountofofhorizontal
horizontalfencing
fencingused.used. on some simple algebra, we T  mx
supplemented that
were observed the only thing Solution: that would In this version of the problem, T = mx + ny or y =
be proven , and the goal is to m
algebra with words that explainn to the reader
is that the rule holds for those specific numbers T T mxmxwhat we were doing. In the end, we were rewarded
olution:
Solution: InInthis
chosen. thisversion
version
To ofofthe
prove the
problem,
that problem,
it always T T= =mx mx+ +nyny
holds orory it
true, y= =is , and
, andthe thegoalgoalis istotomaximize
maximize
n nwith a remarkably simple  T  mxyet  powerful result
necessary to solve the related general problem by that precludes the need A ( x ) = x  .
forn repeating those same
replacing the numbers with variables. With this in calculations in the future. For example, we now 
T T
 mx
 mx 
mind, the original problem can now ) =xrestated.
A(Ax()x=be x

 . . know that if 300 meters of fencing is(3)(3)
  nn   available, m
We observe that A(x ) is a then quadratic the optimum enclosure uses 150 metersof is neg
function and that the leading coefficient
Problem A n
before that the function takesfencing in
on a maximum m
each m direction.
value at the vertex. as This
We Weobserve
observethat thatA(Ax()x is) isa aquadratic
quadraticfunction function (Takeandandthat
2)thatthe theleading
leading coefficient
coefficient   is isnegative,
negative, indicating
indicating as function has zeros
T n n T
efore
beforethat thatthethefunction
functiontakes takesonona amaximum maximum
x =A total , andofthe Tthe vertex lies In
meters midway Problembetween A, these
proving zerosthe
x x 0at xgeneral
0 = atat . Thus, case the
wastotal hori
mvalue
valueatat the vertex.
vertex.This This function
function hashas zeros
zeros at at = = andand 2 m
TT of fencing is to be T Trelatively easy because we were able to use the
x= = , and, andthe thevertex
vertexlies liesmidway
midwaybetween between Tthese
these T zeros atatx x= =
 = zeros . .Thus,
Thus, the
the total
total horizontal
horizontal fencing
fencing used
usedis isproblem
  to exact same methodology for the general
mm mused , precisely
enclose one-half
2 2
m m of the total fencing, thereby completing this proof.
a2m  rectangular
2 that we used in the specific numerical problem. The
 T T  T T
mm  = = , precisely
Figure , precisely one-half
one-halfofofthe thetotaltotal fencing,
fencing,
region andthereby
thereby
divide completing
only difference
completingthis thisproof.
proof.
was that in the general problem we
2
 m2 m 2 2 2 Let's take a
it into smaller moment and review
used what
variableswe just
in did.formula
the After solvingrathera than routine specific numeri
numbers.
General case of the fencing observed an interesting result. Although this is sometimes possible, solvingwould
Attempting to prove a general theorem that the encom
rectangular
specific problems,
regions
we restated the original problem in more general
et's
Let'stake
elated take a amoment
problem
general moment
problem andand
byreview
review
replacingwhatwhat we
the we just
justdid.
numbers did. After
After solving
solving a aroutine
routine specific
specific numerical
numerical problem,
problem, we we terms and tried to pr
proof using mwith
depended
variables.
horizontal
mostly on
With
general
some
this in
problem
simple
will sometimes
algebra, we supplemented
require
that
visualizing
algebra wor
bserved
observed
w be restated.
he related
anan interesting
interesting
general problem
result.
result.
by
Attempting
Attempting
true,
replacing itnreader
totoprove
isvertical
the
prove
necessary
numbers
a ageneral
general
towith
solve
theorem
theorem that
that
a completely
the related
variables. With
would
would
general
this in
encompass
encompass
different
problem
many
approach many
by replacing
such
such
to the
the numbers
problem,with aswith
variabl
pecific pieces,
specificproblems,
problems,weeach of length
werestated
restatedthe x and
theoriginal
originalproblem what we
pieces,
problemininmore wereeach
moregeneraldoing.
general In
terms the
terms end,
andand we
tried were
triedtoto proverewarded
proveit.it.Although with
Althoughoura remarkably
our simple yet
nroofnow be restated. mind, the original
precludes theproblem
need can
for nowthe
repeating be restated.
nextthose
example
same
illustrates.
calculations in the future. For example, we
proofdepended
meters
ssary depended
to
of
ofsolvelength
fencing mostly
the mostly
y
isrelated
to be onon
(see some
some
Figure
used
general tosimple
simple
2).
enclose
problem algebra,
Show algebra,
aby we
that
rectangularwesupplemented
replacing supplemented
when region
the
the andthat
numbers that
divide algebra
with algebra
it
variables. withwords
with words
With that
this that
inexplain
explaintotothe the
eader
reader
sing mwhatwhat we
horizontalwewere
were doing.
pieces, doing.
each InInthe
ofthe end,
end,
length meters
wex we were
and of
were fencing
rewarded
rewarded
n total
vertical is available,
with
withaeach then
aremarkably
remarkably the optimum
simple
simple yetyetenclosure
inpowerful
powerful uses
result 150
resultthatthat meters of fencing in
2): Apieces, ofseveral
nal problemenclosed
can area
now be isrestated.
maximized that the amount
T
frecludes
meters of fencing is to be used Problem
to enclose A (Takea rectangular of T At
totalregion meters
and divide ittimesis
of fencing to bethe used secondary
to enclosecurriculum,
a rectangular regio
tprecludes
when the ofthetheneed
needfor
enclosed
vertical forrepeating
area
fencingrepeating
is used those
maximized thosesame
equals same
that
thethecalculations
calculations
totaltotal amount
amount ininthe
ofthe
of future.
future.
vertical ForForexample,
m case
example, we wenow
tonow know
know that
thatif if300300the areas
xwe n vertical
ns
meters using
meters of m
of horizontal
fencing
fencing is is pieces, then
available,
available, each
then
into
the
smaller
ofthe
length
In
optimum
optimum xrectangular
Problem and n vertical
A,
enclosure proving
enclosure
regions
pieces,
uses the
uses
usingeach
students
general
150150 meters
meters
horizontal
of are
ofof
exposed
was
fencing
pieces,
relatively
fencing in in
each
aeach
each
unit
easy ofbecause
on
direction.
length
direction.
finding andwere able topiece
use
unt2): of horizontal fencing
T used.
wke that whenAhorizontal
total
theofenclosed
meters
fencing ofused.
area fencing
is length is to
maximized be that
y (see used
Figure
theto total
enclose
2). Showamounta rectangular
thatof when
of vertical region
the
triangles. and
enclosed
The divide
area is
following
methodology for the general problem that we used in the specific numerical problem. Th
it is
maximized
a problem thatthat
the they
total amount
ctangular
amount of regions
horizontal using m horizontal
fencing used. fencing pieces, used eachequals of length the total x and amount nmight
vertical pieces, each
of encounter
horizontal fencing of used.
nigure
InProblem
Problem A,A,proving
proving the thegeneral
general case
case that
was was in the
relatively
relatively general easy easy problem
because
because wewe weusedwere
were variables
ableable toininuse
tovertical
use the
such
the formula
the aexact
unit.
exact rather
same than numbers. Altho
same
2).Solution:
Show thatIn when T
the  mx
enclosed area is maximized that the total amount of
blem,
methodology
methodologyT = mx forfor
+ ny
the or
thegeneral y
general this
= problem
problem ,
version and
that possible,
thatwethe ofwegoal the
used solving
is
usedininthe to problem, the
maximize
thespecific general
specificnumerical problem
numericalproblem. will sometimes
problem.The Theonly require
onlydifference visualizing
differencewas was a completely diff
quals the total amount of horizontal n T mxfencing used. T mx
hat problem, as the next example illustrates. 
ethat ininthe
the
Tgeneral
general problem
+problem or we ywe =used used variables
variables in inthe the formula
formula rather
rather than
thannumbers. numbers. +Although
Although this
thisis issometimes
sometimes

problem,  = T mx = mx  +  ny ny or Solution: ,, and and Inthethis
the version
goal goal is toisof the
maximize
to Problem
problem, T =B mx(Take ny or 1) y= , and the goal is to maxim
ossible,
possible,solving solvingthe thegeneral
generalproblem problem n will will sometimes
sometimesT mx require
require visualizing
visualizing a acompletely
completely different
different n
approach
approach totothe
the
 Texamplemx T � �
 Find the area of a
is version
roblem,
problem, maximize
asA of
as the the problem,
next
next example mx
 =illustrates. ny
illustrates. Ator y
Tseveral
−=mx times, and in the the secondary
goal is to curriculum,
maximize students are exposed to aatunit
triangle with vertices (2,on1),finding th
(x ) = x . +
 A(x) = xThe following n is a problem (8,
 (3) that3) they
(3)
might encounter
 T  mxin such a unit.
and (4, 6).
 n T  mx  n
tAtseveral
several times
times A(inxin
)the
=thexsecondary
 secondary  .curriculum,
curriculum, students
students are are exposed
exposed to toa a
unitunit Aon
on ( xfinding
)
(3) = x the
finding theareas  . ofoftriangles.
areas triangles.
We observe thatn A(x) is aProblem
T
quadratic
mx B (Take
function 1): Find Solution:
the area of Most
a triangle of
 n vertices
my
with students
 at (2,approach this(4, 6).
1), (8, 3) and
he
Thefollowing
followingis isa aproblem problemthat thatthey theymight might
 encounter
 encounter
m
 ininsuch sucha aunit. unit.
tic function A ( x ) = x   . problem as by using Heron's Formula, which states (3)
andand that thatthetheleading
leadingcoefficient coefficient n  n isisnegative, negative,indicating
roblem
ProblemB B(Take (Take1):1):Find Findthe thearea area ofof
 Solution:
a atriangle
triangle with m
Most of my students that approach
the area this
of6).problem
a triangle by using
with Heron's m
sides ofFormula,
lengths which s
adratic
maximum function valueand at as that
the the leading
vertex. We
This observe
coefficient
function that
has withAvertices
zeros(xisvertices
is
at
atat(2,(2,1),1),(8,
axquadratic
) negative, 0 indicating
and at
(8,3)3)and
function and(4, (4, 6).
as and that the leading coefficient  is negative,
indicating before that the function
a triangle with takes on
n sides of lengths
a= a, ba, b, andc cisis A = s (s  a )(s  b )(s  c ),, where n s sis the se
T m , and where
yat
on A
olution:
Solution:x )maximum
Most
is a
a(maximum Most of of
my
quadratic
value my students
students
atfunction
atthe
approach
xat=vertex.
approach
andbeforethat
This this
the
that this problem
problem
leading
the
function function
has by by using
coefficient
zeros using
takes at Heron's
Heron's
x =is
on a 0maximum
isused
and Formula,
Formula,
negative,
atis tovalue which
which
indicating states
at thestates
as that
of vertex. that thethe area
area ofof
This functiondistancehas zeros at x
between these zerosvalue the .vertex.
Thus, triangle. the
This total
The horizontal
functiondistance has fencing
formula n
theis semiperimeter
used find that a, b,theand triangle.
c are 5, The 40 , and 29 . Sub
atriangle
trianglewith withsides sidesofoflengthslengthsa, 2m TT
a,b,b,and and c c
is is A A= = s ( s ( s
 a
 a
)(s)( s b
 b
)( s)( s c ) c
formula ,) where
, where s s
is
is used to is
thethe semiperimeter
semiperimeter
find Ta, b, and
of of
the the c total
function
dway takesat
between
zeros onxa = 0
these maximum
zeros andatatxvalue =x = at.into ,,theand
Thus,
andHeron's
the
vertex.thevertex
the Formula lies
Thishorizontal
total
vertex function
lies (and has
midway usingbetween
fencingzeros a calculator,
used xis= 0 zeros
at these since
and atat thexthatcalculations
= . Thus, with are radicals
thethose
5,
horizontag
f the total fencing, thereby completing this proof. T
iangle.
triangle. The The distance
distance formula
formula is is
used 2
used m m
to the
to find value
find thatthat of
a, a,the
b, b,
and area
and c c as
are are exactly
5, 5, 40 40 13.
,
, and
, and
and 29 29 . .
Substituting
. Substituting
Substituting 2 m
these
these values
these values
values into
einto
nto vertex
Heron's liesFormula
midway
Heron's midway
Formula between
between
(and(andusing these
using these zerosT  atatTsince
 zeros
a acalculator, xsince
= the .. calculations
Thus,
Thus, the total horizontal fencing used is a
alf of the total fencing, thereby completing mcalculator,
At  , proof.
=thispoint,
this 2mthe
precisely
the calculations
one-half of
observant
with
with
Heron's
the total
student
those
those
Formularadicals
radicals
fencing,
should marvel
get
(and get
thereby
at
pretty
pretty
using
the
messy)
messy)
calculator,
completing
result.
gives
Even
givessince the
thisthough
proof. two of the s
he thevalue
valueofofthe theareaareaasasexactlyexactly13. 13. 2m  2 � � calculations with those radicals get pretty messy)
hat we just
precisely the did.
totalAfter
one-half of thesolving
horizontal afencing
total fencing, routinethereby are
specific
used irrational m values
numerical
iscompleting
T and
T these numbers are plugged into a formula that requires
problem,
=this , proof. we the value of the area as exactly 13. taking a
gives
tempting
tAtthis
thispoint, to prove
point, the a general
theobservant
observant theorem
student
student shouldthat
shouldanswer would
marvel
marvel cameencompass
at at 2m
theout
the aresult.
nice
result. 2many neat
Even Even integer.
suchthough
though This
two two of raises
of
thethe the question
sides
sides hadhad lengths of whether
lengths that
that or not this will
ew whatprecisely we just did. After solving Let's
a routinetake aspecific moment and review
numerical what we
problem, just did. After solving a routine specific numerical pro
wecoordinates
re original
are irrational problem
irrational values
values inone-half
more
andandthese general
these ofnumbers terms
numbers you
are and
are onlytried
plugged
pluggeduse tointeger
prove
into it.values
intoaresult.
aformula Although
formula for
that
that the our
requires
requires taking
taking ofasquare
the vertices. Again, youshould
could try repe
a ageneral
square root,
root,the the
the total fencing, thereby
. Attempting to prove a general theorem observed an
that interesting
would encompass Attempting
At
many such to
this prove
point, the observanttheorem that would
student encompass
imple
nswer
answer algebra,
came
came out we
out supplemented
a what
anice
nice neat
neat integer. thatAfter
integer. algebra
ThisThis raises with
raises the words
athe question that
question explain
of ofwhether
whether to the orat
ornot not this
this will
oment and
d the original review
completing problem inwe
this more just
proof. did.
specific
general terms solving
problems,
and tried routine
we restated
to specific
prove it.the numerical
original
Although
marvel problem,
problem
our the inwill
result. always
we
more always
Even bebe
general thethe case
terms
though case if if tried
and
two of theto prove it.
eou end,
you only
teresting we
only use were
use integer
result. rewarded
integer values
values
Attempting with
forfor aprove
tothe remarkably
the coordinates
proof coordinates
athat
general
depended simple
oftheorem
ofthe yet
the
mostly powerful
vertices.
vertices.
thatsome
on result
Again,
would Again, youthat
encompass
simple you could
couldtry
algebra, try
manywerepeating
repeating
such are
supplemented thethecalculations
calculations
that algebra with words tha
me simple algebra, we supplemented algebra with words that explain
sides had to the
lengths that irrational values and
hose
ms, sameLet's calculations
take thearewarded
momentin the future.
and review inFor example,
what wewe just now did. know that toif prove
300
n thewe restated
end, we were original problem
with reader
a remarkably more
what general
we were
simple terms
doing. and
yet powerful In thetried end,
result wethat it. Although
were rewarded
these numbers are plugged into a formula that
our with a remarkably simple yet powe
nd the
mostlyoptimumon
After someenclosure
solving simplea uses
algebra,
routine 150 meters
we
specific
precludes supplementedof
numerical
the fencing
need in each
that
problem,
for algebra
repeating direction. with
those words
same that explain to the
ing those same calculations in the future. For example, we now know requires 300 a square root, the answer camewe now
that iftaking calculations in the future. For example,
ethenwerethe doing.
weoptimum In the
observed end,
an we were
interesting
enclosure uses rewarded
meters of fencing
result.
150 meters with a
Attempting remarkably
is available,
of fencing to
in each simple
then yet powerful
the optimum enclosure
direction. result that uses 150 meters of fencing in each
case for
need was relativelythose
repeating easy same because we were in
calculations able thetofuture.use the For exact example, same we now know that if 300
em
ng is
neral
that
case
we usedthen
available,
was relatively
in the thespecific
optimum
easy
numerical
becauseInenclosure
Problem we were
problem.
uses
A, proving
able150 The
tometers
the
usegeneral 12exact
only difference
theof fencing
case same
was
in each
was direction.
relatively easy because we were able to use the e
ed variables in the formula rather than numbers. Although this is sometimes
problem that we used in the specific methodology
numerical for the general
problem. The only problem differencethat wewas used in the specific numerical problem. The only
6

A1
5

out a nice neat integer. This raises 4 the question look like Figure 3. We want to tryAto 2
visualize the
of whether or not this will always be the case if area using integers rather than radicals. After a
3 A
you only use integer values for the coordinates of bit of thought, you might envision the triangle
2
the vertices. Again, you could try repeating the enclosed inside of a 6 x 5 rectangle, as shown
A3
calculations using different points. 1 Eventually, in Figure 4 (the gridlines have been deleted for
you would find that sometimesusing the different
area is points.
not Eventually,
clarity.) It you
is nowwouldclearfindthat
that the
sometimes
area Athe area is not an
is simply
1
1 2 3 4 5 6 7 8 9
1 multipleof
30 - A - A - A , where each of the A ( j = 1, 2,
an integer but rather is a multiple of (for
(forexample, if 1the point 2 (8,
3 3) is changed to (7, 3),j the area becomes
Figure2 3 3) is the area of a right triangle, Figure 4 which is half the
example, if the point (8, 3) is changed Specific example
trying several
to (7,sets of the
3), of points,
product youoffind
its that the area
legs.More always
general seems
picture of to be half of an inte
the area becomes 10.5.) However, triangle
actually proved
after problem
tryinganything yet, you are now ready to state
the triangle a hypothesis and try to prov
problem
several sets of points, you find further,
that see theif you
areacan For stateourthe numerical
property that we will need
example, then,towe prove.
haveFinished? My s
problem appears below. Keep in1 mind that 1 there are 1 other correct ways of stating it.
always seems to be half of an integer. Although
For our numerical example, then, we have A = 30  (6  2 ) (3  4 ) (5  2 ) = 13 , as previously found
you haven't actually proved anything Problem yet,B you(Take are2): Prove that 2if all of the2 vertices2of a triangle have integer coordi
now ready to Moreover, though, it triangle's
should
and trynow to be as why
previously
the areafound. Moreover, though, it should
state a hypothesis area isapparent
prove always an integer. must always be half of an integer. Armed wit
new any
it. Before reading insight, we aresee
further, ready to proceed
if you can state with our proof.
now be apparent why the area must always be
the property that we will need toSolution: We could tryhalf
prove. Finished? of anthis
to prove integer. ArmedFormula,
using Heron's with this butnew insight,
it should quickly bec
A proof is a connected series of statements weintended
are to
ready establish
to proceed a proposition.
with our As before,
proof. we ares (going
sa
My statement of the general approach problem would appears be messy. It would be necessary to show that the radicand
have to use words in these statements, and we want to choose those words carefully so that our proof s
below. Keep in mind that there perfect
are square,
other correct
what we intend for it to say. With this in mind,and since we would
let's is
A proof carefullyhave radical
analyze
a connected expressions
eachofofstatements
series for s,
these statementsa, b, and
intended c,
one at a lo
this t
ways of statingWeit.must first get rid nightmare.
of all numbers Consequently,
and replace we
them
to establish
arewith
motivated
variables to find
a proposition. as we a different
As did in our
before,
approach
wefirst
areproof,
to the probl
goingidentifyin
as the given information. This is easy enough. to have to use
Problem B (Take 2) Start by drawing the original triangle (thewords
one within numerical
these statements,
vertices) onand graph pape
Figure 3. We want to try to visualize the area using integers rather than
we want to choose those words carefully radicals. A
so that
Prove that if Statement 1: Let Xof
all of the vertices (a,ab )triangle
might
, Y (c, d ), have
and Z (e,our
envision the triangle
f ) be the says
enclosed
proof coordinates
inside
whatofwe of the vertices
a 6intend for itofto
 5 rectangle,
triangle
assay.
shown
XYZ, where
With in Figure 4
a , b , c , d , e and
integer coordinates, then twice the f are integers.
deleted for clarity.)
triangle's area It isthis
nowinclear thatlet's
mind, the area A is simply
carefully analyze30 each
 A1 ofA2these
 A3 , where
is always an integer. the area of a right triangle, which is half the product of its legs.
statements one at a time. We must first get rid of
Next we have to describe the rectangle enclosing all numbers and replace them with variables as we now, le
the triangle, and this is actually a bit tricky. For
Solution: Weclaim could the following:
try to prove this using 6
did in our first proof, identifying this as the given
Heron's Formula, but it should quickly become
information. This is easy enough.
A1
5
clear that this approach would be messy. It A2
would be necessary to show that the radicand 4
Statement 1
s (s - a)(s - b)(s - c) is always a perfect square, 3 A
and since we would have radical expressions for s, 2 Let X (a,b ), Y (c,d ) and Z (e,f ) be the vertices of
a, b, and c, this looks like an algebraic nightmare. 1 triangle XYZ, where a, b, c, d, e, and f are integers. A3
Consequently, we are motivated to find a different
approach to the problem. Next
1 2 3
we4 have
5
to describe
6 7 8 9
the rectangle enclosing
1 the triangle, and this is actually a bit tricky. For
Start by drawing the original triangle (the one now, Figure 3
let's claim the following: Figure 4
with numerical vertices) on graph paper. It should Specific example of the More general picture o
triangle problem
Statement 2 the triangle problem

1 sides parallel
A rectangle with its 1 to1 the
= 30 can(6be 2drawn
we have A axes
For our numerical example, then, coordinate ) (3so
 4 )that 5  2 ) = 13 ,
 (X,
2 2 2
Y, and Z all lie on the rectangle and at
Moreover, though, it should now be apparent why the area must always be half of an
leastwith
new insight, we are ready to proceed oneourof proof.
the vertices of the triangle
coincides with a vertex of the rectangle.
A proof is a connected series of statements intended to establish a proposition. As
This isand
have to use words in these statements, actually
we want a pretty strong
to choose (and
those not carefull
words
what we intend for it to say. With completely
this in mind, true!)
let's carefully
statementanalyze
andeach of these s
would
We must first get rid of all numbers
beand replace
rather them with
difficult variables
to justify as we did in our
rigorously,
as the given information. This is easy
but enough.
it seems obvious from our existing
Figure 3 Figure 4
drawing, so for now we're going to
Specific example of the triangle Statement
More 1: Let
general X (a, bof), Ythe
picture (c, d ),proceed
and Z (e,with
f ) be the
the coordinates
assumptionof that
the vertices
it of t
a, b, c, d , e and f are integers.
problem triangle problem
13
Next we have to describe the rectangle enclosing the triangle, and this is actually a b
2: A rectangle
Statement 2: A with its sides
rectangle withparallel to the
its sides coordinate
parallel axes can be
to the coordinate drawn
axes cansobethat X, Y,
drawn thatZX,allY,lieand Z all lie
soand
tangle and
is at least
true. one
Bear of
in the
mindvertices
that of
inthe
a triangle
truly coincides
thorough with a vertex
exercise of the
caution rectangle.
when
on the rectangle and at least one of the vertices of the triangle coincides with a vertex of the making
rectangle. this claim.
proof, this statement would require considerable
ually
This isa pretty
actually strong (andstrong
a pretty
justification. not completely true!) statement
(and not completely and would
true!) statement
Onebeof
andrather
woulddifficult
be ratherto justify
my more cynical college professors once
difficult to justify
, rigorously,
but it seems butobvious
it seems from our existing
obvious from our drawing,
existingsodrawing,
for nowso we're goingwe're
for informed
now to proceed
me towith
goingthat the words
the
proceed with the\without loss of
nassumption
that it Clearly,
is true. Bear
that ittherein mind
is true. that
areBear
many in a
in mind truly thorough
that inways
different proof, this
a trulytothorough statement
draw proof, would require
this statement
generality" considerable
would require
are almost always considerable
followed by a
n.justification.
the triangle and the rectangle based only on statement that results in a loss of generality.
Statement 2. We can only deal with one such While he likely overstated the extent of its
ere are many
Clearly, there different
are many ways to draw the triangle and the rectangle based onlybased
on Statement 2. We
eal with
orientation
one such
at adifferent
time,
orientation at a
ways
so weto
time,
draw the
make
so we
the triangle
make
following
the
and the
following
rectangle
misuse,
our next
his overallonly
statement:
on Statement
message to use 2.
thisWestatement
can only deal with one such orientation at a time, so we make the following our next statement:
our next statement: cautiously is worth remembering. For example,
3: Assume3:without
Statement Assume loss of generality
without that vertexthat
loss of generality X is vertex
in the lower
X is inleft
in our corner of the
first problem
the lower left cornerrectangle
you doand not want to claim,
of the rectangle and
es on the
vertex vertical
Statement
Y lies on theside 3to its right.
vertical side toFigure 4, then
its right. represents
Figure this particular
4, then represents \Assume this configuration.
particular configuration. = 5 and n = 3."
WLOG that and m
Assume without loss of generality that vertex X is There can be little doubt that this condition
,Effectively,
we are claiming arethat
welower the proof
claiming thatproceeds
the proof the same regardless
proceeds thevertex of limits
which triangle
same regardless of which
the vertex
triangle
generality coincides
of vertex with
coincides
the proof. with
By contrast, in
angle in the
vertex, as long left
as corner
all of the ofconditions
the rectangle
of and
Statement 2 are met. The disclaimer in bold (often
which rectangle vertex, as long as all of the conditions of Statement our 2current
are met.problem,
The disclaimer we in bold
could list(often
each possible
ed "WLOG")Y lies aon
is"WLOG") thetime-saver,
great vertical side but to
oneits right.exercise
should Figure caution
4, when making thismaking
claim. this claim.
abbreviated is a great time-saver, but one should exercise caution
configuration whenimplied by Statement 2 and work
then represents this particular configuration.
moreofcynical college professors once informed out the area for each, but are such an exercise would
One my more cynical college professors onceme that theme
informed words
that "without
the words loss of generality"
"without loss of generality" are
ways
almostfollowed
alwaysbyfollowed
Effectively,a statement
we arethat resultsthat
claiming
by a statement in aresults
loss ofingenerality.
that the proof While he likely
be
a loss of generality.
pointlessWhileoverstated
and repetitive. the extent
he likely overstated
Hence,the in extent
this case, the
se, hismisuse,
of its overall
proceeds message
his the same
overall to useregardless
message this to
statement cautiously
of which
use this statement is worth remembering.
triangle
cautiously use of \WLOG"
is worth Forseems
remembering. example, Forinexample,
our
justified. in our
em you do not want to
coincidesclaim, "Assume
with which WLOG
rectanglethat m 5
vertex,
= andas n = 3 ." There can
first problem you do not want to claim, "Assume WLOG that m = 5 and n = 3 ." There can be little doubt that
vertex be little doubt that
ion The restweofcould
thislimits the generality of the proof. By of
contrast, Byincontrast,
our 2current problem,
current problem,list weeachcouldpossible
the proof proceeds much as it would if
condition
long limits
as all the generality
of the of the
conditions proof.
Statement arein our list each possible
ion implied by Statement 2 and work out the area for each, but such an
numbers exercise
were
configuration implied by Statement 2 and work out the area for each, but such an exercise would be pointless
met. The disclaimer in bold (often abbreviated would
involved. be pointless
tive. Hence,
and repetitive.
\WLOG")in this case,
Hence, the
thisuse
is aingreat case,of the
"WLOG"
time-saver, use of seems
but onejustified.
"WLOG" seems justified.
should
fThe
the rest
proofofproceeds
the proof much
Statement 4 as it much
proceeds wouldas
if numbers
it would ifwere involved.
numbers were involved.

4: TheThe dimensions
dimensions
Statement of theofrectangle,
4: The dimensions
the rectangle, then,
then, are
of the rectangle, c are
a )are
(then, f (cab) x (
(c( -  f - b). Also,
)a. )Also,
( f  b ). Also,
1
A1 = (e Aa1)(=f 1(be), aA)(2 f=1b()c, Ae2)(=f 1 (dc), eand
)( f
1
(c Aa3)(=d 1 (bc). a )(
A3d=), and  d  b ). (4) (4) (4)
2 2 2 2
 2 2

Thus, Thus,
1
A = (c  a )(Af =(bc)a )([(fe ba))(f1[(be)+a(c)(f e)(bf)+ (dc)+e()( c f a )(dd)+ (bc)] a )(d  b )]  (5) (5)
2 2
(5)
2 A = 2and
and and (c 2aA)(=f 2(bc)a[()(e f a )(b )f[(be)+a(c)(f e)(bf)+ (dc)+e()(c f a )(dd)+ (bc)]. a )(d  b )]. (6) (6) (6)
eed
We to interpret
nowWeneed
nowtothis algebraic
interpret
need thisstatement
algebraic
to interpret (6) to explain
thisstatement
algebraic (6) why it establishes
to explain
statement why
(6) our proposition.
toitexplain
establishes
whyour proposition. our proposition.
it establishes
5: Since a5:
Statement , b,Since
c, d , e and
Statement a5, b,fcare
, d , eintegers and
and f are the setand
integers of integers is integers
the set of closed under the operations
is closed of
under the operations of
ubtraction, and multiplication,
addition, subtraction, it follows that
and multiplication, 2A must
it follows thatbe2A
anmust
integer.
be an integer.
Since a, b, c, d, e, and f are integers and the set of 6
6
uInare not familiar
integers with
is the
closed use
under of the
the word "closed"
operations ofin Statement
addition,
case you are not familiar with the use of the word "closed" in Statement 5, 5, 5 Figure 5
ays that any time you add, subtract, or multiply two integers, the
5
subtraction, and multiplication, it follows
this just says that any time you add, subtract, or multiply two integers, the that 2A An obtuse triangle
always be
result will
must an integer.
always be an anThese
be integer.
integer.closure
These properties justify the justify the
closure properties
4
4

that 2 A is that
that does not conform
nconclusion an integer.
2 A is an integer. 3
3 to the conditions of
In case you are not familiar with the use of the 2 statement 2
ppear that
It might word we are
appear finished
that
\closed" we in
arewith this proof,
finished
Statement with but
thiswe
5, this have
proof,
just buttowe
says tidy up one
have
that to tidy up one
2

. loose
For Statement
end.
any For time2 toyou
be true,
Statement each
2 to
add, be vertex
true,
subtract, oforthe
each triangle
vertex
multiply must
of the have must have
triangle
two
1
1
e coordinate
at leastintegers, that
one coordinate is either a maximum
that will
is either or a
a maximum minimum value
or a minimum within
the result always be an integer. These value within 1 2 3 4 5 6 7 8 9
e.
theThat is always
triangle. That true
is if XYZ
always true isifright
 XYZor acute,
is right but
or if XYZ
acute, butis if XYZ is 1 2 3 4 5 6 7 8 9

sobtuse,
closure
only true if one
properties
of the sides
justify
of the
the conclusion
triangle lies on
that
one
2Athe
of
is sides 1
it isinteger.
only true if one of the sides of the triangle lies on one of theFigure sides 5 -- An obtuse
1
an Figure 5 -- An obtuse
triangle that does not
triangle that does not
It might appear that we are finished with this conform toconform
the conditions of
to the conditions of
proof, but we have to tidy up one loose end. For Statement Statement
2. 2.

14
Statement 2 to be true, each vertex of the triangle obtuse case on your own. Choose your wording
must have at least one coordinate that is either a carefully, and always question whether or not
maximum or a minimum value within the triangle. your statements say what you intend for them to
That is always true if ΔXYZ is right or acute, but say. After writing your proof, have a friend read
if ΔXYZ is obtuse, it is only true if one of the it and give you feedback about how clearly you
sides of the triangle lies on one of the sides of the made your case.
rectangle. Again, this is not an easy statement to
justify rigorously, but a couple of quick sketches The body of knowledge encompassed by
should convince you that it is true. Thus, way mathematics is constantly growing. The impetus
back at the beginning, we need to split this proof for additions to this body of knowledge is curiosity.
into two separate cases. New relationships are constantly being discovered
because someone perceives an interesting result
Statement 0 to a problem and starts wondering how and
if this result can be generalized or extended.
Case 1 | ΔXYZ does not contain an obtuse angle. Although the examples in this article take you
down mathematical paths that have already been
Now that the proof for non-obtuse triangles is trodden many times, they hopefully show you how
complete, Statement 6 should begin the proof of such paths are found in the first place and have
the obtuse case. The picture for the obtuse case helped you develop some of the thinking skills
looks a bit different (see Figure 5), but otherwise required to experience mathematical adventures
the proof should proceed pretty much the same of your own. Happy exploring!
as the one that we just did. Try to prove the
century. Here the book delivers with aplomb:
�This was the first time in the history of
Book Review
mathematics that someone had been bold enough
to prove the existence of a mathematical solution
without illustrating how it was to be Gord Hamilton
constructed.
The article Hilbert published in 1888 created an
For example, to introduce a discussion of prime
In the Autumn of 2010,
uproar.
Dr. Hamilton The numbers,
opened ultraconservative
the author Kronecker, who self-
sacrificed authentic
doubted even
www.MathPickle.com to the existence
reflections of irrational
in order to teach thenumbers
mathematically
provide practical
such as help
thetosquare
unprimed:
root of 2, dismissed the solution
K-12 teachers. He is also
without further \I argument.
knew that primeGordan himself,
numbers were who
those that
a founding member of the
wasofknown
Game Artisans Canada for his be
cannot geniality and his
divided except generosityand the
by themselves
– Canada'stoward young, talented
premier board number one." mathematicians,
P. 19 responded
game design group.
angrily to Hilbert�s paper, saying, �This is the
notnarrator
...But within a page, a question from
mathematics, but theology.� As for Lindemann,
Pythagorean Crimes is written by the Greek ac- shows that he is at a significantly higher level:
Tefcros Michaelides. In a he described his former pupil�s method as
Crimesademic,
is written by the Greek unheimlichcompact \da \...Isn't that Gauss' conjecture? If I'm not
Vinci Code" style, the reader is led helter skel- � mistaken,profane. he Others, however, such as
worked out, but was unable to
os Michaelides. In a compact
ter through historical, mathematical, Arthur Cayleyprove,
and artistic in Cambridge andnumbers
Klein arestudied
how many prime smaller than
� style, the reader is led helter
events of the early 1900s - all to set up an exciting
the proof in detail and, havingP.initially believed it
a given number." 20
historical, mathematical, and to be impossible, ended up congratulating Hilbert
murder trial.
the early 1900s � the
all book
to set up thoroughly
an warmly. The motivation to expand the readership to
By the end of I was Hilbert
satisfied, became
include thosea that
fanatical
need anproponent
explanation of of prime
trial. but for the first part of the book, whereof the
proofs existence. �Inside this hall,� he would
numbers is understandable, but it unfortunately
majority of the mathematical history
of the book I was thoroughly often say in his is found, I lectures,the�there
undermines is atof least
believability one
the narrator.
was distracted: The self-reflections of the narrator,
the first part of the book, where student who has more hair on his head than any
The believability of the narrator doesn't matter
Michael Ingerinos, and the conversations with his
f thefriend,
mathematical historysometimes
Stephanos Kandartzis, is other. We don�t
felt forced. know
if you are who that
reading for person is, nor
the purpose is
of getting a
tracted: The self-reflections of the there any practical way of finding out. But this
el Igerinos, and the conversations doesn�t mean 15 that he doesn�t exist!�� P.59
glimpse of mathematicians at the turn of the last foundations of all these subjects are to be found
century. Here the book delivers with aplomb: in mathematics. What you see is an apotheosis of
algebra, trigonometry, infinitesimal calculus. It is
\This was the first time in the history of mathematics the tower of wisdom! I'm surprised you don't see
that someone had been bold enough to prove the ex- it that way."
istence of a mathematical solution without illustrat-
ing how it was to be constructed. The article Hilbert [Michael:] \It's the tower of Hubris," I said,
published in 1888 caused an uproar. The ultracon- annoyed \The only thing it symbolizes is human
servative Kronecker, who doubted even the existence arrogance. Mathematics can construct bridges,
of irrational numbers such as the square root of 2, houses, trains, and ships. It doesn't need such
dismissed the solution without further argument. contraptions to justify itself." P. 84
Gordan himself who was known for his geniality and
his generosity toward young, talented mathemati- That's an entertaining argument to eavesdrop.
cians, responded angrily to the paper, saying, \This I also like that the narrator's observations are
is not mathematics, but theology." As Lindemann, sometimes flawed. For example, he totally misses
he described his former pupil' method as unheimlich the melancholic mood in Picasso's paintings of
- profane. Others, however, such as Arthur Cayley in harlequins and pierrots - declaring them \cheerful
Cambridge and Klein studied the proof in detail and, stuff" P. 149.
having initially believed it to be impossible, ended
up congratulating Hilbert warmly. Hilbert became However, despite my enjoyment of the book, I
a fanatical proponent of proofs of existence. \In- am uneasy about recommending it for grade 12
side this hall," he would often say in his lectures, students.
\there is at least one student who has more hair
First, the book will not appeal to every top
on his head than any other. We don't know who
mathematics student because of its impressive
that person is, nor is there any practical way of
breadth of subject matters - the reader should
finding our. But this doesn't mean that he doesn't
be interested not only in mathematics but also in
exist" P. 59
history and art. Without these interests, a person
That is the kind of paragraph which made the book is liable to get irritated by the interjections - such
such a satisfying read. My reservations about the as the one page summary of the Dreyfus affair in
development of Stefanos and Michael are also off- the middle of a description of the delegates at a
set by insightful psychological sketches of other mathematics conference:
people:
\French society in 1900 was being torn apart by a
\... The top brass at [military] headquarters scandal revolving around Alfred Dreyfus, a young
considered me and the three other mathematicians lieutenant convicted six years earlier of being a
who worked with the French to be geniuses. This spy. After a farcical court-martial, the military
didn't stop them from burdening us with all sorts had sentenced him to hard labour for life and sent
of chores. Whenever they could, however, just so him to Devil's Island, a penal colony off the coast
we wouldn't forget our place." P. 155 of French Guiana. The royalists, backed by the
Catholic Church, seized the opportunity of the
By the middle of the book Michael and Stefanos conviction to attack the republican constitution.
were more believable and definitely interestingly The fact that Dreyfus was a Jew contributed
opinionated... For example, here is a discussion on to the rekindling of anti-Semitic feeling among
the newly constructed Eiffel Tower: certain sections of French society." P. 20
[Stefanos:] \This moment symbolizes a new era, Second, many high schools will not want to expose
the era of technology. What we have before us their students to some of the passages describing a
is a marvel of statics, dynamics, chemistry, bohemian lifestyle.
electricity - they have all combined to make this
the tallest construction in the world. And the In a nutshell - I thoroughly recommend the book,
but not necessarily for high school student.
16
The Mathematics of Climate Modelling
The Mathematics of ClimateAdam
Modelling
H. Monahan
Adam H. Monahan

Adam Monahan is an associate professor in the Euros or renminbi. Each of these forms of money
School of Earth and Ocean Sciences at the University represents a different “pool”. You can move money
of Victoria. His research focuses on studying interac- from one pool to another - depositing cash to a bank
tions between “large” and “small” scales in the atmo- account, or changing currencies; we can call such
sphere and ocean (the “weather-climate connection”). transfers “fluxes”. Fluxes don’t change the total
amount of money you have (if we neglect service
charges) - they just change the form that it’s in.
Introduction You gain or lose money by earning or spending -
these processes represent net changes in the total
Predicting the evolution of the Earth’s climate amount of money you have. Earning is a “source” of
system is one of the most daunting mathematical money, while spending is a “sink”. Different pools
modelling challenges that humanity has ever un- may have different sources and sinks - if you buy
dertaken. The atmosphere and ocean display vari- something with cash, that’s a sink for that pool; if
ability over a bewildering range of space and time you use your bank card, that’s a sink for your bank
scales - from microns to thousands of kilometers, account. The amount of money you have in a given
and from seconds to millions of years. The challenge pool increases with time if the sum (sources + fluxes
is increased by the fact that the climate system - sinks) is positive; it decreases with time if this
consists not just of the atmosphere and the ocean sum is negative. The process of budgeting involves
but also includes the cryosphere (frozen water in understanding these processes - how much you have
land and sea ice), the biosphere (life on Earth), and in each pool, what the fluxes are between them,
the geosphere (the solid Earth). While descriptive and what the various sources and sinks are. The
climate science is a centuries-old discipline, modern science (and art) of mechanistic modelling involves
climate physics - with a quantitative focus on representing these processes mathematically.
mechanism - is relatively recent. In fact, the first Mechanistic climate modelling is concerned with
global atmospheric circulation model was built only budgets of physical, chemical, and biological quanti-
in the mid-1950’s (a nice discussion of the history ties - energy, momentum, water (in various phases),
of atmospheric modelling for weather forecasting carbon (in various chemical compounds), terrestrial
is given in Harper et al. (2007)). Modern climate vegetation biomass, and many others. Newton’s
physics is a fundamentally mathematical discipline, Second Law of Motion
making use of tools and techniques from across the d
spectrum of modern applied mathematics. This F = (mv)
dt
brief article discusses the interweaving of mathe-
matics with modern climate science. is a mathematical description of the momentum
budget: accelerations are changes in the “mo-
mentum pool” of a physical object resulting from
Climate Modelling and Budgets sources and sinks of momentum which we call
forces. Newton’s Third Law of Motion - “the force
An idea fundamental to climate modelling is that of A on B is equal and opposite to the force of B on
of a “budget”. In everyday life, the budgets we A” - is just a statement that forces acting between
most often think about are budgets of money. The objects are momentum fluxes that may change the
money you have may exist in different forms - in momentum of each object (that is, an individual
cash or in a bank account; in Canadian dollars or momentum pool) but don’t change the total amount
17
ernal energy, etc.). quantity that we have at some time t, then t
Seen in this way, climate modelling seems budget analysis for the time rate of change of Q
aightforward. To build
of momentum a climate
(the sum over allmodel,
pools). allInyou
the
talked about earlier
Mathematics and can be written
Climate mathematica
Modelling
ed to do is
same kind of way, the First Law of Thermodynamics as the differential equation
is a description of the energy budget - energy is Consider the budget of some climatically significant
1. determine what
not created quantities
or destroyed, just are important
transformed to
between quantity - fordQ
example, the amount of CO2 in the
= 1).
sources
If Q is+the
fluxes - sinks
thepools
part(gravitational
of the climate system
potential you’reenergy,
energy, kinetic con- atmosphere (Figure
dt amount of this
internal
cerned withenergy, etc.).
understanding or predicting, quantity that we have at some time t, then the
Seen in this way, climate modelling seems budget
To make analysis for the time
predictions rate of
about howchange of Q we in tim
Q changes
straightforward.
2. decide what the Toimportant
build a climate model,
pools of all you
these talked about earlier can be written mathematically
we need to understand the fluxes, sources, and sin
need to do is as the differential equation
quantities are, and represent these mathematically. For exampl
1. determine what quantities are important to what aredQ atmospheric CO2 sources and sinks, a
3. come upthewith = sources + fluxes - sinks
part mathematical descriptions
of the climate system of
you’re con- what determines
dt how strong these are? An i
the fluxes between
cerned these pools, or
with understanding ofpredicting,
the sources, portant natural source is respiration
To make predictions about how Q changes in bytime,organis
and of2. the sinks, and - the
decide what the important pools of these we needrelease of COthe
to understand 2 as a byproduct
fluxes, sources, and of extracti
sinks
quantities are, energy
and fromthese
represent organic carbon (what
mathematically. we normally
For example - c
4. study the properties of the resulting mathe-
what are atmospheric
“food”). The flip side CO 2 sources and sinks, and
of this is a major sink
matical models
3. come up to make
with predictions
mathematical about the
descriptions of what determines how strong these are? An im-
climate.the fluxes between these pools, of the sources, atmospheric CO2 : photosynthesis, by which CO2
portant natural source is respiration by organisms
and of the sinks, and -transformed
the release of into organic
CO2 as carbon.
a byproduct To model th
of extracting
this is 4.
a breeze,
study theright? Not
properties really
of the - none
resulting of
mathe- processes
energy from mathematically,
organic carbon (what weweneed to express
normally call ra
ese steps ismatical
easy. models
As always,
to makethe devil isabout
predictions in the
the of photosynthesis
“food”). and
The flip side of respiration
this is a major(insink
vegetation,
of
tails. All four steps require expertise in physics,
climate. atmospheric CO2 : photosynthesis,
soils, in animals) as functionsby of which CO2 is we a
the variables
transformed into organic carbon. To model these
emistry, biology, or geology - and (particularly in modelling - a wonderful problem of mathematic
So this is a breeze, right? Not really - none of processes mathematically, we need to express rates
ps 3 and
these4)steps
mathematics.
is easy. As always, the devil is in the biology.
of photosynthesis and respiration (in vegetation, in
details. All four steps require expertise in physics, soils, in animals) as functions of the variables we are
chemistry, biology, or geology - and (particularly in modelling - a wonderful problem of mathematical
steps 3 and 4) mathematics. biology.

gure 1:Figure
Estimate
while
1: Estimate of global carbon cycle pools (or reservoirs) and fluxes. Black numbers are estimates of pre-industrial (e.g. natural) values,
of global
red numbers carbon
represent cycleofpools
estimates (or
changes reservoirs)
caused by humanand fluxes.(asBlack
activities of thenumbers are Chapter
1990’s). From estimates
7 ofofSolomon
pre-industrial (e.g. natural) val
et al. (2007).
le red numbers represent estimates of changes caused by human activities (as of the 1990’s). From Chapter 7 of Solomon et al. (2007).

18
2
which is the burning of organic carbon stored in (evaporation and precipitation) - that are the do-
geological reservoirs (coal, oil, natural gas). The main of the physical disciplines of mechanics and
strength
Thereofare thisother
source – thatofis,CO
sources the2 -rate
notat which
the leastwe
of thermodynamics.
of energy (for temperatures), budgets of freshwater
burnwhich
fossil isfuels
the –burning
is largelyof determined
organic carbon by economic
stored in And whatand
(evaporation determines the concentration
precipitation) - that are the of do-CO2
geological
decisions regardingreservoirs (coal, extraction
resource oil, naturaland gas). The
energy main of theWell,
in water? physicalthisdisciplines
is a wholeofnew mechanics and
budget problem
strength of this source – that is, the
generation; modelling these emissions represents an rate at which we thermodynamics.
- one with all the complications of the budget
burn in
exercise fossil fuels – is largelyeconomics.
mathematical determined by economic for And what determines
atmospheric the concentration
CO2 . Currents move CO of2COaround
2
decisions
And whatregarding resource extraction
about atmospheric CO2 fluxes?and energy
Well, in water?
along with Well,
thethis is a whole
water, new budgetinproblem
photosynthesis the ocean
generation; modelling these emissions
CO2 dissolves in water - in fact, the oceans contain represents an -consumes
one withit,all the complications of the budget
respiration releases it, CO2 reacts with
exercise in mathematical economics. for atmospheric CO2 . Currents move−CO2 around
more than 50 times the amount of carbon that seawater to form bicarbonate HCO3 and carbonate
And what about atmospheric CO2 fluxes? Well, along 2− with the water, photosynthesis in the ocean
the CO
atmosphere does - and CO2 is constantly being CO 3 ions - and all of these processes need to be
2 dissolves in water - in fact, the oceans contain consumes it, respiration releases it, CO2 reacts with
exchanged
more than between thesethe
50 times atmospheric
amount of and oceanic
carbon that represented
seawater to formmathematically.
bicarbonate HCOThis − involves more
3 and carbonate
poolsthe(Figure
atmosphere 2). This
doesCO 2 flux
- and CO2between the atmo-
is constantly being budgets,
CO 2−
ionsmore
- andequations,
all of theseand more complications;
processes need to be
3
sphere and ocean
exchanged is very
between important
these atmospheric– butand
howoceanic
do we as should be clear, even a straightforward
represented mathematically. This involves more ques-
represent it mathematically?
pools (Figure 2). This CO2 Physical
flux between chemistry
the atmo- tion likemore
budgets, “howequations,
much CO and is exchanged
2 more complications;between
tellssphere
us that andthe fluxisshould
ocean be proportional
very important – but howto dothe
we theshould
as ocean be andclear,
atmosphere?” can represent
even a straightforward a very
ques-
air-water COit
represent 2 concentration
mathematically? difference
Physical chemistry involved
tion like problem
“how much of physics,
CO2 is chemistry,
exchanged and betweenbiology
tells us that the flux should  be proportional to the  the ocean and atmosphere?” can represent
- and of the mathematics needed to model it all. a very
CO flux water to air = k [COwater ] − [COair ]
2
air-water CO2 concentration difference involved problem of physics, matters
chemistry,isand
2 2
Further complicating thebiology
fact that
  -very
and of
This
CO equation
flux wateris to
2
deceptively water ] − [COairthe
simple-looking:
air = k [CO 2 ] 2
often, we’re interested not just init the
the mathematics needed to model all. total
coefficient k bundles together a lot of very compli- Furtherof complicating
amount some quantity matters is the
in the fact that or
atmosphere
This equation is deceptively simple-looking: the very often, we’re interested not just in the total
cated physics. The air-water interface consists of ocean, but also in how the abundance varies from
coefficient k bundles together a lot of very compli- amount of some quantity in the atmosphere or
two boundary layers (one in each fluid) which are place to place. We don’t experience globally-
cated physics. The air-water interface consists of ocean, but also in how the abundance varies from
oftentwo
turbulent.
boundary Understanding and fluid)
layers (one in each modelling
whichthis
are averaged
place temperature:
to place. We don’t weexperience
experienceglobally-
what the
turbulence - the strength
often turbulent. and character
Understanding of which
and modelling this temperature
averaged is right here.
temperature: Therefore,
we experience whatwetheneed
are responsible
turbulence - for
themediating the exchanges
strength and character ofofwhich
CO2 mathematicalis models
temperature right here.of how quantities
Therefore, vary in
we need
(andareother quantities)
responsible betweenthe
for mediating theexchanges
fluids - isof “the
CO2 both space and
mathematical time -ofthat
models howis,quantities
partial differential
vary in
great unsolved
(and problem ofbetween
other quantities) classical
thephysics”.
fluids - is This
“the equations.
both space and These
time represent local budgets,
- that is, partial in which
differential
problem
great involves more budgets
unsolved problem - budgets
of classical of This
physics”. mo- the abundance
equations. These ofrepresent
our quantity of interest
local budgets, at every
in which
problem
mentum (forinvolves
the windsmore
andbudgets - budgetsbudgets
the currents), of mo- the
pointabundance
in space of our quantity
represents of interest
a different at every
pool.
mentum (for the winds and the currents), budgets point in space represents a different pool.

Figure
Figure 2: Air-sea
2: Air-sea CO2 fluxes (in moles of carbon per square meter per year) for the year 1995 as computed from a Global Climate Model (left
CO2 fluxes (in moles of carbon per square meter per year) for the year 1995 as computed from a Global Climate Model (left
panel) and as estimated from observations (right panel). Positive values indicate a CO2 flux from the ocean to the atmosphere; negative values
panel) and as estimated from observations (right panel). Positive values indicate a CO2 flux from the ocean to the atmosphere; negative values
indicate fluxes from the atmosphere to the ocean. While there is broad agreement between the observed and modelled fluxes, there are many
indicate fluxes from
differences the There
in detail. atmosphere to theofocean.
is still plenty work toWhile there
be done is broad
in climate agreement
modelling. between
Adapted fromthe observed
Zahariev et al.and modelled fluxes, there are many
(2008).
differences in detail. There is still plenty of work to be done in climate modelling. Adapted from Zahariev et al. (2008).

19
For example, the partial differential equation for physics. In particular, an emerging school of
the temperature field T (x, y, p, t) (in which pressure thought takes advantage of the essentially random
rather than altitude is used as the vertical coordi- character of small-scale turbulence to make use of
nate, following common meteorological practice) is tools from probability theory and stochastic
  processes.
∂T ∂T ∂T ∂T
cp ρ +u +v +ω − ω = 2 − ∇ · q
∂t ∂x ∂y ∂p
Such probabilistic perspectives are also useful
In this equation - which is just a mathematical
for appreciating the difference between “weather”
statement of the first law of thermodynamics - (u, v)
and “climate”. Robert Heinlein put it best: “Cli-
is the horizontal velocity vector, ω is the vertical
mate is what you expect, weather is what you
velocity (in pressure coordinates), cp is the specific
get”. This perspective is fundamentally proba-
heat capacity of air at constant pressure, ρ is the
bilistic: “climate” is the probability distribution of
air density, and 2 is the heating rate associated
outcomes for many rolls of dice, and “weather” is
with viscous dissipation in a turbulent flow. The
the outcome of any particular throw. Changes in
vector q is the spatial flux of heat energy through
climate are changes in the weighting of the dice.
the atmosphere due to the emission and absorp-
From a dynamical systems perspective, we can
tion of electromagnetic radiation, phase changes of
speak of the slowly-evolving “climate attractor” in
water, chemical reactions, and conduction. Each
contrast to the fast “weather trajectory”.
of u, v, ω, ρ, 2 , and q are themselves (x, y, p, t)-
dependent fields with their own budgets (expressed
as partial differential equations). Clearly, modelling Of course, all of these mathematical models
climate variability in both space and time can be would only be so much science fiction without obser-
very complicated indeed. vations to test them and to help estimate parameter
Once we’ve got these partial differential equa- values - so statistical analyses play a fundamental
tions written down, how do we use them to study role in modern climate science. These analyses
the climate system? Some things we can learn by help us understand our budgets. For example, the
looking at the form of the equations. In particular, increasing trend in atmospheric CO2 concentration
by considering the relative sizes of different terms is a result of sources being stronger than sinks and
we can gain some insight into the relative impor- fluxes (as we burn fossil fuels); statistical analyses
tance of different processes. We can even carry out of these trends help us understand just how out
formal asymptotic analyses of the equations, to of balance these budgets are. Statistical analyses
systematically “throw away” less important terms also help us understand how different parts of
to simplify the models. the climate system are related - for example, how
But even these simplified equations are still very variability in the Eastern Pacific sea surface temper-
complicated in general, and can’t be solved by ature is related to large scale pressure distributions
hand to make predictions. In this case, all we can in the El Niño - Southern Oscillation phenomenon.
do is replace the original equations with discrete Understanding these relationships in observations
approximations appropriate for simulation by com- helps us both build and assess our mechanistic
puters. That is, we need to make use of numerical models. Furthermore, all of these mathematical
analysis. In these approximate models, only space representations of sources, sinks, and fluxes in
and time scales above a particular threshold are mechanistic models have various parameters that
explicitly modelled. However, the smaller scales need to be set. Some of these - the speed of light,
cannot generally be ignored - the nonlinearity of the for example, or the acceleration of gravity - are
original partial differential equations leads to im- well known. Others, such as the dependence of soil
portant interactions between large and small scales. respiration rate on soil temperature and moisture,
The effect of the smaller unresolved scales must be are not so well constrained. Inverse modelling
parameterised in terms of the larger resolved scales. provides a systematic framework for estimating
There are different approaches to addressing the these parameters so as to bring mechanistic models
parameterisation problem, but some of the most into closest accord with observations – and thereby
promising make use of perspectives from statistical to make better predictions.
20
Conclusions Sarmiento and Gruber (2006) present a detailed
discussion of the ocean’s role in the global carbon
While the Earth’s climate system is intimidatingly cycle from a modelling perspective, and an excellent
complex, the reality of climate change compels us introduction to statistical analyses of climate data
to build models for understanding the system and is provided in Wilks (2005). Unfortunately out
predicting its future behaviour. These models, of print, Haltiner and Williams (1980) is a classic
expressed in the language of mathematics, range introductory text on numerical methods for meteo-
across a hierarchy of complexity. At one end rological modelling.
of this hierarchy are idealised conceptual models
that are useful for developing understanding but
cannot be expected to be quantitatively accurate. Acknowledgements
At the other end of the hierarchy are the fully
The author would like to thank Jim Christian for
complex Global Climate Models which represent
providing Figure 2, and an anonymous reviewer for
the physics, chemistry, and biology of the climate
very helpful comments on the original version of
system - but which are so complicated that they can
the manuscript. The author is supported by the
only be studied numerically using computers. The
Natural Sciences and Engineering Research Council
world’s most powerful computers require several
of Canada.
months of computational time to perform 1000-year
simulations with these complex models; increases
in computer power are generally consumed by in-
creases in model complexity or resolution. In fact,
References
models from across the entire range of this hierarchy Colling, A., 2001: Ocean Circulation. Butterworth-
play important roles in understanding and predic- Heinemann.
tion, and mathematics provides the fundamental
framework by which these models are constructed, Haltiner, G. J. and R. T. Williams, 1980: Numeri-
studied, and refined. cal Prediction and Dynamic Meteorology. Wiley,
New york, 477 pp.

Further Reading Harper, K., L. W. Uccellini, E. Kalnay, K. Carey,


and L. Morone, 2007: 50th anniversary of opera-
The climate literature is extensive and can be tional numerical weather prediction. Bull. Amer.
intimidating to the uninitiated. An comprehensive Meteor. Soc., 88, 639–650.
overview of the state of the art of climate science
is given in the most recent report of Working Hartmann, D. L., 1994: Global Physical Climatol-
Group I of the Intergovernmental Panel on Climate ogy. Academic Press, San Diego, 411 pp.
Change, available as a book (Solomon et al., 2007)
Holton, J. R., 2004: An Introduction to Dynamic
or online at http://www.ipcc.ch. A more succinct
Meteorology. Academic Press.
introduction to climate modelling is presented by
Thorpe (2005), while Weaver (2008) provides a Peixoto, J. P. and A. H. Oort, 1992: Physics
non-technical discussion of climate change (with of Climate. American Institute of Physics, New
a particular focus on Canada). There too many York.
good books on the physics of the climate to provide
an exhaustive list here; an excellent introductory Sarmiento, J. L. and N. Gruber, 2006: Ocean
text is Hartmann (1994), while a more quantita- Biogeochemical Dynamics. Princeton University
tively detailed treatment is provided in Peixoto Press.
and Oort (1992). Colling (2001) and Wallace and
Hobbs (2006) are good introductions to physical Solomon, S., D. Qin, M. Manning, Z. Chen,
oceanography and meteorology, respectively; deeper M. Marquis, K. Averyt, M. Tignor, and H. Miller,
treatments of modern geophysical fluid mechan- eds., 2007: Climate Change 2007: The Physical
ics are given in Holton (2004) and Vallis (2006). Science Basis. Contribution of Working Group
21
I to the Fourth Assessment Report of the In- Weaver, A., 2008: Keeping Our Cool: Canada in a
tergovernmental Panel on Climate Change 2007 . Warming World . Viking Canada.
Cambridge University Press.
Thorpe, A. J., 2005: Climate change prediction: Wilks, D. S., 2005: Statistical Methods in the
A challenging scientific problem. Institute of Atmospheric Sciences. Academic Press.
Physics.
Vallis, G. K., 2006: Atmospheric and Oceanic
Zahariev, K., J. R. Christian, and K. L. Den-
Fluid Dynamics: Fundamentals and Large-Scale
man, 2008: Preindustrial, historical, and fertil-
Circulation. Cambridge University Press.
ization simulations using a global ocean carbon
Wallace, J. M. and P. V. Hobbs, 2006: Atmo- model with new parameterizations of iron limi-
spheric Science: An Introductory Survey. Aca- tation, calcification, and N2 fixation. Progress in
demic Press. Oceanography, 77, 56–82.

PiPiininthe
theSky
Sky Math Challenges
Math Challenges
Solutions to the th
problems published
Solutions to the problems published in the 13 Issue of Pi in the Sky
in the 13th issue of Pi in the Sky

Problem 1: Find all positive integers n such that log2008 n = log2009 n + log2010 n.

Solution:
A solution for the above equation is n = 1. Let us prove that there are no other solutions.
For any positive integers k > 2, n > 1 we have
1 1 1
logk+2 n + logk+1 n − logk n = + −
logn (k + 2) logn (k + 1) logn k

logn (k + 1) (logn k 2 − logn (k + 2)) + logn (k + 2) (logn k 2 − logn (k + 1))


= >0
2 logn (k + 2) logn (k + 1) logn k
since k 2 > k + 1. Hence, if we take k = 2008 we get log2009 n + log2010 n > log2008 n.

2
Problem 2: Find the smallest value of the positive integer n such that (x2 + y 2 + z 2 ) ≤
n (x4 + y 4 + z 4 ) for any real numbers x, y, z.

Solution: The given inequality can be transformed into an equivalent useful format:
   2  2  2
(n − 3) x4 + y 4 + z 4 + x2 − y 2 + y 2 − z 2 + z 2 − x2 ≥ 0.

Since the minimum value of (x2 − y 2 ) + 22


2 2 2
(y 2 − z 2 ) + (z 2 − x2 ) is 0 , we must have
4 4 4
   2  2  2
(n − 3) x4 + y 4 + z 4 + x2 − y 2 + y 2 − z 2 + z 2 − x2 ≥ 0.
2 2 2
Since the minimum value of (x2 − y 2 ) + (y 2 − z 2 ) + (z 2 − x2 ) is 0 , we must have
(n − 3) (x4 + y 4 + z 4 ) ≥ 0, hence n ≥ 3.

An alternative solution, using a geometric argument was given by Carlo Del Noce, Gen-
ova, Italy.

Problem 3: Let a be a positive real number. Find f (a) = maxx∈R {a + sin x, a + cos x}.

Solution:
This problem is trivial. Since maxx∈R {a + sin x, a + cos x} is clear a + 1 then f (a) = a + 1.

Problem 4: Prove that the equation x2 − x + 1 = p(x + y) where p is a prime number, has
integral solutions (x, y) for infinitely many values of p.

Solution:
Let us assume by contradiction that the equation has integral solutions (x, y) only for
a finite number of prime numbers, among which the greatest is denoted by P. If we take
x = 2 · 3 · 5 · .... · P then x2 − x + 1 = x(x − 1) + 1 is not divisible by any of the prime
numbers which are ≤ P. Hence x2 − x + 1 = Qm, where m is an integer and Q is a prime,
Q > P. So, if we take y = m − x, then (x, y) would be an integral solution of the equation
x2 − x + 1 = Q(x + y), which is a contradiction.

This problem was also solved by Carlo Del Noce,Genova, Italy.

Problem 5: Find all functions f : Z −→ Z such that 3f (n) − 2f (n + 1) = n − 1, for every


n ∈ Z. (Here Z denotes the set of all integers).

Solution:
It is clear that f (n) = n + 1 is a solution of the problem. Let us prove that there is
no other solution. If we set g(n) = f (n) − n − 1 and using the given equation one obtains
3g(n) = 2g(n + 1) for every integer n. Assume by contradiction that there is an integer m
such that g(m) = 0. Then

2g(m) = 3g(m − 1) = 32 g(m − 2) = ... = 3k g(m − k)

for any positive integer k. Hence 3k |g(m) for any positive integer k and therefore we must
have g(n) = 0 for any integer n.

A correct solution was received from Carlo Del Noce,Genova, Italy.


23
Problem 6: In ∆ABC, we have AB = AC and BAC  = 100◦ . Let D be on the extended

line trough A and C such that C is between A and D and AD = BC. Find DBC.

Solution:
Ahmet Arduç from Turkey, submitted five solutions to this problem. The following is his
first solution. In ∆ABC, we have AB = AC and BAC = 100°. Let D be
 on the
Let α = m(CBD). Drawextended line through
the equilateral A and C such that C is
triangle AED.
between Awe
Since ∆ACB ≡ ∆BAE(SAS) andgetDAB
and=AD BE,=m(BC.BAE)
 Find=DBC. = 40◦ , m(ABE)
m(BEA)  =

100 . As AB = BE and AD = DE, ABED is a deltoid with BD the axis of symmetry. i) Dra
 = m(DBA)
Hence m(DBE)  = 50◦ = 40◦ + α. Thus α = 10◦ . ii) Dra
iii) m(B
also solved– this
Carlo Del Noce Solution 1: problem. ⟹m
⟹m
iv) ECB

Solution – 3

i) Draw equilateral triangle AED,


ii) ACB ≌ BAE (SAS congruency)
a. |AB| = |BE|
\It is proven that theb. celebration
m(BAE) = m(BEA) = 40° is healthy. Statistics
of birthdays
show that those peoplec. m(ABE) = 100° the most birthdays become
who celebrate i) Dra
iii) Let m(CBD) = α. ii) We
the oldest." iv) Since |AB| = |BE| and |AD| = |DE| iii) Dra
   S. den Hartog, a. PhD ABED is a deltoid,
Thesis, University of Gronigen iv) EAD
b. [BD] is the symmetry axis of the deltoid a.
ABED b.
c. m(DBE) = m(DBA) = 50° = 40° + α v) Sinc
⟹ α = 10°. 160
24 vi) m(E
vii) m(C
Math Challenges
Problem 1:
Find all the real pairs (x, y) such that
log 3 x + logx 3 ≤ 2 cos πy.

Problem 2:
Determine all the triples (a, b, c) of integers such that
a 3 + b 3 + c 3 = 2011.

Problem 3:
In decimal representation the number 22010 has m digits while
52010 has n digits. Find m + n.

Problem 4:
Find all the polynomials P(x) with real coefficients such that
sin P(x) = P(sin x), for all x ∈ R.

Problem 5:
The interior of an equilateral triangle of side length 1 is
covered by eight circles of the same radius r.
Prove that r ≥ 71– .

Problem 6:
Prove that in a convex hexagon of area S there exist three
consecutive vertices A, B and C such that
Area (ABC ) ≤ S
6

25
Year of Science
Science is everywhere and affects everything we do - from
driving our cars, to how we communicate and how we learn
in schools. Science can improve the quality of our lives at
many different levels - from our everyday activities to global
issues. You would have to live in an exceptionally remote
part of our world not to enjoy at least some offshoot, some
benefit of scientific inquiry. But does the average person
really think about how science affects our daily lives?

In British Columbia and across Canada, scientists and


researchers carry out - through chemistry, biology, physics,
mathematics and other fields - a wide range of work which
is continually leading the way to the development of many
new products and practical solutions that are vital to every
aspect of our daily lives.

Learn more at www.YearOfScienceBC.ca

26

You might also like