Está en la página 1de 182

MATHEMATICS CONTESTS

THE AUSTRALIAN SCENE 2015


PART 1 AND 2
A Di Pasquale, N Do and KL McAvaney
Le ar n f r om ye ste r d a y, l ive f o r to day, ho pe f o r to m o r r ow.
T h e i m p or ta nt th i ng is no t to sto p que stio ning.
Alber t Einstein

A u s t r a l i a n M a t h e ma t i c a l O l y m p i a d C omm i t t e e
A

department of the

A u s t r a l i a n M at h e mat i c s T r u s t

Published by
AMT Publishing

Australian Mathematics Trust


University of Canberra Locked Bag 1
Canberra GPO ACT 2601
Australia
Tel: 61 2 6201 5137
www.amt.edu.au

AMTT Limited
ACN 083 950 341

Copyright 2015 by the Australian Mathematics Trust


National Library of Australia Card Number
and ISSN 1323-6490

SUPPORT FOR THE AUSTRALIAN MATHEMATICAL


OLYMPIAD COMMITTEE TRAINING PROGRAM
The Australian Mathematical Olympiad Committee Training Program is an activity of the Australian Mathematical
Olympiad Committee, a department of the Australian Mathematics Trust.

Trustee
The University of Canberra

Sponsors
The Mathematics/Informatics Olympiads are supported by the Australian Government Department of Education
and Training through the Mathematics and Science Participation Program.
The Australian Mathematical Olympiad Committee (AMOC) also acknowledges the significant financial support
it has received from the Australian Government towards the training of our Olympiad candidates and the
participation of our team at the International Mathematical Olympiad (IMO).
The views expressed here are those of the authors and do not necessarily represent the views of the
government.

Special Thanks
With special thanks to the Australian Mathematical Society, the Australian Association of Mathematics Teachers
and all those schools, societies, families and friends who have contributed to the expense of sending the 2015
IMO team to Chiang Mai, Thailand.

ACKNOWLEDGEMENTS
The Australian Mathematical Olympiad Committee thanks sincerely all sponsors, teachers, mathematicians and
others who have contributed in one way or another to the continued success of its activities.
The editors thank sincerely those who have assisted in the compilation of this book, in particular the students
who have provided solutions to the 2015 IMO. Thanks also to members of AMOC and Challenge Problems
Committees, Adjunct Professor Mike Clapper, staff of the Australian Mathematics Trust and others who are
acknowledged elsewhere in the book.

PREFACE
After last year, there seemed little room for improvement, but 2015 has
been even better, marked particularly by our best ever result at an IMO,
where we were placed 6th out of the 104 competing countries, finishing
ahead of all European countries (including Russia) and many other
traditional powerhouses such as Singapore, Japan and Canada. For the
first time ever, all six team members obtained Silver or better, with two
team members (Alex Gunning and Seyoon Ragavan) claiming Gold.
Alex finished fourth in the world (after his equal first place last year, and
becomes Australias first triple Gold medallist in any academic Olympiad.
Seyoon, who finished 19th, now has three IMOs under his belt with a year
still to go. Once again, we had three Year 12 students in the team, so there
will certainly be opportunities for new team members next year. It was also
pleasing to, once again, see an Australian-authored question on the paper,
this year, the prestigious Question 6, which was devised by Ivan Guo and
Ross Atkins, based on the mathematics of juggling.
In the Mathematics Ashes we tied with the British team; however, we
finished comfortably ahead of them in the IMO competition proper. Director of Training and IMO Team Leader,
Dr Angelo Di Pasquale, along with his Deputy Andrew Elvey Price and their team of former Olympians continue
to innovate and keep the training alive, fresh and, above all, of high quality. The policy of tackling very hard
questions in training was daunting for team members at times but seems to have paid off.
The Mathematics Challenge for Young Australians (MCYA) also continues to attract strong entries, with the
Challenge continuing to grow, helped by the gathering momentum of the new Middle Primary Division, which
began in 2014. The Enrichment stage, containing course work, allows students to broaden their knowledge
base in the areas of mathematics associated with the Olympiad programs and more advanced problem-solving
techniques. We have continued running workshops for teachers to develop confidence in managing these
programs for their more able studentsthis seems to be paying off with strong numbers in both the Challenge
and Enrichment stages. The final stage of the MCYA program is the Australian Intermediate Mathematics
Olympiad (AIMO). It is a delight to record that over the last two years the number of entries to AIMO has
doubled. This has been partly due to wider promotion of the competition, but more specifically a result of
the policy of offering free entry to AMC prize winners. There were some concerns in 2014 that some of the
new contestants were under-prepared for AIMO and there were more zero scores than we would have liked.
However, this year, the number of zero scores was less than 1% and the quality of papers was much higher,
revealing some significant new talent, some of whom will be rewarded with an invitation to the December School
of Excellence.
There are many people who contribute to the success of the AMOC program. These include the Director of
Training and the ex-Olympians who train the students at camps; the AMOC State Directors; and the Challenge
Director, Dr Kevin McAvaney, and the various members of his Problems Committee, who develop such original
problems, solutions and discussions each year. The AMOC Senior Problems Committee is also a major
contributor and Norman Do is continuing with his good work. The invitational program saw some outstanding
results from Australian students, with a number of perfect scores. Details of these achievements are provided in
the appropriate section of this book. As was the case last year, we are producing Mathematics ContestsThe
Australian Scene in electronic form only and making it freely available through the website. We hope this will
provide greater access to the problems and section reports. This book is also available in two sections, one
containing the MCYA reports and papers and the other containing the Olympiad training program reports and
papers.
Mike Clapper
November 2015

CONTENTS
Support for the Australian Mathematical Olympiad Committee Training Program

Acknowledgements 4
Preface 5
Background Notes on the IMO and AMOC

Mathematics Challenge for Young Australians

10

Membership of MCYA Committees

12

Membership of AMOC Committees

14

AMOC Timetable for Selection of the Team to the 2016 IMO 15


Activities of AMOC Senior Problems Committee

16

Challenge Problems

17

Challenge Solutions

26

Challenge Statistics

51

Australian Intermediate Mathematics Olympiad

55

Australian Intermediate Mathematics Olympiad Solutions

57

Australian Intermediate Mathematics Olympiad Statistics

65

Australian Intermediate Mathematics Olympiad Results

66

AMOC Senior Contest

70

AMOC Senior Contest Solutions

71

AMOC Senior Contest Results

79

AMOC Senior Contest Statistics

80

AMOC School of Excellence

81

Australian Mathematical Olympiad

83

Australian Mathematical Olympiad Solutions

85

Australian Mathematical Olympiad Statistics

110

Australian Mathematical Olympiad Results

111

27th Asian Pacific Mathematics Olympiad

113

27th Asian Pacific Mathematics Olympiad Solutions

114

27th Asian Pacific Mathematics Olympiad Results

128

AMOC Selection School

130

IMO Team Preparation School

133

The Mathematics Ashes

134

The Mathematics Ashes Results

135

IMO Team Leaders Report

136

International Mathematical Olympiad

140

International Mathematical Olympiad Solutions

142

International Mathematical Olympiad Results

168

Origin of Some Questions

173

Honour Roll

174

BACKGROUND NOTES ON THE IMO AND AMOC


The Australian Mathematical Olympiad Committee
In 1980, a group of distinguished mathematicians formed the Australian Mathematical Olympiad Committee
(AMOC) to coordinate an Australian entry in the International Mathematical Olympiad (IMO).
Since then, AMOC has developed a comprehensive program to enable all students (not only the few who aspire
to national selection) to enrich and extend their knowledge of mathematics. The activities in this program are not
designed to accelerate students. Rather, the aim is to enable students to broaden their mathematical experience
and knowledge.
The largest of these activities is the MCYA Challenge, a problem-solving event held in second term, in which
thousands of young Australians explore carefully developed mathematical problems. Students who wish to
continue to extend their mathematical experience can then participate in the MCYA Enrichment Stage and pursue
further activities leading to the Australian Mathematical Olympiad and international events.
Originally AMOC was a subcommittee of the Australian Academy of Science. In 1992 it collaborated with the
Australian Mathematics Foundation (which organises the Australian Mathematics Competition) to form the
Australian Mathematics Trust. The Trust, a not-for-profit organisation under the trusteeship of the University
of Canberra, is governed by a Board which includes representatives from the Australian Academy of Science,
Australian Association of Mathematics Teachers and the Australian Mathematical Society.
The aims of AMOC include:
(1) giving leadership in developing sound mathematics programs in Australian schools
(2) identifying, challenging and motivating highly gifted young Australian school students in mathematics
(3) training and sending Australian teams to future International Mathematical Olympiads.

AMOC schedule from August until July for potential IMO team members
Each year hundreds of gifted young Australian school students are identified using the results from the
Australian Mathematics Competition sponsored by the Commonwealth Bank, the Mathematics Challenge for
Young Australians program and other smaller mathematics competitions. A network of dedicated mathematicians
and teachers has been organised to give these students support during the year either by correspondence sets
of problems and their solutions or by special teaching sessions.
It is these students who sit the Australian Intermediate Mathematics Olympiad, or who are invited to sit the
AMOC Senior Contest each August. Most states run extension or correspondence programs for talented
students who are invited to participate in the relevant programs. The 25 outstanding students in recent AMOC
programs and other mathematical competitions are identified and invited to attend the residential AMOC School
of Excellence held in December.
In February approximately 100 students are invited to attempt the Australian Mathematical Olympiad. The
best 20 or so of these students are then invited to represent Australia in the correspondence Asian Pacific
Mathematics Olympiad in March. About 12 students are selected for the AMOC Selection School in April and
about 13 younger students are also invited to this residential school. Here, the Australian team of six students
plus one reserve for the International Mathematical Olympiad, held in July each year, is selected. A personalised
support system for the Australian team operates during May and June.
It should be appreciated that the AMOC program is not meant to develop only future mathematicians.
Experience has shown that many talented students of mathematics choose careers in engineering, computing,
and the physical and life sciences, while others will study law or go into the business world. It is hoped that the
AMOC Mathematics Problem-Solving Program will help the students to think logically, creatively, deeply and with
dedication and perseverance; that it will prepare these talented students to be future leaders of Australia.

The International Mathematical Olympiad


The IMO is the pinnacle of excellence and achievement for school students of mathematics throughout the
world. The concept of national mathematics competitions started with the Etvos Competition in Hungary during
1894. This idea was later extended to an international mathematics competition in 1959 when the first IMO was
8

held in Romania. The aims of the IMO include:


(1) discovering, encouraging and challenging mathematically gifted school students
(2) fostering friendly international relations between students and their teachers
(3) sharing information on educational syllabi and practice throughout the world.
It was not until the mid-sixties that countries from the western world competed at the IMO. The United States of
America first entered in 1975. Australia has entered teams since 1981.
Students must be under 20 years of age at the time of the IMO and have not enrolled at a tertiary institution.
The Olympiad contest consists of two four-and-a-half hour papers, each with three questions.
Australia has achieved varying successes as the following summary of results indicate. HM (Honorable Mention)
is awarded for obtaining full marks in at least one question.
The IMO will be held in Hong Kong in 2016.

SUMMARY OF AUSTRALIAS ACHIEVEMENTS AT PREVIOUS IMOS


Year

City

Gold

Silver

Bronze

HM

Rank

1981

Washington

23 out of 27 teams

1982

Budapest

21 out of 30 teams

1983

Paris

19 out of 32 teams

1984

Prague

15 out of 34 teams

1985

Helsinki

11 out of 38 teams

1986

Warsaw

15 out of 37 teams

1987

Havana

1988

Canberra

1989

Braunschweig

22 out of 50 teams

1990

Beijing

15 out of 54 teams

1991

Sigtuna

1992

Moscow

1993

Istanbul

1994

3
1

15 out of 42 teams
1

17 out of 49 teams

20 out of 56 teams

19 out of 56 teams

Hong Kong

12 out of 69 teams

1995

Toronto

21 out of 73 teams

1996

Mumbai

23 out of 75 teams

1997

Mar del Plata

9 out of 82 teams

1998

Taipei

13 out of 76 teams

1999

Bucharest

2000

Taejon

16 out of 82 teams

2001

Washington D.C.

25 out of 83 teams

2002

Glasgow

2003

Tokyo

2004

Athens

2005

Merida

2006

Ljubljana

26 out of 90 teams

2007

Hanoi

22 out of 93 teams

2008

Madrid

2009

Bremen

23 out of 104 teams

2010

Astana

15 out of 96 teams

2011

Amsterdam

25 out of 101 teams

2012

Mar del Plata

27 out of 100 teams

2013

Santa Marta

15 out of 97 teams

2014

Cape Town

1
Perfect Score
by Alexander
Gunning

11 out of 101 teams

2015

Chiang Mai

13 out of 73 teams

15 out of 81 teams

26 out of 84 teams

26 out of 82 teams

27 out of 85 teams

10

25 out of 91 teams

19 out of 97 teams

6 out of 104 teams

MATHEMATICS CHALLENGE FOR YOUNG AUSTRALIANS


The Mathematics Challenge for Young Australians (MCYA) started on a national scale in 1992. It was set up to
cater for the needs of the top 10 percent of secondary students in Years 710, especially in country schools and
schools where the number of students may be quite small. Teachers with a handful of talented students spread
over a number of classes and working in isolation can find it very difficult to cater for the needs of these students.
The MCYA provides materials and an organised structure designed to enable teachers to help talented students
reach their potential. At the same time, teachers in larger schools, where there are more of these students, are
able to use the materials to better assist the students in their care.
The aims of the Mathematics Challenge for Young Australians include:
encouraging and fostering
a greater interest in and awareness of the power of mathematics
a desire to succeed in solving interesting mathematical problems
the discovery of the joy of solving problems in mathematics
identifying talented young Australians, recognising their achievements nationally and providing support that will
enable them to reach their own levels of excellence
providing teachers with
interesting and accessible problems and solutions as well as detailed and motivating teaching discussion and
extension materials
comprehensive Australia-wide statistics of students achievements in the Challenge.
There are three independent stages in the Mathematics Challenge for Young Australians:
Challenge (three weeks during the period MarchJune)
Enrichment (AprilSeptember)
Australian Intermediate Mathematics Olympiad (September).

Challenge
Challenge now consists of four levels. Upper Primary (Years 56) and Middle Primary (Years 34) present
students with four problems each to be attempted over three weeks, students being allowed to work on the
problems in groups of up to three participants, but each to write their solutions individually. The Junior (Years
78) and Intermediate (Years 910) levels present students with six problems to be attempted over three weeks,
students being allowed to work on the problems with a partner but each must write their solutions individually.
There were 12692 entries (1166 Middle Primary, 3416 Upper Primary, 5006 Junior, 3104 Intermediate) for the
Challenge in 2015. The 2015 problems and solutions for the Challenge, together with some statistics, appear later
in this book.

Enrichment
This is a six-month program running from April to September, which consists of six different parallel stages of
comprehensive student and teacher support notes. Each student participates in only one of these stages.
The materials for all stages are designed to be a systematic structured course over a flexible 1214 week period
between April and September. This enables schools to timetable the program at convenient times during their
school year.
Enrichment is completely independent of the earlier Challenge; however, they have the common feature of
providing challenging mathematics problems for students, as well as accessible support materials for teachers.
Newton (years 56) includes polyominoes, fast arithmetic, polyhedra, pre-algebra concepts, patterns, divisibility
and specific problem-solving techniques. There were 1165 entries in 2015.
Dirichlet (years 67) includes mathematics concerned with tessellations, arithmetic in other bases, time/distance/
speed, patterns, recurring decimals and specific problem-solving techniques. There were 1181 entries in 2015.
Euler (years 78) includes primes and composites, least common multiples, highest common factors, arithmetic

12

sequences, figurate numbers, congruence, properties of angles and pigeonhole principle. There were 1708 entries
in 2015.
Gauss (years 89) includes parallels, similarity, Pythagoras Theorem, using spreadsheets, Diophantine
equations, counting techniques and congruence. Gauss builds on the Euler program. There were 1150 entries
in 2015.
Noether (top 10% years 910) includes expansion and factorisation, inequalities, sequences and series, number
bases, methods of proof, congruence, circles and tangents. There were 818 entries in 2015.
Polya (top 10% year 10) (currently under revision) Topics will include angle chasing, combinatorics, number
theory, graph theory and symmetric polynomials. There were 303 entries in 2015.

Australian Intermediate Mathematics Olympiad


This four-hour competition for students up to Year 10 offers a range of challenging and interesting questions.
It is suitable for students who have performed well in the AMC (Distinction and above), and is designed as an
endpoint for students who have completed the Gauss or Noether stage. There were 1440 entries for 2015
and 19 perfect scores.

12

MEMBERSHIP OF MCYA COMMITTEES


Mathematics Challenge for Young Australians Committee 2015
Director
Dr K McAvaney, Victoria
Challenge
Committee
Adj Prof M Clapper, Australian Mathematics Trust, ACT
Mrs B Denney, NSW
Mr A Edwards, Queensland Studies Authority
Mr B Henry, Victoria
Ms J McIntosh, AMSI, VIC
Mrs L Mottershead, New South Wales
Ms A Nakos, Temple Christian College, SA
Prof M Newman, Australian National University, ACT
Dr I Roberts, Northern Territory
Ms T Shaw, SCEGGS, NSW
Ms K Sims, New South Wales
Dr A Storozhev, Attorney Generals Department, ACT
Mr S Thornton, South Australia
Ms G Vardaro, Wesley College, VIC
Moderators
Mr W Akhurst, New South Wales
Mr R Blackman, Victoria
Ms J Breidahl, Victoria
Mr A Canning, Queensland
Dr E Casling, Australian Capital Territory
Mr B Darcy, South Australia
Mr J Dowsey, Victoria
Ms P Graham, MacKillop College, TAS
Ms J Hartnett, Queensland
Ms N Hill, Victoria
Dr N Hoffman, Edith Cowan University, WA
Ms R Jorgenson, Australian Capital Territory
Prof H Lausch, Victoria
Mr J Lawson, St Pius X School, NSW
Ms K McAsey, Victoria
Ms T McNamara, Victoria
Mr G Meiklejohn, Department of Education, QLD
Mr M OConnor, AMSI, VIC
Mr J Oliver, Northern Territory
Mr G Pointer, Marratville High School, SA
Dr H Sims, Victoria
Mrs M Spandler, New South Wales
Ms C Stanley, Queensland
Mr P Swain, Ivanhoe Girls Grammar School, VIC
Dr P Swedosh, The King David School, VIC
Mrs A Thomas, New South Wales
Ms K Trudgian, Queensland

14

Enrichment
Editors
Mr G R Ball, University of Sydney, NSW
Dr M Evans, International Centre of Excellence for Education in Mathematics, VIC
Mr K Hamann, South Australia
Mr B Henry, Victoria
Dr K McAvaney, Victoria
Dr A M Storozhev, Attorney Generals Department, ACT
Emeritus Prof P Taylor, Australian Capital Territory
Dr O Yevdokimov, University of Southern Queensland
Australian Intermediate Mathematics Olympiad Problems Committee
Dr K McAvaney, Victoria (Chair)
Adj Prof M Clapper, Australian Mathematics Trust, ACT
Mr J Dowsey, University of Melbourne, VIC
Dr M Evans, International Centre of Excellence for Education in Mathematics, VIC
Mr B Henry, Victoria
Assoc Prof H Lausch, Monash University, VIC

14

MEMBERSHIP OF AMOC COMMITTEES


Australian Mathematical Olympiad Committee 2015
Chair
Prof C Praeger, University of Western AustraliaDeputy Chair
Assoc Prof DHunt, University of New South Wales

Executive Director
Adj Prof Mike Clapper, Australian Mathematics Trust, ACT

Treasurer
Dr P Swedosh, The King David School, VIC

Chair, Senior Problems Committee


Dr N Do, Monash University, VIC

Chair, Challenge
Dr K McAvaney, Deakin University, VIC

Director of Training and IMO Team Leader


Dr A Di Pasquale, University of Melbourne, VIC

IMO Deputy Team Leader


Mr A Elvey Price, University of Melbourne, VIC

State Directors
Dr K Dharmadasa, University of Tasmania
Dr G Gamble, University of Western Australia
Dr Ian Roberts, Northern Territory
Dr W Palmer, University of Sydney, NSW
Mr D Martin, South Australia
Dr V Scharaschkin, University of Queensland
Dr P Swedosh, The King David School, VIC
Dr Chris Wetherell, Radford College, ACT

Representatives
Ms A Nakos, Challenge Committee
Prof M Newman, Challenge Committee
Mr H Reeves, Challenge Committee

15

AMOC TIMETABLE FOR SELECTION OF THE TEAM TO THE


2016 IMO
August 2015July 2016
Hundreds of students are involved in the AMOC programs which begin on a state basis. The students are given
problem-solving experience and notes on various IMO topics not normally taught in schools.
The students proceed through various programs with the top 25 students, including potential team members and
other identified students, participating in a ten-day residential school in December.
The selection program culminates with the April Selection School during which the team is selected.
Team members then receive individual coaching by mentors prior to assembling for last minute training before
the IMO.

Month

Activity
Outstanding students are identified from AMC results, MCYA, other competitions
and recommendations; and eligible students from previous training programs

August

AMOC state organisers invite students to participate in AMOC programs


Various state-based programs
AMOC Senior Contest

September

Australian Intermediate Mathematics Olympiad

December

AMOC School of Excellence

January

Summer Correspondence Program for those who attended the School of


Excellence

February

Australian Mathematical Olympiad

March

Asian Pacific Mathematics Olympiad

April

AMOC Selection School

MayJune

Personal Tutor Scheme for IMO team members

July

Short mathematics school for IMO team members


2016 IMO in Hong Kong.

16

ACTIVITIES OF AMOC SENIOR PROBLEMS COMMITTEE


This committee has been in existence for many years and carries out a number of roles. A central role is the
collection and moderation of problems for senior and exceptionally gifted intermediate and junior secondary school
students. Each year the Problems Committee provides examination papers for the AMOC Senior Contest and the
Australian Mathematical Olympiad. In addition, problems are submitted for consideration to the Problem Selection
Committees of the annual Asian Pacific Mathematics Olympiad and the International Mathematical Olympiad.

AMOC Senior Problems Committee October 2014September 2015


Dr A Di Pasquale, University of Melbourne, VIC
Dr N Do, Monash University, VIC, (Chair)
Dr M Evans, Australian Mathematical Sciences Institute, VIC
Dr I Guo, University of Sydney, NSW
Assoc Prof D Hunt, University of NSW
Dr J Kupka, Monash University, VIC
Assoc Prof H Lausch, Monash University, VIC
Dr K McAvaney, Deakin University, VIC
Dr D Mathews, Monash University, VIC
Dr A Offer, Queensland
Dr C Rao, NEC Australia, VIC
Dr B B Saad, Monash University, VIC
Assoc Prof J Simpson, Curtin University of Technology, WA
Emeritus Professor P J Taylor, Australian Capital Territory
Dr I Wanless, Monash University, VIC

1. 2015 Australian Mathematical Olympiad


The Australian Mathematical Olympiad (AMO) consists of two papers of four questions each and was sat on
10 and 11 February. There were 106 participants including 11 from New Zealand, seven more participants than
2014. Two students, Alexander Gunning and Seyoon Ragavan, achieved perfect scores and nine other students
were awarded Gold certificates,15 students were awarded Silver certificates and 26 students were awarded
Bronze certificates.

2. 2015 Asian Pacific Mathematics Olympiad


On Tuesday 10 March students from nations around the Asia-Pacific region were invited to write the Asian
Pacific Mathematics Olympiad (APMO). Of the top ten Australian students who participated, there were 1 Gold,
2 Silver, 4 Bronze and 3 HM certificates awarded. Australia finished in 9th place overall.

3. 2015 International Mathematical Olympiad, Chiang Mai, Thailand.


The IMO consists of two papers of three questions worth seven points each. They were attempted by teams of
six students from 104 countries on 8 and 9 July in Cape Town, South Africa. Australia was placed 6th of 104
countries, its most successful result since it began participating. The medals for Australia were two Gold and
four Silver.

4. 2015 AMOC Senior Contest


Held on Tuesday 11 August, the Senior Contest was sat by 84 students (compared to 81 in 2014). There
were three students who obtained perfect scores and two other students who were also Prize winners. Three
studentent obtained High Distinctions and 14 students obtained Distinctions.

17

CHALLENGE
PROBLEMS
PRIMARY
2015 Challenge
Problems- MIDDLE
Middle Primary
Students may work on each of these four problems in groups of up to three, but must write their solutions individually.

MP1 e-Numbers
The digits on many electronic devices look like this:

When numbers made of these digits are rotated 180 (


), some of them become numbers and some dont. For
example, when 8 is rotated it remains the same, 125 becomes 521, and 14 does not become a number. Except for 0,
no number starts with 0.

All rotations referred to below are through 180.


a List all digits which rotate to the same digit.
b List all digits which rotate to a different digit.
c List all numbers from 10 to 50 whose rotations are numbers.
d List all numbers from 50 to 200 that stay the same when rotated.

MP2 Quazy Quilts


Joy is making a rectangular patchwork quilt from square pieces of fabric. All the square pieces are the same size. She
starts by joining two black squares into a rectangle. Then she adds a border of grey squares.

Next, she adds a border of white squares.


a How many white squares does she need?
Joy continues to add borders of grey and borders of white squares alternately.
b How many grey squares and how many white squares does she need for the next two borders?
c Joy notices a pattern in the number of squares used for each border. Describe this pattern and explain why it
always works.

19

d There are 90 squares in the outside border of the completed quilt. How many borders have been added to Joys
starting black rectangle?

MP3 Egg Cartons


Zoe decided to investigate the number of different ways of placing identical eggs in clear rectangular cartons of various
sizes.

For example, there is only one way of placing six eggs in a 2 3 carton and only two different ways of placing five
eggs.

Reflections and rotations of arrangements are not considered different. For example, all of these arrangements are the
same:

a Here are two ways of arranging four eggs in a 2 3 carton.

Draw four more different ways.


b Draw six different ways of arranging two eggs in a 2 3 carton.
c Draw six different ways of arranging three eggs in a 2 3 carton.
d Draw all the different ways of arranging six eggs in a 2 4 carton.

MP4 Condates
On many forms, the date has to be written as DD/MM/YY. For example, 25 April 2013 is written 25/04/13. Dates
such as this that use all the digits 0, 1, 2, 3, 4, 5 will be called condates.
a Find two condates in 2015.
b Find the first condate after 2015.
c Find the last condate in any century.
d Explain why no date of the form DD/MM/YY can use all the digits 1, 2, 3, 4, 5, 6.

19

CHALLENGE
PROBLEMS
PRIMARY
2015 Challenge
Problems- UPPER
Upper Primary
Students may work on each of these four problems in groups of up to three, but must write their solutions individually.

UP1 Rod Shapes


I have a large number of thin straight steel rods of lengths 1 cm, 2 cm, 3 cm and 4 cm. I can place the rods with their
ends touching to form polygons. For example, with four rods, two of length 1 cm and two of length 3 cm, I can form
a parallelogram which I refer to as (1, 3, 1, 3).

1c
m

1c
m

3 cm

3 cm
a List all the different equilateral triangles I can make using only three rods at a time.
b Using only three rods at a time, how many different isosceles triangles can I make which are not equilateral triangles?
List them all.
c List all the different scalene triangles I can make using only three rods at a time.
d Using only four rods at a time, none of which is 4 cm, how many different quadrilaterals can I make which have
exactly one pair of parallel sides and the other pair of sides equal in length? List them all.

UP2 Egg Cartons


Zoe decided to investigate the number of different ways of placing identical eggs in clear rectangular cartons of various
sizes.

For example, there is only one way of placing six eggs in a 2 3 carton and only two different ways of placing five
eggs.

Reflections and rotations of arrangements are not considered different. For example, all of these arrangements are the
same:

21

a Draw four different ways of arranging two eggs in a 2 3 carton.


b Draw six different ways of arranging three eggs in a 2 3 carton.
c Draw all the different ways of arranging six eggs in a 2 4 carton.
d Zoe discovers that there are 55 ways of arranging three eggs in a 2 6 carton. Explain why there must also be 55
ways of arranging nine eggs in a 2 6 carton.

UP3 Seahorse Swimmers


The coach at the Seahorse Swimming Club has four boys in his Under 10 squad. Their personal best (PB) times for
50 metres are:
Mack
Jack
Zac
Nick

55 seconds
1 minute 8 seconds
1 minute 16 seconds
1 minute 3 seconds

a The coach splits the swimmers into two pairs for a practice relay race. He wants the race to be as close as possible.
Based on the PB times, who should be in each pair?
b Chloes PB time is 1 minute 10 seconds and Sallys is 53 seconds. The coach arranges the two girls and four boys
into two teams of three with total PB times as close as possible.
i Why cant the total PB times be the same for the two teams?
ii Who are in each team?
c The swimmers ask the coach what his PB time was when he was 10 years old. The coach replied that if his PB
time was included with all theirs, then the average PB time would be exactly 1 minute. What was the coachs PB
time?

UP4 Primelandia Money


In Primelandia, the unit of currency is the Tao (T). The value of each Primelandian coin is a prime number of Taos.
So the coin with the smallest value is worth 2T. There are coins of every prime value less than 50. All payments in
Primelandia are an exact number of Taos.
a List all combinations of two coins whose sum is 50T.
b What is the smallest payment (without change) which requires at least three coins?
c Suppose you have one of each Primelandian coin with value less than 50T. What is the difference between the
largest even payment that can be made using four coins and the largest even payment that can be made using three
coins?
d A bag contains six different coins. Alice, Bob and Carol take two coins each from the bag and keep them. They
find that they have all taken the same amount of money. What is the smallest amount of money that could have
been in the bag? Explain your reasoning.

21

CHALLENGE
PROBLEMS
JUNIOR
2015 Challenge
Problems -Junior
Students may work on each of these six problems with a partner but each must write their solutions individually.

J1 Quirky Quadrilaterals
Robin and Toni were drawing quadrilaterals on square dot paper with their vertices on the dots but no other dots on
the edges. They decided to use I to represent the number of dots in the interior. Here are two examples:

I=1

I=6

a Toni said she could draw a rhombus with I = 2 and a kite that is not a rhombus with I = 2. Draw such quadrilaterals
on square dot paper.
b Toni drew squares with I = 4 and I = 9. Draw such squares on square dot paper.
c Robin drew a square with I = 12. Draw such a square on square dot paper.
d Toni exclaimed with excitement, I can draw rhombuses with any number of interior points. Explain how this could
be done.

J2 Indim Integers
Jim is a contestant on a TV game show called Indim. The compere spins a wheel that is divided into nine sectors
numbered 1 to 9. Jim has nine tiles numbered 1 to 9. When the wheel stops, he removes all tiles whose numbers are
factors or multiples of the spun number.

4
5

Jim then tries to use three of the remaining tiles to form a 3-digit number that is divisible by the number on the
wheel. If he can make such a number, he shouts Indim and wins a prize.
a Show that if 1, 2, or 5 is spun, then it is impossible for Jim to win.
b How many winning numbers can Jim make if 4 is spun?
c List all the winning numbers if 6 is spun.
d What is the largest possible winning number overall?

23

J3 Primelandia Money
In Primelandia, the unit of currency is the Tao (T). The value of each Primelandian coin is a prime number of Taos.
So the coin with the smallest value is worth 2T. There are coins of every prime value less than 50. All payments in
Primelandia are an exact number of Taos.
a What is the smallest payment (without change) which requires at least three coins?
b A bag contains six different coins. Alice, Bob and Carol take two coins each from the bag and keep them. They
find that they have all taken the same amount of money. What is the smallest amount of money that could have
been in the bag? Explain your reasoning.
c Find five Primelandian coins which, when placed in ascending order, form a sequence with equal gaps of 6T between
their values.
d The new King of Primelandia decides to mint coins of prime values greater than 50T. Show that no matter what
coins are made, there is only one set of five Primelandian coins which, when placed in ascending order, form a
sequence with equal gaps of 6T between their values.

J4 Condates
On many forms, the date has to be written as DD/MM/YYYY. For example, 25 April 1736 is written 25/04/1736.
Dates such as this that use eight consecutive digits (not necessarily in order) will be called condates.
a What is the first condate after the year 2015?
b Why must there be a 0 in every condate in the years 2000 to 2999?
c Why must every condate in the years 2000 to 2999 have 0 as the first digit of the month?
d How many condates are there in the years 2000 to 2999?

J5 Jogging
Theo is training for the annual Mt Killaman Joggin 10 km Torture Track. This fun run involves running east on a
flat track for 2 km to the base of Mt Killaman Joggin, then straight up its rather steep side a further 3 km to the top.
This is the halfway point where you turn around and run back the way you came. This simplified mudmap should
give you the general idea.
Top of Mt KJ

3k

Start/Finish

2 km

Base of Mt KJ

Theo can run at 12 km/h on the flat. Up the hill he reckons he can average 9 km/h and he is good for 15 km/h on the
downhill part of the course.
a Calculate Theos expected time to complete the course, assuming his estimates for his running speeds are correct.
The day before the race, Theo hears that windy conditions are expected. He knows from experience that this will
affect his speed out on the course. When the wind blows into his face, it will slow him down by 1 km/h, but when it is
at his back, he will speed up by the same amount. This applies to his speeds on the flat and on the slope. The wind
will either blow directly from the east or directly from the west.
b From which direction should he hope the wind blows if he wishes to minimise his time? Justify your answer.
c Theo would like to finish the race in 50 minutes. He decides to run either the uphill leg faster or the downhill leg
faster, but not both. Assuming there is no wind, how much faster would he have to run in each case?

23

J6 Tessellating Hexagons
Will read in his favourite maths book that this hexagon tessellates the plane.
D
E
C

To see if this was true, Will made 16 copies of the hexagon and glued them edge-to-edge onto a piece of paper, with
no overlaps and no gaps.
a Cut out 16 hexagons from the worksheet and glue them as Will might have done so they entirely cover the dashed
rectangle.
b Show that your block of 16 hexagons can be translated indefinitely to tessellate the entire plane.
c Find all points on the hexagon above that are points of symmetry of your tessellation.
d In your tessellation, select four hexagons that form one connected block which will tessellate the plane by translation
only. Indicate three such blocks on your tessellation that are not identical.

25

CHALLENGE
PROBLEMS
2015 Challenge
Problems- INTERMEDIATE
Intermediate
Students may work on each of these six problems with a partner but each must write their solutions individually.

I1 Indim Integers
See Junior Problem 2.

I2 Digital Sums
The digital sum of an integer is the sum of all its digits. For example, the digital sum of 259 is 2 + 5 + 9 = 16.
a Find the largest even and largest odd 3-digit multiples of 7 which have a digital sum that is also a multiple of 7.
A digital sum sequence is a sequence of numbers that starts with any integer and has each number after the first
equal to the digital sum of the number before it. For example: 7598, 29, 11, 2. Any digital sum sequence ends in a
single-digit number and this is called the final digital sum or FDS of the first number in the sequence. Thus the FDS
of 7598 is 2.
b Find the three largest 3-digit multiples of 7 which have an FDS of 7.
c Find and justify a rule that produces all 3-digit multiples of 7 which have an FDS of 7.
d Find and justify a rule that produces all 3-digit multiples of 8 which have an FDS of 8.

I3 Coin Flips
I have several identical coins placed on a table. I play a game consisting of one or more moves. Each move consists of
flipping over some of the coins. The number of coins that are flipped stays the same for each game but may change
from game to game. A coin showing heads is represented by H. A coin showing tails is represented by T . So, when
a coin is flipped it changes from H to T or from T to H. The same coin may be selected on different moves. In each
game we start with all coins showing tails.
For example, in one game we might start with five coins and flip two at a time like this:
Start:
Move 1:
Move 2:
Move 3:

T
T
H
H

T
H
H
H

T
H
T
T

T
T
T
H

T
T
T
H

a Starting with 14 coins and flipping over four coins on each move, explain how to finish with exactly 10 coins heads
up in three moves.
b Starting with 154 coins and flipping over 52 coins on each move, explain how to finish with all coins heads up in
three moves.
c Starting with 26 coins and flipping over four coins on each move, what is the minimum number of moves needed to
finish with all coins heads up?
d Starting with 154 coins and flipping over a fixed odd number of coins on each move, explain why I cannot have all
154 coins heads up at the end of three moves.

I4 Jogging
See Junior Problem 5.

25

I5 Folding Fractions
A square piece of paper has side length 1 and is shaded on the front and white on the back. The bottom-right corner
is folded to a point P on the top edge as shown, creating triangles P QR, P T S and SU V .
T

S
U
V

a If P is the midpoint of the top edge of the square, find the length of QR.
b If P is the midpoint of the top edge of the square, find the side lengths of triangle SU V .
c Find all positions of P so that the ratio of the sides of triangle SU V is 7:24:25.

I6 Crumbling Cubes
A large cube is made of 1 1 1 small cubes. Small cubes are removed in a sequence of steps.
The first step consists of marking all small cubes that have at least 2 visible faces and then removing only those small
cubes.
In the second and subsequent steps, the same procedure is applied to what remains of the original large cube after the
previous step.
a Starting with a 10 10 10 cube, how many small cubes are removed at the first step?
b A different cube loses 200 small cubes at the first step. What are the dimensions of this cube?
c Beginning with a 9 9 9 cube, what is the surface area of the object that remains after the first step?
d Beginning with a 9 9 9 cube, how many small cubes are left after the third step?

27

CHALLENGE
SOLUTIONS
MIDDLE
PRIMARY
2015 Challenge
Solutions - Middle
Primary
MP1 e-Numbers
a The digits which rotate to the same digit are: 0, 1, 2, 5, 8.
b The digits which rotate to a different digit are 6 and 9.
c Only the digits in Parts a and b can be used to form numbers whose rotations are numbers. So the only numbers
from 10 to 50 whose rotations are numbers are: 11, 12, 15, 16, 18, 19, 21, 22, 25, 26, 28, 29.
d From Parts a and b, if a number and its rotation are the same, then the first and last digits of the number must
both be 1, 2, 5, 8, or the first digit is 6 and the last 9 or the first 9 and the last 6. So the numbers from 50 to 200
that stay the same when rotated are: 55, 69, 88, 96, 101, 111, 121, 151, 181.

MP2 Quazy Quilts


a The quilt after adding one grey and one white border:

The number of white squares is 18.


b The quilt after adding another grey and another white border:

There are 26 extra grey squares and 34 extra white squares.


c The table shows the number of squares in each border that we have counted so far.
Border
Squares

1
10

2
18

3
26

4
34

The number of squares from one border to the next appears to increase by 8. We now show this rule continues to
apply.
10

27

Except for the corner squares, each square on the previous border corresponds to a square in the new border. Each
of the corner squares C on the previous border corresponds to a corner square C  in the new border. In addition,
there are two new squares next to each corner.
new border

C

C

previous border

previous border

new border

new border

previous border

previous border

new border

C

C

Thus the number of squares from one border to the next increases by 4 2 = 8.
d Alternative i
The first border has 2 + 8 squares.
The second border has 2 + 8 + 8 squares.
The third border has 2 + 8 + 8 + 8 squares.
And so on.
We want 2 + 8 + 8 + = 90.
Hence the number of 8s needed is 11.
So Joys completed quilt has 11 borders.
Alternative ii
From Part c we have the following table.
Border
Squares

1
10

2
18

3
26

4
34

5
42

6
50

7
58

8
66

9
74

10
82

11
90

So Joys completed quilt has 11 borders.

MP3 Egg Cartons


a

For each of these arrangements, any rotation or reflection is acceptable.


b

For each of these arrangements, any rotation or reflection is acceptable.

11

29

For each of these arrangements, any rotation or reflection is acceptable.


d Each arrangement of six eggs in a 2 4 carton has two empty positions. By systematically locating the two empty
positions, we see that there are ten different ways to arrange six eggs in a 2 4 carton.

For each of these arrangements, any rotation or reflection is acceptable.

MP4 Condates
a Any condate in the year 2015 must be written as DD/MM/15. The remaining digits are 0, 2, 3, 4. So the month
must be 02, 03, or 04.
If the month is 02, then the day is 34 or 43, which are not allowed. If the month is 03, then the day must be 24. If
the month is 04, then the day must be 23. Thus the only condates in 2015 are 24/03/15 and 23/04/15.
b There are no condates in the years 2016, 2017, 2018, 2019 since their last digits are greater than 5.
In 2020, the remaining digits are 1, 3, 4, 5. No two of these form a month.
In 2021, the remaining digits are 0, 3, 4, 5. So the month must start with 0. Then the day must contain two of 3,
4, 5, which is impossible.
The year 2022 duplicates 2.
In 2023, the only digits remaining are 0, 1, 4, 5. The day cannot use both 4 and 5, so the month is 04 or 05. The
earliest of these is 04, in which case the day must be 15. So the first condate after 2015 is 15/04/23.
c The latest possible year ends with 54. The latest month in that year is 12. So the latest day is 30. Thus the last
condate in any century is 30/12/54.
d If DD/MM/YY contains all the digits 1, 2, 3, 4, 5, 6, then no digit can be repeated and no digit is 0. Hence the
month must be 12. So the day must contain two of the digits 3, 4, 5, 6. This is impossible.

12

29

CHALLENGE
SOLUTIONS
UPPER
PRIMARY
2015 Challenge
Solutions - Upper
Primary
UP1 Rod Shapes
a There are four equilateral triangles: (1, 1, 1), (2, 2, 2), (3, 3, 3), (4, 4, 4).
b To form an isosceles triangle that is not equilateral we need two rods of equal length and a third rod with a different
length. Thus we have only the following choices:
(1, 2, 2), (1, 3, 3), (1, 4, 4), (2, 1, 1), (2, 3, 3), (2, 4, 4), (3, 1, 1), (3, 2, 2), (3, 4, 4), (4, 1, 1), (4, 2, 2), (4, 3, 3).
To form a triangle, we must have each side smaller than the sum of the other two sides. This eliminates
(2, 1, 1), (3, 1, 1), (4, 1, 1), (4, 2, 2).
So there are only eight isosceles triangles that are not equilateral triangles:
(1, 2, 2), (1, 3, 3), (1, 4, 4), (2, 3, 3), (2, 4, 4), (3, 2, 2), (3, 4, 4), (4, 3, 3).
c A scalene triangle has three unequal sides. So there are four possibilities: (1, 2, 3), (1, 2, 4), (1, 3, 4), (2, 3, 4).
To form a triangle, we must have each side smaller than the sum of the other two sides. This eliminates
(1, 2, 3), (1, 2, 4), (1, 3, 4).
So there is only one scalene triangle: (2, 3, 4).
d If the pair of parallel sides have equal length, then the other pair of sides would be parallel. So the pair of parallel
sides must have unequal length.
The possible lengths for the parallel sides are (1, 2), (1, 3), (2, 3).
So we have only the following choices for the quadrilateral:
(1, 1, 2, 1), (1, 2, 2, 2), (1, 3, 2, 3), (1, 1, 3, 1), (1, 2, 3, 2), (1, 3, 3, 3), (2, 1, 3, 1), (2, 2, 3, 2), (2, 3, 3, 3).
To form a quadrilateral, we must have each side smaller than the sum of the other three sides.
This eliminates (1, 1, 3, 1). So there are just eight required quadrilaterals.

UP2 Egg Cartons


a Any four of the following arrangements.

For each of these arrangements, any rotation or reflection is acceptable.


b

For each of these arrangements, any rotation or reflection is acceptable.


c By systematically locating the two empty positions, we see that there are ten different ways to arrange six eggs in
a 2 4 carton.

13

31

For each of these arrangements, any rotation or reflection is acceptable.


d Each of Zoes arrangements in the 2 6 carton has 3 eggs and 9 empty places. Suppose the 3 eggs are white and fill
the empty places with 9 brown eggs. Then each placement of 3 white eggs corresponds to a placement of 9 brown
eggs. So the number of ways of arranging 3 eggs equals the number of ways of arranging 9 eggs.

UP3 Seahorse Swimmers


a To compare times, first convert each PB time to seconds.
Mack
Jack
Zac
Nick

55 s
1 min 8 s = 68 s
1 min 16 s = 76 s
1 min 3 s = 63 s

Alternative i
The table shows the total PB times, in seconds, for all pairs that include Mack and for the corresponding pairs that
exclude Mack.
Pair with M
Total PB
Other pair
Total PB

MJ
123
ZN
139

MZ
131
JN
131

MN
118
JZ
144

Thus the pairs that have the same total PB time are Mack with Zac and Jack with Nick.
Alternative ii
The total PB time for the four swimmers is 55 + 68 + 76 + 63 = 262 seconds. It might be possible that the total
PB time for each pair is 262/2 = 131 seconds. To get the units digit 1 in the total we must pair 55 with 76 and 68
with 63. Each pair then totals 131 as required. Thus Mack is paired with Zac and Jack is paired with Nick.

14

31

i Chloes PB time is 70 seconds and Sallys is 53 seconds. So the total of all six PB times is 262 + 70 + 53 = 385
seconds. Since 385 is odd and cannot be divided into two equal amounts, the two teams cannot have the same
total PB times.
ii Alternative i
The total of all six PB times is 262 + 70 + 53 = 385 seconds. As explained in Part i, two teams cannot have
the same total PB times. The next closest would be 1 second apart, in which case the times would be 192 s and
193 s. To see if this is possible, we need to select three PB times and add them to see if they total 192 or 193.
It is easier to just add their units digits to see if we get 2 or 3. The swimmers PB times in increasing order
of their units digits are: 70, 53, 63, 55, 76, 68. The only three that give 3 in the units digit of their total are
70, 55, 68, and these do in fact total 193 s. Thus one team has Chloe, Mack, and Jack with a total PB time of
70 + 55 + 68 = 193 s. The other team has Sally, Nick, and Zac with a total PB time of 53 + 63 + 76 = 192 s.
Alternative ii
The table shows the total PB times, in seconds, for all teams that include Mack and for the corresponding
teams that exclude Mack.
Team with M
MJZ
MJN
MJC
MJS
MZN
MZC
MZS
MNC
MNS
MCS

Total PB
199
186
193
176
194
201
184
188
171
178

Other team
NCS
ZCS
ZNS
ZNC
JCS
JNS
JNC
JZS
JZC
JZN

Total PB
186
199
192
209
191
184
201
197
214
207

Thus the two teams that have the closest total PB time are Mack, Jack, Chloe with 193 seconds and Zac, Nick,
Sally with 192 seconds.
c If the average of seven PB times is 60 seconds, then the total of their PB times is 7 60 = 420 seconds. From Part
b, the total PB time for the six swimmers is 385 seconds. So the coachs PB time was 420 385 = 35 seconds.

UP4 Primelandia Money


a From 50, we subtract primes from 50 down to 25 and check if the difference is prime:
50 47 = 3,
50 37 = 13,

50 43 = 7,
50 31 = 19,

50 41 = 9,
50 29 = 21.

Since 9 and 21 are not prime, the only pairs of primes that total 50 are: (47,3), (43,7), (37,13), (31,19).
b We have the following amounts that can be paid with one or two coins:
2=2
3=3
4= 2+2
5=5
6= 3+3

7=7
8= 3+5
9= 2+7
10 = 3 + 7
11 = 11

12 = 5 + 7
13 = 13
14 = 3 + 11
15 = 2 + 13
16 = 3 + 13

17 = 17
18 = 5 + 13
19 = 19
20 = 3 + 17
21 = 2 + 19

22 = 3 + 19
23 = 23
24 = 5 + 19
25 = 2 + 23
26 = 3 + 23

We now show that 27T requires at least three coins.


Alternative i
From 27, we subtract primes from 27 down to 14 and check if the difference is prime:
27 23 = 4, 27 19 = 8, 27 17 = 10.
None of the differences is prime. Hence 27T is the smallest payment which requires at least three coins.

15

33

Alternative ii
The sum of two odd primes is even. So, if 27 is the sum of two primes, then one of those primes is 2, the only even
prime. Now 27 = 2 + 25 and 25 is not prime, so 27 cannot be the sum of two primes. Hence 27T is the smallest
payment which requires at least three coins.
c The four largest primes less than 50 are 47, 43, 41, and 37. So the largest even amount that can be made from four
coins is 47 + 43 + 41 + 37 = 168. If the sum of three primes is even, then one of those primes must be 2. So the
largest even amount that can be made from three coins is 47 + 43 + 2 = 92. The difference is 168 92 = 76.
d The bag cant contain a 2T coin. If it did, then the sum of the pair of coins that includes 2T would be odd and the
sum of each other pair of coins would be even.
Since all coins are odd, the sum of all six coins is even. Since the three pairs of coins have the same sum, the sum
of all six coins is divisible by 3. The sum of all six coins is at least 3 + 5 + 7 + 11 + 13 + 17 = 56, which is not a
multiple of 6. The next multiple of 6 is 60. The following table lists all possibilities for each sum of six coins up to
72T.
Sum of six
60
66
72

Sum of pair
20
22
24

All possible pairs


3 + 17, 7 + 13
3 + 19, 5 + 17
5 + 19, 7 + 17, 11 + 13

Thus the smallest amount that could have been in the bag is 72T.

16

33

CHALLENGE
SOLUTIONS
JUNIOR
2015 Challenge
Solutions -Junior
J1 Quirky Quadrilaterals
a

Kite

Rhombus

Another kite

I=4

I=9

I = 12

d If I is an odd number we can surround a column of I dots with a rhombus whose vertices are immediately above,
below and either side of the middle of the column as the following examples show. Thus one diagonal of the
rhombus lies on the line through the column of I dots and the other diagonal lies on the line perpendicular to the
first diagonal and passing through the middle dot.

I=1

I=3

17

I=5

35

If I is an even number we can surround a south-west/north-east diagonal of I dots with a rhombus whose vertices
are immediately south-west, north-east, and either side of the middle of the diagonal as the following examples
show. Thus one diagonal of the rhombus lies on the line through the I dots and the other diagonal lies on the line
perpendicular to the first diagonal and bisecting the distance between the two middle dots.

I =2

I =4

I=6
These rhombuses are not unique.

J2 Indim Integers
a Since 1 is a factor of every integer, no tile remains if 1 is spun.
If 2 is spun, then all remaining tiles have an odd number. Hence any 3-digit number made from the remaining tiles
would be odd and not divisible by 2. So there is no winning number if 2 is spun.
If 5 is spun, then tiles 1 and 5 are removed. No number made from the remaining tiles is a multiple of 5 since it
cannot end in 0 or 5. So there is no winning number if 5 is spun.
b The only digits remaining after spinning 4 are 3, 5, 6, 7, 9. A number made from these digits is divisible by 4 if
and only if it ends with a 2-digit number that is divisible by 4. The only such 2-digit multiples of 4 that can be
made are 36, 56, 76, and 96.
Alternative i
So there are 12 winning numbers: 536, 736, 936, 356, 756, 956, 376, 576, 976, 396, 596, 796.
Alternative ii
From each of 36, 56, 76, 96, we form a 3-digit number by choosing one of the remaining three digits as its first digit.
So the number of winning numbers is 3 4 = 12.
c The only digits remaining after spinning 6 are 4, 5, 7, 8, 9. A 3-digit number is divisible by 6 if and only if its last
digit is even and the sum of all its digits is divisible by 3. So the only winning numbers if 6 is spun are: 594, 954,
894, 984, 498, 948, 798, 978.

18

35

d From Part a, there is no winning number if 1, 2, or 5 is spun.


Alternative i
If 3 is spun, then tile 9 is excluded. So any winning number from spinning 3 must be less than 900.
From Part b, the largest winning number from spinning 4 is 976.
From Part c, the largest winning number from spinning 6 is 984.
The only 3-digit multiples of 7 larger than 984 are 987 and 994. If 7 is spun, then 987 is excluded. The multiple
994 is excluded because 9 is repeated.
All 3-different-digit numbers larger than 984 start with 98. So there are no winning numbers larger than 984 if 8
or 9 is spun.
So the largest winning number overall is 984.
Alternative ii
The largest number with three different digits is 987. This cannot be a winning number from spinning 9, 8, or 7.
Since 987 is not divisible by 6 or 4, it cannot be a winning number from spinning 6 or 4. It is divisible by 3 but
cannot be a winning number from spinning 3 since digit 9 would have been eliminated. So 987 is not a winning
number.
The next largest number is 986. This cannot be a winning number from spinning 9, 8, or 6. Since 986 is not divisible
by 7, 4, or 3, it is not a winning number from spinning 7, 4, or 3. So 986 is not a winning number.
The next largest number is 985. This cannot be a winning number from spinning 9, 8, or 5. Since 985 is not divisible
by 7, 6, 4, or 3, it is not a winning number from spinning 7, 6, 4, or 3. So 985 is not a winning number.
The next largest number is 984. This is divisible by 6 and none of its digits is a factor or multiple of 6. So it is a
winning number.
Thus the largest winning number overall is 984.

J3 Primelandia Money
a We have the following amounts that can be paid with one or two coins:
2=2
3=3
4= 2+2
5=5
6= 3+3

7=7
8= 3+5
9= 2+7
10 = 3 + 7
11 = 11

12 = 5 + 7
13 = 13
14 = 3 + 11
15 = 2 + 13
16 = 3 + 13

17 = 17
18 = 5 + 13
19 = 19
20 = 3 + 17
21 = 2 + 19

22 = 3 + 19
23 = 23
24 = 5 + 19
25 = 2 + 23
26 = 3 + 23

We now show that 27T requires at least three coins.


Alternative i
From 27, we subtract primes from 27 down to 14 and check if the difference is prime:
27 23 = 4, 27 19 = 8, 27 17 = 10.
None of the differences is prime. Hence 27T is the smallest payment which requires at least three coins.
Alternative ii
The sum of two odd primes is even. So, if 27 is the sum of two primes, then one of those primes is 2, the only even
prime. Now 27 = 2 + 25 and 25 is not prime, so 27 cannot be the sum of two primes. Hence 27T is the smallest
payment which requires at least three coins.
b The bag cant contain a 2T coin. If it did, then the sum of the pair of coins that includes 2T would be odd and the
sum of each other pair of coins would be even.
Since all coins are odd, the sum of all six coins is even. Since the three pairs of coins have the same sum, the sum
of all six coins is divisible by 3. The sum of all six coins is at least 3 + 5 + 7 + 11 + 13 + 17 = 56, which is not a
multiple of 6. The next multiple of 6 is 60. The following table lists all possibilities for each sum of six coins up to
72T.
Sum of six Sum of pair
All possible pairs
60
20
3 + 17, 7 + 13
66
22
3 + 19, 5 + 17
72
24
5 + 19, 7 + 17, 11 + 13
Thus the smallest amount that could have been in the bag is 72T.

19

37

c Examining primes less than 50 for pairs that differ by 6 and starting with small primes, we quickly find the sequence
5, 11, 17, 23, 29.
d Alternative i
Suppose we have a sequence of five prime numbers in ascending order which are 6 apart. Since 6 is even and 2 is
the only even prime, all primes in the sequence must be odd. Hence they only end in 1, 3, 5, 7, or 9. The difference
between any two primes in the sequence is 6, 12, 18, or 24. So no two primes in the sequence can end in the same
digit. There are five primes in the sequence so all of 1, 3, 5, 7, 9 must appear as last digits. The only prime whose
last digit is 5 is 5 itself. So the sequence must be 5, 11, 17, 23, 29.
Alternative ii
Suppose we have a sequence of five prime numbers in ascending order which are 6 apart. Since 6 is even and 2 is
the only even prime, all primes in the sequence must be odd. Hence they only end in 1, 3, 5, 7, or 9. The only
prime ending in 5 is 5 itself. Therefore, if the first prime in the sequence ends in 5, the sequence is 5, 11, 17, 23, 29.
If the first prime ends in 1, then the next four primes end in 7, 3, 9, 5. If the first prime ends in 3, then the next
four primes end in 9, 5, 1, 7. If the first prime ends in 7, then the next four primes end in 3, 9, 5, 1. If the first
prime ends in 9, then the next four primes end in 5, 1, 7, 3. In each of these four cases, the prime ending in 5 is at
least 1 + 6 = 7, 3 + 6 = 9, 7 + 6 = 13, and 9 + 6 = 15 respectively. This is impossible.
So the only sequence is 5, 11, 17, 23, 29.
Alternative iii
Suppose we have a sequence of five prime numbers in ascending order which are 6 apart. Replace each of the five
numbers in the sequence with its remainder when divided by 5.
If the first remainder is 0, then the sequence of remainders is 0, 1, 2, 3, 4. If the first remainder is 1, then the
sequence of remainders is 1, 2, 3, 4, 0. If the first remainder is 2, then the sequence of remainders is 2, 3, 4, 0,
1. If the first remainder is 3, then the sequence of remainders is 3, 4, 0, 1, 2. If the first remainder is 4, then the
sequence of remainders is 4, 0, 1, 2, 3.
In all cases there is a remainder of 0, which represents a multiple of 5 in the original sequence. The only prime that
is divisible by 5 is 5 itself. Since there is no positive number that is 6 less than 5, 5 must be the first number in the
original sequence. Thus the original sequence is 5, 11, 17, 23, 29.

J4 Condates
a Any condate in a year starting with 20 must be DD/MM/20YY. So the month must be 1M. Then there is no
available digit for M. So there is no condate in the years 2000 to 2099.
Any condate in a year starting with 21 must be DD/MM/21YY. So the month must be 0M. Hence the day is 3D.
Then there is no available digit for D. So there is no condate in the years 2100 to 2199.
There is no condate in a year starting with 22, since 2 is repeated.
Any condate in a year starting with 23 must be DD/MM/23YY. So the month is 0M or 10. If the month is 10,
then there is no available digit for the first D in the day. So the month is 0M and the condate is 1D/0M/23YY.
The earliest year is 2345 and the earliest month in that year is 06. Hence the first condate after the year 2015 is
17/06/2345.
b In any year either 0 or 1 must appear in the month. If 0 is not in the month, then the month is 12. Hence the year
cannot start with 2. So in the years 2000 to 2999, 0 must be in the month for any condate.
c From Part b, 0 must be in the month for any condate from 2000 to 2999. If the 0 is not the first digit of the month,
then the month must be 10. Hence the remaining digits are 3, 4, 5, 6, 7, no two of which can represent a day. So
in the years 2000 to 2999, 0 must be the first digit in the month for any condate.
d From Part c, a condate in the years 2000 to 2999 has the form DD/0M/2YYY. So DD is either 1D or 3D.
If the condate is 1D/0M/2YYY, then each of the letters can be replaced by any one of the digits 3, 4, 5, 6, 7
without repeating a digit. There are 5 choices for D, then 4 choices for M, and so on. So the number of condates
is 5 4 3 2 1 = 120.
If the condate is 3D/0M/2YYY, then D is 1 and the month has 31 days. The remaining available digits are 4, 5,
6, 7. So M is 5 or 7. Thus there are 2 choices for M, then 3 choices for the first Y, and so on. So the number of
condates is 2 3 2 1 = 12.
Thus the number of condates in the years 2000 to 2999 is 120 + 12 = 132.

20

37

J5 Jogging
4
a On the flat, Theo will run a total of 4 km at 12 km/h and this will take 12
60 = 20 minutes. He will run 3 km
3
uphill at 9 km/h and this will take 9 60 = 20 minutes. He will run 3 km downhill at 15 km/h and this will take
3
60 = 12 minutes. So the total time taken for Theo to complete the course will be 20 + 20 + 12 = 52 minutes.
15

b Alternative i
If the wind blows from the west, then Theos speed and times for the various sections of the course will be:
Section

Distance

Speed

Flat east
Uphill
Downhill
Flat west

2 km
3 km
3 km
2 km

13 km/h
10 km/h
14 km/h
11 km/h

2
So the total time to complete the course would be ( 13
+

3
10

3
14

Time
2
13
3
10
3
14
2
11

2
11 )

h
h
h
h

60 51.0 minutes.

If the wind blows from the east, then Theos speed and times for the various sections of the course will be:
Section

Distance

Speed

Time

Flat east
Uphill
Downhill
Flat west

2 km
3 km
3 km
2 km

11 km/h
8 km/h
16 km/h
13 km/h

2
11 h
3
8 h
3
16 h
2
h
13

2
So the total time to complete the course would be ( 11
+

3
8

3
16

2
13 )

60 53.9 minutes.

So Theos time will be less if the wind blows from the west.
Alternative ii
Since Theo runs both ways on the flat, he gains no advantage on the flat from either wind direction.
3
3
+ 14
hours. If the wind blows from the east,
If the wind blows from the west, then his time on the hill will be 10
3
3
then his time on the hill will be 8 + 16 hours. To simplify comparison of these times we divide both by 3 and
1
1
12
multiply both by 2. So we want to compare 15 + 17 = 12
35 with 4 + 8 = 32 . Thus the first time is shorter.

So Theos time will be less if the wind blows from the west.
c Alternative i
From Part a, Theo has to gain 2 minutes.
From Part a, he ran 3 km uphill at 9 km/h in 20 minutes. So to gain 2 minutes uphill, he will need to run 3 km in
3
18 minutes. Therefore his speed needs to be 18
60 = 10 km/h, which is 1 km/h faster.
From Part a, he ran 3 km downhill at 15 km/h in 12 minutes. So to gain 2 minutes downhill, he will need to run
3
3 km in 10 minutes. Therefore his speed needs to be 10
60 = 18 km/h, which is 3 km/h faster.
Alternative ii
Theo wants to complete the course in 50 minutes. From Part a, the flat takes 20 minutes. This leaves 30 minutes
for the hill.
3
If Theos speed uphill is r km/h and downhill is 15 km/h, then the time taken for the hill in hours is r3 + 15
=
3
1
1
3
So r = 2 5 = 10 . Hence r = 10 km/h. This is 1 km/h faster.

If Theos speed downhill is r km/h and uphill is 9 km/h, then the time taken for the hill in hours is
So 3r = 12 13 = 16 . Hence r = 18 km/h. This is 3 km/h faster.

21

3
9

+ 3r =

30
60

= 12 .

30
60

= 12 .

39

J6 Tessellating Hexagons
a The flipped hexagons are shaded.
D

E
F

E
F

C
B

A
F

A C
D

B
F E

A C
D

B
D
C A

D
C A

F
E

F
A

F
E

C
F

A C
D

B
F E

A C
D

B
D
C A

D
C A

A
C

F
E

F
E

b Call the block of hexagons in the solution to Part a the 16-block.


Alternative i
We see from the 16-block that AB = ED, BC = F A, the sum of the three interior angles A, C, D is 360 , and the
sum of the three interior angles B, E, F is also 360 .
Indicating a hexagon side with a dash, the bottom boundary of the 16-block is
A BF E D CA BF E D and the top boundary is
D E F B AC D E F B A.
So these boundaries are identical. Hence the 16-block can be translated vertically to form an infinite column without
gaps and without overlap.
The right boundary of this column is
DA F EB C DA F EB C DA repeated indefinitely.
The left boundary of the column is
C BE F AD C BE F AD C repeated indefinitely.
So these boundaries are identical. Hence the column can be translated horizontally to tessellate the plane.
Alternative ii
Make several photocopies of the 16-block. Place them side by side and top to bottom to show how they fit together.

22

39

c There is no point of symmetry for the tessellation inside a hexagon because the hexagon has no point of symmetry.
So all points of symmetry for the tessellation must be on the edges of the hexagons. If a point of symmetry for the
tessellation is on an edge of a hexagon, then it must be at the midpoint of that edge.
From the block of hexagons in Part a, we see that each of the edges AB, BC, DE, F A, has the smallest edge EF
attached at one end but not the other. So the midpoints of edges AB, BC, DE, F A are not points of symmetry
for the tessellation.
To check the midpoint of edge CD, trace a copy of the block and place it exactly over the original block. Then
place a pin through the midpoint of edge CD, rotate the traced copy of the block through 180 and notice that it
again fits exactly over the original block (except for overhang). Thus the midpoint of CD is a point of symmetry
for the tessellation. Similarly for the midpoint of EF .
So the only points of symmetry for the tessellation are the midpoints of edges CD and EF .

23

41

d The block of hexagons in Part a is reproduced below with hexagons labelled H1 , H2 , H3 , H4 , H1 , H2 , H3 , H4 , and
so on as shown.
D

E
F

C
B

H3

A
A C
D

H4

H4

H3

A C
D

B
F E
E

B
D
C A

H1

C A

H1
A

H2

F
E

F
E

H4

H3

A C
D

H4

B
F E

H3

A C
D

H1

D
C A

C A
A
C

H2
A

H2

H1

F
E
B

E
F

H2

F
E

F
E

F
E

Hexagons with the same subscript translate to one another. So there are several blocks of four hexagons that
tessellate by translation. Three such blocks are: {H1 , H2 , H3 , H4 }, {H1 , H2 , H3 , H4 }, {H1 , H2 , H3 , H4 }.

24

41

CHALLENGE
SOLUTIONS
INTERMEDIATE
2015 Challenge
Solutions - Intermediate
I1 Indim Integers
See Junior Problem 2.

I2 Digital Sums
a Alternative i
The table lists 3-digit numbers n in decreasing order whose digit sums s are multiples of 7.
n
993
984
975
966
957
950
948
941
939
932
923
914

s
21
21
21
21
21
14
21
14
21
14
14
14

7 divides n?
no
no
no
yes
no
no
no
no
no
no
no
no

n
905
894
885
876
867
860
858
851
849
842
833

s
14
21
21
21
21
14
21
14
21
14
14

7 divides n?
no
no
no
no
no
no
no
no
no
no
yes

So the largest even and odd n that are multiples of 7 and have a digital sum that is also a multiple of 7 are 966 and
833 respectively.
Alternative ii
The table lists 3-digit numbers n and their digital sums s. The n are multiples of 7 in decreasing order.
n
994
987
980
973
966
959
952
945
938
931
924
917

s
22
24
17
19
21
23
16
18
20
13
15
17

7 divides s?
no
no
no
no
yes
no
no
no
no
no
no
no

n
910
903
896
889
882
875
868
861
854
847
840
833

s
10
12
23
25
18
20
22
15
17
19
12
14

7 divides s?
no
no
no
no
no
no
no
no
no
no
no
yes

So the largest even and odd n that are multiples of 7 and have a digital sum that is also a multiple of 7 are 966 and
833 respectively.
b Alternative i
The largest 3-digit number with a digital sum of 7 is 700. The digital sum s of a 3-digit number is at most 39 = 27.
So, if the FDS of a 3-digit number n is 7 and n > 700, then the digital sum of n is 16 or 25. The table lists, in
decreasing order, the first few 3-digit numbers n with digital sum 16 or 25.

25

43

s
16

n
970
961
952
943
934
925
916
907
880
871
862
853
844
835
826

7 divides n?
no
no
yes
no
no
no
no
no
no
no
no
no
no
no
yes

s
25

n
997
988
979
898
889
799

7 divides n?
no
no
no
no
yes
no

So the three largest 3-digit multiples of 7 that have FDS 7 are: 952, 889, 826.
Alternative ii
The table lists, in decreasing order, multiples of 7 and their digital sum sequences (DSS).
n
994
987
980
973
966
959
952
945
938
931
924
917
910

DSS
22, 4
24, 6
17, 8
19, 10, 1
21, 3
23, 5
16, 7
18, 9
20, 2
13, 4
15, 6
17, 8
10, 1

n
903
896
889
882
875
868
861
854
847
840
833
826

DSS
12, 3
23, 5
25, 7
18, 9
20, 2
22, 4
15, 6
17, 8
19, 10, 1
12, 3
14, 5
16, 7

So the three largest 3-digit multiples of 7 that have FDS 7 are: 952, 889, 826.
c Alternative i
The table lists, in decreasing order, the 3-digit integers n that are multiples of 7 together with their FDSs (F).

26

43

n
994
987
980
973
966
959
952
945
938
931
924
917
910
903

F
4
6
8
1
3
5
7
9
2
4
6
8
1
3

n
896
889
882
875
868
861
854
847
840
833
826
819
812
805

F
5
7
9
2
4
6
8
1
3
5
7
9
2
4

497
490
483
476
469
462
455
448
441
434
427
420
413
406

2
4
6
8
1
3
5
7
9
2
4
6
8
1

399
392
385
378
371
364
357
350
343
336
329
322
315
308
301

3
5
7
9
2
4
6
8
1
3
5
7
9
2
4

n
798
791
784
777
770
763
756
749
742
735
728
721
714
707
700
294
287
280
273
266
259
252
245
238
231
224
217
210
203

F
6
8
1
3
5
7
9
2
4
6
8
1
3
5
7
6
8
1
3
5
7
9
2
4
6
8
1
3
5

n
693
686
679
672
665
658
651
644
637
630
623
616
609
602

F
9
2
4
6
8
1
3
5
7
9
2
4
6
8

196
189
182
175
168
161
154
147
140
133
126
119
112
105

7
9
2
4
6
8
1
3
5
7
9
2
4
6

n
595
588
581
574
567
560
553
546
539
532
525
518
511
504

F
1
3
5
7
9
2
4
6
8
1
3
5
7
9

Thus the only 3-digit multiples of 7 that have FDS 7 are: 133, 196, 259, 322, 385, 448, 511, 574, 637, 700, 763, 826,
889, 952. These have a common difference of 63. So the only 3-digit multiples of 7 that have FDS 7 are the integers
n = 7 + 63t where t = 2, 3, . . ., 15.
Alternative ii
The three numbers from Part b, 952, 889, 826, have a common difference of 63. This suggests that all 3-digit
multiples of 7 that have FDS 7 have the form 7 + 63m. These numbers are: 133, 196, 259, 322, 385, 448, 511, 574,
637, 700, 763, 826, 889, 952. Since 7 divides 7 and 63, all these numbers are multiples of 7. By direct calculation,
each has FDS 7. But are there any other 3-digit multiples of 7 that have FDS 7?
Extending the table in the first solution to Part b shows that there are no other n besides those of the form 7 + 63m
that have digital sum 16 or 25. The following table lists in decreasing order all 3-digit integers with digital sum 7.
n
700
610
601
520
511
502
430
421
412
403

7 divides n?
yes
no
no
no
yes
no
no
no
no
no

n
340
331
322
313
304
250
241
232
223
214

7 divides n?
no
no
yes
no
no
no
no
no
no
no

n
205
160
151
142
133
124
115
106

7 divides n?
no
no
no
no
yes
no
no
no

So there are no other 3-digit multiples of 7 with FDS 7 besides those of the form 7 + 63m.

27

45

Alternative iii
From the table in the first solution to Part b, it appears that integers with the same FDS differ by a multiple of
9. This is equivalent to saying they have the same remainder when divided by 9. We now show that this is always
true.
Any positive integer n can be written in the form
n = a + 10b + 100c +
where a, b, . . . are the digits of n. Thus
n =
=

a + (1 + 9)b + (1 + 99)c +
(a + b + c + ) + 9(b + 11c + )

So n and its digital sum have the same remainder when divided by 9. Therefore all digital sums in the digital sum
sequence for n, including its FDS, have the same remainder when divided by 9. Hence, if n is a multiple of 9, then
the FDS of n is 9, and if n is not a multiple of 9, then the FDS of n is the remainder when n is divided by 9.
So, the FDS of n is 7 if and only if n has the form n = 7 + 9r. Such an n is a multiple of 7 if and only if 7 divides
r. So n is a multiple of 7 and has FDS 7 if and only if n has the form n = 7 + 63t. Hence the only 3-digit multiples
of 7 that have FDS 7 are the integers n = 7 + 63t where t = 2, 3, . . ., 15. These are: 133, 196, 259, 322, 385, 448,
511, 574, 637, 700, 763, 826, 889, 952.
d Alternative i
The table lists, in decreasing order, the 3-digit integers n that are multiples of 8 together with their FDSs (F).
n
992
984
976
968
960
952
944
936
928
920
912
904

F
2
3
4
5
6
7
8
9
1
2
3
4

496
488
480
472
464
456
448
440
432
424
416
408
400

1
2
3
4
5
6
7
8
9
1
2
3
4

n
896
888
880
872
864
856
848
840
832
824
816
808
800
392
384
376
368
360
352
344
336
328
320
312
304

F
5
6
7
8
9
1
2
3
4
5
6
7
8
5
6
7
8
9
1
2
3
4
5
6
7

n
792
784
776
768
760
752
744
736
728
720
712
704

F
9
1
2
3
4
5
6
7
8
9
1
2

296
288
280
272
264
256
248
240
232
224
216
208
200

8
9
1
2
3
4
5
6
7
8
9
1
2

n
696
688
680
672
664
656
648
640
632
624
616
608
600
192
184
176
168
160
152
144
136
128
120
112
104

F
3
4
5
6
7
8
9
1
2
3
4
5
6
3
4
5
6
7
8
9
1
2
3
4
5

n
592
584
576
568
560
552
544
536
528
520
512
504

F
7
8
9
1
2
3
4
5
6
7
8
9

Thus the only 3-digit multiples of 8 that have FDS 8 are: 152, 224, 296, 368, 440, 512, 584, 656, 728, 800, 872,
944. These have a common difference of 72. So the only 3-digit multiples of 8 that have FDS 8 are the integers
n = 8 + 72t where t = 2, 3, . . ., 13.
Alternative ii
From the third solution to Part c, the FDS of a positive integer n is 8 if and only if n has the form n = 8 + 9r.
Such an n is a multiple of 8 if and only if 8 divides r. So n is a multiple of 8 and has FDS 8 if and only if n has the
form n = 8 + 72t. Hence the only 3-digit multiples of 8 that have FDS 8 are the integers n = 8 + 72t where t = 2,
3, . . ., 13. These are: 152, 224, 296, 368, 440, 512, 584, 656, 728, 800, 872, 944.

28

45

I3 Coin Flips
a Here are three moves that leave exactly 10 coins heads up.
Start:
Move 1:
Move 2:
Move 3:

T T T T T T T T T T T
HHHHT T T T T T T
HHHHHHHHT T T
HHHHHHHT HHH

T
T
T
T

T
T
T
T

T
T
T
T

b Since there are 52 coins flipped in each move, the total number of coin flips is 3 52 = 156. We start with 154 T s
and want to finish with 154 Hs. We can achieve this by flipping 153 coins exactly once and flipping one coin three
times over the three moves. Here are three moves that leave all coins heads up.
154

Start:

  
T . . .T
52

102

     
Move 1: H . . . H T . . . T
51

51

51

        
H . . .H H T . . .T T . . .T
51

51

51

        
Move 2: H . . . H T H . . . H T . . . T
51

51

51

        
Move 3: H . . . H H H . . . H H . . . H
c Since there are 4 coins flipped in each move, the total number of coins flipped in 6 moves is at most 24. So it takes
at least 7 moves to finish with all coins heads up. We now show that this can actually be done in 7 moves.
After three moves, each flipping 4 Ts to 4 Hs, we get exactly 12 coins heads up. Then applying the three moves in
Part a to the 14 coins that are tails up, gives us exactly 12 + 10 = 22 coins heads up. One more move, flipping 4
Ts to 4 Hs, results in all coins heads up. Thus seven moves suffice to get all coins heads up.
So 7 is the minimum number of moves to get all coins heads up.
d Alternative i
Suppose m coins are flipped on each move where m is odd. Two moves give the following results.
154

Start:

  
T . . .T
m

154m

     
Move 1: H . . . H T . . . T
mn

mn

1542m+n

           
Move 2: H . . . H T . . . T H . . . H T . . . T

2m2n

1542m+2n

  
H . . .H

  
T . . .T

where 0 n m and
m n 154 m

To have all 154 coins with heads up after Move 3, we must have m = 154 2m + 2n. Thus 3m = 154 + 2n. But
3m is odd and 154 + 2n is even. So we cannot have 154 coins heads up after just three moves.

29

47

Alternative ii
Three moves give the following results. Note that in a move, an even number of coins may mean 0 coins.
154

Start:

  
T . . .T
odd

odd

     
Move 1: H . . . H T . . . T
even

odd

even

odd

           
Move 2: H . . . H T . . . T H . . . H T . . . T or
odd

even

odd

even

           
H . . . H T . . . T H . . . H T . . . T giving
even
even
     
H . . .H T . . .T
even

even

odd

odd

           
Move 3: H . . . H T . . . T H . . . H T . . . T or
odd

odd

even

even

           
H . . . H T . . . T H . . . H T . . . T giving
odd

odd

     
H . . .H T . . .T
Since 154 is even, we cannot have 154 coins heads up after just three moves.

I4 Jogging
See Junior Problem 5.

I5 Folding Fractions
a Let QR = x. Then P R = 1 x.
T

1
2

1
2

1x

1x

U
V

In P QR Pythagoras theorem gives


Hence 2x = 34 and x = 38 .

1
4

+ x2 = (1 x)2 = x2 2x + 1.

b Alternative i
From Part a, we have:
1
2

1
2

b
5
8

3
8

a
R

b
S
U

b
a
V

30

47

Since angles RQP , RP S, P T S, SU V are right angles, triangles RQP , P T S, V U S have the same angles as shown.
Hence they are similar and we have: V U :U S:SV = P T :T S:SP = RQ:QP :P R = 3:4:5.
Hence P S = 35 P T =
Then SV = 54 U S =

5
1
5
3 2 = 6 . So U S = U P SP
5
5
16 = 24
and V U = 34 U S = 34
4

5
1
6 = 6.
1
= 18 .
6

=1

Alternative ii
As in Alternative i, V U :U S:SV = P T :T S:SP = RQ:QP :P R = 3:4:5. So ST = 43 P T =
From the side of the square, we have: 1 = U V + V S + ST = 53 V S + V S +
5
So V S = 13 58 = 24
.
From V U S, we have: V U = 35 SV =

3
5

5
24

1
8

and U S = 45 SV =

4
5

2
3

5
24

4
3

1
2

= 23 .

= 85 V S + 23 .
= 16 .

Alternative iii
Draw RW perpendicular to T S and let T S = t and V U = r. From Part a, we have:
T

1
2

1
2

3
8

3
8

5
8

3
8

S
U

r
V r

Now apply Pythagoras theorem to the following triangles.


In RW S, RS 2 = RW 2 + W S 2 = 1 + (t 38 )2 .
In RP S, RS 2 = RP 2 + P S 2 = ( 58 )2 + P S 2 .
In P T S, P S 2 = P T 2 + T S 2 = ( 12 )2 + t2 .
Hence

1 + (t 38 )2
9
1 + t2 + 64
34 t
3
t
4
t

( 58 )2 + ( 12 )2 + t2
25
+ 14 + t2
64
9
1 14 + 64
25
64
3
16
3
1

4
64
4
4 =
4
1
2
3 2 = 3

=
=
=
=
=

1
2

5
So P S 2 = ( 12 )2 + ( 23 )2 = 14 + 49 = 25
36 and P S = 6 .
5
1
Hence U S = U P SP = 1 6 = 6 .
Since 1 = T S + SV + r = 23 + SV + r, SV = 31 r.

In V U S, Pythagoras theorem gives V S 2 = V U 2 + U S 2 . Hence

1
9

Thus V U =

1
8

and SV =

1
3

1
8

( 31 r)2
+ r 2 23 r
2
3r
r

=
=
=
=

r 2 + ( 16 )2
1
r 2 + 36
1
1
9 36 =
3
1
2 12 =

3
36
1
8

1
12

5
24 .

31

49

c Since angles RQP , RP S, P T S, SU V are right angles, triangles RQP , P T S, V U S have the same angles as shown.
Hence they are similar and we have RQ:QP :P R = P T :T S:SP = V U :U S:SV = 7:24:25 or 24:7:25.
T

a
R

b
S

b
a

Alternative i
If P Q/QR = 7/24, let P Q = 7k. Then QR = 24k and P R = 25k. Since P R = 1 QR, we have 25k = 1 24k. So
1
49k = 1 and P Q = 7 49
= 17 .
If P Q/QR = 24/7, let P Q = 24k. Then QR = 7k and P R = 25k. Since P R = 1 QR, we have 25k = 1 7k. So
1
32k = 1 and P Q = 24 32
= 34 .
Alternative ii
If P Q/QR = 7/24, then P R/QR = 25/24. Since P R = 1 QR, we have 24 24QR = 25QR, QR = 24/49, and
P R = 25/49. From Pythagoras, P Q2 = P R2 QR2 = 1 2QR = 1 48/49 = 1/49. So P Q = 17 .
If P Q/QR = 24/7, then P R/QR = 25/7. Since P R = 1 QR, we have 7 7QR = 25QR, QR = 7/32, and
P R = 25/32. From Pythagoras, P Q2 = P R2 QR2 = 1 2QR = 1 7/16 = 9/16. So P Q = 34 .
Alternative iii
Let T P = x. If V U/U S = 7/24, let V U = 7k. Then U S = 24k and V S = 25k and we have:
T

1 32k

1 24k
R

24k

25k

7k

7k

So (1 24k)/25k = x/7k = (1 32k)/24k.


Hence 25x = 7(1 24k) and 24x = 7(1 32k).
Subtracting gives x = 7(32 24)k = 56k.
Substituting gives 25(56k) = 7(1 24k).
Hence 1 = 224k and x = 56/224 = 8/32 = 1/4.

32

49

Alternatively, in P T S, Pythagoras theorem gives


(1 24k)2
(24k)2 48k
16k
1
x

=
=
=
=
=

(1 32k)2 + (56k)2
(32k)2 64k + (56k)2
(562 + 322 242 )k 2
(142 + 82 62 )k = 224k
56/224 = 8/32 = 1/4

If V U/U S = 24/7, let V U = 24k. Then U S = 7k and V S = 25k. We have:


T

1 49k

1 7k
R

7k

25k

24k

24k

So (1 7k)/25k = x/24k = (1 49k)/7k.


Hence 25x = 24(1 7k) and 7x = 24(1 49k).
Subtracting gives 18x = 24(49 7)k = 24(42)k, hence x = 56k.
Substituting gives 7(56k) = 24(1 49k).
Hence 49k = 3(1 49k), 4(49k) = 3, x = 56(3)/4(49) = 6/7.
Alternatively, in P T S, Pythagoras theorem gives
(1 7k)2
(7k)2 14k
84k
12
3
x

=
=
=
=
=
=

(1 49k)2 + (56k)2
(49k)2 98k + (56k)2
(562 + 492 72 )k 2
(82 + 72 12 )7k = (112)7k
(28)7k
6/7

I6 Crumbling Cubes
a The small cubes have at most three faces exposed. The only small cubes that have exactly three faces exposed
are the 8 on the corners of the 10 10 10 cube. The only small cubes that have exactly two faces exposed are
the 8 on each edge of the 10 10 10 cube that are not on its corners. All other cubes have less than two faces
exposed. There are 12 edges on the 10 10 10 cube, so the number of small cubes removed at the first step is
8 + (12 8) = 104.
b Alternative i
At the first step the 8 small cubes at the corners are removed. The other 192 cubes come equally from the 12 edges
but not from the corners. So the number of non-corner small cubes on each edge is 192/12 = 16. Hence each edge
in the original cube had a total of 18 small cubes.
Alternative ii
A cube with n small cubes along one edge will lose 8 + 12(n 2) small cubes at the first step. So 8 + 12(n 2) = 200,
12(n 2) = 192, n 2 = 16, n = 18. Thus the original cube was 18 18 18.

33

51

Alternative iii
From Part a, a 10 10 10 cube loses 104 small cubes at the first step. Increasing the cubes dimension by 1
increases the number of lost cubes by 12, one for each edge. Since 200 104 = 96 and 96/12 = 8, the cube that
loses 200 small cubes at the first step is 18 18 18.
c Alternative i
At the first step, each of the 6 faces of the 9 9 9 cube are converted to a 7 7 single layer of small cubes. The
exposed surface area of this layer is 7 7 small faces plus a ring of 4 7 small faces. So the surface area of the
remaining object is 6 (49 + 28) = 6 77 = 462.
Alternative ii
First remove the middle small cube on one edge of the 9 9 9 cube. This increases the surface area by 2 small
faces. Then remove the small cubes either side. This does not change the surface area. Continue until only the two
corner cubes remain. At this stage the surface area has increased by 2 small faces. Repeating this process on all
12 edges increases the surface area by 12 2 small faces. At this stage all 8 corner cubes have 6 exposed faces. So
removing the 8 corner cubes reduces the surface area by 8 6 small faces. The surface area of the original 9 9 9
cube was 6 81 = 486. Hence the surface area of the remaining object is 486 + 24 48 = 462.
d At the first step, the original 9 9 9 cube is reduced to a 7 7 7 cube with a 7 7 single layer of small cubes
placed centrally on each face. At this stage the number of small cubes removed is 8 + (12 7) = 92.

77

At the second step the only small cubes removed are the edge cubes in the 7 7 single layers. This leaves a 7 7 7
cube with a 5 5 single layer of small cubes placed centrally on each face. So in this step, the number of small
cubes removed is 6(4 + 4 5) = 144.

55

At the third step the only small cubes removed are the edge cubes in the 5 5 single layers and the edge cubes in
the 7 7 7 cube. So in this step, the number of small cubes removed is 6(4 + 4 3) + 8 + (12 5) = 164.
Thus the number of small cubes remaining is (9 9 9) 92 144 164 = 329.

34

51

CHALLENGE STATISTICS MIDDLE PRIMARY


MEAN SCORE/SCHOOL YEAR/PROBLEM
Mean
Year

Number of
Students

Overall

Problem
1

486

8.8

2.1

2.4

2.3

2.1

653

10.6

2.6

2.9

2.7

2.6

All Years*

1166

9.8

2.3

2.7

2.6

2.4

Please note:* This total includes students who did not provide their school year.

SCORE DISTRIBUTION %/PROBLEM


Challenge Problem
Score

1
e-Numbers

Quazy Quilts

Egg Cartons

Condates

Did not attempt

0%

1%

2%

3%

7%

6%

10%

10%

15%

12%

11%

18%

29%

20%

18%

20%

32%

29%

33%

24%

16%

32%

26%

25%

Mean

2.3

2.7

2.6

2.4

Discrimination
Factor

0.5

0.6

0.6

0.6

Please note:
The discrimination factor for a particular problem is calculated as follows:
(1) The students are ranked in regard to their overall scores.
(2) The mean score for the top 25% of these overall ranked students is calculated for that particular problem
including no attempts. Call this mean score the mean top score.
(3) The mean score for the bottom 25% of these overall ranked students is calculated for that particular problem
including no attempts. Call this mean score the mean bottom score.
(4) The discrimination factor = mean top score mean bottom score
4
Thus the discrimination factor ranges from 1 to 1. A problem with a discrimination factor of 0.4 or higher is
considered to be a good discriminator.

53

CHALLENGE STATISTICS UPPER PRIMARY


MEAN SCORE/SCHOOL YEAR/PROBLEM
Mean
YEAR

Number of
Students

Overall

Problem
1

1363

8.6

1.8

2.8

2.4

1.7

1910

9.8

2.0

3.0

2.8

2.0

129

10.9

2.3

3.3

3.2

2.2

All Years*

3416

9.4

2.0

2.9

2.7

1.9

Please note:* This total includes students who did not provide their school year.

SCORE DISTRIBUTION %/PROBLEM


Challenge Problem
3

Egg Cartons

Seahorse
Swimmers

Primelandia
Money

1%

0%

1%

3%

3%

3%

5%

11%

40%

6%

14%

29%

24%

20%

20%

26%

21%

38%

29%

20%

11%

33%

31%

12%

Mean

2.0

2.9

2.7

1.9

Discrimination
Factor

0.5

0.4

0.6

0.6

Score

Rod Shapes
Did not attempt

Please note:
The discrimination factor for a particular problem is calculated as follows:
(1) The students are ranked in regard to their overall scores.
(2) The mean score for the top 25% of these overall ranked students is calculated for that particular problem
including no attempts. Call this mean score the mean top score.
(3) The mean score for the bottom 25% of these overall ranked students is calculated for that particular problem
including no attempts. Call this mean score the mean bottom score.
(4) The discrimination factor = mean top score mean bottom score
4
Thus the discrimination factor ranges from 1 to 1. A problem with a discrimination factor of 0.4 or higher is
considered to be a good discriminator.
53

CHALLENGE STATISTICS JUNIOR


MEAN SCORE/SCHOOL YEAR/PROBLEM
Mean
Year

Number of
Students

Problem

Overall

2607

11.0

2.6

2.3

1.9

2.0

2.2

1.1

2373

13.4

3.0

2.7

2.2

2.3

2.7

1.4

All Years*

5006

12.2

2.8

2.5

2.0

2.1

2.4

1.3

Please note:* This total includes students who did not provide their school year.

SCORE DISTRIBUTION %/PROBLEM


Challenge Problem
1

Quirky
Quadrilaterals

Indim
Integers

Did not
attempt

3%

Score

Primelandia
Money

Condates

Jogging

Tessellating
Hexagons

3%

7%

8%

9%

23%

7%

6%

8%

15%

10%

28%

8%

17%

24%

14%

18%

20%

15%

23%

30%

19%

16%

13%

36%

26%

22%

31%

21%

10%

31%

26%

10%

13%

27%

5%

Mean

2.8

2.5

2.0

2.1

2.4

1.3

Discrimination
Factor

0.5

0.6

0.6

0.7

0.7

0.5

Please note:
The discrimination factor for a particular problem is calculated as follows:
(1) The students are ranked in regard to their overall scores.
(2) The mean score for the top 25% of these overall ranked students is calculated for that particular problem
including no attempts. Call this mean score the mean top score.
(3) The mean score for the bottom 25% of these overall ranked students is calculated for that particular problem
including no attempts. Call this mean score the mean bottom score.
(4) The discrimination factor = mean top score mean bottom score
4
Thus the discrimination factor ranges from 1 to 1. A problem with a discrimination factor of 0.4 or higher is
considered to be a good discriminator.

55

CHALLENGE STATISTICS INTERMEDIATE


MEAN SCORE/SCHOOL YEAR/PROBLEM
Mean
Year

Number of
Students

Problem

Overall

2038

15.2

3.1

2.6

3.0

3.0

2.0

2.5

10

1055

16.6

3.3

2.9

3.2

3.2

2.2

2.7

All Years*

3104

15.7

3.2

2.7

3.0

3.1

2.0

2.6

Please note:* This total includes students who did not provide their school year.

SCORE DISTRIBUTION %/PROBLEM


Challenge Problem
5

Jogging

Folding
Fractions

Crumbling
Cubes

5%

3%

14%

9%

6%

4%

4%

14%

11%

7%

11%

6%

9%

21%

7%

15%

26%

15%

12%

21%

21%

25%

16%

28%

22%

8%

25%

50%

38%

42%

49%

22%

27%

Mean

3.2

2.7

3.0

3.1

2.0

2.6

Discrimination
Factor

0.4

0.6

0.6

0.6

0.7

0.7

Score

1
Indim
Integers

Digital Sums

Coin Flips

Did not
attempt

1%

3%

2%

Please note:
The discrimination factor for a particular problem is calculated as follows:
(1) The students are ranked in regard to their overall scores.
(2) The mean score for the top 25% of these overall ranked students is calculated for that particular problem
including no attempts. Call this mean score the mean top score.
(3) The mean score for the bottom 25% of these overall ranked students is calculated for that particular problem
including no attempts. Call this mean score the mean bottom score.
(4) The discrimination factor = mean top score mean bottom score
4
Thus the discrimination factor ranges from 1 to 1. A problem with a discrimination factor of 0.4 or higher is
considered to be a good discriminator.
55

2015 Australian INTERMEDIATE


Intermediate Mathematics
Olympiad OLYMPIAD
- Questions
AUSTRALIAN
MATHEMATICS
Time allowed: 4 hours.

NO calculators are to be used.

Questions 1 to 8 only require their numerical answers all of which are non-negative integers less than 1000.
Questions 9 and 10 require written solutions which may include proofs.
The bonus marks for the Investigation in Question 10 may be used to determine prize winners.
1. A number written in base a is 123a. The same number written in base b is 146b . What is the minimum value of
a + b?
[2 marks]

2. A circle is inscribed in a hexagon ABCDEF so that each side of the hexagon is tangent to the circle. Find the
perimeter of the hexagon if AB = 6, CD = 7, and EF = 8.
[2 marks]

3. A selection of 3 whatsits, 7 doovers and 1 thingy cost a total of $329. A selection of 4 whatsits, 10 doovers and 1
thingy cost a total of $441. What is the total cost, in dollars, of 1 whatsit, 1 doover and 1 thingy?
[3 marks]
a
2
1
, can also be written in the form + 2 , where n is a positive integer.
b
n n
If a + b = 1024, what is the value of a?
[3 marks]

4. A fraction, expressed in its lowest terms

5. Determine the smallest positive integer y for which there is a positive integer x satisfying the equation
213 + 210 + 2x = y2 .
[3 marks]

6. The large circle has radius 30/ . Two circles with diameter 30/ lie inside the large circle. Two more circles
lie inside the large circle so that the five circles touch each other as shown. Find the shaded area.

[4 marks]

7. Consider a shortest path along the edges of a 7 7 square grid from its bottom-left vertex to its top-right vertex.
How many such paths have no edge above the grid diagonal that joins these vertices?
[4 marks]

8. Determine the number of non-negative integers x that satisfy the equation


   
x
x
=
.
44
45
(Note: if r is any real number, then r denotes the largest integer less than or equal to r.)

[4 marks]

57

9. A sequence is formed by the following rules: s1 = a, s2 = b and sn+2 = sn+1 + (1)n sn for all n 1.

If a = 3 and b is an integer less than 1000, what is the largest value of b for which 2015 is a member of the sequence?
Justify your answer.
[5 marks]

10. X is a point inside an equilateral triangle ABC. Y is the foot of the perpendicular from X to AC, Z is the foot
of the perpendicular from X to AB, and W is the foot of the perpendicular from X to BC.
The ratio of the distances of X from the three sides of the triangle is 1 : 2 : 4 as shown in the diagram.
B

W
Z

2
X
1

If the area of AZXY is 13 cm2 , find the area of ABC. Justify your answer.

[5 marks]

Investigation
If XY : XZ : XW = a : b : c, find the ratio of the areas of AZXY and ABC.

[2 bonus marks]

57

2015 AustralianINTERMEDIATE
Intermediate Mathematics
OlympiadOLYMPIAD
- Solutions
AUSTRALIAN
MATHEMATICS
SOLUTIONS
1. Method 1
123a = 146b a2 + 2a + 3 = b2 + 4b + 6

(a + 1)2 + 2 = (b + 2)2 + 2
(a + 1)2 = (b + 2)2

a + 1 = b + 2 (a and b are positive)

a = b + 1

Since the minimum value for b is 7, the minimum value for a + b is 8 + 7 = 15.
Method 2
Since the digits in any number are less than the base, b 7.
We also have a > b, otherwise a2 + 2a + 3 < b2 + 4b + 6.
If b = 7 and a = 8, then a2 + 2a + 3 = 83 = b2 + 4b + 6.
So the minimum value for a + b is 8 + 7 = 15.

2. Let AB, BC, CD, DE, EF , F A touch the circle at U , V , W , X, Y , Z respectively.


E
X
D
Y

F
C
Z
V
A

Since the two tangents from a point to a circle have equal length,
U B = BV , V C = CW , W D = DX, XE = EY , Y F = F Z, ZA = AU .
The perimeter of hexagon ABCDEF is
AU + U B + BV + V C + CW + W D + DX + XE + EY + Y F + F Z + ZA
= AU + U B + U B + CW + CW + W D + W D + EY + EY + Y F + Y F + AU
= 2(AU + U B + CW + W D + EY + Y F )
= 2(AB + CD + EF ) = 2(6 + 7 + 8) = 2(21) = 42.

3. Preamble
Let the required cost be x. Then, with obvious notation, we have:
3w + 7d + t = 329
4w + 10d + t = 441
w+d+t = x

(1)
(2)
(3)

Method 1
3 (1) 2 (2): w + d + t = 3 329 2 441 = 987 882 = 105.

59

Method 2
(2) (1): w + 3d = 112.
(1) (3): 2w + 6d = 329 x = 2 112 = 224.
Then x = 329 224 = 105.

Method 3
10 (1) 7 (2): w = (203 3t)/2
3 (2) 4 (1): d = (7 + t)/2

Then w + d + t = 210/2 2t/2 + t = 105.

2
1
2n + 1
+
=
.
n n2
n2
Since 2n + 1 and n2 are coprime, a = 2n + 1 and b = n2 .
So 1024 = a + b = n2 + 2n + 1 = (n + 1)2 , hence n + 1 = 32.

4. We have

This gives a = 2n + 1 = 2 31 + 1 = 63.

5. Method 1

213 + 210 + 2x = y2

210 (23 + 1) + 2x = y2
(25 3)2 + 2x = y2
2x = y2 962
2x = (y + 96)(y 96).

Since y is an integer, both y + 96 and y 96 must be powers of 2.


Let y + 96 = 2m and y 96 = 2n . Then 2m 2n = 192 = 26 3.
Hence 2m6 2n6 = 3. So 2m6 = 4 and 2n6 = 1.
In particular, m = 8. Hence y = 28 96 = 256 96 = 160.
Method 2
We have y2 = 213 + 210 + 2x = 210(23 + 1 + 2x10 ) = 210 (9 + 2x10 ).
So we want the smallest value of 9 + 2x10 that is a perfect square.
Since 9 + 2x10 is odd and greater than 9, 9 + 2x10 25.

Since 9 + 21410 = 25, y = 25 5 = 32 5 = 160.


Comment

Method 1 shows that 213 + 210 + 2x = y2 has only one solution.

6. The centres Y and Y  of the two medium circles lie on a diameter of the large circle. By symmetry about this
diameter, the two smaller circles are congruent. Let X be the centre of the large circle and Z the centre of a small
circle.

Y
X
Z

Y

Let R and r be the radii of a medium and small circle respectively. Then ZY = R + r = ZY  . Since XY = XY  ,
triangles XY Z and XY  Z are congruent. Hence XZ XY .

59

By Pythagoras, Y Z 2 = Y X 2 + XZ 2 . So (R + r)2 = R2 + (2R r)2 .


Then R2 + 2Rr + r 2 = 5R2 4Rr + r 2 , which simplifies to 3r = 2R.
2
So the large circle has area (30/
) = 900,
each medium circle has area (15/
)2 = 225,
and each small circle has area (10/ )2 = 100.
Thus the shaded area is 900 2 225 2 100 = 250.

7. Method 1
Any path from the start vertex O to a vertex A must pass through either the vertex L left of A or the vertex U
underneath A. So the number of paths from O to A is the sum of the number of paths from O to L and the paths
from O to U .

There is only one path from O to any vertex on the bottom line of the grid.
So the number of paths from O to all other vertices can be progressively calculated from the second bottom row
upwards as indicated.

429

132 429

42

132 297

14

42

90

165

14

28

48

75

14

20

27

Thus the number of required paths is 429.


Method 2
To help understand the problem, consider some smaller grids.

Let p(n) equal the number of required paths on an n n grid and let p(0) = 1.

61

Starting with the bottom-left vertex, label the vertices of the diagonal 0, 1, 2, . . . , n.

i
1
0

Consider all the paths that touch the diagonal at vertex i but not at any of the vertices between vertex 0 and
vertex i. Each such path divides into two subpaths.
One subpath is from vertex 0 to vertex i and, except for the first and last edge, lies in the lower triangle of the
diagram above. Thus there are p(i 1) of these subpaths.

The other subpath is from vertex i to vertex n and lies in the upper triangle in the diagram above. Thus there are
p(n i) of these subpaths.
So the number of such paths is p(i 1) p(n i).

Summing these products from i = 1 to i = n gives all required paths. Thus


p(n) = p(n 1) + p(1)p(n 2) + p(2)p(n 3) + + p(n 2)p(1) + p(n 1)
We have p(1) = 1, p(2) = 2, p(3) = 5. So
p(4) = p(3) + p(1)p(2) + p(3)p(1) + p(3) = 14,
p(5) = p(4) + p(1)p(3) + p(2)p(2) + p(3)p(1) + p(4) = 42,
p(6) = p(5) + p(1)p(4) + p(2)p(3) + p(3)p(2) + p(1)p(4) + p(5) = 132, and
p(7) = p(6) + p(1)p(5) + p(2)p(4) + p(3)p(3) + p(4)p(2) + p(5)p(1) + p(6) = 429.

8. Method 1
   
x
x
Let
=
= n.
44
45
Since x is non-negative, n is also non-negative.
If n = 0, then x is any integer from 0 to 44 1 = 43: a total of 44 values.

If n = 1, then x is any integer from 45 to 2 44 1 = 87: a total of 43 values.

If n = 2, then x is any integer from 2 45 = 90 to 3 44 1 = 131: a total of 42 values.

If n = k, then x is any integer from 45k to 44(k + 1) 1 = 44k + 43: a total of (44k + 43) (45k 1) = 44 k
values.
Thus, increasing n by 1 decreases the number of values of x by 1. Also the largest value of n is 43, in which case
x has only 1 value.
Therefore the number of non-negative integer values of x is 44 + 43 + + 1 = 21 (44 45) = 990.

Method 2


  
x
x
Let n be a non-negative integer such that
=
= n.
44
45
 
 
x
x
= n 44n x < 44(n + 1) and
= n 45n x < 45(n + 1).
Then
44
45
   
x
x
=
= n 45n x < 44(n + 1) 44n + n x < 44n + 44.
So
44
45
This is the case if and only if n < 44, and then x can assume exactly 44 n different values.

Therefore the number of non-negative integer values of x is

(44 0) + (44 1) + + (44 43) = 44 + 43 + + 1 = 21 (44 45) = 990.

61

Method 3
  
x
x
=
= n.
44
45
Then x = 44n + r where 0 r 43 and x = 45n + s where 0 s 44.
Let n be a non-negative integer such that

So n = r s. Therefore 0 n 43. Also r = n + s. Therefore n r 43.

Therefore the number of non-negative integer values of x is 44 + 43 + + 1 = 21 (44 45) = 990.

9. Working out the first few terms gives us an idea of how the given sequence develops:
n
1
2
3
4
5
6
7

s2n1
a
ba
b
2b a
3b a
5b 2a
8b 3a

s2n
b
2b a
3b a
5b 2a
8b 3a
13b 5a
21b 8a

It appears that the coefficients in the even terms form a Fibonacci sequence and, from the 5th term, every odd
term is a repeat of the third term before it.
These observations are true for the entire sequence since, for m 1, we have:
s2m+2
s2m+3
s2m+4

=
=
=

s2m+1 + s2m
s2m+2 s2m+1
s2m+3 + s2m+2

=
=

s2m
s2m+2 + s2m

So, defining F1 = 1, F2 = 2, and Fn = Fn1 + Fn2 for n 3, we have s2n = bFn aFn2 for n 3. Since a = 3
and b < 1000, none of the first five terms of the given sequence equal 2015. So we are looking for integer solutions
of bFn 3Fn2 = 2015 for n 3.

s6 = 3b 3 = 2015, has no solution.


s8 = 5b 6 = 2015, has no solution.
s10 = 8b 9 = 2015 implies b = 253.

For n 6 we have b = 2015/Fn + 3Fn2 /Fn . Since Fn increases, we have Fn 13 and Fn2 /Fn < 1 for n 6.
Hence b < 2015/13 + 3 = 158. So the largest value of b is 253.

10. Method 1
We first show that X is uniquely defined for any given equilateral triangle ABC.
Let P be a point outside ABC such that its distances from AC and AB are in the ratio 1:2. By similar triangles,
any point on the line AP has the same property. Also any point between AP and AC has the distance ratio less
than 1:2 and any point between AP and AB has the distance ratio greater than 1:2.
B

P
Q
1

1
A

63

Let Q be a point outside ABC such that its distances from AC and BC are in the ratio 1:4. By an argument
similar to that in the previous paragraph, only the points on CQ have the distance ratio equal to 1:4.
Thus the only point whose distances to AC, AB, and BC are in the ratio 1:2:4 is the point X at which AP and
CQ intersect.
Scaling if necessary, we may assume that the actual distances of X to the sides of ABC are 1, 2, 4. Let h be the
height of ABC. Letting | | denote area, we have
|ABC| = 21 h AB and
|ABC| = |AXB| + |BXC| + |CXA| = 12 (2AB + 4BC + AC) = 12 AB 7.
So h = 7.

Draw a 7-layer grid of equilateral triangles each of height 1, starting with a single triangle in the top layer, then a
trapezium of 3 triangles in the next layer, a trapezium of 5 triangles in the next layer, and so on. The boundary
of the combined figure is ABC and X is one of the grid vertices as shown.
B

W
4

2
X
1
Y

There are 49 small triangles in ABC and 6.5 small triangles in AZXY . Hence, after rescaling so that the area of
AZXY is 13 cm2 , the area of ABC is 13 49/6.5 = 98 cm2 .
Method 2
Join AX, BX, CX. Since  Y AZ =  ZBW = 60, the quadrilaterals AZXY and BW XZ are similar. Let XY be
1 unit and AY be x. Then BZ = 2x.
B

2x
W
Z

X
1

By Pythagoras: in AXY , AX = 1 + x2 and in AXZ, AZ = x2 3. Hence BW = 2 x2 3.

Since AB = AC, Y C = x + x2 3.

By Pythagoras: in XY C, XC 2 = 1 +(x + x2 3)2 = 2x2 2 + 2x x2 3


and in XW C, W C 2 = 2x2 18 + 2x x2 3.


Since BA = BC, 2x + x2 3 = 2 x2 3 + 2x2 18 + 2x x2 3.




So 2x x2 3 = 2x2 18 + 2x x2 3.

63

Squaring gives 4x2 + x2 3 4x x2 3 = 2x2 18 + 2x x2 3, which simplifies to 3x2 + 15 = 6x x2 3.

5
Squaring again gives 9x4 + 90x2 + 225 = 36x4 108x2 . So 0 = 3x4 22x2 25 = (3x2 25)(x2 + 1), giving x = .
3
x 2
4
13
5
Hence, area AZXY = + x 3 = + = and
2
3
2 3
2 3

4 2
49
3
3 10
2
2
+
(2x + x 3) =
= .
area ABC =
4
4
3
3
3
13
49
Since the area of AZXY is 13 cm2 , the area of ABC is ( / ) 13 = 98 cm2 .
3 2 3
Method 3
Let DI be the line through X parallel to AC with D on AB and I on BC.
Let EG be the line through X parallel to BC with E on AB and G on AC.
Let F H be the line through X parallel to AB with F on AC and H on BC.
Let J be a point on AB so that HJ is parallel to AC.
Triangles XDE, XF G, XHI, BHJ are equilateral, and triangles XDE and BHJ are congruent.
B

E
Z

W
4

X
I

D
1
A

F Y G

The areas of the various equilateral triangles are proportional to the square of their heights. Let the area of
F XG = 1. Then, denoting area by | |, we have:

|DEX| = 4, |XHI| = 16, |AEG| = 9, |DBI| = 36, |F HC| = 25.

|ABC| = |AEG| + |F HC| + |DBI| |F XG| |DEX| |XHI| = 9 + 25 + 36 1 4 16 = 49.


|AZXY | = |AEG| 12 (|F XG| + |DEX|) = 9 12 (1 + 4) = 6.5.

Since the area of AZXY is 13 cm2 , the area of ABC is 2 49 = 98 cm2 .

Method 4
Consider the general case where XY = a, XZ = b, and XW = c.
B

W
Z
60
A

c
b
120

65

Projecting AY onto the


line through ZX gives AY sin
60 a cos 60 = b.
Hence AY = (a + 2b)/ 3. Similarly, AZ = (b + 2a)/ 3.

Letting | | denote area, we have


|AZXY |

=
=
=
=
=
=

Similarly, |CY XW | =
Hence |ABC| =

3
(2a2
6

3 2
6 (a

|Y AZ| + |Y XZ|
1
1

2 (AY )(AZ) sin 60 + 2 ab sin 120

3
4 ((AY )(AZ) + ab)

3
12 ((a + 2b)(b + 2a) +

3
2
2
12 (2a + 2b + 8ab)

3 2
(a + b2 + 4ab)
6

+ c2 + 4ac) and |BW XZ| =

+ 2b2 + 2c2 + 4ab + 4ac + 4bc) =

3 2
6 (b

3
(a
3

3ab)

+ c2 + 4bc).

+ b + c)2 .

So |ABC|/|AZXY | = 2(a + b + c)2 /(a2 + b2 + 4ab).

Letting a = k, b = 2k, c = 4k, and |AZXY | = 13 cm2 , we have |ABC| = 26(49k 2)/(k 2 + 4k 2 + 8k 2 ) = 98 cm2 .

Investigation
Method 4 gives |ABC|/|AZXY | = 2(a + b + c)2 /(a2 + b2 + 4ab).
Alternatively, as in Method 3,

|ABC| = |AEG| + |F HC| + |DBI| |F XG| |DEX| |XHI|


= (a + b)2 + (a + c)2 + (b + c)2 a2 b2 c2 = (a + b + c)2 .

Also |AZXY | = |AEG| 12 (|F XG| + |DEX|)


= (a + b)2 12 (a2 + b2 )
= 2ab + 12 (a2 + b2 ).

So |ABC|/|AZXY | = 2(a + b + c)2 /(a2 + b2 + 4ab).

10

65

AUSTRALIAN INTERMEDIATE MATHEMATICS OLYMPIAD


STATISTICS
DISTRIBUTION OF AWARDS/SCHOOL YEAR
Number of Awards

Year

Number of
Students

Prize

High
Distinction

Distinction

Credit

Participation

341

17

39

86

196

414

45

61

99

201

10

462

11

52

89

139

171

Other

221

16

41

151

Total

1438

26

123

205

365

719

NUMBER OF CORRECT ANSWERS QUESTIONS 18


Year

Number Correct/Question
1

119

231

282

164

128

79

48

50

144

298

347

224

187

138

82

86

10

176

341

377

297

219

208

103

104

Other

66

132

176

81

75

34

33

21

Total

505

1002

1182

766

609

459

266

261

MEAN SCORE/QUESTION/SCHOOL YEAR


Mean Score
Year

Number of
Students

Question

Overall Mean

18

10

341

10.1

0.5

0.2

10.9

414

11.8

0.9

0.5

13.2

10

462

13.0

1.1

0.6

14.6

Other

221

8.6

0.4

0.2

9.3

All Years

1438

11.3

0.8

0.4

12.5

67

AUSTRALIAN INTERMEDIATE MATHEMATICS OLYMPIAD


RESULTS
Name

School

Year

Score

Prize
Matthew Cheah

Penleigh and Essendon Grammar School, VIC

10

35

Puhua Cheng

Raffles Institution, Singapore

35

Ariel Pratama
Junaidi

Anglo-Chinese School, Singapore

10

35

Evgeni Kayryakov

Childrens Academy 21st Century, Bulgaria

35

Wei Khor Jun

Raffles Institution, Singapore

35

Jack Liu

Brighton Grammar, VIC

35

Jerry Mao

Caulfield Grammar School, Wheelers Hill,VIC

35

Liao Meng

Anglo-Chinese School, Singapore

10

35

Nguyen Hoai Nam

Anglo-Chinese School, Singapore

10

35

Aloysius Ng Yangyi

Raffles Institution, Singapore

35

Kohsuke Sato

Christ Church Grammar School, WA

10

35

Yuelin Shen

Scotch College, WA

10

35

Chen Tan Xu

Raffles Institution, Singapore

35

Kit Victor Loh Wai

Raffles Institution, Singapore

35

Jianzhi Wang

Raffles Institution, Singapore

35

Zhe Xin

Raffles Institution, Singapore

35

Austin Zhang

Sydney Grammar School, NSW

10

35

Yu Zhiqiu

Anglo-Chinese School, Singapore

10

35

Lin Zien

Anglo-Chinese School, Singapore

35

Bobby Dey

James Ruse Agricultural High School, NSW

10

34

Goh Ethan

Raffles Institution, Singapore

34

Yulong Guo

Hwa Chong Institution, Singapore

34

Edwin Winata
Hartanto

Anglo-Chinese School, Singapore

10

34

Hristo Papazov

Childrens Academy 21st Century, Bulgaria

10

34

Zhang Yansheng

Chung Cheng High School, Singapore

34

Guowen Zhang

St Joseph's College, QLD

34

HIGH DISTINCTION
Ivan Ganev

Childrens Academy 21st Century, Bulgaria

10

33

Theodore Leebrant

Anglo-Chinese School, Singapore

33

Yu Peng Ng

Hwa Chong Institution, Singapore

33

Cheng Shi

Hwa Chong Institution, Singapore

33

Kean Tan Wee

Raffles Institution, Singapore

33

Sharvil Kesarwani

Merewether High School, NSW

32

Hong Rui Benjamin


Lee

Hwa Chong Institution, Singapore

10

32

67

Name

School

Year

Score

Chenxu Li

Raffles Institution, Singapore

32

William Li

Barker College, NSW

32

Han Yang

Hwa Chong Institution, Singapore

32

Stanley Zhu

Melbourne Grammar School, VIC

32

Anand Bharadwaj

Trinity Grammar School, VIC

31

William Hu

Christ Church Grammar School, WA

31

Xianyi Huang

Baulkham Hills High School, NSW

10

31

Wanzhang Jing

James Ruse Agricultural High School, NSW

10

31

Yuhao Li

Hwa Chong Institution, Singapore

31

Steven Lim

Hurlstone Agricultural High School, NSW

31

John Min

Baulkham Hills High School, NSW

31

Elliott Murphy

Canberra Grammar School, ACT

10

31

Longxuan Sun

Hwa Chong Institution, Singapore

31

Boyan Wang

Hwa Chong Institution, Singapore

31

Sean Zammit

Barker College, NSW

10

31

Atul Barman

James Ruse Agricultural High School, NSW

10

30

Atanas Dinev

Childrens Academy 21st Century, Bulgaria

30

Bill Hu

James Ruse Agricultural High School, NSW

10

30

Phillip Huynh

Brisbane State High School, QLD

10

30

Wei Ci William Kin

Hwa Chong Institution, Singapore

10

30

Yang Lee Ker

Raffles Institution, Singapore

30

Forbes Mailler

Canberra Grammar School, ACT

30

Moses Mayer

Surya Institute, Indonesia

30

Zlatina Mileva

Childrens Academy 21st Century, Bulgaria

30

Kirill Saulov

Brisbane Grammar School, QLD

10

30

Yuxuan Seah

Raffles Institution, Singapore

30

Kieran Shivakumaarun Sydney Boys High School, NSW

10

30

Liang Tan Xue

Raffles Institution, Singapore

10

30

Nicholas Tanvis

Anglo-Chinese School, Singapore

30

An Aloysius Wang

Hwa Chong Institution, Singapore

10

30

William Wang

Queensland Academy for Science,


Mathematics and Technology, QLD

10

30

Joshua Welling

Melrose High School, ACT

30

Seung Hoon Woo

Hwa Chong Institution, Singapore

10

30

Chen Yanbing

Methodist Girls' School, Singapore

10

30

Guangxuan Zhang

Raffles Institution, Singapore

10

30

Chi Zhang Yu

Raffles Institution, Singapore

30

Keer Chen

Presbyterian Ladies College, NSW

10

29

Linus Cooper

James Ruse Agricultural High School, NSW

29

Liam Coy

Sydney Grammar School, NSW

29

69

Name

School

Year

Score

Rong Dai Xiang

Raffles Institution, Singapore

29

Hong Pei Goh

Hwa Chong Institution, Singapore

10

29

Tianjie Huang

Hwa Chong Institution, Singapore

29

Ricky Huang

James Ruse Agricultural High School, NSW

10

29

Tianjie Huang

Hwa Chong Institution, Singapore

29

Yu Jiahuan

Raffles Girls' School, Singapore

29

Tony Jiang

Scotch College, VIC

10

29

Charles Li

Camberwell Grammar School, Vic

29

Steven Liu

James Ruse Agricultural High School, NSW

10

29

Hilton Nguyen

Sydney Technical High School, NSW

29

James Nguyen

Baulkham Hills High School, NSW

10

29

Trung Nguyen

Penleigh and Essendon Grammar School,


VIC

10

29

James Phillips

Canberra Grammar School, ACT

29

Ryan Stocks

Radford College, ACT

29

Hadyn Tang

Trinity Grammar School, VIC

29

Stanve Avrilium
Widjaja

Surya Institute, Indonesia

29

Wang Yihe

Anglo-Chinese School, Singapore

29

Sun Yue

Raffles Girls' School, Singapore

29

Wang Beini

Raffles Girls' School, Singapore

10

28

Chwa Channe

Raffles Girls' School, Singapore

28

Keiran Hamley

All Saints Anglican School, QLD

10

28

Lee Shi Hao

Anglo-Chinese School, Singapore

28

Zhu Jiexiu

Raffles Girls' School, Singapore

28

Jodie Lee

Seymour College, SA

10

28

Yu Hsin Lee

Hwa Chong Institution, Singapore

28

Phillip Liang

James Ruse Agricultural High School, NSW

28

Anthony Ma

Shore School, NSW

10

28

Dzaki Muhammad

Surya Institute, Indonesia

28

Daniel Qin

Scotch College, VIC

10

28

Sang Ta

Randwick Boys High School, NSW

28

Ruiqian Tong

Presbyterian Ladies College, VIC

28

Hu Xing Yi

Methodist Girls' School, Singapore

10

28

Zhao Yiyang

Methodist Girls' School, Singapore

10

28

Claire Yung

Lyneham High School, ACT

10

28

Xuan Ang Ben

Raffles Institution, Singapore

27

Hantian Chen

James Ruse Agricultural High School, NSW

10

27

Jasmine Jiawei Chen

Pymble Ladies College, NSW

10

27

Harry Dinh

James Ruse Agricultural High School, NSW

10

27

Tan Jian Yee

Chung Cheng High School, Singapore

10

27

69

Name

School

Year

Score

Daniel Jones

All Saints Anglican Senior School, QLD

10

27

Winfred Kong

Hwa Chong Institution, Singapore

10

27

Adrian Law

James Ruse Agricultural High School, NSW

10

27

Jason Leung

James Ruse Agricultural High School, NSW

27

Sabrina Natashya
Liandra

Surya Institute, Indonesia

27

Angela (Yunyun) Ran

The Mac.Robertson Girls High School, VIC

27

Yi Shen Xin

Raffles Institution, Singapore

27

Peter Tong

Yarra Valley Grammar, VIC

27

Jordan Truong

Sydney Technical High School, NSW

10

27

Andrew Virgona

Concord High School, NSW

27

Tommy Wei

Scotch College, VIC

27

Tianyi Xu

Sydney Boys High School, NSW

27

Wu Zhen

Anglo-Chinese School, Singapore

10

27

Gordon Zhuang

Sydney Boys High School, NSW

27

Zhou Zihan

Raffles Girls' School, Singapore

27

Amit Ben-Harim

McKinnon Secondary College, VIC

26

Hu Chen

The King's School, NSW

10

26

Merry Xiao Die Chu

North Sydney Girls' High School, NSW

26

Li Haocheng

Anglo-Chinese School, Singapore

26

William Hu

Rossmoyne Senior High School, WA

10

26

Laeeque Jamdar

Baulkham Hills High School, NSW

26

Yasiru Jayasoora

James Ruse Agricultural High School, NSW

26

Arun Jha

Perth Modern School, WA

10

26

Tony Li

Sydney Boys High School, NSW

10

26

Zefeng Jeff Li

Glen Waverley Secondary College, VIC

26

Adrian Lo

Newington College, NSW

26

Lionel Maizels

Norwood Secondary College, VIC

26

Marcus Rees

Taroona High School, TAS

26

Elva Ren

Presbyterian Ladies College, VIC

10

26

Aidan Smith

All Saints' College, WA

26

Jacob Smith

All Saints' College, WA

26

Keane Teo

Hwa Chong Institution, Singapore

10

26

Jeffrey Wang

Shore School, NSW

10

26

Xinlu Xu

Presbyterian Ladies College, Sydney, NSW

10

26

Jason (Yi) Yang

James Ruse Agricultural High School, NSW

26

Shukai Zhang

Hwa Chong Institution, Singapore

10

26

Yanjun Zhang

Hwa Chong Institution, Singapore

26

Kevin Zhu

James Ruse Agricultural High School, NSW

10

26

Jonathan Zuk

Elwood College, VIC

26

THE
2015SENIOR
AMOC SENIOR
CONTEST
AMOC
CONTEST
Tuesday, 11 August 2015
Time allowed: 4 hours
No calculators are to be used.
Each question is worth seven points.
1. A number is called k-addy if it can be written as the sum of k consecutive positive
integers. For example, the number 9 is 2-addy because 9 = 4 + 5 and it is also 3-addy
because 9 = 2 + 3 + 4.
(a) How many numbers in the set {1, 2, 3, . . . , 2015} are simultaneously 3-addy,
4-addy and 5-addy?
(b) Are there any positive integers that are simultaneously 3-addy, 4-addy, 5-addy
and 6-addy?
2. Consider the sequence a1 , a2 , a3 , . . . defined by a1 = 1 and
am+1 =

1a1 + 2a2 + 3a3 + + mam


am

for m 1.

Determine the largest integer n such that an < 1 000 000.


3. A group of students entered a mathematics competition consisting of five problems.
Each student solved at least two problems and no student solved all five problems.
For each pair of problems, exactly two students solved them both.
Determine the minimum possible number of students in the group.
4. Let ABCD be a rectangle with AB > BC. Let E be the point on the diagonal AC
such that BE is perpendicular to AC. Let the circle through A and E whose centre
lies on the line AD meet the side CD at F .
Prove that BF bisects the angle AF C.
5. For a real number x, let x be the largest integer less than or equal to x.
Find all prime numbers p for which there exists an integer a such that
     
 
a
2a
3a
pa
+
+
+ +
= 100.
p
p
p
p

c 2015 Australian Mathematics Trust




71

THE 2015 AMOC SENIOR CONTEST


Solutions and cumulative marking scheme
c 2015 Australian

Mathematics
Trust
AMOC
SENIOR

CONTEST SOLUTIONS

1. A number is called k-addy if it can be written as the sum of k consecutive positive integers.
For example, the number 9 is 2-addy because 9 = 4 + 5 and it is also 3-addy because
9 = 2 + 3 + 4.
(a) How many numbers in the set {1, 2, 3, . . . , 2015} are simultaneously 3-addy, 4-addy
and 5-addy?
(b) Are there any positive integers that are simultaneously 3-addy, 4-addy, 5-addy and
6-addy?

Solution (Angelo Di Pasquale)


Since (a 1) + a + (a + 1) = 3a, the 3-addy numbers are precisely those that are divisible
by 3 and greater than 3.
1
Since (a 2) + (a 1) + a + (a + 1) + (a + 2) = 5a, the 5-addy numbers are precisely those
that are divisible by 5 and greater than 10.
Since (a 1) + a + (a + 1) + (a + 2) = 4a + 2, the 4-addy numbers are precisely those that
are congruent to 2 modulo 4 and greater than 6.
2
(a) From the observations above, a positive integer is simultaneously 3-addy, 4-addy and
5-addy if and only if it is divisible by 3, divisible by 5, divisible by 2, and not divisible
by 4. Such numbers are of the form 30m, where m is a positive odd integer.
3
Since 67 30 = 2010, the number of elements of the given set that are simultaneously
3-addy, 4-addy and 5-addy is 68
4
2 = 34.
(b) Since (a 2) + (a 1) + a + (a + 1) + (a + 2) + (a + 3) = 6a + 3, all 6-addy numbers
are necessarily odd.
5
On the other hand, we have already deduced that all 4-addy numbers are even.

Therefore, there are no numbers that are simultaneously 4-addy and 6-addy.

72

2. Consider the sequence a1 , a2 , a3 , . . . defined by a1 = 1 and


am+1 =

1a1 + 2a2 + 3a3 + + mam


am

for m 1.

Determine the largest integer n such that an < 1 000 000.

Solution 1 (Norman Do)


First, we note that all terms of the sequence are positive rational numbers. Below, we
rewrite the defining equation for the sequence in both its original form and with the value
of m shifted by 1.
am am+1 = 1a1 + 2a2 + 3a3 + + mam
am+1 am+2 = 1a1 + 2a2 + 3a3 + + mam + (m + 1)am+1

Subtracting the first equation from the second yields


am+1 am+2 am am+1 = (m + 1)am+1

am+2 am = m + 1,

for all m 1.

So for m = 2k + 1 an odd positive integer,


a2k+1 a1 = (a2k+1 a2k1 ) + (a2k1 a2k3 ) + + (a3 a1 )
= 2k + (2k 2) + + 2
= 2 [k + (k 1) + + 1]
= k(k + 1).

Similarly, for m = 2k an even positive integer,


a2k a2 = (a2k a2k2 ) + (a2k2 a2k4 ) + + (a4 a2 )
= (2k 1) + (2k 3) + + 3

= k 2 1.

Using a1 = 1 and a2 = 1, we obtain the formula

m2 ,
if m is even,
4
am =
2 +3
m

4 , if m is odd.
If m is odd, then

am+1 am =

2m 2
(m + 1)2 m2 + 3

=
,
4
4
4

am+1 am =

2m + 4
(m + 1)2 + 3 m2

=
.
4
4
4

and if m is even, then


6

In particular, it follows that a2 < a3 < a4 < . Since a2000 = 1 000 000, the largest
integer n such that an < 1 000 000 is 1999.
7
2

73

Solution 2 (Angelo Di Pasquale)


We will prove by induction that a2k1 = k 2 k + 1 and a2k = k 2 for each positive integer
k. The base case k = 1 is true since a1 = a2 = 1. Now assume that the two formulas hold
for k = 1, 2, . . . , n. We will show that they also hold for k = n + 1.
2
Consider the following sequence of equalities.
2n


iai =

i=1

=
=

n

i=1
n

i=1
n

i=1

n


(2i 1) a2i1 +

2i a2i

i=1

(2i 1) (i2 i + 1) +

n


(2i) (i2 )

(by the inductive assumption)

i=1

4i3 3i2 + 3i 1
3i3 + (i 1)3

i=1
n


=3
=

n


i +

i=1
3n2 (n

+
4

n1


i3

i=1
1)2

(n 1)2 n2
4

= n (n + n + 1)
= (n2 + n + 1)a2n

n


n2 (n + 1)2 
since
i3 =
4
i=1

(by the inductive assumption)

It follows that
a2n+1 =

1a1 + 2a2 + 3a3 + + 2na2n


= n2 + n + 1.
a2n

Now consider the following sequence of equalities, which uses the facts derived above that

2 2
2
state that 2n
i=1 iai = n (n + n + 1) and a2n+1 = n + n + 1.
2n+1


iai = n2 (n2 + n + 1) + (2n + 1)a2n+1

i=1

= n2 a2n+1 + (2n + 1)a2n+1


= (n + 1)2 a2n+1

It follows that
a2n+2 =

1a1 + 2a2 + 3a3 + + (2n + 1)a2n+1


= (n + 1)2 .
a2n+1

So we have shown that the two formulas a2k1 = k 2 k + 1 and a2k = k 2 hold for
k = 1, 2, . . . , n + 1. This completes the induction and the rest of the proof follows Solution
1.
7

3
74

3. A group of students entered a mathematics competition consisting of five problems. Each


student solved at least two problems and no student solved all five problems. For each
pair of problems, exactly two students solved them both.
Determine the minimum possible number of students in the group.

Solution (Norman Do)


It is possible that the group comprised six students, as demonstrated by the following
example.
Student 1 solved problems 1, 2, 3, 4.

Student 4 solved problems 2, 4, 5.

Student 2 solved problems 1, 2, 3, 5.

Student 5 solved problems 3 and 4.

Student 3 solved problems 1, 4, 5.

Student 6 solved problems 3 and 5. 2

Suppose that a students solved 4 problems, b students solved 3 problems, and c students

solved 2 problems. Therefore, a students solved 42 = 6 pairs of problems, b students


solved 32 = 3 pairs of problems and c students solved 22 = 1 pair of problems. Since we
have shown an example in which the number of students in the group is 6, let us assume
that a + b + c 5.

There are 52 = 10 pairs of problems altogether and, for each pair of problems, exactly
two students solved them both, so we must have
6a + 3b + c = 20.

Reading the above equation modulo 3 yields c 2 (mod 3). If c 5, then we have
a + b + c 6, contradicting our assumption. Therefore, we must have c = 2 and 2a + b = 6.
For a + b + c 5, the only solution is given by (a, b, c) = (3, 0, 2).
5
However, it is impossible for 3 students to have solved 4 problems each. That would mean
that each of the 3 students did not solve exactly 1 problem. So there would exist a pair of
problems for which 3 students solved them both, contradicting the required conditions.
In conclusion, the minimum possible number of students in the group is 6.

75

4. Let ABCD be a rectangle with AB > BC. Let E be the point on the diagonal AC such
that BE is perpendicular to AC. Let the circle through A and E whose centre lies on the
line AD meet the side CD at F .
Prove that BF bisects the angle AF C.

Solution 1 (Alan Offer)


Let the circle through A and E whose centre lies on AD meet the line AD again at H.

E
D

H
Since F D is the altitude of the right-angled triangle AF H, we have AF H ADF .
Since triangles AEH and ADC are right-angled with a common angle at A, we have
AEH ADC. Since BE is the altitude of the right-angled triangle ABC, we have
ABC AEB. These three pairs of similar triangles lead respectively to the three
pairs of equal ratios
AF
AH
=
AD
AF

AE
AH
=
AD
AC

AB
AC
=
.
AE
AB

Putting these together, we have


AF 2 = AD AH = AC AE = AB 2 .

So triangle BAF is isosceles and we have


AF B = ABF = 90 CBF = CF B,
where the last equality uses the angle sum in triangle BCF . Since AF B = CF B, we
have proven that BF bisects the angle AF C.
7

Solution 2 (Angelo Di Pasquale)


First, note that the circumcircles of triangle AEF and triangle BEC are tangent to the
line AB at A and B, respectively. Considering the power of the point A with respect to
the circumcircle of triangle BEC, we have
AB 2 = AE AC.
5

76

Next, by the alternate segment theorem and the fact that AB  CD, we have EF A =
EAB = ECF . Hence, by the alternate segment theorem again, the circumcircle of
triangle EF C is tangent to the line AF at F . Considering the power of the point A with
respect to the circumcircle of triangle EF C, we have
AF 2 = AE AC.

Comparing the two equations above, we deduce that AB = AF .

Hence, AF B = ABF = CF B, as required.

Solution 3 (Ivan Guo)


Let the circle through A and E whose centre lies on AD meet the line AD again at H.
Then AEB = AEH = 90 , so the points B, E and H are collinear. Consider the
inversion f with centre A and radius AF .
2
Note that the circumcircle of the cyclic quadrilateral AEF H must map to a line parallel
to AB through F . Thus,
f (circle AEF H) = line CD.
4
This immediately gives f (C) = E and f (H) = D. Now the circumcircle of ABCD must
map to a line passing through f (C) = E and f (D) = H. This implies that f (B) = B.
Hence, AB = AF .
6
Therefore, we have BF C = ABF = AF B.

Solution 4 (Chaitanya Rao)


Consider the following chain of equalities.
DF 2 = AD DH
= AD (AH AD)


AB 2
AD
= AD
BC
= AB 2 AD2

(ADF F DH)

(ABC HAB)

(AD = BC)

Hence, AB 2 = DF 2 + AD2 = AF 2 , which implies that AB = AF .

So triangle BAF is isosceles and we have


AF B = ABF = 90 CBF = CF B,
where the last equality uses the angle sum in triangle BCF . Since AF B = CF B, we
have proven that BF bisects the angle AF C.
7

77

5. For a real number x, let x be the largest integer less than or equal to x.
Find all prime numbers p for which there exists an integer a such that
 
     
2a
3a
pa
a
+
+
+ +
= 100.
p
p
p
p
Solution 1 (Norman Do)
The possible values for p are 2, 5, 17 and 197.
We divide the problem into the following two cases.
The number a is divisible by p.
If we write a = kp, the equation becomes
a(p + 1)
= 100
2

kp(p + 1) = 200.

So both p and p + 1 are positive divisors of 200. However, one can easily see that
there are no such primes p.
1
The number a is not divisible by p.
For a real number x, let {x} = x x. Then we may write the equation as

      
 
a 2a 3a
pa
a
2a
3a
pa
+
+
+ +

+
+
+ +
= 100.
p
p
p
p
p
p
p
p
Summing the terms of the arithmetic progression on the left-hand side yields
     
 
a
2a
3a
pa
a(p + 1)

+
+
+ +
= 100.
3
2
p
p
p
p
     3a 
 pa 
We will prove that the sequence of numbers ap , 2a
is a rearrangep , p ,..., p
p1
0 1 2
ment of the sequence of numbers p , p , p , . . . , p .
4
 ka 
p1
0 1 2
Observe that if k is a positive integer, then p is one of the numbers p , p , p , . . . , p .
     3a 
 
So it suffices to show that no two of the numbers ap , 2a
, p , . . . , pa
are equal.
p
p
 ia   ja 
Suppose for the sake of contradiction that p = p , where 1 i < j p. Then

a(ji)
ia
must be an integer. It follows that either a is divisible by p or j i is
p =
p
divisible by p. However, since we have assumed that a is not divisible by p and that
1 i < j p, we obtain the desired contradiction. Hence, we may conclude that the
     3a 
 pa 
sequence of numbers ap , 2a
is a rearrangement of the sequence
p , p ,..., p
p1
0 1 2
of numbers p , p , p , . . . , p .
5
ja
p

We may now write the equation as


a(p + 1) p 1

= 100
2
2

(a 1)(p + 1) = 198.

Therefore, p + 1 is a positive divisor of 198 in other words, one of the numbers


1, 2, 3, 6, 9, 11, 18, 22, 33, 66, 99, 198.
Since p is a prime, it follows that p must be equal to 2, 5, 17 or 197. This leads to
the possible solutions (p, a) = (2, 67), (5, 34), (17, 12), (197, 2). All four of these pairs
satisfy the given equation with a not divisible by p, so we obtain p = 2, 5, 17, 197. 7
7

78

Solution 2 (Ivan Guo, Angelo Di Pasquale and Ian Wanless)


The possible values for p are 2, 5, 17 and 197.
The case where a is divisible by p is handled in the same way as Solution 1.

Furthermore, one can check that the pair (p, a) = (2, 67) satisfies the conditions of the
problem. So assume that p is an odd prime and that a is not divisible by p.
For any integers 1 r, s p 1 with r + s = p, we have p  ra and p  sa. Therefore,
   
   
ar
as
ar
as
ar as
ar
as
1+
1<
+
<
+
a2<
+
< a.
p
p
p
p
p
p
p
p
   
as
But since ar
is an integer, we conclude that
p + p


  
as
ar
+
= a 1.
p
p

The numbers {1, 2, . . . , p 1} can be partitioned into

p1
2

3
pairs whose sum is p.

Using the equation above for each such pair and substituting into the original equation,
we obtain


p1
(a 1) + a = 100

(a 1)(p + 1) = 198.
6
2
Therefore, p + 1 is a positive divisor of 198 in other words, one of the numbers
1, 2, 3, 6, 9, 11, 18, 22, 33, 66, 99, 198.
Since p is a prime, it follows that p must be equal to 2, 5, 17 or 197. This leads to the
possible solutions (p, a) = (2, 67), (5, 34), (17, 12), (197, 2). All four of these pairs satisfy
the given equation with a not divisible by p, so we obtain p = 2, 5, 17, 197.
7

79

AMOC SENIOR CONTEST RESULTS


Name

School

Year

Score

Prize
Yong See Foo

Nossal High School VIC

11

35

Kevin Xian

James Ruse Agricultural High School NSW

11

35

Wilson Zhipu Zhao

Killara High School NSW

11

35

Ilia Kucherov

Westall Secondary College VIC

11

34

Seyoon Ragavan

Knox Grammar School NSW

11

34

High Distinction
Jongmin Lim

Killara High School NSW

11

32

Matthew Cheah

Penleigh and Essendon Grammar School VIC

10

29

Jerry Mao

Caulfield Grammar School (Wheelers Hill) VIC

29

Distinction
Alexander Barber

Scotch College VIC

11

28

Michelle Chen

Methodist Ladies College VIC

11

28

Thomas Baker

Scotch College VIC

11

27

Linus Cooper

James Ruse Agricultural High School NSW

27

Steven Lim

Hurlstone Agricultural High School NSW

27

Eric Sheng

Newington College NSW

11

27

William Song

Scotch College VIC

11

27

Jack Liu

Brighton Grammar VIC

26

Leo Li

Christ Church Grammar School WA

11

25

Bobby Dey

James Ruse Agricultural High School NSW

10

24

Michael Robertson

Dickson College ACT

11

24

Charles Li

Camberwell Grammar School VIC

22

Isabel Longbottom

Rossmoyne Senior High School WA

10

22

Guowen Zhang

St Josephs College, Gregory Terrace QLD

22

80

AMOC SENIOR CONTEST STATISTICS


SCORE DISTRIBUTION/PROBLEM
Number of Students/Score

Problem
Number

46

31

6.2

16

15

25

4.1

25

11

25

3.5

54

13

15

1.5

65

1.2

81

Mean

SCHOOL
OF EXCELLENCE
2014 AMOCAMOC
School of
Excellence
The 2014 AMOC School of Excellence was held 110 December at Newman College,
University of Melbourne. The main qualifying exams to be invited to this are the AIMO
and the AMOC Senior Contest.
This year AMOC went ahead with a new initiative in an attempt to widen the net of
identification. In particular, any prize winner in the Australian Mathematics Competition
(AMC) would be given free entry into the AIMO. In this way we hoped to identify top
students from among schools that may not normally enter students in the AIMO. This
turned out to be quite successful and prompted us to increase the number of invitations
we normally make. Consequently, 28 students from around Australia attended the school.
A further student from New Zealand also attended.
The students are divided into a senior group and a junior group. There were 17 junior
students, 16 of whom were attending for the first time. There were 12 students making
up the senior group.
The program covered the four major areas of number theory, geometry, combinatorics
and algebra. Each day would start at 8am with lectures or an exam and go until 12
noon or 1pm. After a one-hour lunch break they would have a lecture at 2pm. At
4pm participants would usually have free time, followed by dinner at 6pm. Finally, each
evening would round out with a problem session, topic review, or exam review from 7pm
until 9pm.
Another new initiative we tried was to invite two of our more experienced senior students
to give a lecture. The rationale behind this is that teaching a subject is highly beneficial to the teacher because it can really solidify the foundations of the teachers own
understanding. A further fringe benefit is that they also get to hone their LATEX skills.
Alexander Gunning was assigned the senior inequalities lecture, and Seyoon Ragavan was
assigned the senior transformation geometry lecture. I did some trial runs with them
prior to the lectures so as to trouble shoot any problems, as well as to give them some
practice for the real thing. Overall it was successful, and I will likely try it again in the
future.
Many thanks to Andrew Elvey Price, Ivan Guo, Victor Khou, and Sampson Wong, who
served as live-in staff. Also my thanks go to Adrian Agisilaou, Norman Do, Patrick He,
Alfred Liang, Daniel Mathews, Konrad Pilch, Chaitanya Rao, and Mel Shu who assisted
in lecturing and marking.
Angelo Di Pasquale
Director of Training, AMOC

1
82

PARTICIPANTS AT THE 2014 AMOC SCHOOL OF EXCELLENCE


Name

School

Year

Seniors
Thomas Baker

Scotch College VIC

10

Matthew Cheah

Penleigh and Essendon Grammar School VIC

Yong See Foo

Nossal High School VIC

10

Alexander Gunning

Glen Waverley Secondary College VIC

11

Leo Li

Christ Church Grammar School WA

10

Allen Lu

Sydney Grammar School NSW

11

Seyoon Ragavan

Knox Grammar School NSW

10

Kevin Shen

St Kentigern College NZ

11*

Yang Song

James Ruse Agricultural High School NSW

11

Kevin Xian

James Ruse Agricultural High School NSW

10

Jeremy Yip

Trinity Grammar School VIC

11

Henry Yoo

Perth Modern School WA

11

Juniors
Adam Bardrick

Whitefriars College VIC

Rachel Hauenschild

Kenmore State High School QLD

William Hu

Christ Church Grammar School WA

Shivasankaran Jayabalan

Rossmoyne Senior High School WA

Tony Jiang

Scotch College VIC

Sharvil Kesarwani

Merewether High School NSW

Ilia Kucherov

Westall Secondary College VIC

10

Adrian Law

James Ruse Agricultural High School NSW

Charles Li

Camberwell Grammar School VIC

Jack Liu

Brighton Grammar School VIC

Isabel Longbottom

Rossmoyne Senior High School WA

Hilton Nguyen

Sydney Technical High School NSW

Madeline Nurcombe

Cannon Hill Anglican College QLD

10

Zoe Schwerkolt

Fintona Girls' School VIC

10

Katrina Shen

James Ruse Agricultural High School NSW

Eric Sheng

Newington College NSW

10

Wen Zhang

St Joseph's College QLD

* Equivalent to year 10 in Australia.

83

THE 2015 AUSTRALIAN MATHEMATICAL


OLYMPIAD
AUSTRALIAN
MATHEMATICAL OLYMPIAD
DAY 1
Tuesday, 10 February 2015
Time allowed: 4 hours
No calculators are to be used.
Each question is worth seven points.
1. Define the sequence a1 , a2 , a3 , . . . by a1 = 4, a2 = 7, and
an+1 = 2an an1 + 2,

for n 2.

Prove that, for every positive integer m, the number am am+1 is a term of the sequence.
2. For each positive integer n, let s(n) be the sum of its digits. We call a number nifty if it
can be expressed as n s(n) for some positive integer n.
How many positive integers less than 10,000 are nifty?
3. Let S be the set of all two-digit numbers that do not contain the digit 0. Two numbers
in S are called friends if their largest digits are equal and the difference between their
smallest digits is 1. For example, the numbers 68 and 85 are friends, the numbers 78
and 88 are friends, but the numbers 58 and 75 are not friends.
Determine the size of the largest possible subset of S that contains no two numbers that
are friends.
4. Let be a fixed circle with centre O and radius r. Let B and C be distinct fixed points
on . Let A be a variable point on , distinct from B and C. Let P be the point such
that the midpoint of OP is A. The line through O parallel to AB intersects the line
through P parallel to AC at the point D.
(a) Prove that, as A varies over the points of the circle (other than B or C), D lies
on a fixed circle whose radius is greater than or equal to r.
(b) Prove that equality occurs in part (a) if and only if BC is a diameter of .

c 2015 Australian Mathematics Trust




84

OLYMPIAD
DAY 2
Wednesday, 11 February 2015
Time allowed: 4 hours
No calculators are to be used.
Each question is worth seven points.
5. Let ABC be a triangle with ACB = 90 . The points D and Z lie on the side AB such
that CD is perpendicular to AB and AC = AZ. The line that bisects BAC meets CB
and CZ at X and Y , respectively.
Prove that the quadrilateral BXY D is cyclic.
6. Determine the number of distinct real solutions of the equation
(x 1) (x 3) (x 5) (x 2015) = (x 2) (x 4) (x 6) (x 2014).
7. For each integer n 2, let p(n) be the largest prime divisor of n.
Prove that there exist infinitely many positive integers n such that



p(n + 1) p(n) p(n) p(n 1) > 0.

8. Let n be a given integer greater than or equal to 3. Maryam draws n lines in the plane
such that no two are parallel.
For each equilateral triangle formed by three of the lines, Maryam receives three apples.
For each non-equilateral isosceles triangle formed by three of the lines, she receives one
apple.
What is the maximum number of apples that Maryam can obtain?

c 2015 Australian Mathematics Trust




85

AUSTRALIAN MATHEMATICAL OLYMPIAD SOLUTIONS


1. For reference, the sequence a1 , a2 , a3 , . . . is given by a1 = 4, a2 = 7, and
an+1 = 2an an1 + 2,

for n 2.

Solution 1 (Linus Cooper, year 9, James Ruse Agricultural High School, NSW)
First we prove the following formula by induction.
an = n2 + 3,

for n 1.

We require two base cases to get started. The formula is true for n = 1 and n = 2
because a1 = 4 = 12 + 3 and a2 = 7 = 22 + 3.
For the inductive step, assume that the formula is true for n = k 1 and n = k.
Then for n = k + 1, we have
ak+1 = 2ak ak1 + 2
(given)
2
2
= 2(k + 3) ((k 1) + 3) + 2 (inductive assumption)
= k 2 + 2k + 4
= (k + 1)2 + 3.
Hence the formula is also true for n = k + 1. This completes the induction.
Using the formula, we calculate
am am+1 = (m2 + 3)((m + 1)2 + 3)
= (m2 + 3)(m2 + 2m + 4)
= m4 + 2m3 + 7m2 + 6m + 12
= (m2 + m + 3)2 + 3
= am2 +m+3 .
Hence am am+1 is a term of the sequence.

17

86

Solution 2 (Jack Liu, year 9, Brighton Grammar School, VIC)


The given rule for determining an+1 from an and an1 can be rewritten as
an+1 an = an an1 + 2,

for n 2.

Thus the difference between consecutive terms of the sequence increases by 2 as we


go up. Hence starting from a2 a1 = 3, we may write down the following.
a2 a1 = 3
a3 a2 = 5
a4 a3 = 7
..
.
am am1 = 2m 1
If we add all these equations together, we find that almost everything cancels on
the LHS, and we have
am a1 = 3 + 5 + + 2m 1.
Using the standard formula11 for summing an arithmetic progression, we find
(2m + 2)(m 1)
2
2
= m 1.

am a1 =

Since a1 = 4, it follows that am = m2 + 3 for each m 1.

It is now a simple matter to calculate

am am+1 = (m2 + 3)((m + 1)2 + 3)


= m4 + 2m3 + 7m2 + 6m + 12
= (m2 + m + 3)2 + 3.
Hence am am+1 = an , where n = m2 + m + 3. Therefore, am am+1 is a term of the
sequence.


1
1

The standard formula for the sum a + (a + d) + (a + 2d) + + (a + kd) is


a = 3, d = 2 and k = m 2.

18
87

(2a+kd)(k+1)
.
2

In our case

Solution 3 (Michael Robertson, year 11, Dickson College, ACT)


Starting from am and am+1 , let us compute the next few terms of the sequence in
terms of am and am+1 .
am+2 = 2am+1 am + 2
= 2(am+1 + 1) am
am+3 = 2am+2 am+1 + 2
= 2((2am+1 + 1) am ) am+1 + 2
= 3(am+1 + 2) 2am
Similarly, am+4 = 4(am+1 + 3) 3am .

We prove the following general formula by induction.


am+k = k(am+1 + k 1) (k 1)am ,

for k = 0, 1, 2, . . ..

(1)

By inspection, formula (1) is true for the base cases k = 0 and k = 1.


For the inductive step, assume that formula (1) is true for k = r 1 and k = r.
Then for k = r + 1, we have
am+r+1 = 2am+r am+r1 + 2
= 2(r(am+1 + r 1) (r 1)am )
((r 1)(am+1 + r 2) (r 2)am ) + 2
= (r + 1)(am+1 + r) ram .
Hence the formula is also true for k = r + 1. This completes the induction.
Let us substitute k = am into formula (1). We have
am+am = am (am+1 + am 1) (am 1)am
= am am+1 .
Hence am am+1 = ak , where k = m + am . Therefore, am am+1 is a term of the
sequence.

Comment This proof shows that the conclusion of the problem remains true for
any starting values a1 and a2 of the sequence, provided that m + am 0 for all m.
This is certainly true whenever a2 a1 0.

19
88

2. Solution 1 (Yang Song, year 12, James Ruse Agricultural High School, NSW)
Answer: 1000
For each positive integer n, let f (n) = n s(n). We seek the number of different
values that f (n) takes in the range from 1 up to 9999.
Lemma The function f has the following two properties.
(i) f (n + 1) = f (n) if the last digit of n is not a 9.
(ii) f (n + 1) > f (n) if the last digit of n is a 9.
Proof If the last digit of n is not a 9, then s(n + 1) = s(n) + 1. From this it easily
follows that f (n + 1) = f (n).
If the last digit of n is a 9, then suppose that the first k (k 1) digits from the
right-hand end of n are 9s, but the (k + 1)th digit from the right-hand end of n
is not a 9. In going from n to n + 1, the rightmost k digits all change from 9
to 0, and the (k + 1)th digit from the right-hand end of n increases by 1. Hence
s(n + 1) = s(n) 9k + 1, and so
f (n + 1) = n + 1 s(n + 1)
= n + 1 (s(n) 9k + 1)
= n s(n) + 9k
= f (n) + 9k
> f (n).

From the lemma it follows that


f (1) = f (2) = = f (9) < f (10) = f (11) = = f (19)
< f (20) = f (21) = = f (29)
..
.
< f (10000) = f (10001) = = f (10009)
< f (10010).
Since, f (9) = 0, f (10) = 9, f (10000) = 9999 and f (10010) = 10008, the required
number of different values of f in the range from 1 up to 9999 is simply equal
to the number of multiples of 10 from 10 up to 10000. The number of these is
10000 10 = 1000.


20
89

Solution 2 (Seyoon Ragavan, year 11, Knox Grammar School, NSW)


Suppose that the decimal representation of n is ak ak1 . . . a1 a0 where ak = 0. Then
we have the following.

n = 10k ak + 10k1 ak1 + + 101 a1 + a0


s(n) = ak + ak1 + + a1 + a0

n s(n) = (10k 1)ak + (10k1 1)ak1 + + (101 1)a1

If k 4, then n s(n) (10k 1)ak 9999. Note that 9999 is nifty, because if
n = 10000 then n s(n) = 9999. So it only remains to deal with positive integers
that are less than 9999.
If k 3, then n s(n) = 999a3 + 99a2 + 9a1 . It follows that the nifty numbers
less than 999 are precisely those numbers of the form 999a3 + 99a2 + 9a1 , where
a1 , a2 , a3 {0, 1, 2, . . . , 9}.
Lemma If 999a + 99b + 9c = 999d + 99e + 9f where a, b, c, d, e, f {0, 1, 2, . . . , 9},
then a = d, b = e and c = f .
Proof If 999a + 99b + 9c = 999d + 99e + 9f , then this can be rearranged as
111(a d) + 11(b e) + (c f ) = 0.

(1)

If a > d, then since b e 9 and c f 9, we have LHS(1) 111 99 9 > 0.


A similar contradiction is reached if a < d. Hence a = d and equation (1) becomes
11(b e) + (c f ) = 0.

(2)

If b > e, then since cf 9, we have LHS(2) 119 > 0. A similar contradiction


is reached if b < e. Hence b = e, from which c = f immediately follows.

Returning to the problem, the number of combinations of a1 , a2 , a3 is 103 = 1000.
The lemma guarantees that each of these combinations leads to a different nifty
number. However, we exclude the combination a1 = a2 = a3 = 0 because although
0 is nifty, it is not positive.
In summary, there are 999 nifty numbers from the case k 3, and one nifty number
from the case k 4. This yields a total of 1000 nifty numbers.


21
90

3. Solution 1 (Alexander Gunning, year 12, Glen Waverley Secondary College, VIC)
Answer: 45
Call a subset T of S friendless if no two numbers in T are friends. We visualise a
friendless subset T as follows. Draw a 9 9 square grid. If the number 10a + b is in
T we shade in the square that lies in the ath row and bth column.
First we exhibit a friendless subset T of size 45.
Let T consist of all two-digit numbers whose smaller digit is odd, as depicted in the
first diagram below. No two numbers in T are friends because their smaller digits
are both odd and hence cannot differ by 1. Since there are 45 shaded squares, we
have shown that |T | = 45 is possible.
1 2 3 4 5 6 7 8 9

1 2 3 4 5 6 7 8 9

1
2
3
4
5
6
7
8
9

1
2
3
4
5
6
7
8
9

Finally, we show that all friendless subsets T of S have at most 45 numbers.


To see this we pair up some of the squares in the 9 9 square grid as shown in the
second diagram above. Observe that the numbers in each pair are friends. Hence T
can have at most one number from each pair. Since there are 9 unpaired numbers
in the top row and 36 pairs, this shows that T has at most 45 numbers.


22
91

Solution 2 (Anand Bharadwaj, year 9, Trinity Grammar School, VIC)


Consider the following nine lines of numbers.

11
21, 22, 12
31, 32, 33, 23, 13
41, 42, 43, 44, 34, 24, 14
51, 52, 53, 54, 55, 45, 35, 25, 15
61, 62, 63, 64, 65, 66, 56, 46, 36, 26, 16
71, 72, 73, 74, 75, 76, 77, 67, 57, 47, 37, 27, 17
81, 82, 83, 84, 85, 86, 87, 88, 78, 68, 58, 48, 38, 28, 28
91, 92, 93, 94, 95, 96, 97, 98, 99, 89, 79, 69, 59, 49, 39, 29, 19
Each pair of neighbouring numbers on any given line are friends. So a subset T of S
that contains no friends cannot include consecutive numbers on any of these lines.
On the ith line there are exactly 2i 1 integers. The maximum number of integers
we can choose from the ith line without choosing neighbours is i. This shows that
T contains at most 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 = 45 integers.
Furthermore, |T | = 45 only if we choose exactly i numbers from the ith line without
choosing neighbours. There is only one way to do this, namely take every second
number starting from the left of each line.
It is still necessary to verify that T contains no friends because even some nonadjacent numbers from the same line, such as 31 and 23, can be friends. To do this,
note that the smaller digit of each number in T is odd. Thus for any pair of integers
in T , the difference between their smaller digits is even, and thus cannot be equal
to 1. Hence T contains no friends.

Comment This proof shows that there is exactly one subset of S of maximal size
that contains no friends.

23
92

4. Comment All solutions that were dependent on how the diagram was drawn received a penalty deduction of 1 point. The easiest way to avoid diagram dependence
was to use directed angles as in the three solutions we present here.
For any two lines m and n, the directed angle between them is denoted by (m, n).
This is the angle by which one may rotate m anticlockwise to obtain a line parallel
to n.22
Solution 1 (Yang Song, year 12, James Ruse Agricultural High School, NSW)
Let Q be the intersection of lines OC and P D. Since AC  P Q and A is the
midpoint of OP , it follows that C is the midpoint of OQ.
P

D
A

O
B

(a) We know OD  BA and DQ  AC. It follows that (OD, DQ) = (BA, AC),
which is fixed because A lies on . This implies D lies on a fixed circle, say,
through O and Q.
Let s be the radius of . Since the diameter is the largest chord length in a
circle, we have 2s OQ. Since OQ = 2OC = 2r we have s r, as desired. 

(b) From the preceding analysis, we have s = r if and only if OQ is a diameter of


. This is achieved if and only if OD DQ, which is equivalent to BA AC.
But the chords BA and AC of are perpendicular if and only if BC is a
diameter of .


22 For

more details on how to work with directed angles, see the section Directed angles in chapter 17 of
Problem Solving Tactics published by the AMT.

24
93

Solution 2 (Ilia Kucherov, year 11, Westall Secondary College, VIC)


Let E be the intersection of lines AC and OD. Since AC  P D and A is the
midpoint of OP , it follows that E is the midpoint of OD.
P

O
B

C
Q

(a) As in solution 1, we use directed angles.


Since OE  BA, we have (OE, EC) = (BA, AC), which is fixed because
A lies on . This implies that E lies on a fixed circle, say, through O and
C. Since E is the midpoint of OD, the point D is the image of E under the
dilation centred at O and of enlargement factor 2. The image of under the
dilation is a circle that is twice as large as . Thus D lies on .
Let s be the radius of . Then has radius 2s . But OC is a chord of ,
hence its length r is less than or equal to the diameter s of . Hence s r, as
required.

(b) From the preceding analysis we have s = r if and only if OC is a diameter of
. This is achieved if and only if OE EC, which is equivalent to BA AC.
But the chords BA and AC of are perpendicular if and only if BC is a
diameter of .


25
94

Solution 3 (Kevin Xian, year 11, James Ruse Agricultural High School, NSW)
The motivation for this solution comes from considering a couple of special positions
for A.
If A is diametrically opposite C, then D = O. If A is diametrically opposite B, then
this gives a second position for D. With this in mind, let X be the point on that
is diametrically opposite B, and let Y be the point such that X is the midpoint of
OY .
P

Y
X
O
B

(a) Using directed angles, we have


(OY, OD) = (BO, BA) (OD  BA)
= (AB, AO) (OB = OA)
= (OD, OP ). (BA  OD)
We also have OY = 2OX = 2OA = OP . Therefore, Y and P are symmetric
in the line OD. It follows that
(Y D, OD) = (OD, P D)
= (OD, OP ) + (OP, P D)
= (BA, OA) + (OA, AC)
(OD  BA and P D  AC)
= (BA, AC).
Hence (Y D, OD) is fixed because B, C and are fixed.
Since O and Y are also fixed points, it follows that D lies on a fixed circle,
say, that passes through O and Y .
Let s be the radius of . Then by the extended sine rule applied to ODY

26
95

we have
OY
sin (Y D, OD)
2r
=
sin (BA, AC)
2r.

2s =

Therefore s r, as required.

(b) From the preceding analysis, we have s = r if and only if


sin(BA, AC) = 1.
This is equivalent to BA AC, which occurs if and only if BC is the diameter
of .


27
96

5. This was the easiest problem of the competition. Of the 106 students who sat the
AMO, 77 found a complete solution. There were many different routes to a solution,
and we present some of them here.
Solution 1 (Zoe Schwerkolt, year 11, Fintona Girls School, VIC)
We are given ACZ is isosceles with AC = AZ. By symmetry, the angle bisector
at A is also the altitude from A, and so AX CZ. Thus CY A = 90 = CDA.
It follows that ADY C is a cyclic quadrilateral.
C

X
Y

x
A

x
D

Since ADY C is cyclic, we may set Y DC = Y AC = x, as in the diagram. Hence


BDY = 90 x.

(1)

But from the exterior angle sum in CAX, we have AXB = 90 + x, and so
Y XB = 90 + x.

(2)

Adding (1) and (2) yields


BDY + Y XB = 180 ,


and so BXY D is a cyclic quadrilateral.

28
97

Solution 2 (Alan Guo, year 12, Penleigh and Essendon Grammar School, VIC)
We deduce that ADY C is cyclic as in solution 1.
It follows that
AY D = ACD
= 90 DAC
= 90 BAC
= CBA.

(ADY C cyclic)
(angle sum CAD)
(angle sum ABC)

Hence AY D = XBD, and so BXY D is cyclic.


C

X
Y

29
98

Solution 3 (Wen Zhang, year 9, St Josephs College, QLD)


We are given ACZ is isosceles with AC = AZ. Hence by symmetry, the angle
bisector at A is both the median and the altitude from A. Thus Y is the midpoint
of CZ.
Since CDZ is right-angled at D, the point Y is the circumcentre of CDZ. Hence
we have Y C = Y D = Y Z.
Let CBA = XBD = 2x. We now compute some other angles in the diagram.

BAC = 90 2x
ZAY = 45 x
Y ZA = 45 + x
ZDY = 45 + x
DY Z = 90 2x
DY X = 180 2x

(angle sum ABC)


(AY bisects BAC)
(angle sum AZY )
(Y D = Y Z)
(angle sum Y DZ)
(AX CZ)

Hence XBD + DY X = 180 , and so BXY D is cyclic.


C

X
Y

2x
A

30
99

Solution 4 (Seyoon Ragavan, year 11, Knox Grammar School, NSW)


As in solution 1, we deduce that AY CZ.

Let lines AY and CD intersect at H. Since AY CZ and CD AZ, it follows that


H is the orthocentre of CAZ. Hence ZH AC. But since AC BC, we have
ZH  BC. Observe also that HDZY is cyclic because ZDH = 90 = HY Z.

We now have

CXH = ZHX
= ZHY
= ZDY.

(BC  ZH)
(HDZY cyclic)

Hence CXY = BDY , and so BXY D is cyclic.


C

X
Y
H

31
100

Solution 5 (Kevin Xian, year 11, James Ruse Agricultural High School, NSW)
As in solution 1, we deduce that AY CZ. Hence using the angle sums in CAY
and CAX, we have
ACY = 90 XAC = CXY.
Hence by the alternate segment theorem, line AC is tangent to circle CY X at C.
Using the power of point A with respect to circle CY X, we have
AC 2 = AY AX.

(1)

In a similar way we may use the angle sums in CAD and ABC to find
ACD = 90 BAC = CBD.
Hence by the alternate segment theorem, line AC is tangent to circle CDB at C.
Using the power of point A with respect to circle CDB, we have
AC 2 = AD AB.

(2)

Combining (1) and (2), it follows that


AY AX = AD AB,
and so BXY D is cyclic.
C

X
Y

32
101

6. Solution 1 (Michelle Chen, year 11, Methodist Ladies College, VIC)


Answer: 1008
Let f (x) = p(x) q(x), where
p(x) = (x 1)(x 3)(x 5) (x 2015),
q(x) = (x 2)(x 4)(x 6) (x 2014).
We seek the number of distinct real solutions to the equation f (x) = 0. Note that
f (x) is a polynomial of degree 1008 because p(x) has degree 1008 while q(x) has
degree 1007.
Observe that p(x) is the product of an even number of brackets and q(x) is the
product of an odd number of brackets. Therefore,
f (0) = (1) (3) (2015)
(2) (4) (2014)
= 1 3 2015 + 2 4 2014
> 0.
We also have
f (2016) = 2015 2013 1 2014 2012 2
> 0,
where the last line is true because 2015 > 2014, 2013 > 2012, and so on down to
3 > 2.
Next, for x = 2, 4, 6, . . . , 2014, we have q(x) = 0, and so f (x) = p(x). Hence
f (2) = 1 (1) (3) (2013) < 0.
Note that by increasing the value of x from x = 2 to x = 4, we reduce the number
of negative brackets in the product for q(x) by 1. This has the effect of changing
the sign of f (x). In general, each time we increase x by 2 up to a maximum of
x = 2014, we change the sign of f (x). Therefore, we have the following inequalities.
f (0) > 0
f (2) < 0
f (4) > 0
f (6) < 0
..
.
f (2014) < 0
f (2016) > 0
Since f (x) is a polynomial it is a continuous function. Hence, by the intermediate value theorem, f (x) = 0 has at least one solution in each of the intervals
(0, 2), (2, 4), (4, 6), . . . , (2014, 2016). Therefore, there are at least 1008 different solutions to f (x) = 0.
Finally, since f (x) is a polynomial of degree 1008, the equation f (x) = 0 has at
most 1008 roots. Hence we may conclude that f (x) = 0 has exactly 1008 distinct
real solutions.

33
102

Solution 2 (Yong See Foo, year 11, Nossal High School, VIC)
The polynomials p(x), q(x) and f (x) are defined as in solution 1. We also deduce
f (2016) > 0, as in solution 1.
For any x {1, 3, 5, . . . , 2015}, we have p(x) = 0, and so f (x) = q(x). Therefore,
for any x {1, 3, 5, . . . , 2015}, we have
f (x) = (x 2)(x 4) (x (x 1)) (x (x + 1)) (x 2014).

(We use the symbol merely as a placeholder for future reference.)

All bracketed terms to the left of are positive while the remaining bracketed terms
to the right of are negative. Hence the total number of bracketed terms that are
negative is equal to 2014(x+1)
+ 1 = 2015x
.33 Taking into account the minus sign at
2
2
the front, it follows that for each x {1, 3, 5, . . . , 2015}, we have
f (x) > 0 for x 1 (mod 4)

and

Hence we have the following inequalities.

f (x) < 0 for x 3 (mod 4).

f (1) > 0
f (3) < 0
f (5) > 0
..
.
f (2013) > 0
f (2015) < 0
f (2016) > 0
Since f (x) is a polynomial, it is a continuous function. Hence by the intermediate value theorem, f (x) = 0 has at least one solution in each of the intervals
(1, 3), (3, 5), . . . , (2013, 2015), (2015, 2016). Therefore, there are at least 1008 different solutions to the equation f (x) = 0.
However, as in solution 1, f (x) is a polynomial of degree 1008, and so f (x) = 0 has
at most 1008 roots. Thus we may conclude that f (x) = 0 has exactly 1008 distinct
real solutions.

Comment Some contestants found it helpful to do a rough sketch of p(x) and q(x).
This helps to determine the signs of p(x), q(x) and f (x) for x = 0, 1, 2, . . . , 2016.
From this one can further narrow down the location of the solutions to f (x) = 0.
They are found in the intervals (1, 2), (3, 4), (5, 6), . . . , (2013, 2014), (2015, 2016).
y

y = p(x)
y = q(x)
...
1

3
3

2013 2014 2015 2016

This is all still true in the extreme cases x = 1 and x = 2015. The case x = 1 corresponds to putting the
to the left of the first term (x 2) but to the right of the negative sign. The case x = 2015 corresponds
to putting the to the right of the last term (x 2014).

34
103

7. There are many different ways to solve this problem. Over 10 different methods of
solution were found among the 24 contestants who completely solved the problem.
Some of these are presented here.
Solution 1 (Charles Li, year 9, Camberwell Grammar School, VIC)
We proceed by contradiction. Assume that only finitely many integers n satisfy



p(n + 1) p(n) p(n) p(n 1) > 0.

Observe that p(n) = p(n + 1) for all positive integers n 2 because adjacent
numbers cannot be divisible by the same prime. Hence there is a positive integer N
such that



p(n + 1) p(n) p(n) p(n 1) < 0 for all integers n > N .
(*)

Thus p(n) > p(n 1) if and only if p(n + 1) < p(n) for all n > N . It follows that
the graph of p(n) versus n alternates between peaks and valleys once n > N . This
can be schematically visualised as follows.
p(n)

n
Consider the number 2m where m > 1 is large enough so that 2m > N . Note that
p(2m ) = 2. Since 2m 1 and 2m +1 are odd we have p(2m +1) > 2 and p(2m 1) > 2.
Hence the above graph has a valley at 2m . Since peaks and valleys alternate for
every n > N , it follows that there is a valley at every even number and a peak at
every odd number once we pass N .
Consider the number 3m . Note that p(3m ) = 3. However, from the preceding
paragraph, there is a peak at 3m because it is odd and greater than N . Hence
p(3m 1) = 2 and p(3m + 1) = 2. This is possible if and only if 3m 1 and 3m + 1
are powers of 2. But the only powers of 2 that differ by 2 are 21 and 22 . This implies
m = 1, which is a contradiction.

Comment 1 Some contestants found a second way to deduce that the graph of
p(n) versus n (for n > N ) has peaks at odd n and valleys at even n.
They observed that if q > 3 is a prime number greater than N , then there is a peak
at q. This is because p(q) = q, while p(q + 1) q+1
< q and p(q 1) q1
< q.
2
2

Comment 2 All solutions used the method of indirect proof.44 They all established
that the graph of p(n) versus n (for n > N ) has peaks at odd values of n and valleys

4
4

Also known as proof by contradiction.

35
104

at even values of n. We present some of the variations that contestants used to


derive a contradiction from this point on.
Variation 1 (Isabel Longbottom, year 10, Rossmoyne Senior High School, WA)
Consider the number 32m+1 , where m 1 satisfies 32m+1 > N .

Since 32m+1 is odd and p(32m+1 ) = 3, it follows that 32m+1 1 is a power of 2.

Since m > 1, the number 32m+1 1 is a power of 2 which is greater than 2. Hence
32m+1 1 (mod 4).


But 32m+1 (1)2m+1 1  1 (mod 4), which is a contradiction.

Variation 2 (Richard Gong, year 10, Sydney Grammar School, NSW)


Consider the number 34m , where m 1 is large enough so that 34m > N .
Since 34m is odd and p(34m ) = 3, it follows that 34m 1 is a power of 2.

But 34m = 81m 1m (mod 5).

So 5 | 34m 1, which is a contradiction.

Variation 3 (Anthony Pisani, year 8, St Pauls Anglican Grammar School, VIC)


By Dirichlets theorem there are infinitely many primes of the form q = 6k + 1.
Choose any such prime q > N .
There is a valley at 2q because it is even. Hence p(2q + 1) > p(2q) = q.
But 2q + 1 = 2(6k + 1) + 1 = 3(4k + 1), and so 2q + 1 is a multiple of 3.
Hence p(2q + 1)

2q+1
3

< q, which is a contradiction.

Variation 4 (Jack Liu, year 9, Brighton Grammar School, VIC)


Consider the number 2q, where q > 3 is prime greater than N . Note that p(2q) = q.
There is a valley at 2q because it is even.
However, 2q 1, 2q and 2q + 1 are three consecutive numbers, and so one of them
is a multiple of 3. Since q > 3 we know 3  2q.
If 3 | 2q 1, then p(2q 1)

2q1
3

< q, and if 3 | 2q + 1, then p(2q + 1)

Either way, this contradicts that there is a valley at 2q.

2q+1
3

< q.


Variation 5 (Kevin Shen, year 1255 , Saint Kentigern College, NZ)


Let q > N be any prime. There is a valley at 2q because it is even.
So we have p(2q 1) > p(2q) = q. This implies that p(2q 1) = 2q 1.

Thus 2q1 is prime whenever q is prime. Therefore, we also have 2(2q1)1 = 4q3
is prime.
A simple induction shows that 2r (q 1) + 1 is prime for r = 0, 1, 2, . . ..
In particular 2q1 (q 1) + 1 is prime.

By Fermats little theorem, 2q1 (q 1) + 1 1(q 1) + 1 0 (mod q).

Hence q | 2q1 (q 1) + 1. Since 2q1 (q 1) + 1 > q, we have contradicted that


2q1 (q 1) + 1 is prime.


5
5

Equivalent to year 11 in Australia.

36
105

Variation 6 (Thomas Baker, year 11, Scotch College, VIC)


Let q1 < q2 < be the list of all odd primes in order starting from q1 = 3.

By Dirichlets theorem there are infinitely many primes of the form 4k + 3. So we


may choose j > 3 such that qj > N and such that the list q1 , q2 , . . . , qj contains an
odd number of primes of the form of 4k + 3.
Consider the number m = q1 q2 qj . Note that p(m) = qj . Furthermore, there is a
peak at m because it is odd.
Note m 3 (mod 4). Thus m 1 = 2r for some odd number r.

Furthermore, r is not divisible by any qi (1 i j). Hence p(m 1) is a prime


that is greater than pj . Thus p(m 1) > p(m), which contradicts that there is a
peak at m.


37
106

Solution 2 (Jeremy Yip, year 12, Trinity Grammar School, VIC)


We shall prove that p(n 1) < p(n) < p(n + 1) for infinitely many n. In particular,
for each odd prime q, we show there exists a positive integer k such that
k

p(q 2 1) < p(q 2 ) < p(q 2 + 1).


Lemma 1
1 a < b.

(1)
b

For any odd prime q, we have gcd(q 2 + 1, q 2 + 1) = 2 for all integers


a

Proof Let d = gcd(q 2 + 1, q 2 + 1). Using the difference of perfect squares factorisation we know x 1 | x2 1 and x + 1 | x2 1. From this we deduce the following
chain of divisibility.
a

a+1

2a+1

2a+2

2a+2

q2 + 1 | q2

q2
a

b1

1|q
1|q
..
.

2a+3

1
1

1 | q2 1

It follows that q 2 + 1 | q 2 1, and so d | q 2 1. Combining this with d | q 2 + 1


a
b
yields d | 2. Since q 2 + 1 and q 2 + 1 are even, we conclude that d = 2.

Lemma 2

If (1) is false for all positive integers k, then


k

p(q 2 1) < q

and p(q 2 + 1) < q

for k = 1, 2, . . . .

Proof We proceed by induction under the assumption that (1) is false.


 
. Hence
For the base case k = 1, since q is odd we have q 2 1 = 2(q 1) q+1
2
2
2
p(q 1) < q. However, since (1) is false for k = 2, we also have p(q + 1) < q.
For the inductive step, suppose we know that
j

p(q 2 1) < q

and p(q 2 + 1) < q,


j+1

for some positive integer j. Then since q 2 1 = (q 2 1)(q 2 + 1), we have


j+1
j+1
p(q 2 1) < q. However, since (1) is false for k = j+1, we also have p(q 2 +1) < q.
This completes the induction.

To complete the proof of the problem, let us focus on the following list of numbers.
L = p(q 2 + 1), p(q 4 + 1), p(q 8 + 1), . . .
From lemma 1, q 2 + 1, q 4 + 1, q 8 + 1, . . . all have pairwise greatest common divisor
equal to 2. Hence at most one of the numbers in L is equal to 2, while the others
are distinct odd primes. This means that L contains infinitely many different prime
numbers. This is a contradiction because lemma 2 implies that L contains only
finitely many different numbers.


38
107

Comment Jeremys proof is rather impressive because it establishes a much


stronger result than what was required. The AMO problem only required a proof
that at least one of p(n 1) < p(n) < p(n + 1) and p(n 1) > p(n) > p(n + 1)
occurs infinitely often, but without pinning down which of the two alternatives does
occur infinitely often. Jeremys proof establishes that p(n 1) < p(n) < p(n + 1)
definitely occurs infinitely often.66

66 Erd
os

and Pomerance proved this result in their 1978 research paper On the largest prime factors of n
and n + 1. The question of whether p(n 1) > p(n) > p(n + 1) could occur infinitely often was finally
resolved in the positive by Balog in his 2001 research paper On triplets with descending largest prime
factors.

39
108

8. This was the most difficult problem of the 2015 AMO. Just four contestants managed
to solve it completely.
Solution (Seyoon Ragavan, year 11, Knox Grammar School, NSW)


Answer: n n1
2

We shall refer to a triangle ABC as being isosceles with base BC if AB = AC.


Note that if ABC is equilateral then it is isosceles with base AB, isosceles with
base BC, and isosceles with base AC. In this way an equilateral triangle counts as
three isosceles triangles depending on which side is considered the base. Thus the
number of apples Maryam receives is equal to the number of base-specified isosceles
triangles.
Let S denote the set of lines that Maryam draws. Consider a line a S. We
determine how many ways a can be the base of an isosceles triangle. Observe that
if a, b, c and a, d, e both form isosceles triangles with base a, then since no two lines
are parallel,

 the pairs {b, c} and {d, e} are either equal or disjoint. Thus there are at
most n1
pairs {b, c} from S such that a, b, c forms an isosceles triangle with base
2
a. Since there are n possibilities
 fora, we conclude that the number of base-specified
isosceles triangle is at most n n1
.
2

It remains to show that this is attainable. Take n equally spaced lines passing

through the origin so that the angle between consecutive lines is 180
. Next translate
n
each of the lines so that
no
three
of
them
are
concurrent.
We
claim
this
configuration
 n1 
of lines results in n 2 base-specified isosceles triangles.

For any line x, let x denote the line after it has been translated. Consider any
triangle bounded by lines a , b , c . It is isosceles with base a if and only if before the
translations, the lines b and c were symmetric in a. There
are n ways of choosing
 n1
a. Once a is chosen, then by our construction there are 2 pairs b, c such that
they are symmetric in a. Indeed, if n is odd, all the other lines are involved in
a symmetric pair about a. If n is even then exactly one line is not involved in a
symmetric pair about a because it is perpendicular to a. Thus

the total number of
base specified isosceles triangles for this construction is n n1
, as desired.

2

The following diagram illustrates the construction for n = 6. Each line is a base for
two isosceles triangles, one of which is equilateral. In total there are two equilateral
triangles and six isosceles triangles, making twelve base-specified isosceles triangles
in all.

40
109



are possible. The
Comment Other constructions that achieve the bound n n1
2
following one is by Andrew Elvey Price, Deputy Leader of the 2015 Australian IMO
team.
If n is odd, Maryam can achieve this by drawing the lines to form the sides of a
regular polygon with n sides. If n is even, Maryam can achieve this by drawing the
lines to form all but one of the sides of a regular polygon with n + 1 sides.


41
110

AUSTRALIAN MATHEMATICAL OLYMPIAD STATISTICS


SCORE DISTRIBUTION/PROBLEM
Problem Number

Number of
Students/Score

10

14

29

80

17

60

47

86

30

18

10

13

10

11

17

26

11

50

28

23

77

25

24

5.2

4.6

3.4

1.2

5.4

2.5

2.0

0.6

Average Mark

111

AUSTRALIAN MATHEMATICAL OLYMPIAD RESULTS


Name

School

Year
Perfect Score and Gold

Alexander Gunning

Glen Waverley Secondary College VIC

12

Seyoon Ragavan

Knox Grammar School NSW

11

Gold
Jeremy Yip

Trinity Grammar School VIC

12

Yong See Foo

Nossal High School VIC

11

Ilia Kucherov

Westall Secondary College VIC

11

Yang Song

James Ruse Agricultural High School NSW

12

Allen Lu

Sydney Grammar School NSW

12

Alan Guo

Penleigh and Essendon Grammar School VIC

12

Thomas Baker

Scotch College VIC

11

Richard Gong

Sydney Grammar School NSW

10

Henry Yoo

Perth Modern School WA

12

Silver
Matthew Cheah

Penleigh and Essendon Grammar School VIC

10

Kevin Xian

James Ruse Agricultural High School NSW

11

Kevin Shen

Saint Kentigern College NZ

12 NZ

George Han

Westlake Boys High School NZ

13 NZ

Wilson Zhao

Killara High School NSW

11

Michelle Chen

Methodist Ladies' College VIC

11

Leo Li

Christ Church Grammar School WA

11

Jack Liu

Brighton Grammar School VIC

Charles Li

Camberwell Grammar School VIC

Anthony Pisani

St Paul's Anglican Grammar School VIC

Ivan Zelich

Anglican Church Grammar School QLD

12

Michael Robertson

Dickson College ACT

11

Edward Chen

Palmerston North Boys High School NZ

William Hu

Christ Church Grammar School WA

Martin Luk

King's College NZ

11 NZ
9
13 NZ

112

Name

School

Year
Bronze

Jerry Mao

Caulfield Grammar School VIC

Zoe Schwerkolt

Fintona Girls' School VIC

11

Harish Suresh

James Ruse Agricultural High School NSW

11

Justin Wu

James Ruse Agricultural High School NSW

11

Austin Zhang

Sydney Grammar School NSW

10

Isabel Longbottom

Rossmoyne Senior High School WA

10

Anand Bharadwaj

Trinity Grammar School VIC

Devin He

Christ Church Grammar School WA

11

Miles Yee-Cheng Lee

Auckland International College NZ

Bobby Dey

James Ruse Agricultural High School NSW

10

Tony Jiang

Scotch College VIC

10

Nitin Niranjan

All Saints Anglican School QLD

12

William Song

Scotch College VIC

11

Wen Zhang

St Joseph's College, Gregory Terrace QLD

Prince Michael Balanay

Botany Downs Secondary College NZ

Michael Chen

Scotch College VIC

William Wang

King's College NZ

12 NZ

Xuzhi Zhang

Auckland Grammar School NZ

13 NZ

Keiran Lewellen

Home Schooled NZ

10 NZ

Eryuan Sheng

Newington College NSW

11

Linus Cooper

James Ruse Agricultural High School NSW

David Steketee

Hale School WA

12

Alexander Barber

Scotch College VIC

11

Matthew Jones

All Saints Anglican School QLD

12

Madeline Nurcombe

Cannon Hill Anglican College QLD

11

Steven Lim

Hurlstone Agricultural High School NSW

12 NZ

13 NZ
12

113

27TH ASIAN PACIFIC MATHEMATICS OLYMPIAD

March, 2015
Time allowed: 4 hours
No calculators are to be used
Each problem is worth 7 points
Problem 1. Let ABC be a triangle, and let D be a point on side BC. A line
through D intersects side AB at X and ray AC at Y . The circumcircle of triangle
BXD intersects the circumcircle of triangle ABC again at point Z = B. The lines
ZD and ZY intersect again at V and W , respectively. Prove that AB = V W .
Problem 2. Let S = {2, 3, 4, . . .} denote the set of integers that are greater than or
equal to 2. Does there exist a function f : S S such that
f (a)f (b) = f (a2 b2 ) for all a, b S with a = b?

Problem 3. A sequence of real numbers a0 , a1 , . . . is said to be good if the following


three conditions hold.
(i) The value of a0 is a positive integer.
(ii) For each non-negative integer i we have ai+1 = 2ai + 1 or ai+1 =

ai
.
ai + 2

(iii) There exists a positive integer k such that ak = 2014.


Find the smallest positive integer n such that there exists a good sequence a0 , a1 , . . .
of real numbers with the property that an = 2014.
Problem 4. Let n be a positive integer. Consider 2n distinct lines on the plane, no
two of which are parallel. Of the 2n lines, n are colored blue, the other n are colored
red. Let B be the set of all points on the plane that lie on at least one blue line,
and R the set of all points on the plane that lie on at least one red line. Prove that
there exists a circle that intersects B in exactly 2n 1 points, and also intersects R
in exactly 2n 1 points.
Problem 5. Determine all sequences a0 , a1 , a2 , . . . of positive integers with a0 2015
such that for all integers n 1:
(i) an+2 is divisible by an ;
(ii) |sn+1 (n + 1)an | = 1, where sn+1 = an+1 an + an1 + (1)n+1 a0 .

114

27TH ASIAN PACIFIC MATHEMATICS OLYMPIAD


SOLUTIONS
1. Solution

(Alan Guo, year 12, Penleigh and Essendon Grammar School, VIC)

Applying the Pivot theorem to AXY we see that circles ABC, XBD and CDY
are concurrent at point Z.11 Hence CDZY is cyclic.
A

V
X

Let (m, n) denote the directed angle between any two lines m and n.22 We have
(W Z, V Z) = (Y Z, DZ)
= (Y C, DC) (CDZY cyclic)
= (AC, BC).
Therefore V W and AB subtend equal or supplementary angles in . It follows that
V W = AB.

Comment All solutions that were dependent on how the diagram was drawn received a penalty deduction of 1 point. The easiest way to avoid diagram dependence
was to use directed angles as in the solution presented above.

11

The point Z is also the Miquel point of the four lines AB, AY , BC and XY . It is the common point of
the circumcircles of ABC, AXY , BDX and CDY . See the sections entitled Pivot theorem and
Four lines and four circles found in chapter 5 of Problem Solving Tactics published by the AMT.
22
This is the angle by which one may rotate m anticlockwise to obtain a line parallel to n. For more
details, see the section Directed angles in chapter 17 of Problem Solving Tactics published by the AMT.

43
115

2. Solution 1 (Henry Yoo, year 12, Perth Modern School, WA)


Answer: No
Assume such a function exists. For any positive integer n we have


f (2n )f (2n+4 ) = f 22(2n+4) = f (2n+1 )f (2n+3 ),
and so

Similarly,
and so

f (2n+3 )
f (2n )
=
.
f (2n+1 )
f (2n+4 )

(1)



f (2n+1 )f (2n+4 ) = f 22(2n+5) = f (2n+2 )f (2n+3 ),
f (2n+3 )
f (2n+1 )
=
.
f (2n+2 )
f (2n+4 )

(2)

Combining (1) and (2), we find that


f (2n+1 )
f (2n )
=
.
f (2n+1 )
f (2n+2 )

(3)

Since (3) is true for all positive integers n, it follows that


f (21 ), f (22 ), f (23 ), . . .
is a geometric sequence. Thus f (2n ) = arn1 for some positive rational numbers a
and r.
Since f (2)f (22 ) = f (26 ) we have
a ar = ar5

a = r4 .

(4)

And since f (2)f (23 ) = f (28 ) we have


a ar2 = ar7

a = r5 .

(5)

Comparing (4) and (5) we have r4 = r5 . Since r > 0, we have r = 1. But then from
(4) we have a = 1. Thus f (2) = 1  S, a contradiction. Hence there are no such
functions.


44

116

Solution 2 (Linus Cooper, year 9, James Ruse Agricultural High School, NSW)
Suppose such a function exists. Then for any integer b > 2 we have
f (8b)f (b) = f (64b4 ) = f (2)f (4b2 ) = f (2)f (2)f (b),
and so
f (8b) = f (2)2 .
Thus f (c) = f (2)2 whenever c 24 and c is a multiple of 8.
Next we have

f (24)f (32) = f (242 322 )

f (2)4 = f (2)2 .
Since f (2) > 0, we have f (2) = 1  S. This contradiction shows that no such
function exists.


45
117

Solution 3 (Jeremy Yip, year 12, Trinity Grammar School, VIC)


Suppose such a function exists. Let a < b < c be positive integers. We work with
the quantity f (2a )f (2b )f (2c ) in two different ways.
 a

f (2 )f (2c ) f (2b ) = f (22a+2c )f (2b ) (since c > a)
= f (24a+2b+4c )

(since 2a + 2c > b)

(1)

f (2b )f (2c ) f (2a ) = f (22b+2c )f (2a ) (since c > b)


= f (22a+4b+4c )

(since 2b + 2c > a)


f (2a )f (2b ) f (2c ) = f (22a+2b )f (2c ) (since b > a)
= f (24a+4b+2c )

(if c = 2a + 2b)

(2)

(3)

From (1), (2) and (3) it follows that


f (42a+2b+c ) = f (42a+b+2c ) = f (4a+2b+2c ),

(4)

for any three positive integers a < b < c satisfying c = 2a + 2b.

In what follows we use (4) repeatedly.


(a, b, c) = (1, 2, 3)
(a, b, c) = (1, 2, 4)
(a, b, c) = (1, 3, 5)
(a, b, c) = (1, 4, 7)

f (49 ) = f (410 ) = f (411 )


f (410 ) = f (412 ) = f (413 )
f (413 ) = f (415 ) = f (417 )
f (417 ) = f (420 ) = f (423 )

Thus f (49 ) = f (420 ).


However, f (49 )f (4) = f (420 ), and so f (4) = 1  S. This contradiction shows that
no such function exists.


46
118

Solution 4 (Angelo Di Pasquale, Director of Training, AMOC)


For any a, b S, choose an integer c > a, b. Then since bc > a and c > b, we have
f (a4 b4 c4 ) = f (a2 )f (b2 c2 ) = f (a2 )f (b)f (c).
Furthermore, since ac > b and c > a, we have
f (a4 b4 c4 ) = f (b2 )f (a2 c2 ) = f (b2 )f (a)f (c).
Comparing these two equations, we find that for all a, b S,
2

f (a )f (b) = f (b )f (a)

f (b2 )
f (a2 )
=
.
f (a)
f (b)

It follows that there exists a positive rational number k such that


f (a2 ) = kf (a),

for all a S.

(1)

Substituting this into the functional equation yields


f (ab) =

f (a)f (b)
,
k

for all a, b S with a = b.

(2)

Now combine the functional equation with (1) and (2) to obtain for all a S,
f (a)f (a2 ) = f (a6 ) =

f (a)f (a)f (a4 )


f (a)f (a5 )
f (a)f (a)f (a2 )
=
.
=
k
k2
k

It follows that f (a) = k for all a S.

Putting this into the functional equation yields k 2 = k. But 0, 1  S and hence
there is no solution.


47
119

3. Solution 1 (Alexander Gunning, year 12, Glen Waverley Secondary College, VIC)
Answer: n = 60
Clearly all members of the sequence are positive rational numbers. For each nonnegative integer i let ai = pqii where pi and qi are positive integers with gcd(pi , qi ) = 1.
Therefore,
pi+1
2pi + qi
pi+1
pi
=
or
=
.
(1)
qi+1
qi
qi+1
pi + 2qi
If each of pi and qi are odd, then so are 2pi + qi , qi , pi , and pi + 2qi . Thus when the
RHSs of (1) are reduced to lowest terms, the numerators and denominators are still
odd. Hence pi+1 and qi+1 are odd. It follows inductively that if pi and qi are odd,
then pk and qk are odd for all k i. Since pqnn = 2014 we cannot have pi and qi both
being odd for any i n. Since gcd(pi , qi ) = 1, it follows that
pi and qi are of opposite parity for i = 0, 1, . . . , n.

(2)

Suppose pi is odd for some i < n. We cannot have the second option in (1) because
that implies pi+1 and qi+1 are both odd, which contradicts (2). So we must have
i+1
i
the first option in (1), namely, pqi+1
= 2piq+q
. From (2), qi is even, and so we have
i
gcd(2pi + qi , qi ) = gcd(2pi , qi ) = 2. Hence
pi odd

pi+1 = pi +

qi
2

and qi+1 =

qi
2

for i < n.

(3)

On the other hand, a similar argument shows that if pi is even for some i < n, then
pi
i+1
we must take the second option in (1), namely pqi+1
= pi +2q
, and gcd(pi , pi +2qi ) = 1.
i
Hence
pi even

pi+1 =

pi
2

and qi+1 =

pi
+ qi
2

for i < n.

(4)

In both (3) and (4), we have pi+1 + qi+1 = pi + qi . Since pn + qn = 2015, we have
pi + qi = 2015 for i 2015.
If pi is odd, we may combine (3) and (5) to find

qi qi 
(pi+1 , qi+1 ) = pi + ,
2 2
p q 
i
i

,
(mod 2015).
2 2

(5)

(6)

A very similar calculation using (4) and (5) shows that (6) is also true if pi is even.
A simple induction on (6) yields,
(pn , qn )

p
2

0
n,

q0 
2n

(mod 2015).

We require qn = 1. However, from (5), this is true if and only if


qn 1 (mod 2015).
48
120

(7)

Note that q0 = 1 because a0 is a positive integer. Thus from (7) we require


2n 1 (mod 2015).
Since 2015 = 3 13 31, we require 5, 13, 31 | 2n 1. Computing powers of 2, we see
that
5 | 2n 1 4 | n
13 | 2n 1 12 | n
31 | 2n 1 5 | n.
Since 60 = lcm(4, 12, 5) we conclude that 2015 | 2n 1 if and only if 60 | n. The
smallest such positive integer is n = 60.


49
121

Solution 2 (Henry Yoo, year 12, Perth Modern School, WA)


Clearly all members of the sequence are positive rational numbers. For each positive
ai+1 1
2ai+1
or ai =
. Since ai > 0 we deduce that
integer i, we have ai =
2
1 ai+1

ai+1 1

if ai+1 > 1

2
ai =
(1)

2ai+1

if ai+1 < 1.
1 ai+1

Thus ai is uniquely determined from ai+1 . Hence we may start from an = 2014 and
simply run the sequence backwards until we reach a positive integer.
From (1), if ai+1 = uv , then

uv

2v
ai =

2u

vu

if u > v
if u < v.

Consequently, if we define the sequence of pairs of integers (u0 , v0 ), (u1 , v1 ), . . . by


(u0 , v0 ) = (2014, 1) and

(ui vi , 2vi ) if ui > vi
(2)
(ui+1 , vi+1 ) =
(2ui , vi ui ) if ui < vi ,
then ani =

ui
vi

for i = 0, 1, 2, . . ..

Observe from (2) that ui+1 + vi+1 = ui + vi . So since u0 = 2014 and v0 = 1, we have
ui + vi = 2015 for i = 0, 1, 2, . . ..

(3)

Suppose that d is a common factor of ui+1 and vi+1 . Then (3) implies d | 2015, and
so d is odd. If ui > vi , then from (2) we have d | ui vi and d | 2vi . This easily
implies d | ui and d | vi . If ui < vi , we similarly conclude from (2) that d | ui and
d | vi . This inductively cascades back to give d | u0 and d | v0 . Since gcd(u0 , v0 ) = 1,
we deduce that d | 1. Therefore,
gcd(ui , vi ) = 1 for i = 0, 1, 2, . . ..

(4)

From (3) and (4), we have ui = vi . Hence (2) yields ui+1 , vi+1 > 0, and so from (3)
we have
ui , vi {1, 2, . . . , 2014} for i = 0, 1, 2, . . ..
(5)
Next we prove by induction that
(ui , vi ) (2i , 2i ) (mod 2015) for i = 0, 1, 2, . . ..

(6)

The base case is immediate. Also, if (ui , vi ) (2i , 2i ) (mod 2015), then using (2)
we have
ui > vi

(ui+1 , vi+1 ) (2i 2i , 2 2i )


i+1

(2
50
122

,2

i+1

(mod 2015)
(mod 2015)

and
ui < vi

(ui+1 , vi+1 ) (2(2i ), 2i (2i )) (mod 2015)


(2i+1 , 2i+1 )

(mod 2015).

Either way, this completes the inductive step.


We are given that a0 = uvnn is a positive integer. We know that un and vn are positive
integers with gcd(un , vn ) = 1. Thus we seek n such that vn = 1. From (5) and (6),
this occurs if and only if
2n 1 (mod 2015).


As in solution (1), the smallest such positive integer is n = 60.

Comment 1 In this solution, if we were to run the sequence backwards from


an = 2014 until a positive integer is reached, the result would be
2014
,
1
1918
,
97
666
,
1349
122
,
1893

2013
,
2
1821
,
194
1332
,
683
244
,
1771

2011
,
4
1627
,
388
649
,
1366
488
,
1527

2007
,
8
1239
,
776
1298
,
717
976
,
1039

1999
,
16
463
,
1552
581
,
1434
1952
,
63

1983
,
32
926
,
1089
1162
,
853
1889
,
126

1951
,
64
1852
,
163
309
,
1706
1763
,
252

1887
,
128
1689
,
326
618
,
1397
1511
,
504

1759
,
256
1363
,
652
1236
,
779
1007
,
1008

1503
,
512
711
,
1304
457
,
1558
2014
.
1

991
,
1024
1422
,
593
914
,
1101

1982
,
33
829
,
1186
1828
,
187

1949
,
66
1658
,
357
1641
,
374

1883
,
132
1301
,
714
1267
,
748

1751
,
264
587
,
1428
519
,
1496

1487
,
528
1174
,
841
1038
,
977

959
,
1056
333
,
1682
61
,
1954

Since there are 61 terms in the above list, this also shows that n = 60.
Comment 2 A corollary of both solutions is that the only value of a0 which yields
a good sequence is a0 = 2014.

51
123

4. Solution 1 (Jeremy Yip, year 12, Trinity Grammar School, VIC)


Let be the maximal angle that occurs between a red line and a blue line. Let r
be a red line and b be a blue line such that the non-acute angle between them is .
Note that r and b divide the plane into four regions.
In the following diagrams, the two regions that lie within the angular areas defined
by are shaded. Also we orient our configuration so that the x-axis is the angle
bisector of r and b that passes through the two shaded regions. Let O be the
intersection of lines r and b.
Let  be another line in the configuration. Four options arise that need to be
considered.

In the first two diagrams,  does not pass through O. Observe that , r and b enclose
a triangle which is either completely shaded or completely unshaded. If the triangle
is completely shaded, as in the first diagram, then 1 = + > and 2 = + > .
But this is impossible because it implies that  cannot be blue or red. Hence the
triangle is completely unshaded, as in the second diagram.
In the last two diagrams,  passes through O. Observe that apart from the point O,
either all of  lies in the unshaded regions, or all of  lies in the shaded regions. If
 were to lie in the unshaded regions, as in the third diagram, then since + >
and + > , it would follow that  could not be red or blue. Hence  lies in the
shaded regions, as in the fourth diagram.
Let S be the set of intersection points of lines in the configuration that lie on r or b.
Consider any circle that lies to the right of all points of S and is tangent to r and
b in the right-hand shaded region. We claim that has the required property. It
suffices to show that every line in the configuration, apart from r and b, intersects
in two distinct points.

Let R and B be the points of tangency of with r and b, respectively, and let T be
the union of the segments OB and OR.
52
124

If  ( = r, b) is any line of the configuration, then the part of  lying in the right
shaded region is an infinite ray  .
Let F be the figure enclosed by T and the minor arc RB of . Then  passes into
the interior of F , and so intersects the boundary of F at least twice. Since  cannot
intersect T twice, it must intersect the minor arc RB.
Finally,  cannot be tangent to because that would imply that  intersects both
OB and OR (not at O). It follows that  intersects at two distinct points, as
desired.


53
125

Solution 2 (Yang Song, year 12, James Ruse Agricultural High School, NSW)
We may rotate the plane so that no red line or blue line is vertical. Let 1 , 2 , . . . , 2n
be the lines listed in order of increasing gradient. Then there is a k such that lines
k and k+1 are oppositely coloured. By rotating our coordinate system and cyclicly
relabelling our lines we can ensure that 1 , 2 , . . . , 2n are listed in order of increasing
gradient, 1 and 2n are oppositely coloured, and no line is vertical. Without loss of
generality 1 is red and 2n is blue. Let O = 1 2n .

Let  (not one of the 2n lines) be a variable vertical line to the right of O. Let
R = 1  and B = 2n . Since the lines 2 , 3 , . . . , 2n1 have gradients lying
in between those of 1 and 2n , we can move  far enough to the right so that all
the intersection points of 2 , 3 , . . . , 2n1 with  lie between R and B. Let be the
excircle of ORB opposite A. We claim that has the required properties. To
prove this, it suffices to show that each line i (2 i 2n 1) intersects twice.
B

O
R

R

Let  be the vertical line which is tangent to and lying to the right of . Let
R =  1 and B  =  2n . Then is the incircle of BRR B  . The result now
follows because any line that intersects the opposite sides of a quadrilateral having
an incircle, must also intersect the incircle twice.


54
126

5. Solution (APMO Problem Selection Committee)


There are two families of answers.
an = c(n + 2)n! for all n 1 and a0 = c + 1 for some integer c 2014.

an = c(n + 2)n! for all n 1 and a0 = c 1 for some integer c 2016.


Let a0 , a1 , a2 , . . . be a sequence of positive integers satisfying the given conditions.
We can rewrite (ii) as
sn+1 = (n + 1)an + hn ,
where hn {1, 1} . Substituting n with n 1 yields
sn = (n 1)an + hn1 ,
for n 2. Adding together the two equations we find
an+1 = (n + 1)an + nan1 + n

(1)

for n 2 and where n {2, 0, 2}.

We also have |s2 2a1 | = 1, which yields a0 = 3a1 a2 1 3a1 , and therefore
a1 a30 671. Substituting n = 2 in (1), we find that a3 = 3a2 + 2a1 + 2 . Since
a1 | a3 , we have a1 | 3a2 + 2 , and therefore a2 223. Using (1), we obtain that
an 223 for all n 0.
Lemma 1

For n 4, we have an+2 = (n + 1)(n + 4)an .

Proof For n 3 we have

an = nan1 + (n 1)an2 + n1
> nan1 + 3.

(2)

By applying (1) and (2) with n substituted by n 1, we have for n 4,


an = nan1 + (n 1)an2 + n1
< nan1 + (an1 3) + n1
< (n + 1)an1 .

(3)

Using (1) to write an+2 in terms of an and an1 along with (2), we have for n 3,
an+2 = (n + 3)(n + 1)an + (n + 2)nan1 + (n + 2)n + n+1
< (n + 3)(n + 1)an + (n + 2)nan1 + 3(n + 2)
< (n2 + 5n + 5)an .
Also for n 4,
an+2 = (n + 3)(n + 1)an + (n + 2)nan1 + (n + 2)n + n+1
> (n + 3)(n + 1)an + nan
= (n2 + 5n + 3)an .
Since an | an+2 , we have an+2 = (n2 + 5n + 4)an = (n + 1)(n + 4)an , as desired. 
Lemma 2

For n 4, we have an+1 =

(n+1)(n+3)
an .
n+2

55
127

Proof Using the recurrence an+3 = (n + 3)an+2 + (n + 2)an+1 + n+2 and writing
an+3 , an+2 in terms of an+1 , an according to lemma 1 we obtain
(n + 2)(n + 4)an+1 = (n + 3)(n + 1)(n + 4)an + n+2 .
Hence n + 4 | n+2 , which yields n+2 = 0 and an+1 =
Lemma 3

For n 1, we have an+1 =

(n+1)(n+3)
an ,
n+2

as desired.

(n+1)(n+3)
an .
n+2

an . By lemma 2, there
Proof Suppose there exists n 1 such that an+1 = (n+1)(n+3)
n+2
is a greatest integer 1 m 3 with this property. Then am+2 = (m+2)(m+4)
am+1 .
m+3
If m+1 = 0, then am+1 =
m+1 = 0.

(m+1)(m+3)
am ,
m+2

which contradicts our choice of m. Thus

Clearly m + 3 | am+1 . So we have am+1 = (m + 3)k and am+2 = (m + 2)(m + 4)k,


for some positive integer k. Then
(m + 1)am + m+1 = am+2 (m + 2)am+1 = (m + 2)k.

So, am | (m + 2)k m+1 . But am also divides am+2 = (m + 2)(m + 4)k. Combining
the two divisibility conditions, we obtain am | (m + 4)m+1 .

Since m+1 = 0, we have am | 2m + 8 14, which contradicts the previous result


that an 223 for all non-negative integers n.


an for n 1. Substituting n = 1 yields 3 | a1 . Letting a1 = 3c,


So, an+1 = (n+1)(n+3)
n+2
we have by induction that an = n!(n + 2)c for n 1. Since |s2 2a1 | = 1, we then
get a0 = c 1, yielding the two families of solutions.
Finally, since (n+2)n! = n!+(n+1)!, it follows that sn+1 = c(n+2)!+(1)n (ca0 ).
Hence both families of solutions satisfy the given conditions.


56
128

27TH ASIAN PACIFIC MATHEMATICS OLYMPIAD RESULTS


TOP 10 AUSTRALIAN SCORES
Name

School

Year

Total

Award

Jeremy Yip

Trinity Grammar School VIC

12

28

Gold

Alexander Gunning

Glen Waverley Secondary College VIC

12

28

Silver

Yang Song

James Ruse Agricultural High School NSW

12

23

Silver

Henry Yoo

Perth Modern School WA

12

20

Bronze

Thomas Baker

Scotch College VIC

11

20

Bronze

Allen Lu

Sydney Grammar School NSW

12

20

Bronze

Ilia Kucherov

Westall Secondary College VIC

11

19

Bronze

Seyoon Ragavan

Knox Grammar School NSW

11

17

Yong See Foo

Nossal High School VIC

11

16

Alan Guo

Penleigh and Essendon Grammar School VIC

12

14

COUNTRY SCORES
Rank

Country

Number of
Contestants

Score

Gold

Silver

Bronze

Hon.Men

USA

10

298

Korea

10

279

Russia

10

266

Singapore

10

259

Japan

10

256

Canada

10

237

Thailand

10

228

Taiwan

10

222

Australia

10

205

10

Brazil

10

202

11

Peru

10

185

12

Mexico

10

169

13

Hong Kong

10

167

14

Kazakhstan

10

163

15

Indonesia

10

161

16

Malaysia

10

134

17

India

10

127

129

Rank

Country

Number of
Contestants

Score

Gold

Silver

Bronze

Hon.Men

17

Tajikistan

10

127

19

Bangladesh

10

122

20

Philippines

10

105

21

Turkmenistan

10

99

22

Saudi Arabia

10

94

23

New Zealand

10

86

24

Argentina

10

73

24

Colombia

10

73

26

Syria

52

27

Sri Lanka

48

28

El Salvador

47

29

Trinidad and
Tobago

10

31

30

Ecuador

10

27

31

Costa Rica

20

32

Panama

12

33

Cambodia

299

4583

11

42

102

81

Total

130

AMOC SELECTION
SCHOOL
2015 IMO Team Selection
School
The 2015 IMO Selection School was held 514 April at Robert Menzies College, Macquarie
University, Sydney. The main qualifying exams are the AMO and the APMO from which
25 students are selected for the school.
The routine is similar to that for the December School of Excellence; however, there is
the added interest of the actual selection of the Australian IMO team. This year the IMO
would be held in Chiang Mai, Thailand.
The students are divided into a junior group and a senior group. This year there were 10
juniors and 15 seniors. It is from the seniors that the team of six for the IMO plus one
reserve team member is selected. The AMO, the APMO and the final three senior exams
at the school are the official selection criteria.
My thanks go to Andrew Elvey Price, Ivan Guo, Victor Khou, and Konrad Pilch, who
assisted me as live-in staff members. Also to Peter Brown, Vaishnavi Calisa, Mike Clapper,
Nancy Fu, Declan Gorey, David Hunt, Vickie Lee, Peter McNamara, John Papantoniou,
Christopher Ryba, Andy Tran, Gareth White, Rachel Wong, Sampson Wong, Jonathan
Zheng, and Damon Zhong, all of whom came in to give lectures or help with the marking
of exams.

Angelo Di Pasquale
Director of Training, AMOC

2015 Australian IMO team


Name
Alexander Gunning
Ilia Kucherov
Seyoon Ragavan
Yang Song
Kevin Xian
Jeremy Yip
Reserve
Yong See Foo

Year
12
11
11
12
11
12
11

School
State
Glen Waverley Secondary College
VIC
Westall Secondary College
VIC
Knox Grammar School
NSW
James Ruse Agricultural High School NSW
James Ruse Agricultural High School NSW
Trinity Grammar School
VIC
Nossal High School

3
131

VIC

2015 AUSTRALIAN IMO TEAM


Name

School

Year

Alexander Gunning

Glen Waverley Secondary College VIC

12

Ilia Kucherov

Westall Secondary College VIC

11

Seyoon Ragavan

Knox Grammar School NSW

11

Yang Song

James Ruse Agricultural High School NSW

12

Kevin Xian

James Ruse Agricultural High School NSW

11

Jeremy Yip

Trinity Grammar School VIC

12

Nossal High School VIC

11

Reserve
Yong See Foo

2015 Australian IMO Team, from left, Jeremy Yip, Alexander Gunning, Yang Song, Kevin Xian, Ilia
Kucherov and Seyoon Ragavan.

132

PARTICIPANTS AT THE 2015 IMO SELECTION SCHOOL


Name

School

Year

Seniors
Thomas Baker

Scotch College VIC

11

Matthew Cheah

Penleigh and Essendon Grammar School VIC

10

Michelle Chen

Methodist Ladies' College VIC

11

Yong See Foo

Nossal High School VIC

11

Alexander Gunning

Glen Waverley Secondary College VIC

12

Alan Guo

Penleigh and Essendon Grammar School VIC

12

Ilia Kucherov

Westall Secondary College VIC

11

Leo Li

Christ Church Grammar School WA

11

Allen Lu

Sydney Grammar School NSW

12

Seyoon Ragavan

Knox Grammar School NSW

11

Kevin Xian

James Ruse Agricultural High School NSW

11

Jeremy Yip

Trinity Grammar School VIC

12

Henry Yoo

Perth Modern School WA

12

Wilson Zhao

Killara High School NSW

11

Juniors
Bobby Dey

James Ruse Agricultural High School NSW

10

Rachel Hauenschild

Kenmore State High School QLD

10

William Hu

Christ Church Grammar School WA

Tony Jiang

Scotch College VIC

10

Charles Li

Camberwell Grammar School VIC

Jack Liu

Brighton Grammar School VIC

Isabel Longbottom

Rossmoyne Senior High School WA

10

Hilton Nguyen

Sydney Technical High School NSW

Tommy Wei

Scotch College VIC

Wen Zhang

St Joseph's College, Gregory Terrace QLD

133

TEAM School
PREPARATION SCHOOL
IMO Team IMO
Preparation
The pre-IMO July school is always a great reality check when it comes to our perception
of the teams ability. This is of course because we train with the UK team. Our joint
training school was held 28 July at Nexus International School, Singapore.
The routine for the teams each day consisted of an IMO trial exam in the morning, free
time in the afternoon while their papers were being assessed, a short debrief of the exam
late in the afternoon followed by going out to dinner each evening. The results of training
showed that both teams were quite strong, with the UK having the edge.
The final exam also doubles as the annual Mathematics Ashes contest. Australia won
the Ashes in its inaugural year, lost them the next year, and have not been able to win
them back since. There have been a few close calls, and even a tie in 2011. In another
heart-breaking nail biter, the UK again retained the Ashes after both teams tied on 84
points apiece.
Angelo Di Pasquale
IMO Team Leader

1
134

THE MATHEMATICS
ASHES
The 2015 Mathematical
Ashes: AUS v UK
Exam
1. Does there exist a 2015 2015 array of distinct positive integers such that the sums
of the entries on each row and on each column yield 4030 distinct perfect squares?
2. Let and O be the circumcircle and the circumcentre of an acute-angled triangle
ABC with AB > BC. The angle bisector of ABC intersects at M = B. Let
be the circle with diameter BM . The angle bisectors of AOB and BOC intersect
at points P and Q, respectively. The point R is chosen on the line P Q so that
BR = M R.
Prove that BR  AC.

(In this problem, we assume that an angle bisector is a ray.)


3. We are given an infinite deck of cards, each with a real number on it. For every
real number x, there is exactly one card in the deck that has x written on it. Now
two players draw disjoint sets A and B of 100 cards each from this deck. We would
like to define a rule that declares one of them a winner. This rule should satisfy the
following conditions:
1. The winner only depends on the relative order of the 200 cards: if the cards are
laid down in increasing order face down and we are told which card belongs to
which player, but not what numbers are written on them, we can still decide
the winner.
2. If we write the elements of both sets in increasing order as A = {a1 , a2 , . . . , a100 }
and B = {b1 , b2 , . . . , b100 }, and ai > bi for all i, then A beats B.

3. If three players draw three disjoint sets A, B, C from the deck, A beats B and
B beats C, then A also beats C.
How many ways are there to define such a rule?
(In this problem, we consider two rules as different if there exist two sets A and B
such that A beats B according to one rule, but B beats A according to the other.)
Results

AUS
AUS
AUS
AUS
AUS
AUS

1
2
3
4
5
6

Q1
7
7
7
7
3
6
37

Q2
7
0
7
5
6
7
32

Q3
7
7
0
0
1
0
15

21
14
14
12
10
13
84

UNK
UNK
UNK
UNK
UNK
UNK

2
135

1
2
3
4
5
6

Q1
7
7
5
3
7
7
36

Q2
7
2
0
7
6
7
29

Q3
5
0
0
7
7
0
19

19
9
5
17
20
14
84

THE MATHEMATICS ASHES RESULTS


The 8th Mathematics Ashes competition at the joint pre-IMO training camp in Putrajaya, was tied; the results for
the two teams were as follows, with each team scoring a total of 84:

AUSTRALIA
Code

Name

Q1

Q2

Q3

Total

AUS1

Alexander Gunning

21

AUS2

Ilia Kucherov

14

AUS3

Seyoon Ragavan

14

AUS4

Yang Song

12

AUS5

Kevin Xian

10

AUS6

Jeremy Yip

13

37

32

15

84

Q1

Q2

Q3

Total

TOTAL

UNITED KINGDOM
Code

Name

UNK1

Joe Benton

19

UNK2

Lawrence Hollom

UNK3

Samuel Kittle

UNK4

Warren Li

17

UNK5

Neel Nanda

20

UNK6

Harvey Yau

14

36

29

19

84

TOTAL

136

The 56th International Mathematical Olympiad, Chiang Mai,


IMO TEAM LEADERS
REPORT
Thailand
The 56th International Mathematical Olympiad (IMO) was held 416 July in Chiang Mai,
Thailand.
This was the largest IMO in history with a record number of 577 high school students
from 104 countries participating. Of these, 52 were girls.
Each participating country may send a team of up to six students, a Team Leader and a
Deputy Team Leader. At the IMO the Team Leaders, as an international collective, form
what is called the Jury. This Jury was chaired by Soontorn Oraintara.
The first major task facing the Jury is to set the two competition papers. During this
period the Leaders and their observers are trusted to keep all information about the contest
problems completely confidential. The local Problem Selection Committee had already
shortlisted 29 problems from 155 problem proposals submitted by 53 of the participating
countries from around the world. During the Jury meetings one of the shortlisted problems
had to be discarded from consideration due to being too similar to material already in
the public domain. Eventually, the Jury finalised the exam questions and then made
translations into the more than 50 languages required by the contestants. Unfortunately,
due to an accidental security breach, the second days paper had to be changed on the
night before that exam was scheduled. This probably resulted in a harder than intended
second day.
The six questions that ultimately appeared on the IMO contest are described as follows.
1. A relatively easy two-part problem in combinatorial geometry proposed by the
Netherlands. It concerns finite sets of points in the plane in which the perpendicular bisector of any pair of points in such a set also contains another point of the
set.
2. A medium classical number theory problem proposed by Serbia.
3. A difficult classical geometry problem in which one is asked to prove that a certain
two circles are mutually tangent. It was proposed by Ukraine.
4. A relatively easy classical geometry problem proposed by Greece.
5. A medium to difficult functional equation proposed by Albania.
6. A difficult problem in which one is asked to prove an inequality about a sequence
of integers. Although it does not seem so at first sight, the problem is much more
combinatorial than algebraic. It was inspired by a notation used to describe juggling.
The problem was proposed by Australia.
These six questions were posed in two exam papers held on Friday 10 July and Saturday
11 July. Each paper had three problems. The contestants worked individually. They were
allowed four and a half hours per paper to write their attempted proofs. Each problem
was scored out of a maximum of seven points.
For many years now there has been an opening ceremony prior to the first day of competition. HRH Crown Princess Sirindhorn presided over the opening ceremony. Following
the formal speeches there was the parade of the teams and the 2015 IMO was declared
open.
5
137

After the exams the Leaders and their Deputies spent about two days assessing the work
of the students from their own countries, guided by marking schemes, which had been
discussed earlier. A local team of markers called Coordinators also assessed the papers.
They too were guided by the marking schemes but are allowed some flexibility if, for
example, a Leader brought something to their attention in a contestants exam script
that is not covered by the marking scheme. The Team Leader and Coordinators have to
agree on scores for each student of the Leaders country in order to finalise scores. Any
disagreements that cannot be resolved in this way are ultimately referred to the Jury.
The IMO paper turned out to be quite difficult. While the easier problems 1 and 4
were quite accessible, the other four problems 2, 3, 5 and 6 were found to be the most
difficult combination of medium and difficult problems ever seen at the IMO. There were
only around 30 complete solutions to each of problems 2, 3 and 5. Problem 6 was very
difficult, averaging just 0.4 points. Only 11 students scored full marks on it.
The medal cuts were set at 26 for gold, 19 for silver and 14 for bronze.11 Consequently,
there were 282 (=48.9%) medals awarded. The medal distributions22 were 39 (=6.8%)
gold, 100 (=17.3%) silver and 143 (=24.8%) bronze. These awards were presented at the
closing ceremony. Of those who did not get a medal, a further 126 contestants received
an honourable mention for solving at least one question perfectly.
Alex Song of Canada was the sole contestant who achieved the most excellent feat of
a perfect score of 42. He now leads the IMO hall of fame, being the most decorated
contestant in IMO history. He is the only person to have won five IMO gold medals.33 He
was given a standing ovation during the presentation of medals at the closing ceremony.
Congratulations to the Australian IMO team on an absolutely spectacular performance
this year. They smashed our record rank44 to come 6th, and they also smashed our record
medal haul, bringing home two Gold and four Silver medals.55 This is the first time that
each team member has achieved Silver or better. The team finished ahead of many of
the traditionally stronger teams. In particular, they finished ahead of Russia, whom we
would have considered as untouchable.
Congratulations to Gold medallist Alexander Gunning, year 12, Glen Waverley Secondary
College. He is now the most decorated Australian at the IMO, being the only Australian
to have won three Gold medals at the IMO. On each of these occasions he also finished
in the top 10 in individual rankings.66 He is now equal 17th on the IMOs all-time hall of
fame.
Congratulations also to Gold medallist Seyoon Ragavan, year 11, Knox Grammar School.
Seyoon solved four problems perfectly and was comfortably above the Gold medal cut.
He was individually ranked 19th.
11

This was the lowest ever cut for gold, and the equal lowest ever cut for silver. (This was indicative of
the difficulty of the exam, not the standard of the contestants.)
2
2 The total number of medals must be approved by the Jury and should not normally exceed half the
total number of contestants. The numbers of gold, silver and bronze medals must be approximately in
the ratio 1:2:3.
3
3
In his six appearances at the IMO, Alex Song won a bronze medal in 2010, and followed up with gold
medals in 2011, 2012, 2013, 2014 and 2015.
4
4 The ranking of countries is not officially part of the IMO general regulations. However, countries are
ranked each year on the IMOs official website according to the sum of the individual student scores
from each country.
5
5
Australias best performance prior to this was the dream team of 1997. They came 9th, with a medal
tally of two Gold, three Silver and one Bronze.
6
6
In his four appearances at the IMO, Alexander won a bronze medal in 2012, and followed up with gold
medals in 2013 (8th), 2014 (1st) and 2015 (4th).

6
138

And congratulations to our four Silver medallists: Ilia Kucherov, year 11, Westall Secondary College; Yang Song, year 12, James Ruse Agricultural High School; Kevin Xian,
year 11, James Ruse Agricultural High School; and Jeremy Yip, year 12, Trinity Grammar
School.
Three members of this years team are eligible for selection to the 2016 IMO team. So
while it is unlikely we will be able to repeat this years stellar performance, the outlook
seems promising.
Congratulations also to Ross Atkins and Ivan Guo, who were IMO medallists with the
Australian team when they were students.77 They were the authors of the juggling-inspired
IMO problem number six. In fact Ross is a proficient juggler.
The 2015 IMO was organised by: The Institute for the Promotion of Teaching Science
and Technology, Chiang Mai University, The Mathematical Association of Thailand under
the Patronage of His Majesty the King, and The Promotion of Academic Olympiad and
Development of Science Education Foundation.
The 2016 IMO is scheduled to be held July 6-16 in Hong Kong. Venues for future IMOs
have been secured up to 2019 as follows.
2017
2018
2019

Brazil
Romania
United Kingdom

Much of the statistical information found in this report can also be found at the official
website of the IMO.
www.imo-official.org

Angelo Di Pasquale
IMO Team Leader, Australia

7
7

Ross and Ivan won Bronze at the 2003 IMO, and Ivan won Gold at the 2004 IMO.

7
139

Ross Atkins demonstrates his juggling skills. (Photo credit: Gillian Bolsover)

84
140

INTERNATIONAL MATHEMATICAL OLYMPIAD


IMO Papers

Language:
English
Day: 1
Day: 1

Friday, July 10, 2015

IMO Papers

Problem 1. We say that a finite set S of points in the plane is balanced if, for any two
different points A and B in S, there is a point C in S such that AC = BC. We say that
Day:P 1in
S is centre-free if for any three different points A, B and C in S, there is no point
S such that P A = P B = P C.
Friday, July 10, 2015
(a) Show that for all integers n 3, there exists a balanced set consisting of n points.
S set S of points in the plane is balanced if, for any two
Problem 1. We say that a finite
(b) Determine
integers
n C
3 for
balanced
set that
conA
B points
S Aalland
Sis which
=
S We say
different
B in S,
there
a pointthere
CAC
in exists
S BC
sucha that
AC =centre-free
BC.
sisting
of
n
points.
A
B
C
S
P
S
P
A
=
P
B
=
S is centre-free if for any three different points A, B and C in S, there is no point PP Cin
S such that P A = P B = P C.
n3
n
Problem 2. Determine all triples (a, b, c) of positive integers such that each of the
(a) Show that for all
integers
n 3, there exists a balanced set consisting of n points.n
n
3
numbers
c, which
bc there
a, ca
b a balanced centre-free set con(b) Determine all integers n ab3for
exists
sisting
is a power
of of
2. n points.

(a, b, c)
(A power of 2 is an integer of the form 2n , where n is a non-negative integer.)
c, bc
ca positive
b
Problem 2. Determine allabtriples
(a,b,a,c) of
integers such that each of the
Problem
3.
Let
ABC
be
an
acute
triangle
with
AB
>
AC. Let be its circumcircle,
numbers
2
H its orthocentre, and F the foot
abof
the
c, altitude
bc a, from
ca A.
b Let M be the midpoint of BC.
n

n
2
2
Let Q be the point on such that HQA = 90 , and let K be the point on such that
is a power of 2.
HKQ = 90 . Assume that the points A, B, C, K and Q are all different, and lie on
(A
power
of ABC
2 is an integer of the form 2n , where
is a non-negative
integer.)
AB >nAC

H
in this
order.
F
A
M
BC
Q
Prove that the circumcircles
of triangles KQH and F KM are tangent to each other.

= 90be
K triangle with AB > AC. LetHKQ
= circumcircle,
90
Problem
3. HQA
Let ABC
an acute
be its
A
B
C
K
Q

H its orthocentre, and F the foot of the altitude from A. Let M be the midpoint of BC.
Let Q be the point on such that
HQA F
=KM
90 , and let K be the point on such that
KQH
HKQ = 90 . Assume that the points A, B, C, K and Q are all different, and lie on
in this order.

Prove that the circumcircles of triangles KQH and F KM are tangent to each other.

Language: English

Time: 4 hours and 30 minutes


Each problem is worth 7 points
141

Language: English
Day: Day:
2
Saturday, July 11, 2015

Day: 2

Problem 4. Triangle ABC has circumcircle and circumcentre O. A circle with


centre A intersects the segment BC at points D and E, such thatSaturday,
B, D, E July
and C
all
11,are
2015
different and lie on line BC in this order. Let F and G be the points of intersection of
and , such that
A, F , B, C and G lie
this order.OLet K be the
second point
ABC
on in
A
Problem 4. Triangle
ABC has circumcircle
and circumcentre
O.
A circle with
of intersection ofBC
the circumcircle
triangle BDFB and
the
AB. Let L be the
D BCofat
E
D such
E segment
C B, D,
centre A intersects the segment
points D and E,
that
E and C are all
second
of intersection of the circumcircle of triangle CGE and the segment CA.
BC point
A F
different
and lie on lineFBC inGthis order. Let F and G be thepointsof intersection
of
B C Suppose
G

K
that
the
lines
F
K
and
GL
are
different
and
intersect
at
the
point
X.
Prove
that
and , such that A, F , B, C and G lie on in this order. Let K be the second point
AB
L of triangle BDF and the segment AB. Let L be the
X
lies on the line
AO.circumcircle
ofBDF
intersection
of the
CGE point of intersection
CA of the circumcircle of triangle CGE and the segment CA.
second
Problem 5. Let
R denote the set of real numbers. Determine all
functions f :XR R
X point
Suppose that theF K
lines F GL
K and GL are different and intersect at the
X. Prove that
satisfying
the equation
AO
X lies on the line AO.
f (x + f (x + y)) + f (xy) = x + f (x + y) + yf (x)
R f: R R
Problem 5.R Let R denote the set of real numbers. Determine fall: R
functions
satisfying
equation
 x and y. 
for all realthe
numbers
f x + f (x + y) + f (xy) = x + f (x + y) + yf (x)
f (x + f (xa+, ay)), .+
= x +satisfies
f (x + y)
+ following
yf (x)
Problem 6. x The ysequence
. . fof(xy)
integers
the
conditions:
1

for all real numbers x and y.


a1 ,all
a2 ,j. .
. 1;
(i) 1 aj 2015 for

The jsequence
a1 , a2 , . . . of integers satisfies the following conditions:
1
1Problem

aj k+2015
(ii)
a6.
k = + a for all 1 k < .
+
a
1 all
k<
1;
k +(i)
ak =
2015 for
j
ajthere
Prove1that
exist two
positive
integers b and N such that
b
N 
(ii) k + ak = + a for all 1 k <
n .

 
 n 

 
(a

b)
10072
j
 
 
2
(aj b)b and
1007
 j=m+1
Prove that there exist two positive
integers
N such that


j=m+1

 n
 
for all integers m and n satisfying
n > m N .

m
n
n > m  N (aj b) 10072


j=m+1

for all integers m and n satisfying n > m N .

Language: English

Language: English

Time: 4 hours and 30 minutes


Each problem is worth 7 points
142

Time: 4 hours and 30 minutes

INTERNATIONAL MATHEMATICAL OLYMPIAD


SOLUTIONS
1. Solution (All members of the 2015 Australian IMO team solved this problem using
similar methods. Here we present the solution by Yang Song, year 12, James Ruse
Agricultural High School, NSW. Yang was Silver medallist with the 2015 Australian
IMO team.)
(a) For n odd, we may take S to be the set of vertices of a regular n-gon P.
It seems obvious that S is balanced but we shall prove it anyway. Let A and
B be any two vertices of P. Since n is odd, one side of the line AB contains an
odd number of vertices of P. Thus if we enumerate the vertices of P in order
from A around to B on that side of AB, one of them will be the middle one,
and hence be equidistant from A and B.
For n even, say n = 2k we may take S to be the set of vertices of a collection
of k unit equilateral triangles, all of which have a common vertex, O say, and
exactly one pair of them has a second common vertex. Note that apart from
O, all of the vertices lie on the unit circle centred at O.
The reason why this works is as follows. Let A and B be any two points in S.
If they are both on the circumference of the circle, then OA = OB and O S.
If one of them is not on the circumference, say B = O, then by construction
there is a third point C S such that ABC is equilateral, and so AC = BC.
The cases for n = 11 and n = 12 are illustrated below.

(b) We claim that a balanced centre-free set of n points exists if and only if n is
odd.
Note that the construction used in the solution to part (a) is centre-free. We
shall show that there is no balanced centre-free set of n points if n is even.
For any three points A, X, Y S, let us write A {X, Y } to mean AX = AY .
We shall estimate the number of instances of A {X, Y } in two different ways.
First, note that if A {X, Y } and A {X, Z} where Y = Z, then S cannot
be centre-free
 n1  because AX = AY = AZ. Hence for a given point A, there are
at most 2 pairs {X, Y } such that A {X, Y }. Since there are
 n choices
for A, the total number of instances of A {X, Y } is at most n n1
.
2
On the other hand, since S is balanced, for each pair of points X, Y S, there
is at leastone
 point A such that A {X, Y }. Since the number ofnpairs

n
{X, Y } is 2 , the total number of instances of A {X, Y } is at least 2 .

 

n2 , which simplifies to
If
we combine our estimates, we obtain n n1
2
 n1
n1
. This final inequality is impossible if n is even.

2
2
58

143

Comment 1 A nice graph theoretical interpretation of the solution to part (b) was
given by Kevin Xian, year 11, James Ruse Agricultural High School, NSW. Kevin
was a Silver medallist with the 2015 Australian IMO team.
Let S  denote the set of unordered pairs of elements of S. Form a directed bipartite
graph G as follows. The vertex set of G is S S  . The directed edges of G are
simply all the instances of A {X, Y} asper the solution to part (b) above. Then
n1
the
 total outdegree of G is at most n 2 , while the total indegree of G is at least
n
. The inequality found in the solution to part (b) is simply a consequence of the
2
total indegree being equal to the total outdegree.
Comment 2 Alternative constructions for n odd in part (a) were found by Yang
Song, year 12, James Ruse Agricultural High School, NSW, and Jeremy Yip, year
12, Trinity Grammar School, VIC. Yang and Jeremy were Silver medallists with the
2015 Australian IMO team.
For n = 2k + 1, let S be the set of vertices of a collection of k unit equilateral
triangles, all of which have exactly a common vertex, O say, and no two of which
have any other common vertices besides O. Note that apart from O, all of the
vertices lie on the unit circle centred at O. The case for n = 13 is illustrated below.

Comment 3 At the time of writing, the classification of balanced sets is an open


problem. From the foregoing, we see that for n odd there are two infinite families of
balanced sets, and for n even there is one infinite family. It is unknown if there are
any other infinite families which do not fit into this scheme. Note however, there
are sporadic constructions which do not fit into either of these families. Some of
these are shown below. (All drawn segments are of unit length.)

59
144

2. Solution 1 (Jeremy Yip, year 12, Trinity Grammar School, VIC. Jeremy was a
Silver medallist with the 2015 Australian IMO team.)
The answers are (a, b, c) = (2, 2, 2), (2, 2, 3), (2, 6, 11), (3, 5, 7) and their permutations. It is straightforward to verify that they all work.
Note that ab c = bc a (b + 1)(a c) = 0 a = c.

We divide our solution into five cases as follows.


Case 1 At least two of a, b, c are equal.

Case 2 All of a, b, c are different and none of them are even.


Case 3 All of a, b, c are different and one of them is even.
Case 4 All of a, b, c are different and two of them are even.
Case 5 All of a, b, c are different and all three of them are even.
Case 1 Without loss of generality a = b. Then b2 c and bc b are powers of 2.
That is,
b2 c = 2x
b(c 1) = 2y

(1)
(2)

for some non-negative integers x and y. Equation (2) implies that b = 2p and
c 1 = 2q for some non-negative integers p and q. Putting these into (1) we find
22p = 2x + 2q + 1.

(3)

Since RHS(3) 3, we have p 1, and so 4 | 2x + 2q + 1.

If q = 0, then (3) becomes 22p 2x = 2. The only powers of two that differ by 2 are
2 and 4. Hence x = p = 1, which quickly leads to (a, b, c) = (2, 2, 2).

Similarly, if x = 0, then (3) becomes 22p 2q = 2. Hence q = p = 1, and so


(a, b, c) = (2, 2, 3).
Finally, if q, x 1, then (3) cannot hold because the LHS is even while the RHS is
odd.
Case 2 Without loss of generality we may suppose that a > b > c, where a, b, c
are all odd.
It follows that ab c > ac b and a 3. Since ab c and ac b are powers of two,
the smaller divides the larger. Therefore,
ab c 0 (mod ac b)

a(ac) c 0 (mod ac b)
(a 1)(a + 1) 0 (mod ac b). (c is odd)
Since a 1 and a + 1 are consecutive even positive integers, exactly one of them is
divisible by 4, while the other is even but not divisible by 4. Since ac b is a power
of two, it follows that either ac b | 2(a 1) or ac b | 2(a + 1). Either way, we
have
ac b 2(a + 1).
Since a > b, we have ac a < 2(a + 1), which can be rearranged as
(c 3)a < 2.
60
145

(4)

Observe that a 5, b 3 and c 1. This is because a, b, c are all odd and


a > b > c 1.
If c > 3 then (4) implies a < 2. This contradicts a 5.

If c = 1, then a b and b a are powers of two, which is impossible because they


sum to zero.
If c = 3, then 3b a and 3a b are powers of two. Since 3b a < 3a b, we have
3a b 0 (mod 3b a)
3(3b) b 0 (mod 3b a)

8 0 (mod 3b a). (b is odd)


Hence 3b a {1, 2, 4, 8}.

Certainly 3b a = 1, because a and b are odd.

If 3b a = 2, then a = 3b 2, and so 3a b = 8b 6 2 (mod 4). Hence 8b 6 = 2,


and b = 1. But this impossible because b > c 1.

If 3b a = 4, then a = 3b 4, and so 3a b = 8b 12 4 (mod 8). Hence


8b 12 = 4, and b = 2, which contradicts that b is odd.

If 3b a = 8, then a = 3b 8, and so 3a b = 8(b 3). Hence b 3 is a power of


two. Thus b = 3 + 2k for some positive integer k. Also
ab c = b(3b 8) 3
= (3b + 1)(b 3)
is a power of two. Hence 3b+1 = 32k +10 is also a power of two. This is impossible
for k 2 because 3 2k + 10 2 (mod 4) and 3 2k + 10 = 2. However, for k = 1,
we find 3b + 1 = 16. Thus b = 5 and a = 3b 8 = 7, and we have found the solution
(a, b, c) = (3, 5, 7).
Case 3 Without loss of generality a is even, while b and c are odd.
Then ab c and ac b are both odd. Hence ab c = ac b = 1. This implies b = c,
which is a contradiction. So this case does not occur.
Case 4 Without loss of generality c is odd, while a and b are even with a > b.
This immediately implies that a 4. Also, since ab c is odd, we have ab c = 1.
Thus c = ab 1 3 (mod 4). Hence c 3.

Let 2k (note k 1) be the greatest power of two dividing both a and b. Then
a = 2k m and b = 2k n for some integers m > n 1.

We have ac b = 2k (mc n). Thus mc n is a power of two. Thus m and n are of


the same parity because otherwise mc n would be an odd number that is greater
than 1. From the choice of k, this implies that m and n are both odd.
Since a > b we have bc a < ac b. Because they are both powers of two, the
smaller divides the larger. Therefore,

ac b 0 (mod bc a)
(bc)c b 0 (mod bc a)

2k n(c2 1) 0 (mod 2k (nc m))


(c 1)(c + 1) 0 (mod nc m).
(n is odd)
61
146

(5)

(6)

Using similar reasoning as in case 2, since nc m is a power of two, and c 1 and


c + 1 are consecutive even positive integers, we deduce that
nc m 2(c + 1)

n(4 mn 1) m 2 4k mn (c = ab 1 = 4k mn 1)
n

4k (n2 2n)
m
< 1.
(m > n)
Since n is odd, this implies n = 1. Putting this into (6), we find
(c 1)(c + 1) 0 (mod c m)
(m 1)(m + 1) 0 (mod c m).
Using similar reasoning as in case 2, since c m is a power of two, and m 1 and
m + 1 are consecutive even positive integers, we deduce that
c m 2(m + 1)

4k m m 2m + 2

4k m < 3m + 2
k = 1.

(c=4k m 1)

To summarise, we have found (a, b, c) = (2m, 2, 4m 1). In particular, c = 2a 1.


Putting this into (5) we find,
a(2a 1) 2 0 (mod 3a 2)

3a(6a 3) 18 (mod 3a 2)

2 18 (mod 3a 2).
Hence 3a 2 | 16. Since a 4, we have 3a 2 = 16. Thus a = 6, and so
(a, b, c) = (6, 2, 11).
Case 5 Without loss of generality we may suppose that a = 2 A, b = 2 B and
c = 2 C, where 1, and A, B and C are odd positive integers. It
follows that
ab c = 2 (2+ AB C) and ac b = 2 (2+ AC B).
Since 1, we have that 2+ AB C and 2+ AC B are odd.
But since ab c and ac b are powers of two, it follows that
ab c = 2

and ac b = 2 .

Since 2 | c, we have 2 c, and similarly 2 b. Hence


ab 2c and ac 2b.

(7)

Multiplying these inequalities together yields


a2 bc 4bc.
Hence a 2, and thus a = 2 because a is even. But now the two inequalities in
(7) become 2b 2c and 2c 2b. Hence b = c, which contradicts that a, b, c are all
different. So this case does not occur, and the proof is complete.

62
147

Solution 2 (Alex Gunning, year 12, Glen Waverley Secondary College, VIC. Alex
was a Gold medallist with the 2015 Australian IMO team.)
The case where two of the variables are equal is handled in the same way as in
solution 1.
Without loss of generality we may assume that a > b > c. If c = 1, then we would
require both a b and b a to be powers of two. But this is impossible because
their sum is zero.
Hence a > b > c 2.

It easily follows that ab c ca b bc a. Since a smaller power of two always


divides a larger power of two, we have
bc a | ca b | ab c.
Thus ca b | (a(ab c) (ca b)), that is, ca b | (a + 1)(a 1)b.

If a is even, then since ca b is a power of two and (a 1)(a + 1) is odd, we must


have ca b | b. Thus ca 2b < 2a, which yields the contradiction c < 2. Henceforth
we may assume that a is odd. Also a > b > c 2 implies a 4, and hence a 5.

If 4 | a 1, then since ca b is a power of two and a + 1 is divisible by 2 but not


by 4, we have
ca b | 2(a 1)b
ca b | 2(a 1)b 2(ab c) (ca b | ab c)
ca b | 2(b c)
ca b 2(b c).
(b = c)
If 4 | a + 1, then since a 1 is divisible by 2 but not by 4, we have
ca b | 2(a + 1)b
ca b | 2(a + 1)b 2(ab c) (ca b | ab c)
ca b | 2(b + c)
ca b 2(b + c).
Thus whatever odd number a is, we can be sure that ca b 2(b + c). That is,
ca 3b + 2c.
Since b, c < a, we have ca < 3a + 2a = 5a, and so c < 5.
If c = 4, then 4a < 3a + 8. Thus a < 8. But a > b > c = 4 implies a 6. Thus
a = 7 because a is odd. But since cb a = 4b 7 is a power of two and it is odd,
we must have 4b 7 = 1. Thus b = 2, which contradicts b > c.
If c = 3, then 3a 3b + 6 < 3a + 6. Thus a b + 2 < a + 2.

If b = a 1, then ca b = 3a (a 1) = 2a + 1 cannot be a power of two because


it is odd and greater than 1.

If b = a 2, then bc a = 3(a 2) a = 2a 6 and ca b = 3a (a 2) = 2a + 2


are two powers of two that differ by 8. The only such powers of two are 8 and
16. This quickly yields a = 7 and implies (a, b, c) = (7, 5, 3).
63
148

If c = 2, then bc a = 2b a = 1 because it is odd (a is odd) and is a power of two.


But
ca b | ab c
3b 2 | (2b 1)b 2
(a = 2b 1, c = 2)
3b 2 | 3((2b 1)b 2) 2b(3b 2)
3b 2 | b 6.
Since 3b 2 > b 6 > 0 for b 7, we must have b 6.

Checking 3b 2 | b 6 directly for b = 3, 4, 5, 6, we find that only b = 6 works. This


implies (a, b, c) = (11, 6, 2).


64
149

3. Solution 1 (Alex Gunning, year 12, Glen Waverley Secondary College, VIC. Alex
was a Gold medallist with the 2015 Australian IMO team.)
Let A be the point diametrically opposite A on and let E be the second point of
intersection of the line AHF with .
Lemma The points A , M , H and Q are collinear.
Proof First note that QA QA because AA is a diameter of . Also since
QA QH, it follows that QHA is a straight line. Thus Q lies on the line A H.
A
Q

A

Next we show that CHBA is a parallelogram.


Since AA is a diameter of we have AE A E. However AE BC. Thus
BC  A E. Since A EBC is cyclic, it follows that A ECB is an isosceles trapezium
with CE = A B. Thus
BCA = EBC
= EAC
= 90 ACB
= CBH.

(A B = CE)
(ECAB cyclic)
(AE BC)
(BH AC)

Thus BH  A C. A similar argument shows that CH  A B, and so CHBA is a


parallelogram.
Finally, since the diagonals of any parallelogram bisect each other, and M is the
midpoint of BC, it follows that M is also the midpoint of A H. Thus M lies on the
line A H. Since we already know that Q lies on the line A H, it follows that A , M ,
H and Q are collinear.

Let R be the intersection of the line A E with the line through H that is perpendicular to A Q. Observe that RH is tangent to circles KQH, F HM and EHA . This
is because each of these circles has a diameter which lies on the line A Q. Applying
the radical axis theorem to the three circles , KQH and EHA , we deduce that
lines RH, A E and QK are concurrent. Hence R lies on QK.

65
150

A
Q

S
B

A

Recall that M is the midpoint of A H and that BC  A E. Thus the midpoint, S


say, of RH lies on the line BC. Since HK KR we have that S is the centre of
circle RKH, and so SH = SK. But since SH is tangent to circle KQH at H and
SH = SK, it follows that SK is tangent to circle KQH at K. Considering the
power of point S with respect to circle F HM , we have
SF SM = SH 2 = SK 2 .
It follows that SK is tangent to circle F KM at K. Since circle KQH is also tangent
to SK at K, we conclude that circles KQH and F KM are tangent to each other
at K.


66
151

Solution 2 (Andrew Elvey Price, Deputy Leader of the 2015 Australian IMO
team)
Point A is defined as in solution 1. Furthermore, as in solution 1, we establish that
M is the midpoint of A H, and A , M , H and Q are collinear.
Since A AK = A QK = HQK and AKA = 90 = QKH, we have
KAA KQH. Hence there is a spiral symmetry, f say, centred at K, such
that f (A ) = A and f (H) = Q. Note that f is the composition of a 90 rotation
KA
about K with a dilation of factor KA
 about K.
Let M  and F  be the respective images of M and F under f . Thus M  F  M F .
But since AH M F , it follows that M  F   AH.

Since M is the midpoint of A H, it follows that M  is the midpoint of AQ. Since


M  F   AH, it follows that M  F  passes through the midpoint, S  say, of HQ. Note
that S  is the centre of circle KQH.
A
M
Q

F
S

H
M
B

A

Let S be the preimage of S  under f . (Note that S happens to lie on the line M F ,
as shown in the diagram, because S  lies on line M  F  . But we will not need this
fact.)
Since S  F  Q S  QM  (AA), we have
S Q
S F 
=
.
S Q
S M 
It follows that
S  F  S  M  = S  Q2 = S  K 2 .

Therefore, S  K is tangent to circle F  KM  at K.

Applying the inverse of f , we have SK is tangent to circle F KM at K. But since


S is the preimage of S  under f , we also have S  KS = 90 , and so SK is tangent
to circle KQH. Hence circles F KM and KQH are tangent at K.

67
152

Solution 3 (IMO Problem Selection Committee)


Points A and E are defined as in solution 1. Moreover, as in solution 1, we establish
that M is the midpoint of A H, and A , M , H and Q are collinear. Note that since
M F  A E, it follows that F is the midpoint of EH.
Let Q be the point diametrically opposite Q on . Then KQ KQ because QQ
is a diameter of . Also since KQ KH, it follows that KHQ is a straight line.

Let T be a point on the tangent to circle KQH at K, such that T and Q lie
on the same side of the line KH. By the alternate segment theorem we have
T KQ = KHQ.
By the alternate segment theorem, it is sufficient to prove that
KF M = T KM.
We have
KF M = T KM
90 + KF A = T KQ + 90 + HKM

KF A = Q HA + HKM.

(1)

Thus it suffices to establish (1).


Let J be the midpoint of HQ . Observe that triangles KHE and AHQ are similar
with F and J being the midpoints of corresponding sides. Hence KF A = HJA.

Q
B

A

Observe also that triangles KHA and QHQ are similar with M and J being the
midpoints of corresponding sides. Hence HKM = JQH.
Thus our task is reduced to proving
HJA = Q HA + JQH.
Let us draw a new diagram that will help us focus on the task at hand.
68
153

O
J

Q

A

Note that QAQ A is a rectangle. Let O be its centre. We also know that H lies
on side A Q and that J is the midpoint of Q H. Thus J and O both lie on the
mid-parallel of QA and Q A. Hence
HJA = HJO + OJA
= Q HA + Q AJ. (A Q  JO  Q A)
Thus it suffices to prove that JQH = Q AJ. However, this is an immediate
consequence of the fact that JO a line of reflective symmetry of the rectangle. 

69
154

Solution 4 (Jacob Tsimerman, Leader of the 2015 Canadian IMO team)


Points A , E and Q are defined as in solution 3. Furthermore, as in solution 3, we
have M is the midpoint of A H; points A , M , H and Q are collinear; points Q , H
and K are collinear; and F is the midpoint of HE.
Since the three chords AE, QA and KQ are concurrent we have
HA HE = HQ HA = HK HQ = r2 ,
for some positive real number r. Let f be the inversion with centre H and radius
1 1 in H. Let A and Q be the reflections of H in A and
r, followed by the reflection
Q, respectively. Then f has the effect of exchanging the following pairs of points.
AE

A F

Q A

Q M

K Q

A

A
Q

Q
K

H
Q
B

M
A

Hence circle KQH is transformed into line Q A , and circle F KM is transformed


into circle A Q Q . So it suffices to show that Q A is tangent to circle A Q Q . Since
Q A  AQ  A Q , it suffices to show that A Q Q is isosceles with Q A = Q Q .

Observe Q A  A Q because AQA Q is a rectangle. So Q A A Q . But A is the


centre of circle A HQ because it is the midpoint of A H and A Q H = 90 . Thus
Q A is the perpendicular bisector of A Q . From this it follows that Q A = A Q ,
as desired.


Reflection in a point is the same as a 180 rotation about that point.

70
155

Comment Solutions 1 and 3 first establish that Q, H, M and A are collinear, and
that M and F are the respective midpoints of A H and EH. After this the points
B and C are no longer relevant to the solution. The crux of matter boils down to
the following.
Let AQA Q be a rectangle inscribed in a circle . Let H be any point on
the line A Q. Let E and K be the respective second points of intersection
of the lines AH and Q H with . Let M and F be the midpoints of A H
and EH, respectively. Then circles KQH and KM F are tangent to each
other.

H
Q
M

A

71
156

Here is an even more minimal way of looking at the same thing.


Let A EKQ be a cyclic quadrilateral. Let H be a point on the line A Q
such that HEA = QKH = 90 . Let M and F be the midpoints of
A H and EH, respectively. Then circles KQH and KM F are tangent to
each other. (*)

A

72
157

Solution 5 (Panupong Pasupat, one of the Coordinators at the 2015 IMO)


It is sufficient to prove statement (*) found in the comments on the previous page.
Let T be a point on the tangent to circle KQH at K, such that T and Q lie on the
same side of the line KH. Let W be the midpoint of QH. Since QKH = 90 , it
follows that W is the centre of circle KQH. Hence W H = W K, and we may let
W KH = KHW = .
From the angle sum in KQH, we have HKQ = 90 . Since A EKQ is cyclic,
we have AEK = 90 + , and so KEH = . Thus by the alternate segment
theorem, circle HEK is tangent to line A Q at H. Then since M H = M E, it
follows by symmetry that M H and M E are the common tangents from M to circle
HEK.

W
K

M
F

A

The configuration where M H and M E are common tangents to the circumcircle of


HEK, is a standard one. In particular it yields an alternative characterisation of
the symmedian.22 Thus we may let
HKM = F KE = .
From the exterior angle sum in EKF , we have

KF H = + = W KM.

(1)

Since W K is the radius of circle KQH, we have T K KW . Thus adding 90 to


both sides of (1) yields
KF M = T KM.
Hence from the alternate segment theorem, T K is tangent to circle F KM at K.
Thus circles F KM and KQH are tangent at K.

2
2

For more details see the section Alternative characterisations of symmedian found in chapter 5 of Problem
Solving Tactics published by the AMT.

73
158

4. Solution 1 (Kevin Xian, year 11, James Ruse Agricultural High School, NSW.
Kevin was a Silver medallist with the 2015 Australian IMO team.)

X
K
L

Observe that AF = AG (radius of ), and OF = OG (radius of ). So F and


G are symmetric in AO. Thus lines F K and GL intersect on AO if and only if
KF A = AGL. We have,
KF A = F KB F AB
= F DB F GB
= F GE F GB
= BGE
= CEG CBG
= CLG CAG
= AGL.

(exterior angle AF K)


(BDKF and AF BG cyclic)
(F DEG cyclic)
(exterior angle GEB)
(ECGL and ABCG cyclic)
(exterior angle GLA)

Comment A careful analysis of this solution shows that the result is still true if
we only assume that the centre of lies on the line AO.

74
159

Solution 2 (Found independently by Ilia Kucherov, year 11, Westall Secondary


College, VIC, and Jeremy Yip, year 12, Trinity Grammar School, VIC. Ilia and
Jeremy were Silver medallists with the 2015 Australian IMO team.)
As in solution 1 it suffices to show that KF A = AGL.
Let the lines DF and EG intersect for a second time at points P and Q, respectively. Then
P DE = F GQ (F DEG cyclic)
= F P Q. (F QP G cyclic)
Thus BC  QP . Since BQP C is cyclic, it follows that BQP C is an isosceles
trapezium with BQ = P C. Hence BGQ = P F C, that is,
BGE = DF C.

(1)

F
G

75
160

X
K
L

Since ABCG is cyclic, we have LAG = CAG = CBG = EBG. Since ECGL
is cyclic, CLG = CEG, and so GLA = GEB. Thus GAL GBE (AA).
Hence
AGL = BGE.
(2)
Similarly

KF A = DF C.

(3)

Combining (1), (2) and (3), it follows that


KF A = AGL.

Comment There are two pairs of similar triangles associated with circles ECGL
and . They are GAL GBE and GLE GAB. This is a standard
configuration which can help fast track the route to a solution.33

3
3

For more details see the section Similar Switch found in chapter 5 of Problem Solving Tactics published
by the AMT.

76
161

5. Solution 1 (Seyoon Ragavan, year 11, Knox Grammar School, NSW. Seyoon was
a Gold medallist with the 2015 Australian IMO team.)
We show that the only answers are: f (x) = x for all x R and f (x) = 2 x for all
x R.

We are asked to find all functions f : R R such that for all x, y R,


f (x + f (x + y)) + f (xy) = x + f (x + y) + yf (x).

(1)

Set y = 1 in (1) to find that for all x R,


x + f (x + 1) is a fixed point of f .

(2)

With (2) in mind, set x = 0 and y = z + f (z + 1) in (1) to find that for all z R,
f (0) = f (0)(z + f (z + 1)).

(3)

Case 1 f (0) = 0

Equation (3) implies z + f (z + 1) = 1 for all z R. Putting z = x 1, this may be


rearranged to f (x) = 2 x for all x R. We verify this is a solution.
LHS = 2 (x + 2 (x + y)) + 2 xy = 2 + y xy
RHS = x + 2 (x + y) + y(2 x) = 2 + y xy = LHS
Case 2 f (0) = 0
Set x = 0 in (1) to find that for all y R,
f (f (y)) = f (y).

(4)

Set y = 0 in (1) to find that for all x R,


x + f (x) is a fixed point of f .

(5)

Let S denote the set of fixed points of f . Suppose that u S. Then (5) with x = u
tells us that 2u S. And (2) with x = u 1 tells us that 2u 1 S. Hence
u S 2u, 2u 1 S.

(6)

Since 0 S, applying (6) tells us that 1 S. Applying (6) again tells us that
2, 3 S. Continuing inductively, we find that all negative integers are in S.

On the other hand, if x is any positive integer, choose an integer y such that y < 0,
x + y < 0, 2x + y < 0 and xy < 0 (any y < 2x will do). Using these values in (1)
yields f (x) = x. Hence
Z S.
(7)
Since f (1) = 1, we may set x = 1 in (1) to find that for all y R,
f (1 + f (y + 1)) + f (y) = y + 1 + f (y + 1).

(8)

Suppose that u, u + 1 S. Then setting y = u in (8) tells us that u + 2 S. Also


setting y = u 1 in (8) tells us that u 1 S. An immediate corollary of this is
u, u + 1 S u + n S
77
162

for any n Z.

(9)

Let y R. Then from (2) with x = y 1 we have y + f (y) 1 S, and from (5) we
have y + f (y) S. Hence with u = y + f (y) 1 in (7), we find that for each n Z,
y + f (y) + n S.

(10)

Using the change of variables y = x + m and n = m, we have for each m Z and


x R,
x + f (x + m) S.
(11)
If we set y = m in (1) and use (11), then for each m Z and x R we have,
f (mx) = mf (x).

(12)

If we replace y with f (y) in (10) and remember that f (f (y)) = f (y) from (4), we
find
2f (y) + n S.
(13)
Let y R and let y = 2x. Then applying (12) and (13), we are able to deduce that
f (y) + 1 = f (2x) + 1 = 2f (x) + 1 S. Hence for all y R we have
f (y) + 1 S.

(14)

Finally, put x = 1 in (1). Then using f (1 + f (1 + y)) = 1 + f (1 + y) from (14), and


f (1) = 1 from (7), we deduce that f (y) = y for all y R. This is easily seen to be
a solution.


78
163

Solution 2 (Based on the presentation by Alex Gunning, year 12, Glen Waverley
Secondary College, VIC. Alex was a Gold medallist with the 2015 Australian IMO
team.)
We are asked to find all functions f : R R such that for all x, y R,
f (x + f (x + y)) + f (xy) = x + f (x + y) + yf (x).

(1)

The case f (0) = 0 is handled as in solution 1.


It remains to deal with the case f (0) = 0. All equations that follow will hold true
for all z R.
Replacing x with y and y with x in (1) yields

f (y + f (x + y)) + f (xy) = y + f (x + y) + xf (y).

(2)

Computing the difference (1) (2), we have for all x, y R,


f (x + f (x + y)) f (y + f (x + y)) = x y + yf (x) xf (y).

(3)

Putting y = x in (3) yields


f (x) f (x) = 2x xf (x) xf (x).

(4)

Suppose that f (x) = x. Then (4) can be rearranged to yield


(x 1)(f (x) + x)) = 0.
If x = 1, then f (x) = x. If x = 1, then putting (x, y) = (1, 1) in (1), yields
f (1) = 1. Hence if S is the set of fixed points of f , we have shown that
x S x S.

(5)

Putting (x, y) = (z, 0) in (1), we find,


z + f (z) S.

(6)

Putting (x, y) = (0, z) in (1), we find,


f (z) S.

(7)

Putting (x, y) = (f (z), 0) in (1) and using (7), we find,


2f (z) S.

(8)

Put (x, y) = (z + f (z), f (z)) in (3). The LHS of (3) is f (z + 2f (z)). We also
have x = z + f (z) S from (6). And y = f (z) S from (7) and (5). Thus
yf (x) xf (y) = yx xy = 0, and so the RHS of (3) is just x y = z + 2f (z).
Hence
z + 2f (z) S.
(9)
79
164

Put (x, y) = (z, z f (z)) in (3). Note that y = z f (z) S from (6) and (5).
And x + y = f (z) S from (7) and (5). Thus (3) simplifies to
f (2x + y) f (x + 2y) = x y + yf (x) xy.
However, x + 2y = z 2f (z) S from (9) and (5). So (3) simplifies further to
f (2x + y) = 2x + y + yf (x) xy.
Writing x and y in term of z, and tidying up yields
f (z f (z)) = z f (z) + z 2 f (z)2 .

(10)

Put (x, y) = (z f (z), f (z)) in (3). The LHS of (3) equals f (z) f (2f (z)) = f (z)
from (8). Hence
f (z) = x y + yf (x) xf (y)
= z 2f (z) + yf (x) xy
f (z) z = f (z)(f (x) x)
= f (z)(z 2 f (z)2 ).

(since y = f (z) S)
(y = f (z))
(by (10) as x = z f (z))

This can be written in the form


(f (z) z)(f (z)2 + zf (z) + 1) = 0.

(11)




2
2
Solving for f (z) in (11), we find f (z) z, z+ 2 z 4 , z 2 z 4 . Since f (z) R,
we have f (z) = z for all |z| < 2. Hence (2, 2) S. Finally, from (8) we see that
z S implies 2z S. This allows us to deduce that (2k , 2k ) S for all positive
integers k. Hence S = R, and f (x) = x for all x R.


165

80

6. This was the hardest problem of the 2015 IMO. Only 11 of the 577 contestants were
able to solve this problem completely.
The authors of this problem were Ross Atkins and Ivan Guo of Australia. Ross and
Ivan were Bronze medallists with the 2003 Australian IMO Team and Ivan was a
Gold medallist with the 2004 Australian IMO Team. The problem was inspired by a
notation for juggling (Ross is also a juggler) in which each ai represents the airtime
of a ball thrown at time i, and b is the total number of balls.
Solution 1 (Alex Gunning, year 12, Glen Waverley Secondary College, VIC. Alex
was a Gold medallist with the 2015 Australian IMO team.)
Let S be the set of positive integers which are not of the form n + an for some
positive integer n. Note that S is nonempty because 1  S. Let s1 < s2 < be
the elements of S listed in increasing order.
Lemma |S| 2015.
Proof Assume that |S| 2016. Choose n so that an + n s2016 . Since an 2015,
this implies that s1 , s2 , . . . , s2016 {1, 2, . . . , n + 2015}. However, the n numbers
1 + a1 , 2 + a2 , . . . , n + an are not equal to any si and are also members of the set
{1, 2, . . . , n + 2015}. Hence {1, 2, . . . , n + 2015} contains at least n + 2016 different
numbers, contradiction.

We claim that if b = |S| and if N is larger than all members of S, then the inequality
posed in the problem statement is true.
n
Let n be any integer
satisfying
n

N
.
We
shall
find
bounds
for
j=1 (j + aj ) and

hence also for nj=1 (aj b). In what follows, let L be the following list of n + b
distinct positive integers.
1 + a1 , 2 + a2 , . . . , n + an , s1 , s2 , . . . , sb
For the lower bound,
n+bsince the n + b numbers in L are distinct, their sum is greater
than or equal to j=1 j. Hence we have
n


b


(j + aj ) +

j=1

j=1

n

j=1

sj
aj

n+b


j=1

n+b


j=n+1

b


sj

j=1

b(2n + b + 1)
s
2
n

b2 + b
s,
(aj b)
2
j=1
=

where s =

b

j=1

(1)

sj .

For the upper bound, observe that s1 , s2 , . . . , sb are b members belonging to the set
T = {1, 2, . . . , n + 1}. The remaining n + 1 b members of T must be of the form
j + aj where j n, and so are in L. The sum of these n + 1 b numbers is exactly
n+1

j=1

81
166

b

j=1

sb .

All together there are exactly n numbers of the form j + aj in L and so far we have
accounted for n + 1 b of them.

Consider the remaining b1 numbers of the form j +aj which are in L. When listed
in decreasing order, they can be no larger than
b1n+2015, n+2014, . . . , n+2015b+2,
respectively. Hence their sum is at most j=1 (n + 2016 j). Thus
n

j=1

(j + aj )

n

j=1

n

j=1

n+1

j=1

aj n + 1 +

(aj b)

b


sb +

j=1

b1

j=1

(n + 2016 j)

(b 1)(2n + 4032 b)
s
2

4033b b2 4030
s.
2

(2)

Summarising (1) and (2), we have established the following bounds for any n N .
n


4033b b2 4030
b2 + b
s
s.
(aj b)
2
2
j=1

(3)

Now let m, n be any two integers satisfying n > m N . Since also m N , (3) is
also satisfied if n is replaced by m. Thus

  n

n
m
 
 



 

(aj b) =  (aj b)
(aj b)


 

j=m+1

j=1

j=1

 2
 


 b +b
4033b b2 4030

s
s 

2
2
= (b 1)(2015 b)
2

(b 1) + (2015 b)

(AMGM)


= 10072 .

82
167

Solution 2 (A juggling interpretation solution by the authors Ross Atkins and


Ivan Guo)
Suppose you are juggling several balls using only one hand. At the ith second, if a
ball lands in your hand, it is thrown up immediately. If no ball lands, you instead
reach for an unused ball (that is, a ball that has not been thrown yet) and throw it
up. In both cases, a ball is thrown so that it will stay in the air for ai seconds. The
condition ai + i = aj + j ensures that no two balls land at the same time.
Let b be the total number of balls used. If b > 2015, then eventually a ball must
stay in the air for more than 2015 seconds, contradicting ai 2015. So b is finite
and bounded by 2015.

Select N so that all b balls have been introduced by the N th second. For all i N ,
denote by Ti the total remaining airtime of the current balls, immediately after the
ith throw is made. (That is, we calculate the remaining airtime for each current
ball, and add these values together.) Consider what happens during the next second.
The airtime of each of the b balls is reduced by 1. At the same time a ball is thrown,
increasing its airtime by ai+1 . Thus we have the equality Ti+1 Ti = ai+1 b. This
gives a nice representation of the required sum,
n


i=m+1

(ai b) = Tn Tm .

To complete the problem, it suffices to identify the maximal and minimal possible
values of the total remaining airtime Ti . Since no two balls can land at the same
time, the minimal value is 1 + 2 + + b. On the other hand, the maximal value
is 1 + 2015 + 2014 + + (2015 b + 2). (Note that there must be a ball with a
remaining airtime of 1 since something must be caught and thrown every second.)
Taking the difference between these two sums, we find that
(4032 b)(b 1) (b + 2)(b 1)

2
2
= (2015 b)(b 1)
10072
(GM AM)

|Tn Tm |

as required.

83
168

INTERNATIONAL MATHEMATICAL OLYMPIAD


RESULTS
MARK DISTRIBUTION BY QUESTION
Mark

Q1

Q2

Q3

Q4

Q5

Q6

93

256

408

91

153

521

89

151

122

36

255

11

77

12

61

34

15

21

27

18

90

72

11

12

13

20

14

265

31

30

351

30

11

Total

577

577

577

577

577

577

Mean

4.3

1.4

0.7

4.8

1.5

0.4

AUSTRALIAN SCORES AT THE IMO


Name

Q1

Q2

Q3

Q4

Q5

Q6

Score

Alex Gunning

36

Gold

Ilia Kucherov

19

Silver

Seyoon Ragavan

29

Gold

Yang Song

20

Silver

Kevin Xian

21

Silver

Jeremy Yip

23

Silver

Totals

42

26

11

42

20

148

Australian average

7.0

4.3

1.8

7.0

3.3

1.2

24.7

IMO average

4.3

1.4

0.7

4.8

1.5

0.4

13.0

The medal cuts were set at 26 for Gold, 19 for Silver and 14 for Bronze.

169

Award

SOME COUNTRY TOTALS


Rank

Country

Total

United States of America

185

China

181

South Korea

161

North Korea

156

Vietnam

151

Australia

148

Iran

145

Russia

141

Canada

140

10

Singapore

139

11

Ukraine

135

12

Thailand

134

13

Romania

132

14

France

120

15

Croatia

119

16

Peru

118

17

Poland

117

18

Taiwan

115

19

Mexico

114

20

Hungary

113

20

Turkey

113

22

Brazil

109

22

Japan

109

22

United Kingdom

109

25

Kazakhstan

105

26

Armenia

104

27

Germany

102

28

Hong Kong

101

29

Bulgaria

100

29

Indonesia

100

29

Italy

100

29

Serbia

100

170

DISTRIBUTION OF AWARDS AT THE 2015 IMO


Country

Total

Gold

Silver

Bronze

HM

Albania

37

Algeria

60

Argentina

70

Armenia

104

Australia

148

Austria

63

Azerbaijan

73

Bangladesh

97

Belarus

84

Belgium

67

Bolivia

Bosnia and Herzegovina

76

Botswana

Brazil

109

Bulgaria

100

Cambodia

24

Canada

140

Chile

12

China

181

Colombia

72

Costa Rica

53

Croatia

119

Cuba

15

Cyprus

58

Czech Republic

74

Denmark

52

Ecuador

27

El Salvador

14

Estonia

51

Finland

26

France

120

Georgia

80

Germany

102

Ghana

Greece

71

171

Country

Total

Gold

Silver

Bronze

HM

Hong Kong

101

Hungary

113

Iceland

41

India

86

Indonesia

100

Iran

145

Ireland

37

Israel

83

Italy

100

Japan

109

Kazakhstan

105

Kosovo

24

Kyrgyzstan

17

Latvia

36

Liechtenstein

18

Lithuania

54

Luxembourg

12

Macau

88

Macedonia (FYR)

45

Malaysia

66

Mexico

114

Moldova

85

Mongolia

74

Montenegro

19

Morocco

27

Netherlands

76

New Zealand

72

Nicaragua

26

Nigeria

22

North Korea

156

Norway

54

Pakistan

25

Panama

Paraguay

53

Peru

118

Philippines

87

Poland

117

172

Country

Total

Gold

Silver

Bronze

HM

Portugal

70

Puerto Rico

18

Romania

132

Russia

141

Saudi Arabia

81

Serbia

100

Singapore

139

Slovakia

97

Slovenia

46

South Africa

68

South Korea

161

Spain

47

Sri Lanka

51

Sweden

63

Switzerland

74

Syria

69

Taiwan

115

Tajikistan

57

Tanzania

Thailand

134

Trinidad and Tobago

26

Tunisia

41

Turkey

113

Turkmenistan

64

Uganda

Ukraine

135

United Kingdom

109

United States of America

185

Uruguay

16

Uzbekistan

64

Venezuela

13

Vietnam

151

39

100

143

126

Total (104 teams, 577 contestants)

173

ORIGIN OF SOME QUESTIONS

ORIGIN OF SOME QUESTIONS

Senior Contest
Question 1 was submitted by Angelo Di Pasquale.
Questions 2, 3 and 5 were submitted by Norman Do.
Question 4 was submitted by Alan Offer.
Australian Mathematical Olympiad
Questions 1, 2, 5 and 6 were submitted by Norman Do.
Question 3 was submitted by Andrei Storozhev.
Questions 4 and 7 were submitted by Angelo Di Pasquale.
Question 8 was submitted by Andrew Elvey Price.
Asian Pacific Mathematical Olympiad 2015
Question 2 was composed by Angelo Di Pasquale and submitted by the AMOC Senior Problems Committee.
International Mathematical Olympiad 2015
Question 6 was composed by Ross Atkins and Ivan Guo, and submitted by the AMOC Senior Problems
Committee. Ivan provided the following background information on the problem.
The original idea for this problem came about while Ross was reading the paper Positroid
Varieties: Juggling and Geometry by Knutson, Lam and Speyer, in which the excitation
number of a periodic juggling sequence was discovered. It seemed obvious that this was
similar to some specific elementary result that could be proven using elementary methods.
We had some difficulties in phrasing the problem in a concise self-contained way. Intuitively,
each term ai in the sequence corresponds to throwing a ball at the ith second with an air time
of ai . The inequality condition ensures that no two balls land simultaneously.
The first formulation of the problem was to show that the long-term average of the sequence
converges to an integer b, which is the total number of balls. However, the usage of limits was
inappropriate for an olympiad problem. We then came up with three more versions which
involved bounding the partial
sums. Eventually we settled on the most difficult version, with

the explicit bound of | (ai b)| 10072 . Interestingly, the term ai b can be interpreted
as the change in the total air time on the ith second, while 10072 is the difference between
maximal and minimal possible total air times, after the introduction of all b balls. The final
wording may be a little difficult for students who are unfamiliar with the construct: there
exists an N such that for all m > n > N .
It is possible to solve the problem combinatorially without invoking any physical interpretations, juggling or otherwise. Furthermore, as demonstrated by some at the IMO, the problem
can also be tackled using purely algebraic approaches. Overall, we are very happy with the
problem and we hope everyone enjoyed it.
It is worth noting that one of the authors of the paper that inspired this problem was Thomas Lam, a
member of the 1997 Australian IMO team and recipient of an IMO gold medal.

174

HONOUR ROLL
Because of changing titles and affiliations, the most senior title achieved and later affiliations are generally used,
except for the Interim committee, where they are listed as they were at the time.

Interim Committee 19791980


Mr P J OHalloran
Prof A L Blakers
Dr J M Gani
Prof B H Neumann
Prof G E Wall
Mr J L Williams

Canberra College of Advanced Education, ACT, Chair


University of Western Australia
Australian Mathematical Society, ACT,
Australian National University, ACT,
University of Sydney, NSW
University of Sydney, NSW

Mathematics Challenge for Young Australians


Problems Committee for Challenge
Dr K McAvaney
Victoria, (Director)

Mr B Henry
Victoria (Director)

Prof P J OHalloran
University of Canberra, ACT
Dr R A Bryce
Australian National University, ACT
Adj Prof M Clapper
Australian Mathematics Trust, ACT
Ms L Corcoran
Australian Capital Territory
Ms B Denney
New South Wales
Mr J Dowsey
University of Melbourne, VIC
Mr A R Edwards
Department of Education, QLD
Dr M Evans
Scotch College, VIC
Assoc Prof H Lausch Monash University, VIC
Ms J McIntosh
AMSI, VIC
Mrs L Mottershead
New South Wales
Miss A Nakos
Temple Christian College, SA
Dr M Newman
Australian National University, ACT
Ms F Peel
St Peters College, SA
Dr I Roberts
Northern Territory
Ms T Shaw
SCEGGS, NSW
Ms K Sims
New South Wales
Dr A Storozhev
Attorney Generals Department, ACT
Prof P Taylor
Australian Mathematics Trust, ACT
Mrs A Thomas
New South Wales
Dr S Thornton
South Australia
Miss G Vardaro
Wesley College, VIC

Visiting members
Prof E Barbeau
University of Toronto, Canada
Prof G Berzsenyi
Rose Hulman Institute of Technology, USA
Dr L Burjan
Department of Education, Slovakia
Dr V Burjan
Institute for Educational Research, Slovakia
Mrs A Ferguson
Canada
Prof B Ferguson
University of Waterloo, Canada
Dr D Fomin
St Petersburg State University, Russia
Prof F Holland
University College, Ireland
Dr A Liu
University of Alberta, Canada
Prof Q Zhonghu
Academy of Science, China
175

9 years; 20062015;
Member
1 year 20052006
17 years; 19902006;
Member
9 years 20072015
5 years; 19901994
23 years; 19902012
3 years; 20132015
3 years; 19901992
6 years; 20102015
8 years; 19952002
26 years; 19902015
6 years; 19901995
24 years; 19902013
14 years; 20022015
24 years; 19922015
23 years; 19932015
26 years; 19902015
2 years; 1999, 2000
3 years; 20132015
3 years; 20132015
17 years; 19992015
22 years; 19942015
20 years; 19952014
18 years; 19902007
18 years; 19982015
22 years: 19932006,
20082015
1991,
1993,
1993
1993
1992
1992,
1994
1994
1995,
1995

2004, 2008
2002

2005

2006, 2009

Dr A Gardiner
University of Birmingham, United Kingdom
Prof P H Cheung
Hong Kong
Prof R Dunkley
University of Waterloo, Canada
Dr S Shirali
India
Mr M Starck
New Caledonia
Dr R Geretschlager
Austria
Dr A Soifer
United States of America
Prof M Falk de Losada Colombia
Mr H Groves
United Kingdom
Prof J Tabov
Bulgaria
Prof A Andzans
Latvia
Prof Dr H-D Gronau
University of Rostock, Germany
Prof J Webb
University of Cape Town, South Africa
Mr A Parris
Lynwood High School, New Zealand
Dr A McBride
University of Strathclyde, United Kingdom
Prof P Vaderlind
Stockholm University, Sweden
Prof A Jobbings
United Kingdom
Assoc Prof D Wells
United States of America

Moderators for Challenge


Mr W Akhurst
Ms N Andrews
Prof E Barbeau
Mr R Blackman
Ms J Breidahl
Ms S Brink
Prof J C Burns
Mr A. Canning
Mrs F Cannon
Mr J Carty
Dr E Casling
Mr B Darcy
Ms B Denney
Mr J Dowsey
Br K Friel
Dr D Fomin
Mrs P Forster
Mr T Freiberg
Mr W Galvin
Mr M Gardner
Ms P Graham
Mr B Harridge
Ms J Hartnett
Mr G Harvey
Ms I Hill
Ms N Hill
Dr N Hoffman
Prof F Holland
Mr D Jones
Ms R Jorgenson
Assoc Prof H Lausch
Mr J Lawson
Mr R Longmuir
Ms K McAsey

New South Wales


ACER, Camberwell, VIC
University of Toronto, Canada
Victoria
St Pauls Woodleigh, VIC
Glen Iris, VIC
Australian Defence Force Academy, ACT
Queensland
New South Wales
ACT Department of Education, ACT
Australian Capital Territory
South Australia
New South Wales
Victoria
Trinity Catholic College, NSW
St Petersburg University, Russia
Penrhos College, WA
Queensland
University of Newcastle, NSW
North Virginia, USA
Tasmania
University of Melbourne, VIC
Queensland
Australian Capital Territory
South Australia
Victoria
Edith Cowan University, WA
University College, Ireland
Coffs Harbour High School, NSW
Australian Capital Territory
Victoria
St Pius X School, NSW
China
Victoria

176

1996
1997
1997
1998
1999
1999,
2000
2000
2001
2001,
2002
2003
2003,
2004
2007
2009,
2014
2015

2013

2010

2011

2012

Moderators for Challenge continued


Dr K McAvaney
Ms J McIntosh
Ms N McKinnon
Ms T McNamara
Mr G Meiklejohn
Mr M OConnor
Mr J Oliver
Mr S Palmer
Dr W Palmer
Mr G Pointer
Prof H Reiter
Mr M Richardson
Mr G Samson
Mr J Sattler
Mr A Saunder
Mr W Scott
Mr R Shaw
Ms T Shaw
Dr B Sims
Dr H Sims
Ms K Sims
Prof J Smit
Mrs M Spandler
Mr G Spyker
Ms C Stanley
Dr E Strzelecki
Mr P Swain
Dr P Swedosh
Prof J Tabov
Mrs A Thomas
Ms K Trudgian
Prof J Webb
Ms J Vincent

Victoria
AMSI, VIC
Victoria
Victoria
Queensland School Curriculum Council, QLD
AMSI, VIC
Northern Territory
New South Wales
University of Sydney, NSW
South Australia
University of North Carolina, USA
Yarraville Primary School, VIC
Nedlands Primary School, WA
Parramatta High School, NSW
Victoria
Seven Hills West Public School, NSW
Hale School, WA
New South Wales
University of Newcastle, NSW
Victoria
New South Wales
The Netherlands
New South Wales
Curtin University, WA
Queensland
Monash University, VIC
Ivanhoe Girls Grammar School, VIC
The King David School, VIC
Academy of Sciences, Bulgaria
New South Wales
Queensland
University of Capetown, South Africa
Melbourne Girls Grammar School, VIC

Mathematics Enrichment Development


Enrichment Committee Development Team (19921995)
Mr B Henry
Victoria (Chairman)
Prof P OHalloran
University of Canberra, ACT (Director)
Mr G Ball
University of Sydney, NSW
Dr M Evans
Scotch College, VIC
Mr K Hamann
South Australia
Assoc Prof H Lausch Monash University, VIC
Dr A Storozhev
Australian Mathematics Trust, ACT
Polya Development Team (19921995)
Mr G Ball
University of Sydney, NSW (Editor)
Mr K Hamann
South Australia (Editor)
Prof J Burns
Australian Defence Force Academy, ACT
Mr J Carty
Merici College, ACT
Dr H Gastineau-Hill
University of Sydney, NSW
Mr B Henry
Victoria
Assoc Prof H Lausch Monash University, VIC
Prof P OHalloran
University of Canberra, ACT
177

Dr A Storozhev
Australian Mathematics Trust, ACT
Euler Development Team (19921995)
Dr M Evans
Scotch College, VIC (Editor)
Mr B Henry
Victoria (Editor)
Mr L Doolan
Melbourne Grammar School, VIC
Mr K Hamann
South Australia
Assoc Prof H Lausch Monash University, VIC
Prof P OHalloran
University of Canberra, ACT
Mrs A Thomas
Meriden School, NSW
Gauss Development Team (19931995)
Dr M Evans
Scotch College, VIC (Editor)
Mr B Henry
Victoria (Editor)
Mr W Atkins
University of Canberra, ACT
Mr G Ball
University of Sydney, NSW
Prof J Burns
Australian Defence Force Academy, ACT
Mr L Doolan
Melbourne Grammar School, VIC
Mr A Edwards
Mildura High School, VIC
Mr N Gale
Hornby High School, New Zealand
Dr N Hoffman
Edith Cowan University, WA
Prof P OHalloran
University of Canberra, ACT
Dr W Pender
Sydney Grammar School, NSW
Mr R Vardas
Dulwich Hill High School, NSW
Noether Development Team (19941995)
Dr M Evans
Scotch College, VIC (Editor)
Dr A Storozhev
Australian Mathematics Trust, ACT (Editor)
Mr B Henry
Victoria
Dr D Fomin
St Petersburg University, Russia
Mr G Harvey
New South Wales
Newton Development Team (20012002)
Mr B Henry
Victoria (Editor)
Mr J Dowsey
University of Melbourne, VIC
Mrs L Mottershead
New South Wales
Ms G Vardaro
Annesley College, SA
Ms A Nakos
Temple Christian College, SA
Mrs A Thomas
New South Wales
Dirichlet Development Team (20012003)
Mr B Henry
Victoria (Editor)
Mr A Edwards
Ormiston College, QLD
Ms A Nakos
Temple Christian College, SA
Mrs L Mottershead
New South Wales

Australian Mathematical Olympiad Committee


The Australian Mathematical Olympiad Committee was founded at a meeting of the Australian Academy of
Science at its meeting of 23 April 1980.
* denotes Executive Position
Chair*
Prof B H Neumann
Prof G B Preston
Prof A P Street
Prof C Praeger
Deputy Chair*
Prof P J OHalloran
Prof A P Street
Prof C Praeger,

Australian National University, ACT


Monash University, VIC
University of Queensland
University of Western Australia

7 years; 19801986
10 years; 19861995
6 years; 19962001
14 years; 20022015

University of Canberra, ACT


University of Queensland
University of Western Australia

15 years; 19801994
1 year; 1995
6 years; 19962001

178

Assoc Prof D Hunt


Executive Director*
Prof P J OHalloran
Prof P J Taylor
Adj Prof M G Clapper

University of New South Wales

14 years; 20022015

University of Canberra, ACT


University of Canberra, ACT
University of Canberra, ACT

15 years; 19801994
18 years; 19942012
3 years; 20132015

Australian Defence Force Academy, ACT



Victorian Chamber of Mines, VIC

9 years; 19801988
4 years; 19891992
6 years; 19931998

Australian Defence Force Academy, ACT


University of Canberra, ACT
CPA
Monash University, VIC
Australian National University, ACT
The King David School, VIC

8
2
5
8
6
7

Secretary
Prof J C Burns
Vacant
Mrs K Doolan

Treasurer*
Prof J C Burns
Prof P J OHalloran
Ms J Downes
Dr P Edwards
Prof M Newman
Dr P Swedosh

years;
years;
years;
years;
years;
years;

19811988
19891990
19911995
19952002
20032008
20092015

Director of Mathematics Challenge for Young Australians*


Mr J B Henry
Dr K McAvaney

Deakin University, VIC


Deakin University, VIC

17 years; 19902006
10 years; 20062015

Chair, Senior Problems Committee


Prof B C Rennie
Mr J L Williams
Assoc Prof H Lausch
Dr N Do

James Cook University, QLD


University of Sydney, NSW
Monash University, VIC
Monash University, VIC

1 year; 1980
6 years; 19811986
27 years; 19872013
2 years; 20142015

Director of Training*
Mr J L Williams
Mr G Ball
Dr D Paget
Dr M Evans
Assoc Prof D Hunt
Dr A Di Pasquale

University of Sydney, NSW


University of Sydney, NSW
University of Tasmania
Scotch College, VIC
University of New South Wales
University of Melbourne, VIC

7 years; 19801986
3 years; 19871989
6 years; 19901995
3 months; 1995
5 years; 19962000
15 years; 20012015

University of Sydney, NSW


University of New South Wales

5 years; 19811985
9 years; 1986, 1989, 1990, 1996

Monash University, VIC


University of Tasmania
University of Melbourne, VIC
University of New South Wales

2 years; 1987, 1988


5 years; 19911995
13 years; 20022010, 20122015
1 year; 2011

Team Leader
Mr J L Williams
Assoc Prof D Hunt
2001
Dr E Strzelecki
Dr D Paget
Dr A Di Pasquale
Dr I Guo

Deputy Team Leader


Prof G Szekeres
Mr G Ball
Dr D Paget
Dr J Graham
Dr M Evans
Dr A Di Pasquale
Dr D Mathews
Dr N Do
Dr I Guo

University of New South Wales


University of Sydney, NSW
University of Tasmania
University of Sydney, NSW
Scotch College, VIC
University of Melbourne, VIC
University of Melbourne, VIC
University of Melbourne, VIC
University of New South Wales
179

2
7
1
3
3
5
3
4
4

years; 19811982
years; 19831989
year; 1990
years; 19911993
years; 19941996
years; 19972001
years; 20022004
years; 20052008
years; 200910, 20122013

Mr G White
Mr A Elvey Price

University of Sydney, NSW


Melbourne University, VIC

1 year; 2011
2 years; 20142015

State Directors
Australian Capital Territory
Prof M Newman
Australian National University
Mr D Thorpe
ACT Department of Education
Dr R A Bryce
Australian National University
Mr R Welsh
Canberra Grammar School
Mrs J Kain
Canberra Grammar School
Mr J Carty
ACT Department of Education
Mr J Hassall
Burgmann Anglican School
Dr C Wetherell
Radford College
New South Wales
Dr M Hirschhorn
University of New South Wales
Mr G Ball
University of Sydney, NSW
Dr W Palmer
University of Sydney, NSW
Northern Territory
Dr I Roberts
Charles Darwin University
Queensland
Dr N H Williams
University of Queensland
Dr G Carter
Queensland University of Technology
Dr V Scharaschkin
University of Queensland
Dr A Offer
Queensland
South Australia/Northern Territory
Mr K Hamann
SA Department of Education
2013
Mr V Treilibs
SA Department of Education
Dr M Peake
Adelaide
Dr D Martin
Adelaide
Tasmania
Mr J Kelly
Tasmanian Department of Education
Dr D Paget
University of Tasmania
Mr W Evers
St Michaels Collegiate School
Dr K Dharmadasa
University of Tasmania
Victoria
Dr D Holton
University of Melbourne
Mr B Harridge
Melbourne High School
Ms J Downes
CPA
Mr L Doolan
Melbourne Grammar School
Dr P Swedosh
The King David School
Western Australia
Dr N Hoffman
WA Department of Education
Assoc Prof P Schultz University of Western Australia
19961999
Assoc Prof W Bloom Murdoch University
Dr E Stoyanova
WA Department of Education
Dr G Gamble
University of Western Australia

1 year; 1980
2 years; 19811982
7 years; 19831989
1 year; 1990
5 years; 19911995
17 years; 19952011
2 years; 20122013
2 years; 20142015
1 year; 1980
16 years; 19811996
19 years; 19972015
2 years; 20142015
21 years; 19802000
10 years; 20012010
4 years; 20112014
1 year; 2015
19 years; 19801982, 19912005,
8 years; 19831990
8 years; 20062013
2 years; 20142015
8 years; 19801987
8 years; 19881995
9 years; 19952003
12 years; 20042015
3 years; 19801982
1 year; 1982
6 years; 19831988
9 years; 19891998
18 years; 19982015
3 years; 19801982
14 years; 19831988, 19911994,
2 years; 19891990
7 years; 1995, 20002005
10 years; 20062015

Editor
Prof P J OHalloran
Dr A W Plank
Dr A Storozhev
Editorial Consultant
Dr O Yevdokimov

University of Canberra, ACT


University of Southern Queensland
Australian Mathematics Trust, ACT

1 year; 1983
11 years; 19841994
15 years; 19942008

University of Southern Queensland

7 years; 20092015

180

Other Members of AMOC (showing organisations represented where applicable)


Mr W J Atkins
Dr S Britton
Prof G Brown
Dr R A Bryce

Mr G Cristofani
Ms L Davis
Dr W Franzsen
Dr J Gani
Assoc Prof T Gagen
Ms P Gould
Prof G M Kelly
Prof R B Mitchell
Ms Anna Nakos
Mr S Neal
Prof M Newman


Prof R B Potts
Mr H Reeves

Mr N Reid
Mr R Smith
Prof P J Taylor
Prof N S Trudinger
Assoc Prof I F Vivian
Dr M W White

Australian Mathematics Foundation


University of Sydney, NSW
Australian Academy of Science, ACT
Australian Mathematical Society, ACT
Mathematics Challenge for Young Australians
Department of Education and Training
IBM Australia
Australian Catholic University, ACT
Australian Mathematical Society, ACT
ANU AAMT Summer School
Department of Education and Training
University of Sydney, NSW
University of Canberra, ACT
Mathematics Challenge for Young Australians
Department of Education and Training
Australian National University, ACT
Mathematics Challenge for Young Australians
(Treasurer during the interim)
University of Adelaide, SA
Australian Association of Maths Teachers
Australian Mathematics Foundation
IBM Australia
Telecom Australia
Australian Mathematics Foundation
Australian Mathematical Society, ACT
University of Canberra, ACT
IBM Australia

18 years; 19952012
8 years; 19901998
10 years; 1980, 19861994
10 years; 19911998
13 years; 19992012
2 years; 19931994
4 years; 19911994
9 years; 19901998
1980
6 years; 19931998
2 years; 19951996
6 years; 19821987
5 years; 19911995
13 years; 20032015
4 years; 19901993
15 years; 19861998
10 years; 19992002,
20092014
1 year; 1980
11 years; 19881998
20142015
3 years; 19881990
5 years; 19901994
6 years; 19901994, 2013
3 years; 19861988
1 year; 1990
9 years; 19801988

Associate Membership (inaugurated in 2000)


Ms S Britton
Dr M Evans
Dr W Franzsen
Prof T Gagen
Mr H Reeves
Mr G Ball

16
16
16
16
16
16

years;
years;
years;
years;
years;
years;

20002015
20002015
20002015
20002015
20002015
20002015

AMOC Senior Problems Committee


Current members
Dr N Do

M Clapper
Dr A Di Pasquale
Dr M Evans
Dr I Guo
Dr J Kupka
Dr K McAvaney
Dr D Mathews
Dr A Offer
Dr C Rao
Dr B B Saad
Dr J Simpson
Dr I Wanless

Monash University, VIC (Chair)


(member)
Australian Mathematics Trust
University of Melbourne, VIC
Australian Mathematical Sciences Institute, VIC
University of Sydney, NSW
Monash University, VIC
Deakin University, VIC
Monash University, VIC
Queensland
IBM Australia
Monash University, VIC
Curtin University, WA
Monash University, VIC

181

2 years; 20142015
11 years; 20032013
3 years; 2013-2015
15 years; 20012015
26 years; 19902015
8 years; 20082015
13 years; 20032015
20 years; 19962015
15 years; 20012015
4 years; 20122015
16 years; 20002015
22 years; 19942015
17 years; 19992015
16 years; 20002015

Previous members
Mr G Ball
Mr M Brazil
Dr M S Brooks
Dr G Carter
Dr J Graham
Dr M Herzberg
Assoc Prof D Hunt
Dr L Kovacs
Assoc Prof H Lausch

Dr D Paget
Prof P Schultz
Dr L Stoyanov
Dr E Strzelecki
Dr E Szekeres
Prof G Szekeres
Em Prof P J Taylor
Dr N H Williams

University of Sydney, NSW


LaTrobe University, VIC
University of Canberra, ACT
Queensland University of Technology
University of Sydney, NSW
Telecom Australia
University of New South Wales
Australian National University, ACT
Monash University, VIC (Chair)
(member)
University of Tasmania
University of Western Australia
University of Western Australia
Monash University, VIC
University of New South Wales
University of New South Wales
Australian Capital Territory
University of Queensland

16 years; 19821997
5 years; 19901994
8 years; 19831990
10 years; 20012010
1 year; 1992
1 year; 1990
29 years; 19862014
5 years; 19811985
27 years; 19872013
2 years; 2014-2015
7 years; 19891995
8 years; 19932000
5 years; 20012005
5 years; 19861990
7 years; 19811987
7 years; 19811987
1 year; 2013
20 years; 19812000

Mathematics School of Excellence


Dr S Britton
Mr L Doolan
Mr W Franzsen
Dr D Paget
Dr M Evans
Assoc Prof D Hunt
Dr A Di Pasquale

University of Sydney, NSW (Coordinator)


Melbourne Grammar, VIC (Coordinator)
Australian Catholic University, ACT (Coordinator)
University of Tasmania (Director)
Scotch College, VIC
University of New South Wales (Director)
University of Melbourne, VIC (Director)

2 years; 19901991
6 years; 1992, 19931997
2 years; 19901991
5 years; 19901994
1 year; 1995
4 years; 19961999
16 years; 20002015

International Mathematical Olympiad Selection School


Mr J L Williams
Mr G Ball
Mr L Doolan
Dr S Britton
Mr W Franzsen
Dr D Paget
Assoc Prof D Hunt
Dr A Di Pasquale

University of Sydney, NSW (Director)


University of Sydney, NSW (Director)
Melbourne Grammar, VIC (Coordinator)
University of Sydney, NSW (Coordinator)
Australian Catholic University, ACT (Coordinator)
University of Tasmania (Director)
University of New South Wales (Director)
University of Melbourne, VIC (Director)

182

2 years; 19821983
6 years; 19841989
3 years; 19891991
7 years; 19921998
8 years; 19921996, 19992001
6 years; 19901995
5 years; 19962000
15 years; 20012015