The Back Files: Published by The Canadian Mathematical Society
The Back Files: Published by The Canadian Mathematical Society
The Back Files: Published by The Canadian Mathematical Society
Mathematicorum
Published by the Canadian Mathematical Society.
http://crux.math.ca/
Special EUREKA
MORLEY
Issue
Vol. 3, No. 10
December 1977
Sponsored by
Carleton-Ottawa Mathematics Association Mathematique d'Ottawa-Carleton
A Chapter of the Ontario Association for Mathematics Education
Publie par le College Algonquin
EUREKA is published monthly (except July and August). The yearly subscription
rate for ten issues is $8.00 in Canadian or U.S. dollars ($1.50 extra for delivery
by first-class mail). Back issues: $1.00 each. Bound volumes: Vol. 1-2 (combined),
$10.00; Vol. 3, $10.00. Cheques or money orders, payable to EUREKA, should be sent
to the managing editor.
All communications about the content of the magazine (articles, problems,
solutions, permission to reprint, etc.) should be sent to the editor. All changes
of address and inquiries about subscriptions and back issues should be sent to the
managing editor.
Editor: Leo Sauve, Mathematics Department, Algonquin College, 281 Echo Drive,
Ottawa, Ontario, K1S IN3.
Managing Editor: F.G.B. Maskell, Mathematics Department, Algonquin College,
200 Lees Ave., Ottawa, Ontario, K1S 0C5.
Typist-compositor: Nancy Makila.
CONTENTS
A Topical Integral Leon Bankoff 272
Presenting the Morley Issue of EUREKA 272
On the Intersections of the Trisectors of the Angles of a Triangle . F. Morley 273
Notes on Morley's Proof of his Theorem on Angle Trisectors Dan Pedoe 276
Robson's Proof of Morley*s Theorem 280
A List of References to the Morley Theorem C O . Oakley and C.W. Trigg 281
An Elementary Geometric Proof of the Morley Theorem Dan Sokolowsky 291
The Beauty and Truth of the Morley Theorem Leon Bankoff 29'4
Problems - Problemes 297
Solutions 298
La mort de Condorcet Alain Decaux 300
The Twelve Days of Christmas Charles W. Trigg 300
- 271 -
- 272 -
A TOPICAL INTEGRAL
LEON BANKOFF, Los Angeles, California
The readers of EUREKA may find it interesting at this time to evaluate with me
the following integral:
[2fe (r ) [f8
I = M — xdx\y + x dx.
a
I h J Jo
I found it in an old notebook of mine and no longer recall its source.
We have
J
a
Because of its historical interest, Morley's own proof of his theorem is given
in the first of the articles that follow, exactly as it appeared in [2J. (The
printing is too poor to reproduce it photographically.) It is an understatement to
say that Morley's proof is hard to understand. In the article which immediately
follows, Dan Pedoe attempts to shed some light on Morley's convoluted thinking.
The remaining articles on Morley's Theorem culminate in what is almost certainly
the most complete list of Morley references in existence.
REFERENCES
1. H.S.M. Coxeter and S.L. Greitzer, Geometry Revisited, Random House and The
L.W. Singer Co., 1967, p. 47.
2. Frank Morley, On the intersections of the trisectors of the angles of a
triangle, Journal of the Mathematical Association of Japan for Secondary Education,
6 (December 1924), 260-262.
j'c ;'; s's
- 273 -
2
6o ON THE INTERSECTIONS OF THE TRISECTORS
OF THE ANGLES OF A TRIANGLE.
By
Professor FRANK MORLEY.
(From a letter directed to Prof. T. Hayashi.)
Whence if 9^ are the angles which these tangents make with any
fixed line, and $ the angle of x itself,
34) = B 1 + e 2 + Q3 (i)
261
The image y of any points x in the tangent is given by
y - 3t + 3t2 -xt3 = 0.
y = 3(t-tA).
Hence, if
y = 2pe so that y = 2pe"^(i),
we have
e2™ -- -t\
t + l/£ = -2 COS 2o)/3,
and p = 3 sinw/3 (2)
This is the line-equation of the
cardioid. The equation p~a sin yw
for any cycloidal curve is given
in some of the older books (for
instance, in Edwards, Differential
Calculus), so that we might begin
with equation (2).
If then p 1 5 p 2 , p 3 are perpendiculars from the centre on 3
tangents, and u) l9 w 2 , o)3 the angles of these perpendiculars, since
I sin— 1 sin,002-^3 = o,
3 "'" 3
we have
,^2-033 _
IPlsini 3
0.
i\ - Ai 7T - Ao -IT -A3
Pl sin- + p 2 sin- + p 3 sin 0,
3
262
30 = e1 + e 2 + e 3 .
They are thus 3 sets
A
of 3 parallel lines, i
forming equilateral
triangles. The centre
changes from one
line to another when
A A
one of the lines is a 2 3 I
double tangent.
Consider in particular the cardioids which lie inside the
triangle. Let ol be the centre of a cardioid with double tangent
A2A^. We have from (l) I
angle A3A2Ox = A2/3,
and we have seen that the 3 lines Ox02, 0203, 03Ox form an I
equilateral triangle. I
That was the argument. Verification is naturally a much simpler I
matter. If you think above worth printing I shall be very pleased I
to have it appear in a Japanese journal. I
Further should the matter of the memoir referred to be of I
interest I shall be glad to send a copy, with a correction, for the I
use of "direction lines" there is not clear. I
. _ , I
- 276 -
Frank Morley was a remarkable geometer, but there cannot be many who have
found it easy to follow his thoughts. His Inversive Geometry (Frank Morley and
F.V. Morley, Chelsea, 1954) is a book with the most remarkable insights, but I,
at any rate, find it almost incomprehensible. The following notes on Morley's
proof of his celebrated theorem are very tentative first steps towards the
elucidation of his work.
In this paper, written for a Japanese journal devoted to secondary education,
Morley changes his notation a number of times. He uses 9 for the angle made by a
tangent to the cardioid with the ic-axis (p. 260), then changes to to on p. 261, and
he then uses w for the angle made with the #~axis by the perpendicular from the
centre of the cardioid onto a tangent. I shall use the notation I am accustomed
to, and shall then identify my results with Morley's.
We consider the first statement, p. 260: "The cardioid is mapped on the unit
circle by an equation x = 2 t - t 2 , x a complex number, t a complex number such that
I*I = i."
We suppose that a cardioid is
traced out by a point P fixed on the
circumference of a unit circle (see
figure) which rolls without slipping
on a fixed unit circle. The centre 0
of the fixed circle, taken as origin
of coordinates, is called the centre
of the cardioid. If N is the point
of contact, and ON meets the rolling
circle again in A, then N is the
instantaneous centre of motion for the
rolling circle, so that the tangent
at P to the cardioid is perpendicular
to NP, and is therefore PA.
If 0 is the angle H0x9 the coordinates (xp9 yp) of P are:
X = 2 COS 0 - COS 20,
z/p = 2 sin 0 - sin 20,
- 277 -
so that
x = x~ +iyp = 2(cos 0 + i sin 0) - (cos 0 +i sin 0 ) 2
I X X 1 I
\ 3t 3t 1 = 0 ,
I 2t - t2 2t -i2 1 I
2
where X is any point on the tangent at x = 2t - t . (Morley uses the same x for
both points.) After expansion, and use of the equation tt = l, which leads to
The three tangents to the cardioid from a point X correspond to points tl9
t2, t3 on the unit circle which are such that
X
Hence, if X = r(cos cj) + i sin (j>),
tjt 2 t 3 = cos 2<J> + i sin 2(j>,
and therefore
0j + 0 2 + 0 3 = 2<|).
If \\) is the angle made by the tangent at P with Ox, the figure shows that this
angle is 30/2, the tangent being the line PA, and hence the angles if;, made by the
tangents which pass through the point X satisfy the relation:
3
=
$i +^2+^3 2"(ei +e
2+ e a )
= 3*.
This is Morley's fundamental equation (l), p. 260.
If we now turn to p. 262 of Morley's paper, and consider the locus of the
centre of cardioids which touch the sides of a given triangle ABC, we now prove
that the points at infinity on this locus coincide with the points at infinity on
the sides of a determinate equilateral triangle.
- 278 -
Let 0 be the centre of a cardioid touching the sides of triangle ABC which
is at a great distance from the triangle, and let J be a point near the triangle.
The tangents from x to the mammoth cardioid must be very nearly parallel to the
sides of triangle ABC ("so that the triangle behaves like a point"). Hence, in
equation (l), the angle <f> which determines the direction of Ox is one-third of
the sum of the \p., where the ip. correspond to the sides of triangle ABC.
Since anyty.is taken modulo 2TT, and we are dividing by 3, we obtain 3
directions for <J> which differ by 2TT/3.
If we know that the centre locus consists entirely of straight lines, these
must form sets which are parallel to the sides of a determinate equilateral
triangle.
To prove that the centre locus consists of lines, Morley obtains the pedal
(p, \p) equation of a cardioid. From the figure, if OQ is the perpendicular from
0 onto AP,
p = OQ = 0 A s i n | = 3sin|. (2)
Morley remarks that this can be obtained directly.
Now there is some very clever manipulation! Morley wishes to move from
equation (2) to a trilinear equation for the locus of the centre. It is easily
verified that for any P, Q and R,
sin P sin (Q - R) + sinQ sin (R - P) + sinR sin (P - Q) = o,
so that
sinfesin^li+ ... + ... = o.
Using (2), we can write this
where p x , p 2 , p 3 are the perpendiculars from the centre of the cardioid onto
three tangents which make angles \pl9 ty2 and \p with Ox.
This must be Morley*s equation (3), the (3) having been omitted in his paper.
These perpendiculars p. are the trilinear coordinates (see Appendix) of 0 with
respect to the triangle ABC, and the (ij>. - ij; .)/3 are related to the angles of the
triangle. The fundamental relation connecting trilinear coordinates is
apx +bp2 \-cp3 = 2A,
where a, b and c are the sides of the triangle, and A its area.
The equation (3) in trilinear coordinates is that of a line, and hence
- 279 -
Morley has proved that the locus of the centre of a cardioid which touches the
sides of a given triangle is a set of lines.
In the last paragraph of p. 261 Morley interchanges the angles made by the
tangents with the ;c-axis with the angles made by the perpendiculars onto the
tangents, and remarks: "...bearing in mind that in (3) the angles must have a
sum congruent to o..." (which is not too clear), he obtains 9 lines for the locus
of the centre.
Verifying, in trilinear coordinates, that these 9 lines are parallel in sets
of 3 would be a formidable task, since lines are parallel in these coordinates if
they intersect on the line at infinity,
apl +bp2 +op3 = 0.
But, as we have seen, Morley overcomes this difficulty by using equation (l) in a
remarkable manner.
Finally, how does all this fit in with the trisectors of the angles of
triangle ABC? Once again, on p. 262, a brilliant geometrical statement: "The
centre changes from one line to another when one of the lines is a double tangent."
The use of "line" is ambiguous, since it is used for the centre locus and for the
lines forming the triangle of tangents, but the figure shows what he means. Using
equation (1) again, Morley shows that the vertices of his equilateral triangle are
the intersections of trisectors of the angles of triangle ABC.
Morley himself adds the final touch: "Verification is naturally a much
simpler matter."
Appendix. All that need be said about "trilinear coordinates" is that it is
not difficult to show that a linear equation in (p is p 2 , p 3 ) represents a line.
In fact, if the sides of the triangle are taken in the normal form as
x c o s a . + usina. - a . - 0 (i - 1, 2, 3),
then we can take
=
Vi tfpcosou + i/pSincu - qi
for the trilinear coordinates (p 1? p 2 , p 3 ) of P = (xp9 yp)9 and therefore a linear
homogeneous equation in the p. produces a linear equation in the (xp, y p ) .
Where else?
The Managing Editor of EUREKA, F.G.B. Maskell, resides at 1332 Morley Blvd.,
Ottawa, Ontario.
- 280 -
nB. David C. Kay, College Geometry, Holt, Rinehart & Winston, Inc., New
York, 1969, pp. 15-17.
12B. E.H. Lockwood, A Book of Curves, Cambridge Univ. Press, New York, 1971,
p. 78.
13B. F. Morley and F.V. Morley, Inversive Geometry, Ginn & Co., Boston, 1933;
reissued by Chelsea Publishing Co., Bronx, N.Y., 1954, p. 244.
1MB. William Schaaf, A Bibliography of Recreational Mathematics, v. 2, The
National Council of Teachers of Mathematics, Reston, VA., 1970, pp. 61-62.
15B. F. Schuh, Leerboek der vlakke driehoeksmeting, The Hague, 1939.
16B. James R. Smart, Modern Geometries, Brooks/Cole Publishing Co., Monterey,
CA., 1973, pp. 152-154.
17B. J. Steiner, Gesammelte Werke, v. 1, 2nd ed., Chelsea Publishing Co.,
Bronx, N.Y., 1971.
18B. K. Strubecker, Einfuhrung in die hohere Mathematik, Bd. 1 , Gvundlagen,
R. Oldenbourg, Munchen, 1956, p. 595 (Proof by H. Dorrie).
97G. Haim Rose, A simple proof of Morley's theorem, Amev. Math Monthly*
71 (1964) 771-773.
98PP. Charles Salkind, Problem E 1030 [1952, 465; 1974, m o ] , Morley
polygons, Amev. Math. Monthly* 82 (1975) 1010-1011.
99R,CS. K.R.S. Sastry, Constellation Morley, Math. Mag., 47 (1974) 15-22.
Only outline of proofs suggested.
looT,R. John Satterly, The Morley triangle and other triangles, Sch. Sci.
and Math., 55 (1955) 685-701.
101T. M. Satyanarayana, Solution to problem 16381 (Morley's theorem), The
Educational Times, New Series, Vol. 61, July 1 (1908) 308 (possibly the earliest
proof (along with [36, 42])); Mathematical Questions and Solutions from "The
Educational Times", New Series, 15 (1909) 23.
102R,CV. R. Sibson, Cartesian geometry of the triangle and hexagon, Math.
Gas., 44 (1960) 83-94.
103R. James R. Smart, The n-sectors of the angles of a square, Math. Teacher,
60 (1967) 459-463.
104R. , Eight new Morley-type theorems, Jour. California Math. Coun.,
2 (1977) 10-15.
105T. W.R. Spickerman, An extension of Morley's theorem, Math. Mag., 44
(1971) 191-192.
106PG. J. Strange, A generalization of Morley's theorem, Amer. Math. Monthly,
81 (1974) 61-63.
107R. F. Glanville Taylor, The relation of Morley's theorem to the Hessian
axis and the circumcentre, Pvoc. Edinburgh Math. Soc, 32 (1913-1914) 132-135.
108CS. F.G. Taylor and W.L. Marr, The six trisectors of each of the angles of
a triangle, Proc. Edinburgh Math. Soc, 32 (1913-1914) 119-131. (Possibly first to
give complete solution.)
109G,T. V. Thebault, Recreational geometry: the triangle, Scripta Math., 22
(1956) 14-30, 97-105.
110R. A. Vandeghen, A note on Morley's theorem, Amer. Math. Monthly, 72 (1965)
638-639.
lllG,IP. K. Venkatachaliengar, An elementary proof of Morley's theorem, Amer.
Math. Monthly, 65 (1958) 612-613.
112CC. P.S. Wagner, An extension to Clifford's chain, Amer. Jour, of Math.,
51 (1929) 473-481.
113T. R.J. Webster, Morley's triangle theorem, Math. Mag., 43 (1970) 209-210.
- 289 -
CHARLES W. TRIGG
Professor Emeritus, Los Angeles City College
117. Anon., Morley*s Theorem, Indiana School Mathematics Journal, 10, No.3
(February 1975) 1-3.
118. H.F. Baker, A Theorem due to Prof. F. Morley, Mathematical Gazette,
24, No. 261 (October 1940) 284-286.
119. F.C. Boon, Morley's triangle, Mathematical Gazette* 17 (May 1933)
126-127.
120. A.G. Burgess, Concurrencies of lines joining vertices of a triangle
to opposite vertices of triangles on its sides, Proceedings Edinburgh Mathematical
Society, 32 (1914) 58-64.
121. T. Dantzig, An elementary proof of a theorem due to F. Morley, American
Mathematical Monthly, 23 (September 1916) 246-248.
122. U.P. Davis, [Solution of Problem 581], School Science and Mathematics,
19 (1919) 563-564.
123. H.G. Forder, A School Geometry, Cambridge, London, 1930, p. 178.
124. B. Gambier, Bulletin des sciences mathematiques, 61 (1937) 360-368.
125. Bertrand Gambier, Trisectrices des angles d'un triangle, Mathesis,
(1949) 174-208.
126. Martin Gardner, New Mathematical Diversions from Scientific American,
Simon and Schuster, New York, 1966, pp. 198, 206.
127. N.M. Gibbins, The Non-Equilateral Morley Triangles, Mathematical Gazette,
26, No. 269 (May 1942) 81-86.
128. R. Goormaghtigh, [Bibliography], Sphinx, 9 (1939) 46.
129. A.M. Harding, [Trigonometric solution of geometry problem 431], American
Mathematical Monthly, 21 (June 1914) 193-194.
- 290 -
The sum of these angles is 3 6 0 ° - 2 ( 3 + y ) ; hence / PXQ = 2(3 + y) = 120° - 2a. Since
the angles at P and Q are equal, and that at 0 is 6a, it follows that the angles at
P, Q, X are all equal to 120° - 2a. Hence we state our lemma as follows:
LEMMA. In quadrilateral OPXQ (see Figure 2), suppose OP = OQ,, XP = XQ, and the
angles at 0 and X are 6a and 120° - 2a respectively. Let K denote the circle with
center 0., radius OP,, and let w denote the arc of K subtended by the angle at 0.
Then the perpendicular bisectors of XP, XQ meet w at points Y, 1, respectively3
such that Y_, Z trisect w and A XYZ is equilateral.
P,lt
- 294 -
Finally, PY = XY and ZQ = ZX, so that XY = YZ = ZX and A XYZ is equilateral. This
completes the proof of the lemma, and Theorem l follows.
Now, referring back to Figure 1, since /YXZ = 60° we have ZPXY = 30° - a, and
hence /BXY = (90° - B) + (30° - a) = 120° - (a + 3). Since ZPBX = 23, XY meets AB at
an angle of 2a + y, BC at an angle of a + 2y, and AC at an angle of |a-y|- Similarly
XZ meets AC at an angle of 2a+ 3, etc.
The same method as was used to prove Theorem l will also verify it when
"interior angles" is replaced by "exterior angles" or "reflex angles." The reader
will have no difficulty arriving at a suitably modified lemma for these two cases,
and the proofs are virtually the same. Also, an angle count similar to that in
the preceding paragraph will show that in all cases the corresponding sides of the
three Morley triangles meet the sides of A ABC at the same angles, thereby proving
the following
emotional excitement subsides, reason steps in and demands some verification for
what our eyes and brain seem to perceive. Going along with Keats, we believe
that "Beauty is truth, truth beauty 9 " and we feel the need for cerebral evidence
to confirm intuitive suspicions. Is the inner triangle really equilateral? Does
this happen with all parent triangles? Is a convincing proof available? Does one
need a familiarity with higher mathematics to achieve the intellectual satisfaction
complementing our esthetic pleasure? While the mathematical sophisticate may revel
in proofs by complex numbers, involution, or Brianchon's Theorem, we wonder what
sort of proof we can offer the high school student, the bright and eager neophyte
on the threshold of mathematical exploration and enlightenment.
These questions can now be answered more readily than ever before. The current
issue of EUREKA presents the most comprehensive collection of references on the
Morley Theorem ever to be assembled and the interested researcher should have no
trouble gleaning from the list enough accessible sources to furnish a large variety
of proofs. A few comments on the diversity of the demonstrations and the relative
elegance of the various types of proofs may be helpful.
As one would expect, the long list of references contains many duplications
of previously published proofs, either through rediscovery or by worthy attempts
at refinement and clarification. Some of the proofs suffer from excessive brevity
and insufficient clarity; others are unwieldy and cumbersome and offend our sense
of elegance. Some proofs are direct, others indirect. Some are primarily trig-
onometrical, others purely geometrical (elementary or advanced). Preference for
one type of proof or another becomes a matter of individual taste.
A preliminary classification of proofs results in the establishment of two
basic categories, direct and indirect. Here we use the term indirect not in its
customary meaning of "reductio ad absurdum" but in the sense of a reversal of the
usual sequence of steps from hypothesis to conclusion. Examples of indirect proofs
are those by Naraniengar, Chepmell, Boon, Grossman, Davis, Dobbs and Child, to name
a few. These proofs start with the foreknowledge that the internal triangle is
equilateral and, with subsequent constructions based on the known values of the
surrounding angles, lead to the demonstration that certain rays emanating from
the vertices of the inner triangle actually converge to form the outer triangle.
This procedure is essentially a proof not of the theorem as stated by Morley but
of its converse. An analogous situation occurs with the Steiner-Lehmus Theorem
regarding two equal base angle bisectors [see EUREKA 2 (1976) 19-24]. The easily
established converse does not provide a legitimate proof of the main theorem.
- 296 -
Among the direct proofs, the preponderance of those published are trigonomet-
rical. We can account for this top-heaviness by noting that trigonometry is the
ideal tool for handling submultiple angles. Theoretically we should be able to
convert any trigonometrical argument to one involving synthetic geometry, but in
practice this would surely result in a muddy proof. We find considerable variation
in the trigonometrical approaches. Some are directed toward the computation of the
sides of the equilateral triangle (e.g., Neuberg, Letac, Thebault), while others
by-pass this method and go directly to the calculation of the angles surrounding
the three pertinent intersections, leaving 60° for each vertex of the Morley tri-
angle (e.g., Satyanarayana, Bankoff and others). Letac's solution (Sphinx, 1939)
achieves a deceptive simplicity by merely omitting steps essential for a clear
understanding—again a case of brevity at the expense of clarity.
Excellent direct geometrical solutions have been constructed using isogonal
conjugates, cross ratios, Desargues' theorem, Menelaus' theorem, and complex
numbers. More advanced mathematicians find these proofs particularly attractive
since they combine concision with precision. Examples of these techniques may
be found among the proofs devised by Neuberg, Thebault, Robson, Ghiocas and Lubin.
The ideal proof yet to be discovered should be one that adheres to synthetic
geometry, that follows a direct path from hypothesis to conclusion, that is rela-
tively easy to follow and is neither too long nor too involved. The existing
proofs that approximate these ideals most closely are those by W.E. Philip and by
B. Niewenglowski.
Numerous authors have reached out for extensions of Morley's original theorem
and have wandered into ramifications involving external angle bisectors, angle
quintisectors and nonequilateral Morley triangles. Vandeghen, for example, corre-
lates the basic Morley Triangle with those constructed within the orthic triangle,
foot-median triangle, excenter triangle and triangles formed by the excircle and
incircle contacts, as well as the circumcircle tangential triangle at the vertices
of the parent triangle.
The reader is encouraged to explore the various references with the hope of
devising improvements, shortcuts and possibly new methods of attack. Morley's
Theorem is still yery young and we can surely expect novel methods of proof to loom
up in the future as more and more geometrical aficionados try their hand at further
refinement and clarification of this beautiful theorem.
* & &
Because the Morley material has expanded to fill nearly all the space
available, the problem section is rather rudimentary in this issue. We will
make up for it in the next few issues.
- 297 -
P R O B L E M S — P R O B L E M ES
Problem proposals and solutions should be sent to the editor, whose address
appears on the front page of this issue. Proposals shouldr whenever possible, be
accompanied by a solution, references, and other insights which are likely to be of
help to the editor. An asterisk (*) after a number indicates a problem submitted
without a solution.
Original problems are particularly sought. But other interesting problems may
also be acceptable provided they are not too well known and references are given as
to their provenance. Ordinarily, if the originator of a problem can be located, it
should not be submitted by somebody else without his permission.
To facilitate their consideration, your solutions, typewritten or neatly
handwritten on signed, separate sheets, should preferably be mailed to the editor
before March 1, 1978, although solutions received after that date will also be
considered until the time when a solution is published.
show that Fl9 F2 and F 3 are functionally dependent and find their functional
relationship. Also, reduce the five-parameter representation of F , F2 and Fs
to one of two parameters.
SOLUTIONS
No problem is ever permanently closed. The editor will always be pleased to
consider for publication new solutions or new insights on past problems.
199, [1976: 220; 1977: 112] Proposed by H.G. Dworschak, Algonquin College.
If a quadrilateral is circumscribed about a circle, prove that its
- 299 -
diagonals and the two chords joining the points of contact of opposite sides are
all concurrent.
Editor's comment.
I give below eight additional references to this problem. References [l]-[7]
were sent in by Charles W. Trigg, San Diego, California, and [8] came from Murray
S, Klamkin, University of Alberta.
REFERENCES
1. M.N. Aref and William Wernick, Problems and Solutions in Euclidean Geometry,
Dover, 1968, p. 209.
2. John Casey, A Sequel to Euclid, Longmans-Green, 1884, p. 3 44.
3. E.A. Maxwell, Geometry for Advanced Pupils, Oxford, 1949, pp. 140-141.
4. William J. McClelland, The Geometry of the Circle, Macmillan, 1891, p. 126.
5. Charles Pierson, Solution of Problem 85, The Pentagon, 18 (Fall 1958) 42-43.
6. J.M. Wilson, Solution of Problem 2310 on p. 34 of Vol. VIII, Mathematical
Questions from the Educational Times, edited by W.J. Miller, C.F. Hodgson & Son,
London, 1868.
7. Solution of Geometry Problem 196, American Mathematical Monthly, 10 (May
1903) 138.
8. Mathematics Magazine, Nov-Dec 1962, pp. 289-294. (This contains nine
proofs of the theorem.)
234. [1977: 104, 154, 257] Late solution: DANIEL ROKHSAR, Susan Wagner
H.S., Staten Island, N.Y.
CARL
x F .
MJS^
Editor's comment.
R. Robinson Rowe and the proposer have both confirmed that the three solutions
given in [1977: 264] are the only ones. The details of their calculations will be
left to enrich the archives of this journal. This problem thus has the unique dis-
tinction of having exactly three unique solutions.
241 i [1977: 130, 265] Late solution: DANIEL ROKHSAR, Susan Wagner H.S.,
Staten Island, N.Y.
- 300 -
0
CHARLES W. TRIGG 2
i— .r——ii i
TF?|BG
Extrait d'une critique litteraire parue dans Histoire Pour Tous (fevrier 1963).
Reprinted by permission from the Wesley Palms Breezes (Nov-Dec 1975), p. 12.
- 301 -