The February 2001 cryptogram was:
UNFORTUNATELY,
UNLIKE
BONES,
BEHAVIOR
DOES
NOT
BECOME
FOSSILIZED.
RICHARD
LEAKEY
Kenyan paleoanthropologist and political
activist (1944-).
Correct answers came from
Delmar Burkitt,
Andrzej Derdzinski,
Jeff Gerken,
Bob Kaplan,
Alma Litten,
Julia Minturn,
Carla Nuenke,
and
Sally Yocom.
The second puzzle proved a bit deceptive.
Each merger, no matter how done, decreases the number
of companies by one.
Hence 105 mergers are needed.
However, the first merger begins on the first Friday
and the last (#105) occurs on Friday 104 weeks later.
So 104 is the answer.
Only Andrzej and Bob had
the correct answer.
The correct answer to the third puzzle was 1500 Democrats,
1300 Republicans, 1100 Independents
with 6100 not voting.
Correct were: Delmar, Andrzej, Jeff, and Carla.
Finally, Andrzej, Jeff, and Steve submitted answers to the overlapping circles
puzzle submitted
by Doug McDonald. The challenge was to find when the area of overlap of
two circles is equal to half the area of one of the circles.
Here is an analysis of the solutions:
From Andrzej J. Derdzinski:
I took a look at the overlapping circles problem.
The answer is both simple and disappointing.
Simple because it can be obtained by setting up an equation in a most
straightforward fashion, and disappointing because the answer (the distance
d between the centers or, more precisely, the ratio d/r of that distance to the common
radius r) does not have an explicit expression in terms of
elementary functions.
One way of stating the answer is:
Take the smallest positive angle a (measured in radians) such that a = cos a.
This number must exist, and lie between 0 and /2, since
the value of y=x - cos x is
negative (y=-1) at x=0 and positive (y= /2) at
x= /2, and so, by
reasons of continuity, we must have y=0 at some intermediate x=a.
The ratio d/r at which the overlapped and nonoverlapped parts have equal areas
then equals d/r=cos b, with b=a/2 + /4.
To see that this is the case, one may denote P,Q,R the center of Circle "B", one
of the intersection points of the two circles and, respectively, the midpoint
between their centers, and denote b the angle QPR expressed in radians.
Then one-quarter of the overlap area equals the difference
between the area of the circular wedge
QPR and the area of the triangle
QPR, i.e., br2/2-r2(cos b sin b)/2, and
we want this number to be
one-eighth of r2, which gives the equation
b-cos b sin b = /4.
Rewritten in terms of a=2b + /2, this equation
reads a=cos a.
The fact that we do not have an elementary formula for this angle a
(or the ratio d/r) makes it no less "explicit" than other
numbers that are already entrenched in common usage
(such as , e, etc.).
We could imagine giving it a name, using standard procedures to
find its decimal approximations, etc.,
just like it once happened with (when
someone asked the similar question:
what exactly is the area of a circle of radius one?).
Jeff Gerken sent this solution:
For the circle problem, the area of overlap will be equal to the non-overlapped
area when the arc (theta) subtended by overlap satisfies the equation (with theta in
degrees):
( * )/180 - sin( )= /2.
The solution to this equation is approximately 132.348 degrees.
Steve Herrick's solution:
The area of the overlap is the sum of the areas of the arcs described
by DAE and DBE minus the area of the rhombus ADBE.
The area of arc DAE is r2*DAE/2 in radians.
The area of arc DBE is equal to DAE so their sum is r2*DAE/ or r2DAE.
The area of the rhombus is side BE*height AF.
To determine length of AF, we use the sin of angle AEF which equals AF/AE.
Angle AEB is equal to -DAE thus we get r2sin( -DAE).
So we want to know the size of angle DAE when r2DAE-r2sin( -DAE)= *r2/2.
Dividing by r2, we get
DAE-sin( -DAE)= /2.
This solves out at app. 2.3 radians or 132 degrees.
What I was looking for was the length of AB.
We know that the cos(DAC)=AC/AD and AB=2AC and DAC=DAE/2
So AB=2r*cos(DAE/2)
So AB=.808*r
|