Two segments parallel to the sides of a parallelogram are drawn through a point on the parallelogram's diagonal.
The area of the entire figure is 2009, and the areas of the red, blue, and green sections are distinct integers which form a geometric progression in that order.
What is the maximum possible area of the red section?
Note: The diagram is not drawn to scale.
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
I initially assumed that "in order" meant that the red area was the smallest. When it dawned on me that was not what was stated, I was able to get the correct answer.
The “in that order” descriptor is very misleading. What exactly could it imply besides the red area being the smallest?
Log in to reply
It seems to mean that the blue area comes in the middle of the other two.
I was just as confused. I knew I had gotten the right answer and kept double checking, but it turns out that the red triangles didn't have the smallest areas as depicted in the image.
Isn't it by definition a geometric progression in that order, due to the construction? I also had difficulty with this phrase, and initially assumed the intended solution was red:blue:green = 369:492:656.
Of course the factor in a geometric progression can indeed be smaller than 1, which resulted in the areas red:blue:green = 1476:246:41.
I find the whole "geometric progression in that order" phrase confusing, unless you can give a reason why it is necessary. As far as I can see the blue area is always the geometric mean of the other areas, which apparently is the thing you wanted to say with the in that order part...
Log in to reply
Honestly, I didn't create this problem. It was just in a book I had. That confused me as well, and since my brain skipped over the word integer, I was completely stuck for a while. It could either be a red herring or a hint to solve the problem. Either way, the point is that you have to prove that the blue area is the geometric mean of the other two before continuing.
A lot of people are complaining about the "in that order" phrasing, so I might delete it.
Log in to reply
I didn't find it confusing. When we come across problems other than in refined text books, the problem is rarely precisely described. Here all you need is the parallelogram, the parallel lines, the integer restriction and the area total to define the problem. Telling us the areas are in a geometric series is nice, even better that they are consecutive terms in one, but this is a consequence of the construction.
You don't really have a responsibility to only put clean questions up. If people are complaining about the assumptions they themselves made, that should be a lesson for them to learn. Reducing assumptions is critical to mathematical thinking.
Log in to reply
@Chris Maitland – So what do the words "in that order" actually mean, when there is (apparently) a clear indication of the relative sizes of the segments in the diagram (the most famous "not to scale" diagram of them all, H C Beck's London Underground creation which is the model for maps of urban public transport systems worldwide, still has the central area smaller than the outlying area, just not as much smaller as it actually is)? I do not expect diagrams to be in the correct absolute proportions, but keeping the relative proportions correct (i.e the largest single area in reality is also the largest single area in the diagram) is surely a must. I am not a habitual complainer (as I mentioned in my earlier and as yet unaddressed comment this is the first brilliant.org problem about which I have seen fit to complain) so the fact that I am on the side of the complainers should be seen as fairly damning of the problem.
I'm new here, but this was very confusing. The depicted construction always leads to a geometric progression, thus the "in that order" precision should mean red is smallest - it isn't. At no point was it specified to only use integers.
I agree with those complaining about this problem. The picture shows the red section as the smallest, with blue next smallest and green largest, and the only geometric progression that can link them is that each is one third of the area of the other (green = 3 x blue = 3 x red). The "in that order" that is included in the setting of the question further reinforces this view. Given the answer that you were looking for the diagram is an absolute disgrace - "not to scale" is fine, but even so the relative sizes of the segments should be in proper order - the segment that is largest should appear so in the diagram. Please note that this is the first time since I started solving these problems that I have formally complained about one.
I totally agree with those comments. The words "in that order" mean that the red area is the smallest. In fact my solution was 41. You need to reformulated the quiz because this way it is not a mathematical quiz.
Log in to reply
It doesn't mean that. It simply means that red comes first in the geometric progression, blue second, and green third.
Let the green area be a positive integer a . To maximise the red area, we'll need to minimise a with the restriction that the coloured areas are distinct positive integers that sum to 2009.
The the total area is a ( 1 + 2 r + r 2 ) = a ( 1 + r ) 2 = 2 0 0 9 = 7 2 ⋅ 4 1 .
This means r = 7 a 4 1 − 1 and so a r = 7 a a 4 1 − a = 7 4 1 a − a .
As a is a positive integer and a r is a distinct positive integer, a must contain 4 1 as a prime factor.
So the minimum solution has a = 4 1 and 1 + r = 7 , meaning r = 6 . Therefore, the red area is a r 2 = 4 1 ⋅ 3 6 = 1 4 7 6 .
Good solution, except a = 4 1 isnt the only solution for a . a can be any value less than 2 0 0 9 which is 4 1 times a perfect square.
Log in to reply
@Zain Majumder has it on the nose: a is any 4 1 ⋅ y 2 < 2 0 0 9 . Specifically, if y is an integer, y ⋅ r is an integer, and y ⋅ ( 1 + r ) = 7 . Another way to think of it, is to imagine that a is the red area (the way I solved it). Solutions include: ( y , r ) = ( 1 , 6 ) , ( 2 , 2 5 ) , ( 3 , 3 4 ) , ( 4 , 4 3 ) , ( 5 , 5 2 ) , ( 6 , 6 1 )
Ah! You're correct! But I want to minimise a meaning that perfect square is 1 . I've edited the answer to include this.
General solution
The sides of the parallelogram are divided in the same ratio a : b (assume no common factors). Thus the areas have ratios red : blue : green = c a 2 : c a b : c b 2 for some constant c . If a , b , c = 0 and a = b this is a geometric progression with factor b / a .
The total area of the parallelogram is A = c ( a + b ) 2 . In order to have integer areas, we need a , b , c to be integers. Thus we find the following recipe for solving the problem:
Factor the given area A in the form A = c x 2 , with x as large as possible.
Let a = x − 1 and b = 1 .
Then the red area is c a 2 .
In this case, A = 2 0 0 9 = 4 1 ⋅ 7 2 , so that x = 7 , a = 6 , b = 1 , c = 4 1 , and red = c a 2 = 4 1 ⋅ 6 2 = 1 4 7 6 .
Bugger! Same reasoning but I thought that "in that order" meant that the red area should be the smallest of the three. Then I got 2009*(3/7)^2 = 369 ... Obviously not accepted ;-(
Let red area be A 1 , blue area A 2 , green area A 3 . We know that:
A 2 A 1 = A 3 A 2 ⇒ A 2 = A 1 A 3
And also:
A 1 + 2 A 2 + A 3 = 2 0 0 9
Therefore:
A 1 + 2 A 1 A 3 + A 3 = 2 0 0 9
( A 1 + A 3 ) 2 = 2 0 0 9
A 1 + A 3 = 7 4 1
So A 1 and A 3 must be multiples of 4 1 , and, since A 3 can't be zero, the maximum of A 1 must be 1 4 7 6 = 3 6 ⋅ 4 1 , in order to have:
6 4 1 + 4 1 = 7 4 1
Very neat answer!
Great answer!
Since the diagonal divides the big, red and green parallelograms into halves, it is obvious that the area of the blue parallelogram is same as that of the white parallelogram.
Let the height of the big parallelogram be h and that of the red parallelogram be x h , where x < 1 . Since the shapes of the red and green parallelograms are similar to that of the big parallelogram their areas are directly proportional to their heights. That is:
⎩ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎧ A b i g = 2 0 0 9 A r e d = 2 0 0 9 x 2 A g r e e n = 2 0 0 9 ( 1 − x ) 2 A b l u e = 2 2 0 0 9 ( 1 − x 2 − ( 1 − x ) 2 ) = 2 0 0 9 x ( 1 − x )
Note that A r e d A b l u e = A b l u e A g r e e n = x 1 − x , the common ratio.
We note that 2 0 0 9 = 7 2 × 4 1 . For A r e d = 2 0 0 9 x 2 to be an integer, x = 7 n < 1 , where n is a positive integer. Then we have:
⎩ ⎪ ⎨ ⎪ ⎧ A r e d = 4 1 n 2 A g r e e n = 4 1 ( 7 − x ) 2 A b l u e = 4 1 n ( 7 − n ) , which are all integers.
We note that when n = 7 , A r e d = 2 0 0 9 = A b i g and A g r e e n = A b l u e = 0 . Therefore, A r e d is maximum when n = 6 ; that is max ( A r e d ) = 4 1 × 6 2 = 1 4 7 6
Also remember that all the possible values of A r e d are when x = 7 n for n ∈ { 1 , 2 , 3 , . . . 6 } . In retrospect, I could have asked for the second-largest value of A r e d to make the problem a bit trickier!
Log in to reply
Thanks, I failed to see that. I will change my solution.
Because the figure isn't drawn to scale, there is no reason that the red area must be smaller than the blue area, which is our key to solving the problem. Say the area of the red section is equal to a, and the ratio of the blue section's area to the red section's area is equal to d. Because the areas are in a geometric progression, we must have a + a d + a d 2 + a d = 2 0 0 9 . Note that the lefthand side is equal to a ( d + 1 ) 2 , so a = ( d + 1 ) 2 2 0 0 9 , and because a is an integer, the expression on the right-hand side must also be an integer. 2 0 0 9 = 4 1 ∗ 4 9 = 4 1 ∗ 7 2 , so we must have d + 1 = n 7 for some integer n, and thus a = 4 1 n 2 . Because a < 2 0 0 9 , n < 7 , and the value n = 6 maximizes a. So a = 4 1 ∗ 3 6 = 1 4 7 6 .
R:=red area, B:=blue area, G:=green area. r is the common ratio. R,B,G are integers. R, B and G form a geometric progression, so: B = r R, G = r B = r^2 R. The area of the entire figure is 2009 = R + 2 B + G = R + 2 r R + r^2 R = R (1 + 2 r + r^2).
The primes of 2009 are: 7 and 41. If If you try R = 7 you get: 1 + 2 r + r^2 = 2009 / 7 = 287. If you transform the equation you get: 1 + 2 r + r^2 = (r+1)^2 = 287. So r = 15,941... That doesn't work because r has to be an integer.
Do the same with R = 41 and you get (r+1)^2 = 2009 / 41 = 49. So r = 6. Now R = 41, B = 41 * 6 = 246 and G = 246 * 6 = 1476.
To answer the question you have to switch the green area an the red area and you see that the maximum possible area of the red (or the green) section is 1476 .
You've started with an argument equivalent to "an integer divided by an integer is another integer". This isn't true and r could be a rational number.
You've cleverly used symmetry in your argument but this should have shown you your mistake! 1476 is not a factor of 2009.
Don't worry. I'm not judging. I made the exact same mistake and had to edit my answer after someone pointed it out to me.
Log in to reply
Hi Chris, you are right. But "Hence r is an integer." is not necessary for the proof and I correct it. If you think of R<B<G then R is a factor of 2009. If you switch R and G then R is not a factor. So thank you for your comment! KR Volker
Let r, g, b be the area of the colored shapes, and k be the constant in the geometric progression.
then we easily get these equations:
r+2b+g=2009
r = kb =kkg
and thus when we substitute
kkg+2kg+g=2009=g(kk+2k+1) <=>2009/(kk+2k+1)=g
we know g is an integer and thus (kk+2k+1) must be a factor of 2009:
2009
287
49
41
7
1
We want r to be the biggest so we brute force these value to find a positive rational as solution to (kk+2k+1)={factor} where 2009k/{factor}=b is an integer and 2009kk/{factor}=r is an integer (if there were multiple solutions take the one which yield the biggest r).
so we find the solution {factor} = 49, k=6, g=41, b=246 and finally r = 1476
The matematics should be elegant and simple. Your explanation is the most elegant and simple by so far.
Let the red, blue, and green areas be r , b , g , respectively. They are in a geometric progression in that order, but, because we are maximizing r , that progression will be decreasing. To make things easier, let's reverse the order of the geometric progression, so that we have r = g k 2 , b = g k , for some k > 1 . Note that, because the three areas are integers, k is rational.
Also, note that the white area and the blue area are equal, so the total area of the parallelogram (which is 2009) can be written as 2 0 0 9 = r + 2 b + g = g k 2 + 2 g k + g = g ( k 2 + k + 1 ) = g ( k + 1 ) 2 . Now we compare the factorization of 2 0 0 9 = 4 1 ⋅ 7 2 with this final expression.
Non-Rigorous Way
In the factorization of 2009, there's an integer times a square, and in our expression, we also have that. Wouldn't it be pretty cool if those components lined up? Let's try it: g = 4 1 , ( k + 1 ) 2 = 7 2 ⇒ k = 6 . If this is the green area, the red area is just g k 2 = 4 1 ⋅ 6 2 = 1 4 7 6 , which is the right answer.
Actual Justification
The method above doesn't show that 1476 is the maximum value of the red area, just that it's a possible value. To do that, we can go back to the equation 4 1 ⋅ 7 2 = g ( k + 1 ) 2 and notice that, if we divide both sides by g, we get g 4 1 ⋅ 7 2 = ( k + 1 ) 2 ⇒ k + 1 = g 4 1 ⋅ 7 2 = 7 g 4 1 ⇒ 7 k + 1 = g 4 1
Because k is rational, so is the left side of the final expression. Therefore, the right side is also rational. But, the right side is only rational if and only if g 4 1 is the square of a rational. However, because 41 is prime and g is an integer, we must have g = 4 1 ⋅ c 2 for some positive integer c . Plugging in, that gives 7 k + 1 = 4 1 ⋅ c 2 4 1 = c 1 .
Rearranging, we have k = c 7 − 1 . Remember, g = 4 1 c 2 . So, plugging all this into r , we get r = g k 2 = ( 4 1 c 2 ) ( c 7 − 1 ) 2 = ( 4 1 c 2 ) ( c − 2 ) ( 7 − c ) 2 = 4 1 ( 7 − c ) 2 . It is clear that the minimal value of c gives the maximum value of r . c is a positive integer, so c = 1 ⇒ r = 1 4 7 6
As an aside, the reason we can assume c is also positive is because it's squared
Area(Total)=2009
define as follows from geometric progression (Ai=x n^(i-1), from i = 1 to 3) for some integer x and constant n. 1st Term: A(1)=Area(Red) = x 2nd Term: A(2)=Area(Blue) = n x 3rd Term: A(3)=Area(Green) = n^2*x
we know that the total area is the sum of its regions. A(T) = A(R)+2*A(B)+A(G)
substitute in the terms defined A(T) = x+n x+n^2 x+n*x 2009 = x(1+2n+n^2)
break down 2009 into its prime factors and group the terms to the same form. The integer solutions for x and n are now obvious. 41*7^2=x(n+1)^2 thus x=41 and (n+1)^2=7^2 -> n+1=7 -> n=6 we know this as the only other integer solution is x=2009,y=0.
this progression is increasing positively so x=41=A(R)=A(Rmin) is the minimum area satisfying the requirements. thus it is evident that A(Rmax) = A(G) = x n^2 = 6 6 41 = 1476 or similarly, A(Rmax)=A(T)-A(Rmin)-2 A(B)=2009-x-2nx=2009-41-2 6 41 = 1476
Log in to reply
n doesn't have to be an integer for it to be a geometric progression of integers. For instance, 16,20,25 form a geometric progression of integers with common ratio 5/4. Actually, the equation g(k+1)^2 = 2009 has two other solutions for k aside from k=1. Provided that k>1, you can derive that c (the positive integer factor in g that is squared) must be less than 3.5, so it can only be 1,2, or 3. In the maximum, it's 1. However, letting it equal 2 or 3 yields another valid, non-maximum, distribution of the areas. In those cases k would be 5/2 or 4/3, the red area would be 1025 or 656, respectively.
Problem Loading...
Note Loading...
Set Loading...
The red and green parallelograms are similar since they are each made of two congruent triangles, and a red triangle is similar to a green triangle. Therefore, if we let the base of the red parallelogram be b , and its height be h , then the base and height of the green parallelogram are b k and h k for some number k .
Now we can calculate the areas of each colored region, so the sequence b h , b h k , b h k 2 forms a geometric progression. This means that k is the common ratio for this progression.
The base of the entire parallelogram is b + b k = b ( k + 1 ) , and the height is h + h k = h ( k + 1 ) . Therefore, its area is b h ( k + 1 ) 2 = 2 0 0 9 ⟹ b h = ( k + 1 ) 2 2 0 0 9 . Since b h and b h k are both integers, k must be a rational number in the form y x , where x and y are positive integers.
b h = ( y x + 1 ) 2 2 0 0 9 = y 2 ( x + y ) 2 2 0 0 9 = ( x + y ) 2 2 0 0 9 y 2
2 0 0 9 = 7 2 ∗ 4 1 , and x + y ≥ 2 . Therefore, for this fraction to be an integer, x + y = 7 . We have b h = 7 2 2 0 0 9 y 2 = 4 1 y 2 . The largest value of y is 6 , so the maximum value of b h is 4 1 ∗ 3 6 = 1 4 7 6 .
Note: There is no way to get the fraction to be a multiple of 4 9 . For this to happen, y 2 must contain a factor of 4 1 , so y 2 = 4 1 n for some integer n . ( x + y ) 2 > 4 1 n , so x must be chosen to make ( x + y ) 2 be a multiple of 4 1 n , which means there is now an extra factor in the denominator. Therefore there is no way to make this work.