Let be a triangle with sides and . For , if and and are the points of tangency of the incircle of to the sides and respectively, then is a triangle with side lengths and if it exists. What is the perimeter of the last triangle in the sequence ?
- A)
- B)
- C)
- D)
- E)
Answer
D
Key insight
Tangent lengths are s - a, so sides x-1, x, x+1 become x/2-1, x/2, x/2+1; the middle side halves until it drops below 2.
Solution
Tangent lengths. The two tangent segments from a vertex to the incircle are equal, so with , , and semiperimeter ,
The pattern. Suppose has sides . Then and the new sides are
So again has three sides spaced by , now centered at . Starting from , the middle side of is , and the perimeter of is .
When does it stop? Sides form a triangle exactly when , i.e. . We need the last with , i.e. . Since , the last valid triangle is , with middle side . (For the middle side would be about , so its sides would fail the triangle inequality.)
Perimeter.
The answer is .
Why this works
The identity "tangent length " turns a geometric construction into pure arithmetic on the side lengths, and the special shape of the sides (consecutive, spaced by ) is preserved because subtracting each side from the same just reverses and shifts them. The perimeter of is , half the previous perimeter, so the perimeters form a geometric sequence and the only real work is finding where the triangle inequality first breaks.
Alternative approach
Each perimeter is half the one before: , so the answer is for some , matching the form of every choice. The stopping condition needs the smallest side positive and the two smallest to beat the largest: sides and work while . Counting halvings from down past gives nine, so after the first three perimeters, i.e. .
The trap
Halving one step too many (or too few): the triangle with middle side 2012/512 is valid but the next, with middle side about 1.96, violates the triangle inequality.
Common mistakes
- Halving one step too many (or too few): the triangle with middle side is valid but the next, with middle side about , violates the triangle inequality.
- Using the condition "smallest side positive" () instead of the triangle inequality (), which incorrectly allows one more triangle and gives choice (E).
- Mixing up which tangent length goes with which vertex; since only the multiset of new side lengths matters, this does not change the answer, but it can derail the derivation of the pattern.
Techniques
Apply an identity: SFFT, sum of squares, difference of cubes, Vieta · Bound the quantity above/below or estimate to pin it down · Compute small cases, spot the pattern, generalize