Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Cevian triangles

6 views
Skip to first unread message

quasi

unread,
May 25, 2012, 5:16:31 PM5/25/12
to
Conjecture:

If a triangle with integer side lengths a,b,c is such that,
for some interior point P of the triangle, the 3 Cevians
through P partition the triangle into 6 smaller triangles
with integer side lengths, then gcd(a,b,c) > 1.

quasi

1treePetrifiedForestLane

unread,
May 25, 2012, 5:38:27 PM5/25/12
to
thank you, quasibird;
does it really matter, if it is in- or out-side?

> If a triangle with integer side lengths a,b,c is such that,
> for some interior point P of the triangle, the 3 Cevians
> through P partition the triangle into 6 smaller triangles
> without all integer side lengths, then gcd(a,b,c) <= 1.

quasi

unread,
May 25, 2012, 8:00:13 PM5/25/12
to
1treePetrifiedForestLane wrote:
>quasi wrote:
>>
>>Conjecture:
>>
>> If a triangle with integer side lengths a,b,c is such that,
>> for some interior point P of the triangle, the 3 Cevians
>> through P partition the triangle into 6 smaller triangles
>> with all integer side lengths, then gcd(a,b,c) > 1.
>
>thank you, quasibird;
>does it really matter, if it is in- or out-side?

Maybe not, but I'm not sure.

I certainly don't want to let P be on the boundary of the
big triangle, otherwise the 6 small triangles would not all
be non-degenerate.

quasi

quasi

unread,
May 29, 2012, 3:56:56 AM5/29/12
to
There is exactly one triangle ABC with perimeter less than
1000 satisfying the specified conditions.

It has sides

BC, CA, AB = 195, 244, 329

and Cevians AD, BE, CF meeting at point P such that

BD, CE, AF = 39, 183, 188

AC, BE, CF = 304, 282, 144

AP, BP, CP = 190, 141, 126

thus yielding a counterexample to the stated conjecture.

quasi
0 new messages