0% found this document useful (0 votes)
78 views4 pages

Sharygin Solutions

The document presents a proof by contradiction that the number of "bad links" in a closed broken line that touches a circle is always even. It defines a "bad link" as one that does not touch the circle. The proof considers all possible configurations where a single bad link could occur and shows that in each case, another bad link must also exist, making the total number even. It analyzes four cases for where the first bad link could be located and considers subcases for each. This exhaustive case analysis demonstrates that a single bad link is not possible.

Uploaded by

sameer chahar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
78 views4 pages

Sharygin Solutions

The document presents a proof by contradiction that the number of "bad links" in a closed broken line that touches a circle is always even. It defines a "bad link" as one that does not touch the circle. The proof considers all possible configurations where a single bad link could occur and shows that in each case, another bad link must also exist, making the total number even. It analyzes four cases for where the first bad link could be located and considers subcases for each. This exhaustive case analysis demonstrates that a single bad link is not possible.

Uploaded by

sameer chahar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Sharygin P10

Shamim Akhtar
March 3, 2020

Problem Statement: Given are a closed broken line A1 A2 . . . An and a


circle ω whoch touches each of the lines A1 A2 , A2 A3 , . . . , An A1 . Call the link
good, if it touches ω and bad otherwise. Prove that the number of bad links is
even.
Nice Problem!
Proof : Notice that it suffices to show that Bad segments occur in pairs,
hence for one bad segment there is a unique(distinct) bad segment.
We will assume for the sake of contradiction that this is not the case, and
considering the problem modulo 2, we will show that if there is only one bad
segment we cannot achieve a valid configuration.
For simplicity, we will refer to the diagram as position of the first vertex
does not matter.
Let A1 be the first vertex, and let the two tangents from A1 to ω be `1 , `2
respectively and let A2 ∈ `1 .
Thus it is necessary that the last vertex An in the chain ∈ `2 .
Assume that the ”badness” occurs at the segment Aj−1 Aj with 1 < j < n.
(But we will not consider when badness occurs at A1 A2 for better understanding,
but it is similar to handle)

1
Let this be the starting configuration, let `2 ∩ ω = T
Assume now that badness occurs at segment Aj−1 Aj .

2
Now for termination of this chain, we must have the node An ∈ `2 , thus the
intersection of last segment with one endpoint An−1 and `2 MUST be An , the
last vertex.
Now we show that even if we continue drawing good segments from now
onwards, one segment HAS TO BE bad to make the configuration valid.

CASE 1 - Aj = An−1 implies second tangent from Aj 6= Aj Aj−1 intersects


`2 at An , but then An A1 is a bad segment, contradiction.
CASE 2 - Aj = An−2 implies second tangent from Aj+1 6= Aj+1 Aj intersects
`2 at An .
We need two subcases-
Subcase 1-
Notice that if Aj+1 lies beyond `2 from Aj (i.e, Aj+1 lies on the the plane
different from the plane containing Aj , where the two planes considered here
are the planes which are formed by dividing the original plane by `2 ), then
An−1 An = Aj+1 An is a bad segment as the point of tangency of An An−1 with
ω is on the lower part of ω divisioned by T (as Aj Aj+1 was extended beyond `2 ,
which is tangent to ω at T ). Hence, Aj+1 An is the bad segment, contradiction.

Subcase 2 (depicted in the figure) - If Aj Aj+1 is not extended beyond `2 then


An lies in the interior of A1 T , and thus A1 An is a bad segment, contradiction.

3
CASE 3 - Aj = An−3 implies tangent from Aj+2 to ω 6=Aj+2 Aj+1 intersects
`2 at An .
We divide again into two subcases - Subcase 1 (depicted in the diagram) -
Aj+2 lies beyond (same meaning as mentioned before) A1 T (seen from Aj+1 ),
then obviously An lies between Aj+2 and the tangency point of ω with the
tangent from Aj+2 to ω 6= Aj+2 Aj+1 (depicted in the figure by point ’K’), thus
Aj+2 An = An−1 An is a bad segment, contradiction.
Subcase 2 -
If Aj+2 is not beyond A1 T from Aj+1 , then An lies in the interior of segment
A1 T , thus A1 An is a bad segment, contradiction.
CASE 4 - Aj = An−4
Let N be the antipode of T wrt ω. Construct the tangent `3 from N to ω.
Now we divide it into two cases - Subcase 1 -
When Aj+2 did not lie beyond A1 T from Aj+1 , we divide this further into
two cases-
Subcase (i) - Aj+3 lies beyond A1 T , then An lies below T on `2 thus making
Aj+3 An = An−1 An a bad segment, contradiction.
Subcase (ii) - Aj+3 does not lie beyond A1 T , then An lies above T on `2
thus making A1 An a bad segment, contradiction.
Subcase 2 -
When Aj+2 lies beyond A1 T
Again, we divide even this subcase into two further cases -
Subcase (i) (depicted in the diagram) -
When Aj+3 = An−1 lies beyond `3 seen from Aj+2 , we have that the point
of intersection of the tangent to ω from Aj+3 and `2 is above T on `2 , thus
making An A1 a bad segment, contradiction.
Subcase (ii) -
When Aj+3 = An−1 does not lie beyond `3 wrt Aj+2 , then the point of
intersection of the tangent to ω from Aj+3 and `3 is below T on `2 , thus making
An Aj+3 = An An−1 a bad segment, contradiction.
Now, this case work is sufficient as the configuration after CASE 4 repeats
and thus changing the required indices, it becomes similar to one of these four
possible configurations, and thus shows that bad segments occur in pairs, and
we are done.

You might also like