Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

The following is a geometry theorem whose proof is examinable in the Irish 'High School' Exam.

Let $\Delta ABC$ be a triangle. If a line $L$ is parallel to $BC$ and cuts $[AB]$ in the ratio $s:t$, then it also cuts $[AC]$ in the same ratio.

The proof required is the commensurable case and appeals to a previous theorem: If three parallel lines cut off equal segments on some transversal line, then they will cut off equal segments on any other transversal.

I have been struggling to find the 'standard' proof in the incommensurable case (i.e. without using continuity). I have a proof that I want to compare with the 'standard' (I use an area argument to show that similar triangles are enlargements of each other and work from there). I have tried and failed to find it online, the closest I've come is p. 30 of --- but note that I want to define two triangles as similar if they have equal angle measures and this assumes something different.

Any help would be greatly appreciated.

EDIT: This is my own proof:

Theorem Let $\Delta ABC$ be a triangle and $DE$ be a line segment such that $DE\parallel BC$ as shown

enter image description here

Then $$\frac{|AD|}{|BD|}=\frac{|AE|}{|CE|}.$$

Clearly we can reduce to the case of a right-angled triangle.

Lemma 1: Suppose $\Delta ABC$ is similar to triangle $\Delta A'B'C'$ and suppose further that both of a right-angle at $B$. Then if $|A'B'|=k|AB|$ then $|B'C'|=k|BC|$; i.e. similar triangles are equivalent modulo enlargement.

Proof of L1: Suppose without loss of generality that $|AB|<|A'B'|$ so $k>1$. Hence we can place $\Delta ABC$ inside $\Delta A'B'C'$ as shown:

enter image description here

Suppose $|B'C'|=\alpha|BC|$. Now we compute the area of $\Delta A'B'C'$ in two different ways:

$$\text{area}(\Delta A'B'C')=\frac1{2}|A'B'||B'C'|=\frac1{2}k\alpha |AB||BC|.$$

Also, $$\text{area}(\Delta A'B'C')=\text{area}(\Delta ABC)+\text{area}(\Delta CEC')+\text{area}(BCDB')$$ $$=\frac1{2}|AB||BC|+\frac{1}{2}|CD||DC'|+|BC||BB'|.$$

Note that

$$|CD|=|BB'|=|A'B'|-|AB|=(k-1)|AB|\text{ , and}$$ $$|DC'|=|B'C'|-|BC|=(\alpha-1)|BC|.$$ Hence $$\text{area}(\Delta A'B'C')=\frac1{2}|AB||BC|+\frac1{2}(k-1)|AB|(\alpha-1)|BC|+(k-1)|AB||BC|.$$ Now set these two equal: $$\frac1{2}k\alpha |AB||BC|=\frac1{2}|AB||BC|+\frac1{2}(k-1)|AB|(\alpha-1)|BC|+(k-1)|AB||BC|,$$ $$\Rightarrow k\alpha =1+(k-1)(\alpha-1)+2(k-1),$$ $$\Rightarrow k\alpha=1+\alpha k-k-\alpha+1+2k-2,$$ $$\Rightarrow \alpha=k\,\,\,\bullet$$

Lemma 2: Suppose $a,\,b>0$. Then the only positive solutions to $$\frac{a+x}{b+y}=\frac{a}{b}$$ occur when $x/y=a/b$

Proof: $$\frac{a+x}{b+y}=\frac{a}{b}\Leftrightarrow ab+bx=ab+ay\Leftrightarrow \frac{x}{y}=\frac{a}{b}\,\,\,\bullet$$

Proof of Theorem: Take $\Delta ABC$ to be a right-angled triangle and translate $CE$ to $D$ to form $DF$ as shown $(*)$:

enter image description here

Now $\Delta ADE\sim \Delta BDF$ so by Lemma 1 $$|AD|=k|BD|\text{ and }|DE|=k|BF|,$$ $$\Rightarrow \frac{|AD|}{|BD|}=\frac{|DE|}{|BF|}\,\,(**).$$ Now by Pythagoras: $$|AE|^2=|AD|^2+|DE|^2\text{ and }|CE|^2\underset{(*)}{=}|BD|^2+|BF|^2,$$ $$\Rightarrow \frac{|AE|^2}{|CE|^2}=\frac{|AD|^2+|DE|^2}{|BD|^2+|BF|^2}\underset{(**)\text{ and Lemma 2}}{=}\frac{|AD|^2}{|BD|^2}\,\,\bullet$$

My happiness at finding this proof is singularly dampened by seeing how much slicker the 'standard' proof is!

share|cite|improve this question
Is this what you want? – David Mitra Dec 20 '11 at 11:51
The proof they use is on p.62 here… – Jp McCarthy Dec 20 '11 at 11:52
@David That's the one thank you very much. Might throw my own proof in here later. – Jp McCarthy Dec 20 '11 at 11:54
I can't see how the very first line of the proof is true... – Jp McCarthy Dec 20 '11 at 12:02
I find the wiki proof somewhat confusing too. My answer below is essentialy the same as the one at PlanetMath (…), but with added detail. – David Mitra Dec 20 '11 at 12:19
up vote 1 down vote accepted

You want the:

Intercept Theorem:

The parallel lines $\color{maroon}{\ell_1}$ and $\color{maroon}{\ell_2}$ split the sides of the angle $\angle BAC$ into equal proportions: $$ \tag{1} {AD\over DC}={AE\over EB},\quad{\rm and}\quad{AD\over AC}={AE\over AB} $$ Also: $$\tag{2} {AD\over AC}={DE\over BC}$$

enter image description here

Proof: Observe:

1) $\color{darkgreen}{\triangle ADE}$ and $\color{darkblue}{\triangle BDE}$ have the same height, $\color{gray}{h_1}$, with respect to the baseline $AB$. Therefore $${AE\over EB} = {{\rm area}(\color{darkgreen}{\triangle ADE}) \over{\rm area}(\color{darkblue}{\triangle BDE})}.$$

2) The triangles $\color{darkblue}{\triangle BDE}$ and $\color{pink}{\triangle DEC}$ have a common base $DE$. Moreover, since $\color{maroon}{\ell_1}$ is parallel to $\color{maroon}{\ell_2}$, the heights of these triangles, $\color{darkblue}{h_3}$ and $\color{pink}{h_4}$, with respect to the base $DE$ are equal. Thus, $${\rm area}(\color{darkblue}{\triangle BDE})= {\rm area}(\color{pink}{\triangle DEC}).$$

3) $\color{darkgreen}{\triangle AED}$ and $\color{pink}{\triangle DEC}$ have the same height, $\color{gray}{h_2}$, with respect to the baseline $AC$. Therefore $${{\rm area}(\color{darkgreen}{\triangle AED}) \over {\rm area}(\color{pink}{\triangle DEC})}={AD\over DC }.$$

The first part of $(1)$ follows immediately from these observations.

The second part of $(1)$ can now be established by using the first part and a bit of algebra.

$(2)$ can be established by applying $(1)$ to the angle $ \angle ABC$, the line $AC$, and the line parallel to $AC$ passing through $E$.

share|cite|improve this answer
I followed this a lot easier than the Wikipedia effort thank you. – Jp McCarthy Dec 20 '11 at 12:24
I don't know why the powers that be restrict to the commensurable case --- this proof is more fundamental than their's and proves more. Odd that. – Jp McCarthy Dec 20 '11 at 12:32
@Jp McCarthy I agree. I also don't understand why the (lovely) Intercept Theorem is never mentioned in the standard curriculum here in the US. Similar triangles get all the credit... – David Mitra Dec 20 '11 at 12:38

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.