2
$\begingroup$

I need to prove that 4 points of the complex plane lie on a same line or a same circle if and only the following is right for corresponding complex numbers:

$$\frac{z_1-z_3}{z_2-z_3} \div \frac{z_1-z_4}{z_2-z_4} \text{ is real.}$$

$\frac{z_1-z_3}{z_2-z_3} \div \frac{z_1-z_4}{z_2-z_4} \in \mathbb{R} \Leftrightarrow \frac{z_1-z_3}{z_2-z_3}$ and $\frac{z_1-z_4}{z_2-z_4}$ both real or both imaginary. If they are both real, then we have $z_1-z_3 = r_1(z_2-z_3)$ and $z_1-z_4 = r_2(z_2-z_4), r_1,r_2 \in \mathbb{R}$. So vectors $\vec{Z_1}-\vec{Z_3},\vec{Z_2}-\vec{Z_3}$ and $\vec{Z_1}-\vec{Z_4},\vec{Z_2}-\vec{Z_4}$ are collinear.

And what if $\dfrac{z_1-z_3}{z_2-z_3}$ and $\dfrac{z_1-z_4}{z_2-z_4}$ are both imaginary?

$\endgroup$
  • $\begingroup$ It is not correct that $a/b\in\mathbb R \Longleftrightarrow{}$ $a$ and $b$ are either both real or both imaginary. It is true that $a/b\in\mathbb R \Longleftarrow{}$ $a$ and $b$. However, suppose $a = 2+6i$ and $b=1+3i$. Then $a/b$ is real even though neither $a$ nor $b$ is real and neither $a$ nor $b$ is imaginary. ${}\qquad{}$ $\endgroup$ – Michael Hardy Sep 3 '15 at 12:29
  • $\begingroup$ $\uparrow$ I didn't finish a sentence: I meant $a/b\in\mathbb R \Longleftarrow{}$ $a$ and $b$ are both real or $a$ and $b$ are both imaginary. ${}\qquad{}$ $\endgroup$ – Michael Hardy Sep 3 '15 at 12:46
1
$\begingroup$

I would try this:

  • Show that for any three distinct complex numbers $z_1,z_2,z_3$ there is a linear fractional transformation $z\mapsto\dfrac{az+b}{cz+d}$ for which $\dfrac a b \ne \dfrac c d$ that maps $z_1,z_2,z_3$ to distinct real numbers.

  • Show that if $z_4$ is on the same circle or the same line as $z_1,z_2,z_3$ then $z_4$ is also mapped to a real number. I.e. show that the image of a line or a circle under a linear fractional transformation is a line or a circle.

  • Show that if $w_i$ is the image of $z_i$ under the linear fractional transformation for $i=1,2,3,4$, then $$\frac{w_1-w_3}{w_2-w_3} \div \frac{w_1-w_4}{w_2-w_4}= \frac{z_1-z_3}{z_2-z_3} \div \frac{z_1-z_4}{z_2-z_4}.$$ I.e. linear fractional transformations do not change cross-ratios.

Maybe the second item is where most of the work needs to be done.

$\endgroup$
  • $\begingroup$ The "cross ratio" comes up in projective geometry. A very tiny nit to pick with Michael Hardy ; Say " $ ad \ne bc $ ", not "$ a/b \ne c/d$ " because $ b $ or $ d $ can be $ 0$. $\endgroup$ – DanielWainfleet Sep 3 '15 at 17:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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