proof of Pascal’s mystic hexagram

We can choose homogeneous coordinates $(x,y,z)$ such that the equationof the given nonsingular conic is $yz+zx+xy=0$ , or equivalently

z(x+y)=-xy

(1)

and the vertices of the given hexagram $A_{1}A_{5}A_{3}A_{4}A_{2}A_{6}$ are

$A_{1}=(x_{1},y_{1},z_{1})$	$A_{4}=(1,0,0)$
$A_{2}=(x_{2},y_{2},z_{2})$	$A_{5}=(0,1,0)$
$A_{3}=(x_{3},y_{3},z_{3})$	$A_{6}=(0,0,1)$

(see Remarks below).The equations of the six sides, arranged in opposite pairs, are then

$A_{1}A_{5}:x_{1}z=z_{1}x$	$A_{4}A_{2}:y_{2}z=z_{2}y$
$A_{5}A_{3}:x_{3}z=z_{3}x$	$A_{2}A_{6}:y_{2}x=x_{2}y$
$A_{3}A_{4}:z_{3}y=y_{3}z$	$A_{6}A_{1}:y_{1}x=x_{1}y$

and the three points of intersection of pairs of opposite sides are

A_{1}A_{5}\\cdot A_{4}A_{2}=(x_{1}z_{2},z_{1}y_{2},z_{1}z_{2})

A_{5}A_{3}\\cdot A_{2}A_{6}=(x_{2}x_{3},y_{2}x_{3},x_{2}z_{3})

A_{3}A_{4}\\cdot A_{6}A_{1}=(y_{3}x_{1},y_{3}y_{1},z_{3}y_{1})

To say that these are collinear is to say that the determinant

D=\\left|\\begin{array}[]{ccc}x_{1}z_{2}&z_{1}y_{2}&z_{1}z_{2}\\\\x_{2}x_{3}&y_{2}x_{3}&x_{2}z_{3}\\\\y_{3}x_{1}&y_{3}y_{1}&z_{3}y_{1}\\end{array}\\right|

is zero. We have

$D=$	$x_{1}y_{1}y_{2}z_{2}z_{3}x_{3}-x_{1}y_{1}z_{2}x_{2}y_{3}z_{3}$
	$+z_{1}x_{1}x_{2}y_{2}y_{3}z_{3}-y_{1}z_{1}x_{2}y_{2}z_{3}x_{3}$
	$+y_{1}z_{1}z_{2}x_{2}x_{3}y_{3}-z_{1}x_{1}y_{2}z_{2}x_{3}y_{3}$

Using (1) we get

(x_{1}+y_{1})(x_{2}+y_{2})(x_{3}+y_{3})D=x_{1}y_{1}x_{2}y_{2}x_{3}y_{3}S

where $(x_{1}+y_{1})(x_{2}+y_{2})(x_{3}+y_{3})\eq 0$ and

$\\displaystyle S$	$\\displaystyle=$	$\\displaystyle(x_{1}+y_{1})(y_{2}x_{3}-x_{2}y_{3})$
	$\\displaystyle+$	$\\displaystyle(x_{2}+y_{2})(y_{3}x_{1}-x_{3}y_{1})$
	$\\displaystyle+$	$\\displaystyle(x_{3}+y_{3})(y_{1}x_{2}-x_{1}y_{2})$
	$\\displaystyle=$	$\\displaystyle 0$

QED.

Remarks: For more on the use of coordinates in a projectiveplane, see e.g. http://www.maths.soton.ac.uk/staff/AEHirst/ ma208/notes/project.pdfHirst(an 11-page PDF).

A synthetic proof (without coordinates) of Pascal’s theoremis possible with the aid of cross ratios or the related notionof harmonic sets (of four collinear points).

Pascal’s proof is lost; presumably he had only the real affine planein mind. A proof restricted to that case, based on Menelaus’s theorem,can be seen athttp://www.cut-the-knot.org/Curriculum/Geometry/Pascal.shtml#wordscut-the-knot.org.