Let’s do something simple today, something rational. We all know what a circle is, and most should know that there are points on the unit circle with rational coordinates, like (3/5, 4/5). This is because of the birational map called stereographic projection, known since antiquity:


Take a point t on the real axis, connect it to the north pole (0,1) on the unit circle with a straight line, and find the second intersection of that line with the circle. This is a rational expression in t. So when t is rational, you get a point on the circle with rational coordinates. This is, of course, also a quick way to get (all) Pythagorean triples. But today we are going elsewhere. Now that we have many points with rational coordinates on a circle, we can make rational polygons, like the one below.

9 gon

This 9-gon is not only rational, but super-rational in the sense that all its edge lengths are rational numbers. Try it out. Even better: All the diagonals are rational as well. Is this a miracle? Are there others? No and yes, of course. Let’s get started:


Using rational versions of the sine and cosine functions, we can write down rational rotation matrices. They will (for rational t) rotate any point with rational coordinates on a unit circle to another point with rational coordinates. What we are interested in are superrational rotations: Those that rotate a point to any other point. The example above suggests that there are many of those.

I will give the answer next time. For the moment, only a hint: The superrational rotations form a subgroup of the group of rotations. Which is it?


Odd Angles

For a while, this will be my last post about conformal spiderwebs. Today, we will still look at circular quadrilaterals that are conformal images of squares, but allow the angles to be multiples of 90 degrees. Like so:

2quarter a 01

Let’s call this a square of type (1,1,3,3). Multiply the numbers by 90 and you get the angles at the vertices. I have again employed Möbius to place three corners at (1,0), (0,1), and (-1,0). The fourth vertex is again moving cautiously along the unit circle. Below is a square of type (1,3,3,3), and here the fourth vertex is on the x-axis, the second possible case we noticed for right angled circular quadrilaterals.

Mixed31b 0

Similarly, here is a square of type (1,1,1,3), also with the fourth vertex on the x-axis.

1quarter a 01

Missing are squares of type (1,3,1,3). While there are quadrilaterals of this type, all conformally correct squares I could find were only immersed (i.e. overlapping).

Then one can also have squares of type (2,2,2,2), for instance. The circle would be an example, with artificial vertices at (1,0),(0,1),(-1,0) and (0,-1), but there are also bean shaped squares like the one below.


Finally, the square with zero angles, in its most regular form.

Zero 4 01

Not Being Square

I meant to post today a sequel to the circular triangles from last week, but I got carried away looking at right angled quadrilaterals bounded by circular arcs. Like the pillows, but more general. Like so:

Circulon 2

The question arises for what choices of four points we can find a right angled quadrilaterals bounded by circular arcs?

By the way, how do we call these? I thought about circulons (taken) and horny squares (oops). For now, I call them circulions (like centurions), to avoid a lawsuit about trade marks. Above you see a solution that is not a square but where the vertices are at the corners of a square. There are more like these, in fact a 1-parameter family.

Circulon 1

Below you can see the entire family at once, you just have to follow all dots with the same color.


Can we do that for any choice of four points? Not so, but: Möbius allows us to move three points anywhere we like (and he will send circulions to circulions), so we can ask: where are we allowed to place a fourth point so that there is a circulion through all of them?

Möbius also tells us that this is easy if we place all four points on a circle (by sending that circle to a line, and then connecting the four points on the line alternatingly by segments of the real line and half circles, for instance). Here is an example where the first three points are at the corners of an equilateral triangle, and the fourth point is on the circle through them.

Circulon 3

Again, there is a 1-parameter family of such circulions through these points.


Pretty, isn’t it?

Now, surprisingly to me, for each choice of three points, there is a second circle on which the fourth point can reside: Take the circle that contains the given three points, and construct the circle orthogonal through it that passes through the two points between which we want to put the fourth vertex. You can put the fourth point anywhere on that new circle. Here is an example, with the first three points again at the corners of an equilateral triangle.

Circulon 4

Below is again an entire family, color coded and adorned with moiré.


Next week you’ll see conformally correct squares. Promised.