# Hyperbolic inversions

 In this section, we establish basic results about hyperbolic lines and angles. This includes results illustrated in the experimental pages. Taking the kleinian view, we base our development of the theory on a group of transformations of the hyperbolic plane. Suppose that H is an h-line. Then H = LnD, where L is an i-line orthogonal to C. From the Mirror Property, inversion with respect to L, iL, maps D to D. We refer to the restriction of iL to D as h-inversion with respect to H, denoted by iH. The h-inversions play a role in hyperbolic geometry very similar to the role of reflections in euclidean geometry. For example: The h-inversions generate the hyperbolic group H(2) whose elements are called hyperbolic transformations. Theorem 1 If t is a hyperbolic transformation, then t maps h-lines to h-lines, t preserves angles between h-lines. Proof of Theorem 1 In our Poincare disk model, O appears to have a special status, since the h-lines through O are euclidean segments rather than arcs of circles. In fact, it is typical, as we shall see. Origin Lemma For any point P ε D, there is an h-inversion mapping P to O and O to P. If P ≠ 0, then the h-inversion is unique. The CabriJava pane on the right illustrates the Lemma. You can move P and note how the green h-line varies. Of course, if you drag P outside the circle, the is no h-line. More generally, we have the Interchange Lemma If P and Q ε D, there is an h-inversion mapping P to Q and Q to P. If P ≠ Q, then the h-inversion is unique. Since the h-inversion, and hence the h-line, in the Interchange Lemma are unique, we make the Definitions If P and Q are distinct points of D, then (1) the h-line H for which iH interchanges P and Q is the h-bisector of the h-segment PQ. (2) the point where the h-bisector of PQ meets PQ is the h-midpoint of PQ. The justification for the names h-bisector and h-midpoint will become clear when we discuss hyperbolic distance. These Lemmas can be used to prove fundamental results on hyperbolic geometry. The Origin Lemma is often used to map a chosen point to O to simplify the figure. The Incidence Theorem two distinct points define a unique h-line, two distinct h-lines meet in at most one point. To illustrate the use of the Lemma, we give the proof in full. Proof (1) Suppose that P and Q are distinct points of D. By the Origin Lemma, there is a t in H(2) with t(P) = O. Now, any h-line through O is a diameter of C. There is a unique diameter H through t(Q). Then t-1(H) is the required h-line. (2) If two h-lines meet in two (or more) points, then, by (1), they are identical. Thus, distinct h-lines meet at most once. Note that these are exact analogues of incidence results in euclidean geometry. However, the possibilities for non-intersecting h-lines are rather different. This has been discussed in the experimental pages. There are many analogues of euclidean results. We give some which will be useful later. The Angle Bisector Theorem If P, Q and R are points of D, with R not on ther h-line PQ, then there is a unique h-line H bisecting