At the University of Wisconsin Oshkosh, the mathematics department teaches two half-semester courses in geometry, one in Euclidean geometry and one in hyperbolic geometry. Together, these courses make up a four-credit sequence, designed for prospective secondary teachers who are in their junior year of college. The sequence has both content and pedagogical goals, including the development of

- access to the major ideas of hyperbolic geometry,
- an enriched understanding of Euclidean geometry,
- a deeper conceptualization of the nature of proof and axiom systems,
- exposure to a practice of teaching consistent with the NCTM's vision of mathematics education.

It is with the above goals in mind that the hyperbolic toolbox has been completed. The models considered -- the Beltrami-Klein disk, the Poincaré disk, and the Poincaré half-plane -- provide a context for visualizing the theorems of hyperbolic geometry. By giving students examples of universes where Euclid's parallel postulate fails, they give students a chance to experience hyperbolic geometry in a concrete way: to see how it is possible for two lines through a point to be parallel to a third line, to appreciate a world where rectangles do not exist, and to understand how it is possible for the angle sum in a triangle to fall short of 180 degrees. The **Teaching Examples** on page 8 illustrate several ways in which the tools accomplish this.

Most of the students taking the geometry sequence at the University of Wisconsin Oshkosh have previously completed a course on the nature of proof. For many, however, the geometry course provides the first opportunity to apply their background in a specific subject. The geometry course emphasizes proper justification, proof, and the nature of axiom systems. A benefit of courses in Euclidean and hyperbolic geometry is that students can clearly see the role of axioms, for the only difference between the two geometries is the parallel postulate. Students learn to be careful about what statements are theorems in Euclidean geometry, which are theorems in hyperbolic geometry, and which are theorems in both. Constructions in the models aid in making these distinctions in two ways. For example, consider the statement "*Two lines perpendicular to a third line are parallel to each other*." After exploring that statement in the hyperbolic models, a student should get a sense that it is a theorem in hyperbolic geometry as well as Euclidean geometry. The student then knows that proving the statement will not require a parallel postulate. This is indeed a powerful connection for a student to make. Second, a compelling aspect of straightedge and compass constructions is the clear necessity of justification. When a student finishes a construction, only rarely does an instructor need to ask, "How do you know you are correct?" Students naturally ask that question themselves.

By examining hyperbolic geometry we obtain a deeper understanding of Euclidean geometry. The study allows for a contrast; we are able to see what makes Euclidean geometry "Euclidean" by negating a single axiom. Working within the hyperbolic models provides us with another quite different way of enriching our understanding of the Euclidean world. Though each of the models is "hyperbolic", the underlying space for each is the Euclidean plane. For example, the Klein model consists of a fixed circle and its interior, and "lines" in the model are merely chords of the circle. It is a purely Euclidean object. When viewed in this manner, the models can be used as a means of better understanding Euclidean geometry. One useful construction in the Klein model, for example, is the *pole* of a hyperbolic line: it is the intersection of tangents drawn to the Klein disk at the endpoints of the chord defining the line. Constructing the pole of a Klein line thus involves standard Euclidean straightedge and compass constructions (specifically, raising a perpendicular to a line from a point on the line). In this way, the models offer an opportunity to understand Euclidean geometry through the window of hyperbolic geometry. This argument is fundamental to the nature of the geometries, for the relative consistency of hyperbolic geometry is demonstrated by the existence of models of hyperbolic geometry in Euclidean geometry.

It should be noted that the fourth goal is an implicit one. The course is content-driven, and discussion of teaching methodology is not an explicit part of that content. However, the NCTM's Professional Standards for Teaching tells us, "Learning is an active, dynamic, and continuous process that is both an individual and a social experience." (see Professional Development Standard 3.) We would like our future teachers to support such a learning environment, and so providing them with a model of such a classroom is vital. The **Teaching Examples** on page 8 offer possible ways in which the tools can be used to stimulate an active learning environment.

As noted in the **Background** section (Page 5), others have produced *Geometer's Sketchpad* construction tools for the Poincaré disk and (in part) for the Poincaré half-plane models. Completing the tools for the half-plane and contributing tools for the Beltrami-Klein disk rounds out a package that allows students to analyze hyperbolic geometry in a concrete way from several different perspectives. As the NCTM notes in the Principles and Standards for School Mathematics , "It is important for teachers to highlight ways in which different representations of the same objects can convey different information and to emphasize the importance of selecting representations suited to the particular mathematical tasks at hand." (p. 362) The different models do indeed offer different representations of hyperbolic geometry. The Poincaré models are conformal, and their faithful representations of angle measure make them attractive for this reason. On the other hand, the hyperbolic axiom is most immediate to visualize in the Klein disk, and constructing the common perpendicular between two divergently parallel lines in the Klein model is an accessible problem. The three models together provide a powerful means of considering hyperbolic geometry from different perspectives.