126 S. L. Penfield — Stereographic Projection. 



poles (the parallel 40° South, plate IY) has an infinite radius, 

 and appears as a straight line in the projection. Upon this 

 straight line are located the centers of all the stereographically 

 projected meridians, for they are all arcs of circles running 

 through the north and south poles, hence having their centers 

 on a line crossing the middle point of the ~N. S. diameter at 90°. 

 The points of intersection of the meridians with the equator 

 may be found by drawing an arc, with a radius like that of the 

 equator, on an impression of protractor No. IJ, plate I. Then, 

 by means of dividers, the points 5°, 15°, 25°, etc., from the 90° 

 line of the protractor are transferred to the shifted equator. 

 It took very little time to construct the parallels and meridians 

 of plate IY, and, considering the size of the projection, they are 

 accurately drawn. The outlines of the continents were sketched 

 upon the chart by Mr. H. H. Robinson of Yale University. 



Professor Andrew W. Phillips of Yale University has devised 

 a machine consisting of jointed rods, by means of which the 

 pole of any stereographic projection can be shifted to any 

 desired position. Thus an equatorial projection, the easiest of 

 all to make, can be transformed into a meridian projection, or 

 into one like plate IY, having some desired point at the center. 

 This machine was exhibited in 1884 before the British Asso- 

 ciation for the Advancement of Science, at their summer 

 meeting in Montreal,* and is described by Professor Phillips 

 in his geometry. f 



Map Projection. — The method of projection almost uni- 

 versally employed by geographers for representing hemi- 

 spherical surfaces is the so-called Globular Projection, invented 

 in 1660 by the Italian Nicolosi.J In this method the equator 

 is divided into equal parts, and the meridians are circular arcs 

 uniting these points with the poles ; the parallels are likewise 

 circular arcs, dividing the extreme and central meridians into 

 equal parts. Figure 24 shows the meridians and parallels in 

 globular projection. Compare this figure with the stereo- 

 graphic projection of the meridians and parallels, plate III, and 

 a marked difference is at once apparent. The stereographic 

 projection is correct in every particular, the parallels intersect 

 the meridians at right angles, as on a globe, and, as has been 

 shown, distances and directions can be accurately measured and 

 plotted on such a projection. In the globular representation, 

 on the other hand, nothing is correct except the graduation of 

 the outer circle and the directions of the two diameters ; dis- 

 tances and directions can be neither measured nor plotted. 



* Report of the British Association for the Advancement of Science, 1884, p. 649. 

 f Phillips and Fisher's Geometry, 1899, p. 510. 



X Germain, Traite de Projections des Cartes Geographique, p. 127, Paris, about 

 1865. 



