440 SUPPLEMENT. 



The brass plate A D, at the top, is parallel to the 

 equator, and the axis E F G is perpendicular to it. If 

 right lines be drawn from the center of this plate, to 

 the upper ends of the equidistant parallel lines on the 

 outside of the tube ; these right lines will be the hour- 

 lines on the equinoctial dial A D, at 15 degrees dis- 

 tance frotn each other, and the hour-letters may be set 

 to them as in the figure. Then, as the shadow of the 

 axis within the tube comes on the hour-lines of the 

 tube, it will cover the like hour-lines on the equinoctial 

 plate A D. 



If a thin horizontal plate ef be put within the tube, 

 so as its edge may touch the tube all around ; and right 

 lines be drawn from the center of that plate to those 

 points of its edge which are cut by the parallel hour- 

 lines on the tube; these right lines will be the hour- 

 lines of a horizontal dial, for the latitude to which the 

 tube is elevated. For, as the shadow of the axis comes 

 successively to the hour-lines of the tube, and covers 

 them, it will then cover the like hour-lines on the hori- 

 zontal plate ef, to which the hours may be set, as in 

 the figure." 4 



If a thin vertical plate g C, be put within the tube, 

 so as to front the meridian or 12 o'clock line thereof, 

 and the edge of this plate touch the tube all around ; 

 and then, if right lines be drawn from the center of the 

 plate to those points of its edge which are cut by the 

 parallel hour-lines on the tube; these right-lines will be 

 hour-lines of a vertical south-dial : and the shadow of 

 the axis will cover them at the same times when it 

 covers those of the tube. 



sphere, so that the axis of both solids should coincide, the projections 

 of the hour-lines of the sphere, upon the surface of the cylinder, would 

 be at equal distances from each other. In other words, the hour-lines 

 of the cylinder would be equidistant. 



Note 134. The dialing cylinder, like the globe, most be considered 

 as useful rather as an illustration than an a method of describing dials. 



