224 Scientific Proceedings, Royal Dublin Society. 



This system, while giving even greater freedom of motion than the old 

 single roller, has the additional advantages that it rarely requires lubrication, 

 and that the weight being distributed among several rollers there is not so 

 much tendency to wear grooves on the rolling surfaces. 



Declination Axis. — The relief of friction for the declination axis of an 

 equatorial telescope mounting suitable for any latitudes except those near the 

 Pole is a much more complicated and difficult problem, as a little consideration 

 will show. 



Suppose the latitude to be, say, 45° north, and that fig. 1, Plate XVITI., 

 represents an end view of declination axis in its bearings viewed from the east 

 in north latitude — A and B being the bearings — it will be evident that in this 

 position the bearing B is taking the entire weight, while A has none to 

 bear ; while if the telescope is turned half round on axis P, P', so that the 

 tube be on the other side of the pier, the bearing A will be receiving all the 

 weight and B none, as in fig. 2, Plate XVIII. While in the six-hour position, 

 half-way between, the weight will be equally distributed between A and B. 

 Any arrangement, therefore, to relieve the weight in the first instance by an 

 upward thrust from B to C would act in a totally wrong direction when the 

 telescope is turned to the other side of the pier. The problem is, therefore, 

 much less simple than that of the polar axis, in which case the direction of 

 gravity bore a fixed relation to the position of the Y bearings. 



As far as I am aware, my father was the first to attempt to provide means 

 for satisfying these conditions, and on this principle : — In fig. 3, Plate XVIII., 

 the force of gravity acting in the direction C TF, instead of being resisted by 

 an upward thrust represented by the diagonal D C oi the parallelogram, could 

 be equally well resisted by two forces represented respectively by the sides of 

 the parallelogram A C and B C, of which one ^ C is constant in its direction, 

 irrespective of the revolution of the polar axis. He tlierefore used an arrange- 

 ment of this nature : — figs. 4, 5, and 6, Plate XVIII. : — The upper end of the 

 • polar axis is closed by a plate e e, having a hole in the centre ; G,G is the cross- 

 head; i^the declination axis ;/',/is a strong, steel, fork-shaped casting of a 1' 

 section, and partially surrounding the declination axis ; this tbrk,//^ is carried 

 on the end of a steel bar X which passes through the hole in the plate e e 

 loosely, and, projecting down through the polar axis, carries a series of heavy 

 weights, W. This steel fork, //, carries on its extremities two gun-metal 

 blocks which serve as carriers for three rollers each ; two of these rollers, r r, 

 at each side roll against turned rings on the declination axis provided for the 

 purpose, while the two rollers, y y, roll freely in a groove prepared in the axis 

 between the two rings. 



In the case of an equatorial at the Pole, of course, this apparatus would 



