276 M. R. Ruhlmann on the Alteration produced by Heat 



the two sharp edges of the opening and took the mean (fig. 4). 

 At all events, by this means greater accuracy is secured than 

 was attained by Schmidt, who measured this distance from the 

 axis of rotation to the face of the prism next to the telescope. 



(4) The hollow prism. — Two conditions are necessary in re- 

 gard to this : first, that it be easily and surely warmed ; and se- 

 condly, that it be kept at as constant a temperature as possible 

 during the experiment, I had the arrangement, at first, as fol- 

 lows : — A glass tube of about 2 centims. internal diameter, and 

 2 or three millims. thickness of glass, was ground smoothly at 

 the ends in such a manner that the sides were inclined to one 

 another at an angle of about 60°. Exactly in the middle of this 

 tube was a circular opening of about 1 centim. radius, into 

 which another glass tube was cemented watertight for the recep- 

 tion of a thermometer. The ground tube was placed in a 

 prism-shaped sheet-metal box open above, and whose section 

 was an equilateral triangle. Through the side walls the glass 

 tube projected about 1 millim., and was fitted and cemented as 

 closely as possible into the box. Upon the two flat glass edges 

 were laid two plane glass plates which were taken from an old 

 artificial horizon by Troughton, and which had been previously 

 tested by means of reflexion in regard to their flatness and the 

 parallelism of their faces. By means of a brass diaphragm, 

 which was fastened by three screws to the metallic box, the 

 glass plates were pressed upon the tube by turning the screws. 

 This entire box was placed in a second similar one, which was 

 only so much greater than the first as to receive it without 

 sticking or touching. The second box had a triangular hole cut 

 out of its bottom, so that a lamp standing on the theodolite 

 could directly warm the bottom of the smaller prism. The 

 entire arrangement was closed by a cover which had only a small 

 hole corresponding to the glass tube. At the places where the 

 introduced tube enters the inner box and is closed by the plates 

 A, openings are formed in the outer box P 2 which can be closed 

 by flaps K (fig. 5). The liquid to be examined is introduced 

 into the glass tube G ; and Pj is filled with any other liquid. A 

 thermometer is introduced at R, its bulb reaching exactly into 

 the middle of the glass tube G ; and it carries a cardboard screen 

 immediately below the scale, in order to protect the projecting 

 part of the scale from radiant heat. The liquid under examina- 

 tion was surrounded by another liquid, in order to warm it more 

 generally, and to avoid thereby quick and sudden changes of 

 temperature. The object of the external box was to diminish 

 the outward radiation of heat by interposing a layer of air as a 

 bad conductor. In order to hinder the mobility of the air and 

 to diminish the effect of upward currents, the intervening space 



