1128 
MR. G. F. RODWELL OH THE EFFECTS OF HEAT OH CERTAIH 
supports for tlie rod of iodide of silver, and a system of vanes for agitating the liquid 
during heating. The rod of iodide was cast, sawn plane at the ends, smoothed by 
sand-paper, and furnished at one end with a cross wire arrangement for viewing in the 
field of the microscope. A cover fitted upon B through which a thermometer passed, 
also the head of the cross wire arrangement. The microscope was horizontally attached 
to a massive iron pillar, which was fixed in a block of white marble. Of course nume¬ 
rous precautions were taken to avoid the acquisition of heat by any portion of the 
microscope or stand, and every attempt was made to eliminate errors. But the 
method was unsatisfactory and had to be abandoned. Walking on the floor, resting 
an arm on the table, and other inevitable disturbing causes, produced tremors of such 
magnitude that all accurate observation came to an end. 
y. Method of the expansion apparatus .—I finally decided to attempt to enormously 
multiply a small motion by means of a system of levers; to cause bars of the iodide of 
silver, or other substance under examination, to act upon these levers while being 
heated through various known ranges of temperature; and to measure the movements 
by means of a micrometer-screw. This plan was adopted for all the determinations 
hereinafter given. 
The following apparatus was devised in order to carry out the idea. The box A, 
fig. 1 (drawn to scale), contains the levers, shown in plan and section in figs. 2 and 3 ; 
B is an index moving through 180° of arc over the graduated half-circle C. B is 
attached to the a.x is X, which is in connexion with the second lever U. D is a 
