98 Proceedings of tJie Royal Society of Victoria. 



the second vessel was similarly hung inside a third vessel, and 

 this finally hung in a cylindrical outer shell of galvanised iron. 

 The whole formed a '-nest" of calorimeters very much like a nest 

 of beakers, but each vessel was in contact with the next only 

 along a small flange, and to minimise conduction at these points 

 strips of asbestos-paper were interposed. The calorimeter itself 

 held about 200 grams of water, and the system of air-spaces and 

 polished metal shields protected it very effectively from i-adiation 

 and convection from the sides and bottom. The top, however, 

 could only be protected by a small, moveable lid to allow of the 

 manipulations. The temperature of the water in the calori- 

 meter Avas read on a large mercury thermometer divided to 

 one-hundredth of a degree Centigrade. Attached to the outer 

 shield was a small piece of clock-work driving a spindle carry- 

 ing a small screw-propeller immersed in the calorimeter. The 

 liquid to be experimented upon was enclosed in a small, flat, silver 

 flask, holding about twenty cubic centimetres. The mouth of 

 this flask was closed with a tightly-fitting ebonite stopper ; 

 through two suitable holes in this stopper passed a fine copper 

 and german-silver wire I'espectively ; inside the flask these were 

 soldered together and coated with a thin coat of paratiin, thus 

 forming one thermo-electric junction. A second and similar 

 junction was placed in the calorimeter close to the bulb of the 

 large thermometer, and the pair of junctions were connected with 

 a low-resistance reflecting galvanometer showing a spot of light 

 on a scale. The procedure was then as follows — first both cal- 

 orimeter and silver receptacle were weighed both empty and then 

 with their contents ; everything being put in its place, the small 

 silver receptacle was warmed, in an air-bath, to a temperature 

 considerably above that required for the experiment — then it was 

 wrapped in cotton-wool and allowed to cool slowly, the observer 

 keeping it agitated and in the meantime noting the reading of 

 the thermometer in the calorimeter. At the first heating, the 

 spot of light was generally driven completely ofi" the scale, but as 

 the contents of the receptacle cooled the spot re-appeared. As 

 sooia as the spot approached to within a few divisions of one 

 particular reading, the stirrer was started, the large thermometer 

 read, and the silver flask taken from its wrapj^ings was brought 

 close to the calorimeter, and at the instant that the spot of light 



