On the Method of Condensation in Calorimetry. 307 



experiment on the constancy of weight has been extended to over 

 half an hour on many occasions. 



The nse of plaster of Paris in forming the aperture avoids what 

 uncertainty and want of sensibility would be introduced into the 

 weighing by water permitted to accumulate in the orifice. Until this 

 plan of catting the aperture in a bibulous material was resorted to, the 

 accumulation of water gave much trouble. The plaster remains dry 

 when in use for an indefinite time, and wears well. That now in the 

 calorimeter has served through more than 300 experiments. It is also 

 very easily renewed. 



In order to avoid as far as possible escape of steam up the wire and 

 ensure constancy of pressure, I have lately adopted a plan which acts 

 automatically in keeping the calorimeter fall of steam sensibly at 

 atmospheric pressure. The arrangement consists in locating the 

 escape valve of the boiler at a level so much below the level of admis- 

 sion into the calorimeter, that a small and constant pressure — due to 

 the levity of the steam — is established, urging the gas to ascend the 

 tube leading to the calorimeter. This end is secured by affixing the 

 escape-valve to the end of a wide tube which is taken from the boiler 

 bent twice at right angles and carried down to a level outside the 

 boiler sufficient to secure the requisite pressure ; a couple of deci- 

 metres "head" will in general be sufficient. 



The effect on the weighing of steam draught within the calorimeter 

 can be entirely avoided by shielding or jacketing. In the form 

 already described, indeed, it seldom gave trouble, but in this respect 

 I have in a recent form improved the apparatus. It is worthy of 

 notice, also, that the conditions prevailing when we weigh a substance 

 at a temperature different from that of the air within the balance 

 case do not obtain here, the substance in the calorimeter being 

 surrounded by an atmosphere at its own temperature. 



Description of an Improved Gravimetric Calorimeter. 



I have designed an apparatus intended to carry out such conditions 

 as have suggested themselves as conducing to accuracy and to ready 

 working. Its advantages have already to a great extent been prac- 

 tically tested in a recently completed instrument, but this instrument 

 serving itself to suggest improvements is not the same as that now 

 described. Plate 7 shows a vertical section and plan. 



The apparatus it will be seen consists of a cylindrical chamber with 

 conical ends. This chamber is divided in a vertical plane passing- 

 through its axis, the separate halves being firmly hinged on a tripod 

 borne on three adjusting screws. The chamber is thus readily opened 

 by turning back the halves, and closed by pressing them together, 

 when they meet on a ground edge. It is made of brass and kept as 



