1904.] Cotton 'by Water and ly Water Vapour. 253 



two experiments, and this was reached in 2 minutes, as nearly as possible. 

 The first test was continued for 40 minutes, when about 17 milli- 

 grammes of vapour was found to have been condensed ; the second 

 was stopped at 5 minutes, when about] 7 was the weight found. The 

 temperature curve resembled that of cotton wool in a general way, for 

 it rose steeply to the maximum and showed a slow descent ; but its 

 form was quite different in this, that it gave not even a rough 

 approximation to those rules which have been found to govern the 

 other case. This fact and the relatively small absorption (allowance 

 being made for difference of surface area) are just what might be 

 expected from the presumably impermeable character of glass, as 

 here probably the condensed water remains as a mere surface film. 



According to the views already put forward as to the cause of the 

 temperature change when cotton is immersed in water itself, it seemed 

 probable that little or no effect of the kind would be produced by 

 immersing cotton in absolute alcohoLor glass wool in water, though in 

 both cases notable effects followed immersion in the bath of air 

 saturated with vapour. As a fact, no change could be observed. In 

 the case of glass wool and water, the test was very carefully carried 

 out, and there was not the least sign of departure from the initial 

 temperature, which was that of the' thermostat surrounding the water. 

 A change of 0*02 could hardly have escaped notice. Apparently, 

 then, the effect does not occur when the liquid is capable of rapidly 

 and completely wetting the solid, and thus destroys the air insulation. 

 In confirmation, it was found here, as in the case of " absorbent " 

 cotton wool already described, that^the thermometer when lifted from 

 the bath carries with it a much larger weight of mechanically adhering 

 water than comes away with a wrapping of cotton of the ordinary 

 variety. 



There is not the least doubt that the thermal effect dealt with 

 in this paper is radically different from that investigated by 

 Dr. G. J. Parks by the use of precipitated silica and of glass wool* 

 and attributed by him to a pressure at the surface of contact of solid 

 and liquid. In the first place, the Parks effect is manifested as a 

 general rise of temperature on mixing the solid and water, and he 

 detected it and determined its value by employing the immersion vessel 

 as a calorimeter; whereas the cotton wool effect is essentially local 

 and would be altogether prevented by any mixing process efficient 

 enough to make complete contact between the solid and liquid. In 

 the next place, Parks took the maximum temperature, reached in 

 about 3 minutes, as indicative of the completion of the action, while 

 it has been proved in the other case that the action, though leading 

 to a maximum temperature in about the same time, has then but 

 started on an apparently endless career. Finally Parks estimates the 

 * ' Phil. Mag.,' August, 1902, p. 240, and May, 1903, p. 517. 



