4 SCHMITZ, On a Method of Ice Calorimetry. 



amount, causes trouble, not from any direct calorimetrical 

 effect, but from the fact that it prevents the formation of 

 an ice-coating completely enclosing the piece of metal. 

 The direct calorimetrical effect is exceedingly small : in 

 fact, if we attempt to make it large by allowing the object 

 to carry a comparatively large amount of liquid air into 

 the calorimeter, we find that the bulk of the liquid air 

 immediately rises to the surface of the water, where it 

 sometimes forms its own independent ice-capsules before 

 evaporating. 



As regards the time necessary to cool the object to 

 the temperature of the liquid air, this is easily settled, 

 either by observing the behaviour of the liquid air or 

 by quantitative experiments. The necessary time, in the 

 case of the objects I used, was five minutes or less, but 

 my practice was to allow the object to remain in the bath 

 of liquid air for a quarter of an hour or longer. 



In a few experiments I cooled the piece of metal by 

 allowing it to remain for a long time in a corked test-tube 

 immersed to a considerable depth in liquid air. On 

 plunging the cold metal into the calorimeter, an ice-jacket 

 was formed completely enclosing the piece of metal, of 

 very uniform thickness when the piece of metal was 

 spherical, and in some other respects superior to the ice- 

 jacket formed in my usual method of procedure. I made 

 however few quantitative experiments using this method 

 of cooling, for the following reasons. It seemed doubtful 

 whether even a prolonged cooling would reduce the metal 

 to the temperature of the liquid air.* The temperature 



* A platinum thermometer, protected by a corked glass tube with a metal 

 base from direct contact with the liquid air, showed a temperature far from 

 steady and differing by several degrees from the temperature of the liquid air, 



during observations continued for ninety minutes. But in this case it is 

 probable that the thick wire terminals had an appreciable effect in keeping 



up the temperature of the coil. 



