82 Dr. Meyer Wilde rman on the Velocity of 



salt from supersaturated solutions) 36 movements of the 

 stirrer per minute more than amply provided for the instanta- 

 neous mixing of the liquid : a perfect curve was obtained. On 

 the contrary, when the reaction took place only in a few planes 

 of the liquid, e. g. when an ice cube melted in the warmer 

 water, there was no possibility of getting a regular curve, 

 however rapid and energetic the stirring might be. Every 

 time the stirrer comes to the top or to the bottom it stops for 

 a great part of a second ; and since the stirrer alone provides 

 for the equalization of temperature, the photographic curve 

 indicates all these interruptions of stirring. The curve con- 

 sists of a series of smaller curves, every one belonging to one 

 movement of the stirrer up or down. To get these curves as 

 Mnail as possible, and so to approximate to the curve which 

 is the true one, the process of ice melting had to be investi- 

 gated first of all at lower temperatures, where the velocity of 

 reaction is considerably slower, and then a more rapid and 

 effectual stirring had to be used*. 



The upper part of the rod of the nickel stirrer was covered 

 with a thick layer of asbestos, notably a very bad conductor of 

 heat, and through this either fastened to the guide (g) moved 

 by the electrometer or held by hand. The rate of stirring was 

 kept at 36 per minute. 



D. During the experiment the heat taken by the liquid from 

 the surrounding medium or given to it can be neglected. 



To reduce the velocity of cooling or heating of the liquid 

 in the metal beaker by the surrounding medium, 4^ litres of 

 liquid instead of 1250 c.c. and an air-bath were used. The 

 velocity constant, C, in the equation of cooling, when the 

 Jiquid is removed one degree from the convergence-tempe- 

 rature is about o, 004 per minute. The experiment was 

 always so conducted as to get the point of equilibrium removed 

 from the convergence-temperature only by about o, l. The 

 photographic curves show that after a few stirrings the tem- 

 perature of the liquid becomes constant at T 0tf (the line runs 

 parallel to the edge of the paper) , so that even at T „, where 

 the liquid is most removed from the convergence -temperature 

 and the heating or cooling of the liquid by the surrounding 

 medium is the greatest, the effect of the latter during the 

 photographed minute cannot be clearly established. As now 



* These observations explain to a great extent the reason of the known 

 failure in getting good concordant results for almost all chemical reactions 

 in heterogeneous systems. This failure is chiefly due to the difficulty in 

 getting a sufficiently rapid equalization of the concentration throughout 

 the liquid and on the surface of the liquid in contact with the solid. 



