DILUTION OF CONCENTRATED SOLUTIONS. 321 



the concentrated solutions without making them dilute in the process ; thus about 

 20 gr. of water were added to 300 gr. of solution. The heat generated 

 or absorbed in this case divided by the number of gr. molecules of water added, 

 gives the differential heat of dilution sufficiently accurately for a concentration which 

 is the mean of the initial and final concenti^ations. 



The differential heat of dilution d Q/rZN may be defined as the quantity of heat 

 liberated or absorbed per molecule of water added to an infinite quantity of solution, 

 whose concentration will thereby be unaffected. This condition can only be 

 approximated to, but the error in the above estimate will be small. 



THOMSEN added large quantities of water and so obtained a total heat 



Q =. [ ' Sg . 3N. This involved a large thermal equivalent and a small rise in 



Jy, 



temperature. 



By the method here described the thermal capacity is kept sufficiently low to 

 give a fair rise in temperature on dilution, and the relatively small quantity of 

 water employed can be placed in a vessel completely immersed in the solution. 

 Temperatures can thus be well equalised before dilution takes place. 



The following record describes work extending from October, 1912, to December, 

 1914. The accurate determination of heat of dilution is necessarily a tedious business, 

 owing to the fact that only one observation can be taken in an ordinary working day 

 of six hours. 



(2) DESCRTPTFON OF APPARATUS. (Diagram T.) 



Measurement of Temperature, 



In most of the previous researches on heat of dilution a mercury thermometer was 

 employed. Here temperature changes were measured with certainty by a platinum 

 thermometer with fundamental interval of 12 - 8 ohms mounted in a very thin- 

 walled cylindrical bulb, whose length is not far short of the depth of the solution. 

 Long experience with the working of this thermometer in other researches and 

 careful application of all the corrections peculiar to it, enable the author to guarantee 

 its readings. (Diagram I., T.) 



Importance of Stirring. 



One of the most important features of the experiment is the stirrer. Different 

 types were tested, and the most efficient one was found to be of the screw propeller 

 type turned by a high speed motor. The process of mixing was rapidly performed 

 even with some of the very viscous solutions employed, and a careful experiment was 

 performed to see if any appreciable temperature rise was indicated during the process. 

 After long periods, less than xth} C. rise was noticeable. (Diagram I., S.) 



2 u 2 



