CLARK. — SURFACE TENSION BETWEEN LIQUIDS AND VAPORS. 377 



This was interpreted as follows. The effect of Cantor comes first, form- 

 ing a film of ether on the glycerine surface. Owing to the concavity of 

 this surface, the ether vapor is not in equilibrium with an ether surface 

 at this temperature and is consequently condensed. The presence of a 

 small quantity of glycerine vapor undoubtedly facilitates this condensa- 

 tion. It will be noticed that the capillary tube contains the hottest 

 ether vapor while the whole is cooling off, and here the vapor should 

 condense last. This excludes the possibility of unequal heating being 

 the reason for this effect. 



In order to measure the surface tension between liquid glycerine and 

 the vapor of ethyl ether, a special method was employed to keep the tem- 

 perature of the glycerine above that of the ether so as to avoid the above- 

 mentioned condensation. The glycerine tube was surrounded by a small 

 coil of German silver wire and a separate heating current sent through 

 this. The method of procedure was as follows. When the desired 

 temperature of the heater was obtained, a measured current was sent 

 through this auxiliary coil for a measured time, and then when the con- 

 densed liquid had disappeared the current was shut off and measurements 

 taken of the height of the meniscus until the shape of the meniscus 

 changed due to the redeposit of ether. After the completion of the 

 experiments, enough of the tops of the two tubes was removed to insert 

 the two junctions of an iron and German silver thermo-couple. Then the 

 same current that was used previously was sent through the auxiliary coil 

 for the same time and the change in temperature inside the capillary tube 

 measured. Of course the thermo-couple was calibrated by a separate 

 experiment. In this way a rough determination of the rise in tempera- 

 ture was made. It is in error, for the specific heat of ether under 

 such circumstances is unknown. I have used the value for ordinary 

 temperatures. 



This temperature of equilibrium between a vapor of one liquid and 

 another liquid has apparently not been studied except by Cantor at room 

 temperature. This will be taken up again by a new method, and 

 instructive results are expected. 



Ethyl alcohol and olive oil belong to the first class of liquids, and no 

 precautions were necessary to prevent distillation. Here, however, there 

 is considerable absorption of the vapor by the oil especially at high tem- 

 peratures. The mixture becomes homogeneous at about 240°. 



The alcohol was the purest obtainable and distilled into the tube as 

 described, and the oil was freed from volatile matter by boiling at high 

 temperature. 



