LIQUIDS MAY BE MEASURED. 25 



The following arrangement (Fig. 6) 

 will probably be found preferable to the 

 one just described, in many cases. Its 

 construction depends on the observation, 

 that for the phenomenon itself, and for 

 the result of the experiment, it is entirely 

 a matter of indifference whether the tube 

 be closed with a single, double, or treble 

 layer of bladder.* For experiments on 

 very thin membranes which are permeable 

 to liquids under a very low pressure, the 

 apparatus (Fig. 5) is obviously better 

 adapted. For the explanation of the phe- 

 nomenon we have to distinguish 



1. The mixture of different liquids. 



2. The change in their volume. 



As to the mixture of two liquids of dissimilar Causes of 



the mixture 



nature and characters, this always depends on a ofdissimi- 



. lar liquids. 



chemical attraction. In a mixture of alcohol and 

 water, or of brine and water, there is in every part 

 the same proportion of particles of alcohol and 

 water, or of salt and water. If in the former, the 

 lighter particles of alcohol lying at the bottom of 

 the vessel were not retained, in the place and 



* In these experiments membranes of all kinds may be used. 

 With the thinner membranes, such as the bladder of the calf and 

 the pig, the experiments are more rapidly completed than with 

 the thicker, such as the gall-bladder and urinary bladder of the 

 ox. The peritoneum of the ox and calf is preferable to all 

 others. The tube c is tied with bladder under water. 



