THE HYDRATE THEORY OF SOLUTIONS 131 



exists, and which Pickering himself recognised, 1 that the careless 

 use of a bent lath may lead to the recognition of imaginary 

 breaks in a continuous (but non-parabolic) curve, reference may 

 be made to the viscosity curves for mixtures of pyridine and 

 water. These are very similar in form to the inflected con- 

 ductivity-temperature curves, and might in all probability be 

 represented by similar equations. It is, however, noteworthy 

 that whilst Hartley, Thomas, and Appleby {Trans. Client. Soc. 

 1908, 93, 546) were able to represent their experimental points 

 in a perfectly satisfactory manner by a single continuous curve, 

 the previous observers (Dunstan, Thole, and Hunt, Trans. 

 Chcm. Soc. 1907, 91, 1728) were led by the use of a bent lath to 

 plot their points as lying on a series of no less than seven 

 intersecting parabolas. The discontinuity in this case appears 

 to have been due in part to small experimental errors ; but even 

 if these were eliminated, it would be necessary to resolve the 

 curve into almost as many separate sections before it could be 

 distorted successfully into a parabolic form. 



But whilst the general acceptance of the hydrate theory of 

 solutions was undoubtedly retarded by the extreme complexity 

 of the hydrates postulated and by the uncertain character of the 

 methods used in detecting them, a far more serious obstacle 

 arose from the incompatibility of Mendeleef's theory and method 

 with the ideas of reversibility and mass-action which were 

 gradually permeating the whole of chemical science. The in- 

 spiration which led him to look for, and to detect, abrupt 

 changes in his experimental curves was derived from a view 

 of the process of hydrate formation in solution, which finds a 

 close analogue in the actual behaviour of solid hydrates. Pareau 

 had shown in the case of copper sulphate {Ann. Phys. Chem. 

 1877 (3), 1, 39) that each solid hydrate possessed a definite 

 vapour-pressure increasing with the temperature in the same 

 way as in the case of water itself. At 50 the values were : 



CuSO, o 



CuS0 4 .H 2 4-4 mm. 



CuS0 4 .3H 2 30 mm. 



CuS0 4 .5H 2 47 mm. 



Water 91 mm. 



1 " The application of the graphic method requires a great amount of care and 

 a close attention to experimental and other conditions, and it is to be feared that 

 the hurried use of it by those who have not taken the trouble to master the 

 necessary details, or to acquire the requisite amount of skill, may bring it into 

 undeserved disrepute " {Phil. Mag. 1892, 33, 451). 



