744 Mr. Gervaise Le Bas on the 



easily be shown that considerable irregularities occur in the 

 data, so that we are under the necessity of smoothing the 

 curves. 



The accompanying diagrams show how the values o£ V/W 

 vary in 



R_C— 0— R' 

 II 

 



(a) according as R is increased in complexity, 



(b) according as R' is increased in complexity. 



Diagram A gives the methyl, ethyl, and propyl salts of 

 the different fatty acids. 



Diagram B gives the different salts of the various acids, 

 formic and acetic. 



There is a striking difference between them. 



Diagram A shows regular open curves which are similar 

 to that for the paraffins. This is the effect one would expect, 

 seeing that the radicle of increasing complexity R is attached 

 to a carbon atom (the C radicle). The similarity of the 

 two series is increased by the fact that in the methyl curve 

 we have successively 



H— C — , CII3— C — , C 2 H 5 - C-&c. 

 li II II 







combinations similar to H— CH 3 , CH 3 — CH 3 , 2 H 5 — 0H 3 . 

 In consequence of this we find that the initial part of the 

 methyl curve shows a rapid decrease before the minimum. 

 The actual values are of course depressed owing to the 

 influence of — OCH 3 . 



All the methyl salts are smaller in volume than the corre- 

 sponding (metameric) ethyl salts, and the latter slightly 

 smaller than the propyl. This question will be considered 

 later. 



Turning to Diagram B we find quite a different result. The 

 increase in complexity is connected with the radicle — that 

 is, the curves show the variation of V/W when — — R 

 increases in complexity. The C radicle remains constant. 



We note first the unique character of the curve for the 

 esters of formic acid. 



If we compare this in its earlier stages with the acetic 

 ester curve we note how the influence of H — C is exerted in 



II 





 raising the value of V/W just as the value for H — CH S is 

 exceptionally high. This is in spite of the presence of the 



