OF ORGANIC SUBSTANCES ON THE RAYS OF THE SPECTRUM. 
265 
Propionic Acid. Specimen No. 1.-—This was a very good sample of the acid, 
boiling at 140°-141° after rectification. 
Specimen No. 2.—This was prepared from the nitrile. Boiling point 141°. 
Specimen No. 3.—This was a fraction of the distillate from No. 2. It boiled at 
139 -141°. These three specimens do not greatly differ in the length of transmitted 
spectrum; the two acids prepared from the nitrile are more diactinic than the other. 
The acid examined by M. So ret corresponds well with these. 
Butyric Acid. —No. 1 boils at 161°‘5-162°, No. 2 at 160°-161 o, 5. Both portions are 
very nearly the same in diactinic property, and equal thicknesses of the liquid being 
taken they do not differ much from the propionic acids. 
The specific volumes of these four fatty acids are the following : — 
Formic acid.42 
Acetic.64 
Propionic.86 
Butyric.108 
Consequently we must examine thicknesses of liquid proportional to these numbers 
in order to see the change in diactinic property corresponding to molecular differences. 
Since, however, we have not been able to obtain pure formic acid we have very 
carefully crystallised formate of sodium and compared it with sodium salts of those 
acids, molecular weights of the salts being dissolved and made up to equal volumes of 
liquid. One of the purest specimens of the sodium formate was prepared from a 
beautifully crystallised copper salt which had been completely dried and heated to 
near the temperature at which it decomposes. This was dissolved in water and 
mixed with a solution of carbonate of* soda in just sufficient quantity to precipitate 
the copper. A little cupric hydrate remained dissolved, and was separated by electro¬ 
lysis in a platinum dish with the aid of one of Grove’s cells. The aqueous solution 
of the sodium salt was slightly yellowish, but a beautifully white product was obtained 
by precipitation from a highly concentrated liquid by the addition of pure absolute 
alcohol, in which the formate is much less soluble than in water. The crystalline 
precipitate was recrystallised from alcohol. Notwithstanding, however, the excessive 
care with which sodium formate was prepared, it was found practically impossible to 
obtain it in a state of sufficient purity to make it available for trustworthy obser¬ 
vation. Crystals of snow-like whiteness yielded yellowish solutions soon becoming 
yellow ; various crops of such crystals gave spectra of different lengths ; sometimes 
the portion giving the longest spectrum, after the most careful recrystallisations would 
transmit only half the rays, and by no artifice could a fraction be made to transmit 
the original spectrum of the salt. A very careful examination showed us that formate 
of soda is changed with extreme ease into oxalate, and there is no doubt that from 
this fact arises the unsuccessful issue of our experiments on this salt. The solution 
of the four sodium salts contained the following weights of the substances :—Formate, 
MDCCCLXX1X. 
2 M 
