62 GIBBS. 
The methyl benzoate was dried over fused calcium chloride and purified by 
distillation. About 2 kilograms of a constant boiling product were obtained. 
This was hydrolyzed in 500-gram lots with fused potassium hydroxide 
and the methyl alcohol distilled. The distillate was fractioned until a 
portion of fairly constant boiling point was obtained. 
The final drying was accomplished by means of metallic calcium.“ The 
methyl alcohol and calcium were heated on a water bath with reflux con- 
denser for a number of hours and distilled. This operation was repeated 
with fresh calcium four times. 
Second method—Pure methyl oxalate was prepared from the same 
alcohol, Kahlbaum’s, and pure, dried oxalic acid. The purified crystals 
were hydrolyzed, and the methyl alcohol obtained and dried by the same 
procedure as that employed in the first method. 
Third method—One liter of methyl alcohol (Kahlbaum) was treated 
with 10 grams of calcium and allowed to stand about ten days. It was 
then distilled with a fractioning column and the distillate collected in 5 
fractions of about 150 cubic centimeters each. The boiling point was 
constant throughout the distillation. The residue in the distilling flask 
was discarded. Each fraction was tested for formaldehyde by Remini’s, 
Leach’s, and Hehner’s methods™ with negative results. The middle frac- 
tions only were used. 
Fourth method.—A portion of the product from the third method was 
further purified by treating with m-phenylenediamine hydrochloride ~™ 
(3 grams per liter). After standing five days at 30° it was fractioned 
with a glass fractioning column. Only the middle fractions were em- 
ployed. 
No traces of formaldehyde could be discovered in any of the 
samples of alcohol employed in the following investigations. 
EXPOSURE TO SUNLIGHT. 
With one exception, all of the samples exposed to sunlight 
were sealed in tubes of very thin glass, the absorption spectrum 
of which showed that all of the short wave lengths found in 
sunlight, and even considerably shorter waves, were transmitted. 
The preliminary experiments, performed more than two years 
prior to the bulk of this experimental work, showed that formal- 
dehyde was produced in samples of the purest methyl alcohol 
prepared by Kahlbaum, when exposed to the sunlight in the 
presence of air. In the following Table II, the character of the 
mixtures and the length of time during which they were exposed 
are tabulated. 
~ Gyr, Ber. d. deutschen chem. Ges. (1908), 41, 4322. 
* Bull. U. S. Dept. Ag., Bu. Chem. (1908), 107, (revised) 185. 
* Thid., 96. 
