pwned 
142 BOTANICAL GAZETTE [SEPTEMBER 
so as to have comparable analytical results. The hard coats were 
separated and discarded, just as in the collection of the seed material. 
ANALYTICAL METHODS AND RESULTS 
The material for analysis was prepared according to LOWEN- 
STEIN (14) and MILter (15), except for slight modifications. The 
collected seeds and seedlings were thoroughly ground with 95 per 
cent alcohol in a mortar; then the material was transferred to 
evaporating dishes and the alcohol evaporated. After thus treat- 
ing the material three times with 95 per cent and twice with absolute 
alcohol,*it was dried in a vacuum at 75°C. for one hour. The 
material was then powdered and placed in the desiccator until 
analyzed. When analyzed the material was in perfect condition, 
and showed no signs of oxidation. 
The method followed in the analysis was outlined by Kocu (11). 
As it was necessary to make both fat and protein analysis on the 
same sample, the acid precipitation of the lipoid fraction, as earlier 
described by Kocu, could not be used because of possible protein 
hydrolysis. The lipoids, therefore, were extracted by an 18-hour 
continuous extraction with hot absolute anhydrous ether. Cal- 
cium chloride tubes were used to protect the material from moisture 
during the extraction. Lipoid or ether soluble material is 
referred to as F;. The whole of the lipoids were not dried to get 
the true weight, because of the danger of oxidizing the unsaturated 
compounds. This lipoid weight was derived by subtracting the 
weight of the dry lipoid-free material from the original dry weight. 
Such a change made it possible to analyze the lipoids at once and 
avoid oxidation (table II); then the lipoid-free material was 
extracted with hot 50 per cent alcohol for 12 hours. This 50 per 
cent alcohol soluble material is indicated as F,, or extractives. 
These extractives were dried to constant weight in vacuum, dis- 
solved in hot water, and portions taken for the analysis. The 50 
per cent alcohol insoluble material (or F;) was dried in vacuum, 
weighed, powdered, and portions taken for analysis (table IV). 
Table I gives some general data. The amount of water and 
solid material found in the air-dry seeds, after-ripened seeds, and 
seedlings at the time the material was prepared for analysis, is 
