IN THE SEED WHICH ACCOMPANY GERMINATION. 
53 
A 10 per cent. NaCl extract of the seeds was found to give similar reactions, but to 
contain more of the body precipitable by dialysis o'r dilution of its solution. There 
was also extracted by this fluid a greater quantity of the globulin, as shown by the 
greater amount of coagulum produced by boiling. 
As already mentioned, very careful and long-continued diatysis showed that neither 
of the proteids of the seed was capable of dialysis. The dialysate of the boiled watery 
extract, on concentration and slow evaporation on a glass slide, showed only here and 
there a needle-shaped crystal. The seed hence contains, before germination begins, two 
albumoses (protalbumose and heteroalbumose) and a globulin, but no peptone, nor 
any crystalline derivative of proteids. 
Before discussing the behaviour of the ferment when mixed with these bodies for 
purposes of experiment, it may be well to mention certain changes which took place in a 
watery extract of the germinating seeds, prepared at the same time as the glycerine 
one used in the experiments already detailed. This watery extract when first made 
was turbid, and would not filter clear. It contained albumoses, as evinced by a 
precipitate given on addition of HN0 3 or acetic acid. After standing some days at 
the temperature of the laboratory, the turbidity had partially disappeared, and addition 
of acid failed to produce a precipitate. The ferment extracted by the water from the 
germinating seed had evidently performed a process of digestion of the proteids 
extracted coincidently with itself. A quantity of it was then submitted to dialysis for 
some days, and the dialysate gave a slight biuret reaction, much masked by a colour¬ 
ing matter that diffused out; when concentrated and evaporated on a slide, it slowly 
deposited crystals resembling those of asparagin. 
The composition of the stored proteids having been ascertained, a quantity of the 
mixed albumoses was prepared from the salt extract of the resting seeds. This was 
chosen, as it seemed to contain the two albumoses more in the proportion in which 
they existed in the seed than did the watery one. The heteroalbumose is insoluble in 
water, and so only so much of it dissolved in the extract as was enabled to do so by 
the small amount of salts in the seed. This salt extract was boiled to separate the 
globulin, which coagulated and was filtered off. The liquid was then completely 
precipitated by acetic acid, and the precipitate separated by decantation and sub¬ 
sequent filtration, washed, and dried. It was then dissolved again and precipitated by 
alcohol. After standing under this for some time, it was dried and used for the 
purposes of the experiment, being mixed with a quantity of the ferment in new tested 
dialysers. The solution was made acid to an amount equal to ‘2 per cent. HC1, and 
outside the dialysers acid of the same strength was used. The relative quantities of 
the two albumoses used in different experiments varied considerably, but neither was 
absolutely isolated in any. In all the experiments the course of the digestion was the 
same, and a mixture of both albumoses with the globulin behaved in the same way. 
The first body formed during the digestion of the albumoses was an acid-albumin, 
exactly like the parapeptone described as occurring when the ferment was acting on 
