IN THE SEED WHICH ACCOMPANY GERMINATION. 
43 
found to answer well with Vetch seeds, but did not, for some reason, extract anything 
from the Lupin. Instead of following it, I only made a glycerine extract of the ground 
germinating seeds, which I used after dialysing for some time. In consequence of 
this mode of extraction it was necessary to modify the ordinary method of testing its 
activity. This glycerine extract would not only contain any ferment present, but 
would have taken up such of the proteids of the seed as were soluble in water. 
According to Vines, in the Lupin there occur hemialbumose and a form of globulin. 
The latter is soluble only in salt solutions, but the former dissolves in water readily. 
As this hemialbumose also gives the biuret reaction, which is always relied on as the 
characteristic reaction of peptone, it is evident that its presence in the extract con¬ 
taining the ferment would render it very difficult to say that the latter formed peptone 
at the expense of the fibrin, unless some method could be devised which should 
separate the albumose from any peptone that might be formed, or a reaction discovered 
which peptone gives and albumose does not. Such a method of separation, neglected 
apparently by both v. Gorup-Besanez and by Krauch, is furnished by dialysis. 
According to Vines, hemialbumose does not dialyse, while peptone does so readily. It 
is evident, therefore, that if a mixture of the two are at any time present in the same 
dialyser the peptone will pass through and be found in the dialysate, while the 
hemialbumose will not. Before trusting to this method, I carefully tested this asserted 
indiffusibility, and found I could confirm Vines completely. After more than a week’s 
exposure of a solution of hemialbumose in a parchment-paper dialyser, the liquid 
outside failed to give a biuret reaction, while the solution inside did so readily. 
Instead of using glass vessels, therefore, for my digestive experiments, I carried them 
all out in well tested dialysing tubes, through the walls of which would pass, as 
formed, the peptones and nitrogenous crystalline bodies, should such be produced, 
while any hemialbumose or globulin present in the extract would be retained in the 
vessel. In all cases careful control experiments were carried out, all the conditions 
being the same in both sets, except that one set contained the ferment extract and 
the other did not. The dialysers, too, were carefully tested as to their intactness at 
frequent intervals. The fluid outside was made of the same reaction as that inside. 
The experiments on which 1 base my conclusions were carried out in nearly every 
case with extracts which had themselves been dialysed with care. The reason for this 
was that, as germination had begun in the seeds, it was probable that in the extracts 
there would be a small amount of leucin or asparagin. The dialysis was continued till 
the dialysate gave no proteid reaction, and on concentration and evaporation on a glass 
slide deposited no crystals. In one or two cases extracts were used wdthout such 
dialysis, but when this was the case precautions were taken against error, which will 
be detailed in giving the results. 
In selecting the medium in which to conduct the experiment on fibrin, attention 
and is not coagulated by exposure to alcobol, remaining under it unchanged for a considerable period. 
This gives the biuret reaction, and explains, therefore, Krauch’s criticism. 
