284 
Notes . 
motion of the protoplasm has been arrested, either by limitation of 
oxygen or by chloroform-treatment. 
In the case of a cell the vitality of which is arrested or destroyed, 
the enzyme, if it is to reach the starch at all, must do so by a process 
of diffusion through the protoplasm, and more or less of the chloro- 
plast itself. Now it seemed quite possible, in lack of evidence to the 
contrary, that the negative results we had so far obtained with killed 
leaves might, after all, be due to the inability of the highly colloidal 
diastase to gain ready access to the starch by diffusion. 
As no determination of the diffusion-rate of diastase had been made, 
it was necessary to do this in order to see how much value has to be 
placed on this objection. 
This was done in the following manner : — 
A 3 per cent, gelatine-solution, when just on the point of setting, 
was mixed with a little solid buckwheat-starch which remained sus- 
pended in the solidified gelatine. On the top of this mixture, which 
was contained in a small beaker or other suitable vessel, there was 
then run a 3 per cent, solution of gelatine containing diastase. When 
this had perfectly set, we had the starch-gelatine below and the 
diastase-gelatine above with a perfectly sharp line of separation be- 
tween the two. After a suitable time the gelatine-mixture was made 
as hard as possible by immersion in a freezing mixture and was then 
divided into vertical slices for examination with the microscope. The 
extent of the diffusion of the diastase from the upper into the lower 
layer was indicated by the distance through which action on the starch- 
granules could be detected. 
In the case of the diastase of malt-extract the diffusion of the enzyme 
went on for several days at a very uniform rate of 0-145 mm - P er hour, 
whilst precipitated and re-dissolved diastase diffused at only about one 
half of this rate, viz. 0-061 mm. per hour. 
Now if we consider the size of a palisade-cell of the leaf of 
a Tropaeolum , which is about o-i mm. long, and 0-025 mm. in width, 
we see very clearly that, if the rate of diffusion of diastase through dead 
protoplasm is anything like it is in gelatine-jelly, the non-disappearance 
of starch in a chloroformed leaf cannot be due to any inability of the 
diastase to reach the starch by the ordinary process of diffusion. 
Taking all the evidence which we have been able to collect into 
consideration we are compelled to admit that the first stage of dis- 
solution of the starch-granule in the leaf is in some way or other 
