Stoward. — On Endospermic Respiration in Certain Seeds. 435 
an almost complete envelope to the starch-bearing cells is the aleurone 
layer, consisting of a three-celled layer in Hordeum , and a one-celled layer 
in Zea. The living nature of the cells of this layer, with their rich protein 
and protoplasmic contents and apparently active nuclei, has been based 
hitherto mainly on cytological evidence. 
If, as already indicated in the preceding section, our comparisons be 
made on the output of C 0 2 per gram of moist material per hour, then the 
relative magnitude of C 0 2 production yielded by the seed and its parts 
assumes the following order : — 
1, isolated embryos; 2, intact seeds; 3, endosperms; 4, pure endo- 
sperms. 
If average values are taken for the purpose of instituting these com- 
parisons (vide Table, ‘ Comparative Average Output of C 0 2 ’), then, confining 
ourselves to the summer experiments with Hordeum , the following state- 
ment may be advanced : — 
1 gram of embryonic tissue respires 
1-2315 
0*1689 
1-2315 
0*07210 
1-2315 
0*02012 
I- 23 I 5 
0-0542I 
= 7*3 times more actively than 1 gram of intact seed. 
= 17*0 times more actively than 1 gram of endosperm. 
= 6 1 *2 times more actively than 1 gram of pure endosperm. 
= 22*71 times more actively than 1 gram of pure endosperm 
(re-established experiments). 
Similarly, from the average values given for Zea , the following data 
are derived : — 
1 gram of embryonic tissue respires 
Q - 444 I 
0*03346 
0*4441 
0*006948 
0*4441 
=13*27 times more actively than 1 gram of intact seed. 
= 63*9 times more actively than 1 gram of endosperm (low 
temperature experiments). 
- =31.3 times more actively than 1 gram of endosperm (high 
001 42 1 temperature experiments). 
0*4441 
— = 92*2 times more actively than 1 gram of pure endosperm 
0*004819 IA ' A C S A\ 
(derived from resting seeds). 
0*4441 
0*002661 
= 1 66*8 times more actively than 1 gram of pure endosperm 
(derived from germinated seeds). 
From the average values given in the table referred to, still other 
comparisons involving the relative outputs of C 0 2 by endosperms, pure 
endosperms, and intact seeds may be made. In all such comparative 
