390 Prof. J. R. Green. On the 



resting endosperm, when extracted with salt solution or with glyce- 

 rine, yields up to the solvent something which, though possessing no 

 ferment power, yet is capable of having this developed in it by 

 warming with dilute acid. The cotyledons, on the other hand, are 

 found, even when germination is active, to contain no ferment. 



Yet it seems improbable that the function of the young embryo hi 

 its relation to the endosperm is one of absorption only. Van 

 Tieghem found that the young endosperm, apart from the embryo, 

 increased in bulk with extreme slowness, not doubling its size till a 

 month had elapsed. In normal germination this result is attained in 

 five or six days. This is not due simply to slowing of the process, 

 owing to elaborated materials not being absorbed. In some of my 

 experiments I laid the flat surfaces of the isolated endosperms upon 

 dialysing paper exposed upon moist sand, so that absorption could 

 take place, but even then the rate of development was scarcely 

 accelerated. 



On the other hand, the germination was much more rapid when 

 a small piece of the cotyledon was left adhering to the endosperms, 

 though no removal of products could, under the conditions, take 

 place. The first set of experiments alluded to (p. 375), when seeds 

 with all the embryo taken away were germinated side by side with 

 those that had only lost its axis, shows, too, that the mere presence 

 of the cotyledons, apart from their absorbing power, had a con- 

 siderable influence on the progress of the germination. 



It" is difficult to suggest an adequate explanation of this action. 

 As already shown, it is not caused by the formation and subsequent ex- 

 cretion of the glyceride ferment. Nor does it appear that the embryo 

 excretes anything that may bring about the later changes which the 

 ferment does not effect. An embryo extracted from a germinating 

 seed, and washed from adherent matter, was found powerless to 

 effect any change in free ricinoleic acid when the two were placed 

 together in an incubator. Nor would a watery extract of several 

 embryos taken from germinating seeds produce any change in an 

 emulsion of ricinoleic acid in the direction of forming the crystalline 

 dialysable acid found to result from the oxidation of the former. It 

 seems probable that its growth or development acts as a stimulus 

 to the protoplasm of the endosperm, cells in which it is embedded, 

 whereby these are caused to undergo their metabolic changes more 

 rapidly than they do in the absence of such stimulus. The processes 

 which are most affected seem to be those dependent on a supply of 

 free oxygen, and not those of the ferment action, a point which 

 supports the idea of the former ones being the expression of the 

 vital activity of the protoplasm of the endosperm cells. Such a 

 stimulus is probably of a physiological character, and is not the mere 

 increase of pressure on the endosperm as the embryo grows. 



