EMBBYO, SEED AND CARPEL IN THE DATE. 155 



be extracted from the endosperm, and he thought that it was 



present "in the softened layer of endosperm against the 

 cotyledon" but it is clear from the context that he did not 



enzyme to originate in the endosperm but in the 



The point however being moot, Pond (1906) 



aclusivelv that cvtase does not originate in the 



cotyledon. 



m 



Sachs' reasoning is thus justified. It is a well-known 



fact 



haustorium 



while this does not vitiate the reasoning it calls for explana- 

 tion. Figure 25 shows that the digestion begins at the end 

 of the cotyledon, where the endosperm cells are isodiametric. 

 It is clear that there is a localization of enzyme secretion in 

 this region of the cotyledon. The movement of the fer- 

 ments into the endosperm from this on must be due either 

 to the anatomical conditions in the endosperm, or to a con- 



* 



tinuation of the localization of more active ferment secre- 

 tion. If one follows the movement of the haustorium, one 



< 



it it is not related to the shape or 3 

 endosperm (text f. B). Its more 



be 



amount 



regions where the cells are approximately isodiametric. 

 There is no evidence, moreover, that there is any more 



;e-way through the endosperm 

 more pores. We must conclu 



determined 



by the localization of greater activity of ferment secretion 

 in the haustorium. We are urged to the conclusion, fur- 

 ther, that the greater secretion of ferments is at first at the 

 end of the cotyledon, but that it becomes relatively less 

 active, the scene of greater activity passing over to the sides 

 of the haustorium. In this connection, the finding of a seed 

 in which the embryo had suffered abnormal displacement, 

 so that it lay in a position almost antipodal to the normal 

 one. is of interest. It was discovered only after it had been 



for some time ( f . B ) . 



