60 WRIGHT : THE GENUS DIOSPYROS 



is laid down in the epicotyledonary leaves long before they 

 are exposed to light. 



As to the causes which lead to the detachment of the 

 cotyledons, it is very difficult to make any definite state- 

 ments. The regularity of the occurrence in seeds of all 

 sizes, under natural and artificial conditions, is suggestive of 

 the possibility of a definite cellular activity cutting off the 

 cotyledons at the node in a manner similar to that in which 

 old leaves are cut off from the parent tree. This, however, 

 cannot be the case, as in all experiments where the coty- 

 ledons have been set free prior to detachment their dead and 

 shrivelled tissues have persisted at the cotyledonary node 

 for a considerable length of time. 



Hence we are driven to the conclusion that the detachment 

 is probably a case of separation under conditions of strain, 

 and several possible factors may be quoted in support of 

 this contention. The first is the pressure of the endosperm. 

 In those species where the cotyledons are persistent the 

 endosperm may be equable (D. Ebenum) or strongly ruminate 

 (D. sylvatica) ; similarly it may be markedly ruminate in 

 species with detached cotyledons (D. hirsute^ or equable 

 (D. msignis). Hence, the pressure of the endosperm, as 

 indicated by the absence or presence of ruminations of this 

 substance, has probably very little to do with the detach- 

 ment of the cotyledons. A second factor is probably to be 

 obtained in the bulk of the endosperm, testa, &c. ; for 

 though D. ovalifolia may prove exceptional, it is a fact that 

 those species with persistent cotyledons have relatively 

 a smaller quantity of endosperm which is necessarily 

 exhausted early, resulting in a shrinking of the testa and 

 liberation of the cotyledons at an early date. 



However, neither of the foregoing, even if coupled with 

 the increased resistance of the irregular surface of the 

 mature cotyledons, can completely explain the cause of 

 detachment. 



The chief cause, in my opinion, is to be found in the 

 insufficient splitting of the testa ; in D. pruriens and D. 



