56 



process of germination, the seedling assumes the characteristic 

 form of a bulb. The accompanying figures are drawn from 

 specimens grown at the Empress Gardens Poena. 



Remarks on the facts narrated above regarding seed structure 

 and germination in Trapa and Crinum. 



The germination of Trapa has such a close resemblance to 

 the same process in Crinum that, it may be regarded as a 

 case of convergence of a dicotyl to the monocotyl-type. A 

 process analogous to that in Crinum has been observed in a 

 number of other monocotyls, e. g. "sedges and rushes, irises, 

 snowdrops, narcissus, aloes, butcher's broom, bananas, palms" '), 

 which all at any rate begin life in marshy surroundings. Struc- 

 turally two seeds could not differ more than these two do. 

 The one has two cotyledons the other one only. The one is 

 exalbumiuous, the other has a copious quantity of albumen. 

 Again, the pericarps of the two differ. Trapa does not liberate 

 its seed ; in Crinum we get a capsule, which ruptures. Yet in 

 the process of germination all these differences are got over. 

 The testa in Crinum is as impervious to water as the endocarp 

 is in Trapa. The albumen in Crinum gives up its store of 

 nutriants to the embryo, just as well as the fleshy cotyledon 

 does in Trapa. Though Trapa has two cotyledons only one is 

 functional and the other is reduced to a vestige, of only mor- 

 phological value, without any functional importance. 



Regarding differences between the two there are two calling 

 for observation ; one of specific gravity and the other of geo- 

 tropism. Both are distinctly correlated with the differences 

 between the ultimate habitats of the two plants. 



i) The Natural History of Plants by Kerner and Oliver P. 1, 604. 



