ORIGIN OF ENDOGENS FROM EXOGKNS. 487 



some respects, but it was of a higher type, possessing a stem or 

 distinct axis bearing leaves of a very delicate nature and pre- 

 senting a venation like that of dicotyledons. Another plant, 

 named Protorrhipis Ckoffbti, had a limb with deep indentations, 

 and a palmately-nerved venation with many irregularly- formed 

 meshes, recalling the structure of various organs of dicotyle- 

 donous plants. He concludes with the following observations : 

 " II me semble qu'il en resulte .... que les Angiospermes 

 auraient traverse un etat primitif de faiblesse et de subordina- 

 tion, en rapport avec le point de depart originaire de toute la 

 classe. Les Dicotylees de Cereal, encore faiblement differenciees, 

 moins eloignees des Monocotylees qu'elles ne le furent a la suite 

 de l'extension rapide qu'elles prirent un peu plus tard, affectant 

 une nervation dont les cotyledons, les bractees et les parties 

 stipulaires ou involucrales offrent encore des exemples, auraient 

 ete plus rapprochees des types herbaces de la classe que des 

 types ou ordres arborescents, plus elabores et plus arretes aussi 

 dans les traits decisifs de leurs organes appendiculaires." 



M. Saporta here inclines to the opinion that the simplicity of 

 structure observable in these fossil plants, of which there were 

 about tw T enty Angiosperms, indicates a more primitive condition; 

 but from the nature of the plants described, I think there can 

 be no doubt that they were partly if not all aquatic, and the 

 " feebly differentiated " structures to have been due to a con- 

 sequent degeneration, and are not necessarily indicative of a 

 higher ancestry at all. 



Survivals. — The belief, however, that endogens are of very 

 early origin — though I shall give reasons for thinking that they 

 were not antecedent to exogens — is supported by the fact that 



so many orders of this class include very few genera. Thus, 

 according to the ' Genera Plantarum ' of Bentham and Hooker, 

 there are thirteen orders out of one hundred and sixty-six which 

 have only four or a less number of genera, while two orders 

 have six. Taken together, therefore, these amount to nearly 

 8 per cent, of the whole class. It need hardly be added that 

 monotypic animals and plants, as well as those orders and genera 

 with but few members, are always regarded as survivals on the 

 principles of evolution and represent a lost ancestry. 



Percentages of Aquatic Orders.— SuppoM'ng the supposition to 

 be true that endogens were originally derived from exogens 

 through an aquatic habit of life, and that at least mauy of them 



