120 BOTANICAL MAOAZ/NE. [vol. xix, 



circumstances of their habitats. The most reliable character is, no doubt, 

 as the fundamental principle of the leading classification of the other algfe 

 depends upon, the mode of propagation. Next comes the internal structure 

 of the frond, and the external form the last. I shall try to remark on 

 these subject in connection with the fronds of the Corallinae. 



1. Propagating Organs. The process of fertilization in the family 

 Corallinacem has been carefully studied by Kosanoflf, 13 Bornet and Thuret, 2 ' 

 Solms-Laubach/" Heydrich," etc. We cannot expect a new fact to add 

 to the results of their investigations. According to their descriptions the 

 process of fertilization in both the Melobesias and the Corallina? are com- 

 mon in the essential characters. The morphological characters of pollinoids, 

 however, differ in both large groups. 



The pollinoids of the Melobesias have two minute globular appendages 

 at the opposite poles of the main spherical body/ In the Corallime they 

 have considerably long filamentous appendages at the posterior ends, ap- 

 parently assuming the shape of some animal spermatozoa. This has already 

 been described and illustrated by Bornet and Thuret/ who have observed 

 them in Jania rubens Lamx. and Corallina officinalis var. mediterranea 

 Hauek ; and by Solins-^Laubach' 5 in Corallina mediterranea. I was able 

 also actually to examine the autheridial conceptacles full of matured pol- 

 linoids of this sort in Cheilosporum McMillani.® But the material was 

 not sufficient for cytological purpose. About the nature of iunction of the 

 tail-like appendages there yet remain some points to be solved in future. 



As far as my present knowledge extends the observations on the 

 antheridial conceptacles are limited to Jania, Corallina and Cheilosporum. 

 The pollinoids in these start out from the whole inner wall of the concep- 

 tacles which are lined by the medullary cells. The relative position of the 

 generating cells differs widely from those of tetraspores and the carpo- 

 spores. Yet they are all equally of medullary origin. We have no know- 

 ledge concerning the pollinoids of Amphiroa, Lithothrix, etc. A study of 

 the antheridial conceptacles in the representatives of these genera may 



]) Sur los Melobesias 



1') Ktuile Phycologique. 



:>) ( 'orallina. 



4) Litliothamnion von Helgoland. 



5i Sur les Melobesise. PL 11. lig. 15. 



(\) 1. c. p. 9">. 



7) 1. c. p. 158. 



s) Cor. verse of Port Renfrew. PL LVI. rte. 14. 



