76 



ON THE CYSTOCABPS OF CATENELLA OPUNTIA. 



so numerous on the exterior of the cystocarp ramulus are physio- 

 logically and morphologically trichogynia, and that the carpo- 

 spores arise as the result of the transmission of a fertilizing 

 stimulus to the medullary weft from them. My reasons for 

 believing this to be so are : — (1) The obvious similarity of these 

 trichophoric systems to those of other FloridesB ; (2) the fact 

 that bodies identical with the pollinoids formed in the antheridia 



i 



can be seen attached to the trichogynie processes ; (3) that the 

 carpospores are formed only subsequently to the fusion of the 

 pollinoids with the trichogynia, and that the trichogynia wither 

 after the carpospores begin to develop ; (4) that I have seen no 

 evidence whatsoever of any other organ by which fertilization 

 might be effected; (5) and lastly, that there exists direct commu- 

 nication between the trichophoric cells and the cells from which 

 the carpospores arise. 



EXPLANATION OF THE PLATES. 



Plate XIII. 



Fig. 1. Catenella Opuntia, with terminal and lateral cystocarps. Nat. size. 



X350. 



Figs. 2, 3, 4. Stages in the development of a tetrasporangiuin 



Fig. 5. Vertical section of a young antheridium. x 350. One " nest " only is 



shown. 



1 



Fig. G. Fully developed "nest " of pollinoids in vertical section. xMO. 



Figs- 7, 8. Trichogynia and trichophoric cells arising from the subcortica 



cell-network, x 350. 



Fig. 7 shows a trichogyne which has not yet pierced the cuticle, 

 and which is provided with only one trichophoric cell. Fig. 8 shows 

 two trichophoric cells surrounded by a number of vegetative cells. 



Plate XIV. 



Fig. 9. Peripheral layer of a young cystocarp, with pollinoids attached to two 



trichogynia. x 500. 

 Fig. 10. Vertical longitudinal section of a young cystocarpic ramulus. X350. 

 Fig. 11. Frond with antheridium (b) and nearly mature cystocarp (a). x20. 

 Fig. 12. Development of carpospores from the medullary network. A few 



carposporic chains only are represented. X350. 



