504 



JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



case with them all — e.g. Dancea dmplicifolia, Rudge, in which 

 the leaves are simple, and bear on their lower surfaces the 

 synangia, or sori, composed of coalescent sporangia, which 

 dehisce by terminal pores ; and I do not feel that the difficulties 

 are greater in regarding this as a relatively primitive form than 

 if we place one of the simpler Hymenophyllacece in that position. 

 The view that the Eusporangiate Ferns, including the Ophio- 

 glossacece, may represent a relatively ancient form as compared 

 with the Lcptosporangiatce is thus capable of defence — I am even 

 disposed to think the weight of evidence is in its favour. At the 

 same time, however, it is to be constantly remembered how great 

 the gulf is between the FilicinecB and the Bryopliyta, so great 

 indeed as to make it difficult to conceive by what steps the 

 transition came about, and to leave it a matter for speculation, 

 rather than of demonstration, which are the nearest points of con- 

 tact of die two series. But, as above stated, the choice now seems 

 to be between an affinity of the Musci and Hymenophyllacece , 

 or of the Eusporangiates and the Liverworts, and, as at 

 present advised, I am disposed to think the latter the more 

 probable. 



The effect of this, if established, w T ill be in some measure to 

 modify the general view of the relations of the main series of 

 plant-forms. The Bryopliyta and Filicinece might be viewed as 

 two series which show a certain parallelism of adaptation to a 

 moist habit. In each case the series would start with a relatively 

 bulky character of all the parts, and especially with the sexual 

 organs more or less completely immersed ; the progression would 

 be in either case towards forms with filmy, foliar characters, a 

 filamentous stage of the gametophyte, and with the sexual 

 organs not deeply immersed. Both these series would be so-called 

 " blind " branches of development, leading on to no higher forms. 

 As regards the geological evidence, the more delicate representa- 

 tives of both phyla probably represent forms of relatively recent 

 origin, and the absence of Mosses from the earlier strata is to be 

 noted in this connection. 



Secondly, it seems very probable that the Marattiacece are 

 related to the Cycads ; accordingly, if the former be viewed as 

 more primitive Ferns, the developmental series leading to the latter 

 is proportionally telescoped, while Isoetes, which has of late been 

 a subject of frequent discussion, would probably find a place as 



