BOTANICAL SOCIETY OF EDINBURGH. 565 



branched Japanese species which is also apparently very 

 widely distributed in those seas. The specimens of this 

 species bearing sporangia, which I was able to examine, 

 never showed any indication of paraspore fruits. 



The same is true also of G. dura (Eupr.), J. Ag., a 

 species which has hitherto been known only from the 

 coasts of Kamtschatka ^ and the neighbouring Behring 

 Island."' In this species, ^ which closely resembles G. 

 capillaris in the anatomical structure of the thallus, I 

 found procarps, as in the last-named species, embedded in 

 the inner part of the wall of the thallus and enclosed in 

 its mucilage. The shape and mode of development of 

 these procarps was the same as in G. capillaris and 

 G. furmta. The mature cystocarps formed small pro- 

 jections on the surface of the fertile shoot, the gonimoblast 

 of each cystocarp was moreover embedded in the loose 

 inner part of the wall of the thallus, which was locally 

 very much thickened. I have not seen paraspore fruits of 

 this species, but as I have only been able to examine very 

 little correctly named material, I cannot say whether the 

 formation of paraspore fruits is quite absent ^ or only 

 occurs occasionally. 



^ Ruprecht. Algae Ochotenses, p. 118. 



- Kjellman Om Beringhafvets Alg. Flora, p. 28. 



^ I examined authentic fertile material of this species (^Dumontia 

 dura Rupr. cum tetrasp. Herb. Acad. Petrop — Kamtschatka ad portum 

 St. Petri et Pauli) — from the Hamburg Herbarium. 



■* I have not attempted to determine definitely whether this G. 

 dura (Rupr.), J. Ag.. from Kamtschatka. represents an independent 

 species or not. The habit of the specimens, which I examined, recalled 

 the simpler forms of G. capillaris (Suringar Illustr.. pi. ii. fig. 1-13). 

 The form and arrangement of the procarps and cystocarps appeared 

 also to correspond closely with that in G. cajyillaris. The only con- 

 stant difference lay in the anatomical structure of the thallus. and only 

 to this extent that in G. dura the cavity of the inner space of the 

 shoot was somewhat wider than in G. capillaris. The wall-forming 

 filaments, together with their branches, do not all enter into the composi- 

 tion of the thallus wall from their very bases. The lower parts of the 

 basal branches of these filaments rather run through a tract continuous 

 with the internal cavity, and only subsequently become united by 

 mucilage to form the wall of the shoot. Moreover, on the inner limit 

 of the wall, separate short rhizoids directed towards the iuterior of the 

 thalhis arise, and here and there reach the inner space of the shoot. I 

 must leave it to others to determine whether a special significance for 

 the recognition of species is to be attached to these small anatomical 

 differences, especially as Suringar (lUustr., p. 11-12) has already called 

 attention to the occurrence of similar variations of anatomical structure 

 in G. capillaris. My experience with G. dura has hitherto been too 



