114 KJELLMAN, THE ALGyE OF THE ARCTIC SEA. 



or sickle-shaped, erect ramuli, which are thickest a little below the middle and at- 

 tenuated towards the base. They occur without any apparent order, sometimes very 

 numerous, sometimes very few, and are, as far as 1 have been able to see, a sort 

 of adventive branches. A good figure is to be found in J. G. Ag. Spetsb. Alg. Bidr., 

 pi. 2, fig. 2. In Rhodomela subfusca there are certainly branches, that remind one of these, 

 viz. those above indicated as elements 1 of the ramification, but these are subulate or 

 cylindrically subulate, issuing from broad bases and always developed in strictly acro- 

 petal order. I have never seen any formations resembling those of Rh. lycopodioides 

 on the considerable number of Rh. subfusca from Bohuslan, that I have examined. 

 These circumstances seem to me to imply, that Rh. li/cojjodioides and Rh. subfusca are 

 two distinct, although feebly differentiated species, which have possibly once sprung 

 from one type, but afterwards developed differently. 



Gobi's opinion that different species should be perceptibly unlike in anatomical 

 structure can scarcely be regarded as defensible. If such a condition should be car- 

 ried out in algology, a considerable number of species, constant, easily recognizable, 

 and regarded as good, must be suppressed and subsumed under others in long series. 

 External morphological diversity ought certainly even here to be considered valid as a 

 character of species. Small anatomical diversities are indeed to be found even between 

 Rh. lycopodioides and Rli. subfusca, but the anatomical structure of both species being 

 essentially different in different, older or younger, portions of the frond, and the diffe- 

 rent forms of what is undoubtedly Rh. lycopodioides being also somewhat different from 

 one another in this respect, it is necessary to examine a great many specimens of 

 different ages and places of growth, in order to be able to ascertain what is essential or 

 unessential. 1 have not had an opportunity to undertake such an examination, and I am 

 thus obliged to confine myself to stating that, with regard to structure, Rh. subfusca 

 and Rh. lycopodioides are very similar to each other, and differ essentially from Rh. virgata. 



Remark on the form. As appears from the list of synonyms, I have united in 

 one species all that Rhodomela which has been reported from the Arctic Sea. I 

 have thus regarded as variations of the same type a great many forms that, at 

 first sight and in their extremes, differ most considerably from the typical form. 

 As far as I can judge, there is ta present no other course left. The principal 

 forms, that I have tried to discern, do not stand isolated, but are combined with one 

 another by more or less numerous intermediate forms. The two forms which I have 

 called f. flagellaris and f. tenuissima (i glacialis are those most unlike the typical one. 

 To the former one of these I have not before been able to assign a place, but have 

 mentioned it as an alga incertce sedis in my description of the marine alg89 of Spitz- 

 bergen; cf. Kjellm. Spetsb. Thall. 1. p. 33. Having since that time become acquainted 

 with the form named above f. setacea, which, as is shoAvn by the figures given, stands 

 undoui)tedly near f. flagellaris, I do not hesitate to regard it as a peculiarly deve- 

 loped Rh. lycoijodioides. That f. setacea belongs to the series of forms of Rh. lycopodioides, 

 is shown b}'^ a comparison of figures 1 and 3 on plate 9. 



I formerly regarded Rh. lycopodioides f. tenuissima (i glacialis as a good species, 

 but, having found its subform prolifera, which merges in Rh. lycopioides f. typica (i 



