99 



description (cfr. Postels and Ruprecht, III. Alg., p. n, 1840), it is 

 credited to Unalaska. Ruprecht says (cfr. Tange d. Och. Meeres, 

 p. 355, 1 85 1 ) that the specics is rare and had occurred to him la- 

 tely only from Fort Ross on the Californian coast. Harvey (cfr. Proc. 

 Linn. Soc, Botany, voi. 6, p. i65 5 1862) refers doubtfully to this 

 species, specimens collected at Esquimalt, B. C. No other writer has 

 attempted to place any specimens under it, but the common Maria 

 of the Californian coast has passed current in locai collections under 

 this name, on whose authority, I do not know. Kjellman mentions 

 the species in the discussion of his A. laticosta (cfr. Om Beringhaf- 

 vets Algflora, p. 40, 1889), but without committing himself as to 

 whether any one of the several broad ribbed species described by 

 him may be nearest to it unless it may be his A. laticosta, for he 

 quotes Harvey (loc. cit.) to the effect that his Esquimalt plants have 

 a costa of 1-1 V 2 inches broad. In the Harvey Collection in Trinity 

 College, Dublin, are four plants under A. marginata, three of which 

 are young Cymathaere triplicata (P. & R.) J. Ag. while the fourth is 

 Pleurophycus Gardneri Setchell and Gardner. Hence we can see how 

 the « midrib » may have been considered to be broad. In the Ru- 

 precht Collections in the Imperiai Academy of Sciences at St. Pe- 

 tersburg, I found two specimens under this name, one young and 

 incomplete, consequently uncertain, but the other of broad midrib, 

 biade with long tapering base, and few, not over long pinnae, ap- 

 parently the type specimen, for it was labelled Unalaska, though also 

 credited to Wosnessenski who is not mentioned as one of the col- 

 lectors on the Luetke Expedition. This specimen is clearly the plant 

 which Kjellman (cfr. Setchell and Gardner, loc. cit., p. 274) referred 

 to the neighborhood of his A. praelonga. It seems to me that there 

 is little doubt but that the plants Gardner and I have referred to A. 

 praelonga are really to be placed under A. marginata, as well as 

 those from California, including the A. curtipes Saunders (Minnesota 

 Botan. Studies; 2d. ser., p. 5ói, pi. 33, 1901). The Californian plants 

 vary very considerably in the breadth of the costa, even in speci- 

 mens growing in the same dump and there may be found in the 

 same place, dose to one another, ali transitions from a moderately 

 to a very wide costa. It seems as if A. marginata may replace both 

 the names of A. praelonga Kjellm. and A. laticosta Kjellm. The 



