1920] Setchell-Gnrdner : Chlorophyceae 195 



posterior extremities. It seems possible that there may be some error 

 of interpretation about this and Collins (1918, p. 86) suggests, with 

 good reason, as it seems to us, that the supposed gametes may be 

 microzoospores and that "figure 20, 'gametes fusing,' would seem 

 rather to represent two imperfectly separated microzoospores." The 

 matter of microzoospores and gametes especially needs further inves- 

 tigation in this as well as in other species of Hornviscia. 



The holdfast of H. tetracilmta is, as represented by Frye and 

 Zeller {loc. cit., p. 9, pi. 2, f. 3), composed of intramatrical rhizoids 

 arising from a considerable number of cells and simple or branched, 

 descending within the outer wall, or sheath, of the filament, but 

 emerging and spreading to some extent at the very base and becom- 

 ing trumpet-shaped. This combination of intra- and extramatrical 

 rhizoids seems like a combination of the characters of the species of 

 the Pemcilliformes section and those of the Wormskiold<ii formes sec- 

 tion. 



4. Hormiscia grandis (Kylin) S. and G. 

 Plate 9, fig. 3 



Filaments attached by intramatrical rhizoids from 8-14 segments 

 above the base, flaccid, 8-10 cm. long; segments cylindrical, con- 

 stricted at the joints, 45-70ju thick at the base, usually 2 times as long 

 as broad, 125-175/x, up to 200jti thick at the upper end, 1-3.5 times as 

 long as broad, walls 12-18/^, thick, lamellate, fertile segments slightly 

 swollen ; chromatophore a thin, parietal, reticulate band, with numer- 

 ous small pyrenoids. 



Growing on rocks in the upper littoral belt. West shore of Amak- 

 nak Island, Bay of Unalaska, Alaska, to Puget Sound, Washington. 



Setchell and Gardner, in Gardner, New Pac. Coast Mar. Alg. IV, 

 1919, p. 494. Urospora penicilliformis Setchell and Gardner, Alg. 

 N.W. Amer., 1903, p. 220 (in part). Urospora Wormskioldii Setchell 

 and Gardner, Alg. N.W. Amer., 1903, p. 221 (in part). Urospora 

 grandis Kylin, Studien ueber Algenflora, etc., 1907, p. 18, f. 3. 



In reexamining the specimens previously referred by us (1903, 

 p. 220) to Urospora penicilliformis and U. Wormskioldii, we found 

 several which differed in having the fertile segments long and not 

 swollen. They also have thick, lamellate walls. These characters 

 agree so well with those given for Urospora grandis that we feel 

 reasonably safe in referring them to that species and transferring 



