41'4 University of California Publications in Botany [Vol. 7 



Ectocarpus confervoides f. parvus (Saunders) comb. nov. 



Fronds <j:rc^Mri()ii.s. r<»i-niin<^ tufted or extended masses. 1-2 em. 

 liiixh. more or le.ss alternately branched; erect filaments 21-30/x broad, 

 .slifxhtly constricted at the joints, much attenuated at the tips or end- 

 injr in hairs; cells 1-3 times as long as broad below, shorter above; 

 chroiiiatophoi-es irrefruhii'ly band-shaped; zoosporanfiia ovoid or ellip- 

 .soid, .35-5;V long, 20-27/x broad, usually sessile, sometimes on the same 

 filament with the gametangia; gametangia narrowly lanceolate-conical, 

 gradually long-attenuated above, 120-400ju, long, 20-27/a broad, borne 

 on a longer or shorter pedicel, occasionally seriate and intercalary 

 (a,s in rylaidla). 



On sand-covered rocks (type!), logs, or on the larger Melano- 

 phyceae. Central California (near San Francisco) and southern 

 California (San Pedro, the type locality, and San Diego). 



Ectocarpus siliculosus parvus Saunders, Phyc. Mem., 1898. p. 153, 

 pi. 22. figs. 1-9. 



The f. parvus, as defined above, presents certain resemblances to 

 both Ectocarpus siliculosus and E. confervoides. Its slender elongated 

 gametangia resemble those of E. siliculosus and much might be said 

 in favor of the opinion of Saunders in making it a "variety" of that 

 species. On the other hand, very few, or none, of the typical game- 

 tangia end in a hair and we have therefore preferred to place it rather 

 with E. confervoides. There are often found in this form, as well as 

 in forms of E. siliculosis, very curious, elongated, intercalary game- 

 tangia which seem to be seriate in that \\\ey open laterally in sections, 

 as do those of Pylaiclla. Such gametangia are very puzzling, but may 

 possibly be regarded as indicating hybridization between Pylaiella and 

 Ectocarpus or a mutation of some sort. Speculation, however, can 

 scarcely explain such anomalies. The majority of the gametangia are 

 typically of Ectocarpus and while ajiproaching in shape those of E. 

 silicidosus, are less slender and less elongated than is typical for that 

 species and are not ordinarily prolonged into terminal hairs. 



So far as we may judge from the scanty specimen in our copy. no. 

 358 of Tilden's American Algae (from Pacific Grove) belongs rather 

 under f. parvus than f. variahilis, although the determination as f. 

 variahili.s is attributed to Saunders. 



