70 ERYTHEA. 



The material was collected near Pasadena by Professor A. 

 J. McClatchie, who also sent some of the same material to 

 the writer. Miss Tilden has also very kindly communicated 

 some of the type specimens. 



Miss Tilden says, in her article, that O. trapezoidea is 

 larger than O. chalyhea, and gives the diameter of O. trape- 

 zoidea as ranging from 12/i to 15yM. ■ Gomont (Monogr. des 

 Oscill., p. 253) gives the diameter of O. chalybea as from 8/i 

 to 13/f. The material in the writer's possession, shows 

 filaments almost uniformly 12/^ in diameter. Certainly an 

 increase in diameter of 2/^ might readily be allowed in a 

 species ranging from 8/x to IS pi. 



O. trapezoidea is said, by Miss Tilden, to show no spiral 

 arrangement. Gomont says of O. chalybea; — " recta, vel in- 

 terdum in spiram laxam contorta." 



Finally O. trapezoidea "is clearly distinguished by the 

 shape of the terminating cells of the filament." The end cells 

 in the filaments of the writer's material agree well with Miss 

 Tilden's figure^ and both agree extremely well with the end- 

 cell figured by Gomont for O. chalybea (loc. cit., PL 7, fig. 

 19). Furthermore the shape of the terminal cell has been 

 proven by Gomont to be a very important character in dis- 

 tinguishing the species of this genus. 



But Miss Tilden evidently refers, not to the terminal cell 

 itself, but to "the mature cells near the extremity of the fila- 

 ment." These cells are wider at one end than at the other 

 giving them a trapezoidal shape as seen in optical-section. 

 This character however is neither constant nor particularly 

 distinct in the writer's specimens. In some filaments a num- 

 ber of such cells may be seen, but this shape generally 

 disappears upon the application of weak swelling reagents 

 such as very dilute acetic acid. 



The same appearance is seen in most specimens of Oscilla- 

 toria, which have either been dried or preserved in fluid^ 

 (aqueous or alcoholic) and the writer has been in the habit 

 of considering this appearance due to the greater shrinking 

 of the newer cell- walls, (intercalated between each two 



