H. H. GRAN AND K. YENDO. 



M.-N. Kl. 



in the former. Taking this into consideration we identified our specimens 

 with the present species. Okamura's determination coincides with ours. 

 It should be noted, however, that the shape of the foramen is often 

 unreliable for specific distinction, and further study on a larger material 

 might perhaps show the transitions between them. 



We mention under this heading ScHRciDER's statement of having 

 found 6^ curvisetiim Cleve in Japan, although not without hesitation. 



Loc. Boshu Prov. (Okamura). Misaki; Otaru Bay (Yendo) Akashi; 

 Enoshima (Schröder). 



Chaetoceras pseudocurvisetam Maxgin. 



Mangix: 19 io, p. 350, figs. 3 II, 4 II. 



This peculiar species v^as collected by one of us at Penang, Malay 

 Peninsula, and was also found in the material from southern Japan. 

 Maxgin compared this species with C. currisetum Cleve, and gave the 

 specific name from its similarity. But the present species has remarkable 

 peculiarities among the Chaetoceras species. The frustules are characterized 

 b}^ having four protuberances on each valve, in contact with the corres- 

 ponding protuberances on the adjacent frustules. Chaetoceras hacteriasiro- 

 ides Karsten (1905 — 07, p. 390, Taf. XLIV, fig. 2) is the only species 

 known which has a somewhat similar kind of connecting warts. 



The lateral view of the frustules illustrated by Mangix seems by no 

 means to have made clear the real form of the cells. The characteristic 



processes on both sides of the insertions of the 

 setae are not shown by him. 



We found in one of the chains an aberrant 

 form of foramen. The optical section seen from 

 the broader surface of the frustule showed a 

 narrow, slit-like aperture, while in the surface- 

 view it was a circular opening, much smaller than 

 the others. There were no characteristic pro- 

 tuberances, and the frustules touched one another 

 by amply broad ends about the insertions of the 

 setæ,the setæ^being much shorterand more slender 

 than those belonging to the normal frustules. We 

 regard this as a stage of development shortly 

 after the cell-division. 



Loc. Misumi (Yendo). Penang, Malay Pen- 

 insula (Yendo), 



Fig, 12. Clinetoceraspsen- 



dociirvisetimi Maxgi.v. 

 a. Portion of a chain 

 seen from the broader 



surface, X 360. 



h. The same chain in a 



side view. X 360. 



