MARINE DIATOMS OF THE PHILIPPINE ISLANDS 161 



those on either side of it; no markings except a very obscure rugos- 

 ity spread over the entire surface of the frustule, exclusive of the 

 arms; the valve view discloses no fact of structure except that the 

 frustules are strictly cylindrical. 



This remarkable genus bears only a slight resemblance to other 

 chain-forming genera, like SJceletonema, Corethron, Stephanopyxis, and 

 Greville's grotesque genus, Thaumatonema. The arms are so strong 

 and their ends so firmly welded with those of the next frustule that 

 any break in the chain invariably occurs across the cylinder, rather 

 than in the arms. 



TRIBRACHIA PELLUCIDA, new species 



Plate 37, fig. 2 



General characters those of the genus. The perfectly hyaline arms 

 and the merely slight rugosity of the cylinders give to this diatom a 

 remarkably crystalline clearness like finely molded glass. It is infre- 

 quent, even in the few dredgings where it occurs; for although sev- 

 eral specimens were secured, they are rarely met with, those secured 

 probably representing all that were present, as their robust resistance 

 to breakage and their striking appearance would insure their being 

 found. 



Length of cylinder, 0.061-0.071; width, 0.015-0.025; length of cyl- 

 inder with arms, 0.153-0.166 mm. 



Type.— Cat. No. 43692, U.S.N.M. 



Genus TRIGONIUM Cleve 



TRIGONIUM ARCTICUM (Brightwell) Cleve 



This is Cleve's type species of the present genus. For a discussion 

 of the necessity of this genus to accommodate forms once included in 

 the impossible genus Triceratium and not assignable to other genera, 

 see under Trigonium in Mann's Diatoms of the Albatross Voyages, 

 page 289. In the present species is included Brun's T. cyclamen 

 (Schmidt, Atlas, pi. 165, fig. 5) probably also the following species. 



TRIGONIUM BALAENA (Ehrenberg) Cleve 



(Peragallo, Diat., France, pi. 105, figs. 1-3; Micro. Journ., 1.859, pi. 9, fig. 15.) 



This is possibly a two-angled form of T. arctium. Bailey gives a 

 good figure of it in his New Species and Localities of Microscopical 

 Organisms, plate 1, figure 29, with the name Zygoceros radiatus. 

 Brightwell in the second reference above puts it in BiddulpTiia. 



TRIGONIUM BICORONATUM (Castracane) Mann 



Plate 37, fig. 4, and plate 38, figs. 1-3 



For a discussion of this see under Trigonium eulensteinii (Grunow) 

 Mann. 



