232 GREVILLE, ON NEW DIATOMS. 



third of the length from the upper end; striae very fine. 

 Length -0060" to -0193". (PI. X, figs. 1—3.) 



Campylostylus striatus, Shadbolt^ MS. 



Hab. On mahogany logs imported from Honduras; G. 

 Shadbolt, Esq. On mahogany logs from the same country ; 

 George Norman^ Esq. 



This most singular diatom resembles a long, curved club in 

 shape, the line of the curve being somewhat modified by the 

 length of the frustule. In the small variety (fig. 3) it was 

 aptly compared by Mr. Norman to the bill of the common 

 curlew. The valve is nearly linear for about a third of its 

 length from the base, when it begins to widen, rapidly in the 

 small variety, more gradually in the longer one. The greatest 

 breadth in the former is about '0005", in the latter about 

 •0006"; the extraordinary difierence in the length not appear- 

 ing to have much effect on the breadth. The upper extremity 

 is less rapidly attenuated, and is never so narrow as the lower. 

 The strise are very fine, but with careful manipulation 

 come sharply out. There is no trace whatever of a pseudo- 

 nodule. 



I had prepared a description of this species for publication 

 in my last communication, from specimens presented to me 

 by Mr. Norman, when, through the kindness of Mr. Roper, 

 I received a slide of the same diatom which had been dis- 

 covered on mahogany logs in the docks by Mr. Shadbolt as 

 long ago as in 1849, and had been provisionally named by 

 him. Mr. Shadbolt^s gathering is quite pure, while that 

 made by Mr. Norman is less so, but extremely interesting on 

 account of the very dwarf habit of the frustules. I am not 

 aware that Mr. Shadbolt ever prepared a definition of his 

 proposed genus Campylostylus. The prima facie aspect of 

 the frustule is so very peculiar that no one can be surprised 

 at its being hastily taken for a new genus ; but on a rigid 

 examination I cannot perceive any good ground for separating 

 it from Synedra, which contains a number of arcuate species. 

 The inequality of the curve and the club-shaped form, cannot 

 be regarded as alone suflficient to furnish a generic character. 



Triceratium. 



Triceratium Davyanum, n. sp., Grev. — Valve with slightly 

 convex sides, rounded angles, and large, punctate pseudo- 

 nodules ; border and the central triangular space largely 

 cellulate, the former divided into compartments by trans- 

 verse lines. Distance between the angles -0068" to -0080". 

 (Fig. 4.) 



