GREVILLE, ON DIATOM A CE:E. 11 



nodule. Length of fnistnle 0-0024" to 0-0038"; breadth 

 0-0010" to 0-OOU". Stride 25 in 0-001". (PI. Ill, fig. 7.) 



A very striking and interesting species, of which I have 

 seen two individuals of the sizes indicated in the specific 

 character. Like others of the group to which it belongs, it 

 would appear to vary greatly in its dimensions. It seems to 

 be most nearly related to N. clavata of Gregory {' Trans. Micr. 

 Soc./ vol. iv. p. 46, PI. V, fig. 17), which it closely resembles 

 in general outline ; but, as Professor Gregory has already re- 

 marked, there can be no doubt regarding the distinctness of 

 the present form. The most conspicuous difference lies in 

 the blank space which interrupts the striation on each side 

 throughout the whole length of the valve. In N. clavata 

 this space follows nearly the same rule as in N. Hennedyi, 

 Sm., the outer boundary of the space corresponding with the 

 curve of the valve. Whereas, in the Diatom under con- 

 sideration, the blank space constitutes a mere linear band, 

 rmming parallel with the median line, except at the centre, 

 where it bends slightly towards the nodule. Between this 

 space and the median line the interrupted striae form also 

 linear bands, which are continued into the produced extremi- 

 ties. The strise are more numerous than in N. clavata, and 

 very obscurely moniliform. Another species to which our new 

 Diatom bears, at first sight, no inconsiderable resemblance, is 

 Pinnularia {Navicula) Couperi of Professor Bailey, described 

 in his microscopical observations made in South Carolina, 

 Georgia, and Florida. {' Smithsonian Contributions to Know- 

 ledge,^ vol. ii.) In that form, however, the ends are not suddenly 

 produced, the sides are somewhat constricted, the striation is 

 conspicuously moniliform, and the blank spaces, instead of 

 running parallel with the median line, and terminating at 

 the angle where the produced ends spring from the lateral 

 curve of the valve, are gradually attenuated, and converge 

 and terminate with the median line itself at the nodule. I 

 have much pleasui'e in dedicating this fine species to my 

 friend Professor Gregory, whose acuteness and perseverance 

 have converted the unpromising sand of Glenshira into 

 " diggings " rich in new and curious forms. 



8. Navicula compacta, Grev. Valve broadly oblong, con- 

 stricted at the sides, the shoulders much rounded, ends sud- 

 denly produced, obtuse; striae becoming faint towards the 

 middle, Avhere they are stopped by a line running close to 

 and parallel with the median line. Length of frustule 

 0-0010"; breadth 0-0006". Stride about 12 in 0-001". (PI. 

 Ill, fig. 8.) 



