6 Transactions of the Royal Microscopical Society. 
from a photograph. But they accurately represent the general 
facts. 
I next took a frustule of the same (reputed) form (“ N. crass¢- 
nervis”) which was of an average size, and placed it under the 
same lens, with the same illumination. The result is given at B, 
which represents the frustule magnified as before—800 diameters. 
Precisely the same “resolution” is given; but faint longitudinal 
lines are seen, in addition to the far more plainly visible hemispheres, 
placed precisely as in A. 
An average specimen of what was then labelled “ F’. Sawonica,” 
was now made to replace the above, all the conditions remaining 
intact. The drawing C, magnified 800 diameters, represents the 
result.. I have obtained more general “ resolution” into hemi- 
spheres than is here shown; but the specimen is typical, and I 
preferred to retain it. Transverse striz will be seen to be uni- 
versal; longitudinal striz are here and there faintly visible, and 
the rectangular rows of hemispheres are unmistakably manifest. 
Finally, a fair specimen of what I had long known as N. rhomb- 
oides was placed on the stage of the instrument, and, everything 
remaining as before, was examined. The result — magnified as 
before—is given at D. What was now aimed at was not to get 
the largest surface of hemispheres, but to get such a general result 
as would show the exact correspondence, in ultimate structure, of 
the preceding forms with this. The transverse striz, the longitu- 
dinal striz, and the rectangularly placed rows of hemispheres, are 
all most distinctly seen. 
But during the last three years some remarkable specimens of 
Rhomboides have come into my hands—specimens in which the 
frustules reach an immense proportional development ; and in which 
the striz, or rows of hemispheres, are very much coarser in rela- 
tion to Mr. Hardman’s coarsest frustules than the -latter are in 
relation to the finest and smallest of the forms so recently called 
“ N. crassinervis.” The largest of these have been found by me in 
some mountings received from a friend, and marked as having 
come from “Cherryfield, America.” The variation in their sizes, 
on the same slide, is very great, and the consequent difficulty of 
resolution ; but I selected one of the larger forms, and employing 
the same illumination as in the instances given above, put on the 
4th immersion of Powell and Lealand (new formula) in place of the 
4th, and worked up to 600 diameters. The result is given at E, Plate 
CLXYVI., which may be taken at once as the interpreter of all the rest, 
and the witness of their oneness. The striation in this case is every- 
where resolved into rows of hemispheres, placed in rectangular 
order ; and I know of few more beautiful objects in “still” micro- 
scopy than this. The frustule is of exquisitely perfect form, 
slightly tinted with yellow-brown : the hemispheres are everywhere 
