DoNKiN, on Marine Diatomacece. 
29 
keel. S, V. narrow^ linear-lanceolate^ acute, exceedingly 
convex ; keel strongly sigmoid. Length from 0-002" to 
0-0026" ; breadth variable. 
This form bears a strong resemblance to Apr. alata, in the 
contour of its F. V. But the absence of strise in the 
valve and of punctse on each side of the keel, together Avith 
the narrow linear shape of the S. V., prove it to be quite dis- 
tinct from, though nearly allied to this species. The endo- 
chrome also is differently arranged from that of Apr. alata. 
The frustule undergoing division presents a very beautiful 
appearance, from the interlacing of the external surfaces of 
the contiguous and newly formed valves ; as seen in fig. 13 c. 
A double appearance is thus produced. I may mention, as 
a curious fact, that nine tenths of all the specimens contained 
in a gathering from Druridge Bay, in which this species 
occurred abundantly, were in this double condition. This 
would show that the frustules remain adherent for a consi- 
derable period after being fully developed, by the process of 
self- division, which, in the present species, progresses with 
great rapidity. 
Epithemia. 
14. E. marina, n. sp. — Form on F. V. rectangular, elon- 
gated ; hoop on dorsal or convex surface of frustule orna- 
mented with several longitudinal lines of round, distinct, and 
widely set punctse; on ventral siirface, hyaline. Length 
from 0-004" to 0-007"; breadth from 0-001" to 0-0018". S. V. 
inflated, gently arcuate on outer or dorsal margin, on 
inner or ventral nearly linear, but slightly constricted in the 
middle ; extremities suddenly produced, acute. Canaliculi 
conspicuous, 11 in O'OOl". Striae 11 in 0-001", widely punc- 
tate ; punctse large and rather inconspicuous. Dry valve a 
bright blue, in balsam colourless. 
The fact of the hoop being hyaline in texture, on the 
ventral or concave surface of the frustule, and on its opposite, 
convex or dorsal surface, ornamented with six or more rows 
of large, round, widely set punctse, as may be easily seen by 
carefully focusing, together with the peculiar incurved longi- 
tudinal line observable near the inner margin of each valve, 
shows that it is nearly allied to Professor Gregory^ s group of 
complex AmphorcB."^ But the outline of the entire frustule 
with its inflated valves, which appear to me to possess canali- 
culi and striae which are widely punctate, have induced me, 
for the present at least, to include this large and beautiful 
^ ' Trans. Royal Soc. Ediub./ vol. xxi, psirt iv, p. 47. 
