om 
324 Proceedings of the Royal Irish Academy. 
as such, inasmuch as he regarded the inner layer of the valves on 
which the characteristic coste are situated as an annulus or dia- 
phragm of the same structure as that which we find in the Striatellez : 
also he considered the costz to be canaliculi, which does not corres- 
pond with the actual facts of the case. Grunow has the merit of 
having been the first to point out the error of the opinion of Smith 
above referred to.’”’—De Danske Diat., p. 92. Whether the plate bear- 
ing the locul is more intimately associated with the connecting 
membrane, as Smith thought, or with the valve itself, as Grunow and 
Heiberg are of opinion it is, this is certain, so far as my observation 
extends, that, as Grunow remarks, the plate seems to attach itself more 
frequently to the valve than to the connecting membrane; but as the 
valves frequently occur without the plate, and the plate is often 
found detached, I am disposed to consider it not so much an inner 
layer of the silicious epicterm as a separate formation, and much more 
intimately related to the diaphragms of the Striatellee than to the 
inner layer which bears the costz in the Epithemia. Smith describes 
the loculi as opening by foramina along the line of suture, a statement 
which Ralfs repeats. I have however failed to notice any such open- 
ings, the plate having ever appeared to be perfectly solid. Inasmuch 
as Thwaites considered the occurrence of the frustules in gelatinous 
cushions the distinctive character of the genus Mastogloia, and other 
distinguished writers have entertained the same opinion, Grunow’s 
remark on the subject is deserving of attention: ‘‘ Whether the species 
of the genus Mastogloia occur invariably in a gelatinous investment, is 
a matter concerning which I am very doubtful, as in a fresh collection 
I observed Mastogloia Smithu free, while I found no specimens in a 
gelatinous cushion’’—Verhand. der K. K. Zool. Bot. Gesel., Band x., 
1860, p. 575. However this mav be, the occurrence of the plate with 
loculi in the perfect frustule is a mark of distinction which identifies 
the genus. Further, it was considered by Grunow, that the occurrence 
of the inner layer with its costate striation, so different from the 
sculpture of the valve, constitutes a strong bond of affinity between 
Mastogloia and Cocconeis. If, however, the opinion I have expressed as 
to the distinctness of the plate from the valve be correct, this resem- 
blance fails, and in the general details of structure the two genera are 
widely distinct. The process of reproduction in this genus has been 
observed by Liiders: according to his observations, two mother cells 
produce two auxospores. Pfitzer, Untersuchungen tiber Bau und 
Entwicklung der Bacillariaceen, p. 74, remarks, ‘‘that in this 
feature the genus corresponds with the Naviculee, and not with the 
Cocconeidee in which Grunow placed it; for the latter, out of two 
mother cells, develop but a single auxospore.” 
Mastogloia lanceolata, (Thwaites.) Marine or brackish water. 
Valves lanceolate ; marginal plate wide at middle, and gradually 
tapering to the ends; loculi narrow, and numerous; median line 
slightly undulate; strongly marked at the central nodule, and greatly 
