24 ROPER, ON BRITISH MARINE DIATOMACE, 
and extremities, and rather above that length m the inter- 
mediate space. 
Amphora sulcata, Bréb.—Valves oblong, with truncate 
extremities; the entire surface covered with longitudinal 
bands, formed of short transverse strie. Length -00266” ; 
breadth -001”; striz 14 and 20 in ‘001” (fig. 7). 
Marine. Caldy, Pembrokeshire, Rev. J. Guillemard. 
This species differs from any figured in vol. i. of the 
‘Synopsis,’ and though it approaches in structure some of the 
peculiar forms described by Professor Gregory m vol. v. of 
the ‘ Microsc. Journal,’ I cannot refer it satisfactorily to either 
of the species he has figured. M. De Brébisson, in his 
‘Memoir on the Marine Diatomacez of Cherbourg,’ gives a 
figure and description of Amphora sulcata, which appears 
only to differ in being rather more elliptical than the present 
species. I have therefore adopted his appellation, rather than 
make a further addition to our long list of native species. It 
appears to be rare, as I have only at present met with a 
single specimen. 
It differs from A. costata in the absence of the distinct 
longitudinal coste and moniliform puncta, and from the 
extremities being truncate and not produced as in that 
species ; and from A. affinis, to which the outline of the valve 
more nearly approaches, by the peculiar structure of its longi- 
tudinal bands. 
Amphora membranacea (fig. 8 a and b). 
Brackish water. Pembroke Harbour. Barking Creek. 
This species occurs abundantly in the mud from Pembroke 
Harbour, but does not appear to be common in many other 
localities, and I meet with it but rarely in the Thames and its 
tributaries. I merely give a figure, as that in vol. i. of the 
‘Synopsis’ appears to be taken from a frustule shortly after 
self-division, and gives an erroneous impression of the full- 
grown valve. The longitudinal striz are so marked a feature, 
and the breadth between the central nodules so much greater 
than in the specimen figured by Professor Smith, that the 
form now given might readily be mistaken for a distinct 
species. Fig. 8 a@ may be considered as fairly representing 
the state in which 4. membranacea usually occurs. Fig. 8 6 
is a frustule in process of self-division. 
Cocconeis scutellum, var. y (fig. 9). 
Marine. Lyme Regis. 
I figure this species as a variety of C. scutellum, as at 
