CRITICAL NOTES ON SOME SPECIES OF DIATOMACE^. 39 



a transverse section of a series of frustules, the thick lines indicat- 

 ing the valves and the dots the cingulum. 



In the nneqnal development of the cingnlum, this species re- 

 sembles genns AmjDhora, but it possesses neither a median line or 

 central nodule. Its proper place is doubtless in the genus Pal- 

 meria (Greville in the " Ann. and Mng. of Nat. History," 1865), 

 in which genus I propose to place it with the following specific 

 characters : — 



FalmeTia vulgare (Kitton) = Amphiprora complexa, Gregory, 

 Frustule cuneate in f. v., frequently cohering after self-division. 

 Valve broadly lunate, convex margin, generally constricted at the 

 centre, stride radiant and slightly curved towards the rounded apices. 



Marine or brackish water. The Clyde and Loch Fenn. Dr. 

 Gregory. Sharks Bay, Australia, M. J. Norman. Harwich and 

 Felixstowe, F. Kitton. Dundee, Mr. Rattray. Peruvian guano, 

 and fossil in the Baldjik deposit. 



It will be seen from the above list of habitats that this form is 

 widely distributed, but with the exception of the Dundee gathering 

 (for which I am indebted to Mr. Rattray) I have never found it in 

 abundance. 



Grammatophora islandica, Ehr. — I have lately detected this 

 species in a gathering made- off the coast of Scotland, mixed with 

 G. mauva and G. serpentina. I do not feel sure that it is specifically 

 distinct from the latter form ; it is probably only a variety with 

 more distant striae. 



Nitzschia curvula, Smith. — Some confusion seems to exist 

 amongst foreign observers as to the form described in the 

 Synopsis. The species so named in the " Typen Plate" list is not 

 Smith's N. curvula, but his N. sigma. In a paper on the structure 

 of Diatoms, by G. W. Morehouse, U.S.A. (" Monthly Mic. 

 Journal," vol.-xii., p. 23), he alludes to the extreme difficulty of 

 resolving the longitudinal, as compared with the transverse striae of 

 N. curvula, Sm. I suspect that he is also unacquainted with the 

 true form. 



This form is, however, not a Nitzschia, but a Surirella {Surirella 

 intermedia of Dr. Lewis) ; the surface is striate, but the stri« are 

 not difficult to resolve. The striae on N. sigma, like those on 

 Grammatophora, are more easily resolved when mounted in balsam 

 tlian when mounted dry. 



Cymbella Ehxenbexgii, Agaridh. — Professor Smith, in his 

 Synopsis, remarks that " It is difficult to discriminate between the 

 Cocconemeta and Cymbellce when the specimens are prepared. In 

 a living state the presence of a stipes at once removes the difficulty." 



Recent observations prove, however, that their presence or 

 absence are of no generic or specific value. In a gathering from 

 Aberdeen I found Cocconema lanceolatnm living, and in a vigorous 

 state, without the vestige of a stipes. 



