12 Prof. W. Smith on the Diatomacese of the Pyrenees. 



Kew Gardens iu April 1858, and again by myself in the July of 

 the same year; it is a larger form than the present, with a 

 broader and more regularly elliptical valve and more distinct 

 strise. I propose to describe it in a future edition of the ' Syn- 

 opsis ' under the name of Diadesmis peregrina. 



Plate II, fig. 16. Diadesmis Gallica : a & 6, S. V. of frustule; c, F. V. 

 of filament. 



Genus Gomphogramma, Braun. 



Filaments compressed, continuous, of few frustules, free; 

 frustules annulate, indefinite; septa alternate, nearly equal; 

 valves elliptical, costate. 



Gomphogramma rupestre, Braun. S. V. irregularly elliptical, 



orbicular in young specimens. Costse direct, J to 5. Width 



of filament -0002" to -0007". Breadth of valve -0001" to 

 •0003". v.v. 



Fresh water. Moist surface of high rocks : Gabas near Eaux 

 Chaudes, elevation 3362 feet ; near the Mahourat Source at Cauterets, 

 elevation 3804 feet ; old road near Pierrefitte, elevation 2006 feet ; 

 Lac de Gaube, elevation 5866 feet. 



The structure of the frustules in this species bears so close a 

 resemblance to that of Tetracyclus lacustris, that I should have 

 referred them to the latter genus, had not M. de Brebisson 

 informed me that the present species had been discovered by 

 Professor Braun near Freiburg in Breisgau, and placed in the 

 genus " Gomphogramma." In Tetracyclus the valve is cruciform 

 and the costre arched ; in Gomphogramma the valve is elliptical 

 and the costse direct ; but these seem rather to belong to specific 

 than geneiic characters, and the propriety of uniting these genera 

 hardly admits of a question. Should this course be adopted, it 

 would be necessary to alter the name Tetracyclus, which would 

 no longer be appropriate, and place the species of the two genera 

 under the common name of Gomphogramma ; but until British 

 specimens of our present species be discovered, I do not propose 

 to make this change of nomenclature. The generic characters 

 of Gomphogramma will be better understood by the reader if he 

 refers to the observations I have given under the genus Rhab- 

 donema (Synopsis of British Diatomacese, vol. ii. p. 32). 



Plate I. fig. !?• Gomphogramma rupestre : ahb, front view of frustules ; 

 d &c k, side view of ditto ; e, f, g, h, I, m, & n, annuli with their septa. 



The above descriptions include the principal novelties both in 

 species and varieties noticed in the fresh gatherings made during 

 this interesting excursion. The list might be greatly increased, 



