178 P. T. CLEVE, SYNOPSIS OF THE NAVICULOII) DIAToMS. 



C. lanceolafa var. fossilis Pant. III Pl. XXIII f. 344. 

 _ _ var. robusta Pant. III Pl. XXIII f. 3.50. 

 C. marina Pant. III Pl. XIX f. 274 (Amphora angnsta var.?). 

 C. ohiusa Pant. III Pl. V f. 79. 

 C. pachyptera Pant. Pl. XXI f. 304, 316. 

 C. Fcragalli Pant. III Pl. XLII f. 584. 



C. perfeda Pant. III Pl. XVII f. 249 (C. fymbiformis var.?). 

 C. plutonica Pant. III Pl. XX f. 297 (C. aspera var.?). 



C. praclara Pant. Ill Pl. XXXVI f. 512 (resemble.'? C. lanceolata but lias closer .striae). 

 C. RaJcoczya»a Pant. III Pl. VIII f. 121 (C. Cistula vai'. maculata?). 

 C. simplex Pant. III Pl. XXI f. 308 (C. ventricosa?), 

 C. Stauhii Pant. III Pl. VIII f. 131 (C. leptoceras var.?). 

 C. S.vontaghii Pant. III Pl. X f. 161 (Anipliora augusta var.?). 

 C. suavi.iVmT. III Pl. XV f. 229 (akin to C. leptoceras?). 

 C. turgida Pant. III Pl. VI f. 103 (C. parva?). 

 C. raiiiJa Pant. III Pl. IX f. 154 (C. aspera var.?). 

 C. regcta Pant. III Pl. XXIV f. 359. 



Gomphoiiema Agardh (1824). 



Val ve more or le.ss elongated, clavate, or asymmetrical to tlie traiisverse axis. Structure: 

 transverse slightly radiate strife or rows of punota. Coiiiiecting zone not complex, broader in tlie 

 upper tlian in the lower end. CeU-contents a single chromatophore, leaving only a narrow parietal 

 plasmaband along one side of the zone, deeply sinnose below the median line. On conjugation two 

 auxoppores are formed by two mother-cells, parallel to them. The plane of division of the primordial 

 cell is at right angles to the plane of division of the mother-cells (Pfitzer Ban u. Entw. p. 88). 



The large G. genmiafum was observed as early as 1773 bj- O. F. Muller, who naraed it 

 Vorfirclki pyraria. The genus Gomphonema was established 1824 by Agardh for two species, and 

 since then a large number of species have been formed by Ehrenberg, Kötzing and others, unfort- 

 unately fonnded on trifling characteristics. For forms living, as Diadesmis, in bands of closely 

 connected frustules, Ehrenberg ereated (1843) the genus Sphctio.siru. Kutzing founded (1844) the 

 genus Splienella for free-living forms and Rabeniiorst (1853) the genus Gomplionella for forms, 

 which live in gelatinous masses. Heiberg (1863) maintained, with justice, that these genera are 

 not admissible, as they are founded on characteristics, which occur in the same species. An 

 attempt to arrange systematically the known species of Gomphonema was made (1878), by GRrxcw 

 in his description of the algte of the Caspian Sea, and since then he has given a number of most 

 valuable figures in Van Heurck's Synopsis Plates XXIII, XXIV and XXV. As among them there 

 are several species from America, which, while having the general outline of Gomphonema, ditfer in 

 their structure, and in the presence of longitudiual lines, similar to those of Scoliotropis and 

 Caloneis, I now exclude these forms, and place them in a .-separate genus Gomphomis. Eecently 

 Brun has described, as G. cantalieum, a species, which seems to be allied to Gomphoneis, as it 

 shews across the strias a longitudiual line, but the striation of this species is iu all other respects 

 quite different from that of Gomphoneis. 



The valve of Gomphonema is asymmetrical to the transverse axis, and is usually liroader 

 iu tlie upper portion, with a wedge-shaped lower end. The central nodule is uearer to the upper 

 end, or apcv, thau the lower, or basi-i. Manj' species are asymmetrical also to tlie longitudiual 

 axis, but iu a less visible degree, as Pfitzer first pointed out, with a parallel asymmetry, like that 

 of the Cymbell*. Some varieties are even slightly cymbelliform. In many species there is on 

 oue side of tlie central nodule an isolated puuctum, or fifigriia, as in several Cymbelkt', and in others 



