Ross : Classification of Diatoms 405 



correctly calls S. gregorii Ralfs) and .S". salina are quite distinct species is con- 

 firmed, 



Ultrastructure confirms the view that S. phoenicenteron , S. anceps, and 

 S. acuta should be placed in the same genus. S. amphioxys, S. saliua, and 

 S. smithii differ considerably from these and from one another. Meresch- 

 kowsky (1903a), on the basis of endochrome structure, removed .S\ amphioxys 

 and .5. salina from Slauroneis and created a new genus, Slaurophora, for the 

 two species. Although their ultrastructure indicates that they should perhaps 

 be removed from Slauroneis, it provides no contirmation for grouping them 

 together. Information about many more species is needed before any firm 

 conclusions can be drawn about the correct position of these species. S. 

 smithii, however, seems to be close to Capartogranima both in the structures 

 associated with the central nodule and in the ultrastructure of the chambers, 

 and S. salina bears some resemblance. It is noteworthy that Frustulia rhom- 

 boides var. saxonica (Rabenh.) De Toni (Helmcke et al., 1961, plates 279 to 

 280) has an ultrastructure almost identical with that of Caparlogramma and 

 S. smithii, and so also has Scoliopleura tumida (Breb.) Rabenh. (Helmcke and 

 Krieger, 1954, plate 177), a species grouped by Cleve (1894) not with the 

 other members of that genus but in his Naviculae Microstigmaticae, in which he 

 also included Slauroneis. This ultrastructure has certain similarities to that 

 found in most of the species of Pleurosigma and Gyrosigma examined. Whether 

 the species that possess this type of ultrastructure in common form a group of 

 genera more closely related to one another than to the rest of the Naviculaceae 

 is a question that can only be determined as more knowledge is accumulated, 

 but it seems that it is a possibiUty. 



Amphora. As mentioned, there is a species which differs from the others 

 placed in the genus Caparlogramma only in shape of frustule; it has that char- 

 acteristic of the genus Amphora although the other species of the genus are, 

 like most Naviculaceae, symmetrical about the apical, transapical and pervalvar 

 planes. Cleve, in 1896, (p. 99) made the suggestion that the species placed in 

 the asymmetric genera Amphora and Cymbella Ag. were more closely related to 

 symmetrical species of similar valve structure than they were to one another. 

 The discovery of this new species of Caparlogramma adds further evidence for 

 the view that symmetry by itself is not a proper basis for delimiting genera. 

 The only species of the large and variable genus Amphora the ultrastructure of 

 which is known are A. cofeiformis (Ag.) Kiitz. (Helmcke and Krieger, 1953, 

 plate 76), A. deUcalissima Krasske (Helmcke el al., 1961, plate 294) and A. 

 ovalis (Kiitz.) Kiitz. (Helmcke and Krieger, 1953, plate 77, 1954, plate 181). 

 In A. cofeiformis and A. ovalis the ultrastructure resembles that found in 

 Anomoeoneis exilis (Helmcke and Krieger, 1954, plate 169) and A. serians 

 (Breb.) Cleve (Helmcke and Krieger, 1953, plate 68), which may indicate 

 relationship. Amphora deUcalissima has a cjuite different structure. 



Cymbella. This is another genus which, like Amphora, is distinguished from 

 Navicula solely on the basis of asymmetry. Cleve (1894, p. 157) considered 

 that its species were most closely related to those of Navicula subgen. Navicula 

 (his Naviculae Lineolatae) . As far as ultrastructure is concerned, this is true 

 of C rabenhorslii Ross (Kolbe and Golz, 1943, plate 1, fig. 3, Helmcke and Krie- 

 ger, 1953, plate 75, as C. gracilis (Rabenh.) Cleve), C. turgida Greg. (Desika- 



