of the genera navicula and cymbella. 415 



The Genealogical Ladder 



Cynibella microcephala (Plate 28, figs. 18, 19). 



This small Cymbella is common in European deposits. The 

 striae are very faint and dense (25 to 0*01 mm.), parallel. The 

 ends are rostrate-capitate. The areas indistinct. The outline 

 slightly asymmetrical. Length about '03 mm. The typical form 

 is a little smaller and broader than the one figured on the sketches. 

 The characteristic features are : the parallelism of the striae and 

 the absence of areas. 



Cymbella Cesatii (Plate 28, fig. 17). 



I doubt whether this curious diatom ought not to come much 

 higher up the genealogical scale, anyway after Cymbella angustata, 

 I place it here only to follow Cleve. It has been assigned to three 

 genera : to Cymbella, to Encyonema, and to Navicula. It is 

 very narrow and nearly symmetrical. Its chief characteristic is 

 the distant terminal nodules. The ends are more or less acute. 

 Areas indistinct, but the central area is sometimes slightly 

 rounded. The margins begin undoubtedly to show a tendency 

 to triundulation, a feature which seems to be transitional to 

 the next form of our series. 



Cymbella gracilis- Cesatii (Plate 28, figs. 14, 15, 16). 



This is a very curious form in the Herkimer slide, where it is 

 fairly numerous, showing rather variable outlines. The areas 

 begin to be distinct. The terminal nodules are distant from the 

 ends, perhaps even more so than in Cymbella Cesatii. It is 

 rhomboidal and sometimes cymbelloid, bent along the longitu- 

 dinal axis. The striae are radiate, though not so radiate as in 

 Cymbella Cesatii, and not so dense. It usually shows on the^ 

 median striae a distinct stigma. This gives it a false appearance 

 of a Gomphonema. It is shorter than Cymbella Cesatii and still 

 more slender. As I have already mentioned, this may be a new 

 species. 



Cymballa deiicatula (Plate 28, fig. 13). 



This form is very similar to the next one, but it is not an 

 Encyonema. The terminal nodules are not distant and the ends 

 are acute. The Kev. Wm. Smith points out that it comes very 

 near to Encyonema Scotica, and so it does. 



