Navicula serians, N. rhomboides, and Pinnularia gibba. 165 



Navicula rhomboides, Ehr. (Smith, Synopsis, pi, 16. fig. 129). 

 PI. IV. fig. 13. Longest diameters 22 by 5 6000ths of an inch. 



The figure of this frustule, which has a shade over its surface 

 like ground glass, as it appears under a magnifying power of 

 800 diameters, admits of having this shade resolved into trans- 

 verse lines by a higher power (see Smithes figure). Two con- 

 jugations only of it are given, although several more were 

 sketched, and still many more examined; but to these are added 

 two other conjugations (figs. 14 & 15), of smaller individuals of 

 the same species, whose specific characters, although identified 

 by myself, cannot be represented upon the scale on which they 

 have been drawn for publication, on account of their extreme 

 minuteness on the smaller frustule. 



These conjugations, it will be observed, produced sporangia! 

 frustules of the size of fig. 16, which, in its turn, will be observed 

 to be about the size of the conjugating frustules in figs. 11 and 

 12, from which a frustule about the size of fig. 13 would be 

 produced; while I subsequently found a group where the con," 

 jugating frustules were nearly as large as the latter, and which 

 therefore would have produced a frustule still larger than fig. 13. 

 The conjugating frustules of these three groups, beginning with 

 the smallest, which was TrVoth of an inch long, gave approxi- 

 matively the proportions of 4, 11, and 17; while those of the 

 sporangial frustules which they respectively produced afforded 

 the proportions of 10, 18, and 26. Thus we are presented with 

 four successively larger sizes of the same species, three of which 

 were produced by conjugation, while these are only to be con- 

 sidered average measurements of the elements of the conjuga- 

 tions witnessed, there being every variety of size both of conju- 

 gating and sporangial frustules between them. 



Another observation made on the smallest of the conjugations 

 was that which may be partially seen in fig. 14, viz. that the 

 empty valves of the conjugating frustules appeared to be borne 

 out upon the ends of the sheaths, instead of lying parallel to 

 them as in the other conjugations. Whether this imports that 

 it may be the case in the conjugation of small frustules gene- 

 rally, and not with large ones, is a question that I am not pre- 

 pared to answer. 



Pinnularia gibba, Ehr. (Smith, Synopsis, pi. 19. fig. 180). PI. IV. 

 fig. 21. Longest diameters 22 by 2 6000ths of an inch. 



Two conjugations only of this species are delineated, viz. 

 figs. 17 and 18. In the latter the valves of one of the conju- 

 gating frustules had been rubbed off, while the sheath' in fig. 20 

 and the conjugating frustule in fig. 19, which was in company 



