86 REV. E. o'meara. 



slight indentation is observable at the ends of the plate, in 

 which there is an approach to the Cocconeideee, similarly 

 geniculated in the axis of division. The division of the 

 endochrome plate proceeds just as in the case of Gomphonema, 

 towards which genus many transition-forms from Rhoicos- 

 phenia tend. 



The formation of auxospores was observed by Thwaites 

 in the case of R. curvata so early as 1847. This process 

 goes on precisely, as in Gomphonema, only the plasm-sac, ac- 

 cording to Thwaites, does not emerge at the side, but from an 

 opening at the end of the cell. Smith found the same species 

 as well as R. marina in conjugation. In the case of the form 

 last-named, Lliders has added the remark that the auxo- 

 spores before they have attained the length of the older cells 

 are invested with a very fine silicious coating, which exhibits 

 broad transverse striae, This investment is at first cy- 

 lindrical, but becomes bent in the firstling-cells, which issue 

 from it. 



VII. 



The forms embraced in the four groups to be now con- 

 sidered are placed by Dr. Pfitzer, in reference to themselves 

 and to other groups, in a relationship very different from 

 that assigned to them by most writers of authority on the 

 subject. The groups referred to are Amphipleurea, Plagio- 

 tropide(B, Amphitropidece, and Niizschieee. Heiberg indeed 

 places the genus Amphipleura under that of Nitzschia, and 

 that because the species of the former family, which he had 

 specially examined, was that named by Smith A. sigmoidea 

 — a form which I believe to be identical with Nitschia 

 sigmoidea. So under the sujiposition that the form named 

 is to be properly regarded as really belonging to the genus 

 Amphipleura, he was quite right in the position he assigned 

 to it; but regarding Amphipleura pellucida as a genuine 

 type of the family, its position, according to Heiberg's 

 system of classification, is widely apart. Grunow at first 

 placed Amphipleura among the Surirellea, but afterwards 

 made it the type of a distinct group, in Avhich he included 

 Berkeleya. Pfitzer agrees with Grunow as to the compre- 

 hension of Berkeleya in the group, though he places the 

 group itself in a very different relation. The position of 

 Amphipleura, according to Ralfs, is between the Nitzschiem 

 and SurirellecB. Rabenhorst placed it under the Navicu- 

 lacece in his Siissw. Diat., but subsequently, in Fl. Eur. 

 Alg., ranges it between the Synedrece and Niizschie(S, 



