DUBLIN NATITEAL HISTORY SOCIETY. 23 



history is deficient, that it is more advisable to allow the puzzling forms 

 to remain combined with such groups as may appear temporarily the 

 most convenient ; nevertheless, if any organism is found really not to 

 agree with the characters which are common to and appear to pervade 

 an apparently perfectly natural assemblage, it would seem to me to be 

 repugnant to a proper classification, if it could be avoided, that it 

 should be therewith associated. 



I shall, then, venture to delay, before entering on the subject pro- 

 per of this communication by drawing attention to the diagnosis of this 

 family, as given in Ralfs' monograph, " The British Desmidife :" — 

 " Fresh- water figured mucous and microscopic algae of a green colour. 

 Transverse division mostly complete, but in some genera incomplete. 

 Cells or joints of two symmetrical valves, the junction always marked 

 by the division of the endochromc, often also by a constriction. Spo- 

 rangia formed by the coupling of the cells and union of their contents." 

 Although I have no new observation in regard to the history of Pe- 

 diastrum to add, I shall just briefly compare that genus with the fore- 

 going definition. 



That Pediastrum agrees with the first clause of Ralfs' diagnosis, is 

 indeed apparent. 



In regard to the second clause, so far as I can make out, I believe the 

 complete fission into two distinct cells of any of the component ceUs has 

 not been observed ; that is to say, I believe the number of component 

 cells in any particular frond is not increased after their first formation; 

 in other words, there does not appear to be any extension of the cell ■ 

 wall of any cell accompanied by a transverse fission. Mr. Ealfs men- 

 tions that he did not see cell-division. I have certainly myself, so far 

 as my own limited experience in this genus goes, never noticed anything 

 to indicate the mode of division characteristic of the Desmidiaceae. By 

 this, of course, is not meant to be denied the subdivision of the endo- 

 chrome within the parent cell, — the necessary prelude to its conversion 

 into zoospores. The number of constituent cells in a frond, of often 

 indeed even the same species, seems, therefore, to depend on the number 

 of times the original endochrome of the parent ceU had become seg- 

 mented, and the consequent number of zoospores. Occasionally a frond 

 may be met with in which one of the component cells is about double 

 the dimensions of the others, while the normal number is deficient by 

 e ; indicating, not the special increase in size of one of the cells, as 

 compared with the others, but rather that the ultimate segmentation of 

 the endochrome within the original parent cell, preparatory to being 

 converted into zoospores, was in this one instance not fully carried out. 

 Sometimes a few marginal cells ai'e wanting, which may, perhaps, be 

 explained in the same manner ; sometimes, however, they become acci- 

 dentally removed by external forces. Indeed, it is hard to suppose an 

 increase in number of the constituent ceUs of a frond without its be- 

 coming altered from a plane to an irregular structure, such as takes place 

 in Monostroma, Ulva, &c., the dimensions of the frond itself, however, 



