of the Desmidiese and Diatomese. 15 



when seen from the side (fig. 28) . It would be no great stretch 

 of imagination to regard these cells as the conjugation-spores of 

 a large Desmidiean. But these spores are all spiny, with the 

 single exception of those of Xanthidium armatum. This very- 

 striking form occurs but rarely with us, having hitherto been 

 found only in a single locality, while these globules are as com- 

 mon as they are abundant, and are often found in great numbers 

 in forest pools, which harbour, in addition to them, only very 

 small Desmidiese. But such a supposition is still more decidedly 

 negatived by the circumstance that the cells in question are 

 sometimes found dividing into two (fig. 29). This renders 

 it in the highest degree probable that they are independent 

 organisms, — Desmidiese without a central constriction, which 

 may form the commencement of a series of forms terminating 

 in Micrasterias. 



These cells frequently appear surrounded by a wider coat, 

 inside which the cell then floats freely, enclosed by its own closely 

 investing coat (figs. 26, 27). Several such empty coats are 

 often met with, even as many as six sticking one inside another. 

 Close investigation shows that the broader empty coats have an 

 orifice, towards the border of which the membrane grows gra- 

 dually thinner. These holes have not the aspect of perforations 

 of the outer walls through external injury ; they rather resemble 

 the orifices of the walls of Cladophoraj through which the 

 swarming-spores escape. It might be conjectured that the 

 plant multiplied by swarming-spores, and that solitary ones be- 

 coming developed inside the empty coat of the mother-cell gave 

 rise to that appearance. But this is contradicted by the great 

 frequency of their occurrence, as also by the circumstance that 

 we never find a number of green cells inside one cell-coat. It 

 is more probable that the contents of the cell contract, and 

 become coated with a new membrane, when the old one is per- 

 forated, — by unknown causes, which perhaps lie in the course of 

 development of the species. 



If we seek to bring the phsenomena introductory to vegetative 

 cell-multiplication under one point of view with the preparations 

 for conjugation, we find that, in the Desmidiese, in both cases a 

 new membrane is formed around the total contents of the cell, 

 which indeed lies close upon the old coat at all points, but by 

 no means adheres to it, as we are accustomed to conceive of the 

 so-called layers of thickening of the cell-wall. The growth of 

 the young membrane cracks the stronger old one — in vegetative 

 cell-multiplication always in an annular form, in conjugation 

 mostly in a one-sided manner, with a valve-like slit {Hyalotheca 

 dissiliens, figs. 5-8; Closterium, fig. 10). At this stage first 

 occurs a distinction between the two processes of development — 



