10 W. Hofmeister on tite Propagation 



The establishment of the assertion that the commencement of 

 conjugation in the Desmitliese and Diatomese is but little distin- 

 guished from the commencement of vegetative cell-division, 

 renders some discussion of the latter requisite. Pringsheim* has 

 already directed attention to the resemblance of this process in 

 the Desmidiese to the vegetative cell-multiplication of the joints of 

 (Edoffonium. In fact, it is an absolutely general phenomenon in 

 the true Desmidiese, so far as observation reaches, that the older 

 parts of the membrane of a cell about to divide, do not, as in 

 other cases (for example, in Zygnemese), regularly increase in 

 size with the parent-cell by growth in all directions; but the 

 older, outer layers of the integument split open with an annular 

 crack at the equator of the cell, shortly after (or during ?) the 

 division. They still remain sticking on, covering the ends of the 

 cell with a thick envelope, but become removed gradually further 

 apart by the interposition of new cellulose between their frac- 

 tured edges. The interposed new coat is the direct continuation 

 of that which lines the internal surface of the cracked halves of 

 the old shell. It is the margins of the half-shells which consti- 

 tute the rings, parallel to the end-surfaces, upon the cylindrical 

 lateral surfaces of the cells of Hyalotheca dissiliens and mucosa, 

 the wrinkled projections of the membrane in the middle of the 

 deep constriction of the cell oi Micrasterias and the large Euastra, 

 of the flat constriction of the cell of Docidium, as also the ring 

 at the equator of the external surface of Closterium : in Clo- 

 sterium and in Docidium, frequently as many as six may be 

 counted, — a phsenomenon which, in Docidium truncatum and the 

 large Closteria, may be recognized at first sight as dependent 

 upon a number of halves of cracked cells regularly encasing their 

 successors. 



The dehiscence of the coat of the dividing cell is, in all ob- 

 served cases, preceded by the formation of the septum dividing 

 the cell into two halves (fig. 30, Cosmariuin margaritifei'um). 

 The gradual development of this from the margin of the cell- 

 wall inwards as a gradually-widening, annular fold of the inner- 

 most layer of the integument, has not yet been observed, and, 

 from analogy with the processes in (Edogonium, is scarcely pro- 

 bable f. But, as in (Edogonium, the contents of the cell may 



* Pringsheim, Pflanzenzelle, p. 38, note. 



t In judging of the equivocal appearances in the cell-division of (Edo- 

 gonium, it seems to me of slight importance that the pellicular coat of the 

 contracted, already-divided contents of the cell which has not dehisced, 

 does not acquire a hlue colour when treated with the iodized chloride-of- 

 zinc solution, or with sul])huric acid and tincture of iodine. It is more 

 than probable that the cellulose is not secreted by the cell-contents in 

 lamellae of firm substance, but as a semifluid substance ; that in this latter 

 state of aggregation it does not take a blue coloiur with iodine. I men- 



