ZOOLOGICAL AND BOTANICAL ASSOCIATION. 127 



least evident that they resemble each other very much, and are closely 

 allied. • 



M. Hofmeister has already compared his plant to a Desmidian. At 

 first sight ours has some resemblance to Cosmarium Ralfm; but, being 

 a perfect sphere, it, of course, wants the constriction and elliptic ends of 

 that species. I hardly think, in our plant, that there is a central suture, 

 though the cast-off coats have a tendency to split into two hemispherical 

 portions : they often display, after collapse, a flattened or depressed cir- 

 cular portion (when viewed sideways, almost as if a small segment had 

 been abruptly cut off the sphere), possibly representing the somewhat 

 flattened surface of contact with the companion cell, just after the 

 complete shutting off of the two cells or entire formation of the double 

 wall. "We have seen that there is not an interruption of the endochrome 

 until the commencement of the process of division. As to the mode of cell- 

 increase which prevails amongst theDesmidiaceae, it does appear probable 

 that during the division and formation of the constriction, the original 

 hemispheres may be pushed asunder, by new growth, without their mate- 

 rially altering. But, even so, is this an exclusively Desmidian (and Diato- 

 macean) characteristic ? An apparently similar mode of division seems 

 to hold good with the greatly more minute cells of the moniliform fila- 

 ments of the Nostochaceous Algaj (e. g. Dolichospermum). One of these 

 globular cells appears to elongate, to become constricted into a figure of 8 

 form, deeper and deeper, until two new globular cells grow out of one, 

 during which process the opposite hemispheres of the original cell appear 

 to remain unchanged. So in the plant in question. The main distinction 

 appears to be, as I apprehend, that in neither of these organisms is the 

 first step of the process of division — the formation of a septum between 

 the halves of the cell-contents, as appears to prevail in the Desmidiaceae. 

 So far as I can make out, the halves of the contents of the cell about to 

 divide (in the plants alluded to) merely become retracted from each 

 other, separated by a sharp, smooth line of demarcation, and eventually 

 become shut off by an addition to, and external gradual constriction of, 

 the original outer cell- wall, as well as, of course, the original pri- 

 mordial utricle, afterwards producing each its own proper cell-membrane. 

 In other words, the new and intervening growth appears to be an ex- 

 tension and continuation of the original outer wall of the dividing cell, 

 — still a single cavity only, until the constriction becomes shut off, or 

 until the halves of the cell-contents have withdrawn and become invested 



