134 THE PHENOMENON OP 



imperceptible internal processes, they remain wholly 

 unchanged. 



To distinguish these from the direct germ-cells (gonidia), 

 I shall call them seed-cells (spores}.* The development 

 of these spores has not yet been observed, but it may be 

 assumed as certain that they do not pass, as such, into 

 the primary generation, but produce this at the period 

 of germination by an internal transformation of their 

 contents, and bring these to light as a new generation, 

 with a dehiscence of the old envelope. Certain early 

 conditions observed in Closterium and Euastrum, namely, 

 families of unusually small individuals, enclosed in 

 transparent, colourless vesicles,! render it even probable 

 that in certain genera of this family, a number of indi- 

 viduals are produced from one spore, by a formation of 

 transitory generations occurring already within the spore. 

 The enclosing vesicle is probably the dissolved and 

 swollen-up internal cell-coat of the spore, which holds 

 the young individuals combined for some time after the 

 outer coat of the spore has been thrown off. The be- 

 haviour of the spores of the family of the Zygnemaceae, 

 which appears to be closely allied to that of the Desmi- 

 diaceae, confirms the conjecture as to the development of 

 the spores of the latter. The filaments of the Zygnemaceae 

 are composed, like those of the Desmidiaceae with con- 

 nected cell-generations, merely of one kind of vegetative 

 cells, which increase by repeated halving.^ The last 



* Many authors call these cells sporangia; but if we compare their 

 behaviour with that of the spores of the higher orders of Cryptogamia, in 

 which, in like manner, only the internal substance is developed into the 

 germ-plant, the coat being thrown aside, we must regard them as true 

 spores, while, on the other hand, those spores of the Alga; passing directly 

 into germination, to which belong in particular all s warming-spores, differ 

 importantly from ordinary spores, and would be better termed Gonidia. 



t Yide Ralfs, 1. c. t. xxvii, and Focke, ' Physiologische Studien,' (1847,) 

 t. iii, fig. 27. 



f So far as my observations extend, the filaments of the Zygnemacese 

 appear, like those of Desmidium, to be formed merely by repeated division 

 into equivalent cells, thus to possess neither apical growth nor distinction 

 into an upper and lower end. I have never found the radical structure by 

 which they are often attached otherwise than lateral. Unfortunately I 



