256 CASTRACANE, ON DIATOMACE^. 



consequently giving off oxygen, is sufficient to show that 

 organisms of such excessive minuteness must be endowed 

 Avith an extraordinary reproductive capacity in order to 

 supply, by their number, the vast scoj^e of the office they are 

 destined to fulfil. Their most obvious mode of reproduction 

 or multiplication is by a process of spontaneous division or 

 fissiparity, similar to that which is seen to take place in the 

 unicellular algge and protophyta generally, and as may also 

 be said to be universal in the vegetable cell. This process of 

 division is effected in the same way as in the Desmidiece, 

 commencing with an internal movement in the granular sub- 

 stance or endocrome, which exhibits a tendency to separate 

 into two portions. These separate portions become applied 

 to the extremities of the cell, that is, to the two valves, whilst 

 at the same time may be observed the secretion of two siliceous 

 lamella or valves, which are probably invested with a delicate 

 mucous layer (or membrane) on either surface. These two 

 siliceous lamella are the counterparts of the two primitive 

 valves, and exhibit the same markings and structural pecu- 

 liarities. In this way the primitive cell ultimately becomes 

 divided into two cells, each formed of an old and new valve, 

 and each having a siliceous border or c'mgulum, in the way 

 I have on another occasion observed, at any rate, in the 

 genera Navicula, Pinnularia, Stauroneis, Eunotia, and 

 Grammatophora. 



In some species the two frustules or individuals after divi- 

 sion remain free, and enjoy an individual, independent life, 

 and in turn undergo a new division. In many other species 

 the two new frustules continue more or less adherent to each 

 other at one of the angles, as takes place in Diatoma, Gram- 

 matophora, Tabellaria, Isthmia, and Biddulphia ; or closely 

 applied side to side, as in Odontidium, Himantidium, Denti- 

 cula, Meridian ; or, finally, remain imbedded in an amorphous 

 mucuous substance, or disposed in tubes or fronds. 



This process of multijilication in the Diatomacece is a 

 generation and an extension of the individual life, of which 

 an infinitude of instances will at once j^resent themselves to 

 any one accustomed to consider the general laws of the vege- 

 table kingdom. But every plant Avhich is capable of multi- 

 plication, by gemmation or offsets, is more conmionly repro- 

 duced by seed. Tt cannot, therefore, be supposed that the 

 highly interesting class of the Uiatomacete is not also capable 

 of true and projier reproduction by seeds or by germs. With 

 respect to this, we may refer to the statement contained in 

 the classical work of Mr. W. Smitli, ' Synopsis of British 

 Diatomacea?,' founded on his own observations, and on 



