DIATOMACEiE. 



[ 201 ] 



DIATOMACE^. 



Diatomacese have been kept a long time. 

 Perhaps the most remarkable development 

 of the silicified hoop occm^s in Biddulphia 

 (PI. 14. fig. 9), Isthmia, and similar forms; 

 the new half-frustules formed inside the 

 "hoop" of these genera slip out from it 

 like the inner tubes from the outer case of 

 a telescope. In Melosira (PI. 6. fig. 8) the 

 hoops appear to keep the new frustules 

 united too-ether for some time. It seems 



CD 



probable that the " hoop"-structure is not 

 always silicified, but that it is the source 

 from which, by softening and swelling up, is 

 derived the gelatinous envelope of some 

 kinds, as is the case with the gelatinous 

 sheath of Hyalotheca, &c. among the Des- 

 midiaceae, and Nostoc, &c. among the fila- 

 mentous Confervoids. The hoop appears 

 to be a provision for the protection of the 

 nascent half-frustules, which probably do not 

 become silicified until full-grown, and w ould 

 thus be liable to be injured or disturbed by 

 the movements of the rigid and heavy parent 

 half-frustules, if the centre of the frustule in 

 process of division was naked as in the Des- 

 midiaceae. 



The development of the stipes upon w hich 

 the frustules of many genera are attached, is 

 at present altogether a mystery. 



The only mode of reproduction (besides 

 the division) kno\Mi certainly to exist in the 

 Diatomaceae, is one in which the operation 

 of conjugation takes place. This has been 

 observed in a number of genera, and presents 

 considerable variation in its details. In Fra- 

 gilaria (PL 6. fig, 4) and Surirella (PI. 6. 

 fig. 5) the conjugation takes place between 

 two free frustules lying near together, each 

 of which opens at the suture and extrudes 

 its contents in a mass (probably enclosed in 

 the primordial utricle) ; the masses of con- 

 tents coalesce, the whole meanwhile beco- 

 ming involved in a mass of gelatinous sub- 

 stance. After a while, the body resulting 

 from the conjugation is seen to assume the 

 form of a fi-ustule, of larger size than the 

 parents, which the discoverer, Mr. Thwaites, 

 called a sporangial frustule. In the majority 

 of cases, however, as in Eunotia (PI. 6. fig. 

 6), Gomphonema, Cocconema, Sec, the con- 

 jugation is double, as is the case in Closte- 

 rium lineatum (Conjugation); the con- 

 tents of the parent-frustules apparently divide 

 into two portions (as if for cell-division) 

 before conjugating, and then there is a col- 

 lateral conjugation of the two pairs, two 

 sporangial frustules being the result. In 

 Melosira (PL 6. fig. 8) and Orthosira (PL 6. 



fig. 9) the conditions are different, and even 

 more curious, if the received view be correct. 

 The appearances presented seem to indicate 

 that the conjugation takes place between 

 two segments of a frustule which have sepa- 

 rated as if for ordinar}^ cell-division, but 

 instead of forming new half-frustules, have 

 coalesced again and secreted a coat over the 

 entire surface, thus constituting one new 

 independent "sporangial" frustule of larger 

 size. In Melosira (PL 6. fig. 8) this has 

 been observed to increase by cell- division, 

 and form a new filament of far greater dia- 

 meter than that to which it owed its birth. 

 The "sporangial" frustules of the free forms 

 doubtless increase by cell-division in the 

 usual way. (See Conjugation.) 



A great difliculty meets us here. The 

 necessary consequence of the conjugation 

 just described is, that every species in which 

 it occurs must be represented by two forms, 

 one small and the other large, between which 

 a gap exists, over which we have at present 

 no means of bridging, except by supposing 

 that the two new halves formed in cell-divi- 

 sion need not always be equal, and that by 

 a dwindling away through a succession of 

 steps of this kind, the progeny of the spo- 

 rangial frustules may be reduced to the 

 original size. The size of the frustules is said 

 also to vary with the depth of the sea, in 

 marine species. The efi'ect of all this seems to 

 have been disregarded in systematic treatises 

 on the Diatomaceae. Some of the book-species 

 appear to produce other book-species by 

 conjugation; according to Focke, Surirella 

 splendida produces S. hifrons, a very di- 

 stinct form, and it is not improbable that 

 <S. splendida is produced by the conjugation 

 of S. Microcora (Focke). There is great 

 probability, however, that the observations 

 recently made by Focke upon the contents 

 of certain species will lead to the discovery 

 of another mode of increase, a reproduction 

 by gonidia, either active or quiescent, such 

 as occurs in the Desmidiaceae and the other 

 Confervoids. Indeed the contents of the 

 cells of Melosira have been observed to 

 display a motion like ' swarming.' Such 

 spores or gonidia discharged from the large 

 ' sporangial ' frustules might reproduce the 

 small form, just as the young filaments deve- 

 loped from the zoospores of Chcetophora, &cc., 

 are veiy slender compared with those of full- 

 grow'n filaments. Focke describes and figures 

 appearances on the contents of the frustides 

 of Pinnularia viridis, Surirella hifrons, and 

 others, very like what occur occasionally in 



