DIATOMACE^. 



[ 248 ] 



DIATOMACE^, 



these frustules we shall distinguish as com- 

 pound (PI. 18. fig. 69). 



During the process of multiplication by 

 division, which is almost always going on, 

 the annular siliceous, narrower or broader 

 band, or hoop, imdergoes an increase of 

 width, and the two valves are removed 

 some distance apart (PI. 15. figs. 7, 45 ; 

 PI. IG. fig. 1). Sometimes it consists of 

 two pieces, one overlapping the other. 

 Some of the valves are furnished with pro- 

 cesses, called cornua or tubuli (PI. 16. fig. 

 30 b) ; the surfaces of others are undulate, 

 producing the appearance of dark curved 

 or wavy lines or bands (PI. 16. figs. 22, 23, 

 24) ; sometimes curiously arranged lines, 

 (vittfe) indicate either imperfect internal 

 septa, the internal margins of the flattened 

 hoops, or certain inflections of the margins 

 of the valves (PL 1. figs. 14, 15 ; PI. 17. 

 figs. 17, 18). 



In the young state of these organisms, 

 the endochrome or cytioplasm is imiformly 

 distributed; but after a certain time, it 

 becomes accumulated into various, usually 

 very regular and often elegant forms, and 

 minute granular globules are formed, trans- 

 parent vesicles become visible, drops of oil, 

 and vesicles tilled with granides, which at 

 first are motionless, but afterwards move 

 about as in the swanniug motion of the 

 Algse. Frequently a considerable nucleus- 

 like body is present in the middle of the frus- 

 tule (PL 15. fig. 33 a). As we have seen it, 

 delicate processes were visible arising from it. 



The frustules of the Diatomacese are 

 sometimes surrounded by a transparent ge- 

 latinous sheath, frequently of great delicacy, 

 or contained in gelatinous simple or branched 

 tubes ; in some genera they are attached by 

 a stipes or stalk to water-plants &c. 



Those Diatomacese which are not fixed 

 by a stipes, and especially such as are linear 

 or spindle-shaped, are capable of sponta- 

 neous motion ; they may be constantly seen 

 slowly moving across the field, or now and 

 then starting somewhat suddenly forwards, 

 moving mostly in the direction of their 

 length, sometimes receding, sometimes per- 

 forming a rotatory movement on their axis. 

 Those which are contained in numbers in 

 gelatinous tubes, like Mict/07iema (PL 19. 

 fig. 10), are capable of moving backwards 

 and forwards in these; and Thwaites de- 

 scribed a curious movement of the frustules 

 of Bacillaria parndoxa, where the frustules, 

 united in a band, slid backwards and for- 

 wards over one another. 



The cause of this motion is undetermined. 

 It has been supposed to be produced by the 

 endosmotic changes connected with the 

 nutrition of the organisms ; but this is very 

 improbable, otherwise it woidd be met with 

 frequently in other minute unicellular or- 

 ganisms; moreover, the diffiisi on -move- 

 ments are far too slow and gentle to explain 

 them. No true vibratile ciUa have been 

 detected upon the Diatomaceee, although 

 Thwaites imagined, from the appearance of 

 currents in the water, that they exist on 

 Bacillaria. Some are not unfrequently 

 found bearing tufts of or fringed with rigid 

 cilia, like those often seen at the ends of 

 the filaments of Oscillatorics ; these woidd 

 seem to be formed like the fringes met with 

 in the Desmidiace?e, by a modification of 

 the gelatinous envelope; they never exhibit 

 motion. 



It has more recently been supposed to 

 arise from the action of a delicate layer of 

 protoplasm investing the frustules, or pro- 

 truding through chinks between the valves 

 of the frustules at the suture. (H. L. Smith, 

 Jn. Mic. Soc. i. 1878-79.) 



In placing the Diatomaceae among plants, 

 we assume an agreement between the frus- 

 tule of a Diatomacean and an individual cell 

 of any undoubted vegetable, such as Proto- 

 coccus ; and between the series of frustules 

 such as we find in Fragilaria (PL 16. fig. 33), 

 or Melosira (PL 17. fig. 5 a), and the cellu- 

 lar filament of a Conferva or a Zygnema. 

 This agreement does imdoubtedly exist ; and 

 the siliceous shell is really only a result of 

 the incrustation or permeation by silica of a 

 true vegetable cell-membrane, just in the 

 same way as takes place in the epidermis 

 of Equisetum. It is not yet ascertained in 

 either case whether the silica is outside or 

 in the substance of the cell-membrane ; 

 certainly it is not inside, as that would be 

 incompatible with the known phenomena of 

 division. It may be removed by hj-drofluoric 

 acid, leaving the basement membrane in 

 situ ; but this proves nothing. The proba- 

 bilities are that the substance of the mem- 

 brane is imbued with it. 



In regard to the cell-contents of the Dia- 

 tomacese, there exists a layer of protoplasm 

 forming a primordial utricle, inside the cell- 

 membrane, and enclosing the rest of the 

 contents. In the coloured substance con- 

 stituting the mass of th^ endochrome the 

 chlorophyll is obscured by a yellow colour- 

 ing-matter, diatomine or phycoxanthine ; it 

 takes a green or greenish-blue tint with 



