lO SOCIÉTÉ BELGE DE M1CROSCOP1E. 



mose movement, scarccly merits separate mention. Et 

 is of too ambiguous a nature to furnish data for the 

 motile agency observed in Diatoms. 



The movements of the endoehrome referred lo bv 

 Nâgeli, are necessarily confined to the interior of the 

 valve, and althongh globules are occasionally seen to 

 move languidly along the marginal portions, in some 

 species, they are never seen to pass to the extremity, 

 nor make a complète circuit of the frustule. Nothing 

 approaching to cyclosis, as seen in Desmids, takes place 

 at the extremities of Diatoms. Siebold's theory is equally 

 unsatisfactory. 



By attentively following a Diatom, under high power 

 magnification and careful illumination, its movements 

 are seen to be under its own control. It will attack a 

 body relatively larger than itself ; it will force the sharp 

 or taper end of its frustule into a mass of matter, and 

 receed from it, with a jerky motion. This action it will 

 repeat many times over or until it bas opened a way for 

 itself. Sueh movements will be explained by the alter- 

 nate extension and retraction of a delicately constructed 

 préhensile organ, or organs, contractile préhensile fila- 

 ments, protruding through an opening or operating on 

 the external surface of the siliceous frustule. It is but 

 necessary to hâve préhensile filaments capable of exten- 

 sion, in the transverse direction of each half of the 

 frustule, to which they may be attached, to further coirï- 

 prehend the foi'ward and backward movements perform- 

 ed by the Diatom. For a nearly parallel example I may 

 refer to the pediculate Rotifer : the pedicle of which 

 consists of a highly contractile spiral style. This the 

 Rotifer alternately expands and contracts at will ; and 



