Bacillariece 267 



occurring in the larger species of Navicula (Pinnularia), even 

 though the movements of the frustules, the structure of the valves, 

 the system of fissures and pores, and the arrangements of the 

 protoplasmic currents are the same. In some of the small species 

 of Navicula the movements are extremely active, comparing not 

 unfavourably with the slow movements of certain of the Infusoria. 

 If the movements of such Diatoms be due to the secretion of 

 gelatinous material, then the amount secreted in a short space of 

 time must be relatively enormous. 



After a careful consideration of the facts which have been 

 brought forward with a view to the elucidation of this most 

 interesting problem, there appears to be no doubt that the move- 

 ments are connected with the raphe, and the balance of evidence 

 indicates that in some Diatoms at least they are due to an exuda- 

 tion of mucilage. 



The multiplication of Diatoms takes place by successive 

 bipartitions, each division resulting in a gradual reduction in the 

 size of the individuals. A slight increase in the volume of the 

 frustule is the first appreciable change, after which the nucleus 

 divides karyokinetically. A division of the cell-protoplasm now 

 takes place, and a new siliceous valve is formed over each divided 

 surface. These new valves are situated within the girdle of the 

 original frustule, and the connecting-bands of the new valves are 

 soon developed, sometimes making their appearance before and 

 sometimes after the separation of the individuals. Thus, each 

 individual consists of a new valve and an old one, the connecting- 

 band of the old valve overlapping that of the new valve. Some- 

 times the cells do not separate, but remain in contact after division, 

 successive bipartitions resulting in a chain of individuals. Owing 

 to the formation of a pair of new valves within the girdle of the 

 old ones, and since the cells when once formed are incapable of 

 growth, every succeeding generation becomes reduced in size by 

 the double thickness of a connecting-band. This statement is 

 not strictly true, however, in the case of some of the filamentous 

 species, and possibly in many others. It has been shown that 

 daughter-cells are often produced of larger size than the parent- 

 cells, such daughter-cells being recognizable by the thickened rim 

 of the" valves. This fact has a retarding influence on the diminu- 

 tion o'f the size of the cells, the reduction in size not being in 

 strict proportion to the number of bipartitions ; and, concerning 



