INTRODUCTION. XIX 



The complex structure of the siliceous valves which enclose 

 the frustules of the Diatomaceae has also been regarded as in- 

 compatible with their location in the vegetable kingdom. Much 

 of this complexity is apparent rather than real, and is due to 

 the refractive nature of the material which enters into combina- 

 tion with the outer membrane of the cell- wall. In every case 

 this membrane is more or less penetrated or imbued with silex ; 

 and the presence of this substance appears to have modified the 

 intimate structure of the membrane, and induced great variety 

 in the mode and character of its formation in different genera 

 accompanied by great regularity in the individual species. 



These variations exhibit themselves in the different modifica- 

 tions of structure which constitute the markings of the valves, 

 appearing under the form of ribs and nodules, costae, striae, or 

 cellules of an elliptical, circular, or hexagonal outline. A wide 

 comparison of specimens seems to me to prove that these various 

 markings originate in the tendency impressed upon all organized 

 structure to develope itself upon the type of the cell, and that 

 the presence of the siliceous constituent in the cell-membrane of 

 the Diatom gives a fixedness to this tendency, which, in ordi- 

 nary cases, is either not discernible in the structure of the mem- 

 brane, or whose effect is obliterated by the coalescence of the 

 softer material which constitutes its substance. However this 

 may be, it appears to me certain that the structure of the sili- 

 ceous valve in the Diatomaceae is invariably cellulate, the cellules 

 being more or less modified according to the peculiar require- 

 ments of each species, and that no other explanation of their 

 characteristic markings seems consistent with the facts which are 

 established by a careful examination and comprehensive know- 

 ledge of Diatomaceous structure. That this explanation does 

 not involve considerations at variance with the conditions of uni- 

 cellular vegetable life will be obvious to any one familiar with 

 the structure of the siliceous epiderm in the JEguisetaeece and 

 Graminacece, and the distinctly cellulate structure of many 

 pollen-grains ; while this very presence of silex as a constituent of 

 the cell-wall in the Diatomaceae appears to be wholly unaccount- 

 able except on the supposition of the vegetable nature of these 

 organisms. In no instance do we find a parallel condition in 



b2 



