194 ROMANCE OF LOW LIFE AMONGST PLANTS. 



with b. We have now two perfect cells, one of which 

 is smaller than the other, by the thickness of the 

 band. In the free forms the cells usually separate, 

 but in many genera the cells continue attached, thus 

 forming a filament or chain. This process goes on 

 until the productive energy is destroyed, but the 

 number produced by the parent cell is probably very 

 large. We have many times measured the thickness 

 of this band, and never found it exceed the ten- 

 thousandth of an inch in thickness. 



"The shape of the diatom cell is very variable: they 

 occur circular, oval, triangular, square, five to eight 

 or ten-sided, boat-shaped, wandlike, sometimes curved 

 in opposite directions, sometimes only in one. But 

 the structure is always the same, viz. the two lids 

 and the band connecting them. The former are 

 usually termed valves^ and the latter the connecting 

 zone, or cingidimi. It is from the form of, and the 

 markings on, the valves that the generic and specific 

 distinctions are principally constituted. These mark- 

 ings are generally of great beauty, and usually require 

 the highest microscopic powers to show all their 

 details, particularly those forms on which the mark- 

 ings appear as fine lines. One species in particular 

 has lines so close that as many as ninety-four in the 

 one-thousandth of an inch have been counted, and 

 by certain arrangements of the light these lines are 

 found not to be continuous, but composed of minute 

 dots. The nature of these markings has always been 

 a disputed question, some asserting that they are 

 elevations on the surface, whilst others have been 

 equally positive that they are minute cells in the 



