4i6 THE BEGINNINGS OF II FE. 



passed through one or other of various changes. Occa- 

 sionally two or three of them, after having very slightly 

 increased in size, became ovoid and rather paler at 

 one extremity, before protruding a flagellum and 

 moving about actively as green Monads. Other of the 

 ovoidal corpuscles continued to increase in size, at the 

 same time becoming more and more fusiform, whilst 

 their green contents became granular (g-h'). The 

 most elongated of these were subsequently bisected 

 by a delicate partition; they also developed a greenish 

 nuclear-like body in each segment, and soon began to 

 grow into unmistakeable filamentous Desmids, of which 

 many otherwise similar specimens were seen in all 

 stages of growth {h'-k). But other representatives of the 

 minute ovoid corpuscles assumed a paler colour, and 

 then a slightly olive tint, whilst their colouring matter 

 became in part metamorphosed into two comparatively 

 large, rounded, nuclear corpuscles. These bodies in- 

 creased in size, and it soon became obvious that 

 they were young Naviculis (/,/')■ The exact pattern 

 assumed in the early stages is subject to much varia- 

 tion, and several different kinds of Diatom.s seemed 

 to be produced corresponding to these different initial 

 forms [m^ ?n). At first no striation was observable, 

 but gradually their envelope became more and more 

 differentiated — silica appearing to be assimilated from 

 the water in which they were immersed — and some of 

 these Diatoms exhibited a well-marked striation. 



Occasionally the individualizing contents of one of 



