392 F. W. GAMBLE AND FREDERICK KEEBLE. 



inclusions are visible. It is therefore probable that they act 

 as reserve materials. The figures 8, 12, 13, and 19 illustrate 

 the changes which accompany the formation of the green cells 

 from colourless ones. 



In addition to the structures we have mentioned, the 

 occurrence of a curved crimson band, an eye-spot, or "stigma," 

 is noteworthy, and has only previously been observed by 

 Famintzin in the green cells of Stentor. It is placed on the 

 surface of the colourless cells, and persists for a long time after 

 the green colour is established (PI. 31, figs. 8 C, 19 B, 20). This 

 little stigma, so similar to the organ of Volvocinea3 and many 

 other Chloropliycea3, and so utterly unknown in animal tissue 

 cells, is in itself almost sufficient to establish the immigrant 

 character of the colourless and green cells. 



5. The Influence of Light. 



The first appearance of chlorophyll in the colourless cells 

 is influenced to a remarkable extent by the conditions of 

 illumination, though, when once fully established, it is 

 extremely stable, and in the majority of chloroplasts is but 

 little affected by continued darkness or dim light. A very 

 bright light (direct sun) is required to quickl}^ effect the 

 production of chlorophyll in such amount as to be noticeable 

 to the eye under high microscopic powers. The change may 

 then be produced in twenty-four hours. Otherwise', in diffuse 

 light, sufficient chlorophyll may not appear in the colourless 

 cells to produce a green effect even after a fortnight. Like 

 every other feature in the development of the green cells, the 

 influence of light is variable ; some Convoluta become 

 rapidly green in diffuse light, and others, to all appeai'ances 

 as favourably situated, remain colourless. The exj^lanation of 

 this variability appears to be that Convoluta makes a pure 

 culture from a mixed infection. The infecting organism is 

 ingested in all stages, and some of these stages develop less 

 readily into the grccn-cell condition than do others. 



It isj indeed^ known that one and the same alga exhibits 



