PROTOZOA AS CELLS 45 



and by the septum across its upper end, where central fibers arise. 

 Soon after liberation, the flagellum lifts from the cell surface, the 

 membrane closing around it and over the cell below it. This 

 emergence appears to occur progressively from the distal tip 

 toward the base, so that intermediate stages can be recognized in 

 which the basal part of the flagellum lies within a slight ridge on 

 the cell surface. It would be interesting to learn when the first 

 movements of the flagellum can be detected in this cell. 



A very similar observation, again by Manton (1959a), concerns 

 the new flagellum in dividing cells of Mieromonas pusilla. In a set 

 of serial sections, the kinetosome and a short length of flagellum 

 are clearly seen beneath the cell surface ; only the two central fibers 

 project as a minute membrane-covered finger. Emergence of the 

 flagellum must involve a movement of the whole structure 

 toward the surface, or a shrinking back of the cytoplasm around it. 



Most interestingly, examples are known of non-motile cilia, 

 associated with sensory structures in metazoa, in which no central 

 fibers appear at all. This is true of the cilia connecting the inner 

 and outer segments of the rod and cone cells of the vertebrate 

 eye (DeRobertis, 1956), including the third or pineal eye of 

 reptiles (Eakin and Westfall, 1959). The kinetosome here is 

 distinguishable as a more compact proximal segment of the 

 fibrous cylinder. During rod cell development (DeRobertis, 1956 ; 

 Lasansky and DeRobertis, 1960; Tokuyasu and Yamada, 1959), 

 the slender cilium grows out from the primordium of the inner 

 segment, quite like a free cilium, except that central fibers are not 

 present ; the bulk of the cilium rapidly expands to form the distal 

 segment of the rod. On the other hand, in another instance of a 

 receptor-associated flagellum, the invaginated second flagellum 

 of Chromulina psammobia (Rouillet and Faure-Fremiet, 1958a), the 

 flagellum is complete with central fibrils. 



That animal centrioles are constructed on a plan closely similar 

 to that of kinetosomes, as had long been suspected by light 

 microscopists, was first shown to be true by de Harven and 

 Bernhard (1956) for a number of kinds of vertebrate cells and has 

 subsequently been demonstrated in innumerable animal species, 

 invertebrate as well as vertebrate. Nine skewed triplet fibrils 

 show very clearly in a centriole in a chick spleen cell pictured by 

 Bernhard and de Harven (1960) and interconnections between 



