268 University of California. Publications in Zoology [Vol. 16 



each. In the last case all four flagella ordinarily project from 

 the same end, but three cases have been found in which they come off 

 from nearly opposite ends, making the flagellate somewhat triangular 

 in shape. One of these was watched until it turned back into an amoeba 

 and it proved not to be a division stage, for it was .still binucleate as 

 an amoeba. 



(a) Process of enflagcllation. — In enflagellation there is a progres- 

 sive change in form of the individual which may be slightly modified. 

 Differing from the characteristic liraax form, the amoeba first has short 

 pseudopodia which are constantlj' being withdrawn. It gradually 

 rounds up either before (pi. 22, figs. 92. 93) or after the flagella are 

 formed (pi. 22, fig. 94). The flagella appear in the first individuals in 

 about an hour, depending on the temperature, and with their appear- 

 ance a side-to-side motion at the anterior end begins and increases 

 until the wbole flagellate is involved. Finally an oval flagellate swims 

 away with a movement that is at first uncertain in appearance, but 

 soon becomes the characteristic side-to-side, forward spiral flagellate 

 motion and its shape is then pyriform. 



In a preparation fixed when flagellation is beginning the karyosome 

 of some of the individuals has a .small projection at one end and is 

 slightly elongated (pi. 22, fig. 91), differing from a reorganization 

 nucleus only in the absence of peripheral chromatin. Later it is 

 rounded and at the nuclear membrane there is a dark-stained granule 

 which is connected with the karyosome (pi. 22, fig. 92). This granule 

 in others is found to be located at varying distances from the nucleus 

 to the periphery but always with a plastin connection with the karyo- 

 some (pi. 22, fig. 93). No stage has been foimd in which it is at the 

 periphery without flagella being present. 



Individuals are found with short, blunt flagella and the nucleus at 

 a distance from their bases (pi. 22, figs. 94, 99"), individuals with 

 nuclear membrane pulled out in the direction of the blepharoplast (pi. 

 22, fig. 99), and individuals with long flagella and nucleus close to the 

 blepharoplast (pi. 5, figs. 95, 98). The increase of flagellar length and 

 coincident movement of the nucleus to a position near the blepharoplast 

 has been observed in living material. 



Judging from these conditions the basal granule comes out from 

 the karyosome and is always connected with it. It may be derived 

 from the centriole or it may be budded off from the karyosome around 

 the centriole. Our data do not enable us to determine this. It takes 

 its position at the periphery of the cytoplasm and coincident with that 



