TRICHOMONAD FLAGELLATES. 321 



augusta. There seems to be a subsequent but temporary telosynaptic 

 fusion of the split chromosomes at the time of the first appearance of 

 the equatorial plate (PI. 2, Fig. 20). 



Metaphase and Amphiaster. 



No arrangement in an equatorial plate occurs until the daughter 

 blepharoplasts have migrated to the poles of the ellipsoidal nucleus 

 (PI. 2, Figs. 20-21). Such a plate is suggested in Figure 13, but only 

 four pairs are here in line, and no spindle is present to contain the plate. 

 This is only a chance grouping and not a true equatorial plate. 



Amphiaster Stage. 



The equatorial plate is formed apparently only after all of the 

 chromosomes are in this "telosynaptic" relation (PI. 2, Figs. 20, 21). 

 In the few spindles we have found in this stage there are but five 

 chromatin masses, rather stout (Fig. 20) or tapering (Fig. 21) towards 

 either end, thickest at the middle, and showing there no trace of any 

 central zone of fusion or separation where from previous conditions 

 we may suppose "telosynaptic" fusion to have occurred. There is 

 for this condition of the chromosomes, if it be the normal and regular 

 sequence following that of parallel split chromosomes, but one expla- 

 nation, namely that the early splitting is followed by a later "telo- 

 synaptic" fusion in the equatorial plate. The relations of the daughter 

 chromosomes in Figures 12 to 19 support this interpretation. Coupled 

 with this is the fact that the individuals with split chromosomes 

 present collateral evidence in other organelles of an early phase of 

 mitosis, whereas this evidence in Figures 20 and 20a, indicates that 

 these are later stages. There is, however, another possible sequence, 

 namely that in some individuals represented by Figures 20 and 21 

 there has been no previous splitting and the elongated chromosomes 

 part transversely. If this be true we have two types of division of 

 chromosomes in the one species and one of these is of the normal 

 type by longitudinal splitting and the other by an exceptional and 

 unique type (in non-maturation divisions) of transverse division. 

 This is possible especially in view of the fact that we are dealing with 

 parasitic organisms subject to the action of the antitoxins of their 

 hosts and therefore to disturbances in the normal processes of growth. 



