328 KOFOID AND SWEZY. 



this view. The second is that there are among flagellates both 

 methods of origin of the axostyle in fission. On purely a -priori grounds 

 that is improbable, though possible. The third is that the two struc- 

 tures, the axial rod of the Trichonymphidae and the axostyle of the 

 trichomonads are not homologous at all, but are morphologically 

 different structures of different origins. The mass of accessory 

 evidence is against this view\ Investigation of the trichonymphids 

 on this point is needed to settle this problem. 



JoUos (1911) finds that in the division of Monocercomonas cetoniae 

 the old axostyle disappears and new ones arise by division of the basal 

 graimle, presumably from the central spindle between the daughter 

 granules. 



Martin and Robertson (1911) do not find in Trichomonas (= Tetra- 

 trichomonas) gallinarum, Trichomonas eberthi, and Trichomastix gal- 

 linarum that the paradesmose becomes the axostyles of the daughter 

 cells. They note particularly, however, that this structure "can 

 hardly have played the important role in nuclear division" which 

 other workers have ascribed to it. The old axostyle according to 

 their observations fades away "by some process of solution starting 

 from the anterior end" and axostyles of the daughter cells are in 

 some unexplained way re-formed. 



Alexeieff (1912) in a discussion of his own results and those of 

 Dobell (1909) and Martin and Robertson (1911) admits the possi- 

 bility of several methods of origin of the axostyle and suggests that 

 the method by which the blepharoplast gives rise to various organelles 

 is of no consequence. The primary question, however, is whether 

 or not the several methods are all correctly interpreted. 



In a later paper (1913) he figures Trichomonas augusta in several 

 stages of mitosis labeling his figure VII, b-d, " trois stades de division; 

 la blepharoplastodesmose axiale on centrale qui s'etend entre les deux 

 blepharoplastes-fils devient I'axostyle (en d) des deux individus-fils." 

 His series is incomplete and does not show the derivation stated for 

 the axostyle. The paradesmose in his figure c is chromatic, and in d 

 achromatic, and the connection between the two quite hypothetical. 

 We interpret his figure c as a phase of multiple fission approaching 

 the second mitosis, and his Figure d as binary fission with parades- 

 mose apparently gone and the two daughter axostyles meeting end 

 to end, but represented as continuous, comparable with our Plate 4, 

 Figure 39, as to stage and arrangement of organelles. His actual data 

 as far as figured are not, except for the continuous structures inter- 

 preted as daughter axostyles in his figure d, contradictory to our 

 results. 



