STUDIES ON AVIAN H.EMOrROTOZOA. 725 



1 admit, nevertheless, that I am now doubtful of such an 

 actual connection, especially since I have been working- at 

 Rovigno. I am more inclined to think that an intra-cellular 

 parasite may exhibit nuclear dimorphism, in certain conditions 

 or phases as a result of a close phylogenetic relationship with a 

 parasitic flagellate (say a trypanosome), without necessarily 

 being any longer ontogenetically connected with one. Put 

 into other words, this is to say that a parasite, such as 

 Halteridium, which shows this feature, is probably derived 

 from a trypanosome which has become adapted entirely to a 

 resting, intra-cellular condition, and has coincidently lost, 

 more or less completely, the ability to develop an active 

 trypaniform phase. 



Berliner, in a recent paper entitled " Flagellaten-Studien " 

 (1), has incidentally corroborated my account of the occur- 

 rence of nuclear dimorphism in Halteridium by describing 

 it in the case of H. nocture, i.e. in the very parasite in 

 which Schaudinn first maintained it was present. Berliner's 

 figures are very striking and interesting. His preparations 

 were stained with iron-hgematoxylin, and another most im- 

 portant point brought out by this method of staining is the 

 close correspondence between the structure of the (chief) 

 nucleus in the Halteridium and that of the trophonucleus 

 of a trypanosome. I need not dwell upon this point here, as 

 Professor Minchin and myself have already referred to it in 

 our paper (13), showing the essential difference which exists, 

 on the other hand, between the nuclear structure of a h«mo- 

 gregarine and of a trypanosome ; and we shall have more to 

 say about it in our own account of the parasites of Athene 

 n o c t u a. 



This fact furnishes, however, strong additional evidence in 

 support of the (modified) view of a close relationship between 

 Halteridium and the hffimoflagellates, which I am inclined 

 to pi'efer. On this view the gradual "Riickbildung " of the 

 kinetouucleus — which is associated principally, of course, with 

 the locomotor activities — can be readily understood, and is, 

 indeed, to be expected. It accounts, further, for the com- 



