202 University of California Publications in Zoology [Vol. 16 



In the publications of Woodcock, Berliner (1909), Hartmann, and 

 many others, this terra is applied to any fibril connecting two granules, 

 with little regard to the origin, function, or fate of the structure. In 

 fact, much of the proof for the so-called binuelear condition of the 

 Haemosporidia rests on the demonstration of this fibril, and a fair 

 example of its use is afforded by the memoir of Woodcock (1912). In 

 his plates 9 and 10, Woodcock has figured two granules which some- 

 times ai'e equal. When these are connected by a darkly staining line 

 it is termed the centrodesmose and a process of division is supposed 

 to have taken place. As noted above, to function as a true centro- 

 desmose it must take pai't in a division of the nucleus and cell, and 

 evidence of this is totally lacking throughout the entire series. Indeed 

 division of the Halteridium during the stage of the life-cycle figured 

 by Woodcock has never been known to occur. An elaborate process 

 of division, involving a centrosome, centrodesmose and central spindle, 

 has before this time not been found necessary to account for the 

 extrusion of a granule of chromatin from the nucleus. Hartmann and 

 JoUos (1910, pi. 10, figs. 5a, 7) figure the same condition as evidence 

 for the origin of the parabasal body by division of the nucleus in 

 Babesia and Plasmodium. Similar figures of Berliner (1909) of 

 Haemoproteus noctuae are cited by Hartmann as proof of the haemo- 

 flagellate affinities of that species. 



On the grounds stated above it is impossible to justify the use of 

 the terms centrodesmose and central spindle as applied here. It is 

 equally impossible to postulate that the mere chance appearance of a 

 granule, which is not constant in position or occurrence, is of great 

 significance from a taxonoraic standpoint. 



As Woodcock (1912, 1914) and Minehin (1912) have pointed out, 

 the relations of Halteridium are rather with the Coccidia than with 

 the Haemoflagellata, judging by the life-cycle as a whole, and by their 

 morphology. 



The linking of the Haemogregarina with the Haemoflagellata has 

 rested on the same sort of evidence as in the ease of Haemoproteus. 

 Minehin and Woodcock (1910) have pointed out the essential differ- 

 ences between the nuclear conditions of the haemogregarines and the 

 trypanosomes. Reichenow (1910) and Robertson (1910) have given 

 accounts of the life-cycles of these forms which agree in all essential 

 details, and which also separate these organisms very clearly from the 

 flagellates. The importance of considering the life-cycle as a whole 

 in the systematic placing of any organism has come to be a truism in 



