THE HvEMOSPORIDIA 389 



than the other Telosporidia, are allied specially to the Haemoflagellates, more 

 so than to other Flagellata ; whether, in short, the Haemosporidia should 

 be removed from the Telosporidia altogether, and should be classified, together 

 with the Hsemoflagellates, in one natural order, family, or other systematic 

 category. In dealing with the Hsemoflagellates in a previous chapter, cause 

 was shown for believing them to have two distinct lines of ancestry, the one 

 from a Cercomonad, the other from a Bodonid type of Flagellate ; in that 

 case it is the Cercomonad section that is to say, the trypanosomes and their 

 allies to which the Haemosporidia must be considered to be specially related 

 on the theory now to be discussed. 



Leger and Duboscq (646), recognizing distinct Bodonid and Cercomonad 

 stems in the Haemoflagellates, derive the Gregarines, Coccidia, and Haemo- 

 gregarines, from the Bodonid stem (trypano plasms), the Hsemamoebae and 

 Piroplasms from the Cercomonad (trypanosome) type. 



The close relationship of the Haemosporidia and the Coccidia 

 seems at first sight so obvious, from a general consideration of the 

 life-histories of typical members of each group, that any theory 

 to the contrary must justify itself by convincing and cogent argu- 

 ments. The chief grounds upon which affinities between Haemo- 

 sporidia and Haemoflagellates are alleged are found, when analyzed, 

 to be of three kinds namely : first, developmental data ; secondly, 

 structural that is to say, mainly cytological peculiarities ; 

 thirdly, resemblances between certain forms which appear to be 

 sufficiently close to link the two groups together by a series of gradual 

 transitions. The evidences of affinity between Haemosporidia and 

 Haemoflagellates based on these three classes of facts must be 

 considered separately. 



1. Developmental Data. Beginning with the first of the five types of 

 Haemosporidia which have been recognized above namely, the haemamcebae 

 or malaria] parasites, it is very evident, as Schaudinn (658) first pointed out, 

 that their life-cycle resembles in the closest manner that of the Coccidia. 

 With one exception, every phase in the life-cycle of a malarial parasite has a 

 corresponding phase in that of a coccidian, and the same terminology can be 

 used throughout for describing the stages of the development ; the one ex- 

 ception to this statement the only phase that requires a special name is 

 the ookintte-stage of the malarial parasites, which is not known to occur 

 in any coccidian. It is clear, however, that the points in which the life- 

 cycles differ from one another in the two cases are such as can be correlated 

 with the differences in the mode of parasitism that is to say, with the fact 

 that in Coccidia, speaking generally, there is a single host, and the mode of 

 infection is contaminative, while in the haEmamcebae there are two hosts, 

 and the vertebrate is infected by the inoculative method. Corresponding 

 with this difference, the zygote in the Coccidia prepares at once for leaving 

 the body of the host and passing out into the open, and protects itself by a 

 firm envelope ; while that of the haemamcebae, produced in the body of an 

 intermediate host, does not encyst itself, but is actively parasitic, continuing 

 to absorb nourishment from the host and to grow. Further, in the hsemamcebae 

 the parasite is always in the body of one or the other of its two hosts, and 

 consequently tough, impervious cysts and spores like those of Coccidia are 

 superfluous and are never formed ; the oocyst is a thin membrane through 

 which soluble foodstuffs can diffuse, and sporocysts are not secreted, as is the 

 case also in some Coccidia. The adaptive significance of these differences 

 is so obvious that it does not require further elucidation or discussion. 



The development of the halteridium-type, as described by Aragao, can be 



