THE ELEMOSPORIDIA 365 



ordinary multiple type, taking place in the blood-corpuscles, in H. tropiduri 

 and H. diploglossi. According to Dobell, however, the schizogony of 

 H. simondi consists simply of binary fission as a rule, sometimes of division 

 into four. The male and female gametocytes, sharply differentiated by their 

 staining properties in this as in other species, are stated also to have the 

 nucleus divided into two when mature ; Woodcock (687), however, disputes 

 the correctness of Dobell's interpretations. In no case as yet is the inverte- 

 brate host of any Hoemocystidium known. 



2. The Halteridia. The characteristic form of parasite in this 

 section, only known to occur in the blood of birds, is an organism 

 which is found within the nucleated red corpuscle, and which does 

 not displace the nucleus of the corpuscle, but grows round it into a 

 halter-like form, whence the name Halteridium given to it by 

 Labbe. Hence the parasite is easily distinguished from Proteosoma, 

 which is more compact in form, and which displaces the nucleus of 

 the corpuscle. Halteridium is amoeboid, but the form-changes 

 are generally slight ; it produces the characteristic melanin-pigment 

 in abundance ; and when the blood is drawn, " exflagellation " of 

 the ripe male sporonts takes place very readily. Not merely the 

 gamete-formation, but the subsequent fertilization and the for- 

 mat on of the ookinete, can be observed on the slide. It is in this 

 form that Macallum first followed out the whole process, and so 

 made clear the true significance of the " Polymitus " stage in the 

 malarial parasites. 



The correct generic name for the HaUeridium-pa,Ta,site is believed 

 to be Hoemoproteus. Labbe considered the halteridia of different 

 birds to be all one species, to which he restricted the specific name 

 danilewskyi (Grassi and Feletti). By other naturalists several 

 species have been distinguished and named after the birds in which 

 they occur, as H. noctuce of the little owl, H. columbce of pigeons, 

 etc. The halteridia of different birds show considerable differences 

 in form, structure, and appearance, and there can be no dou bt that 

 there are many species of these parasites ; but it by no means 

 follows that a given species is restricted to a particular host. It is 

 probable that in some cases one and the same species may be 

 capable of infecting several species of avian hosts. The Sergent 

 brothers were unable, however, to infect canaries with H. columbce 

 of pigeons. 



The life-cycle of these parasites has been the subject of con- 

 flicting statements. We shall consider first the type of develop- 

 ment made known by the Sergent brothers (686) in part, and more 

 fully by Aragao (Fig. 157). The development described by Schau- 

 dinn (132), to which the utmost doubt attaches, will be dealt with 

 later (p. 390). 



The invertebrate host of H. columbce is a biting fly of the genus 

 Lynchia, of the dipterous family Hippoboscid<t. These flies, though 



