parasite, keeping up the infection of the invertebrate host. In 

 hunger-periods the flagellates may persist as simple, rounded, 

 Leishmania-like forms. Sooner or later many, it may be the greater 

 number, but not all, of the crithidial forms become modified into the 

 trypaniform individuals, which represent (3) the propagative phase 

 of the parasite, and pass forwards to be inoculated into the verte- 

 brate host. Those crithidial forms which do not become trans- 

 formed into the propagative individuals remain to multiply and 

 replenish the stock. 



A very much debated question in this development is that relating to the 

 occurrence of sexual phases and syngamy, which, purely on the analogy 

 of the malarial parasites, are assumed almost universally to occur in the 

 invertebrate host. Not in a single instance as yet, however, has the sexual 

 act been proved satisfactorily to take place in the development of trypano- 

 somes. The fertilization described by Schaudinn (132) in " T. noctuce " is 

 the well-known conjugation of Halteridium, which can be observed without 

 difficulty ; and though Schaudinn described so-called "male " and " female " 

 types of trypanosomes in the mosquito, he expressly stated that they did not 

 and could not conjugate. The process of syngamy described by Prowazek 

 (497) for T. lewisi in the rat-louse, though " confirmed " by Baldrey (396), 

 Gonder (445*5), and Bodenwaldt, is almost certainly the agglomeration of 

 degenerating forms (Swellengrebel, 516 ; compare Keichenow, 78, p. 268). 

 Less biassed investigators, who have studied the developmental cycle of 

 trypanosomes with great care, such as Chagas, Robertson, and others, have 

 been quite unable to observe sexual processes of any kind. .The liability to 

 error in the interpretation of observations is greatly increased, first by the 

 fact that trypanosomes divide longitudinally and often unequally, secondly 

 by the phenomena of agglomeration (p. 128), which occur readily under un- 

 favourable conditions. Consequently the adhesion together of two trypano- 

 somes may be due to quite other causes than sexual affinity. In some cases 

 the alleged occurrence of syngamy has been based merely on the fact that 

 non- flagellated forms have been seen, which, on the analogy of the malarial 

 parasites (p. 362), are termed " ookinetes " and interpreted as zygotes. 



It is certainly remarkable, in view of the paucity of data, that so many 

 investigators, following Schaudinn's lead, should persist in ascribing all form- 

 differentiation in trypanosomes to sex, and should be unable, apparently, to 

 conceive of any other cause of polymorphism in parasites which have to adapt 

 themselves, in the course of their life-cycle, to a great diversity of conditions 

 (compare also Doflein, 430). It must be emphasized that the only true 

 criterion of sexual polymorphism is sexual behaviour, and until that has been 

 established it is premature to speak of sexual differentiation. 



Some investigators have upheld the unfashionable view that the syngamy of 

 trypanosomes occurs in the vertebrate host ; so Bradford and Plimmer, aud 

 more recently Ottolenghi, who has described in T. brucii, T. ecpiinnm, 

 T. gambiense, and T. equiperdum, the following process of sexual conjugation : 

 Two trypanosomes of very different size and appearance attach themselves 

 to one another by their hinder ends. One, regarded as the microgamete, is 

 more slender, and contains one trophonucleus or a larger nucleus of this kind 

 and two smaller (reduction-nuclei) ; the other, the macrogamete, is much 

 larger, and contains also a larger nucleus near the kinetonucleus and two or 

 more other nuclei in process of degeneration. The macrogamete also has 

 usually three, sometimes two or four, flagella and undulating membranes. 

 After the two gametes have united by their hinder ends, a small nucleus is 

 budded off from the principal nucleus of the microgamete, passes over into the 

 body of the macrogamete, and fuses with its principal nucleus. Subsequently 

 the microgamete appears to degenerate, and the fertilized macrogamete to 



