684 H. M. WOODCOCK. 



end. It generally happens that, as the split extends, the 

 parasites tend to separate from one another, turning out- 

 wards, away from each other as it were (figs. 92, 93) ; eventu- 

 ally the two daughter-individuals come to lie in one line (which 

 may be more or less curved), with the flagella, waving freely 

 at opposite ends, the parasites only remaining connected by 

 what is actually the still undivided aflagellar end (figs. 

 94, 95). The fact, therefore, that we may find either equal 

 or sub-equal cytoplasmic division in which the daughter- 

 flagella differ considerably in length, explains the great 

 variation in this respect which is met with amoug the 

 ordinary trypanomonad individuals. 



In many cases the division of the two nuclear bodies does 

 not take place in a direction quite transverse to the long 

 axis of the body, but in an oblique direction, one pair of 

 daughter-nuclei lying somewhat nearer to the aflagellar end 

 than the other pair. In this manner are produced forms 

 such as are seen in figs. 8, 97, and 100. These individuals 

 in which the nuclei have progressed into the aflagellar 

 part of the body have the undulating membrane very well 

 developed; it may be said that the trypanomonad condition 

 is here accentuated. In such forms of the parasite the 

 mode of division is also distinct, being markedly unequal in 

 character (figs. 98-100). The two resulting individuals are not 

 of the same type (cf. figs. 12, 103, and 107). One, the larger 

 parasite, is of the same type as the parent individual, and 

 possesses fi-om the first a conspicuous membrane, but the 

 other, the smaller daughter-individual, is at first pear-shaped 

 and stumpy, and has only a short, inconspicuous membrane. 

 This mode of division presents a general resemblance, it will 

 be noted, to one of the types of division characteristic of 

 T. lewisi. Indeed, in the present case, the process might 

 also be regarded as a " budding-off " of a daughter- 

 individual from the parent. I have never observed, how- 

 ever, more than one bud formed, i.e. the process appears 

 always to retain its character of binary fission and never 

 to be of the multiple type. AVhen set free, the smaller 



