102 WILLIAM F, FISKE — THE BIONOMICS OF GLOSSINA \ 



must feed freely and without prejudice to themselves on birds or reptiles 

 (monitors or crocodiles) or on both. 



Dr. Duke, in a paragraph in one of his papers in No. XII of the Reports of 

 the Sleeping Sickness Commission, specifically notes that in the open the flies 

 were observed to be attracted in large numbers to a monitor,* but that in the 

 laboratory they did not feed well on this reptile. Some observations are in 

 themselves sufficient to cast some doubt on the validity of Kleine's results, in their 

 broadest application, and to suggest the desirability of further research. 



It has been observed that in captivity the flies gorge themselves with 

 difficulty on reptilian blood, which might be held as indicative of a sluggish 

 peripheral circulation. Perhaps the peripheral blood of a crocodile exposed to 

 the full effect of a tropical sun flows faster, thus permitting quicker engorge- 

 ments. Perhaps if the difference between avian and reptilian blood is largely one 

 of temperature, it may disappear (so far as peripheral blood is concerned) when 

 the reptilian host is exposed to the sun. Perhaps, even, the flies need to expose 

 themselves to radiated heat after engorgement with cold blood, in order that 

 digestion may properly be effected. At all events there are differences enough 

 between laboratory and field conditions to make further and more exhaustive 

 research desirable in an instance like this, where field observations do not 

 support laboratory conclusions. 



Carpenter assumes that some natural enemy attacking the female more freely 

 than the male must occur on the islands and not on the mainland ; at the same 

 time he does not suggest what it may be. Aside from the need of providing 

 some other enemy equally prejudicial to the males in places where the females 

 predominate, this theory is against precedent, so far as the writer is aware, in the 

 world of insects. There are precedents enough for a greater proportionate 

 mortality among individuals of one or the other sex, but invariably, so far as any 

 instances of this sort come to mind, the phenomenon is 'preceded by a marked 

 segregation of the sexes. Segregation as a preliminary is thus inferentially 

 necessary and segregation is what we are trying to explain. It is true that the 

 females are found to adopt a somewhat different mode of life from the males, but 

 their feeding habits appear to be markedly similar, and Carpenter himself has 

 followed them to their breeding grounds and observed their habits at the time 

 of parturition without discerning any enemy lying in wait for them there. 



In short, the simplest and most logical of all explanations, which (because of 

 its very simplicity, perhaps) seems to have been overlooked, appears to fit by far 

 the most accurately all the facts in the case : viz. the the migration of either the 

 males or the females to or from localities where males predominate. 



Of the mutual alternatives the writer is well satisfied that the females are the 

 migrants. There are several reasons for adopting this view. 



Obviously, for one, Damba (and not improbably other islands) is actually a 

 very favourable breeding ground, since puparia are more readily collected there 

 than on the mainland.! On this account it would seem that migration of females 



* [A similar observation has been made more recently by Dr. G. D. H. Carpenter. — Ed.] 

 f It is the source of supply of some 2,000 or 3,000 puparia monthly for the laboratory at 

 Mpumu. 



