A REVIEW WITH HYPOTHETICAL CONCLUSIONS. 109 



to drive them away from favourable breeding grounds. It does it, moreover, 

 without wasting the blood of its hosts, to support the individuals destined thus 

 to be sacrified. 



Whether the hypothesis is eventually supported by the facts or not, a more 

 beautiful example of facultative natural control by this highly important and 

 most beneficent of all agencies would be difficult to conceive. 



Possible economic application of female migration hypothesis. 



This is wholly dependant on its being established in the first place ; upon the 

 actual conditions found to exist in any particular fly-infested region in the 

 second ; and finally, and most important of all, in the natural resources awaiting 

 and ripe for economic development in that region. In consequence, any discus- 

 sion of the subject is pure speculation and nothing else, worth less than nothing 

 except as it may be suggestive of lines of research leading to results capable of 

 direct economic application. 



The hypothesis itself, if soundly based, ought easily to be capable of estab- 

 lishment. The actual conditions prevailing in a given infested area, as well as 

 the maximum migrating range of the fly, could then be secured without very great 

 outlay by a systematic survey of the fly spots and determination of the sex 

 disparity prevailing. Localities in which males predominate on an average 

 throughout the year (it would be much better to have figures for several years) 

 would then be held favourable for the fly ; where females predominate, the 

 opposite. That is to say, the aggregate control over increase exerted by 

 unfavourable breeding grounds^by such natural enemies as birds, ants, Bemhex, 

 etc., by unsuitable or irregular food supply, by unfavourable conditions as regards 

 humidity, shelter from the sun, etc., would permit increase in one locality and 

 not in the other. 



The relative suitability of different localities would, moreover, be roughly 

 indicated in the manner already used in the attempt to controvert Kleine's 

 argument that the islands were unfavourable breeding grounds. Thus if the 

 males averaged 80 per cent, of the total, year in and year out, the reproductivity 

 of each female would stand at 400 per cent, or four pairs ; if 75 per cent, were 

 males, at 300 per cent. ; if less than 50 per cent., there would be less than 100 per 

 cent, reproductivity, and a proportionate decrease in abundance. For example, 

 47 per cent, of males, as at the locality near Jinja where Carpenter noted the sex 

 of 12,000 flies over a period of some seven months, would indicate a reproduct- 

 ivity of 90 per cent, according to this method of approximating it. This, if 

 correct, would mean that unless some other agent than migration, and like it, 

 exerting facultative control, were in part responsible for conditions so unfavourable 

 to increase, the species would decrease in the last-mentioned locality at the rate 

 of 10 per cent, of its prevailing numbers for each generation of the fly, and that 

 it would finally cease to exist, provided immigration was prevented. 



If this could be established as a feature in the bionomics of Glossina, it would 

 seem to have great potentialities for economic application, especially in connection 

 with a campaign for the suppression of the fly through clearing methods. 



Another fertile field for bionomic research would be opened up by the 

 establishment on a firm basis of the female migration hypothesis. There has 



