6 9 



in Hong Kong : (i) clearing all long grass and undergrowth, and then 

 (2) destroying all breeding places by draining, etc. Alike from obser- 

 vation and experiment, I would place a cleared area as of immense 

 importance ; no mosquitoes will fly far without vegetation, and old 

 residents in West Africa have by experience found this out, and 

 regularly burn the bush, and this is over and over again emphasised in 

 various localities in Hong Kong ; even a road intervening made all 

 the difference to the health of a community. 



Extensive experiments have been made as to the destruction of 

 mosquitoes and their larvae by the " natural enemies of mosquitoes," 

 dragon flies, tadpoles, small fish ; the whole subject seems to me of 

 little practical importance. I have repeatedly put 20 Culex larvae and 

 20 Anopheles , collected from different pools, together. The effect was 

 astonishing. Within twenty-four hours there was no trace of the 

 Anopheles larvae, and like Pharoah's lean kind the Culexdid not seem 

 any the fatter for having absorbed them ; indeed, no one can realise 

 the intensity of the struggle for existence which goes on in a stagnant 

 pool until he forces himself with a large hand lens to watch the various 

 forms of life with which the water is swarming. Culex larvae, then, 

 are an important factor explaining why in many places where mos- 

 quitoes are more numerous there is no malaria, the weaker Anopheles 

 being overcome in the struggle for existence. 



CONCLUSIONS. 



These results, and the details of other places, lead me to the 

 following conclusions : 



1. That malaria, in every place examined, was invariably associated 

 with Anopheles breeding pools in the immediate vicinity. 



2. That in no case were Anopheles found in the neighbourhood of 

 houses without malarial fever manifesting itself repeatedly, that is, 

 where Anopheles are found, there is fever, and in every district where 

 there is marked malaria there the blood of children examined con- 

 tained the malarial parasite, living Anopheles were caught and their 

 breeding pools localised. 



3. That in no case was there a greater distance from the breeding 

 pools to the houses infected than 1 50 yards. 



4. That the breeding pools were always localised and compara- 

 tively easily treated by surface draining. 



