N. H. SWELLENGREBEL AND J. M. H. SwELLENGREBEL-DE GrRAAF 189 
that M. barbirostris is comparatively negligible as a cause of severe endemic 
or epidemic malaria. 
This method of comparison, in neighbouring regions, of the parasitaemia 
and of the spleen-rate of the inhabitants, with the Anopheline fauna, enables 
us to ascertain the presence or absence of a correlation between the occurrence 
of a severe malaria and that of a particular species of Anopheline, and leads 
us to the determination of the indirect index of injectivity of this species, or 
shortly “indirect index” (I. I.). Taken alone, conclusions drawn from the I. I. 
are open to adverse criticism, but if the N. I. is likewise known, the conjoint 
data will furnish information as to the actual amount of damage done by the 
species. 
Taking into consideration the figures given in Tables III and IV, together 
with the determinations of the 1. I. for each species, we may summarize the 
facts as follows: 
1. M. ludlowi is the most important vector of malaria in the Malay 
Archipelago, and shows the highest degree of N. I.; its importance is enhanced 
by the following propensities: the adult shows a preference for the interior 
of human dwellings; it usually occurs in large numbers; and it is able to cover 
by flight considerable distances. We have observed (1918 c) that large numbers 
of adults in nature habitually cover distances of 1*6 km., as measured in the 
direct line from the place of capture to the nearest breeding-place. We 
excluded all instances in which there was not absolute certainty of the absence 
of nearer breeding-places, and, moreover, all cases were excluded where only 
a fewAnophelines could be found in the place of capture. Some small breeding- 
places might after all have been overlooked, and a few Anophelines might 
have been derived from this hypothetical breeding-place, but a number 
amounting to several hundreds could not be accounted for in this way. The 
majority were certainly derived from the extensive breeding-places under 
observation. 
In two cases we succeeded in infecting Europeans with subtertian, by 
exposing them to the bites of M. ludlowi, which had been fed on a crescent- 
carrier twelve days previously. Ourselves being the subjects of this experi¬ 
ment, we are assured of the absence of any previous malaria infection. 
Moreover, the experiment was performed in a locality some 12 km. from the 
coast, where M. ludlowi was absent and malaria infection rare. 
With few exceptions we found severe malaria to prevail wherever M. ludlowi 
was present. 
2. M. aconita (including M. minima). Some doubt exists as to the 
ability of this species to transmit malaria. Winoto (1918) found it to possess 
the highest N. I. (7 per cent.) of all the species in very malarious hill-districts. 
We found it (1919 b) to be the cause of a rather severe epidemic in a pre¬ 
viously non-malarious district, and exhibiting an N. I. of 2 per cent. In the 
highly malarious regions of Mandailing and Soendatar, however, we failed to 
