MALARIA AND THE COMMUNITY 
149 
anthropophilic index or better, human 
blood ratio, and is admittedly unreliable if 
based on selected groups of insects found 
resting in particular situations such as bed¬ 
rooms or stables, which may not be repre¬ 
sentative of the anopheline population at 
large. Furthermore, engorged insects of 
certain important species are impossible to 
find in sufficient numbers for an index since 
they disappear after feeding. Recent de¬ 
velopments in technique (Hackett and 
Bates 1938) have led to the cultivation of 
captive colonies of many anopheles includ¬ 
ing all the Palearctic vectors, and use of 
such material permits a more accurate de¬ 
termination of instinctive host preferences. 
The extent to which this supposed natu¬ 
ral order of host preference may be altered 
by the local predominance of certain kinds 
of animal is not known. The deviation of 
some anopheline species from man by do¬ 
mestic animals has been considered the 
cause of the “natural” recession of malaria 
from entire regions in Europe (Hackett 
1937). It seems likely that with vectors 
only weakly attracted to man, the presence 
of livestock may considerably reduce the 
transmission of malaria. It is still a ques¬ 
tion what the effect of an animal barrier is 
in the case of a vector species with a strong 
attraction for human blood. (Hackett, 
Russell, Seharff and Senior-White 1938.) 
The terms anthropophilic (or androphil- 
ous) and zoophilic have been loosely used 
to describe mosquitoes which prefer human 
and animal blood respectively. They are 
perhaps useful words to indicate tendencies, 
but they do not divide anopheles into 
groups, because the same mosquito may 
have both characteristics: it may prefer 
certain animals to man, and man to certain 
other animals. In fact this is usually the 
ease even with dangerous vectors, while 
“zoophilic” species such as A. quadrima- 
culatus, A. philippinensis or A. pseudo- 
punctipennis spread malaria by force of 
numbers even though man stands well down 
in the scale of preferred hosts. 
The transmission rate, therefore, which 
is of such fundamental importance to the 
community, will depend as far as the anoph¬ 
eline agent is concerned not only on the 
quantity of mosquitoes produced, or the 
host-parasite relationship within the insect, 
but also in large measure on the degree of 
host relationship to man, all these factors 
combining to determine the local efficiency 
of the species as a vector of malaria. 
The Sources op Infection 
The number and distribution of game- 
tocyte carriers in the population are the 
resultant of another complex of factors 
which have not been found easy to analyze. 
Malaria is characteristically a carrier dis¬ 
ease, the true carrier being an individual 
with gametocytes in his circulating blood, 
hence potentially infective to anopheles. 
Every case, however treated, ordinarily be¬ 
comes a carrier at one time or another. It 
might be expected that the number of car¬ 
riers would bear some fairly constant ratio 
to the incidence of the disease, but there is 
a wide disparity in reports from different 
regions. For example, in Egypt, in un¬ 
treated villages in 1939, we found that 
about 50 per cent of the positive thick-film 
blood specimens from apparently healthy 
children showed gametocytes, but Clark re¬ 
ports from Panama that thick-film blood 
examination of an infected community in 
an untreated area showed gametocytes pres¬ 
ent in from 8 to 15 per cent of those carry¬ 
ing malarial parasites in their peripheral 
blood. Such gross differences are due to 
the operation of a number of factors, many 
of which we can identify although we may 
not understand their mode of action. 
1. The effect of clinical attacks. It is 
generally thought that gametocytes are not 
produced continually but in waves asso¬ 
ciated with acute attacks. Their produc¬ 
tion varies with the species of parasite: 
with P. vivax they closely parallel the at¬ 
tack ; with P. falciparum they follow it and 
persist with fluctuating numbers for some 
time; and with P. malariae their appear¬ 
ance is delayed sometimes for months, 
though they continue to turn up at irregu¬ 
lar intervals for a long period. However, 
Swellengrebel and his collaborators were 
impressed by the fact that in Holland cer- 
