280 
MALARIA 
TABLE XI 
Surveys of Infants 12 Months or Less of Age. 
Sept. 1935- Aug. 1940 
Period 
covered 
Chagres 
River 
towns 
Control 
* areas 
Positive during 
first six months 
Positive during 
last six months 
Number 
examined 
Number 
positive 
Number 
examined 
Number 
positive 
1935-36 
53 5 
13 1 
0 
6 
1936-37 
59 8 
4 
4 
1937-38 
53 1 
1 
0 
1938-39 
65 1 
0 
1 
1939-40 
42 1 
7 3 
1 
3 
Totals 
272 16 
20 4 
6 
14 
Per cent 
5.9 
20.0 
30.0 
70.0 
the second to the twelfth. The highest in¬ 
cidence was 4 in the seventh month, but 
too few infections were recorded to give the 
result significance. Of the 20 infants 
found positive, five had P. vivax infections 
and 15 had P. falciparum infections, a 
ratio which is quite comparable to that 
existing for these types in the general 
population. 
The infant rate of infection is a yard¬ 
stick with which to measure transmission. 
Table XI shows an infant rate of 5.9 per 
cent among the babies in the treated villages, 
surveyed 12 times a year, as compared with 
a rate of 20 per cent among infants living 
in the control areas, which are surveyed 
only at infrequent intervals. 
Cyclical Variations in Malaria Inci¬ 
dence in Populations Under 
Observation 
During the 10 years of our observations, 
great fluctuations in the malarial parasite 
rate have been noted, in both our treated 
towns and in the control groups. The chart 
which will be found on page 112 of our 
fifth annual report (1936) indicates these 
fluctuations for the first five-year period, 
and Table I of the present paper shows the 
variations in the rates observed in the 
treated population since 1935. The exist¬ 
ence of such fluctuations is undeniable, as 
shown in the chart and the table, but the 
authors are not in agreement as to their 
causes. The first author (H. C. C.) be¬ 
lieves that they are caused by unusual in¬ 
creases in the numbers of anopheles mos¬ 
quitoes in the villages, owing to favorable 
conditions for larval production caused by 
periodic lowering and stagnation of the 
waters of the Chagres River. Such favor¬ 
able conditions recur annually at the be¬ 
ginning of the dry season (January), and 
have been particularly marked since the 
beginning of impoundment of water behind 
Madden Dam (Sept. 1934). Unfortunately 
no data as to mosquito prevalence at various 
seasons of the year are available, except 
for the last two years of our study (Sept. 
1938-Sept. 1940). The figures for these 
years show that mosquito production is 
heaviest during the dry season (January- 
May), when low water, sluggish current 
and bright sunlight combine to make lar¬ 
val breeding conditions very favorable. In 
support of the idea that increased mos¬ 
quito production is responsible for high 
malarial rates, it is pointed out that two 
notable peaks of incidence occurred in the 
villages along the Chagres River, in the 
midst of the breeding areas, at the begin¬ 
ning of the dry season in 1933, and again 
in 1935. Since 1935, only one such peak 
has occurred, and this increase began in 
August, 1939, and continued for several 
months thereafter. In 1936 and 1937, 
malarial rates were higher in the first 5 
months of the year, corresponding with the 
dry season, but in 1938 and 1939, higher 
rates were obtained in the second half of 
the year, particularly in 1939, when a para¬ 
site rate of 17.5 per cent was found in 
August in the 4 Chagres River villages. 
This was higher than any rate noted in 
these villages since the great epidemic of 
the first 4 months of 1935, when parasite 
rates as high as 28.8 per cent were obtained 
in the 4 Chagres River villages. This 
relatively high rate of 17.5 per cent 
occurred in the early part of the rainy 
season, and was more than three times as 
high as the rates found during the dry 
season months of March, April and May of 
the same year, when they averaged about 
5.4 per cent. 
