CONTROL OP MALARIA IN PANAMA WITH DRUGS 
281 
The evidence that cyclical variations in 
malarial parasite incidence is due to some 
other factor than mosquito abundance is 
contained in the tables showing parasite 
rates obtained during the past 10 years. 
In Table III the high parasite rate of 45.6 
per cent for the initial surveys (including 
New San Juan) is mentioned. These sur¬ 
veys were made in September and Decem¬ 
ber, 1929, during the heavy rainy season, 
5 years before Madden Dam was completed. 
The next surveys were made a year later 
(Sept. 1930), revealing a parasite rate in 
the same population of only 16.2 per cent. 
Meanwhile no treatment had been given. 
This diminution in parasite rate, which 
was not in any way influenced by treat¬ 
ment, was apparently the normal decline 
from a peak of high incidence, in which 
we had unwittingly made our initial sur¬ 
veys. Increases in parasite rate were noted 
during the first year of our treatment, until 
in August 1931 they reached another peak 
with a rate of 27.8 per cent, in spite of 
treatment. They declined thereafter almost 
steadily, even through the dry season, until 
1933, when high rates were again found, 
which continued throughout the year in 
spite of intensive treatment of 4 towns with 
atabrine, personally administered by one 
of us (W. H. W. K.). In 1934, rates 
dropped precipitately, even through the 
dry season, and remained at low figures 
throughout the year. Then in the first 4 
months of 1935 we experienced an epi¬ 
demic, when rates rose to high levels (28.8 
per cent in the treated towns) during the 
dry season (January through March), and 
declined rapidly thereafter (see chart on 
p. 112 of our fifth year’s observation). No 
significant increases were noted in 1936, 
1937, or 1938, and none occurred in 1939 
until August, when the rate increased to 
17.5 per cent from 6.9 per cent in the pre¬ 
ceding month. Thus 4 dry seasons passed 
without any significant rises in parasite 
rates; the first such increase occurred in 
August, 3 months after the rains had begun. 
Apparently, judging from this experience, 
an epidemic may occur at any season of 
the year. 
If river conditions favoring increased 
production of anopheles mosquitoes are the 
cause of increased parasite rates, such in¬ 
creased rates should reoccur each dry 
season, after the completion of the Madden 
Dam (September 1934), which impounded 
water during the dry season. However, 
with the exception of the epidemic in the 
early months of 1935, the next 5 years 
passed with no significant rise in parasite 
rates in any dry season. The first notable 
increase came in the month of August 1939, 
several months after the rains had begun. 
If mosquito density were the answer to 
the problem of cyclical increases, towns 
along the river bank should have the high¬ 
est parasite rates, as they are nearest the 
source of mosquitoes. The town of Las 
Guacas, which is nearly surrounded by 
water and in which the highest density of 
mosquitoes was found in 1932, has had con¬ 
sistently the lowest rate of any of our 
villages. On the other hand, New San 
Juan, which lies some 2 or 3 miles to the 
north of the Chagres River, has had con¬ 
sistently the lowest density of mosquitoes 
of any of our towns. Yet at times the 
parasite rate in New San Juan has been 
higher than in any of the other villages. 
Malarial parasite rates in our area ex¬ 
hibit large variations over periods of vary¬ 
ing lengths. One of us (W. H. W. K.) 
believes that these variations are caused by 
great increases in new infections, following 
a loss of tolerance, which may be due to 
natural causes or to treatment. Relapses 
certainly play a part in the increases noted, 
but they are believed to be significant as 
compared to the numbers of new infections 
found during epidemic periods. 
The same author believes that the in¬ 
creased numbers of “heavy infections” 
noted as occurring since our drug treat¬ 
ment was begun is also due to a decrease 
in tolerance brought about by ridding the 
population of its parasites. When malaria 
attacks such a non-immune population, the 
number of clinical cases is higher than in 
a control group, which has had little or 
no treatment. This point is brought out 
strongly in our fifth annual report (2), and 
