U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content







AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window






Chapter 4



The South Atlantic and Caribbean Areas

Luther S. West, Ph. D.

Part I. General Considerations

    The total geographic area covered by this report extends approximately from 30º north latitude to 40º south latitude and from 30 º west longitude to 95º west longitude. It comprises both the Caribbean Command and the South Atlantic Command at their greatest expansion during the war years, 1941-45. The Caribbean Defense Command consisted of the Panama Canal and Antilles Departments, represented in prewar years by permanent installations in the Canal Zone and the island of Puerto Rico.

    War Department Pamphlet No. 8-2, dated 2 April 1943, indicates that malaria was most prevalent in tropical regions of Central America and in the great river basins of South America, especially those in Brazil, the Guianas, Venezuela, Colombia, and Northwestern Argentina. It is believed to occur in all Caribbean islands except Barbados and the Netherlands Antilles. In some areas, up to 80 percent of the native population may show infection.

    The southern range of malaria in the New World terminates at 32º south latitude, in Cordoba, Argentina.1 The only recorded exception was a brief epidemic in 1935-37 near Mendoza, at 33º south latitude. No cases have been reported in Brazil south of the 30th parallel, and in Chile the disease does not occur south of the Pica stream in the Province of Tarapacá, at 20º30' south.

    The malaria problem in these areas has long been very real. The prevalence of the disease in the native population of various Central American, South American, and West Indian countries has played its part in sapping the energy of these peoples and in preventing the attainment of economic standards which natural resources and commercial opportunities would otherwise have made possible. Conversely, a low economic level, whether due to malaria, hookworm, other diseases, or merely to overpopulation, as in certain of the Antilles, provides little money for expenditure on public health. Thus, most of the governments concerned, though realizing the need for malaria control, were not able to command the funds necessary to utilize approved control procedures.

    The example set for the Central American Republics by the United States, in connection with the sanitation of the Canal Zone, had both a desirable and

1 Hackett, Lewis W.: Distribution of Malaria. In Malariology, edited by Mark F. Boyd. Philadelphia: W. B. Saunders Co., 1949, vol. I, pp. 722-735.


an undesirable effect. The practical elimination of mosquito breeding by the use of oiling and drainage techniques showed what could be done, and stimulated imitation. The enormous expense involved, however, made it abundantly clear that the Republics could not hope to raise, by any system of taxation, sufficient money to parallel the U.S. achievement. Educational effects were nevertheless apparent in the greater use of screens or bed nets, avoidance of unnecessary exposure during the biting hours of vector species, and in other ways. These practices were more closely followed in villages and towns than in open country, as evidenced by a higher spleen rate, for example, among children who were transported to village schools, than among those who lived close by.

    In contrast, however, is the fact that native populations, in all tropical regions, tend to develop a degree of tolerance to malarial infection, so that adults though perhaps exhibiting a parasitemia may be relatively free from morbid symptoms. Except for exacerbations brought on by fatigue, malnutrition, chilling, or illness of some other type, they may carry on effectively for long periods of time and be neither a burden to themselves nor to others. Both natural selection and immune reaction are involved. Natural selection operates largely through infant mortality, the survivors being, it may be assumed, somewhat better fitted by nature to cope with the parasite. This selected group then stands to profit by whatever immunity can be developed through repeated exposure to infection. It should be pointed out that frequent reinfection is necessary to accomplish this. Such an immunity is termed premunition. If the individual remains free of parasites for any considerable period of time, his immunity is lost. Untreated natives have, of course, but little opportunity to become parasite free.

    The introduction of a new, nonimmune population into a malarious area is always a hazardous procedure. As long as the newcomers arrive in relatively small groups, make plans for their own medical protection, and tend to engage in more privileged occupations (executive, clerical, professional), no great challenge of existing practices is likely to occur. If, on the other hand, there should be an influx of some numbers destined to engage in common labor and to live under perhaps less than desirable sanitary conditions, an epidemic among the new arrivals is almost certain to take place. The arrival of troops from a nonmalarious region parallels this example in its first aspect because the duties of military personnel necessarily take them into situations where risk of infection may be very great. The second problem, involving living conditions and especially housing, need not exist if there be sufficient advance knowledge of control procedures, the means to apply them, and effective instruction of the troops. Actual combat conditions may destroy all this, at least temporarily, but since this discussion is concerned with an area in which no fighting took place, it is unnecessary to go further into this aspect of the subject.



Central and South America

    During the year 1941, malaria caused the death of 1,178 persons in the United States. 2 This represented a malaria death rate of 0.9 per 100,000 inhabitants. It will be evident from table 10 that malaria, at the outbreak of the war, was a much more serious matter in most of the other American Republics.

    The statistics in table 10 make clear the extent, as well as the importance, of the malaria problem in selected geographic areas in Central and South America.

TABLE 10.- Malaria as a reported cause of death in Central and South American Republics and neighboring States, with selected examples

2   Informe Biodemogr áfico y Epidemioló gico de las Américas.  Washington:  Officina Sanitaria Panamericana, Publication No. 195, February 1943


U.S. Army, 1942-45

    During the war years, the personnel of the U.S. Army suffered more or less seriously from malaria (tables 11 through 15). From a global standpoint, the year 1943 was certainly the most alarming. In Latin America, however, where troop concentrations were well underway at the outbreak of the war, 1942 became the year of disastrous experience, followed by rather consistent improvement for the remainder of the war period.

    The figures given in the tables are adapted from material supplied by the Medical Statistics Division, Office of The Surgeon General, U.S. Army.

TABLE 11.- Number of cases 1 and attack rates of malaria in the U.S. Army, worldwide and Latin America, by type of Plasmodium and year, 1942-45


TABLE 11.- Number of cases 1 and attack rates of malaria in the U.S. Army, worldwide and Latin America, by type of Plasmodium and year, 1942-45 - Continued

TABLE 12.- Incidence of malaria in the U.S. Army, by area and year, 1940-41

TABLE 13.- Deaths due to malaria in U.S. Army, by area and year, 1940-41


TABLE 14. - Deaths due to malaria in the U.S. Army, worldwide and Latin America, by type of Plasmodium and year, 1942-45


TABLE 15. - Admissions and deaths from fever of undetermined origin1 in the total U.S. Army and in the Latin American area, by year, 1942-45


    The malaria control machinery of the various Republics and European possessions, included in this extensive area, differed greatly in magnitude and efficiency at the outset of the war. Wherever the local government organization was well developed, and more or less experienced, as in Brazil, close cooperation was sought by U.S. authorities, and cooperative administration of control activities in the vicinity of military installations was the rule. This was particularly true near the several bases used for air transport purposes along the Atlantic coast. In other areas, the full responsibility had to be assumed by military authorities. More detailed information will he given in connection with the history of particular installations.

    By 1945, there were 70 malaria survey units and 153 malaria control units working in oversea theaters throughout the world.In addition, nonmedical antimalaria details, consisting of one noncommissioned officer and one or more additional enlisted men, were assigned to serve individual companies, batteries,

3 Simmons, James Stevens; Control of Malaria in the United states Army. In Malariology, edited by Mark F. Boyd. Philadelphia: W. B. Saunders Co., 1949, vol. II, pp. 1455-1468.


or similar units in malarious areas. These carried out minor control measures and watched for breaches in malaria discipline.

    The malaria survey units consisted of 11 enlisted men, directed by 2 officers of the Sanitary Corps, one an entomologist and one a parasitologist. Control units consisted of 1 sanitary engineer and 11 enlisted men. Such units were trained and equipped to carry out and supervise large-scale malaria control programs. Survey units and control units frequently assisted one another in adjacent areas.


    As in other theaters of operations, malaria control activities in the Caribbean and South Atlantic Commands were divisible into three general types:

    1. Permanent improvement of the environment, by drainage, filling, installation of floodgates, and other engineering procedures. These activities were, of course, expensive and were only followed where longtime use of the area by military, industrial, or commercial interests, with attendant concentration of population, seemed assured.

    2. Temporary control measures of various kinds, calculated to protect large numbers of individuals, were employed but were effective for only limited periods of time. If the installation continued to function longer than originally expected, such procedures had to be repeated at required intervals. The use of larvicides in breeding places, the screening of barracks, and later, residual spraying of quarters, inside and out, are examples. The eventual employment of DDT, from the air, to render large areas temporarily free from vector species falls here also. It was justified only in connection with the movement of large bodies of troops in combat areas, or in the carrying out of large-scale maneuvers.

    3. Measures usually classed under the head of malaria discipline. The wearing of protective clothing, remaining in protected buildings during the hours of greatest anopheline activity, and the consistent application of repellents are examples. Proper use of jungle hammocks may be included. These practices were valuable in combat zones, while on maneuvers, and in temporary bivouac areas. For these, the individual soldier was held responsible.

    Malaria discipline included the application of repellents. Granett,4 at the time a research fellow at Rutgers University, New Brunswick, N.J., summarized the attributes of a suitable repellent for use by the Armed Forces:

    The repellent must be effective; i.e., it must be able to prevent biting by many pests for long periods of time even under adverse conditions. It must not be injurious to the user or any articles of clothing or objects he may touch. It must possess certain miscellaneous properties such as an acceptable odor, availability, and stability, under storage or field use.

4 Granett, P. : The Significance of the Development of Mosquito Repellents for the Protection of Military and Civilian Populations. Proc. N.J. Mosq. Extermin. Ass. 30: 203-210, 1943


    The most satisfactory list of insecticides and repellents as developed for the Armed Forces is found in U.S. Department of Agriculture Miscellaneous Publication No. 606 released in August 1946. Rutgers 612 (2-ethyl-1, 3-hexane-diol), Indalone (butopyronoxyl), and dimethyl phthalate, alone and in combination, excelled as repellents.

    A memorandum from Headquarters, South Atlantic Ferrying Wing,5 lists precautions to be taken by the various commands wherever conditions were such as to require them:
    1. Effective screening and mosquitoproofing of all buildings occupied by troops.
    2. Use of mosquito bars or netting for sleeping.
    3. Prohibition of loitering or sitting in unscreened buildings or outdoors when not required by military duties, between dusk and dawn. Wearing of protective clothing by personnel on outdoor duty during these hours.
    4. Systematic spraying of barracks, kitchens, messhalls, storerooms, and recreation buildings where indicated.
    5. Keeping of doors tightly closed. Prompt reporting and repair of holes in screening. Enforcement by periodic inspection of buildings.
    6. Elimination of local breeding places of mosquitoes. Spraying with larvicide recommended where practical.
    7. Spraying of aircraft as directed by regulations.
    8. Use of chemical prophylaxis in areas indicated.

    The protection of personnel from malaria and the disinsectization of aircraft were considered command functions.



    The missions of troops in the Caribbean and South Atlantic were various but not especially complex. They were both offensive and defensive, the latter being more important during the early years of the war. American territory had been attacked at Pearl Harbor, and it was considered highly probable that the enemy might attempt to gain control of the Panama Canal, or at least to cripple its operation. The United States was also interested, jointly with the British Government, in maintaining full, functional use of the West Indies. The greatest menace in the West Indies consisted of German submarine activity against Allied shipping.

    In September 1940, the United States acquired bases on Bermuda, Exuma (Bahama Islands), Jamaica (at Kingston), Antigua (Leeward Islands), Saint Lucia (Windward Islands), Trinidad (Great Britain), and Georgetown (British Guiana).

5 Medical History World War II, U.S. Army Forces, South Atlantic. Appendages thereto. [Official record.]


    As the Allies gathered strength for the African offensive, the importance of these airbases in both the Antilles and northern South America became much greater and necessitated more and larger installations, with emphasis on transport function rather than aerial defense.

    Transport of lend-lease equipment had, of course, been a function of many of these bases from the beginning.

    Meantime, both Puerto Rico and the Canal Zone were being developed as training centers for troops intended for employment in jungle warfare. An important training area was developed and maintained in southern Trinidad (by the 33d Infantry). It was necessary that this training be both realistic and thorough, if survival and success in Asia, South Pacific, and the Philippines were to be achieved. An important phase of such training consisted of instruction concerning the nature and transmission of malaria, as well as practice and discipline in the use of those preventive and suppressive practices which are under the control of the individual soldier. In patrol activity and in actual combat, these are the chief measures available.

    The mission of troops in the Caribbean and South Atlantic may thus be summarized:
        1. Defense of the Panama Canal and adjacent areas.
        2. Defense of the West Indies, particularly with a view to suppressing German submarine activity.
        3. Provision of a chain of bases along which both troops and supplies might be transported to the African theater.
        4. Provision of a training area in which conditions of jungle warfare might be duplicated or at least paralleled.
        5. The disciplining of troops in regard to prevention of tropical and parasitic diseases (including malaria) with which they were sure to be confronted, both in African and in oriental service.

    This last objective was not recognized, at first, as being particularly important, and was not defined in early directives. Experience in the Pacific theater soon forced attention to these problems, however, and the inclusion of malaria discipline became a routine matter in preparing troops for foreign duty. The acceptance of responsibility on the part of officers of various levels of command was a necessary step in making the program effective.

Air Force Installations

    Reference has already been made to bases acquired by the United States in 1940 by arrangement with Great Britain. In July 1941, the Brazilian Government passed legislation permitting the United States to develop a number of airbases in that country. These were at Amapá, Belém, Sâo Luis, Fortaleza, Natal, Recife, Maceió, and Bahia.6 Construction was carried out by the Pan American Airways System, Inc., through its subsidiary, the Air Port Develop-

6 Medical History World War II, U.S. Army Forces, South Atlantic. [Official record.]


ment Project. To conduct operational activities, the Army Air Forces created the Army Air Corps Ferrying Command, which later became the global Air Transport Command. Expansion of transport activities and the necessity of better provision for the defense of Northeastern Brazil brought about the establishment of the South Atlantic Wing, Air Transport Command, in June 1942. Headquarters were established at British Guiana on 9 July, where a U.S. Army base already existed.

    The various airbases established in the area controlled by the Caribbean Defense Command will be listed in separate discussions of the Panama Canal and Antilles Departments.



    The area covered by this discussion includes all islands and coastal regions surrounding the Caribbean Sea, together with such further malarious portions of South America as were occupied or used by U.S. troops. It lies approximately between 35 º and 90º west longitude and extends from a little north of the Tropic of Cancer to 35º south latitude (map 10). From the standpoint of mosquito bionomics and malaria transmission, a considerable variety of ecologic situations exist, altitude as well as prevailing winds being factors in some instances. Since it is difficult to generalize for an area characterized by the division of its landmass into so many widely separated units, specific physiographic information will be reserved for presentation in connection with the discussion of mosquito vectors, some of which breed in peculiar and restricted situations, not always typical of surrounding territory.


    As with physiographic characteristics, the area under consideration is too large to permit very concise statements as to climate. From a general standpoint, one is concerned here with the climate of the Torrid Zone, from the Tropic of Cancer to the Tropic of Capricorn. Seasonal differences are measured largely in terms of rainfall, rather than temperature. The striking climatic differences sometimes encountered as one travels from east to west, or vice versa, are practically all explicable in terms of altitude, as, for example, when one proceeds from the Pacific coast of South America, crosses the Andes, and descends into the valley of the Amazon. Apart from altitude, air movements can be a factor, evidenced, for example, by the reduction in precipitation along the coast of South America, as one proceeds westward from French Guiana to Curacao and Aruba, N.W.I. This is due to loss of moisture by easterly trade winds. Climate tends to favor the occurrence and transmission of malaria when the following conditions are met:


MAP 10.- Caribbean area and South Atlantic theater.

    1. A continuous supply of suitable water for the breeding of the vector. Species in which the eggs can survive temporary "mudbank" conditions are obviously more adaptable in this respect. On the other hand, no anopheline does well when there is an excess of precipitation, as when the terrain is covered with a flowing sheet of water.


    2. Maintenance of water temperatures above the point where immature stages of the vector will be killed or rendered incapable of recovery.
    3. Maintenance of air temperature above the point where adult anophelines are rendered permanently incapable of mating, feeding, or egg laying.
    4. Frequent and prolonged association of warm temperatures with high relative humidity, which, by favoring the uninterrupted breeding of the vector, insures its presence in great numbers in the vicinity of both parasitized and nonparasitized human hosts. The same factors tend to favor the use of little or no clothing on the part of native populations, a practice that exposes them to unrestrained attack by enormous numbers of mosquitoes.
    5. Freedom from strong, persistent winds during the normal biting hours of the vector, as mosquitoes do not readily take to wing when winds are severe. It should be remarked, however, that gentle winds are no barrier to vector activity and may even be a positive factor in malaria dispersal. Schiavi,7 for example, conducted a survey in 1944 in the town of Iguape, situated on the coast in the southern portion of the State of Sâo Paulo, Brazil. Anopheles albitarsis domesticus, captured in houses where cases of malaria had occurred, yielded both oocysts on the stomach and sporozoites in the salivary glands. Local opportunities for breeding were so limited, however, that the mosquitoes were considered to have flown to the mainland from the low, marshy island of Ilha Camprida when a sea breeze was blowing. Specimens were observed in flight, over the channel.

    Although certain species of anophelines are adapted to breeding at higher altitudes than others, even the hardiest eventually reaches a level at which frequency of low temperatures prevents the survival of the species.

    From the foregoing remarks, it is evident, therefore, that the coastal regions are, in general, more hazardous from a malaria standpoint than the interior areas, and that within these interior areas, the great river valleys may be expected to be more malarious than the mountain slopes. The peculiar habits of certain vector species are responsible for occasional exceptions.

    Special relations of climate to the incidence of malaria in limited areas will be treated in another section of this chapter.

Native Population

    The population of the Latin American countries is composed of a mixture of races, proportions differing according to locality. American Indian blood is more persistent in Mexico and South America than in certain of the Central American countries, though isolated tribes survive there practically untouched by outside influence. The use of the Portuguese language in Brazil and of Spanish (with various modifications) in the great majority of political units testifies to the dominant character of the European influence. One should not

7 Schiavi, A.: Nota sóbre mosquitos vetores em Iguape. Arq. hig. e saúde púb. 10: 67-75, September 1945.


forget, of course, the French, Dutch, and British interests in the Guianas.

    British, French, and Spanish bloodlines exist in the West Indies, where, however, as in Panama, the tremendous biological vigor of the African element, introduced in the period of the slave trade, tends to obscure the characteristics of all races with which it interbreeds. East Indian stocks introduced labor and augmented by later arrivals are making themselves felt. Many merchants in Panama, for example, are East Indians. Chinese and other oriental groups are evident in many cities and towns. Superimposed on all this is the constant coming and going of nationals, from almost all the leading countries of the world, who visit the American Republics in the interest of commerce and industrial development and who frequently become residents for varying lengths of time.

    Racial makeup of particular countries or areas will be noted, where significant, in other sections of this chapter.

Mosquito Vectors 8

    The anopheline mosquitoes considered to be important carriers of human malaria in the regions under consideration number eight. These are:
    1. Anopheles albimanus Wiedemann, 1921.
    2. Anopheles albitarsis Lynch Arribálzaga, 1878.
    3. Anopheles aquasalis Curry, 1932 (referred to as Anopheles tarsimaculatus by certain authors).
    4. Anopheles bellator Dyar and Knab, 1906.
    5. Anopheles darlingi Root, 1926.
    6. Anopheles gambiae Giles, 1902 (exterminated in New World).
    7. Anopheles pseudopunctipennis Theobald, 1901.
    8. Anopheles punctimacula Dyar and Knab, 1906.

    In addition to the species just listed, Mexico and Central America, including the Caribbean, support at least 34 anopheline species usually not considered vectors, while South America boasts a total of 52 probably nonvector species and varieties, 9 not counting 3 species of Chagasia and omitting some 13 or more forms of questionable validity. Among potential vectors is Anopheles noroes-

8 (1) Komp, W. H. W.: The Anopheline Mosquitoes of the Caribbean Region. Nat. Inst. Health Bull. No. 179, 1942. (2) Kumm, Henry W.: The Geographical Distribution of the Malaria Carrying Mosquitoes: A Collection of Recorded Material in the Literature, and in Personal Communications to the Author. Baltimore: The Johns Hopkins University Press, 1929. (The American Journal of Hygiene Monographic Series No. 10.)  (3) Kumm, H. W., and Zúniga, H.: The Mosquitoes of El Salvador. Am. J. Trop. Med. 22: 399-415, July 1942. (4) Russell, Paul F., West, Luther S., and Manwell, Reginald D.: Practical Malariology. Philadelphia: W. B. Saunders Co., 1946. (5) Mackie, Thomas T., Hunter, George W., and Worth, C. Brooke: A Manual of Tropical Medicine. Philadelphia: W. B. Saunders Co., 1945. (6) Malariology, edited by Mark F. Boyd. vol. I and II. Philadelphia: W. B. Saunders Co., 1949.
9 (1) Dyar, Harrison G. : The Mosquitoes of the Americas. Publication No. 387, Carnegie Institute. Washington, D.C., 1928. (2) Gast-Galvis, A.: Biologia y distribución geográfica de los anofelinos en Colombia, Rev. Fac. Med. Univ. Nac. 12: 53-103, August 1943. (3) Lane, J.: Catálogo dos Mosquitos Neotrópicos do Institute de Higiene de S. Paulo. Editado pelo Clube Zoológico do Brasil. S. Paulo Brash, Boletin Biologico. Série Monográfica, 1939.


FIGURE 23.- Anopheles albimanus breeding along the edges of a quiescent stream.

tensis Galvâo, which Coutinho 10 found infected in Brazil but which has not been shown to transmit malaria to man.

Habits and distribution of vectors

    Anopheles albimanus . - This mosquito ranges from Texas through Mexico and the West Indies into Venezuela, Colombia, and Ecuador. The larvae are found in a variety of situations, including fresh or brackish lagoons, swamps, marshes, lakes, ponds, pools, puddles, pits, hoofprints, and artificial containers (fig. 23). Exposed, sunlit waters are favored, though some shade is tolerated. This adaptability is the chief reason for its great importance as a vector

10 Coutinho, J. O.: O "Anopheles (N.) oswaldoi metalfi" Galvâo e Lane, 1937 e o "Anopheles (N.) albitarsis" Arribalzaga, 1878 como transmissores de malaria no Distrito Federal. Brasil-med. 56: 52-55, 24-31 Jan. 1942.


both in Central America and in the Caribbean. The adults are nocturnal biters, and prefer man, but will also feed on animals. They will enter houses but in most cases leave after feeding. The adults are very secretive in daylight hours, and even experienced collectors have difficulty in locating them.

    Plant associations have an important bearing both on the suitability of the environment for A. albimanus breeding and on preferred methods of control when the vector is present. Studies carried out by Pratt in Puerto Rico have thrown considerable light on this problem.11 Thus drainage and dusting are most successful in coastal swamps where Phragmites and Typha augustifolia are dominant species. Only the sunlit margins are utilized by A. albimanus . For pools containing tall grasses, drainage is preferable, as larviciding becomes difficult. Sugarcane fields produce few mosquitoes just after the cane is cut, but sinking of the trash in the irrigation ditches soon results in a sunlit surface where A. albimanus is at home. Reflooding intensifies this condition. Anopheles grabhami tends to replace A. albimanus when the canes reach a meter in height, and A. vestitipennis usually takes over when canes are full grown.

    Azolla, Pistia, and other plants which form floating mats tend to inhibit mosquito breeding. On the other hand, low-floating plant growth, such as is formed by Chara, Utricularia, and Ceratophyllum, protect the larvae from their natural enemies, and so favor their development. In Puerto Rico Ceratophyllum demusum is a very common associate of A. albimanus.

    Anopheles albitarsis. -The typical A. albitarsis from Buenos Aires is not domestic, nor does it prefer human blood. 12 On the island of Marajó, however, in the State of Pará, Brazil, the females invade dwellings in large numbers during the evening, even when cattle are conveniently available. This variety has come to be known as A. albitarsis domesticus, a form erroneously referred to by certain authors as Anopheles albitarsis var. braziliensis Chagas and by others as Anopheles argyritarsis Robineau-Desvoidy. Anopheles albitarsis var. limai Galvâo and Lane is possibly identical with the typical form.

    Anopheles albitarsis , which is a vector in Brazil, is found from Guatemala into northeastern Argentina and Uruguay. It also occurs in Trinidad. It breeds chiefly in ponds, ricefields, marshes, lagoons, and the overflows of rivers. Matted vegetation and the availability of a certain amount of shade favor its development. The adults show a preference for human blood in certain areas but not in others, indicating the existence of geographic races.

    Anopheles aquasalis . - The geographic distribution of A. aquasalis in the neotropical region is discussed by Lucena 13 under the name of A. tarsimaculatus

11 Pratt, H. D.: Relation of Plants to Malaria Control in Puerto Rico. Pub. Health Rep., Suppl. No. 200, December 1947.
12 Galvâo. A. L. A., and Damasceno, R. G. Observaçôes sobre anofelinos do complexo albitarsis. An. Fac. de med. Univ. Sâo Paulo 20: 73-87, 1944.
13 Lucena, D. T. Esboço ecológico do Anopheles (N.) tarsimaculatus Goeldi, 1905, no nordeste Brasileiro. Folha med. 27: 140. 1946.


Goeldi. Because extensive areas of salt land stretch inward from the coast, the species occurs munch further inland than was formerly supposed. Larvae have been collected 70 miles inland from the State of Paraíba and 160 miles inland in Pernambuco. In these localities, they breed in ditches, pools, or overflows from rivers in which the water is brackish due to the heavy chloride content of the soil. Coastal breeding goes on in waters ranging from 0.117 to 33.345 gm. chloride per liter. This variation is due to tides. Inland, the variation is less extreme, ranging from 5.733 to 29.835 gm. per liter. The concentration at any one spot is usually greater in summer than in winter because of evaporation.

    The adults enter houses in large numbers, being in greatest concentration between 10 p.m. and 10 a.m. They have a strong preference for human blood and will attack man in buildings during the day and in the open air. In those localities where it shares a range with A. albitarsis, A. quasalis is most numerous mn January and February, A. albitarsis, in October.

    Anopheles aquasalis occurs in Nicaragua, Panama, the Lesser Antilles, Trinidad, and on south into Brazil. It is a widespread vector. Favorite breeding situations are brackish lagoons and marshes and the irrigation water in open fields. Strictly fresh water is avoided, hence the species is usually limited to coastal situations. It breeds in either sun or shade. Adults may fly as far as 3 miles.

    Anopheles bellator . - This is one of the most unusual of the vector species. It ranges from Trinidad and parts of Venezuela into Brazil. Eggs are laid and larvae develop in the water found in certain species of bromeliads (wild pineapples), which grow epiphytically in the branches of the immortelle trees (figs. 24 and 25). The adults attack man with avidity, both day and night.

    In survey work, the larvae are collected by breaking the plant loose from the tree, on which it is growing, and by dumping the water into a pan. Larvae which adhere to the leaves are washed off by means of a pipette. Spraying the trees with copper sulfate under high pressure has been used to kill the bromeliads.

    Bromeliads of the genus Gravisia are more typical of the dry, canopy stratum, while species of the genus Guzmania occur chiefly in the lower shade. 14 The latter type has only a small reservoir for holding water (about 80 cc.). Anopheles humunculus, a doubtful vector, shows a strong preference for these forms, while A. bellator favors Gravisia , the capacity of which may exceed 600 cc. Such forms are referred to as "tank" bromeliads.

    Anopheles darlingi. - This species is recorded from British and Spanish Honduras, Guatemala, Venezuela, coastal British Guiana, Brazil, and northern portions of Argentina. As a vector, it is especially important south of the Equator. The larvae, which prefer shade, may usually be found among debris and surface vegetation in fresh marshes, lagoons, and overflows. Water hyacinth (Eichornia ) and waterlettuce (Pistia) frequently shelter the larvae

14 Preliminary Report, C. S. Pittendrigh, 1944, subject: Malaria-Bromeliads and Bremeliad Anophelines, Trinidad, B.W. I.


FIGURE 24 - Bromeliad attached to tree, a typical breeding place for Anopheles bellator Dyar and Knab. (Courtesy of the Rockefeller Foundation.)

of A. darlingi. This is a domestic species. The adults invade houses and show a preference for human blood.

    Anopheles darlingi is the chief vector in the district of Campos, State of Rio do Janeiro, Brazil. 15 Surveys have shown 6.8 percent of captured females to contain oocysts, as opposed to 2 percent for A. albitarsis. Sporozoites were found in A. darlingi only, to the extent of 4 percent.

    Anopheles gambiae . - This species, now believed extinct in South America, is an African species which was accidentally introduced into the New World and became responsible for a great deal of malaria in Brazil between 1930 and 1940.16  Systematic eradication measures accomplished its elimination. It is an adaptable species, breeding in either sun or slight shade, and seems equally

15 Coutinho, J. 0., and Ricciardi, I.: Contribuiçâo para o estudos dos vetores do malaria no Brasil.  Anopheles (Nyssorhynchus)Darlingi Root, em Campos, no Estado do Rio. Mem. Inst. Oswaldo Cruz 42 : 263-280, February 1945.
16 Soper, Fred L., and Wilson, D. Bruce: Anopheles gambiae in Brazil, 1930 to 1940. New York: The Rockefeller Foundation, 1943.


FIGURE 25.- Horizontal section through a bromeliad to snow water-holding area among leaf bases.

at home in puddles, pools, shallow ponds, burrow pits, hoof prints, ditches, and overflows. Rain barrels, cisterns, and other peridomestic containers are sometimes used. The adults prefer human blood and are usually abundant in huts and houses.

    Anopheles gambiae was first discovered in Brazil in March 1930.17 The parasite rate in these mosquitos was 63 percent, including a sporozoite rate of 30 percent. An intense malaria epidemic soon struck suburban Natal. Dry seasons caused a recession in breeding activities for a few years. but normal

17 (1) Shannon, R. C.: Brief History of Anopheles gambiae in Brazil. Caribbean Med. J. 4:123-128, 1942. Abstr. Trop. Dis. Bull 40: 522, July 1943. (2) Shannon, R. C., and Cesar De Andrade, G.: Dry season Observations on the African Mosquito, Anopheles gambiae, in Brazil in 1938. Am. J. Trop. Med. 20 : 641-668, September 1940.


rainfall in 1936 reversed this trend, and in 1938 and 1939, a total of 14,000 deaths from malaria were reported, chiefly from the Açu, Mossoró, and Jaguaribe Valleys. The Brazilian authorities at this point requested aid of the Rockefeller Foundation. By cooperative effort of the Foundation and the Yellow Fever Control Service of Brazil, complete eradication was effected in 19 months, from January 1939. The last living specimen was found in September 1940. The entire cost of this achievement was in the vicinity of approximately $2 million.

    An extended and fully documented account of the joint activities of the U.S. Army and the Brazilian Government in relation to the problem of A. gambiae 18 shows that much anxiety existed and that procedures were not always what they should have been; however, that A. gambiae did not again establish its residence in the New World is distinctly to the credit of all concerned. A brief summary of the manner in which the problem was handled follows:

    Anopheles gambiae had been eradicated in Brazil, in 1940. In 1942, the air traffic across the Atlantic became greatly augmented, thereby vastly increasing the danger of a second colonization in the New World. Increasing numbers of specimens began, in fact, to be found on newly arrived planes. There was some question as to accuracy of identification in all cases, and proof of living specimens seems never to have been certain, but the possibility of a second establishment nevertheless loomed ominously. In an attempt to prevent this, a program was instituted by which all aircraft arriving from Africa were sprayed at once with a mixture of pyrethrum, carbon tetrachloride, and kerosene. 19

    The added complications of military travel soon led to a cooperative program administered jointly by the U.S. Army and Brazilian authorities. In this program, the use of Aerosol bombs became routine (except at the first opening of the doors). It was demonstrated that the spraying of the airplanes in flight had no supplementary value, as most of the insecticide escaped through the tail. In 1943, the U.S. authorities adopted the policy of carrying out sanitation procedures before departure from African airfields. The justification for such routine is made abundantly clear by the insect counts at Brazilian airports between October 1941 and December 1945. Belém, Fortaleza, Natal, and Recife were the bases concerned. A total of 352 specimens of A. gambiae were collected at those points. These were carried by 126 of the 9,229 land-planes and by 16 of the 831 flying boats which landed during the 4-year period.

    In January 1943, it had been made mandatory for all incoming pilots to keep cabin windows closed, and canopies down, until their planes were

18 See footnote 6, p. 122.
19 Carneiro de Mendonça, P., and Cerguerira, N. L. Insects and Other Arthropods Captured by the Brazilian Sanitary Service on Landplanes or Seaplanes Arriving in Brazil between January 1942 and December 1945. Bol. Offic. san. panam. 26 : 22-30, January 1947.


sprayed. 20 Some violations were inevitable, however, especially in the case of pilots who were in ignorance of the regulation, or who, not realizing its importance, attempted to improve visibility for taxiing by manipulating windows. Investigation by a special commission toward the end of 1943 resulted in the conclusion that spraying alone, at takeoff or arrival, did not give sufficient protection to South American ports. The commission therefore recommended that measures should be taken to render the West African airfields wholly free from A. gambiae .

    The War Department, accordingly, assigned a malariologist, together with appropriate mosquito control personnel, to the Middle East theater for the purpose of coordinating mosquito control activities in Sénégal, Liberia, Nigeria, and the Gold Coast. Sixteen officers and ninety enlisted men made up the group. Before this group arrived, the Pan American Airways-Africa, Ltd., and the U.S. Army authorities in West Africa had based their program on the assumption that the flight range of A. gambiae is rarely in excess of 1 mile. Investigations by the new group showed, however, that the essential factor in dispersal was not the flight range of the mosquito but the influence of prevailing winds. In some instances, it became necessary to deepen the control zone to more than 5 miles to counteract this.

    The degree of success achieved at the several airfields was not uniform, due to the great difference in terrain. In Liberia, for example, the location of the airfield in a jungle clearing, close to native settlements, forced reliance chiefly on malaria discipline. In Sénégal, improvement was effected by removal of activities to a more healthy site where control procedures could be more effective. At Accra, an extensive system of drainage ditches was devised, and although far from being completed, this system had so reduced the breeding of A. gambiae, by July 1944, that Brazilian officials expressed complete satisfaction with the results. All things considered, the program was and is highly effective as insurance against a possible second invasion of the Americas by one of the worlds worst vectors.

    Anopheles pseudopunctipennis . - This mosquito has a wide geographic range. It occurs from the south-central United States to Chile and Argentina and is recorded from Grenada, B.W.I. The larvae prefer sunlit waters and may be found in clear seepages, puddles, pits, pools, streams, and springs. An abundance of algae favors their existence. This species is a vector in Mexico, Argentina, Peru, and possibly elsewhere. Adults enter houses and feed on man only in certain areas.

    Anopheles pseudopunctipennis has been shown by Hackett21 to behave rather differently in various portions of its range. Since it breeds in the beds of hill streams when their volume is diminishing, it is well adapted to moun-

20 Farrell, E.: The Anopheles gambiae Problem in Brazil and West Africa, 1941-44. Bull. U.S. Army M. Dept. 8: 110-124, February 1948.
21 Hackett, L. W.: The Malaria of the Andean Region of South America. Rev. Inst. salub. y enferm. trop. 6: 239-252, December 1945.


tainous regions with a long, dry season. During flood periods, it survives in secondary breeding spots, at lower levels. The species reaches its greatest altitude in the Cinti Valley, Bolivia, where it has been recorded from 8,500 feet. In most Andean countries, the incidence of malaria appears directly correlated with the rise and fall of A. pseudopunctipennis , but in Colombia malaria transmission is slight at higher altitudes, even when A. pseudopunctipennis is common. The same is true for Central America. This suggests geographic races, which Vargas 22 has attempted to distinguish. Variety typicus is believed to have been described originally from Grenada, B.W.I., rather than from Granada, Nicaragua, as indicated by certain authors. It occurs in the United States, Mexico, Costa Rica, Panama, and Venezuela, probably also in Argentina and Chile. It is a significant vector of malaria above 2,275 feet but only when present in large numbers. Variety franciscanus McCracken, which is found in the southern United States, is perhaps the correct name for the nonvector form of Central America and Colombia, mentioned previously. Variety boydi Vargas occurs in California, willardi Vargas in Mexico.

    A new variety of A. pseudopunctipennis, Anopheles pseudopunctipennis patersoni, has recently been described from Tucuman, in Argentina. The distinction is based on characteristics on the ova.23

    Anopheles punctimacula . - The species is found from Mexico through Central America, into Brazil and Peru, and also on Trinidad. Breeding occurs only in well-shaded pools, swamps, and sluggish streams. The adults, which are abundant in undrained jungle, are strong fliers and invade houses in order to feed on man. So far as is known, vectorship of malaria is limited chiefly to Panama.

Zoogeographic subdivisions

    Lane 24 makes use of zoogeographic subdivisions in discussing the distribution of neotropical anophelines. The Cariba center involves Venezuela and the Guianas. The Incasic center includes the Andean region from Panama to approximately 20º south latitude. The Central Plateau area lies in interior Brazil and is the source of streams which flow in many directions. The Sêca areas are somewhat isolated in the mountainous portion of Eastern Brazil. The Tupí center extends along the southeastern coast of Brazil from Bahia nearly to the 30th parallel. The Chilean center is confined to the northern two-thirds of Chile. The Patagonian centers are in the area west of Buenos Aires. Anopheline mosquitoes tend to migrate southward from the Cariba,

22 Vargas, L. : Consideraciones sobre el Complejo del Anopheles pseudopunctipennis . Rev. Inst. salub. y enferm. trop. 6: 265-270, December 1945.
23 Alvarado, C. A., and Heredia, R. L.: Observaciones sobre una nueva variedad del Anopheles (A.) pseodopunctipennis Theobald 1901, encontrada en la provincia de Tucuman. An. Inst. med. regi. 2: 73-78, November 1947. Abstr. Trop. Dis. Bull 45: 299, April 1948.
24 Lane, John: Anophelines of the Neotropical Region. In Malariology, edited by Mark F. Boyd. Philadelphia: W. B. Saunders Co., 1949, vol. I, pp. 399-418.


eastward from the Incasic, northwesterly from the Tupí, and centrifugally from the Plateau and Patagonian centers. The Sêca and Chilean are negative areas; that is, they receive species from elsewhere but do not disperse them.

    Using these terms, the distribution of the vectors here concerned is described in table 16.

TABLE 16.- Distribution of major mosquito vectors by zoogeographic subdivisions in the neotropical region

Secondary vectors

    Besides the principal vectors, just discussed, the neotropical region includes a number of additional species which are believed to convey malaria at times, especially when abundant.

    Gabaldon 25 recognizes as secondary or minor vectors, Anopheles crucians, A. grabhami, A. vestitipennis, Anopheles evansi, Anopheles nunez-tovari Gabaldon, Anopheles triannulatus, and Anopheles cruzi. All these have at one time or another been found infected in nature, but it is believed that, if these species were to be eliminated, the total amount of malaria in most localities would remain the same. To these vectors, one might add A. homunculus , 26 a bromeliad breeder, somewhat similar in habits to A. bellator .

25 Gabaldon, Arnoldo: Malaria Incidence in the West Indies and South America. In Malariology, edited by Mark F. Boyd. Philadelphia: W. B. Saunders Co., 1949, vol. I, pp. 764-787.
26 Downs, Wilbur G., and Pittendrigh, Colin S. : Malaria Transmitted by Bromeliad-Breeding Anophelines. In Malariology, edited by Mark P. Boyd. Philadelphia: W. B. Saunders Co., 1949, vol. I. pp. 736-748.


    Anopheles pessoai Galvâo and Lane is considered a secondary vector in Brazil, along with A. albitarsis. 27 Anopheles neivai 28 is also a possible minor vector.

Effect of Military Occupation on Malaria Potentialities

    Increase of mosquito production due to alteration of the terrain by combat was, of course, no factor in these areas. Similar problems arose, however, in connection with the building of roads, landing fields, and port facilities, where excavations and other manmade impoundments due to engineering activities remained to be filled by subsequent rains and thus became potential mosquito breeding situations. In some cases, incidence of malaria among troops was traced definitely to such conditions, and their correction brought about marked improvement in malaria rates.

    Deliberate improvement of terrain either by the military alone or in cooperation with native governments was carried out in the vicinity of practically every installation of any size. In most places, a substantial reduction of mosquito vectors resulted. Exceptions occurred where an attempt was made to apply general measures without knowledge of the biology of specific vectors. For example, no amount of attention to salt marshes could possibly affect the breeding of A. bellator in epiphytic bromeliads.

    In general, it may be said that the longer U.S. troops remained at Caribbean and South Atlantic bases, the lower did the incidence of malaria vectors in the vicinity of such points become. This was especially true after malaria survey and malaria control detachments began to function. It is significant that the bulk of such accomplishment took place before the general availability of DDT.

    Two possibilities existed regarding the enhancement of the gametocyte reservoir. Assuming no essential change in the concentration of native population, one has merely the problem which results through infection of troops and civilian newcomers from the endemic reservoir. To the extent that new arrivals become active cases and eventually carry gametocytes in their blood, these cases add to the local reservoir and increase the possibility of epidemic manifestations. Under routine military procedure, however, such cases were quickly diagnosed, hospitalized, given drug therapy, and though not necessarily cleared of all infection, rarely carried the gametocyte form of the parasite after return to active duty. Suppressive Atabrine therapy particularly tended to eliminate gametocyte forms of Plasmodium vivax and Plasmodium malariae.

    The second aspect of the problem, and the more important, relates to the concentration of native labor for the carrying out of engineering projects. Such labor is often recruited from scattered areas, including some relatively remote

27 Deane, L. M., Causey, O. R, and Deane, M. P.: Studies on Brazilian Anophelines from the Northeast and Amazon Regions. Baltimore: The Johns Hopkins University Press, February 1946. (The American Journal of Hygiene Monograph Series No. 18.)
28 Rey, H. : Revista de nuestros conocimientos sobre vectores de malaria en Colombia. Tijeretazos sobre Malaria 11: 30-32, March and June 1947. Abstr. Trop. Dis. Bull. 45:1053, December 1948.


points. In tropical regions, a certain percentage of these persons is almost certain to be malarious, and a lesser percentage actively infective. If a vector is present, these gametocyte carriers act as a seedbed for infecting the entire labor camp, and a sharp rise in the malaria rate may be expected. The results are serious and manifold: (1) The efficiency of the labor camp is greatly reduced by reason of the hospitalized cases; (2) the hospital and dispensary services of the base or installation may be strained by the double load of military and civilian cases; (3) laborers who are not too ill to work may nevertheless be gametocyte carriers and thus contribute to the further spread of infection; and (4) many mosquitoes then have access to gametocyte. carriers which would normally have bitten either normal or noninfective persons. The number of infective anophelines in the vicinity of military personnel thus becomes tremendously increased.

    All this can take place even when first energies are devoted to terrain improvement with a view to malaria control; that is, while ditching, filling, and the installation of tide gates are in progress. Much trouble can be forestalled, however, if less expensive control procedures are practiced in the beginning. These include administration of prophylactic chemotherapy to all exposed persons, larviciding of obvious mosquito-producing areas, screening and spraying of barracks, and use of repellents in the field. These are the same procedures which constitute the total and permanent program of installations too small in size to warrant large-scale engineering procedures of an expensive nature.

    While these precautionary measures are in operation, mosquito and parasite surveys should be continuous in order to evaluate their efficacy and to form the basis of recommendations for expansion, modification, or abandonment of specific practices. In tropical America, as in other theaters, personnel qualified to conduct such surveys were not available in organized groups (malaria survey detachments) until long after most installations had been fully developed. Military engineers, in cooperation with civilian health authorities, did commendable work, however, in keeping the incidence of malaria among troops and civilian laborers at a desirable low point.