Allochthonous Parasites in the Americas

Previous Syntheses

Some researchers have investigated the introduc­tion of specific diseases to the Americas in general Table V.9.1. Diseases suggested to have been introduced to the Americas

(e.g., Stearn and Stearn 1945), whereas others have studied the introduction of all diseases to a specific region (e.g., Cook 1982).

It will be noted that none of the lists includes epidemic (venereal) syphilis or tuberculosis. The omission of these two major diseases should not be interpreted as supporting the notion that either dis­ease originated in the Americas. Controversies over their origins are extensive (regarding syphilis: Crosby 1972; Ramenofsky 1987; Baker and Arme- Iagos 1988; regarding tuberculosis: Black 1980; Clark et al. 1987; Ramenofsky 1987). Until this ques­tion is resolved, these diseases cannot be considered to be either allochthonous or autochthonous to the New World.

There are some disagreements about specific intro­ductions that stem from the temporal or intellectual focus of the work. In 1972 A. Crosby considered only sixteenth-centmy introductions. In 1986 he adopted a much larger temporal framework; his suggested introductions reflect that change. Although in 1983 H. Dobyns described more introduced infections than those listed in Table V.9.1, the summary is limited to those diseases that arrived during the sixteenth and seventeenth centuries. My own list (1987) is also limited to the first two centuries of European contact.

In developing that list, I weighed evidence from all sources described previously.

Typhoid (enteric) fever was excluded for several reasons. Although native populations of the Ameri­cas certainly suffered from diarrheal infections in the postcontact period, the question becomes whether these infections were new and whether Salmonella typhi caused some of the infections. The first question cannot be answered; the second can be addressed indirectly. The disease was not isolated from typhus or other nonspecific childhood fevers until the nine­teenth century (Creighton 1894; Overturf and Un­derman 1981). In addition, although rare, typhoid fever can take a chronic form and can spread to hu­mans from nonhuman reservoirs including turtles (Youmans et al. 1980). The combination of a chronic state and the indisputable presence of turtles in precontact America suggested that S. typhi could have been present in the Americas before 1492.

The diseases that I listed as introductions, but not treated by Crosby or Dobyns, include anthrax, ru­bella, and pneumonia. Anthrax is typically a rela­tively minor infection in humans. Domestic stock, cattle, sheep, horses, and goats are the primary source of the infection; humans become ill through accidents of proximity (Whiteford 1979). In those areas, such as the Southwest or the pampas of South America, where domestic stock were part of the cul­tural baggage of European colonists, anthrax could have been a source of infection.

Rubella was added to the list for two reasons. First, although a relatively mild disease in adults, it can seriously affect the reproductive fitness of a population. If a woman develops German measles during the first trimester of pregnancy, the fetus may be bom with major congenital defects, includ­ing cataracts, heart disease, microcephaly, and men­tal retardation (Top 1976).

The bacterial pneumonias are caused by a group of unrelated organisms. Some of these (Streptococcus pneumoniae) are part of the normal flora of the hu­man upper respiratory tract. Thus, some causative agents of severe lung infections in humans had to have been present in the Americas before European contact. In contrast, contemporary studies (Brenne­man et al. 1987) suggest that some native American populations are at higher risk from other agents of pneumonia (Hemophilis influenzae). This elevated risk may indicate that not all causative bacteria of pneumonia were present before Europeans arrived in the Americas. Moreover, because pneumonias are major secondary invaders that follow such viral in­fections as measles and influenza, and because both measles and smallpox were allochthonous to the Americas, it is likely that at least some types of lung infection spread to the Americas with Europeans. As health deteriorated from viral infections, bacterial agents could invade and cause death.

Although both Dobyns and I view epidemic typhus as an Old World introduction to the Americas, H. Zinsser (1947) thought that it predated European contact. The vector of typhus is the human body louse. The current worldwide distribution of people means that epidemic typhus is also worldwide (Wisseman 1976). Although it is likely that body lice predated Europeans in the Americas, I think it un­likely that typhus-infected lice were present. The disease thrives under conditions of intense crowding, poor sanitation, and social or economic upheaval asso­ciated with war. Despite the archeological evidence of warfare in the Americas, the distribution of human populations was inappropriate for fixing epidemic typhus as part of the disease load of native popula­tions.

In summary, a minimum of 11 and a maximum of 14 viral, bacterial, or protozoal diseases are sug­gested in Table V.9.1 as having diffused to the Ameri­cas during the first two centuries of European con­tact. There is unanimous agreement on five disease introductions (influenza, measles, smallpox, scarlet fever, and yellow fever). I have increased the total introductions from 11 to 14 by adding anthrax, ru­bella, and pneumonia.

Current Synthesis

After reexamining the evidence for and against the introduction of diseases listed in Table V.9.1,1 have removed one disease. Chickenpox is omitted from Table V.9.2 because it is present in small nucleated populations that lack antibodies for acute, infectious Table V.9.2. Viral, bacterial, and protozoal agents introduced to the Americas

microbes (Black 1980). In addition, Varicella zoster can follow a chronic course, expressing itself as chickenpox in children and as shingles in adults (Brunell 1976).

Table V.9.2 lists viral, bacterial, and protozoal agents, characterized by two modes of transmission. Eight of the 13 viruses (influenza, measles, mumps, rubella, smallpox) and bacteria (pneumonia, scarlet fever, pertussis) colonize humans only and are trans­mitted by them only. Of the remaining 5 diseases, yellow fever is viral; anthrax, plague, and typhus are bacterial; and malaria is protozoal. All 5 of these agents are classified as zoonoses, meaning that the primary reservoir is a nonhuman, invertebrate or vertebrate species (except in rare cases, e.g., pneumonic plague [Benenson 1976b]).

Previously discussed difficulties of reconstructing past epidemics by a single method have direct impli­cations for the current synthesis. I will not attempt to reconstruct the precise date or port of entry of each disease introduction. My goal is to build gen­eral expectations about the postcontact spatial per­sistence of parasites, and to accomplish this I will use data on transmission cycles, evolution, ecology, and history.

For those viruses and bacteria that colonize and reproduce only in humans, density of human popula­tions, regularity of communication, and incubation period of the parasite determined the size of spatial epidemic waves. As population density decreased or communication became irregular relative to the in­cubation period of the parasite, the parasite died out. Whether a second introduction of the same parasite caused another epidemic outbreak varied according to the number of reproductively active survivors and time. Although individuals who escaped infection during the first disease event could be subsequently infected at any time, the size of the susceptible pool determined whether any new outbreak would be lo­cal or regional in scale (Bartlett 1956; Black 1966; Cockbum 1971).

Recently, the 1520 introduction of smallpox to the Americas has been a subject of some interest. It is a classic example of disease transmission and spatial diffusion. On the basis of information drawn from historical documents, Dobyns (1983) has argued that the virus became the initial New World pandemic, spreading as far south as Chile and as far north as the Canadian border. The Caribbean islands were the initial focus of infection in 1518. The virus was then carried to Mexico by a crew member of Panfilo de Narvaez’s expedition. The large size and extreme density of aboriginal populations in the valley of Mexico encouraged the rapid dissemination of the virus.

The question of spatial diffusion beyond the valley of Mexico has been investigated by historians and archeologists. Noble Cook (1981), historian, has not found evidence of spatial diffusion into the Peruvian Empire. A second introduction of smallpox to Pan­ama in 1524 did, however, spread down the Andean chain (Cook 1982). Using climatic data to explain the persistence of smallpox, S. Upham (1986) argued that the smallpox epidemic did spread to southwest­ern groups. Other archeologists (Ramenofsky 1987; Campbell 1989) have relied on archeological indica­tors of population change and time.

In summary, for directly transmitted viruses and bacteria, the distribution of susceptible populations, communication systems, and incubation periods of the parasite determined whether local disease events became regional or multiregional. Even with detailed historical records, it is difficult to define the spatial extent of a specific parasite. Although the scale of resolution currently obtainable from re­gional archeological data bases may be sufficient for concluding that acute infectious microbes caused ca­tastrophic die-offs, current knowledge is simply in­sufficient for stipulating whether one or another epi­demic event was causal.

Transmission cycles of zoonotic infections are fun­damentally different from infections transmitted solely between humans. Because vertebrate or inver­tebrate species other than humans are the primary reservoir of the parasite, niche requirements of the reservoir and microbe determined the postcontact spatial pattern across the American continents. Hu­mans became infected through accidental interac­tions with the reservoir and causative agent. The following focuses on those aspects of reservoir or microbial niches pertinent to reconstructing the postcontact spatial patterning of these infections.

Temperature severely Cintails the distribution of yellow fever. The viral disease is transmitted to hu­mans or other nonhuman primates by several mos­quito vectors, including Aedes and Haemagogus (Taylor 1951). The optimal temperature for incuba­tion of the arbovirus and the transmission ability of the vectors is approximately 30^oC (Whitman 1951). At temperatures less than 24°C Haemagogus, for instance, cannot transmit the disease, and the incu­bation period of the virus lengthens. In addition, the vectors have preferred breeding habitats. Aedes aegypti breeds in clay-bottomed containers; other species of Aedes and Haemagogus prefer breeding in trees of climax rain forests (Carter 1931).

Like yellow fever, malaria is a vectorbome dis­ease that is transmitted to humans by numerous species of Anopheles. The temperature limitations and elevational preferences of these species vary greatly. Whereas Anopheles maculipennis can repro­duce in cold pools along lake margins at elevations greater than 4,000 feet in the western United States, Anopheles quadramaculatis prefers breed­ing in swampy nonbrackish pools or bayous typified by the Lower Mississippi valley (Hackett 1941; Wat­son and Hewitt 1941). In addition, the body tem­perature of the vector affects the reproductive poten­tial of the protozoans. Plasmodium υiυax and Plasmodium malariae will not develop in anophe- Iines if the body temperature is less than 15^oC; Plas­modium falciparum requires an anopheline body temperature of greater than 18^oC for reproduction (Zulueta 1980).

Epidemic typhus is an acute infectious disease for both vector and human populations. The human body louse, Pediculus humanus, is the vector; the causative agent is a bacterium, Rickettsia prowaze- kii. After ingesting the agent, the typhus-infected louse dies within 7 to 10 days. When the louse feeds on human blood, the rickettsial agent is transmitted to humans (Zinsser 1947; Wisseman 1976).

Plague and anthrax are bacterial infections trans­mitted to humans from nonhuman mammalian res­ervoirs. The distribution of the reservoir largely de­termines the distribution of the disease. Wild rodent populations in wooded or desert areas are the true reservoir of the plague bacterium: In these set­tings, the disease is endemic (Meyer 1963; Benenson 1976b). In urban areas, the causative agent, Yer­sinia pestis, is transmitted to humans through a secondary vector, the rat flea, Xenopsylla cheopis. The proximity of humans to rats and their fleas creates a tertiary focus of the disease.

As previously mentioned, anthrax is a bacterial disease of domestic ungulates. The causative agent, Bacillus anthracis, is transmitted to humans through the ingestion of contaminated meat or milk products, contact with infected animals, or the inhalation of viable airborne spores (Brachman 1976; Whiteford 1979).

Because not all New World habitats were appropri­ate for the reproduction of reservoirs, vectors, and parasites, it is likely that zoonotic infections were more limited spatially than were directly transmit­ted viruses and bacteria. Yellow fever became fixed in tropical climates; malaria survived between 60^o north latitude and 40^o south latitude (Watson and Hewitt 1941); epidemic typhus survived where hu­man groups were concentrated; herds of domestic ungulates became the focus of anthrax; bubonic plague was fixed as a disease of ports. As rats mi­grated away from ports, plague also migrated.

The variability in reproductive requirements and, therefore, transmission cycles of parasites has impli­cations for the postcontact spatial patterning in the Americas. Because all native populations were vir­gin soil, the accidental introduction of any parasite could decimate a single population aggregate. Whether the parasite diffused spatially and ac­quired an endemic status in the rich American envi­ronment depended on a number of factors.

The probability of fade-outs of directly transmit­ted viruses and bacteria varied according to stochas­tic contacts between infected and susceptible hosts, the overall distribution of population, and the time separating introductions of the same parasite. As just described, the spatial pattern of zoonoses de­pended on reproduction requirements of reservoirs, vectors, and parasites. Certainly by the seventeenth century, natural selection had fixed the spatial vari­ability of all introduced parasites. Like threads join­ing patches of a quilt, communication mechanisms linked sources of infection to potential recipients. Each epidemic wave further reduced survivors of previous disease outbreaks.