Scientific Evidence for Pre-Columbian Transoceanic Voyages to and from the Americas

John L. Sorenson1 and Carl L. Johannessen2

Introduction

This paper is an expanded version of a presentation given at a conference, "Contact and Exchange in the Ancient World," held at the University of Pennsylvania, Philadelphia, May 5, 2001. The conference was organized by Victor H. Mair of the Department of Asian and Middle Eastern Studies at the University of Pennsylvania. He is also editor of the volume of papers from that conference in press at the University of Hawaii Press in 2004.

Since our initial paper was submitted for inclusion in that volume, we have made further discoveries. The present book incorporates the new materials, constituting a revision and extension of the original paper. Because much of the literature that enters into our argument in the extended paper is interpreted here in ways other than biologists conventionally do, for readers' convenience we give in the appendix précis of our reference materials on each species discussed. Selected illustrations and a bibliography for both the text proper and the appendix follow.

Support for the preparation and presentation of the original paper was provided by the Institute for the Study and Preservation of Ancient Religious Texts, at Brigham Young University, and the Center for Ancient Studies of the University of Pennsylvania. We express gratitude to those organizations, but, of course, we authors alone are responsible for the views expressed and for any errors. Our appreciation also goes to Linda S. McElroy for her helpful editing of the present manuscript.

Abstract

Examination of an extensive literature has revealed conclusive evidence that nearly one hundred species of plants, a majority of them cultivars, were present in both the Eastern and Western Hemispheres prior to Columbus' first voyage to the Americas. The evidence comes from archaeology, historical and linguistic sources, ancient art, and conventional botanical studies. Additionally, 21 species of micro-predators and six other species of fauna were shared by the Old and New Worlds. The evidence further suggests the desirability of additional study of up to 70 other organisms as probably or possibly bi-hemispheric in pre-Columbian times. This distribution could not have been due merely to natural transfer mechanisms, nor can it be explained by early human migrations to the New World via the Bering Strait route. Well over half the plant transfers consisted of flora of American origin that spread to Eurasia or Oceania, some at surprisingly early dates.

The only plausible explanation for these findings is that a considerable number of transoceanic voyages in both directions across both major oceans were completed between the 7th millennium BC and the European age of discovery. Our growing knowledge of early maritime technology and its accomplishments gives us confidence that vessels and nautical skills capable of these long-distance travels were developed by the times indicated. These voyages put a new complexion on the extensive Old World/New World cultural parallels that have long been controversial.

The Problem

In general, scholars concerned with the ancient culture history of the Americas believe that there were no significant connections by voyaging between the Old World and the New World before 1492. To the contrary, our data from an extensive literature that hitherto has been inadequately searched demonstrate that fauna and flora were extensively shared between the Old and New Worlds before Columbus' discovery of the Americas. The only plausible explanation for this bi-hemispheric distribution is that those shared organisms moved across the oceans via intentional voyages that took place during the eight millennia or more immediately preceding Columbus' discoveries. This book presents and documents the evidence for our position. We believe students of the human past are obliged to adopt a new paradigm for the role of long-distance sea communication in history and culture.

In the past, arguments for transoceanic contacts have relied mainly on evidence from cultural parallels (Sorenson and Raish 1996). Some of those parallels are indeed striking, but scholars generally have rejected their value as evidence that significant pre-Columbian contacts took place with the Americas across the oceans (see, e.g., Kroeber 1948, 538—71; Rands and Riley 1958). Over a century ago, Tylor (1896) compared details of the Aztec board game, patolli (e.g., the board's layout, the sequence of moves, and cosmic associations of the pieces and moves), with the game called pachisi in India. Even Robert Lowie, an influential anthropologist who was usually critical of diffusionist (voyage-dependent) explanations for such similarities, accepted that in this case "the concatenation of details puts the parallels far outside any probability [of having been invented independently]" (1951, 13). Still, tentative acceptance by some influential observers, like Lowie, of the possible historical significance of the cultural parallels has always ended up being rebutted by a demand from critics for 'hard,' or 'scientific' evidence for voyaging. Often, the sort of evidence demanded was demonstration that numbers of plants were present on both sides of the oceans before Columbus' day (Kidder et al. 1946, 2).

Data from the life sciences now provide that desired evidence, not only for the flora, but for fauna as well. The largest body of data is on plants. (See Tables 1, 2, and 3, below.) We will also deal with shared infectious organisms as evidence for voyaging (see Tables 4 and 5), as well as some larger animal forms (see Tables 6 and 7).

An example from the history of health-damaging organisms demonstrates our approach while illustrating the power this kind of evidence can provide in the study of human history and prehistory. The hookworm Ancylostoma duodenale causes one of the most widespread human ailments. The long-term prevalence of the hookworm in East and Southeast Asia makes that area the obvious source from which the organism reached the Americas.

A. duodenale was at first assumed to have been introduced by slaves brought from Africa. Early in the 20th century, Fonseca (not fully published until 1970) discovered the parasite in an isolated Amerindian population in the Amazon basin. Shortly afterward, microbiologist Samuel Darling (1920) pointed out that the hookworm apparently had infested South American tropical forest peoples since before Columbus arrived. If a date for the parasite in the Americas before European discovery could be proven, he observed, then the only explanation for the parasite in the New World would have to be that it arrived anciently via infected humans who had crossed the ocean—"storm-tossed fishermen," he ventured.

His reasoning sprang from facts about the life cycle of this worm. In one stage it must inhabit warm, moist soil (in a climate no colder than that of North Carolina today). At a later stage, the worms from the soil penetrate a human host's body and settle in the digestive tract. Immigrants who came to the New World in slow stages via Beringia would have arrived hookworm-free because the cold ambient conditions would have killed the parasite in the soil (Soper 1927; Ferreira et al. 1988).

The hookworm's pre-Columbian presence in the Americas was established authoritatively by Allison et al. (1973), who found traces of the pest in a Peruvian mummy dated about AD 900. Evidence from other mummies and human coprolites has since repeatedly confirmed the initial find (Araújo 1988; Reinhard 1992). In 1988, Brazilian scientists identified this parasite from remains excavated in eastern Brazil. A series of radiocarbon dates fixed the age at about 7,200 years ago. Given the inland remoteness of the site, the organism's arrival on the coast of the continent must have occurred centuries earlier.

Can these findings be said to establish conclusively that early human voyagers crossed the ocean to the Americas? Is there some explanation for the presence of the worm in the New World due to natural forces, independent of human beings? Absolutely not. There is no alternative explanation. Modern microbiologists continue to assure us that Darling's assessment was correct. Ferreira, Araújo, and Confalonieri (1982) say, "Transpacific migrants from Asia by sea must be one component of the ancient American population." Fonseca (1970) asserts, "shared species of parasite . . . make it inescapable that voyagers reached South America directly from Oceania or Southeast Asia." Ferreira and colleagues (1988) agree: "We must suppose that [the human hosts for the parasite] arrived by sea." Araújo (1988) confirms, "The evidence points only to maritime contacts" (emphases added).

This kind of fact, or secure inference, is vital in our present study. First, since A. duodenale could have arrived in the Americas only in the bodies of (presumably) Asians who came by sea, we can be certain that elements of some particular culture, as well as a set of Asian genes, arrived with them. Second, vessels capable of crossing or skirting the Pacific were already in use by the 6th millennium BC, and at least one of those craft actually reached South America, where its occupants passed the hookworm on to subsequent inhabitants.

A second species, Necator americanus, also called 'hookworm,' has been found in Brazil in prehistoric human remains of the same age (Ferreira et al. 1980, 65—7). Its reproductive cycle is similar to that of A. duodenale. Presence of this second organism confirms the fact that millennia ago nautical technology that would allow successful voyages across or around the Pacific existed in Asian waters.

Subsequently, archaeologists and paleo-pathologists have found decisive evidence in the Americas for the presence of 18 other infectious organisms from the Old World (see Table 4). Most of them could not have come with early Bering Strait migrants, and the others very likely did not. They include additional parasites, bacteria, viruses, fungi, and other micro-predators. Beyond the primary 20, there is enough evidence of transfer across the ocean of another 18 disease-causing organisms to call for further research on their pre-Columbian distribution.

The zoological literature also identifies six animal forms (see Table 6) documented as present in both hemispheres, and possibly 6 more (see Table 7).

The volume of evidence about plants obviously greatly exceeds that on fauna. We will first present salient data on the former, after a short methodological orientation.

Organisms, whether plant or animal, have special significance for the history of long-distance human movements. Biologists believe that a given species arises only once in the course of evolution because any new species develops within a unique set of environmental parameters that is found in only a single geographical location (see Zohary 1996, 156; for changes in thought on this topic, see Blumler 1992; 1996). Plant geographer N. Polunin (1960) stated the governing principle very clearly: "The chances that two isolated populations will evolve in exactly the same way are incalculably low," since, as Wulff (1943, 56) put it, "no two localities on earth are exactly alike in all . . . physico-geographical conditions" governing the evolutionary process. Stephen Gould (1994, 3) echoed the thought: "I regard each species as a contingent item of history . . . . [A] species will arise in a single place [and time] . . . ."

When the same species is found to have lived before Columbus in the New World as well as an ocean apart in Oceania, Asia, Europe, or Africa, that departure from the norm demands rational explanation. One might hurriedly conclude that certain seeds moved inter-continentally by the actions of winds or waves; however, few seeds are equipped to survive long while floating or to move great distances via wind (and no disease organism spreads in such a way). The odds for successful natural transport of plants are so slim (Guppy 1906, following De Candolle; Fosberg 1951) that anyone who claims a passive, natural mode of transport is obliged to demonstrate the possibility in the immediate case rather than merely to assume or assert it. By the same token, the explanation that humans transported some plant overseas requires empirical and logical support.

The Plant Evidence

Table 1 lists 98 plant species for which there is what we consider decisive evidence that the organism was present in both Eastern and Western Hemispheres before Columbus' first voyage. Table 2 gives 19 more species for which the evidence is significant, though less than decisive, and Table 3 adds 18 more species that deserve further research to determine their possible bi-hemispheric presence.

What do we consider decisive evidence? It can come from the discovery through archaeology of actual plant remains—macrofossils, pollen, phytoliths, DNA—manifesting the presence before AD 1492 of a particular species in the hemisphere where, according to botanists, it did not originate. A second source of evidence is historical documents—references to or descriptions of plants in ancient texts, explorers' reports, or lexicons of appropriate date that show knowledge of the species in the hemisphere where it did not originate. Detailed representations of plants in ancient art can be determinative also. Conclusive information may also come through well-informed inferences by botanists, based on such data as where a plant's relatives and possible wild ancestor grew (Zohary and Hopf 1993, chap. 1).

Table 1

Plants for Which There is Decisive Evidence of Transoceanic Movement

Species

Common Name

Origin

Moved To

Adenostemma viscosum

Americas

Hawaii,

Agave americana

agave

Americas

India

Agave angustifolia

agave

Americas

India

Agave cantala

agave

Americas

India

Ageratum conyzoides

goat weed

Americas

India, Marquesas?

Alternanthera spp.

Americas

India

Amaranthus caudatus

love-lies-bleeding

Americas

Asia

Amaranthus cruentus

amaranth

Americas

Asia

A. hypochondriacus

amaranth

Americas

Asia

Amaranthus spinosus

spiked amaranth

Americas

South Asia

Anacardium occidentale

cashew

Americas

India

Ananas comosus

pineapple

Americas

India, Polynesia

Annona cherimolia

large annona

Americas

India

Annona reticulata

custard apple

Americas

India

Annona squamosa

sweetsop

Americas

India, Timor

Arachis hypogaea

peanut

Americas

China, India

Argemone mexicana

Mexican poppy

Americas

China, India

Aristida subspicata

Americas

Polynesia

Artemisia vulgaris

mugwort

Eastern Hemisphere

Mexico

Asclepias curassavica

milkweed

Americas

China, India

Aster divaricates

Americas

Hawaii

Bixa orellana

achiote, annatto

Americas

Oceania, Asia

Canavalia sp.

swordbean, jack bean

Americas

Asia

Canna edulis

Indian shot

Americas

India, China

Cannabis sativa

hashish

Eastern Hemisphere

Peru

Capsicum annuum

chili pepper

Americas

India, Polynesia

Capsicum frutescens

chili pepper

Americas

India

Carica papaya

papaya

Americas

Polynesia

Ceiba pentandra

kapok, silk cotton tree

Americas

Asia

Chenopodium ambrosioides

Mexican tea, apazote

Eastern Hemisphere

Mesoamerica

Cocos nucifera

coconut

Eastern Hemisphere

Colombia to Mexico

Couroupita guianensis

cannonball tree

Americas

India

Cucurbita ficifolia

chilacayote

Americas

Asia

Cucurbita maxima

Hubbard squash

Americas

India, China

Cucurbita moschata

butternut squash

Americas

India, China

Cucurbita pepo

pumpkin

Americas

India, China

Curcuma longa

turmeric

Eastern Hemisphere

Andes

Cyperus esculentus

sedge

Americas

Eurasia

Cyperus vegetus

edible sedge

Americas

India, Easter Island

Datura metel

datura, jimsonweed

Americas

Eurasia

Datura stramonium

datura, thorn apple

Americas

Eurasia

Diospyros ebenaster

black sapote

Americas

Eurasia

Erigeron canadensis

Americas

India

Erythroxylon novagranatense

coca

Americas

Egypt

Garcinia mangostana

mangosteen

Eastern Hemisphere

Peru

Gossypium arboreum

a cotton

Eastern Hemisphere

South America

Gossypium barbadense

a cotton

Americas

Marquesas Islands

Gossypium gossypioides

a cotton (genes)

Africa

Mexico

Gossypium hirsutum

a cotton

Mexico

Africa, Polynesia

Gossypium tomentosum

a cotton

Americas

Hawaii

Helianthus annuus

sunflower

Americas

India

Heliconia bihai

balisier

Americas

Oceania, Asia

Hibiscus tiliaceus

linden hibiscus

Americas

Polynesia

Ipomoea batatas

sweet potato

Americas

Polynesia, China

Lagenaria siceraria

bottle gourd

Americas

Asia, East Polynesia

Luffa acutangula

ribbed gourd

Americas

India

Luffa cylindrica

loofa, vegetable gourd

Americas?

India, China

Lycium carolinianum

Americas

Easter Island

Macroptilium lathyroides

phasey bean

Americas

India

Manihot sp.

manioc

Americas

Easter Island

Maranta arundinacea

arrowroot

Americas

Easter Island, India

Mimosa pudica

sensitive plant

Americas

India, China

Mirabilis jalapa

four-o'clock

Americas

India

Mollugo verticillata

carpetweed

Eastern Hemisphere

North America

Monstera deliciosa

a climbing aroid

Americas

India

Morus sp.

mulberry tree

Eastern Hemisphere

Middle America

Mucuna pruriens

cowhage

Americas

India, Hawaii

Musa x paradisiaca

banana, plantain

Eastern Hemisphere

Tropical America

Myrica gale

bog myrtle

Eastern Hemisphere

North America

Nicotiana tabacum

tobacco

Americas

South Asia

Ocimum sp.

basil

Americas

India

Opuntia dillenii

prickly pear cactus

Americas

India

Osteomeles anthyllidifolia

Americas

China, Oceania

Pachyrhizus erosus

jicama

Americas

Asia

Pachyrhizus tuberosus

jicama, yam bean

Americas

India, China, Oceania

Pharbitis hederacea

ivy-leaf morning glory

Americas

India, China

Phaseolus lunatus

lima bean

Americas

India

Phaseolus vulgaris

kidney bean

Americas

India

Physalis sp.

ground cherry

Americas

China

Physalis peruviana

husk tomato

Americas

East Polynesia

Polygonum acuminatum

a knotweed

Americas

Easter Island

Portulaca oleracea

purslane

Americas

Eurasia

Psidium guajava

guava

Americas

China, Polynesia

Salvia coccinea

scarlet salvia

Americas

Marquesas Islands

Sapindus saponaria

soapberry

Americas

India, East Polynesia

Schoenoplectus californicus

bulrush, totora reed

Americas

Easter Island

Sisyrhynchium acre

a 'grass'

Americas

Hawaii

Sisyrhynchium angustifolium

blue-eyed 'grass'

Americas

Greenland

Smilax sp.

sarsparilla

Eastern Hemisphere

Central America

Solanum candidum / S. lasiocarpum

naranjillo

Americas

Oceania, Southeast Asia

Solanum nigrum

black nightshade

Eastern Hemisphere?

Mesoamerica

Solanum repandum / S. sessiflorum

Americas

Oceania

Solanum tuberosum

potato

Americas

Easter Island

Sonchus oleraceus

sow thistle

Americas

China

Sophora toromiro

toromiro tree

Americas

Easter Island

Tagetes erecta

marigold

Americas

India, China

Tagetes patula

dwarf marigold

Americas

India, Persia

Zea mays

corn, maize

Americas

Eurasia, Africa?

Some Particularly Salient Cases from the Flora

We have not treated all the species at equal length. In this section, we present some especially cogent cases. Data of lesser interest are postponed in the interest of clearer presentation of the primary argument, but in the appendix we present full data in our possession for every species.

Amaranthus spp.

The grain amaranths, Amaranthus hypochondriacus and A. caudatus, are both native to the New World (Sauer 1950, 612—13; 1967; Brücher 1989, 54—6), where they constituted important cereal crops. These plants have been grown for a long time in Asia as well.

Some of the significance of this American/Asian distribution was first presented in Jonathan Sauer's monograph on the grain amaranths, published in 1950 (588—614). He concluded (page 613) that ". . . there is a great, vaguely delimited grain amaranth region stretching all the way from Manchuria through interior China and the Himalaya to Afghanistan and Persia. A. leucocarpus [now known as A. hypochondriacus, from Mexico; see Brücher 1989, 56] and A. caudatus [from the Andes] are both grown throughout this [Asian] area." Culinary uses were similar in Asia and the Americas; after being parched or popped, "the seeds were sometimes made into balls, like popcorn, using a syrup or other binder, or the seeds were ground and the flour was stirred into a drink or baked in small cakes."

Sauer also observed (page 588) that, "The crop is scattered so widely through Asia and is so firmly entrenched among remote peoples that it gives a powerful impression of great antiquity in the area;" in fact, it would seem to have been present "from time immemorial." Yet, "the available Old World specimens represent nothing but a small sample of the diversity present in the American grain amaranths." Hence, "The conclusion appears inescapable that the grain amaranths are all of New World origin." However, species of amaranths that are used as potherbs or for decoration (or that are mere weeds) are another matter; those that grow in Asia are Asian natives (Sauer 1967). (In his later publications, Sauer showed reluctance to accept the early transoceanic contact that his first results implied, although the facts of the matter remained unchanged.)

In Mesoamerica, the cultivation of a grain amaranth began as early as 4000 BC (García-Bárcena 2000, 14), and wild species there provide plausible ancestors for the earliest domesticated amaranths. No comparable ancestors for the grain amaranths are known from Asia. While grain amaranths appear to be historically old in Asia on the basis of the evidence mentioned by Sauer, exactly when the transfer of the grain amaranths from their American area of origin took place went unspecified in the literature for many years.

Sauer (1950) was impressed that Bretschneider (1896, 406) presented "what seems to be a clear reference to a grain amaranth" in a Chinese text. The reference is to the Kiu Huang Pen Ts'ao (Jiu huang bencao) (Bretschneider 1882, 49). That volume includes descriptions and woodcut illustrations for 414 plants, based on older printed works, plus the author's own observations. The editor/author, Zhou Ding wang, died in AD 1425. The source of his citation for amaranth was a document written about AD 950 or the Prince of Shu, in modern Sichuan. It recognized six kinds of hien (xian), a generic name for a group of related plants. A "modern record of grain amaranths from the same area . . . gives the same name," Sauer adds. The pre-Columbian presence of grain amaranths in Asia that was indicated by their wide distribution is thus confirmed by this 10th century mention.

Archaeology now has confirmed a dramatically earlier date for the grain amaranths in Asia. Seeds of A. caudatus, along with A. spinosus, a thorny weed that also grew in pre-Columbian Mesoamerica (Sauer 1967, 1007; Miranda 1952—53, I, 215), have been excavated in India. They date to the 1000—800 BC period at the site of Narhan (Gorakhpur Dist., Uttar Pradesh) (Saraswat et al. 1994, 282, 284, 331). Such a date is congruent with the distributional evidence in Asia as interpreted by Sauer.

The idea of an early movement of amaranths to Asia is supported by information on maize (Zea mays). In the 1960s, a "primitive maize" was found in cultivation in the Himalayan country of Sikkim. Botanists judged that it most closely resembled the "wild maize" of "ancient Mexico, a fossil specimen of which was uncovered [in 1960] in a lower level of San Marcos Cave in [the Tehuacán Valley of] Mexico" (Marszewski 1975—78, 132—34, cf. 162; Dhawan 1964; Gupta and Jain 1973). The date for maize of this type in Mexico is not entirely clear; it could be as early as the 5th millennium BC, or it may fall in the 3rd millennium BC (Johnson and MacNeish 1972, 21ff.; Long et al. 1989). (See more on maize in Asia below.)

We have seen above that from the point of view of nautical history we not only can, but must think of watercraft as capable of trans or peri-oceanic sailing as early as this time frame. In recent years, serious proposals about early sailing capability have been made by archaeologists and others involved in research on the question of the early settlement of the Americas. Some see early sailing to be thoroughly plausible (Dixon 1993; Center for the Study of the First Americans 1999). Data on navigation in Near Oceania and Australasia have shown that there was a spectacularly early capability to cross stretches of open ocean. Australia was reached from Papua New Guinea or Timor perhaps more than 50,000 years ago. The Solomon Islands were inhabited nearly 29,000 years BP (before present) after an open-sea crossing of over 100 miles (Gamble 1993, 214—30). Bednarik (1997) has argued convincingly that Homo erectus in island Southeast Asia had "almost habitual use of navigation" of some sort by 840,000 years ago! He and his associates constructed a craft on the island of Timor using only Lower Pleistocene stone-tool technology. They used the craft to cross to Australia (2001). For more recent times, researchers have established that trading voyages thousands of miles in length were carried on in the Pacific millennia ago (Science News 1996; Service 1996; Jett 1998; Dickinson et al. 1999).

Nautical history and modern experimental voyages have demonstrated that oceanic voyaging in early times was not as daunting as many moderns have supposed, at least not to certain types of people (Jett 1971, 16—19; Helms 1988). In modern days, oceans have been crossed hundreds of times in unlikely craft ranging from midget and small boats, rafts, rowboats, and canoes to even less conventional vessels (see Anthony 1930; Barton 1962; Borden 1967). The ocean is not nearly as severe and fearsome a barrier to technologically limited voyagers as landlubbers have been wont to suppose. One experienced sailor of small boats in the tropics went so far as to assert, "It takes a damn fool to sink a [small] boat on the high seas" (Lindemann 1957). Despite well-documented evidence for the ancient capability for oceanic travel, a negative attitude has persisted among scholars that has been called "American thalassophobia" (illogical aversion to {considering} the sea as a route) (Elkin and MacIntosh 1974, 181) and "intellectual mal de mer" (Easton 1992). This phobia "cannot abide sea-travel as a mode of communication" (Meggers 1976); its grip has kept virtually all New World archaeologists from even inquiring whether shipping could have spread ancient cultures and people over long distances (e.g., Chard 1958).

Recently, however, realization has grown among up-to-date scholars that voyagers using simple technology could have reached the New World millennia ago. Over 40 years ago, archaeologist G. Bushnell (1961) granted that there was nothing physically impossible about vessels coasting round the North Pacific at any time after 8000 BC. Since then, Fladmark has argued repeatedly for a similar thesis (1979; 1983; 1986). Dixon considered it "not unreasonable" (1993, 119) to assume that watercraft were capable of moving along the coast from Asia by 13,000 BP. Six years later (1999, 31), he had changed that estimate to 16,000 BP. Nowadays, the coastal voyaging position is supported with increasing frequency (Engelbrecht and Seyfert 1994; Gamble 1993; Borg 1997; Dillehay 2000). A respected archaeologist, Dennis Stanford of the Smithsonian, has even proposed that Late Paleolithic (Solutrean) hunting people from Western Europe made their way around the ice-bound edge of the North Atlantic to settle in Late Pleistocene North America (Holden 1999a).

Support has continued more restrained for the idea of voyages directly across the Pacific. The hypothesis put forward 40 years ago, that voyagers bearing ceramics of the Jomon culture of Japan reached Ecuador around 3000 BC (Estrada and Meggers 1961), was accepted by a number of prominent archaeologists (e.g., Willey 1971, 16; Kidder II 1964, 474; Jennings 1968, 176, but with some subsequent hedging). Edwards (1965; 1969) and Doran (1971; 1978) presented many details about the nautical capability of Chinese sea-going rafts and made patent that the rafts of coastal Peru and Ecuador were explicitly parallel in form and capability to those of China, Indochina, and India (cf. Needham et al. 1985, 48—9). The work of Edwards and Doran has been readily available but widely ignored. There is no question that those rafts (more accurately 'ships') were capable of direct transpacific voyages. Although the date for historical documents on these Chinese and Southeast Asian ocean-going vessels only goes back to the 1st century BC (Ling Shun-sh∞ng 1956), the craft could easily prove to be much older (Edwards 1965, 98—100; Needham et al. 1971, 542—43).

In Peru, balsa rafts were in use along the shore by 2500 BC and ocean-going craft well before the 1st century BC (Norton 1987). Alsar (1973; 1974) demonstrated the feasibility of crossing the Pacific from east to west by sailing a fleet of three Ecuadorean-built rafts with a crew of 12 over 9,200 miles to Australia (the rafts even exchanged crew members at rendezvous points en route). Various forms of such rafts, in addition to large canoes, were used throughout much of Oceania (Clissold 1959). Our present state of knowledge about ancient nautics does not rule out voyages that could account for the early presence of amaranth, maize, the peanut, and other crops, as well as the hookworm, in both Asia and the Americas.

Arachis hypogaea

Early specimens of Arachis hypogaea, the peanut, have been found by archaeologists in China. Because the plant is a South American native, when specimens were found over 40 years ago at Neolithic-age sites in China, some biologists and archaeologists claimed that there was something wrong with the specimens or with the dates attributed to them (Harlan and de Wet 1973; Peng 1961). But the critics failed to examine the evidence carefully enough; the specimens were unquestionably peanuts, and the stratigraphy was sound.

Table 2

Flora for which Evidence is Significant but not Decisive

South America

Species

Common Name

Old World

New World

Acorus calamus

sweet flag

Origin

North America

Amanita muscaria

fly agaric

Origin

Middle America

Chenopodium quinoa

quinoa

Easter Island

Origin

Erigeron albidus

Hawaii

Origin

Gnaphalium purpureum

Hawaii

Origin

Indigofera sp.

indigo

Polynesia, Asia?

Mexico, Peru

Ipomoea acetosaefolia

Hawaii

Origin

Mangifera indica

mango

Origin

Middle America

Musa coccinea

Chinese banana

China

South America

Nelumbo nucifera

East Indian lotus

East Indies, India

Tropics

Nymphaea nouchalli

Indian water lily

India, Egypt

Tropics

Phaseolus adenanthus

Polynesia

Saccharum officinarum

sugar cane

Origin

South America

Salvia coccinea

scarlet salvia

India

Origin

Salvia occidentalis

a salvia

Marquesas

Origin

Smilax sp.

sarsaparilla

Eurasia

Middle America

Triumfetta semitriloba

Easter Island

Origin

Verbesina encelioides

Hawaii

Origin

Vitis vinifera

grape

Eurasia

Mexico

The peanut has since been discovered by archaeological excavation in caves on the island of Timor in Indonesia (Glover 1977, 43, 46). Nuts were found along with two other American plant species, the Annona squamosa fruit (custard apple) and Zea mays (maize), dated after the third millennium BC but before AD 1000, when the caves were no longer inhabited.

Safford had observed (1917, 17) that the kind of peanut found in graves at Ancón, Peru, was the same as that cultivated in China, Formosa, and India. Anderson (1952, 167) noted that "until the Peruvian excavations, the experts were certain that it [the peanut] came from the Old World, so widely is it disseminated there, with every appearance of having been grown for a very long time in Asia and Africa." More specifically, he added that "The most primitive type of peanut, the same narrow little shoestrings which are found in the Peruvian tombs, are commonly grown today, not in Peru, but in South China." Towle (1961, 42—3) later noted that one of two kinds of peanuts from the [Peruvian] tombs found in coastal sites is "similar to one grown in the Orient today."

Any question that Sinologists might have retained about the age of peanuts in China has been put to rest by more recent digging on the mainland. Some "10 or more" additional specimens of nuts have been found in the 3rd-century BC tomb of Western Han emperor Yang Ling in Xianyang, Saanxi (Chen Wenhua 1994, 59—60). Radiocarbon dates associated with the original two finds mentioned above put them as early as around 2800 BC (Chang 1973, 527). The more recent discoveries show the crop was still being grown two and one half millennia later, and use of the cultigen continues into modern times, as Anderson (1945) observed.

Table 3

Flora for which Evidence Justifies Further Study

Species

Common Name

Species

Common Name

Ageratum houstonianum

floss flower

Indigofera tinctoria

indigo

Annona glabra

pond apple

Lonchocarpus sericeus

Cajanus cajan

pigeon pea

Lupinus cruickshanksii

field lupine.

Cassia fistula

Lycopersicon esculentum

tomato

Cinchona officinalis

quinine (bark)

Nicotiana rustica

wild tobacco

Colocasia esculenta

dry-land taro

Ocimum americanum

hoary basil

Cucumis sp.

Paullinia spp.

Cyclanthera pedata

pepino hueco

Sagittaria sagittifolia

wapatoo

Datura sanguinea

Sesamum orientale

sesame

Derris sp.

Sisyrhynchium acre

Dioscorea alata

yam

Spondias lutyea

Dioscorea cayenensis

guinea yam

Spondias purpurea

hog plum

Dolichos lablab

Synedrella nodiflora

Elaeis guineensis

guinea oil palm

Tamarindus indicus

tamarind tree

Gossypium brasiliense

a cotton

Tephrosia spp.

Gossypium drynarioides

a cotton

Trapa natans

water chestnut

Gossypium religiosum

a cotton

Vigna sinensis

Hibiscus youngianus

Krapovickas (1967) compiled names for the peanut from Native American peoples in the Amazon Basin, the area where botanists think the plant was first domesticated. There, it bears such names as Tupí: mandobi, manobi, mandowi, mundubi, and munui; Pilagá: mandovi; Chiriguano: manduvi; and Guaraní: manubi. Black (1988) compared these terms with names for the peanut in India (taken from Kirtikar et al. 1935, 754—65) and found Sanskrit andapi; Hindi munghali; and Gujarati mandavi. These patent lexical parallels taken together with the plant specimens in Asia mean no less than that transoceanic voyaging was surely responsible for the plants and the names reaching Asia.

Erythroxylon novagranatense

The discovery of coca (Erythroxylon sp.) in Egypt was even more shocking. In western South America its leaves have been chewed for its chemical effect for more than 4,000 years (Plowman 1984; Shady S. 1997), although outside the Andean area where it is grown there is little evidence for consuming it. By what route the plant reached Egypt is unclear. Attempts to explain how its chemical signature came to appear in the Egyptian mummies without involving New World contacts seem outlandish.

About as strange as coca's use in the Near East is the fact that Peruvian mummies dating to before AD 200 have been found to contain both physiologically processed residues of tobacco and coca, as well as hashish (Cannabis sativa). Hashish is an ancient in the Old World (Parsche et al. 1993).

Despite the discomfort caused by having a paradigm upset, we must accept that the evidence is convincing for ingesting tobacco, coca, and hashish, at least in Egypt and in Andean South America by the beginning of the 1st millennium BC. The only explanations for that distribution involve voyages across the ocean.

Gossypium spp.

The early history of cotton (Gossypium spp.) in the New World is found, upon cytogenetic analysis, to be tied directly to Asian cotton. For over half a century most botanical models for the origin of the American cottons have depended heavily on the work by a group of cotton scientists led by Joseph Hutchinson with colleagues Silow and Stephens (1947). They argued that in order to rationalize the facts about the chromosome structure of later cottons, we must suppose that a diploid species, either Gossypium arboreum or G. herbaceum, had been carried from India to the Americas at an early date. Here, it hybridized with a Native American diploid lintless species to yield a tetraploid (having doubled the number of chromosomes in the diploid hybrid) cotton that had longer, more useful fibers. The genetic composition of all subsequent domesticated American cottons incorporates the Old World chromosomes of G. arboreum (or, alternatively, G. herbaceum) (the D genome) with A genome chromosomes from the American ancestor. Hutchinson et al. (1947) postulated that the first American hybrid subsequently became extinct, but that two of its descendant species, G. hirsutum and G. barbadense, became the bases for all subsequent domesticated cotton varieties in the New World.

So, when did G. arboreum reach the Americas? Archaeologists have found Mesoamerican specimens of G. hirsutum dating to the 4th or 5th millennium BC (MacNeish et al. 1967, 191), and G. barbadense was being grown in Peru by the mid-3rd millennium BC (Yen 1963, 112; Shady S. 1997). The time of arrival of the ancestral Asian diploid, of course, had to be earlier still, perhaps as early as the 5th millennium BC.

Before Columbus' time, American cottons were carried across the oceans to parts of the Old World. Historical documents from the Cape Verde Islands show that G. hirsutum, the Mexican species, had apparently reached the Guinea Coast of Africa, by means of some unrecorded voyage, before Columbus' first voyage. Cape Verdean records report that a cotton arrived there from West Africa some 25 years prior to 1492. Remnants of that cotton turn out to have the American tetraploid genetic structure (Stephens 1971, 413—14). (Arab sailors may have carried G. barbadense from the Americas to Africa, it was suggested by Jeffreys {1976}, perhaps along with corn, he thought.)

Cottons that evolved from the American tetraploid species were also found by the earliest European explorers when they reached islands in eastern Polynesia (Stephens 1947; 1963; Merrill 1954; Wendel 1989; Langdon 1982). (Johnson and Decker {1980} show that languages across the entire Pacific Basin have terms for cotton that also signify {fish}line.)

Recent genetic research has shown that a portion of the gene sequence of a local cotton species in southern Mexico, G. gossypioides, had one line of its ancestry direct from an African cotton (Wendel et al. 1995, 308—9). The genetic argument holds that some unknown but distinctively African cotton was introduced anciently to Mesoamerica. There, part of its gene sequence was incorporated by introgression into a pre-existing American species to yield G. gossypioides. We do not have direct evidence for a voyage, but no sequence of purely natural events can be conceived of to explain the arrival of the African element. Stephens (1971, 406—8) points out why an accidental, passive scenario for the spread by ocean drift of any variety of domesticated cotton is highly unlikely.

Despite the notable discoveries of the past half-century about the history of cottons, the comings and goings of those plants across the oceans still pose complex questions. The only answers that make sense, we feel assured, involve voyagers crisscrossing the oceans to an extent that no one anticipated 60 years ago.

Ipomoea batatas

The problem of the dispersion of the sweet potato has produced a huge literature over the past 75 years. At long last we have virtual resolution of several facets of the issue.

Ipomoea batatas has been found fossilized in a Peruvian cave dated perhaps as early as 10,000 BP (Brücher 1989, 5). The domesticated sweet potato was being grown in the precocious pre-ceramic city of Caral in the Supe Valley of Peru, a few miles from the coast, between 2700 and 2100 (calibrated radiocarbon dates) (Shady et al. 2001, 725). A name for Ipomoea batatas has been reconstructed for the proto-Mayan language in Guatemala before 1000 BC (Bronson 1966, 262 ff.) Later, common names for the plant in the Peruvian and Ecuadorian Andes, and probably on the coast as well, included variants of cumara or cumal (Patiño 1964, 62; Heyerdahl 1963, 29); speakers of the Chibchan family of languages in Colombia and Panama used a cognate name (see Kelley 1998, 73). Very similar names (see below) were applied later to the sweet potato in Polynesia. Although the claim was made as much as 75 years ago that the sweet potato spread into Polynesia only via Spanish exploring ships (Laufer 1929), traditions, linguistic studies among the island peoples, and explorers' historical records contradicted that position. Rather, those sources painted a picture of early canoe voyagers apparently carrying I. batatas tubers with them from the mainland to the islands (Dixon 1932), or else Amerindians were the carriers, westward. As late as the 1970s, passionate but ill-informed arguments (Brand 1971; O'Brien 1972) were still being offered to stifle the idea that Polynesian voyagers had reached the continent and obtained the sweet potato.

Yen (1974) capped a comprehensive investigation of the dispersion question by concluding that the data require transfer of the sweet potato from South America to Polynesia between AD 400 and 700 to account for its distribution. Since his study, archaeology has confirmed that view. For example, burnt tubers were found on Easter Island dated "early AD" by radiocarbon, while other evidence has also come to light for transfer to Polynesia prior to European exploration (Hather and Kirch 1991, 169; Yen 1998). Moreover, new linguistic studies have shown that American names for the tuber were brought into the eastern Pacific with the plant (Rensch 1991, 108; Kelley 1998). Rensch has found that the sweet potato reached Polynesia at least twice, once via introduction to Hawaii and once through Easter Island. Barthel (1971, 1165—86) reported deciphering an Easter Island text that refers to a directional model where the "path of the sweet potato" was from the east, while a "path of the breadfruit tree" signified from the west. Both fit botanical reality.

The sweet potato also reached Asia long ago, as Baker (1971) had thought. Bretschneider (1882, 38) reported that the Chinese document Nan fang Ts'ao Mu Chang mentions the I. batatas plant. The author was Ki Han during the Tsin (Jin) Dynasty, between AD 290 and 307. We note now complementary evidence from India. Aiyer (1956, 71) cites the Sanskrit name, valli, and Pullaiah (2002, II, 307) gives two more: pindalah and raktaluh. Aiyer (1956, 71) also reports mention of I. batatas in the Hindu text Silappadikaram. Yen, in a 1996 personal communication to Johannessen, reported that the sweet potato had the same name in Sanskrit as it had in northwestern South America. Kelley (1998, 72) has studied plant names for sweet potato and concludes that "An Indonesian word for 'yam,' *kumadjang, appears to have been borrowed by Quechuan and by Chibchan languages (of northwestern South America) and reapplied to 'sweet potato.'" From Quechuan it (the name) was "transferred to southern Polynesia, and from Chibchan it seems to have gone to Hawaii." It would seem that the voyages and botanical movements involving the sweet potato were complex, and that we have detected them only in part so far.

Nicotiana tabacum

The question of whether tobacco (Nicotiana tabacum), a plant native to the Americas, was found in ancient Egypt has arisen in the last quarter-century as a result of museum research. Fragments of tobacco were found about 30 years ago in the abdominal cavity of the mummy of Ramses II in a European museum (Bucaille 1990). In the intervening years, a large literature has arisen about the resulting controversy (the best survey is found in Jett 2003). In 1992, physical scientists in Germany used sophisticated instrumentation to examine nine Egyptian mummies in order to learn about the ancient use of hallucinogenic or narcotic substances. They found chemical residues of tobacco, cocaine, and hashish in the hair, soft tissues, skin, and bones of eight of the nine, including metabolically processed derivatives of the drugs, signifying that the drugs were ingested while the subjects were alive. (Cocaine and hashish, but not tobacco, were found in the ninth mummy.) The historical dates of the mummies ranged from 1070 BC to AD 395 (Balabanova et al. 1992a), indicating that the plants yielding the drugs were apparently continuously available, at least to Egyptian royalty, for over 1,400 years. Investigators have since found evidence of the drugs in additional mummies (Nerlich et al. 1995; Parsche and Nerlich 1995; Balabanova et al. 1997). Despite charges by critics that the analyses must have been faulty, the scientists involved have vigorously defended their work, and recent independent critiques give them good marks (Wells 2000; see also Pollmer 2000). All attempts to explain the unexpected findings without granting the presence of the American plants in the Old World have serious flaws (see, e.g., Buckland and Panagiotakopulu 2001; the critiques in Jett 2003 and 2004, and Wells 2000).

The nub of the controversy is, of course, that according to the standard paradigm of plant history, tobacco "should not" appear at all in Egypt or anywhere else in the Old World until after Columbus. Extravagant claims by Leo Wiener (1920—1922) about pre-Columbian tobacco in the Old World had been dismissed curtly by anthropologists and historians (e.g., Dixon 1920; 1921). But recently, Ashraf (1985) found that textual and artistic materials in India as far back as medieval times witness the presence and use of tobacco. The evidence includes a Sanskrit name, tanbaku (and/or támrakúta; Nadkarni 1914, 257), along with representations or mentions of the water-cooled pipe (huqqa, or hookah) for smoking, but other scholars do not seem to have paid attention to Ashraf's evidence. Now, the facts obtained from study of the Egyptian mummies leave no plausible explanation other than that N. tabacum was indeed used in antiquity in the Old World, and that could only be so if it was taken there at some early point in history by voyagers.

Zea mays

The literature on the question of whether maize (Zea mays) appeared in Eurasia before the time of Columbus has become large and contentious. Few botanists and even fewer archaeologists or historians have combed that literature exhaustively.

As noted above, field investigations have discovered odd sorts of maize growing in Asia (especially Sikkim Primitive in the remote Himalaya and 'waxy' varieties from Myanmar {Burma} all across China to the Korean peninsula), mostly away from coastal areas where 16th-century Iberian sailors are supposed to have first introduced maize. The characteristics and distribution of these grains cannot be explained in terms of post-Columbian introduction, because waxy varieties were not known in the Americas (Marszewski 1975—78, 1987; Johannessen and Wang 1998; Collins 1909; Stonor and Anderson 1949; Suto and Yoshida 1956; Dhawan 1964; Thapa 1966; Chiba 1968, 1969, 1970; Gupta and Jain 1973; The Wealth of India 1974). Yet, some unusual traits exhibited in these Asian maizes have close matches to corn known archaeologically from Peru (Towle 1961, 21—5), or that is still being grown by native groups in Peru, Colombia, Chile, Bolivia, and Argentina (Anderson 1945; Sarkar et al. 1974).

Utterly decisive evidence for the presence of American maize in the Old World has been found in Asian art and archaeology. Johannessen and colleagues (1998a; Johannessen and Parker 1989a) were the first to document extensively that maize ears were represented in sculptures of ears of corn——hundreds of them——on original temple walls in Karnataka State, southern India (details are found in the appendix). This art usually dates from the 11th to the 13th centuries AD, but some representations are much older.

Gupta independently identified maize, as well as a number of other plants of American origin, sculpted on Indian temples and monuments. For example, at the Lakshmi Narasimha temple, Karnataka, "Nuggehalli, the eight-armed dancing Vishnu in his female form Mohini, is holding a corn cob [ear] in one of her left hands, and the other hands hold the usual emblems of Vishnu" (Gupta 1996, 176). At least one maize representation dates from the 1st century AD (Cave Temple III, Badami, where, uniquely, the cob is held horizontally and the stem of the ear shows; Johannessen and Parker 1987b, 4).

In India, four Sanskrit words for maize have been recorded (Watt 1888—1893, VI, Pt. IV, 327; Balfour 1871—73, V, s.v. "Zea"), while the Garuda Purana (1980, 925, 947, 1128), as well as the Linga Purana (1973, 58, 85) texts of the 5th century AD refer to maize. The common name for maize throughout most of India is similar to that for the same plant among Arawak-language speakers of lowland South America (Johannessen 1992).

From near Zhenghou, Henan province, China, comes a ceramic effigy of a bird that was found in an excavation of an imperial tomb of the Han Dynasty. Impressions left on the interior of the figure show that a de-grained corncob had been used as an armature around which wet clay was modeled in avian form and then fired. The cob had, of course, burned up in the kiln. The age of the tomb is about 2000 BP (Johannessen and Wang 1998, 28). The same authors have published a bas-relief showing maize on a wall panel from the Prambanan Temple complex, east of Jogjakarta, Java, dating before AD 1000.

In the Middle East, maize was grown too, as Jeffreys long argued (1953a; 1953b; 1971; 1975; 1976). An Arabian juridical tradition, recorded in Yahya ben Adam's Kitab al Karaj (about AD 800), lists maize among other grain and pulse crops on which a tax could justly be levied (Ben Shemesh 1958, 77). Meanwhile, there is considerable reason to believe that maize was in Europe before Columbus' time (see, e.g., Finan 1948; C. Sauer in Newcomb 1963). Overall, there is not the slightest question remaining that maize was carried from the Americas to Asia millennia ago—at least once and perhaps multiple times.

Other Species of Flora from Table 1

We next present summaries of key evidence for other plants listed in Table 1. For maximum documentation on the entire list, see the appendix.

Agave sp.

The entire genus Agave is of American origin, yet it was represented in Old World biotas of several regions. Already by 1809, Lord Valentia, a traveler in India, observed that agave plants were "in such profusion that it is hardly possible to suppose it could have been introduced from America [i.e., after Columbus]" (Desmond 1992, 201).

When agave plants have been observed growing in Eurasia in recent centuries, scientists have always assumed that they were imports via Iberian ships after AD 1500. Thus, the shock was understandable when Steffy (1985) reported in a premier archaeological journal the discovery of 'agave' fibers mixed with pine resin that served as watertight caulking (between the hull and a sheet of lead lining) in a 4th-century BC Greek ship that had sunk at Kyrenia, Cyprus. Queried by us about the apparent geographical anomaly of agave in the Mediterranean, Steffy responded (2001), "You wouldn't believe how many people have protested that statement, but I was only repeating the identifications made by professionals [botanists] in respectable laboratories," starting with the Royal Botanic Gardens, Kew. (The genus, but no particular species, was identified.) Furthermore, he has been told by other Mediterranean archaeologists that they too have excavated agave specimens but have not reported the discoveries in print (probably because they anticipated hostility to the implications about transoceanic voyaging). Some feel that a plant of this genus must have been growing somewhere in the Mediterranean basin, Steffy continued, although no specific botanical evidence for that view has been put forward of which we are aware.

The plausibility of Steffy's find was recently supported by a datum first reported over 70 years ago. A Mexican archaeologist discovered an apparent Roman figurine head while excavating a site of Aztec age in central Mexico (García-Payón 1961). Two Mexican scholars recently tracked down that object, along with the archived excavation notes. They established beyond question that the head had come from beneath sealed floors where it had been buried no later than the1400s, at least a generation before Cortéz arrived. A thermoluminescence dating test on the terracotta head put its age long before Columbus. Stylistic analysis by experts on Roman art put the date of manufacture around the 2nd century AD (Hristov and Genovés 1999). At the least, this information shows that a (presumably) Roman ship very probably reached Mexico; a return trip could have carried agave plants, or at least the fiber, to the Mediterranean.

Anacardium occidentale

Anacardium occidentale is the cashew nut. This tree, native to the Americas, was introduced to India in the 16th century by the Portuguese, but it is now clear that it was present long before. The distinctive fruit was clearly represented on a bas-relief at the Bharhut Stupa (2nd century BC). Images of the nut had been carved adjacent to renderings of annona (custard apple), another American fruit (Gupta 1996, 17; Watt 1888—1893, I, 259; Cunningham 1879). In addition, the cashew bore a name in Sanskrit, bijara sala (Balfour 1871—73, I, 107; III, 409).

Anana comosus (syn. sativa)

Anana comosus, the pineapple, originated in Brazil. It is shown in a sculpture of the 5th century AD at a cave temple in Madhya Pradesh, India, as well as at a site in Gujarat, also in India (Gupta 1996, 18). A pineapple fruit pictured in an Assyrian bas-relief of the 7th century BC was confidently identified by the Assyriologist Rawlinson ("The representation is so exact that I can scarcely doubt the pineapple being intended") and confirmed by Layard, the excavator of the relief (Collins 1951). The fruit is also represented on artifacts from Egypt (Wilkinson 1879) and in an Ankara museum (seen by Johannessen 1998). Its presence on a Pompeii mural has been claimed; Johannessen's search at the site failed to confirm it.

Annona spp.

At least three species of the annona have been identified in art and referenced in the mythic literature of India. Images of the fruit and leaves of Annona squamosa at Bharhut Stupa place the plant in India by the 2nd century BC (Gupta 1996, 19—20; Cunningham 1879). The fruit is also seen carved on the gateway at Sanchi, on sculptures excavated from Mathura (Pokharia and Saraswat 1998—1999), and at the Ajanta Caves (Watt 1888—1893, I, 259). Johannessen and Wang (1998, 16—17) illustrate an A. squamosa fruit in the hands of a goddess sculpted on the 10th-century AD Durga Temple at Aihole, Karnataka, India. The fruit is also depicted at other Hindu and Buddhist temples in the states of Madhya Pradesh, Karnataka, Bengal, and Andhra Pradesh. For example, Bussagli and Sivartamamurti (1978, 189, Fig. 216) picture an 8th-century sculpture of Varuna, lord of the waters, seated with his consort on a makara monster and holding in his hand an annona fruit. Archaeological discovery of annona seeds in a cave on Timor, prior to AD 1000 at the latest, further confirms the Indian art (Glover 1977, 43, 46).

The plant is mentioned also in the traditional literature of India, where it is widely called sitaphala, 'the fruit of Sita,' because of a popular belief that Sita, wife of Ramachandra of the Ramayana epic, subsisted on the fruit of this tree while in exile (Gupta 1996, 19). In Kerala, the tree is called Ramachakkamaram, "tree bearing the fruit of Lord Rama" (Nicolson et al. 1988, 50). In Sanskrit, a short name of the tree is sita (Watson 1868, 527); it also has a second Sanskrit name, gunda-gutra, or gunda-gatra (Watson 1868, 181, 527), and Nadkarni (1914, 38) lists two other Sanskrit names, shubhâ and suda.

Annona reticulata and A. cherimolia were also present in India. In Sanskrit, the former was also called rama-sita, in addition to three other names (Watt 1888—1893, I, 125; Watson 1868; Torkelson 1999, 1646; Int. Lib. Assoc. 1996, 559). Bishagratna (1907, 72) reads it in a text assigned to the 6th century BC. It is still widely cultivated in Kerala, where it bears the name Ramachakkamaram (Nicolson et al. 1988, 50). Johannessen (Johannessen and Wang 1998, 156—57, Fig. 7) reports that annonas, including A. cherimolia, are shown in the hands of multiple sculpted figures on the walls of Indian temples, including the Hoysala temple at Somnathpur, Karnataka (AD 1268). Yet, as noted earlier, the annonas originated in South America, where A. cherimolia is known archaeologically in Peru before the beginning of our era (Towle 1961, 38—9).

Argemone mexicana

The Mexican prickle poppy is another plant of American origin. Today, it is widespread in India. Saraswat et al. (1994, 262, 333, 334) report mention of this plant in the Sanskrit medical treatise Bhava Prakasha by Sushrutha (1st and 2nd centuries AD). An archaeological find of A. mexicana seeds at Narhan in Uttar Pradesh assures us that the plant was being grown in India, possibly as a medicinal drug, as early as 1100 BC (Pokharia and Saraswat 1998—1999, 90, 100).

Artemisia vulgaris

This fragrant tree was well-known in Europe and Asia anciently. It was also present in Mexico, where it shared parallel cultural meanings (including medicinal use) associated particularly with the goddesses Artemis (Greece) and Chalchiuhtlicue (Mexico). Both of them were especially concerned with women and childbirth, as well as associated with water and marshes (Mackenzie 1924, 201—4; Roys 1931, 310).

Asclepias curassavica

The milkweed, Asclepias curassavica, originated in the Americas. In India, it is thoroughly naturalized and occurs commonly as a weed (Chopra et al. 1956, 28). Some Indianists have considered it to be the soma plant of antiquity (Watt 1888—1893, I, 343). A Sanskrit name attests to its age (Int. Lib. Assoc. 1996, 560). In both areas it is much used medicinally.

Bixa orellana

The small tree, Bixa orellana, serves widely in the tropics as a flavorant for food and as a red colorant for food, body paint, and dye (Donkin 1974). It originated in Brazil (Newcomb 1963, 41). We know it was used earliest in Peru as shown by excavations at the spectacular city of Caral on the central coast, which dates to 2700—2100 BC (Shady et al. 2001). Early botanists in India assumed this was an indigenous plant because it was so completely naturalized (Donkin 1974). It is not surprising that they should think so, for its having a Sanskrit name (Balfour 1871—1873, I, 177) means that it has been present in India for at least 1,000 years.

Cannabis sativa

For a brief discussion of Cannabis sativa, an Old World source for the psychoactive drug hashish/marijuana, see the earlier section on tobacco and coca. Hashish has been discovered in prehistoric Peruvian mummies extending over a number of centuries. See full references for Cannabis in the appendix.

Capsicum spp.

Chile peppers, Capsicum spp., are also American plants. Yet they are mentioned in the Siva Purana and Vamana Purana, Indian sacred texts dated to the 6th through 8th centuries AD (Banerji 1980, 9—10). The Sanskrit name, marichiphalam, was applied to both C. annuum and C. frutescens, says Nadkarni (1914, 86). The C. annuum plant and its fruit are naturalistically depicted in stone carvings both at a Shiva temple at Tiruchirapalli, Tamil Nadu (Gupta 1996, 50), and on a bas-relief dated earlier than the 10th century AD at the Prambanan temple complex east of Yogyakarta, Java (Johannessen and Wang, 1998, 28).

In Oceania, chili peppers were reported growing in Tahiti in 1768 by Bougainville, who reached that island only eight months after Wallis, its European discoverer (Langdon 1988, 334).

Carica papaya

One variety of papaya, Carica papaya, was grown in the Marquesas Islands before the arrival of Europeans (Brown 1935, 190). Its name signified 'papaya of the people,' as distinct from a larger, 'Hawaiian papaya,' that tradition says Christian missionaries brought from Hawaii. Brown was confident that the former fruit was pre-European. Because this species is of American origin (Safford 1905, 215—16; Zeven and de Wet 1982, 188), the implication is that voyagers had brought it to Polynesia.

Several lines of evidence support voyaging as the mechanism for the plant's arrival. The ethnographer Handy (1930, 131) reported a Marquesan legend to the effect that a double canoe of great size left the islands anciently in search of lands to the east. They were said to have reached a large land called 'Jefiti,' where they left some of their crew before returning to the islands. Sinoto's (1983) excavations in the Marquesas revealed that "great vessels" were being built there in the 9th century AD. Gifford (1924) considered that the use in the Marquesas of knotted cords as mnemonic devices is "strongly reminiscent of the Peruvian kipu," an idea previously put forward by von den Steinen (1903). Pérez de Barradas (1954) saw parallels in sculptural styles and in an unusual type of stone bead shared between San Agustín, Colombia, and the Marquesas. Heyerdahl (1996, 149—57) came across a stone statue in the Marquesas (Hivaoa) of a non-Polynesian style (stone sculpture in Polynesia is otherwise virtually unknown) that von den Steinen had earlier discovered there. It showed long-tailed quadrupeds that represent felids (Polynesia, of course, had no cats) and whose nearest analogs were on the monuments of San Agustín. The radiocarbon date on charcoal from beneath the Marquesan statue was ca. AD 1300. We have no hesitancy in accepting the early papaya as having been carried across the 4,400-mile ocean gap from the American mainland.

Ceiba pentandra (syn. Bombax malabaricum)

Ceiba pentandra is the kapok or silk-cotton tree of American origin. Brücher (1989, 146—47) considers the center of radiation for the species to be Central America, where it played a significant role in the cosmological myths of the Maya people. A name for the tree has been reconstructed in the proto-Mayan language of the 2nd millennium BC (Bronson 1966, 262 ff.). It also grew in Peru (Yacovleff and Herrera 1934—1935, 283).

In 1935, De Prez published a representation of the ceiba carved on a monument from Djalatounda, near Sourabaya, Java. This led him to characterize the species as "Indo-American." That monument is considered to date to AD 977 . Moreover, Chinese records report the tree growing during the Tang Dynasty (AD 618—907) on Hainan Island, where its fibers were being woven into fabric by the Li people (Schafer 1970, 64). At least five Sanskrit names for this species have been reported in India (e.g., Nadkarni 1914, 59; Pullaiah 2002, I, 147).

Chenopodium ambrosioides

Commonly known as 'Mexican tea,' Chenopodium ambrosioides has a clear record of ancient cultivation in Asia. When Fa Hien returned to his Chinese homeland in AD 414 from a long journey to Buddhist countries, his spotting this species under cultivation was one of the assurances by which he knew he was actually in China again (Bretschneider 1892, 261—62). But the plant was also ancient in India, for it has three Sanskrit names (Chopra et al. 1956, 61; Torkelson 1999, 1684; Int. Lib. Assoc. 1996, 562). Furthermore, the plant was known in Mesopotamia under the Arabic name, natna (Thompson 1949, I, 416—36). Yet the species was widespread in Mesoamerica. It was found archaeologically in southern Mexico as early as the beginning of our era (Martínez 1978, 123), and it bore a pre-Columbian name in Yucatec Mayan, llucum-xiu (Roys 1931, 262).

Cocos nucifera

The place of origin of the coconut, Cocos nucifera, has long been controversial. Early on, Cook (1901, 261—87) argued that the species was native to America, and other botanists agreed (Guppy 1906, 67; Bailey 1935). Counter arguments, however, later persuaded most plant historians that Asia or the western Pacific was its home (Hill 1929; MacNeish 1992, 259).

Those favoring an Asian homeland extended their argument to claim that the tree was wholly absent from the Western Hemisphere until the Spaniards introduced it. However, Heyerdahl (1965, 461) presented evidence that the tree was being planted for economic purposes on Cocos Island, near the west coast of Panama, before the first Europeans arrived there. Furthermore, traces of coconut fiber had been found in tombs at Ancón, Peru (Harms 1922), and representations of the coconut palm are seen in Peruvian art (Heyerdahl (1965, 461).

Patiño (1976, 54) documented at least five locations on the Pacific coast of the Americas where the coconut was reported by Spanish chroniclers to have been growing shortly after the Conquest and, presumably, before, also. Hernández, the 16th-century Spanish naturalist, casually reported the coconut in Peru and Mexico as being the same as in the Old World (1942—1946 [before 1580], II, 507—10). Balboa found coconuts on the Pacific Ocean side of Panama in 1513 (Mendez P. 1944). Even anti-diffusionist Merrill (1954, 267) judged the presence of the nut in pre-Columbian America to be "most certain," agreeing with cultural historian Julian Steward (1949), who accepted the ancient American occurrence of the coconut to have been "established beyond reasonable doubt." Most recently, and most definitively, Robinson et al. (2000) excavated coconut remains in Guatemala dated around AD 700, confirming a previous archaeological find at Copán, Honduras, that dated three centuries earlier. These data leave no question that the coconut was widely grown in the Americas before Columbus' time.

But might the tree have reached the Americas from across the Pacific by drift nuts that washed ashore and sprouted? Despite off-the-cuff claims that coconuts can drift long distances and remain viable when cast ashore, experimental research has shown that there is a limit of about 3,000 miles (Dennis and Gunn 1971). Pacific islands with coconuts are all more distant from the Americas than that. Ward and Brookfield (1992) did the definitive review of the literature, but it remains inconclusive on the question. They also did an extensive computer simulation of coconut drift that shows that it is highly unlikely that drift nuts would have reached the Americas from the nearest Pacific Island loci. (The coconut was absent from the Atlantic Ocean/Caribbean region until introduced there by the Portuguese—see Sauer 1993, 188). Whether drift nuts could successfully sprout and mature into trees on the strand has also been disputed contentiously.). Heyerdahl carried nuts both atop the raft Kon-Tiki and attached to the raft in the water, but only those that were on top germinated after the voyage (Heyerdahl 1950, 104, 204). Furt5hemore, no cases are known of coconuts having grown on Australian shores (Ward and Brookfield 1992). Carriage of the nuts by Pacific voyagers and intentional planting at an American destination is the explanation preferred by the best-informed scientists, and we think it is the only convincing scenario (Harries 1978, 271; Dennis and Gunn 1971).

Couroupita guianensis

The 'cannon-ball tree,' Couroupita guianensis, is native to South America and the West Indies. It now grows in South India where it is called naga lingam and has a unique meaning in Hindu symbolism. The tree bears flowers that grow directly out of the trunk and main limbs. Stamens and pistil of the blossom fuse in a manner that gives the appearance of a miniature lingam (stylized penis, symbol of cosmic generative power) facing the hovering hood of a naga (cobra). Naga lingam flowers in India today are left as offerings before sacred stone lingams in temples to honor Shiva (Johannessen, personal observation, 1996). The only reason one can imagine for anybody's transporting this New World tree to India would be that a worshipper of Shiva visited the Americas, where he saw the unique flower (no doubt with astonishment) and felt that it ought to grow in India, the homeland of Shiva. No non-Shivaite would have paid particular attention to the blossom, and indeed the plant plays no role in Mesoamerican iconography as far as we know.

Gupta (1996, 58) names four temples in Tamil Nadu state where the naga lingam flower is carved in stone. On the basis of this art, she considers that the plant reached India in "very early times."

Cucurbita spp.

Four species of American cucurbits were present in pre-Columbian Asia: Cucurbita ficifolia, the chilacayote; C. moschata, the crookneck squash; C. pepo, the pumpkin; and C. maxima, the winter, or Hubbard, squash.

That the moschata was an inter-hemispheric transfer has unusually clear evidence. The species' American origin is unquestioned. Archaeological remains in Peru go back to the 5th millennium before the present (Yarnell 1970, 225; Towle 1961, 69), and the plant may be earlier yet in Mexico (Brücher 1989, 26061).

C. moschata (Chinese, nangua) is mentioned in medieval Chinese medical recipes, for example, as recorded by Jia Ming (1966) in a text dated to AD 1473 (see Johannessen and Wang 1998, 25). Furthermore, the moschata squash shows up in Chinese paintings by Shen Zhou, who died in 1509 (before Magellan's arrival in Asia). Effigy pots modeled in the characteristic shape of moschata squash date to the Song (AD 960—1279) and Tang (AD 618—905) Dynasties (Johannessen and Wang 1998, 25, Fig. 10).

Vernacular names in use in India in the 19th century for this economically important plant included kumhra and kumra (Watt 1888—1892, II, 640), while from Yucatan we find the Mayan botanical term "Kum, or Kuum, Cucurbita moschata, Duch." (Roys 1931, 258). Roys quotes an early Spanish record that, "There are [in Yucatan] the calabazas [bottle gourds] of Spain, and there is also another sort of native ones [sic], which the Indians call kum," that is, the cucurbits (Roys 1931, 258). The similarities of names for the same species in Mexico and India strongly suggest, although of course they do not prove, that direct contact tied the areas together linguistically as well as botanically.

For another American cucurbit, C. pepo, the pumpkin, there is also good evidence that it grew in both hemispheres. Levey (1966, 315) discovered in a 9th-century medical text from India mention of the pumpkin as an ingredient (at least the term that was used was the one later applied to the pumpkin). Furthermore, the plant had three Sanskrit names (Watson 1868, 319, 327). Common names for the pumpkin in India also echo the kum root in Mayan: cumbuly, kumbala, kumhra, kúmara (Watt 1888—1893, II, 641; Watson 1868, 92, 119, 310, 311). And texts by the 16th-century Chinese botanist Li Shizhen, which were based on classic Chinese botanical and medical texts (Bretschneider 1882, 59), also mention the pumpkin. The writing of Jia Sixie in the 6th century does so too (Bretschneider 1882, 77—9).

Cucurbita ficifolia was once thought to be of Afro-Asiatic origin, but Whitaker (1947) showed that the species derived from a Central American wild cucurbit, and C. ficifolia also carried Native American names. In fact, Bronson (1966, 262 ff.) reported that a name for the species has been reconstructed for the proto-Mayan language dating to the 2nd millennium BC. Excavated specimens were found dated to the 3rd millennium BC at Huaca Prieta, Peru (Yarnell 1970, 225; Towle 1961, 90).

C.O. Sauer discussed this plant's presence as a crop (often grown for yak feed) in India, Tibet, and western China (Newcomb 1963, 29). In Asia, it is grown over an extraordinary range of environmental conditions, and accordingly its forms vary greatly, suggesting a long period of adaptation. The New World cucurbit does not vary so much. The Old and New World plants are morphologically indistinguishable and are fertile when crossbred. How and when did this cucurbit reach Asia, and how long did it take to adapt to such a wide range of environments there? Quite surely, that process required more than the four or five centuries since it might have been introduced to Asia by 15th- or 16th-century European ships (Newcomb 1963, 29). To us, the most believable explanation for the transfer involves a transpacific voyage to India that carried C. ficifolia along incidentally with more valuable cucurbits.

Cucurbita maxima is another plant that originated in the Americas but was grown in Asia. Its remains have been found by archaeologists in Peru dating before the beginning of our era (Towle 1961, 90), and Brücher (1989, 262—64) speaks of "irrefutable proofs" of its South American origin.

Nadkarni (1914, 129) gives Sanskrit, punyalatha and dadhiphala, for C. maxima, and its widespread cultivation and adaptation in Asia confirm its considerable antiquity. Levey (1966, 315) confirms C. maxima's pre-Columbian presence by finding the species mentioned in another 9th-century Indian medical text.

Mellén Blanco (1986, 211) interprets the account of the 18th-century González expedition to Easter Island as showing that C. maxima was growing there at the time of European discovery.

Curcuma longa

Curcuma longa, the turmeric plant, is of Southeast Asian origin (Newcomb 1963, 61; Brown 1931, 162—63). In Asia, it carried Sanskrit names, as well as terms in Hebrew, Arabic, and Chinese (Watson 1868, 189, 205—6, 248; Watt 1888—1892, II, 659; 1892, 1—3, 231, 417, 432). Nevertheless, the plant was grown and used by people in the remote eastern Andes in South America, where it was used in the same ways as in Asia (Sopher 1950).

Datura

One or more American species of Datura were used in Eurasia from ancient times as a medicinal, aphrodisiac, and hallucinogenic drug. The taxonomy of this genus has been confused, as historical sources have referred to its several species variously and imprecisely. But at least Datura metel (syn. meteloides, syn. innoxia, syn. fastuosa, syn. alba) and Datura stramonium (syn. tatula, syn. patula) were definitely present (Watt 1888, III, 40—1; Burkill 1966, I, 778—79). At least five Sanskrit names are known for one form or another of datura (Nadkarni 1914, 140—45; Chopra et al. 1956, 91; Chopra et al. 1958, 134; Torkelson 1999, 1711—12; Int. Lib. Assoc. 1996, 564; Watson 1868, 257). A fragment of a datura plant was excavated from the site of Sanghol in the Punjab that dates to between the 1st and 3rd centuries AD (Pokharia and Saraswat 1998—1999, 90). The Greeks are thought by some to have used a datura at Delphi to induce oracles, and the Romans are believed to have used D. metel (Burkill 1966, I, 778—79).

Meanwhile, several Datura species were commonplace in Mesoamerica and western North America (Hernandez 1942—43 [before 1580], II, 442; DuBois 1908, 72; Ramírez 2003). At least D. stramonium is thought to have had an American origin (Nicolson et al. 1988, 247). Reko (1919, 115) saw significance in the similarity between the Náhuatl (Aztec) name, toloache, and Chinese tolo-wan for Datura spp., while Kroeber considered that a coastal California cult that involved the practice of ingesting datura might have arrived from "islands of the Pacific" (Kroeber 1928, 395; DuBois 1908, 72).

Helianthus annuus

Helianthus annuus, the sunflower, is an American native, as hinted by the occurrence of a name for sunflower in the proto-Mayan language of the 3rd millennium BC (England 1992, 161). Furthermore, new archaeological finds of plant remains in southern Mexico (of the 3rd millennium BC; see Lentz et al. 2001) fix the domestication of H. annuus in Mesoamerica. Previously it had been supposed first domesticated in eastern North America.

Images of sunflower blossoms have been discovered by Johannnessen and Wang in the sacred sculptural art of India (Johannessen 1998b). In both the art and references in texts—some going back as much as two millennia—there is a strong association of this plant and its flower with the sun.

When sculpted in sacred Hindu art, the flower played a role in the cultural context of ancient Indian astronomy where it marked pivotal moments in the annual cycle of the sun. Given this ritual setting for the sunflower, it is no surprise that a Sanskrit name for the plant, suria-mukhi, was known (Watt 1888—1893, IV, 209; Torkelson 1999, 1749; Chopra et al. 1956, 131). Shivaite Hindu temples, especially those of the 11th to 13th centuries AD, were sometimes oriented to sunrise or sunset points on the horizon on key calendar dates. The flowers were also engraved on temple doorways and on images of Nandi, the bull that served as Shiva's steed, and by implication symbolized Shiva himself (Johannessen 1998b). (The fact that Heyerdahl {1986, 2, 176} found carved sunflowers designs on stones in the Maldives, located precisely at the equator, also could be part of this complex of ideas.)

Live sunflowers were known to Indian artists, which is particularly confirmed by a carving on a temple pillar at Pattadakal, Karnataka State (Johannessen and Wang 1998, 15—6). A long-tailed Indian parrot is shown sitting atop a large seed head, from the edge of which the bird has plucked a few seeds. No other seed-head looks like this nor has a stem capable of supporting a parrot. Gupta (1996) also discusses and illustrates sunflowers decorating Indian temples.

Johannessen has observed that maize images occur in temple statuary along with sunflowers, tying two American species together in a single ritual context. In Karnataka, sculptured female figures on temples are depicted holding ears of maize (see the discussion above), and they also have sunflower blooms below the bottoms of their skirts. (Different figures show whole, half, or quarter images of the flowers; Johannessen suggests that these images communicated information about the ritual calendar—see Johannessen and Wang 1998.) Moreover, a sunflower carved beneath the tail of a sculpted recumbent Nandi in the doorway of Halebid Temple in Karnataka is fully illuminated at each solstice sunrise (from alternate sides, winter and summer). The same Nandi allows dawn light to pass over the sunflower between its horn and its ear so as to penetrate deep into the temple and shine on the Shiva lingam in the inner sanctum, at equinox. Also, the carved blooms on the support for the tail of Nandi are completely shaded on the equinox (Johannessen 1998b, 354—60). At other Shiva temples, the solar azimuths differ, but the sculpted art typically involves the sunflower at sunrise or sunset as a calendrical indicator.

Lagenaria siceraria

The bottle gourd, Lagenaria siceraria, was discovered by Bird (1948) in pre-ceramic levels on the coast of Peru dating to the 3rd millennium BC. Its discovery was hailed by diffusionists as proof that ancient voyagers had crossed the Pacific (Carter 1950). Botanists previously had assumed that the species had not reached the Americas until imported by the Spaniards. Whitaker was the botanist who first identified L. siceraria at the archaeological site of Huaca Prieta, Peru (Whitaker and Bird 1949). He considered that, since bottle gourd plants from both hemispheres are closely comparable and occurred early in Peru, the burden of proof lay on those who, until then, had claimed that it had been colonial-era Europeans who introduced it from the Old World. It is also reported that the bottle gourd appeared in Mexico as early as perhaps 9000 BC; in Peru, by other researchers (Brücher 1989, 265; Heiser 1989, 475).

However, the species was found in 5th-Dynasty Egyptian tombs (Whitaker and Carter 1954, 697—700), and in India it bears Sanskrit names (Nadkarni 1914, 213; Pullaiah 2002, II, 323). Lathrap (1977) and Schwerin (1970) proposed that fishing boats bearing gourds drifted westward from Africa to South America at a very early date (14,000 BC according to Lathrop and 5700 BC according Schwerin). But there is no evidence for domesticated plants in West Africa at times even close to those dates. Patiño (1976, 244) and also Mangelsdorf, MacNeish, and Willey (1964, 441) doubted an Atlantic natural drift hypothesis because the bottle gourd was unknown in the Antilles or eastern South America. Brücher (1989, 265—66) wondered rhetorically "why it did not arrive before the Pleistocene, if the natural dispersal of floating fruits was so easy? Furthermore, "if Lagenaria had in one or another way come from Africa, the puzzling question remains how it crossed the whole American continent and appeared so early [in pre-2500 BC Peru] on the Pacific side?" Clearly, there is no good answer to Brücher's questions, so the plant's arrival via voyaging remains the superior explanation.

Considerable experimentation has gone into the question of whether the bottle gourd could have floated across the ocean, arrived in viable condition, and become self-established on shore (Towle 1952; Carter 1953; Whitaker and Carter 1954). Results of experiments on how long floating gourds remain viable have been inconclusive, perhaps because some experiments were carried out in tanks, where a gourd would not have been subjected to the boring organisms that are encountered in the open sea. The results are summarized by Camp (1954) to the effect that, while a gourd might have crossed from Africa and might have sprouted on shore in the Americas, there would still be no satisfactory explanation of how the offspring became established in the plant's normal habitat, which is inland.

Most recently, Whistler (1990, 1991) concluded that, contrary to the picture that resulted from early misidentifications, L. siceraria never grew in western Polynesia, although it was common in eastern Polynesia. (That the mid-Pacific had no gourds casts doubt on the drift explanation of this plant's dispersal by that means.) "The most likely hypothesis," Whistler concluded, "is that it was introduced [by voyagers] to eastern Polynesia from South America." The coconut "may have moved eastward on the same roundtrip Polynesian voyage as the bottle gourd."

Despite uncertainty as to which ocean the gourd crossed and exactly how, we can be sure the plant was used in both hemispheres long before Columbus. That man played the key role in its ocean-crossing history is the position that best fits the evidence.

Luffa cylindrica (syn. aegyptiaca)

Luffa cylindrica is an Old World plant known as the 'vegetable sponge.' It was grown in South and Southeast Asia (Brücher 1989, 267) where it played a significant role in medicine (Nayar and Singh 1998, 14—15). It has also been cultivated for a long time in the Americas, according to Heiser (1989; 1985); he allows that it might have traveled by ship across the Pacific. Kosakowsky et al. (2000, 199) excavated pottery on coastal Guatemala dating around 1200 BC that had been decorated by daubing the pot's surface with paint using as a tool the unique cut stem end of L. cylindrica.

Macroptilum lathyroides

The minor pulse, Macroptilum lathyroides, has the English vernacular name of 'phasey bean.' It has been considered part of a distinct group within the genus Phaseolus but has recently been made a separate genus. Like several species of Phaseolus, which we will discuss below, the phasey bean is of American origin (Smartt 1969, 452). Along with two other American beans, M. lathyroides has been excavated by Pokharia and Saraswat (1998—1999, 99) at several Neolithic and Chalcolithic sites in India of the 2nd millennium BC. Phasey bean specimens have also been recorded by Vishnu-Mittre, Sharma, and Chanchala (1986) from deposits of the Malwa and Jorwe cultures (1600—1000 BC) at Diamabad in Ahmednagar Dist., Maharashtra state, India.

Manihot sp.

Pullaiah (2002, II, 346—47) points out two Sanskrit names, darukandah and kalpakandah, for Manihot esculenta, i.e., manioc, or cassava. This is surprising since there is no other evidence yet for its transfer to India. To be confident that it was taken there by voyagers will require more data, although no other explanation for the Sanskrit names is apparent.

The first European botanists to discover it growing on Easter Island found that it was called by the same name, yuca, by which it was known widely in the Americas (Langdon 1988, 326—28; Mellén Blanco 1986, 13). For that reason we suppose that a voyage carried the plant there from the continent (Heyerdahl 1964, 126). It may also have been known in the Marquesas, for in 1939, young Heyerdahl and his bride, living on Hivaoa, were taught by relatively unacculturated natives how to grow bitter manioc and process its starch (Heyerdahl 1996).

Maranta arundinacea

The tuberous shrub, Maranta arundinacea, arrowroot, has been found to have grown in Central America up to 6,000 years ago (Piperno 1999, 126). It is a vegetatively reproduced plant, so it could not have crossed the Pacific by floating. Yet in India, Aiyer (1956, 44) has recorded the Sanskrit term, kuvai, for the species, and Pullaiah (2002, 348) gives tavakshiri and tugaksiri. Furthermore, it is mentioned in South India in the Tamil-language texts, Malaipadukadam and Mathuraikanji, that were written long preceding the arrival of Portuguese ships in India.

Mimosa pudica

Mimosa pudica, the 'sensitive plant,' was probably native to Brazil. It is called 'sensitive' because its leaves abruptly fold up when they are touched. It was known to the pre-Spanish Maya (Roys 1931, 267). Yet there were three Sanskrit names for it (Watson 1868, 347; Nadkarni 1914, 233). That means that it was known in India centuries before European voyagers could have brought it from the Americas. The same plant was present in the Marquesas, though perhaps nowhere else in the eastern Pacific; it very likely reached those islands by a separate voyage from the Americas.

Mirabilis jalapa

The widely-appreciated flower, Mirabilis jalapa, is called the 'four o'clock' from its habit of opening its blossoms only in the late afternoon. Zeven and de Wet (1982, 177) report that South America is the center of this genus' diversity and its apparent place of origin. Roys (1931, 291) documented a pre-Columbian Mayan-language name for it, and Hernandez (1942 [before 1580], I, 194—95) identified two varieties that were growing on the Mexican Mesa Central in the 16th century.

Once again, however, we find a plant of American origin in ancient India. Three or four Sanskrit names were in use for the four o'clock. Torkelson (1999, 1786) provides krishnakeli; Balfour (18711873, III, 282) and Watson (1868) give bahu-bumi and sundia-ragum; and Pullaiah (2002, 361) reports sandhya-rága, trisandh, and krsnakeli (sic). To explain such a proliferation of names, we must suppose that the four o'clock was present in India from an early time. It is not credible to suppose that its presence in Asia, only in India, was due to fortuitous oceanic drift from half way around the world.

Mollugo verticillata

The carpetweed, Mollugo verticillata, is shown by Chapman et al. (1974, 411—12) to have been found in archaeological digs in both North America (it was also distributed in Middle and South America) and Europe, yet the plant's origin is said to be China (MOBOT 2003). It is one of those weeds whose favorite habitat is the disturbed ground that results from husbandry.

Monstera deliciosa

The large climbing aroid, Monstera deliciosa, is native to Central America. It is represented on sculptures of medieval age in India. Gupta (1996, 108—9) describes and pictures temple scenes where not only do the distinctive leaves of this giant philodendron provide background for scenes of sacred art, but the distinctive fruit is also visible. In one such scene, the fruit is shown on a plate held by an aide to Vishnu. The fruit tastes like a mixture of pineapple and banana and was much esteemed in both hemispheres (Lundell 1937, 35; Burkill 1966, II, 151).

Morus alba

The white mulberry, Morus alba, has long been one of the plants essential to Chinese civilization. Its leaves have been used to feed silkworms for at least 3,000 years, and mulberry bark was used to manufacture paper (Bretschneider 1892, 128; Watt 1888—1893, V, 280). The genus originated in the Old World.

Soon after the Spaniards subdued the native people of the Antilles, they launched a plan to raise silkworms there in order to take advantage of the strong European market for silk. In central Mexico, too, the conquistadors launched a silk industry, utilizing worms and trees imported from the Mediterranean area (Borah 1943, 5—14). Nevertheless, Las Casas, who died in 1522, said, concerning the Antilles, where he was a planter before his ordination as a priest, that there were "as many mulberries as weeds" already growing in those islands before the intentional importation of the European trees (1875—1876, IV, 379—80). Presumably, they grew on the mainland also. So at least one species of Morus was in Middle America before the Spaniards arrived, although its modern botanical identification is uncertain. Nevertheless, Von Hagen (1944, 67) considered that all three of the Morus species most often recognized, M. alba, M. nigra, and M. rubra, "may . . . have been present in Mexico and used for their bark." Tozzer (1941, 195) translated Bishop Landa's account (ca. 1566) of life in Yucatan as reporting the presence of "two kinds of mulberry plant, very fresh and fine." The two kinds he referred to may have been M. alba (the most common species) and either M. nigra or M. rubra.

Making paper from bark had a long history in Mesoamerica. Von Hagen noted that trees whose bark was made into paper (in colonial days) by the Otomí Indians, a relatively unacculturated native people of central Mexico, included one "which has been identified [in the taxonomy of over 60 years ago] as Morus celtifolia, a paper mulberry similar to the plant used by many Asiatic papermakers" (1944, 50, 51, 58—59, 67).

Making bark cloth and paper of mulberry bark is of particular significance because of a pair of exemplary studies by Paul Tolstoy (1963, 1966). He demonstrated that the bark cloth/paper complex of Mesoamerica was parallel in great detail to bark-processing methods in island Southeast Asia. Meanwhile, MacNeish et al. (1967, 85) excavated a stone bark beater in the Tehuacán Valley of Mexico that they considered so similar to beaters of Java and the Celebes that they found it "extremely difficult" to account for the degree of similarity by "independent invention" (the only alternative explanation for the similarity, of course, being transmission of the artifact form and function across the Pacific). Mulberry tree starts probably accompanied the bark-beater when Asian voyagers carried knowledge of the complex across the Pacific by voyagers in "the early part of the 1st millennium BC" (Tolstoy 1963). (When ancestors of the Maori left central Polynesia to settle New Zealand, they considered bark cloth so essential they carried mulberry starts with them; see Von Hagen 1944, 23.) While the Morus species involved in the Mesoamerican industry still need to be identified definitively, it is evident that at least one species crossed to the Americas with Asian or Oceanian voyagers who not only brought the plant but also carried the complex technology of turning bark into cloth and paper. We are supposing here that the most likely species was the Chinese mainstay, M. alba.

Meanwhile, if, as Brücher (1989,132) suggests, Morus rubra had an American origin, then the additional question arises, how did it reach Asia, where it was not uncommon?

Mucuna pruriens

Mucuna pruriens (syn. M. prurita) is commonly called 'cowhage,' or 'cow-itch' in English (Oxford English Dictionary, 2nd ed., 1082). The pods are covered with barbed hairs that cause severe itching. The hairy part was sometimes ingested as a vermifuge, while the plant's roots served to make a tonic. The seeds have been considered aphrodisiac in India since Sanskritic times (Watt 1888—1893, V, 286; Watson 1868, 263). Balfour (1871—1873, III, 394—95) listed the Sanskrit name, atmagupta, for cowhage; Watson (1868, 11) added alkushee, and Nadkarni (1914, 242—43) gives kapikachchu. Banerji (1980, 26, v, vii) also has adhyanda. Such a varied lexicon indicates that the plant was long familiar in India. That is further confirmed by Banerji's note of its mention in the Satapatha Brahmana, a text dated in its earliest form before the rise of Buddhism.

However, the plant is of tropical American origin. For Mayan, Roys (1931, 235) gives the name, chiican, for it and says, "This is the English cow-itch."

Musa spp.

The terms 'banana' and 'plantain' have been a source of confusion in botanical discussions because the two corresponding 'species,' Musa sapientum and Musa paradisiaca, were never established as bona fide species. Cytogenetic analysis demonstrated in the1950s that most of the bananas or plantains commonly grown contain varying proportions of genomic contributions by hybridization from two base species, M. balbisiana (B-genome) and M. acuminata (A-genome) (Simmonds 1966; Nicolson et al. 1988, 297). In this paper, we speak of a single species, Musa balbisiana x acuminata, covering what the earlier literature termed both 'bananas' and 'plantains.' (Some other 'bananas,' however, did not originate from those two genomes.)

A notion commonly held by scientists and historians is that there were no bananas or plantains in the Americas before the Spaniards introduced them. However, when Sapper (1934) studied Native American names for Musa, he concluded that bananas/plantains had to have been present in South America no later than the 1st millennium AD to account for the diversity and distribution of names. Smole (1980) reached about the same conclusion from ethnographic and ecological study of plantains among peoples of Brazil and Venezuela. Moreover, a word has been reconstructed in the proto-Mayan language dating to the 2nd millennium BC that is glossed by Kaufman and Norman (1984; followed by England 1992, 25) as plátano, banana. Ethno-historical documents also attest to pre-Columbian cultivation of the crop(s) in Mexico and Guatemala (McBryde 1945, 36; Spores 1965, 971—72; Roys 1931, 218). Archaeological finds of probable banana leaves from Peruvian tombs (Harms 1922, 166) seem to confirm these data.

The invalidity of the claimed isolation of the New World in relation to the banana/plantain was clear enough to persuade E.D. Merrill (1954; see also Heyerdahl 1964, 123) to grant that "We may reasonably admit that one, or a few, of the numerous Polynesian plantain varieties may have . . . " reached the Amazon Basin before Columbus. Considering its source, this statement constitutes an admission of the strength of the evidence for the pre-Columbian presence of the genus Musa in the New World.

Myrica gale

Bog myrtle, Myrica gale, is found on both sides of the North Atlantic, although it is now absent from Iceland and Greenland. Its pollen is found in Iceland in excavation layers from saga times, and possibly also in Greenland (Thorarinsson 1942, 46). It is often supposed by biogeographers that various plants growing on both sides of the North Atlantic, including this species, date back to Cretaceous times, before continental drift widened the North Atlantic Ocean gap. Nevertheless, Thorarinsson supposes that the Norse brought this particular plant to North America because they sometimes used it instead of hops to brew their beer. Since no other scenario for the plant's North American distribution makes sense to him, Thorarinsson suggests medieval voyaging as the means. We agree.

Pachyrhizus spp.

Data are unavailable to allow us to distinguish consistently and certainly in the literature between Pachyrhizus erosus (called'yam,' but of the family Fabaceae, not Dioscoreaceae) and P. tuberosus 'yam bean'), both of which are called jícama in Hispanic America. Today's favored view is that P. tuberosus is native to the headwaters of the Amazon in northwestern South America (Brücher 1989, 44). It was commonly cultivated in ancient Peru, fossil remains having come from tombs at Paracas (before our era). The plant is also shown in Nazca-period art (Yacovleff and Herrera 1934—1935, 281—82). Several biotypes of P. erosus grow wild in Mesoamerica (Brücher 1989, 84—85). The Mayan name is chicam (from which the word jícama was derived). Patiño (1976, 33) has mapped the distribution of both species in South America at the time of the Spaniard's arrival from the earliest accounts.

One or the other species is cultivated in the Philippines and China to such an extent that it has been claimed to be of East Asiatic origin (MacNeish 1992, 260). Watt (1888—1893, VI, Pt. 1, 3) said the same thing for P. angulatus (syn. P. erosus), based in part on the fact that the plant bore a Sanskrit name, sankhálu. Both data suggest that transfer from the Americas took place rather anciently.

P. tuberosus is called doushu, or tugua, in modern China (Johannessen and Wang 1998, 26—27). A text dated AD 1736 describes this plant accurately as having a "root . . . quite big, greenish-white in color" which "tastes sweet and fragile or soft-crunchy to eat." The name of the plant was already in use in the Song Dynasty (before AD 1182) to denote a plant whose description was so similar that it cannot be anything but the currently recognized species. Archaeologist K.C. Chang (1970, 177) and other Chinese scholars accept that P. tuberosus was present in China in pre-Columbian times.

Phaseolus spp.

The common, or kidney, bean, Phaseolus vulgaris, and the lima bean, P. lunatus,) also had an American origin. Specimens of P. vulgaris are known at ca. 6000 BP in Mexico (Pickersgill and Heiser 1978, 810—11) and before 4000 BP in Peru (Yen 1963, 112). The orthodox view is that they reached the Old World only when Portuguese traders brought them to Asia around AD 1500.

All botanists have not agreed, however, that these beans were absent from the Eastern Hemisphere. In connection with their landmark research on cotton, Hutchinson, Silow, and Stephens (1947, 138) accepted a bi-hemispheric distribution of Phaseolus and considered the widespread use of beans in Asia to constitute satisfactory evidence for pre-Columbian contact. Hutchinson was reported in 1961 to be still "working on the genetics of American beans" (Bushnell 1961). His research was never published.

However, now the problem is resolved definitively. Phaseolus vulgaris, P. lunatus, and the phasey bean, Phaseolus lathyroides (reclassified as Macroptilium lathyroides, per Walker 1976, 598), have all been discovered in multiple archaeological sites in India of the 2nd millennium BC.

Pokharia and Saraswat (1998—1999, 99) report that Phaseolus ". . . beans of American origin have been encountered from proto-historic sites in peninsular India." P. vulgaris was recorded from the pre-Prabhas and Prabhas cultures at Prabhas Patan, Junagadh Dist., Gujarat, dated from 1800 BC to AD 600. They also came from Chalcolithic Inamgaon (about 1600 BC), Pune Dist., Maharashtra, and from Neolithic Tekkalkota, Bellary Dist., Karnataka, with a radiocarbon date of 1620 BC. P. vulgaris, P. lunatus, and the phasey bean have also been recorded by Vishnu-Mittre, Sharma, and Chanchala (1986) in deposits of the Malwa and Jorwe cultures (1600—1000 BC) at Diamabad in Ahmednagar Dist., Maharashtra. The phasey bean was also found at the Sanghol site (Pokharia and Saraswat 1998—1999, 99), dated in early AD times. Plant names agree. Levey (1973, 55; see also Levey 1966, 16) found that "The medieval Arabic term for kidney bean [i.e., P. vulgaris] is lubiya. It is lubbu in Akkadian and lu.úb in Sumerian . . .. In Sanskrit and Hindustani, however, it is simbi and sim respectively . . .."

It may be noted at this point that we have presented direct evidence for the presence in Asia long before Columbus of the four classic foods in the Mesoamerican diet: maize, cucurbits, chiles, and beans.

Physalis spp.

The ground cherry, or winter cherry, (or husk tomato) seems to refer, at least in the older literature, to more than one species of Physalis. The literature reporting presence of the plant is not consistent in its terminology (see the sources cited under Physalis lanceifolia in the appendix). P. lanceifolia, P. philadelphica (Hernández 1942—1946 [before 1580], I, 283; Brücher 1989, 276), P. pubescens, P. lanceolata (Index Kewensis), P. indica (Nadkarni 1914, 298), P. alkekengi (Bretschneider 1882, 32), P. angula (MOBOT 2003; Roys 1931), and perhaps P. minima (Gunther 1934, 468—71) sometimes may be conspecific. Although the origin of the genus was unquestionably in the Americas, these fruits (if this synonymy is correct) were known in India (Bretschneider 1882, 32; 1892, 43; the plant had a Sanskrit name, rajaputrika, {Chopra et al. 1956, 191; Torkelson 1999, 1808}), China (Bretschneider 1882, 1892), and Greece (Gunther 1934, 468—71). If the synonymies are not all correct, there remains a major question of how the taxonomy and distribution of the 'ground cherry' is to be explained.

Physalis peruviana

Physalis peruviana, the winter cherry, commonly called the 'Cape gooseberry,' or 'Brazil cherry,' seems also to have reached Asia as well as Polynesia before European influence. Balfour (1871—1873, IV, 562) cites P. peruviana as also growing in India, meanwhile noting its American origin (also on origin, see Brücher 1989, 275—77; and Zeven and de Wet 1982, 181). Furthermore, this species had a unique Sanskrit name (Chopra et al. 1956, 192).

Occurrence of P. peruviana in the Marquesas, Easter Island (Heyerdahl 1996; 1964, 126), and Hawaii (Hillebrand 1888, 310) most plausibly was due to a voyage from the Americas to the islands (Brown 1935, 257—58).

Polygonum acuminatum

Polygonum acuminatum is an aquatic species found on Easter Island, where it floats on the surface of the lakes inside two volcanic craters (Heyerdahl 1961, 26). P. acuminatum is used for medicinal purposes on the island, as it is in the Titicaca Basin, Bolivia. It is an American plant. The species occurs nowhere else in Polynesia.

Of the flora, Skottsberg (1920, I, 412) said, "The presence of a neo-tropical element (on Juan Fernández and Easter Island) is surprising." The "mode of occurrence and ecology oblige us to regard" P. acuminatum as "truly indigenous," or else "intentionally introduced in prehistoric time during one of the mythical [sic] cruises which, according to Heyerdahl, put Easter Island in contact with Peru. A direct transport of seeds across the ocean without man's assistance is difficult to imagine. . .." Skottsberg adds (page 425) that, contrarily, for the Marquesas "there is no neo-tropical element [that he detected] in spite of the prevailing direction of winds and currents." Dumont et al. (1998) update the matter by reporting analysis of a core from Rano Raraku crater lake on Easter Island. Five (stratigraphic) zones are identified. The last three of these are separated by waves of immigration. The researchers argue that a first, or South American, wave, dated to the 2nd half of the 14th century by radioactive dating, may represent a visit by South American Indians. They found the top 85 cm. of sediment to include Polygonum acuminatum. Because of the synchronous appearance of multiple floral taxa from the Americas, Dumont rules out passive introduction. "The island is so remote, and such a small target, that mechanisms of passive dispersal were ineffective for populating it" (with flora). Besides, there are no freshwater birds on the island. "We therefore propose that humans introduced these neo-tropical biota, in one single event."

Portulaca oleracea

Purslane, Portulaca oleracea, was common in Roman gardens in Pliny's day (Leach 1982, 2). It grew throughout the warmer parts of the Old World and was also mentioned in Egyptian texts. Nevertheless, Gray and Trumbell (1883, 253) demonstrated over a century ago that the species is actually of American origin. In North America, it was growing as early as 2,500—3,000 years ago (Chapman et al. 1974, 412). It thrives best in the disturbed soils of gardens, which indicates that the plant may well have accompanied a transoceanic transfer of horticulture.

A number of Sanskrit names for purslane are known: lonika, or loonia (Balfour 1871—1873, IV, 660); ghotika (Pullaiah 2002, II, 426); and mansala (Chopra et al. 1958, 521—23). Furthermore, Bretschneider (1882, 49—53, 57—61) notes the mention of the plant in a Chinese treatise by Zhou Ding wang, an imperial prince who died in AD 1425; he had seen the plant growing in Henan province (Bretschneider 1892, 428). Another work, Zhu Zi yulei, by the famous Song-period neo-Confucian scholar, Zhu Xi (AD 1130—1200), mentions purslane as machixian, a term that is still used for P. oleracea today (personal communication from V. Mair, 2002).

Psidium guajava

More than 100 species of the genus Psidium, which includes the guava, are native to Tropical America (Brown 1935, 200). Bailey (1966, III, 284) concurred that the genus originated wholly in the Americas. Towle (1961, 73) documented a fruit of P. guajava in a burial from Ancón, Peru (dated BC), and fruits have also been found in remains from the Gallinazo phase (early centuries AD) and the following Mochica phase.

Historically documented names provide evidence for the pre-Columbian presence of the guava in India. In Sanskrit, it was known as amruta-phalam (cf. Arabic and Persian: both amrúd) (Watt 1888—1893, VI, Pt. 1, 351—53). Pullaiah (2002, 433) has Sanskrit, péràlà, as well as mansala, for this species. Sharma and Dash (1983, 518) identify the guava as the 'Paravata fruit' mentioned in the Caraka Samhita text (between 900 BC and the 4th century AD) (Aiyer 1956, 36; Pullaiah 2002, 2). Of linguistic interest is the fact that the name of the guava in the Mysore area of India involves the root bidji (Watson 1868, 134), while in Yucatec Mayan the same fruit is called by the near-equivalent, pichi (Roys, 1931, 231, 276).

Sapindus saponaria

The soapberry tree, Sapindus saponaria, is another Native American species (Zeven and de Wet 1982, 178; Knoche 1925, 102—23) that has spread throughout the tropics. Its antiquity in the Americas is certain (Tozzer 1941, 197; Hernández 1943 [by 1580], II, 529—30). In the Casma Valley of northern Peru it has been found in middens dating to 1785 BC (Ugent et al. 1986).

An equivalent tree was common in India under the scientific name S. mukorossi in northern India and S. trifoliatus in the south. They are now equated with S. saponaria by Index Kewensis. Three Sanskrit names are known: phenila and arushta for S. trifoliatus (Nadkarni 1914, 350; Int. Lib. Assoc. 1996, 572), and urista for S. mukorossi (Int. Lib. Assoc. 1996, 572). The soapberry trees of India are also distributed in China (Watt 1889, VI, Part II, 468).

The same tree was in use on Easter Island at discovery, and it was widespread in some other (but not all) Polynesian islands as well (Heyerdahl 1963, 31; Langdon and Tryon 1983, 43; Brown 1935, 160—61).

Schoenoplectus californicus

A bulrush, or sedge, Schoenoplectus californicus, grows on Easter Island in close association with Polygonum acuminatum (see above) on the surface of the island's crater lakes. It too, is of American origin, found up and down the Pacific Coast of both American continents. The same species grows in Hawaii with a name nearly parallel phonetically to that on Easter Island (Rapanui, nga?atu; Hawaiian, nanaku) (Langdon 1988, 330, 334; Langdon and Tryon 1983, 43). The uses to which it was put were very nearly the same too in South America and Easter Island (Heyerdahl 1961, 23—5; Towle 1961, 26—7). As with P. acuminatum, Dumont et al. (1998, 410, 418) conclude that the evidence is strong that this plant was brought from the mainland in the 14th century by voyagers, while no other explanation stands up.

Sisyhynchium angustifolium

The small lily (commonly called a 'grass'), Sisyhynchium angustifolium, is fundamentally North American in distribution, although it has also been found at the site of the ruins of Norse settlements in Greenland (Faeggri 1964, 344—51; Polunin 1960, 181; Thorarinsson 1942, 45—6). Here is botanical evidence in support of the historical and archaeological facts on the Norse migration to the American Vinland.

Solanum spp.

Solanum candidum is one of a set of fruit-bearing trees that inhabit Middle and South America, Oceania, and Asia, and overlap taxonomically in an intriguing way. American S. candidum is so close to S. lasiocarpum of Western Oceania that they may be the same species, and the two areas share certain uses. Meanwhile, on the American side, S. candidum is virtually conspecific with S. quitoense. We consider that the similarities are such that two transoceanic transfers were accomplished, either S. candidum or S. lasiocarpum making one voyage and S. repandum or S. sessiliflorum another.

A similar situation is true of S. candidum and S. lasiocarpum of Southeast Asia. In South America, the former has so nearly the same characteristics as the Asian tree that their relationship demands an historical explanation. No plausible one has been offered by those who have studied the botany. We believe that the Pacific voyaging this paper documents can account for the derivation of one species from its near relative across the ocean as no other scenario does.

Solanum repandum, which is spread from Fiji to the Marquesas, is so similar to South American S. sessiliflorum that Whalen et al. (1981; cf. Whistler 1991, 41—66) suspect the two may be conspecific. The uses to which each is put are essentially the same. (See details in the appendix.). Moreover, the name of repandum in the Marquesas, kokoua or koko'u, is enough like a South American name for the species, cocona, that on the basis of the name alone one may be justified in seeing a direct transfer before Europeans came on the scene. We agree with Whalen and Whistler that voyages between the Americas and the islands make more sense than any other possible explanation for these relationships.

The species Solanum nigrum, known as 'black nightshade,' today is widely distributed throughout temperate and tropical regions of the world. In the Americas, the Maya knew S. nigrum as ich-can, or pahal-can (Roys 1931, 248, 272). It was also used medicinally in Peru (Yacovleff and Herrera 1934—1935, 281). After mentioning medicinal uses of the nightshade in ancient Assyria, India, and China, Thompson (1949, 143) observed, "it is obvious that the S. nigrum is a very popular drug in the East." Watt (1888—1893, VI, Pt. III, 263—64) reported S. nigrum growing throughout India and Ceylon. Pokharia and Saraswat (1998—1999, 90) found nightshade seeds in material excavated from the Sanghol site in India (1st to 3rd centuries AD). Its berries are described in Sanskrit works of medicine where the plant's name is kákamáchai (Nadkarni 1914, 373), or a variant thereof. Dioscorides' 1st century AD Greek herbal catalog identified the plant (Gunther 1934, 467). Maimonides (1974; Meyerhof and Sobhy 1932) described it in 12th-century Egypt. We see no way to account for the inter-hemispheric transfer except by voyagers over two millennia ago.

Solanum tuberosum, the potato, was found on Easter Island by early European visitors, as Jeffreys (1963a, 11—23) and Mellén Blanco (1986, 133) demonstrate using explorers' accounts. Of course, the potato is very old on the mainland, but it is absent elsewhere in Polynesia and points west.

Sonchus oleraceus

S. oleraceus in some sources is called 'chicory' although true chicory is Cichorium intybus. It was a potherb and source of medicine among the Maya (Tozzer 1941, 146) and in Peru (Yacovleff and Herrera 1933—1934, 299). Chroniclers' accounts place it so early in the Americas that there is no question of the Spaniards having brought it. Yet, it was an Old World native that was extensively cultivated in Europe and Asia (Balfour 1871—1873, V, 482—83; Bretschneider 1892, 179; Watson 1868, 259; Watt 1888—1893, II, 285). Those facts can only be explained by supposing that it was carried from Eurasia to the Americas before Columbus.

Sophora toromiro

Sophora toromiro was the only wild tree in the flora of Easter Island. Its wood served the islanders for canoes and to make other carved items. Knoche (1919) considered it to have been introduced from the outside and a 'cultivated' plant (syn. S. tetraptera). A native tradition told to the early Christian missionaries (Mellén Blanco 1986, 135; cf. Heyerdahl 1963, 26—7) said that this toromiro tree was among the plants brought by ancestral settler/voyager Hotu Matu'a. It is now extinct. According to Skottsberg (1920, 421) it, or a close relative, has been recorded by botanists only from Chile, Juan Fernandez Island (off Chile), Easter Island, and New Zealand. He also thought its natural transport to Easter Island was not a satisfactory explanation for its presence there.

Tagetes spp.

Tagetes erecta is the common marigold and T. patula the dwarf marigold. They both originated in Mexico (MacNeish 1992). Hernández (1943 [before 1580], II, 644-52) illustrated ten varieties of Tagetes in 16th-century Mexico, including one that Linnaeus later would dub T. erecta. The botanists who prepared the 1943 edition of Hernández suggested two of the other nine varieties that could represent T. patula. Roys (1931, 279) added that the dwarf species grew abundantly in the Maya area as a weed.

Levey and Al-Khaledy (1967, 192) identify one or more of the marigolds in texts from Persia and India dated to the 13th century. The blossom is known as the 'flower of the dead' in both Mexico and India today. Neher (1986) has published an eyewitness account by Harlan of a harvest festival in an Indian village whose ritual and symbolism revolves around the marigold and directly recalls Mexican customs. It is hard to believe that the transfer of postdated European influence on India and, indeed, two Sanskrit names for the marigold, zanduga (Chopra et al. 1956, 239) and sthulapushpa (Int. Lib. Assoc. 1996, 574), could account for the parallels. Rather, they appear to ensure that the genus has been on the subcontinent for many centuries. We suppose that since both Tagetes species are found in India today and had Sanskrit names (Pandey 2000, 271), one or both having become 'naturalized,' that probably both arrived together anciently.

The species listed in Tables 1, 2, and 3 that are not discussed to this point are treated in the appendix.

Microfauna

An earlier discussion under the heading, "The Problem," established that two species of 'hookworm,' Ancylostoma duodenale and Necator americanus, were brought to South America by voyagers in the 6th millennium BC or before. Nineteen other species of infectious organisms that also appear to have crossed with voyagers are shown in Table 4. The occurence of the diseasese which these organisms cause demonstrate that actual human beings, acting as their hosts, landed in the Americas as voyagers and passed on the imported organisms (plus genetic contributions) to their successors. We discuss those here. In addition, 18 further organisms that possibly were transported by ship are listed in Table 5, but they are discussed only in the appendix.

Ascaris lumbricoides

Ascaris lumbricoides is the large roundworm known to have infested Egyptian mummy Pum II; for instance, dating to 170 BC. It has been found also in other Old World locations in antiquity (Cockburn et al. 1998, 79—80). Pictorial evidence exists for ascariasis in ancient Mesopotamia, and prescriptions in ancient Egypt have A. lumbricoides as a target of treatment (Kuhnke 1993, 457). Moreover, this nematode was known to ancient pre-Columbian writers in China, India, and Europe. It was also present in pre-Columbian America (Patterson 1993, 603). Although it was once thought to be a post-Columbian arrival, it has been shown recently to have plagued pre-Columbian South American populations (Verano 1998, 221).

Table 4

Microfauna for which there is Decisive Evidence

Species

Common Name or Caused Disease

Ancylostoma duodenale

a hookworm

Ascaris lumbricoides

roundworm

Bordetella pertussis

whooping cough bacterium

Borrelia recurrentis

relapsing fever spirochete

Entamoeba hystolytica

amoeba that causes dysentery

Human (alpha) herpes virus 3

cause of shingles, chicken pox, etc.

Human (gamma) herpes virus 4

cause of mononucleosis, etc.

Microsporum spp.

cause of ringworm of the body

Mycobacterium tuberculosis

bacterium causing tuberculosis

Necator americanus

a hookworm

Pediculus humanus capitis

louse

Pediculus humanus corporis

louse

Piedreaia hortai.

a fungus that infests the hair

Rickettsia prowazekii

bacterium that causes typhus

Rickettsia rickettsii

bacterium that causes spotted fever

Streptococcus pyogenes

cause of scarlet and rheumatic fever, etc.

Strongyloides sp.

threadworm nematode

T cell lymphotropic (retro) virus (HTLV—I)

lymphotropic virus

Trichosporon ovoides

a fungus infesting hair of scalp or beard

Trichuris trichiura

whipworm

Yersinia pestis

the plague bacillus

Bordetella pertussis

Bordetella pertussis is the bacterium that causes whooping cough. It originated in the Old World, as did all the other species discussed under the "Microfauna" heading. However, antibodies for pertussis bacilli occur in the blood of isolated, unacculturated Brazilian Indians. Furthermore, it "may have been present in the Southwest [of the United States] before arrival of the Spaniards" (Stodder and Martin 1992, 62). Van Blerkom (1985, 46—7) concludes, "If not pertussis, then some close relative of it probably occurred in the New World as well as the Old . . .. Perhaps different strains existed in the two hemispheres." But what "close relative" would that be unless one that also arrived from the Eastern Hemisphere where alone it would have evolved? To Hare (1967, 119, 122), "It is highly improbable that any of these organisms would have become established in a scattered community with a Palaeolithic culture" and thus that they could have crossed to the Americas with early hunters via Beringia. The transfer of B. pertussis would of necessity have waited until an infected person from a more densely populated agricultural society crossed the ocean as a voyager.

Borrelia recurrentis

Relapsing fever is caused by the spirochete, Borrelia recurrentis. It is vector-borne, not transmitted from person to person. The infection is acquired by crushing an infected louse, Pediculus humanus corporis, so that its body fluids contaminate a bite wound or an abrasion of the skin (Chin 2000, 421—22).

Since B. recurrentis occurred in the Old World initially, the primary question becomes, how did the vector organism reach the New World? Van Blerkom (1985, 62—5) thought that louse-borne relapsing fever dates only to the rise of urban centers in the Old World (3000 BC?). Hare (1967, 118) says that cases closely resembling relapsing fever were described by Hippocrates ca. 400 BC. This dating rules out arrival of the disease by early Beringian migrants. Yet Alchon (1991, 22, 25) reports that relapsing fever, both the endemic type (transmitted by ticks) and the epidemic type (carried by lice), was present in pre-Hispanic coastal Ecuador. The only plausible scenario for this occurrence is the arrival in the Americas of lice (and ticks?) with human transoceanic voyagers. We will see below that this immigration is directly confirmed.

Table 5

Microfauna for which Evidence Justifies Further Study

Species

Common Name or Caused Disease

Flavivirus spp.

organism causing yellow fever

Giardia lamblia

protozoan causing giardiasis

Influenza viruses

influenza viruses

Leishmania sp.

protozoa causing Leishmaniasis

Mycobacterium leprae

bacterium causing leprosy

Onchocerca volvulus

nematode causing onchocerciasis

Plasmodium falciparum

sporozoan causing malaria

Rickettsia typhi

bacterium causing typhus murine

Salmonella enterica serovar Typhi

typhoid bacillus

Schistosoma sp.

cause of snail fever

Shigella dysenteriae

causes of bacillary dysentery

Staphylococcus x aureus

bacilli causing impetigo, carbuncles, etc.

Streptococcus pneumoniae

a cause of pneumonia

Treponema pallidum

organism causing yaws, pinta, and syphilis

Trichophyton concentricum

fungus causing ringworm of the body

Trychostrongylus sp.

helminthic parasite

Tunga penetrans

chigger, nigua

Wuchereria bancrofti

nematode causing filiariasis

Entamoeba hystolytica

Amoebic dysentery is caused by Entamoeba hystolytica. Transmission is mainly by ingestion of fecally-contaminated food or water (Chin 2000, 11—13). Van Blerkom (1985, 19) maps amoebic dysentery reaching Persia from India by 480 BC, thence to Africa. Again, the key issue is the presence of the particular amoeba in pre-Columbian America. Stodder and Martin (1992, 62) say that amoebic dysentery was probably present before European contact. Saunders et al. (1992, 118) approvingly cite Newman (1976) who considered that pre-European-contact diseases included both bacillary and amoebic dysentery.

Alchon's (1991, 20) detailed study using ethno-historic sources on diseases in Ecuador concluded that amebiasis was present before the Spaniards arrived. An unidentified (as to species) Entamoeba cyst was found in a Peruvian mummy dating to about AD 1500 (Pike 1967, 185). Van Blerkom considers that this date leaves the case in the range of possible Spanish contamination. Most Andeanists, however, suppose that Spanish influence was unlikely to have had any effect on this desiccated body (Pike 1967). Moreover, Van Blerkom observes (1985, 19), it is now known that the amoeba alone is incapable of inducing the disease state, which requires the concomitant presence of a certain bacterium (citing Schwabe 1967). She thinks it is likely that the amoeba, but not the bacterium, was present in the New World. From our non-epidemiological point of view, the crucial question is only, did any organism reach the New World from the Old, not was the disease manifest? The sources convince us that the amoeba was here. Again, we cannot imagine any means of that transfer except by an infected voyager.

Herpes zoster; varicella-zoster virus VZV

Human (alpha) herpes virus 3 is the cause of chicken pox, shingles, etc. Chicken pox (varicella) is an acute, generalized viral disease. Herpes zoster, or shingles, is a local manifestation of latent varicella infection. The infectious agent for both chicken pox and shingles is human (alpha) herpes virus 3 (Chin 2000, 91—93). Transmission from person to person is by direct contact or airborne (droplet) spread of fluids from an infected person.

Human (gamma) herpes virus 4

This is the Epstein-Barr virus, which is involved in infectious mononucleosis and other disease manifestations such as simplex (cold sores) and cytomegalovirus, a mononucleosis-like infection. Type 4 is closely related to other herpes viruses, similar morphologically but distinct serologically (Chin 2000, 350—51). The agent for both type 3 and 4 can remain latent within the human body for years after the initial attack (Alchon 1991, 23).

To Hare (1967, 121), "It is highly improbable that any of this class of organisms would have become established in a scattered community with a Palaeolithic culture." (In his day herpes viruses had not yet been identified in the Americas.) Actually, while nothing is known about the early history of chickenpox, there is no doubt about its comparative antiquity in the Old World. It was reported by the 1st century AD (Hare 1967, 120). Still, these viruses were endemic in the pre-Spanish Ecuadorean population according to Alchon (1991). They leave no evidence on skeletons; thus their existence cannot be checked archaeologically. But they have been found in isolated populations of Amazonian natives, implying pre-Columbian incidence. Stodder and Martin (1992) also believe that herpes was present in the Southwest of the United States before the Spaniards arrived.

Van Blerkom (1985, 27—9) thought that since varicella is endemic in Brazilian tribes and is also known in wild primates, that herpes zoster in humans in the Americas must have been derived from the ancient primate virus. But because the wild primate population in the Americas has limited interaction with humans, one would hope for real evidence that American monkeys have been shown to be infected and that native populations did, or could, receive it from that source. It is far more persuasive to us that one of the voyages from the Old World to the New, such as that (or those) that brought hookworms, probably tuberculosis, and also Old World plants to South America, was the medium by which humans introduced these organisms from the Eastern Hemisphere.

Microsporum spp.

Several fungi cause ringworm of the human body. The species differ depending on the area of the body infested (head, beard, groin, body, foot, or nails) (Chin 2000, 147—53). For the most common forms of the disease, tinea corporis (ringworm of the body) and tinea cruris (ringworm of the groin and perianal region), the reservoir for these agents is humans. Most species of Microsporum and Trichophyton (as well as Epidermophyton floccosum, Scytalidium dimidiatum, and S. hyalinum) cause 'dry type' tinea corporis in tropical areas of the Americas as well as the Old World; thus they are potential additional species to demonstrate transfer by voyagers.

Fonseca (1970, 40—5, 147 ff.), the discoverer of the disease in South American indigenes, called it tinha (or tinea) imbricada as well as, in the older literature, toquelau, or tokelau (in Oceania), and chimb∞r∞ (in Brazil). European explorers in Brazil reported its presence from the beginning. The disease is also known in central Mexico, Guatemala, and El Salvador, but it is totally absent among native peoples of North America, Alaska, and Canada. The area of incidence in the Old World is most of Polynesia, Micronesia, Melanesia, and Malaysia, as well as Formosa (the indigenous population) and Indochina, and, with less frequency, south China, Burma, Ceylon, and the south of India (but not Africa).

Fonseca (1970, 44—5, 195—96, 216—17) presented a conclusive ten-point argument supporting the proposition that this parasite was introduced to South America by ancient immigrants. The Brazilian tribes who were first discovered (by Fonseca, in 1924) to bear the disease were virtually isolated and untouched by European influence.

Mycobacterium tuberculosis

Tuberculosis is caused by M. tuberculosis. It is one of a complex of organisms that includes M. africanum, primarily from humans, and M. bovis, primarily from cattle (Chin 2000, 523—24). Transmission is by exposure to tubercle bacilli in airborne droplet nuclei produced by people with pulmonary or laryngeal TB when coughing or sneezing.

The origin and history of tuberculosis in the Old World is but dimly understood. Hare (1967, 117) points out that M. tuberculosis has never been isolated from wild animals, nor has it ever become established as a human parasite. It has infected (Old World) dairy herds since before the Christian era. Because Paleolithic societies did not domesticate cattle, it is improbable (or rather, impossible) that this organism caused infection at the time when migrants to the Americas presumably entered via Beringia. But the disease could have existed in Neolithic and more recent societies. The earliest sure evidence for pulmonary tuberculosis in the Old World is late in the 2nd millennium BC—in India, China, and Egypt. Bones with lesions suggestive of tuberculosis may go back as early as 3700 BC, although that date may somehow be deceptive. "It was never found in the thousands of mummies from Egypt and Nubia examined by Dawson, Smith et al. (Hare 1967, 125—26)." Klepinger (1987, 52—3) says that "Current paleo-pathological evidence would suggest that the mycobacteria responsible for the New World disease were not carried over by the Beringian trans-migrants but more likely arose de novo in the Western Hemisphere." However, to accept a second "de novo" origin of a particular species of bacterium requires more faith in parallel evolutionary processes than we can muster.

Allison et al. (1973) first established the presence of M. tuberculosis in a Peruvian mummy. Since then, additional cases in the Americas have been documented beyond any question (Allison et al. 1981; Karasch 1993, 537; Powell 1992; Salo et al 1994; Verano 1998, 217—19). In fact, Alchon (1991, 23—4) maintains that archaeological evidence indicates that acute respiratory infections were the most frequent cause of death among pre-Columbian Andean residents, anciently the same as today.

So how are we to explain tuberculosis in the Americas? Buikstra (1981, 13) says, "In the absence of appropriately-timed migrations from the Old World, we must develop and defend a reasonable model for the origin of this disease in the absence of [New World reservoirs of] domestic herd animals such as cattle." Yet no model has been offered that does not contradict vital facts. So Lovell's (1987, 53) interrogative stands: "But where did that organism come from?" There is no satisfactory answer, unless one accepts, as we have shown that one now must, that voyagers from the Old World intruded into pre-Columbian America and quite probably carried M. tuberculosis bacilli in their lungs.

Pediculus humanus spp.

Two species of lice, P. humanus capitis and P. humanus corporis, were shared by Native American peoples and those of Oceania. That these organisms arrived in the Americas by sea is the only sensible explanation. Lice are host specific; those of lower animals do not infest humans. Transmission is by direct contact with infested persons or objects used by them. Lice can survive for only a week without a food source (Chin 2000, 372—73).

Fonseca (1970) particularly has discussed the distribution of "Pediculus pseudohumanus," as P. humanis was once known. It is "a form of louse found solely in indigenes of the Americas and of Oceania and in American macaques." Fonseca quotes Ferris (1935), who described its distribution as "extremely peculiar" (see also Ferris 1951, 275)—specimens from the Marquesas and Tahiti in Polynesia were identical down to morphological details to lice from Guatemalan and Panamanian villagers and from mummified heads from Ecuador and Yucatan. More modern writers assure us of similarities on an even wider scale. Sandison (1967, 178—83) says lice of the same species are known from pre-Columbian Mexico and Peru, as well as from the Mediterranean through China. Karasch (1993, 538) notes that lice have been identified on mummies from Chile and Peru. According to chroniclers, the poor in the Inca Empire (as a control measure) had to "pay tribute in the form of small containers of lice." "Not surprisingly," Karasch continues, typhus was "a very common disease in ancient Peru." Alchon (1991, 22) believes "One can build a strong case for the existence of both endemic (flea-borne) and epidemic (louse-borne) typhus in the New World, based on lice on Peruvian and Chilean mummies." (Compare Zinsser 1960, 254—61, who also believed that historical evidence involving lice found on mummies suggests that typhus was present in South America.) Meanwhile, Van Blerkom (1985, 4) summarizes: "The lice found on pre-Columbian American mummies are of the same species (with only slight differences, on the order of a subspecies) as those on Old World humans (El-Najjar and Mulinski 1980, 111)."

Linguistic evidence anchors the case for the transfer of lice by direct contact, from Oceania at least. Roys (1931, 341) cites the authoritative 16th-century Mayan Motul dictionary for the term "Uk. The louse found on man and quadrupeds." Schumacher et al. (1992, 18) reports from Oceania that the ethnically-Papuan Austronesian-speaking Buma tribe, on Vanikoro, eastern Solomon Islands, have uka for louse, while in the western Solomons, the Austronesian Ontong Java people call the louse uku.

Piedreaia hortai

This fungus causes piedra (negra), a disease of the hair. "Piedra is characterized by black, hard 'gritty' nodules on hair shafts" (Chin 2000, 147—48). According to Fonseca (1970, 262), the disease is especially characteristic of South American natives in the interior of the continent. It is found very rarely in North America. On page 264, Fonseca observes that this disease is also found in Southeast Asia—Thailand, Vietnam, Burma, Malaya, and Indonesia. In all those regions it presents exactly the same clinical, epidemiological, and parasitological characteristics with which it appears on the South American continent. There are a variety of names for the disease in the Guaraní and Tupí language families in lowland South America. This disease was introduced to the Americas by pre-Columbian migrations of natives from Oceania, he argues, using a variety of evidence. It is missing in northern Asia and North America, so any migration across the Bering Strait could not have brought it. Nor does the disease exist in Europe or Africa. Because it was widely distributed in South America, among many language groups, it must have arrived long ago, Fonseca continues.

Rickettsia prowazekii

The body louse can be infected with R. prowazekii by feeding on the blood of a host with acute typhus fever. People in turn are infected by rubbing louse feces or crushed lice into the bite or into superficial abrasions. The body louse is involved not only in outbreaks of epidemic typhus but also epidemic relapsing fever caused by Borrelia recurrentis (Chin 2000, 372, 541—42).

As noted above, Alchon (1991, 22) has argued that both endemic (flea-borne) and epidemic (louse-borne) typhus probably were present in the New World, based on the fact that Peruvian mummies were infested with lice. Head and body lice were common on mummified remains. Most native households (at least in the Andes) included several guinea pigs in the family's living quarters; these animals can be reservoirs for the typhus rickettsiae. Infected fleas can easily jump from rodent to human, thereby transmitting the endemic form of the disease. There are pre-Conquest traditions of epidemics occurring during periods of social turmoil—wars, famines, and natural disasters—supporting the assertion that typhus existed in the Americas before the 16th century. For example, Peruvian chronicler, Guaman Poma, described two epidemics that took place long before the Spanish Conquest. Cabieses (1979) believes that typhus was common in pre-Columbian Peru. Guerra (1966, 330—32) maintained that the two most important aboriginal Aztec disease entities were matlazahuatl and cocolitztli. They caused major epidemics. Translation of the terms remains unclear; Guerra's analysis of the symptoms indicates that the former was exanthematic typhus. Ackerknecht (1965, 53) interpreted the reported (by tradition) epidemic of 1454 on the plateau of Mexico as in all probability typhus. Goldstein (1969) agreed with Ackerknecht, Bruce-Chwatt, and Sandison, who say that typhus probably occurred in the Americas before Columbus. Villacorta C. (1976) too is confident that exanthematic typhus was present anciently.

Where did the infectious organism come from? It may have been earlier in the Old World than in the New, but that is uncertain. Hare (1967, 118) believes the first outbreaks known in European medical history occurred in Italy only in 1505. Epidemic typhus is therefore a comparatively modern disease and, on chronological grounds alone, could not have come across Bering Strait. However, typhus was much older in Asia and Oceania than in Europe, judging from the wide distribution. Nicolle (1932) drew attention to the Asian/Oceanic distribution of typhus that supports the idea that pre-Columbian migrations reached the Americas from Oceania, bringing typhus (probably along with fleas and lice). Fonseca cites a large literature (Mooser, Gay, Miranda, Gaitán, Nicolle) representing what he (1970) refers to as "most authors" who have written on this point, to conclude that exanthematic typhus indeed existed in pre-Columbian America.

In the nature of the evidence of ancient epidemiology, we cannot be absolutely certain of the presence of many diseases, including typhus, because they leave no physical indication visible on skeletal remains to prove that the corpse had been infected. Nevertheless, we agree with Alchon that the evidence is strong for the presence of this rickettsial organism in the ancient Americas. No alternative explanation for the presence of R. prowazeki, other than voyaging by humans, is apparent. From the plant evidence already cited, it is clear that numerous transoceanic voyages to the Americas took place from the Asian home where so many diseases were endemic.

Rickettsia rickettsii

Rickettsia rickettsii is the infectious agent that causes spotted fever. According to Chin (2000, 372), the disease is one of a group of clinically similar diseases caused by closely related rickettsiae. They are transmitted by ixodid (hard) ticks, the tick species differing by geographic area. Newman (1985; 1976, 669) considered that pre-European-contact diseases in the Americas included rickettsial fevers. Saunders et al. (1992, 118) acquiesce. Ackerknecht (1965) considered that petechial typhus (spotted fever) was at least as old as AD 1083 in Mexico, as he reads the Aztec traditions.

This evidence is not quite as firm as for R. prowazeki (see above), but any alternative explanation for typhus having reached the Americas is impossible. It seems most plausible to us that this rickettsial disease agent was brought across the Pacific Ocean at the same time as R. rickettsii.

Rickettsia typhi (ex. Rickettsia mooseri)

This organism produces typhus murine, i.e., endemic typhus. Murine typhus is similar to louse-borne typhus but milder. It is found where humans and rats live together. Rats, mice, and other small mammals form the reservoir. Infective rat fleas (usually Xenopsylla cheopis) defecate rickettsiae while sucking blood, which contaminates the bite site and other skin wounds. Once infected, fleas remain so for up to their one year of life (Chin 2000, 544—45).

Fonseca (1970, 333—36) said that murine typhus was at first assigned primarily to two distinct geographic areas: 1) certain regions of North America (Mexico, Guatemala, and the southern United States), and 2) the Far East and Pacific—India, Malaysia, China, Manchuria, Formosa, Australia, New Guinea, New Zealand, and Hawaii. Nicolle (1932) recognized the presence of murine typhus in Mexico and Guatemala, the same as in Southeast Asia. He considered the possibility that typhus murine might have come via the Vikings, yet thought it far more logical that it reached the Americas via rats on Polynesian vessels. Alchon (1991, 22) too accepted pre-Columbian presence of the disease in South America, i.e., Ecuador, again based on ethnohistoric records.

Among diseases Newman (1976, 669) thought were "part of man's primate ancestry" and that either crossed the Bering Strait cold-screen or "were acquired" in the Americas were "various rickettsial fevers," including typhus. He thought so because the Aztecs had a name, matlazahuatl, for the disease, and depicted it in conventionalized pictures of suffering Indians. But we cannot see how the typhus vector could have passed the Arctic cold-screen, and neither is it clear from where the supposed settlers of the Americas via the Arctic could have 'acquired' the disease.

Streptococcus pyogenes

Streptococcus pyogenes causes scarlet fever, strep sore throat, and rheumatic fever. Pre-Columbian American skulls have been found showing evidence of acute infection of the mastoid from S. pyogenes. Hare (1967) accepts that streptococcus was present in the New World. To explain the fact is not easy. If the disease came via Bering Strait, as Van Blerkom supposed (1985, 77—8), then the concept of a cold-screen at Beringia becomes of doubtful value. Newman's aside that streptococcus infection was "acquired in the Americas" (1976, 669) is without meaning, absent an explanation of how the organism reached the Americas in order to be "acquired." Transoceanic voyaging as an explanation manages both the mastoid infection in pre-Columbian Amerindian skulls and our extensive evidence for transoceanic journeys (above). In fact, it might seem surprising if the infection had not reached the Western Hemisphere by that means.

Strongyloides sp.

The antiquity and origin of the hair worm, or threadworm, nematode, Strongyloides, as an Old World genus is shown by the fact that an Egyptian mummy had larval forms of the worm in its intestines (Sandison and Tapp 1998, 40). Patterson (1993b, 1016) says it occurred around the world with a range similar to that of the hookworms, although it was once thought to be absent from the New World. However "the presence of Strongyloides is now confirmed from (Peruvian) mummy study" (Verano 1998, 221). Moreover, from Antelope House, New Mexico, traces of this parasite were recovered from a coprolite (Reinhard 1988, 359). Again, the voyaging mechanism is indicated as the only plausible explanation.

T cell lymphotropic (retro)virus (HTLV-I)

Only a small number of peoples in Tropical America show infection with this virus. The group studied in the greatest detail is the Noanama Amerindians in the high mountains of southwestern Colombia; their geographical and social isolation reduces the chance of any contact with the slaves of African origin brought to Colombia by the Spaniards (León et al. 1996). This study combined sero-epidemiologic, genetic, virologic, molecular, anthropological, archaeological, and oceanographic data that led the authors to conclude that this virus could have arrived from Kyushu Island in Japan more than 5,000 years ago through direct voyaging.

An earlier study of 13 genetic markers around the world revealed that the Noanama had very close relations with Samoans on the one hand, and Japanese—especially Ainu—on the other. Furthermore, recent genetic studies on native South Americans showed that their ancestors possessed genetic markers related to the histo-compatible leucocyte antigen (HLA), as do the Japanese of Kyushu. A direct voyaging contact from Japan to Colombia would explain this relationship, because populations of North and Central America are totally without the HLA markers. At a mitochondrial DNA level, study of the deletion 9 bp in the human genome has shown it to be Asiatic in origin; however, it is being found in North American Amerindians and Polynesians (citing Torroni). Yet it is not present in the (Jomon-derived) Ainu people, and the 9 bp deletion is also absent among the Noanama (as well as on the coasts of Chile and Peru a thousand years earlier). This suggests an intrusion of people from Japan. León et al. (1995) cite the proposal of Meggers et al. (1965) concerning the intrusive Valdivia culture of Ecuador as confirming their position about the disease.

Finally, León et al point out that Japanese investigators have voyaged across the Pacific to Colombia by the North Pacific route, which the authors suppose was used anciently; the Japanese researchers used vessels similar to those of prehistoric times. This nautical experiment demonstrates that it was possible to make such a voyage, which is seen as bearing this disease (Errazurriz and Alvarado 1993).

Trichophyton spp.

In the genus Trichophyton, there are eleven species of fungi that are cosmopolitan, seven of which are anthropophilic (Ajello 1960, 30). The Trichophyton concentricum fungus is a cause of ringworm of the body. T. concentricum (the cause of tinea imbricata?, as Fonseca called the disease) is endemic in Southeast Asia and also widespread among the inhabitants of Polynesia and countries bordering the western shores of the Pacific. However, it only occurs sporadically among Indians living in the tropical forests of Brazil, Guatemala, and Mexico. The disparity in prevalence between the Asian endemic areas and those of Latin America has led to the interpretation that the fungus was introduced from Asia into the New World (Ajello 1960, 30). Since the fungus can only have been transferred on a human body, no other explanation than the arrival of voyagers can be accepted for its American incidence.

Trichosporon ovoides

T. ovoides is present in Brazil and Asia. It is a fungus that causes a disease (known in Brazil as piedra branca) consisting of white or clear nodules that develop on individual hairs of the beard or scalp. Fonseca (1970) describes it, gives its distribution, and argues that it had to have arrived by sea-borne travelers across the Pacific.

Trichuris trichiura

The whipworm, Trichuris trichiura, like the hookworm, requires warm, moist climatic conditions for the completion of its life cycle and reproduction. It is particularly incident in Asia and Oceania. Finding the human-specific parasites in the Americas is circumstantial evidence for transpacific contact (Reinhard 1992, 231—45) because the cold of the Bering Strait route would kill the infectious organism in the stage when it is excreted from the human host. Coprolites from coastal Peru show the whipworm present by 2700 BC (Verano 1991, 15—24). According to Ferreira et al. (1988, 65—7), the whipworm has been identified along with the two hookworm species (see above) in human coprolites from Boqueirão do Sitio da Pedra Furada, Brazil, in a stratum dated to 7320±80 BP.

Yersinia pestis

Some epidemiologists might assert that Yersinia pestis, the plague bacillus, is simply zoonotic in origin and as such has no relevance to the history of humans in the Americas. But that dodges the question of how the infectious bacillus came to be in the Americas at all. Either an infected animal or an infected human must have come carrying it from its original home in the Old World. Chin (2000, 381—83) says it is endemic in East and South Asia and sub-Saharan Africa. The reservoir is wild rodents, especially ground squirrels, and also rabbits and hares. Transmission occurs as a result of human intrusion into the zoonotic (sylvan or rural) cycle, or by the entry of sylvatic rodents or their infected fleas into human habitats. Domestic pets may carry plague-infected wild rodent fleas into homes. The most frequent source of exposure that results in human disease has been the bite of infected fleas (especially Xenopsylla cheopis, the oriental rat-flea). Although human plague is commonly zoonotic in origin, it can be transmitted from man to man, with or without the agency of vector fleas, and humans can also act as a reservoir of the disease (Van Blerkom 1985, 48, 50—9). Person to person transmission by Pulex irritans fleas, the 'human' flea, is presumed important in the Andes area (Chin 2000, 381—82).

Many (Schwabe {1969, 282} calls it a consensus) believe that sylvatic plague is indigenous in the Americas. Van Blerkom considers the most compelling evidence in favor of pre-Columbian plague to be the existence of several sylvatic foci in both North and South America, with the largest being in western North America in rodents. Also, it is focused in eastern Siberia and western Canada. This distribution suggests that plague is an ancient and widely distributed disease of rodents diffused across the Bering Strait. But Van Blerkom disagrees. Besides, Hare (1967) reports that there is no evidence that the disease occurred in the Eastern Hemisphere at any time during the pre-Christian era. (Ergo, it could not have reached the Western Hemisphere via early Holocene settlers traversing the Bering Strait.)

How it was transmitted to the Americas is clarified by Van Blerkom's (1985, 58) observation about the distribution of this organism. There are three subspecies: Y. p. orientalis, endemic in India, Burma, and South China; Y. p. antiqua, carried by rodents in Central Asia and Africa; and Y. p. mediaevalis, or the Black Death in Europe, which is today found only in West Africa. If New World sylvatic plague was an indigenous disease of the rodents of this hemisphere, one would expect it to be the same strain found on the other side of the Bering Strait, Y. p. antiqua, the parent of the other strains. However, American plague is derived from the urban strain found in Southeast Asian seaports, Y. p. orientalis (Alland 1970, 101—2; Hull 1963, 534). This suggests that plague was carried by ship to the Americas from Southeast Asia. In fact, any other explanation seems impossible.

Van Blerkom (1985, 58) believes that transmission to the Americas occurred only during the last pandemic. That was in China in 1855. She supposes that only later was plague found in wild American rodents, and it seems to have spread rapidly into wild reservoirs from the original murine foci in seaports (Hull 1963, 547—54). But her scenario presumes an unbelievable rate of spread to a wide variety of rodents (up to 200 species in the New World; so, Van Blerkom 1985, 56) over a wide geographical range. Rather, we suppose that the extensive presence of Y. p. orientalis in the Americas can be explained much more economically by supposing its arrival on a pre-Columbian voyage from some seaport in Southeast Asia with then sufficient time to spread to many rodents. This seems to be the only scenario that takes account of all the facts of the case.

Table 6

Other Fauna for which there is Decisive Evidence

Species

Common Name

Alphitobius diaperinus

lesser mealworm

Gallus gallus

chicken

Littorina littorea

a mollusk

Meleagris gallopavo

turkey

Mya arenaria

American soft-shell clam

Stegobium paniceum

drugstore beetle

Other Fauna

Beyond the flora and the microfauna already discussed, evidence of communication across the oceans also comes from the distribution of various faunal species that are not agents of disease. As a simple term of reference for this miscellany we use "Other Fauna." The 6 species in evidence are listed in Table 6. Six possibilities to be investigated further are listed in Table 7.

Table 7

Other Fauna needing Additional Study

Species

Common Name

Cairina moschata

Muscovy duck

Crax globicera

curassow

Cicada sp.

cicada

Dendrocygna bicolor

fulvous tree duck

Lasioderma serricorne

tobacco, or cigarette beetle

Rhyzopertha dominica

lesser grain borer

Alphitobius diaperinus

This pest of Old World origin has been discovered in a very similar mortuary context in both hemispheres. It is called in the vernacular, the 'lesser meal worm.' Panagiotakopulu (2001, 16) reports it in the British Isles in a 2nd-century AD burial and in Egypt with a mummy at 1350 BC. In Peru, the same worm has been found in an AD 1240 mummy bundle (Riddle and Vreeland 1982).

This dual association of pests with burial practices in both hemispheres appears less startling when we realize that chemicals from the consumption of American tobacco and coca plants were discovered in Egyptian corpses dating from the 2nd millennium BC to the 5th century AD (see above). Further, residues of Cannabis sativa, an Old World plant, has been found in Peruvian mummified corpses. It is not surprising then, that a beetle and a worm from Eurasia should show up in Peru. Since drugs associated with corpses have been found in both hemispheres, logically, contaminant pests could just as well have been shared also. Culture-bearing humans traveling on boats provide the only plausible means by which the two areas, halfway around the world from each other, could have been so linked.

Canis familiaris

The dog is commonly assumed to have been brought to the New World by early hunter-gatherers via Beringia, but there is very slim evidence for this. Mair's (1998) data from Asia suggest that dog-in-the-company-of-humans in the Old World is not very old. The earliest domestication (or taming) occurred in the Near East (during the Natufian era) only around 12,000 years ago. Dogs in the European Mesolithic period date to the order of 9000—6000 BP. The earliest dogs in China are around 6000 BP. Moreover, the common hypothetical root word for dog in ancestral language groupings like Nostratic and Afro-Asiatic appears to date "closer to 6000 BCE than to 10,000 BCE" (Mair 1998, 22—3). Turner (2002, 144) says a dog skull dated to 12,000 BC was found in a cave in the Altai Mountains, but it was not associated with evidence for human presence in the area. These data mean that it is a stretch to imagine domesticated dogs being available in northeastern Asia to accompany the first migrants who came to the New World via Bering Strait. So where did the American dogs come from? (Of course, they might have been independently domesticated from wild canids, although evidence—even hints—of that are all but absent (cf. Turner 2002 144—45).

C. Sauer (1969, 29) says: "The great hunters of the upper Paleolithic had no dogs. It has been noted that these appear archaeologically first with Mesolithic folk." "The Swiss zoologist Studer began a series of studies in the comparative anatomy of the dog and its relatives . . .. These make a strong case for the monophyletic origin of the dog. The conclusion is that the dog, in several ways less specialized than the wolves, cannot be derived from the latter, as on grounds of comparative anatomy man cannot be derived from the apes." (30.) "The dog is considered therefore as originating from a wild dog, native to Southeastern Asia, living in forested monsoon lands . . .."

As many as three varieties, or breeds, of canines might have reached the New World from the Old as a result of ocean travel.

The voiceless, hairless dog. Covarrubias (1957, 93) reported that early in the 1st millennium BC, or before, an edible dog occurred in both China and at Tlatilco, Mexico, of identical appearance. Coe (1968, 59) found physical evidence for consumption of small dogs at the site of San Lorenzo in southern Mexico around 1000 BC. Fiennes and Fiennes (1968, 26, 53—55, 103—110) told of a special breed of hairless, or 'toy,' dogs that were kept and bred in China and also in west Mexico and Peru as temple and sacrificial animals as well as for consumption. Campbell (1989, 360—367, 385) noted the presence of dogs for eating in the Chorrera phase in coastal Ecuador about 1500—500 BC; they appeared alongside such Asiatic traits as house effigies, roller and flat stamps, and ceramic pillows. Tolstoy (1974) considered the hairless dog of Mexico to have been derived directly from Asia (along with chickens and several plants).

A Viking dog? Friant and Reichlen (1950,1—18) concluded that "the Inca dog was not domesticated from a South American wild form but was brought from elsewhere already domesticated." Subsequently, Friant (1964—1965, 130—35) examined mummified dog remains, including skulls in Inca burials, and found that they compared closely with dog remains in Denmark from the late Neolithic. They postulated that the similarity must be due to the hybridization of Viking dogs with those of the 'Indians' (Incidentally, Adelsteinen and Blumenberg {1938} suggest on genetic grounds the possibility that certain cat populations of the northeastern United States originated from a Viking/Norse introduction.) Further research may or may not confirm their speculation.

Dogs kept for wool. Lord (1866, II, 215—17) discovered on the American Northwest Coast a few tribes that kept peculiar dogs "differing in every specific detail from all the other breeds of dogs belonging to either coast or inland Indians" on secluded islands, where they could not interbreed with regular canines. These special dogs had long white hair that was shorn annually. The hair was then woven into rugs, sometimes mixed with wool of the mountain goat, or duck feathers, or finely carded wild hemp. The practice dated from before European contact. Lord was sure these dogs were not indigenous but had been brought from elsewhere. He thought they were likely from Japan, where a small, longhaired dog had been reported. It is not clear to us how more evidence might be obtained to establish this notion.

At least, we believe the most plausible explanation for the 'toy' dog of the Americas is transfer by voyage from Asia. It is highly unlikely that such a breed would have been developed separately in two areas. A mechanism for how transfer could have taken place is suggested by Xu's (2002) report of a Chinese inscription on an artifact from La Venta (the Olmec site in southern Veracruz). This inscription was first noted by Chinese scholars while inspecting artifacts in a museum in Mexico. Xu had previously discovered Chinese characters, also like those of the Shang era in China, on other ceremonial objects from La Venta (Xu 1996; 2002). Chinese experts on Shang-period writing have confirmed a number of Xu's readings (personal communication to Sorenson from M. Xu, 2002).

Gallus gallus

Conventional wisdom among zoologists holds that chickens were absent in the New World until introduced by the Spaniards. If that had been the case, the chickens in the hands of Amerindians after the Conquest ought to have been strictly of the Mediterranean type, as Carter (1971; 1998) points out, but they were not. Many of them looked like Chinese or Malay chickens, very different in appearance, color of eggs, and behavior from the Mediterranean class of chicken. Evidence from physical characteristics of fowls, documentary history and ethnography, the uses to which the fowls were put, and the distribution of vernacular names combine to establish that the reputed introduction of the chicken by the Spaniards is contrary to the facts. Instead, multiple introductions of fowls across the oceans, in addition to the Spanish importation, are indicated.

Latcham (1922, 175) observed that in Chile, Bolivia, and Peru, at least three indigenous domesticated varieties or species were known. The Spanish terms for 'cock,' or 'chicken,' had not been adopted for them by the Amerindians "because they have their own names." Their three kinds were definitely present before the Spanish Conquest and are still represented among the fowls kept by the Araucanian Indians of Chile. Some lay blue and olive-green eggs, are tailless, and have tufts of feathers in the form of a ball at the sides of their heads, as do fowls in China that also lay blue eggs. Finsterbusch (1931) agreed that these were pre-Columbian chickens. Castello (1924) went so far as to identify four types of Chilean chickens which differed from the common European fowl but showed Asiatic features. For instance, blue-egg layers are distributed from Chile to Ecuador and Mexico, he reported. Hartman (1973) surveyed the literature to his point in time, paying particular attention to cultural meanings of fowl and their uses in Asia and throughout the Americas. She concluded that Asiatic voyagers probably introduced black-boned, black-meated (BB/BM) chickens to the New World before Columbus.

C.O. Sauer (1952) summarized some of the evidence for the aboriginal presence in South America of a BB/BM chicken. Its breast meat is dark, a melanotic sheath surrounds the bones, and it bears tufts on the side of its head. It also has raised hackles, a black tongue and legs, and characteristic coloring of the feathers that mark it as distinctively Southeast Asian.

Subsequently, Johannessen's fieldwork found that the BB/BM type was still being kept in several locations (Johannessen 1981; Johannessen and Fogg 1982; Johannessen, Fogg, and Fogg 1984). The recent distribution of the melanotic fowl is from Mexico southward to Brazil, Peru, and Chile. This is a non-flocking chicken, displaying the social psychology characteristic of the chickens of Southeast Asia that leads them to stay apart from others when feeding.

At least among the groups that speak Mayan languages, the BB/BM chicken is not normally eaten but is used ritualistically in divination or medicinal treatments in essentially the same manner as the Chinese recorded in AD 1530 in their first encyclopedia of medicine. These treatments are esoteric. For instance, in highland Guatemala the chicken is cut sagitally, bound against the soles of the feet of an ill person, and left there for two or more hours during which time it is said to absorb the pulmonary problems resulting from an asthma attack. It is also used to cure 'women's problems.' The curer intones specific incantations that are in a non-Mayan jargon recited while candles and copal incense are burned and rum is blown onto the patient's bare skin (providing a shock effect). Other rituals involving the BB/BM chicken take the fowl's blood and life in order to protect a house, family members, tools, and even ships against hexes. These beliefs and practices also correspond to Chinese ways with BB/BM fowls in Asia.

It seems likely that successful transfer of these cultural patterns must have involved explicit teaching over a considerable period by Asian carriers of the original bird stock. Such rites and beliefs could hardly have been imported via ephemeral contact, say with people off the 17th-century Manila galleons. And it is only speakers of Maya who received, or at least have carried on, the practices involving the BB/BM chickens. No Chinese (or any other non-Catholics) were legally allowed to settle in early colonial Mexico or Central America, so these esoteric practices mush have originated earlier.

What appears to be a very ancient presence of chickens in the Americas has recently been pointed out by historical linguistic study (Wichmann 1995, 76). Wichmann has established that the reconstructed proto-Mixe-Zoquean language of southern Mexico, which dates to the 2nd millennium BC in the area occupied at that time by the Olmec civilization, contained the term *ce:we(kv?)(n) 'chicken, hen.' The same term continued in the succeeding proto-Zoquean language in the following millennium, where it meant 'hen.' The linguist also reconstructed the expression *ná'w-ce:wy for 'cock.' (The word for 'turkey' is completely different from either of the above.) Guillemaud's (1947, 112) list of Mixe terms is generally confirmatory, giving tseuk for the Spanish gallina, 'hen,' and tsag-naj for the Spanish gallo, 'cock.'

Littorina littorea

The mollusk, Littorina littorea, is native to Northern European waters. Spjeldnaes and Henningsmoen (1938) suggest that it was introduced to North America by Norse settlers about AD 1000. There is no other way to account for its presence on both sides of the Atlantic. They noted that it is a hardy species that could survive for a long time in water in the bottom of an open boat.

Meleagris gallopavo

It is now clear that that quintessentially American fowl, the turkey, was being kept in Europe in the late Medieval Period. In 1940, an observer claimed that an American turkey could be seen in a painted frieze at Schleswig Cathedral, which had been built about AD 1280 (Hennig 1940). That claim was rebutted by Stresemann (1940; Rieth 1967) who showed that the mural had been restored in the1800s and 1900s; hence the rendering of the turkey as it existed in 1940 could have depended on knowledge acquired after Columbus (Bökönyi and Jánossy 1959; Varshavsky 1961). Nevertheless, findings since that time have restored the plausibility of a pre-Columbian origin for the Schleswig representation. Hungarian archaeologists found bones of a turkey in the 14th-century royal castle at Buda. Turkey bones have also been excavated from a carefully dated 14th-century-site in Switzerland (Bökönyi and Jánossy 1959). Other Hungarian sites of the 10th to 13th centuries have yielded signet rings engraved with images of this fowl, that show the fleshy pendent growth on the turkey's neck. Furthermore, a letter written in 1490 by Hungarian King Matthias, who died later that year, requested through an envoy that the Duke of Milan send him turkeys ("galine de Indie"). The king wanted to acclimatize the bird in Hungary. He also asked that a man who knew how to care for turkeys be sent with the birds (Bökönyi and Jánossy 1959). Obviously, the fowl was in Europe before Columbus' first voyage.

Confirmation of the late medieval European distribution of the turkey appears in a comment in a Relación (report of inspection) from Mérida, Yucatan, from about 1579. "There are many turkeys in the mountains which differ little from those in Spain, very good to eat, very timid birds" (Tozzer 1941, viii, 186; emphasis added). The conventional view is that the turkey was brought from the Americas to Spain by the returning conquistadors probably no earlier than 1523. It seems doubtful that their progeny in Iberia would have multiplied in 50 years to such an extent as to be spoken of in the off-handed manner of this Relación.

Mya arenaria

The case of Mya arenaria, the soft-shelled clam, is probably tied to the Norse voyages in the North Atlantic. Previously thought only to occur in American waters, this species was recently found offshore of Denmark (Peterson et al. 1992). A radiocarbon date on the shells falls in the 13th century, leaving only a slight possibility that the clam could have reached European waters after Columbus. Of course, the transfer to European waters had to have been by ship (perhaps in bilge). Not surprisingly, the investigating scientists decided that the mollusk "could have been transferred from North America to Europe by the Vikings." In the absence of credible alternative scenarios, we think the case should be put more firmly than that.

Stegobium paniceum

The beetle species, Stegobium paniceum, a pest usually found in dried, stored vegetable matter, is documented as discovered in both Egyptian and Peruvian burials. Greater incidence and earlier recognition of the species in the Old World favors that hemisphere as the place of origin. However, in Peru, S. paniceum was also found with mummies dated at least to the 13th century AD (Riddle and Vreeland 1982). Burials in Egypt as early as 3400 BC have revealed this same 'drugstore,' or 'biscuit,' beetle (Buckland and Panagiotakopulu 2001, 6), and it was present also in both Roman (Hall and Kenward 1990) and Bronze Age England (Panagiotakopulu 2000, 16).

Additional evidence for possible bi-hemispheric species of fauna can be found in the appendix.

Turning Parsimony Around

As we view all the evidence, it seems clear that a total approaching 100 plant species were moved across the ocean to or from the Americas before 1492. Furthermore, it is plausible that additional plants will be shown to have crossed as well; we cannot imagine that our present list is exhaustive. Tables 2 and 3 list additional candidates from the flora that ought to be further researched. Of the microfauna, 21 species show what decisive evidence of prehistoric transoceanic movement, and another 18 species deserve more study in that regard. Of other species of fauna, we count 6 as conclusively demonstrated to have been distributed in both hemispheres, and 6 others are possible.

This evidence puts a new complexion on long-standing questions about transoceanic movements of humans and associated flora and fauna. The outdated stance was illustrated by Spinden's statement (1933) that, "the fact that no food plant is common to the two hemispheres is enough to offset any number of petty puzzles in arts and myths" [such as the patolli/pachisi game]. Thirty-eight years later the same argument was still being invoked: "There is no hard and fast evidence for any pre-Columbian human introduction of any single plant or animal across the ocean from the Old World to the New World, or vice-versa" (Riley et al. 1971, 452—53). The logic of those days held that since "the bulk of the evidence" from biology was generally construed as being against any direct inter-hemispheric contact, every item of evidence in apparent contradiction to the orthodox view ought to be discarded or held in abeyance.

As late as 1985, Willey tried to tighten screws on the evidence even further by insisting that, "No Old World manufactured object has yet been found in an indisputable, undisturbed New World context. If nothing concrete can be shown in the next 50 years, proponents should stop talking about it." But no arbitrary stricture like this can be imposed on the question. Evidence speaks for itself, whether it consists of a shared "manufactured object" or a natural feature.

In the light of our findings, parsimony should now be interpreted quite differently from what it formerly was. Given that so many organisms were demonstrably shared between the hemispheres before Columbus, "the bulk of the evidence" today actually supports a voyaging explanation. Not only can we expect additional confirmation to come from further study of the flora and fauna (including DNA studies, archaeology, and more careful study of ancient art), but we may also find that many of the "petty puzzles" in culture, formerly rejected as proof of contact, now will turn out to be in agreement with "the bulk of the (new) evidence." It thus deserves careful consideration instead of perfunctory dismissal.

Cultural Freight

It is obvious that cultural (as well as human genetic) features had to have been transported with the flora and fauna on transoceanic voyages. A full discussion of the significance of the biological facts must take account of concomitant cultural sharing.

Domesticated plants and animals are almost never successfully transplanted by human agency to a strange area without appropriate care being given to the specimens being moved. Cultural norms for the preservation and exploitation of new organisms must be transmitted along with the crop plants if they are to survive and flourish in their new setting. That essential knowledge comprises botanical data, agricultural practices, culinary technology, and other measures needed to ensure that the transported plants are correctly cultivated and usefully employed on the new scene.

Moreover, the skills essential for making ocean voyages generally involve navigational, astronomical, and calendrical lore—concepts that could well survive at the destination. We can be confident also that a substantial body of myth, beliefs, and ritual practices would have accompanied the voyagers. A new linguistic and artistic repertoire would also have been introduced by the newcomers.

Speculation that people arriving from abroad would automatically be killed or their cultural baggage rejected is not supported by historical cases. The notion that such would have been the fate of voyagers probably owes more to Victorian stereotypes about 'cannibals' eating Christian missionaries than to ethnographic reality. Curiosity is the response to new arrivals at least as often as hostility.

Thus, not only does our documentation of the transport of flora and fauna across the oceans open the door for further studies in biological science (for example, we have noted few of the possibly large number of weeds inadvertently transported by voyagers), it also demands reconsideration of cultural parallels that have heretofore been categorically thrown out of court by almost all scholars when treating the issue of Old World/New World contacts.

Let us examine a single geographically focused setting for inter-hemispheric contact in order to appreciate how biological facts might connect to cultural data. The data in this book show that as many as 50 species of plants definitely, or very possibly, were transferred between the American tropics and India, or vice versa, before Columbus' day. While we cannot tell how many voyages this long process involved, there must have been several score—or maybe several hundred—stretched over millennia. Given the apparent scale of biological contact, one would a priori expect substantial cultural interchange as well. For decades researchers have been spelling out data that they consider show a connection between ancient civilizations in India and Mesoamerica, although the nature, timing, and significance of the influences at play have remained vague. (We recognize that considerable cultural evidence that has been offered has been of poor quality and deserves to be ignored.)

As mentioned earlier, Tylor's 19th-century (1896) identification of striking parallels between the South Asian pachisi and the Mexican patolli board games has never had a satisfactory explanation in terms of parallel, independent invention. In the 1920s, G. Elliot Smith added more cultural parallels between the two areas (see especially his 1924 book that treated elephant symbolism; the Mexican and Buddhist 'purgatory' ordeal; the makara, or 'dragon;' and miniature ritual vehicles bearing sacred figures drawn by animals). A series of articles by Milewski (1959, 1960, 1961, 1966) pointed to many conceptual parallels between deity names in Sanskrit on the one hand, and Aztec (Náhuatl) and Zapotec names on the other. Giesing went on (1984) to compile 50 pages of names and epithets for the Hindu god, Siva/Shiva, with which names and titles for the Aztec god, Tezcatlipoca, prove to be congruent. The fire-god complexes of India (Agni) and of central Mexico (Xiuhtecuhtli) were meticulously compared by Cronk (1973), who found extensive and startlingly detailed parallels.

Kelley (1960; Moran and Kelley 1969) argued that much that was basic in Mesoamerican calendrics, cosmology, and mythology is traceable to India of the last centuries BC and to nowhere else as clearly. Durbin (1971) was sufficiently impressed with Kelley's proposals that he suggested a set of lexical links between Prakrit, Sanskrit-derived languages of India, and proto-Mayan in Central America. Mukerji (1936) claimed to demonstrate specific astronomical correlations between the Maya and Hindu calendars. Kirchoff (1964a, 1964b) laid out large blocks of material on conceptual and structural features of the calendars and mythology of Eastern and Southern Asia, also apparently in Mesoamerica. Barthel (1975a, 1975b, 1982, 1985) did a series of intricate studies of Mesoamerican codices and calendars which he believed confirm that a Hindu 'missionary' effort reached Mexico, only to be obscured by a later 're-barbarization' of the transplanted concepts. The sacred figures who hold ears of corn on temple sculptures in India do so with hands in symbolic positions, or mudra gestures, while Mesoamericanists have noted a repertoire of mudras shared by Indian and Mesoamerican art (Martí 1971; Medvedov 1982). And Compton (1997) has pointed out elaborate parallels between Aztec and Buddhist etiological myths involving the rabbit and the moon.

These studies, plus many more that could be cited, have typically been presented by diffusionists at a high level of abstraction, as though disembodied elements of 'Indian culture' or 'Mesoamerican civilization' were somehow wafted across the ocean where they lodged in the minds of the locals. Protagonists of diffusion have rarely proposed, let alone documented, plausible historical scenarios that would account for the parallels they propose. That is, they have not hypothesized actual voyages in which culturally knowledgeable persons with believable motives are supposed to have boarded specific kinds of vessels to travel along nautically feasible routes and then arrive at particular locations in the opposite hemisphere, where they significantly affected existing cultures. But the time is at hand when such plausible scenarios can be proposed.

Concrete data on biology has the potential to help relate cultural features to dates and locations. The degree of concreteness this would furnish to investigations of cultural parallels may allow researchers to formulate focused and convincing hypotheses about when, where, and how sharing took place. For example, the fact that important American crops were represented in Indian art, mentioned in texts, and found in excavations, might provide concrete chronological and material settings to relate to, say, Kelley's, Cronk's, and Barthel's hypotheses about Indian intellectual and religious influence on Mesoamerica in the late BC centuries. Yet India is only one area of influence to which the evidence points.

We emphasize that by momentarily focussing on the India/America interchange, we do not consider other origin/destination pairs non-credible. The evidence is strong for South America/Polynesia, Mesoamerica/Hawaii, several American scenes connected to Southeast (especially Indonesia) and East Asia, and Mediterranean/Mesoamerican links. But those are matters to be delineated elsewhere.

Summary Points

For now, the following summary points are apparent from our analysis of the biological data. Each bypasses old conceptions and opens up new avenues of inquiry.

But we do not intend that our findings spawn new dogma of any kind. Rather, we hope that intellectual curiosity and openness, disciplined by sound research and logic, will prevail among the next generation of investigators, so that they may go beyond, not only where the old paradigm of culture-history allowed, but also beyond the perspective we have reached.

Biographical Notes

1. John L. Sorenson is emeritus professor of anthropology at Brigham Young University, where he founded work in that discipline in the 1950s. He was attracted from the physical sciences (he holds an M.S. degree from the California Institute of Technology) to anthropology by way of archaeology, but after beginning study for the Ph.D. at UCLA, his focus shifted to sociocultural anthropology with emphasis on its applications to problems of modern society.

After completing the Ph. D. degree, he went to BYU. In 1964—1969 he went as head social scientist to General Research Corp., a Santa Barbara, CA, think tank. In 1969, he founded Bonneville Research as a subsidiary of GRC, in Provo, UT, before returning to the BYU faculty in 1971. He served as chair of the department of anthropology for eight years before retiring in 1986. Since then, he has returned to his early interests in Mesoamerican archaeology and transoceanic contacts, seeking to engender in those studies some of the rigor he learned from doing 'hard science.'

2. Carl L. Johannessen is emeritus professor of biogeography in the Department of Geography at the University of Oregon at Eugene. He taught at that university from 1959 to 1994. His M.A. in zoology and Ph.D. in geography came from the University of California at Berkeley.

He studied the distribution of human-modified wild vegetation in Latin America first, but after a decade, the domestication process became his focus. That work led to a search for how human-caused modifications in flora and fauna were accomplished and what were the resultant distributions. By the 1980s, his interest had come to center on evidence for the movement of organisms by voyagers across the oceans before Columbus.

He has made many research trips to India, China, Europe, the Middle East, Polynesia, and Latin America in pursuit of evidence in the literature and in the field of the distribution of plants (domesticates and weeds) and animals transferred long distances by humans. Most recently, he has expanded that topic to include the entire process of transoceanic diffusion in human history.

Go on to part 2 of "Scientific Evidence for Pre-Columbian Transoceanic Voyages to and from the Americas"