Index | Search | Home | Table of Contents

Eshbaugh, W.H. 1993. History and exploitation of a serendipitous new crop discovery. p. 132-139. In: J. Janick and J.E. Simon (eds.), New crops. Wiley, New York.

Peppers: History and Exploitation of a Serendipitous New Crop Discovery*

W. Hardy Eshbaugh

  9. Table 1
  10. Fig. 1

Few could have imagined the impact of Columbus' discovery of a spice so pungent that it rivaled the better known black pepper from the East Indies. Nonetheless, some 500 years later, on the quincentennial anniversary of the discovery of the New World, chili peppers (Capsicum) have come to dominate the world hot spice trade and are grown everywhere in the tropics as well as in many temperate regions of the globe. Not only have hot peppers come to command the world's spice trade but a genetic recessive non-pungent form has become an important "green" vegetable crop on a global scale especially in temperate regions.

The New World genus Capsicum is a member of the Solanaceae, a large tropical family. Various authors ascribe some 25 species to the genus but this is only an estimate with anticipated new species to be discovered and named as exploration of the New World tropics expands. Exploration and plant collecting throughout the New World have given us a general but false impression of speciation in the genus. Humans unconsciously selected several taxa and in moving them toward domestication selected for the same morphological shapes, size, and colors in at least three distinct species. Without the advantage of genetic insight these early collectors and taxonomists named these many size, shape, and color forms as distinct taxa giving us a plethora of plant names that have only recently been sorted out reducing a long list of synonymy to four domesticated species. The early explorations in Latin America were designed to sample the flora of a particular region. Thus, any collection of Capsicum was a matter of chance and usually yielded a very limited sample of peppers from that area. Only with the advent of collecting trips designed to investigate a particular taxon did the range of variation within a species begin to be understood. One needs only to borrow specimens from the international network of herbaria to appreciate what a limited sample exists for most taxa, particularly for collections made prior to 1950. The domesticate Capsicum pubescens, for example, that is widespread in the mid-elevation Andes from Colombia to Bolivia, is barely represented in the herbarium collections of the world. Most herbarium collections of Capsicum, with the exception of Capsicum annuum holdings, are woefully inadequate. Furthermore, besides Capsicum annuum, very little attention has been paid to the many cultivars of each of the domesticated species. Often material is unusable because it was collected only in fruit neglecting the most important and critical characters associated with floral anatomy and morphology. With the advent of germplasm collecting programs during the past three decades, and concomitant improvement in herbarium collections we have come to better understand the nature of variation in the genus Capsicum. The increasing number of Capsicum herbarium specimens permits renewed interest and debate on the proper species classification.


One of the more perplexing questions regarding the taxonomy of Capsicum is defining the genus (Eshbaugh 1977, 1980b; Hunziker 1979). The taxonomy of the genus Capsicum is confounded within certain species complexes, e.g. C. baccatum sensu lato. Major taxonomic difficulties below the species level in other taxa, e.g. C. annuum, also exist. Armando T. Hunziker (unpublished) is currently working on a revision of the genus. What taxa are ultimately included in Capsicum may indeed change if the concept of the genus is broadened to include taxa with non-pungent fruits but with other common morphological and anatomical traits such as the nature of the anther, the structure of nectaries, and the presence of giant cells on the inner surface of the fruit (Pickersgill 1984).

Capsicum, as presently perceived, includes at least 25 species, four of which have been domesticated (Table 1). An understanding of each of these domesticates is instructive when trying to appreciate their origin and evolution. The data from plant breeding and cytogenetics confirm that the domesticated species belong to three distinct and separate genetic lineages. Earlier studies suggested two distinct lineages based upon white and purple flowered groupings (Ballard et al. 1970) but an evaluation of more recent data argues for the recognition of three distinct genetic lineages. Although the barriers between these gene pools may be broken down this rarely, if ever, occurs in nature.


Capsicum pubescens forms a distinct genetic lineage. This pepper, first described by Ruiz and Pavon (1794) never received wide attention from taxonomists until recently (Eshbaugh 1979, 1982). Morphologically, it is unlike any other domesticated pepper having large purple or white flowers infused with purple and fruits with brown/black seeds. Genetically, it belongs to a tightly knit group of wild taxa including C. eximium (Bolivia and northern Argentina), C. cardenasii (Bolivia), and C. tovarii (Peru). Capsicum pubescens is unique among the domesticates as a mid-elevation Andean species. Capsicum pubescens is still primarily cultivated in South America although small amounts are grown in Guatemala and southern Mexico, especially Chiapas. This species remains virtually unknown to the rest of the world. A small export market seems to have reached southern California. Two of the major difficulties in transferring this species to other regions include (1) its growth requirements for a cool, freeze free environment and long growing season and (2) the fleshy nature of the fruit that leads to rapid deterioration and spoilage.


Capsicum baccatum var. pendulum represents another discrete domesticated genetic line. Eshbaugh (1968, 1970) notes that this distinct South American species is characterized by cream colored flowers with gold/green corolla markings. Typically, fruits are elongated with cream colored seeds. The wild gene pool, tightly linked to the domesticate, is designated C. baccatum var. baccatum and is most common in Bolivia with outlier populations in Peru (rare) and Paraguay, northern Argentina, and southern Brazil. This lowland to mid-elevation species is widespread throughout South America particularly adjacent to the Andes. Known as aji, it is popular not only as a hot spice but for the subtle bouquet and distinct flavors of its many cultivars. This pepper is little known outside South America, although it has reached Latin America (Mexico), the Old World (India), and the United States (Hawaii). It is a mystery as to why it has not become much more wide spread, although the dominance of the Capsicum annuum lineage throughout the world at an early date may be responsible.


Pickersgill (1988) has stated that "the status of Capsicum annuum, C. chinense, and C. frutescens as distinct species could legitimately be questioned." Several authors have previously raised this issue culminating in the observation that "at a more primitive level one cannot distinguish between the three species. On the one hand we treat the three domesticated taxa as separate while the corresponding wild forms intergrade to such an extent that it is impractical if not impossible to give them distinct taxonomic names" (Eshbaugh et al. 1983). McLeod et al. (1979, 1983) have argued that isoenzyme data make it impossible to distinguish between these three taxa. From an extensive isoenzyme study of these three taxa and several other species, Loaiza-Figueroa et al. (1989) argue that "thus far, this substitution of alleles constitutes a good argument against the proposal that these species form an allozymically indistinguishable association of a single polytypic species" as advanced by Mcleod et al. (1982, 1983) and Eshbaugh et al. (1983) in their published studies. Nonetheless, Pickersgill (1984) has pointed out that "each domesticate intergrades with morphologically wild accessions by way of partially improved semidomesticates. Any subdivision of the wild complex into three taxa, each ancestral to one of the domesticates, becomes decidedly arbitrary, although clusters corresponding to wild C. annuum, C. chinense, and wild C. frutescens can be detected." Clearly, Loaiza-Figueroa et al. missed the point of these earlier papers which argue for the complexity of the problem noting that the real difficulty comes as one approaches the more primitive forms of these taxa. Furthermore, the Loaiza-Figueroa et al. (1989) dendrogram (p. 183) suggests that the number of C. chinense and C. frutescens taxa included in their study is insufficient to reach any definitive conclusion regarding the status of these three taxa. There is a very close relationship of these three taxa based on crossing data from several studies (Smith and Heiser 1957; Pickersgill 1980). Stuessy (1990) has observed that "the ability to cross does not just deal with a primitive genetic background; it deals with the degree of genetic compatibility developed in a particular evolutionary line." As Stuessy (1990) has inferred there can be no stronger argument for relationship than the data obtained from plant breeding. Regardless of one's viewpoint, it is clear that the C. annuum--C. chinense--C. frutescens complex has been and continues to be a most difficult taxonomic morass. Some preliminary information from the studies of Gounaris et al. (1986) and Mitchell et al. (1989) suggest that molecular data may be useful in resolving this and other taxonomic questions. For the present, I have chosen to recognize the Capsicum annuum complex and the Capsicum chinense complex as two distinct domesticated species. Where C. frutescens fits into this scenario remains to be resolved. William G. D' Arcy, A.T. Hunziker, and others may solve the problem by merging the three taxa under a single taxonomic entity. Taxonomists and formal taxonomy are having a very difficult time coping with what is a complex and dynamic evolutionary process. The problem is heightened by the economic importance of Capsicum and the requirement that not only the domesticated species be named properly but that the several cultivars receive taxonomic recognition.

Capsicum annuum is the best known domesticated species in the world. Since the time of Columbus, it has spread to every part of the globe. The non-pungent form, bell pepper, is widely used as a green vegetable. Another non-pungent form, "pimento," is also present throughout much of the globe. The hot spicy forms of this species have come to dominate the spicy foods within Latin America and the rest of the world. Capsicum annuum probably became the dominant pepper globally in part because it was the first pepper discovered by Columbus and other New World explorers (Andrews in press). This taxon was the first Capsicum species taken to Europe and quickly spread to other regions.

Capsicum chinense was also discovered at an early date and spread globally but to a lesser extent than C. annuum. The more limited global expansion of this species is most probably related to its later discovery in South America and the competitive edge enjoyed by C. annuum which was firmly established in the Old World before C. chinense was introduced there.


A discussion of the geography of Capsicum touches on two questions. The first relates to the origin of the genus Capsicum and the second to the origin of the domesticated taxa. The area of origin of Capsicum cannot be resolved until we understand the nature of the genus. If we accept the genus as currently circumscribed and limited to pungent taxa, then a clear center of diversity is to be found ranging from southern Brazil to Bolivia (McLeod et al. 1982; Eshbaugh et al. 1983; Pickersgill 1984). However, if the genus is reconstituted to include other non-pungent taxa, another center of diversity may be recognized in Central America and southern Mexico. Ultimately, our definition of the genus Capsicum and what species it includes will determine our view of its center of origins and whether the genus is monophyletic or polyphyletic. The emerging molecular studies of J.D. Palmer and R.G. Olmstead should give us a better sense of where Capsicum belongs within the framework of the Solanaceae.

Determining the place of origin of the genus and each of the domesticated species is at best a problematic exercise. In 1983, I stated that "it appears that the domesticated peppers had their center of origin in south-central Bolivia with subsequent migration and differentiation into the Andes and Amazonia." This is a condensation of a highly speculative hypothesis (McLeod et al. 1982). From that hypothesis Pickersgill (1989) later suggested that I (Eshbaugh 1983) argued that all the domesticated taxa arose in Bolivia. Without question, I could have stated this idea more clearly. We (McLeod et al. 1982) have speculatively hypothesized that Bolivia is a nuclear center of the genus Capsicum and that the origin of the domesticated taxa can ultimately be traced back to this area. That does not imply that each of the domesticated species arose in Bolivia. Clearly, evidence supports a Mexican origin of domesticated C. annuum while the other domesticated species arose in South America. Nonetheless, the ancestry of the domesticates can be traced to South America. While McLeod et al. (1982) have hypothesized a Bolivian center of origin for Capsicum there is no evidence for a polyphyletic origin of the genus as now understood.

Evidence suggests that C. annuum originally occurred in northern Latin America and C. chinense in tropical northern Amazonia (Pickersgill 1971). Capsicum pubescens and C. baccatum appear to be more prevalent in lower South America (Fig. 1). Thus, at the time of discovery, the former two species were exploited while the later two species awaited a later discovery and remain largely unexploited outside South America today.

In considering the question of origin of each particular domesticated species two issues must be considered. First, what wild progenitor is the most likely ancestor of each domesticated species and second, where is the most probable site of domestication?

Capsicum pubescens ranges throughout mid-Andean South America. An analysis of fruit size of this domesticate indicates that fruits of a statistically smaller size occur in Bolivia, while fruits from accessions outside Bolivia on the average are somewhat larger suggesting that Bolivian material approaches a more primitive size (Eshbaugh 1979).

Eshbaugh (1979, 1982) has argued that the origin of this domesticate can be found in the "ulupicas," C. eximium and C. cardenasii. Clearly, these two taxa are genetically closely related to each other and C. pubescens. Natural hybrids between these taxa have been reported and evaluated (Eshbaugh 1979, 1982). Furthermore, the two species that show the highest isoenzyme correlation with C. pubescens, C. eximium and C. cardenasii, occur primarily in Bolivia (Eshbaugh 1982; McLeod et al. 1983; Jensen et al. 1979). All three of these taxa form a closely knit breeding unit with the two wild taxa hybridizing to give fertile progeny with viable pollen above the ninety percent level. Crosses between the wild taxa C. eximium and C. cardenasii and the domesticate C. pubescens most often show hybrid pollen viability greater than 55%. These factors lend to the conclusion that domesticated C. pubescens originated in Bolivia and that C. eximium--C. cardenasii is the probable ancestral gene pool. This does not prove that these two taxa are the ancestors of C. pubescens but of the extant pepper taxa they represent the most logical choice. One perplexing question remains to be investigated and that is the origin of the brown/black seed coat in domesticated C. pubescens, a color unknown in any of the other pepper species.

Capsicum baccatum var. pendulum is widespread throughout lowland tropical regions in South America. It ranges from coastal Peru to Coastal, Brazil. The wild form, recognized as C. baccatum var. baccatum, has a much more localized distribution but still ranges from Peru to Brazil. These two taxa have identical flavonoid (Ballard et al. 1970; Eshbaugh 1975) and isoenzyme profiles (McLeod et al. 1979, 1983; Jensen et al. 1979) and are morphologically indistinguishable except for the overall associated size differences found in the various organ systems of the domesticated taxon (Eshbaugh 1970). The wild form of Capsicum baccatum exhibits a high crossability index with domesticated C. baccatum var. pendulum with the progeny typically exhibiting pollen viability in excess of 55 percent (Eshbaugh 1970). The greatest center of diversity of wild C. baccatum var. baccatum is in Bolivia leading to the conclusion that this is the center of origin for this domesticate.

Can we ever unscramble questions about the origin and evolution of the C. annuum--C. chinense--C. frutescens species complex? Pickersgill (1989) states that there is an "overwhelming likelihood of at least two independent domestications of the chile peppers of this complex." She also notes that one "may ... argue about whether wild forms of this complex should really be assigned to different species, and indeed whether domesticated C. annuum and domesticated C. chinense are really conspecific." I would agree that the evolutionary lineage of C. annuum--C. chinense--C. frutescens complex is intimately linked but I would further emphasize that when, where, and how they diverged is obscured in antiquity and that the extant wild forms of these three taxa are so similar as to make them very difficult to separate. One might well ask whether, at a minimum, C. chinense and C. frutescens are conspecific or grades within the same species.

In contrast, a reasonably clear picture emerges on origin and progenitor of C. annuum. Capsicum annuum has its center of diversity in Mexico and northern Central America with a local, and more recent distribution in parts of South America. The wild bird pepper, Capsicum annuum var. aviculare, ranges from northern South America (Colombia) into the southern United States and Caribbean. Crossing studies indicate that the wild bird pepper is genetically the most closely related taxon to domesticated C. annuum (Emboden 1961; Smith and Heiser 1957; Pickersgill 1971). Pickersgill (1971), using karyotype analysis, suggests that the origin of domesticated C. annuum is to be found in southern Mexico. Pickersgill et al. (1979) also provided a detailed phenetic analysis of the C. annuum--C. chinense--C. frutescens complex and the difficulty of separating these taxa at the most primitive level is apparent.

Capsicum chinense remains the least understood of the four domesticated taxa with respect to center of origin and probable progenitor. If one maps the range of forms in C. chinense, it is clear that amazonian South America is the center of diversity of this species. Furthermore, C. chinense does occur sporadically throughout the Caribbean. It is likely that C. chinense spread into the Caribbean at a later date since the diversity of taxa is more limited in that region than in amazonian South America. In considering the progenitor of C. chinense, one is bewildered by the evidence. It has been suggested that C. frutescens, in its primitive form, may be the ancestor of C. chinense (Eshbaugh et al. 1983). However, one needs to ask whether C. frutescens is merely a weedy offshoot of C. chinense or C. annuum. It is clear that the three species, C. annuum, C. frutescens, and C. chinense, hybridize with each other. They form a morphological continuum especially at a primitive level (McLeod et al. 1979). Genetic evidence from isoenzymes also confirms the close relationship of these three taxa (McLeod et al. 1983; Jensen et al. 1979).


The spread of domesticated peppers throughout the world during the 500 years since discovery is truly a phenomenon. Two of the domesticated species, C. annuum var. annuum and C. chinense have been widely utilized on a global scale. Both C. baccatum var. pendulum and C. pubescens have been extensively exploited in South America but remain largely confined to that market. Given both the unique qualities and flavors of these later two species they each represent a potential source for future development.

Of special interest to those working with peppers is the use and exploitation of the wild species. Wherever wild taxa of Capsicum occur, humans use them for their hot properties. In a few cases, exploitation of wild species has reached a commercial level. Capsicum praetermissum is collected and sold commercially in parts of Brazil (reported by correspondents). Capsicum chacoense and C. eximium are collected and bottled and marketed throughout southern Bolivia (pers. observ.). Fresh C. cardenasii is harvested and transported to the La Paz, Bolivia market for sale (pers. observ.). In Mexico and the southwestern United States wild C. annuum var. aviculare, the chiltepin, has been locally used for many years (Nabhan et al. 1989). More recently, a commercial market has developed for chiltepin. A large amount of this wild species is now harvested and sold to the gourmet food market. Nabhan et al. (1989) indicate that "currently chiltepin is almost completely wild harvested." They note that "as much as 12 tons of chiltepines may be harvested from a single Sonoran municipio in a good year, but total harvest may vary from perhaps 8 to as high as 50 tons."

While the quantity of C. eximium, ulupica, being harvested in southern Bolivia is unknown, there is an extensive commercial trade in bottled whole peppers. Bolivians have not attempted to commercially plant wild plants of C. eximium, but Nabhan et al. (1989) indicate that incipient cultivation of C. annuum var. aviculare was initiated with extensive planting of the chiltepin by Sonoran farmers in the 1980s. The manipulation of these two wild species in each setting has led to some significant changes for the wild species. In both the case of C. annuum var. aviculare and C. eximium, larger fruit size has been selected for in the incipient area of cultivation and manipulation. Sonoran farmers are selecting for larger fruit size in the wild chiltepin. In C. eximium, there is a statistically significant larger fruit form of this ulupica in the zone of exploitation when compared to regions where the fruit is not widely collected (Eshbaugh 1979, 1982). In both cases we are witnessing incipient or semi-domestication of the wild species.

Apparently, a market exists for the exploitation of peppers for the medicinal properties of capsaicin and several companies are pursuing such investigations. Two of the more interesting products to come to market in the last five years are the prescription drug Zostrix (Genderm registered trade mark), an analgesic cream, containing 0.025% capsaicin that is used topically to treat shingles and to provide enhanced pain relief for arthritis patients and Axsain (GalenPharma registered trade mark) that contains 0.075% capsaicin and is used for relief of neuralgias, diabetic neuropathy, and postsurgical pain. Both products are believed to work by action on a pain transmitting compound called substance P.

Several pepper species, because of their unique fruit shapes and bright fruit colors, have been widely used as ornamentals. The presence of capsaicin, however is a potential hazard.


In August, 1980, an expert consultative group, under the auspices of the IBPGR (International Board for Plant Genetic Resources), met at CATIE (Centro Agronomico Tropical de Investigacion y Ensenanza) in Turrialba, Costa Rica, to discuss the status of Capsicum germplasm collections and to map a strategy for future collecting and management of these resources (Genetic Resources of Capsicum 1983). The discussions led to a plan to systematically collect Capsicum throughout New World paying particular attention to the wild species most closely related to the domesticated taxa. The efforts of the past decade have resulted in a significant accumulation of pepper germplasm (seeds) that is now stored in various collections. Eshbaugh (1980a, 1981, 1988) has detailed the history of Capsicum germplasm collecting prior to 1980 and discussed the collecting efforts of peppers in Bolivia. Capsicum germplasm collections are now maintained in a number of facilities in the United States, as well as Mexico, Costa Rica, Bolivia, and Brazil.


*The investigations cited and drawn upon in this paper were supported most recently by a grant from the National Science Foundation (BSR 8411136). I am indebted to many colleagues for their thoughts and comments regarding the evolution of the genus Capsicum, although I accept sole responsibility for the comments in this paper. I especially acknowledge Charles B. Heiser, Jr. who started me on a course of study into the evolution of the genus Capsicum that has lasted more than 30 years. My wife, Barbara, has been with me on every step of this journey and deserves a special thanks for all her support, patience, and understanding.

Table 1. Synopsis of the genus Capsicum (Solanaceae)zy.

Capsicum New world distribution
annuum L. Colombia north to southern United States
baccatum L. Argentina, Bolivia, Brazil, Paraguay, Peru
buforum Hunz. Brazil
campylopodium Sendt. South Brazil
cardenasii Heiser & Smith Bolivia
chacoense Hunz. Argentina, Bolivia, Paraguay
chinense Jacq. Latin and South America
coccineum (Rusby) Hunz. Bolivia, Peru
cornutum (Hiern) Hunz. South Brazil
dimorphum (Miers) O.K. Colombia
dusenii Bitter Southeast Brazil
eximium Hunz. Argentina, Bolivia
glapagoensis Hunz. Ecuador
geminifolium (Dammer) Hunz. Colombia, Ecuador
hookerianum (Miers) O.K. Ecuador
lanceolatum (Greenm.) Morton & Standley Mexico, Guatemala,
leptopodum (Dunal) O.K. Brazil
minutiflorum (Rusby) Hunz. Argentina, Bolivia, Paraguay
mirabile Mart ex. Sendt South Brazil
parvifolium Sendt. Colombia, Northeast Brazil, Venezuela
praetermissum Heiser & Smith South Brazil
pubescens Ruiz & Pav. Latin and South America
scolnikianum Hunz. Peru
schottianum Sendt. Argentina, South Brazil, Southeast Paraguay
tovarii Eshbaugh, Smith & Nickrent Peru
villosum Sendt. South Brazil
zThe following Capsicum species have been omitted: C. anomalum, C. breviflorum, and C. ciliatum following the earlier suggestion of Eshbaugh (1983). Also, C. flexuosum Sendt. has been treated as a variety of C. schottianum by Hunziker. The treatment of C. frutescens L. remains to be resolved and some may choose to retain it as a distinct species of Capsicum while others submerge it into C. chinense, as suggested in this paper. Finally, C. eximium var. tomentosum Eshbaugh & Smith is so distinctive that it may deserve species status.
yThis table has been developed, adapted, and modified from Hunziker 1956; Eshbaugh 1977, 1980b; and Pickersgill 1984.

Fig. 1. Distribution of domesticated species of Capsicum (solid circles) and wild progenitor taxa (open circles). Capsicum chinense and C. frutescens are combined together because they may represent the same taxon. Copyright Bul. Torrey Bot. Club, 1975.

Last update September 9, 1997 aw