Selbyana 19(2): 227-235
TAXONOMY OF THE GENERA ANANAS AND PSEUDANANAS-AN HISTORICAL REVIEW
FREDDY LEAL Universidad Central de Venezuela, Facultad de Agronomia, Apartado 4736, Maracay, Aragua, Venezuela
GEORGE COPPENS D'EECKENBRUGGE
CIRAD-FLHOR/IPGRI, % CIAT, AA 6713, Cali, Colombial
BRUCE K. HOLST Marie Selby Botanical Gardens, 811 South Palm Avenue, Sarasota, Florida 34236, USA
ABSTRACT. From the first observations of the pineapple by European explorers to the classification into Pseudananas sagenarius and the seven Ananas species prevailing at present, the evolution of pineapple taxonomy has shown considerable variation. Most early botanists named or renamed species from previous dubious descriptions or from particular horticultural types. More recently, the genus Pseudananas was created, while horticultural types disappeared from the Ananas species. Subsequently, the total number of species increased again as botanical varieties and forms with minor variation were elevated to species rank. The resulting classification is questionable, as neither discontinuous morphological variation nor reproduc tive barriers exist in the genus Ananas.
Ananas and Pseudananas are the only genera peduncle. The dry fibers constitute 6% of plant in the Bromeliaceae whose fused flowers devel weight (Camargo 1943). The absence of spines op into a sorose-type fruit formed by the coa facilitates manual fiber extraction, although lescence of up to 200 berries. The most recent some spiny or partially spiny mutants have been classification (Smith & Downs 1979) established observed. Ananas lucidus has never been found Pseudananas as a monotypic genus, Pseudan in the wild. anas sagenarius (Arruda) Camargo, and divided Ananas bracteatus (Lindl.) Schult. f. comes Ananas divided into eight species, one of which from southern South America (southern Brazil, has been invalidated (Leal 1990). The species Paraguay, and northern Argentina), where it is considered valid now are presented in TABLE 1. always found cultivated as a living hedge or for The best known species of Ananas is the cul fruit juice, or abandoned in ancient settlements. tivated A. comosus (L.) Merr., thanks to its im Its variegated form has been widely diffused as pressive and delicious fruit, which is considered a garden ornamental. The plant is vigorous with an important taxonomic character (more than 15 wide and long leaves, large spines, and suckers cm long according to the key in Smith & Downs abundantly. The inflorescence is characterized 1979). This large fruit is borne on a wide, short by its bright pink to red color and long bracts. to-long peduncle. In the spiny genotypes, spines The fruit and peduncle are medium-sized (ac are antrorse and generally smaller and denser cording to the key presented in Smith and than in other species. Partially spiny genotypes Downs (1979), the syncarp is more than 15 cm also exist as well as completely smooth geno long; however it is often less). Ananas bractea types, characterized by a folding of the lower tus, well adapted to cool conditions and altitude, epidermis over the leaf edge ("piping" character has been observed at 1,000 m in the subtropics. as named by Collins & Kerns 1946). The variability of this species is very limited. Ananas lucidus Mill., the curagua, is cultivat Ananas Jritzmuelleri Camargo is almost identical ed by the peoples of the Orinoco basin and to to A. bracteatus with some retrorse spines (Ca the North of the Amazon River. They use its margo 1943). Formerly included in A. bractea strong and long fibers to make breechcloths, tus (Smith 1939), A. jritzmuelleri has floral hammocks, fishing nets and rods (Leal & Amaya bracts that turn a pale green color at fruit ma 1991). This species is mainly characterized by turity. long, smooth, and erect leaves and a small, fi Ananas ananassoides (Baker) L.B. Sm., A. brous, inedible fruit borne on a medium-to-long parguazensis Camargo and L.B. Sm., and A. nanus (L.B. Sm.) L.B. Sm. are wild species. An anas ananassoides is the most widespread spe 1 Address for correspondence. cies, from southern Brazil to Venezuela and Co-
227 228 SELBYANA [Volume 19(2)
TABLE 1. Current composition of the genera Pseu As commonly found in Bromeliaceae, the ge dananas and Ananas. Based on Smith and Downs nus Ananas is diploid, characterized by having (1979) and Leal (1990). 50 minute and almost spherical chromosomes in both root tips and pollen mother cells (Collins Scientific name Common name & Kerns 1931, Canpinpin & Rotor 1937, Mar Pseudananas sagenarius Gravata de cere a, gravata chant 1967, Sharma & Ghosh 1971, Lin et al. (Arruda) Camargo de rede, yvira, nana ca 1987, Brown & Gilmartin 1989, Dujardin 1991). ",aba, nana brava Giant unreduced gametes may appear and pro Ananas ananassoides Ananas de ramosa, curi duce natural triploids and tetraploids (Collins (Baker) L.B. Sm. bijul, maya pifion, na 1933, 1960). Most genotypes display reduced nlli, pifiuela fertility and a self-incompatibility system with Ananas nanus (L.B. Sm.) Ananai L.B. Sm. considerable variation in its expression (Cop Ananas parguazensis Ca Gravata, pilla montafiera pens d'Eeckenbrugge et al. 1993). Vegetative re margo and L.B. Sm. production is largely dominant over sexuality in Ananas lucidus Mill. Curagua, curana, curaua, Ananas. This is particularly true in domesticated kulaiwat types. Pineapples multiply by suckers (terrestrial Ananas bracteatus Ananas bravo, ananas do and aerial), slips (suckers from the peduncle), (Lindt) Schult. f. mato and crown. Ananas fritzmuelleri Ca Ananas silvestre, gravata Pseudananas sagenarius has been cultivated margo de cerca as a source of fibers, which explains the origin Ananas comosus (L.) Abacaxi, ananas, pilla, Merr. matzatli of its common name, yvira, meaning fiber, as well as of its Latin epithet, which refers to a fishing net (sagena). The fibers constitute 7.6% of the leaf dry weight and reach a length of 1.60 lombia. Although a few genotypes thrive in m (Correa 1952). In contrast to Ananas, P. sa dense rainforest (in the Guianas). it is generally genarius is a tetraploid (211 = 100), character observed in savannas or in low-shade forest, ized by a complete lack of crown and asexual growing well on soils with limited water-holding reproduction by stolons. Its leaf margins bear capacity (such as sand or rocks), and forming strong spines, which are retrorse at the leaf base. populations of variable densities. Most of these Floral bracts are longer than in Ananas como sus. populations are monoclonal, but some are poly The fruits are low in acid. The habitat of P. sa clonal, with variation of recent sexual origin genarius, limited to forest areas, is under semi (Duval et al. 1997). The plant has long and gen dense shade and subjected to a rainy season dur erally narrow spiny leaves and bears a small, ing most of the year, or even to periods of flood globular-to-cylindrical syncarp on a long and ing. This species may be found in southern thin peduncle. The fruit is often seedy, and its South America (southern Brazil, Paraguay, Bo pulp is white, firm and fibrous, with a high sugar livia and northern Argentina), and along the and acidity content, good flavor and aroma, and eastern Brazilian coast, up to Pernambuco. Pop a narrow heart. According to the key, A. anan ulations of P. sagenarius are rare now because assoides is distinguished from A. comosus by the of severely reduced habitat. The few populations size of the fruit (shorter than 15 cm). recently observed, however, showed significant Ananas nanus is characterized by an even variation (Ferreira & Cabral 1993, Ferreira 1996, smaller fruit (shorter than 4 em), In fact, Smith Duval et ai. 1997, Coppens d'Eeckenbrugge et (1939) fonnerly classified it as a dwarf variety al. 1997). of A. ananassoides. Ananas nanus has some times been used as an ornamental. HISTORY OF PINEAPPLE TAXONOMY Ananas parguazensis is also very similar to A. ananassoides but differs in the retrorse ori The main steps in the evolution of pineapple entation of some spines and a wider leaf slightly taxonomy are presented in FIGURE 1. constricted at its base. Ananas parguazensis The Amerindians had domesticated pineap seems less adapted to drought, but an anatomical ples and the curagua for fruit and fiber and had and physiological comparison showed no other dispersed the fonner all over Latin America and differences (Leal & Medina 1995). Its distribu the Caribbean and the latter from the North bank tion is also limited to the northern Amazon (Rio of the Amazon to the Caribbean well before the Negro basin) and Rio Orinoco, with a wider var anival of Columbus (Leal & Coppens d'Eeck iability in the Orinoco region (Duval et ai. enbrugge 1996). They also had domesticated the 1996). It grows in the lowland forests, under yvira, as observed by Thevet (1557) around Rio canopies of variable densities, from clearings or de Janeiro. Native Americans were the first pine riverbanks to dense forest. apple taxonomists. On the island of Hispaniola, ..... Plumier (1755) \0 Pseudananas Hassler ex Harms 1930 Ananas Miller 1754 \0 r< < ------I - Bromelik sylvestris«'
XX Camargo (I956) ~ j j A. lyman-smithii ~ ..,V:l A. parguazenSiSj A. nanus A. Smith (1961, 1962) >- m~lnstrosus Camargo & Smith (1968) >< 0 1 A. tCidUS Smith (1971 Z 1 0 P
FIGURE L A diagrammatic representation of the main changes in pineapple taxonomy during the 18th, 19th, and 20th centuries. tv tv \0 230 SELBYANA [Volume 19(2) they distinguished different bromeliads, such as TABLE 2. The genus Ananas according to P. Miller's "karatas" and "ananas" (Plumier 1755). The The Gardener's Dictionary, 4th ed" 1754, and 8th vernacular "karatas" and related names as "kar ed., 1768. London. agwata" or the Brazilian "gravaHi" are still used by local people throughout South America 4th Edition 8th Edition to name terrestrial bromeliads-in addition to Ananas aculeatus, fructu Ananas (Ovatus). Oval "nana" and "anana" which are used commonly ovato, carne albida. shaped pineapple, with for pineapple throughout the Amazon and Ori Plum. a whitish flesh noco basins, as well as in southern South Amer Ananas aculeatus, fructu Ananas (Pyramidalis). ica. Wild pineapples are often called "nanai" or pyramidata, carne au Pyramidal pineapple, rea. Plum, with a yellowish flesh, "anana!" (Leal & Coppens d'Eeckenbmgge 1996). (called the sugar loaf The word "anana" from the Carib "nana" is pine) also present in the Galibi and Chaima dialects, Ananas folio vix serrato. Ananas (Glabra). Pineap as well as in the Arawak and Tupi. In the latter, Boerh. ple with smooth leaves the form "anana" already exists. In Tupi, "nana" Ananas lucide virens, fo Ananas (Lucidus). Pine is the plant and "anana" is the fruit (Alvarado lio vix serrato. apple with shining 1939). The name Nanas is first mentioned by green leaves and Thevet (1557) and Ananas by de Lery (1580, scarcely any spines on cited by Beer 1856). The Spanish "pifia" and their edges Ananas fructu pyramidata Ananas (Serotinus), Pyra the English "pineapple" came from the com olivae colore, intus au midal olive-colored parison with the exotic pinecone. The Brazilian reo. pineapple, with a yel name "abacaxi," originally designating partic low flesh ular cultivars, is derived from the Guarani word Ananas aculeatus, fructu Ananas (Viridis). The for the maize ear (Bertoni 1919). Amerindians pyramidata eax viridi green pineapple had a thorough knowledge of the plant and its jiavescente, cultivation, differentiating cultivars and wild types (Thevet 1557, Carvajal 1647, Gumilla 1741). Based on their common knowledge, Fernandez of these species is a pineapple with "scarce any de Oviedo (1535) described three cultivars from spines" on the leaf edges. In the eighth edition, La Hispaniola: 'Yayama,' 'Yayagua,' and 'Bon Miller (1768) disputed Linnaeus' inclusion of iama.' the pineapple in the genus Bromelia and reduced When Charles Plumier initiated the taxonomic his former six species to varieties of a single work on the Bromeliaceae at the end of the 17th one. century (Leal 1989), he followed the native clas In 1805, Saint-Hilaire created the Bromeli sification. He created the genus Bromelia for the aceae family, using the terminology established karatas, in honor to the Swedish physician Olaf by Plumier (1755). In 1827, Lindley renamed Bromel, and described the Ananas, using poly the genus as Ananassa and the pineapple as A. nomials such as Ananas aculeatus fructu ovato sativa. He also recognized the species A. lucida, came albida for a pineapple cultivar or Ananas from Miller's Ananas lucidus, and created the non aculeatus pitta dictus for the curagua species A. bracteata. According to Smith and (Plumier 1755). In 1747, Rumphius published Downs (1979), A. lucida corresponds to the cur his Herbarium Amboinense, where he classified agua, and A. bracteata to the form presently the pineapple in three botanical forms: mas, fem named A. bracteatus. The description of A. lu ina, and alba. He then considered femina as An cida, however, indicates that this species corre assa domestica and alba as Anassa silvestris. sponds to smooth-leafed cultivars of pineapple His view was likely biased, for he thought the (see also Beer 1856). In the same way, the spe pineapple originated in the Moluccas islands and cies A. debilis described by Lindley (1827) and described cultivated forms, probably brought by then used by Beer (1856) only corresponds to a the Portuguese when they first settled there. particular pineapple cultivar with undulated Rumphius thus compared cultivars instead of leaves. species and, in addition, observed them at dif Schultes and Schultes (1830) returned to the ferent phenological stages. In his Species Plan original name Ananas, as given by Plumier and tarum, Linnaeus (1753) redesignated Anassa do by Miller, recognizing the cultivated fruit spe mestica and Anassa silvestris as Bromelia an cies as (i) Ananas sativus, (ii) Ananas semiser anas and Bromelia comosa. Quite at the same ratus (instead of Ananassa lucida), and (iii) An time, Miller (1754) published a list of six An anas sagenaria (instead of Ananassa bracteata). anas species in the fourth edition of the Garden According to Camargo (1943) and Smith and er's Dictionary, maintaining Plumier's genus An Downs (1979), this last species also included the anas in his polynomials (TABLE 2). The fourth "yvira," first described with dubious character- 1998] LEAL ET AL.: HISTORY OF ANANAS AND PSEUDANANAS TAXONOMY 231
istics as Bromelia sagenaria in 1810, by Arruda because he supposed that the absence of a crown da Camara, and as Ananas sylvestris in 1825 by in A. macrodontes was only the result of ob Vellozo. Another incomplete and dubious de serving a juvenile inflorescence. The other va scription by Arruda da Camara is that of Bro rieties were: (ii) A. sativus var. lucidus, for melia muricata, possibly a particular form of smooth-leafed pineapples, a synonym for A. lu yvira, which was renamed Ananas muricatus by cidus Mill., A. semiserratus Schult. & Schult. f., Schultes and Schultes (1830). The first clear de A. mordilona Linden, and the curagua described scription of yvira, with an excellent drawing, by Plumier; (iii) A. sativus var. debilis ("only was published by Morren (1878) who named it from European glasshouses"); (iv) A. sativus Ananas macrodontes because of its large and var. muricatus, with doubts on its classification strong spines. within Ananas; and (v) A. sativus var. micro Two new species from Colombia, Ananas stachys for the wild pineapple. mordi/ona and Ananas pancheanus were pro In 1917, Merrill established the binomial An posed respectively by Linden (1879) and by An anas comosus based on Linnaeus' Bromelia dre (1889). Ananas mordilona, with completely comosa, considered synonymous with Bromelia smooth leaves, is strikingly similar to the present ananas. In 1919, Hassler divided the genus An day cultivars 'Perolera' and 'Manzana,' still anas in two sections: Euananas and Pseudoan widely cultivated in what was the territory of the anas. Motilones Indians (mordilona is probably a cor Another division was proposed simultaneous ruption of "Motilona"). Ananas pancheanus is ly by Bertoni (1919), who worked on his col a wild pineapple, with long leaves, minute lection of Paraguayan material, now unfortu spines, antrorse and retrorse, and a long scape nately lost. He divided the genus into five spe bearing a small cylindrical fruit. Smith and cies: Ananas microcephalus, A. bracteatus, A. Downs (1979) stated that it is possibly A. par muricatus, A. sativus, and A. guaraniticus, with guazensis. The exsiccata observed in Kew by F. many botanical varieties for each species. This Leal is too poor on which to offer an opinion. classification is not only confusing but also er Neither A. mordi/ona nor A. pancheanus have roneous. Ananas microcephalus is clearly the been validated later. Andre (1889) collected an yvira, as it lacks a crown. In addition to A. sa other wild pineapple at Panch6, which he clas tivus var. bracteatus (Lindl.) Mez, Bertoni's A. sified as A. sativus. bracteatus includes botanical varieties as diverse Until the end of the nineteenth century, no as those corresponding to Bromelia sagenaria distinct species was recognized for the wild Arruda; Ananas macrodontes E. Morren, a wild pineapples. Beer (1856) had described an exsic pineapple of Honduras described by Hume and cata of a wild type with an erect habit, a limited Miller (1904); and a Paraguayan cultivar. Ber number of thin leaves, spines, and a small fruit, toni maintained A. muricatus Arruda, although corresponding to the present description of An no one had observ\')d it after Arruda da Camara. anas ananassoides (Baker) L.B. Sm. He stated He included in A. sativus the most common that the differences with the cultivated pineapple world cultivars, botanical varieties correspond were only the result of domestication (hypertro ing to wild pineapples, already mentioned in phy of the syncarp) and found no reasons not to valid species (as A. debilis), and A. sativus var. include it in A. sativus. On the other hand, in bracteatus. Ananas guaraniticus clearly corre classifications proposed by early taxonomists for sponds to A. ananassoides (Baker) L.B. Sm. De the cultivated types, confusion persisted regard spite all these mistakes, Bertoni's work and eth ing the levels of genus, species, botanical vari nobotanical arguments on the origin of the genus eties, and cultivars; for this reason, the present had a long-lasting influence on subsequent stud review does not consider subdivisions of the cul ies. Thus, the first systematic collecting of An tivated pineapple. anas and Pseudananas germplasm by Baker and Baker (1889) was the first to give species rank Collins (1939) was based on Bertoni's hypoth to a wild pineapple, using the binomial Acan esis of a southern origin of pineapple. Baker and thostachys ananassoides; and in 1891, Lindman Collins were convinced that pineapple distribu included it in the genus Ananas, as A. micros tion and variability was concentrated under the tachys, a nomen illegitimum-as stated by Smith natural barrier constituted by the Amazonian and Downs (1979) in their list of the synonyms forest, and, impressed indeed by the diversity of A. ananassoides. found in the area, they ignored the possibility of Mez (1892) proposed a first simplification in a wide diversity in the North of the continent. Flora Brasiliensis, with Ananas a monotypic ge In 1930, Harms raised Hassler's section Pseu nus. He considered A. sagenaria together with doananas to the genus rank, with Pseudananas A. macrodontes E. Morren as a variety of A. macrodontes (E. Morren) Harms. Mez (1934) sativus, naming it (i) A. sativus var. bracteatus, did not recognize this new genus, and proposed 232 SELBYANA [Volume 19(2) a new classification into three species: (i) An distinction is disputable because of the paucity anas comosus, including the cultivated forms of Camargo's arguments. Curiously, he under (previously considered botanical varieties) sati lined in his paper that Fritz Muller himself ne vus, lucidus and debilis; (ii) A. sagenaria, cor gated taxonomic value to petal scales in the bro responding to A. bracteatus; and (iii) A. macro meliads. dontes E. Morren. Mez (1934) reiterated doubts In 1956, Camargo presented a new species, about the form debilis, as well about A. muri A;wnas lyman-smithii, to designate crownless catus, an improbable species. pineapples. This species was placed in the syn From 1934 on, the evolution of pineapple tax onymy of A. monstrosus by Smith in 1961. Then onomy has been dominated by the views of L.B. in 1962, Smith further increased the number of Smith in the context of his monumental work on species, with A. nanus (L.B. Sm.) L.B. Sm., cre Bromeliaceae. The Brazilian horticulturist Fel ated from A. ananassoides var. nanus. His ar isberto Camargo also made significant contri gument was that this last change resulted from butions to these studies. Smith (1934) first pro conservation of the original dwarf type through posed to reintegrate the wild pineapple, Ananas prolonged cultivation. Such conservation, how microstachys Lindm., into A. comosus as A. ever, is only normal if plants were propagated comosus var. microstachys. vegetatively. In 1939, Camargo created a new botanical va In 1966, Camargo (Camargo & Smith 1968) riety, Ananas sativus var. duckei, for the cura visited the area of the Parguaza River, an Ori gua. He also introduced a new combination, noco affluent, where Velez and Badillo (1946) changing Pseudananas macrodontes (E. Mor had collected wild pineapples. Similar material ren) Harms to P. sagenarius (Arruda) Camargo had been collected before and classified as An based on the partial descriptions of Arruda da anas ananassoides based on the Smith (1939) Camara (1810) and Correa (1910, cited in Ca key. On the basis of these two collections, how margo 1939). This proposal is not supported be ever, Camargo and Smith (1968) described a cause, as stated by Camargo himself, the de new species, A. parguazensis. Therefore, the scription by Arruda da Camara does not clearly herbarium collections from this region had to be mention the absence of a crown on the fruit, or reclassified and divided between A. ananasso the presence of stolons, which are specific to ides and A. parguazensis. Part of this material, Pseudananas. Indeed this description may cor showing intermediate characteristics, could not respond to either Ananas bracteatus or Pseu be attributed to either species and was left aside. dananas macrodontes and provides no reason to The ideas of Camargo have been published abandon Morren's basionym. separately by Reyes-Zumeta (1967). Apparently, Smith (1939), working in part with the ma Camargo still considered Ananas nanus as a va terial collected by Baker and Collins, revised riety of A. ananassoides. He classified Pseudan Ananas taxonomy and divided the genus into anas sagenarius into three botanical varieties: four species: A. comosus, A. ananassoides, A. thevetii (corresponding to Ananas macrodontes bracteatus, and A. erectifolius. Of these, Ananas E. Morren), bertonii (corresponding to the ananassoides, with two varieties, typicus and southern types described by Bertoni), and dar nanus, corresponds to Smith's previous A. com danensis (a form from northeastern Brazil). osus var. microstachys, and A. erectifolius cor In 1968, Reitz proposed a new species, An responds to Camargo's A. sativus var. duckei, the anas genesio-linsii, from Aguas Emendadas curagua. (Central Brazil), where it is named Ananas dos In 1943, Camargo added a new species, An Indios. The original population, which still ex anas fritvnuelleri, based on specimens collected ists, is a typical wild pineapple, with erect leaves from southeastern Brazil. The botanical features and a fruit of intermediate size borne on a long presented by Camargo were petals bearing ver peduncle, similar to many wild or partially do tical folds, identical to those of Pseudananas; mesticated clones from the Guianas to western long bracts; multiple crown; axillary suckers; Venezuela. Determined to be a triploid (Lin et antrorse or retrorse spines; and pink floral bracts al. 1987, Dujardin 1991), it appears to be a vig getting green at syncarp maturity. This species orous A. ananassoides with larger fruits. No par differs from Ananas bracteatus only by the ori ticular trait justifies the species rank. entation of some spines, the petal folds, and the In 1971, in the Flora de Venezuela, Ananas change in bract color with maturation. Indeed, erectifolius became a synonym of A. lucidus Smith (1939) previously classified it as A. brac (Smith 1971). The most recent revision was by teatus var. albus. Camargo also mentioned syn Smith and Downs (1979), who retained the eight onymy with Fritz Miiller's A. sylvestris. Camar species mentioned. Since then, Leal (1990) in go considered A. JritzmueUeri an intermediate validated A. monstrosus, a nomen nudum syno form between Pseudananas and Ananas. The nym of A. comosus, because the absence of a 1998] LEAL ET AL.: HISTORY OF ANANAS AND PSEUDANANAS TAXONOMY 233 crown is not a pennanent character. Indeed, the and heredity of traits. No differences are appar epithet monstrosus given by Smith was based on ent between the Ananas species either in floral Carriere's description of Ananassa monstrosa. structure and cytology or in chromosome num This was "a simple fonn of Ananas sativa of ber or breeding system. Fertility and self-fertility which it has all characters, except the terminal are often lower in A. comosus than in the other bud" (Carriere 1870). It is amusing that Carriere species. This, however, seems to be the result of insisted his sample could not be considered a artificial selection for reduced fertility and stron new species but just an exception to the common ger self-incompatibility in the course of domes rule for Ananas sativa. Carriere knew that "the tication. Self-incompatibility is present in all Ananas, considered as a type and provided with species (Coppens d'Eeckenbrugge et al. 1993). a crown, sometimes produces, by a kind of slow No interspecific incompatibility has been ob dimorphism, individuals without crown (democ served in the genus .4nanas, neither at the level ratized Ananas, we could say)." of poUen-pistil interaction or in embryogenesis Pineapple taxonomy is not satisfactory yet, and seed development. Interspecific crosses in and the seven remaining species reported in T A volving A. comosus are at least as fertile as in BLE 1 could yet be reduced. Loison-Cabot tercultivar crosses, and the hybrids are fertile (1992), underlining the similarity among Ananas (Collins 1960, pers. obs.). When A. comosus is ananassoides, A. nan us, and A. parguazensis crossed with Pseudananas sagenarius, a few and between A. bracteatus and A. Jritzmuelleri, fertile seeds are produced. Hybrids are tetra has proposed a reduction in the number of spe ploid, vigorous, highly fertile, and self-fertile. cies. Such a reduction would lead back to the Similarly, crossing P. sagenarius with other An first classification of Smith (1939). anas species produces a majority of tetraploids Indeed the Smith and Downs key (1979) is and some smaller and self-sterile triploids (Col not tenable. Mostly based on quantitative traits lins 1960). The isozyme study by Aradhya et al. (e.g., fruit size), it does not consider genetic and (1994) indicated that 86% of variation was strong environmental variations. The best ex found within species, underlining a moderate in ample of the key's problems is the importance terspecific divergence. Despite the geographical assigned to fruit size, which separates Ananas differentiation observed in the genus, either at comosus and A. bracteatuslA. Jritzmuelleri from the molecular level (see Leal and Coppens the other species and A. ananassoides from A. d'Eeckenbrugge 1996 for a review) or at the nan us. The few discriminate qualitative traits, morphological level (Duval et ai. 1997), it such as presence or absence of spines, only de seems that no definitive speciation has yet taken pend on one or two genes (Collins & Kerns place. If the species concept is to be narrowly 1946). Most of the intraspecific variation has applied, taxonomists will need to recognize only been neglected. The presence of some retrorse one from the present Ananas species. spines, generally at the leaf base, in certain spe cies, is not constant. Morphological traits such as petal appendages do not justify the division ACKNOWLEDGMENTS into species, as they show variation within spe cies as well as between species, for instance be This work was partly funded by the European tween cultivars of A. como sus. The same could Commission through the INCO-DC project ER be said for fragility of the fruit on the peduncle, BIC18CT960118. The authors wish to express which contributes to separating A. nanus from their gratitude to Miss Ana Luisa Triana for her A. ananassoides or for bract color at maturity collaboration, and to Eric Gouda, Harry Luther, that contributes to separating A. Jritzmuelleri Edna Sieff, and Barry Walsh for comments on from A. bracteatus. In recent works, pineapple the manuscript. specialists using the Smith and Downs key could not identify intermediate material that combines LITERATURE CITED traits specifically attributed by Smith and Downs to distinct species (e.g., Duval et al. 1997). ALVARADO L. 1939. Glosario de voces indfgenas. Ob Some hybrids between A. comosus cultivars pro ms completas, Vol I. Fundaci6n ia Cas a de Bello, duce a small fruit borne on a long and thin pe Caracas. duncle, resembling A. ananassoides more than ANDRE E. 1889. Description et histoire des Bromeli A. comosus. The genus organization requires acees recoltees dans La Colombie, L' Equateur et simplification. Le Venezuela. G. Masson, Paris, 118 pp. ARADHYA M., F ZEE, AND R.M. MANSHARDT. 1994. Isozyme variation in cultivated and wild pineap CONCLUSIONS ple. Euphytica 78: 87-99. A new classification and resulting key will ARRUDA DA C"MARA M. 1810. Disserta<;1io sobre as need to take into account reproduction biology plantas do Brasil, pp, 1-49. 234 SELBYANA [Volume 19(2)
BAKER J.G. 1889. Handbook of the Bromeliaceae. XIV Congresso Brasileiro de Fruticultura. (in London. press). BAKER K. AND J.L. COLLINS. 1939. Notes on the dis FERREIRA P.R. AND J.R.S. CABRAL. 1993. Pineapple tribution and ecology of Ananas and Pseudananas germplasm in Brazil. Acta Hort. 334: 23~26. in South America. Am. J. Bot. 26: 697~702. GUMTLLA J. 1741. E1 Orinoco ilustrado y defendido. BEER J.G. 1856 [1857]. Die Familie der Bromeliaceen. Bib1ioteca de la Academia Nacional de la Histo Vienna (Tendler & Co.). ria, Caracas. BERTONI M.S. 1919. Contributions 11 l'etude botanique HARMS H. 1930. Bromeliaceae. In: Engler and Prantl. des plantes cultivees. I. Essai d'une monographie Die natiirlichen Pflanzenfamilien. 15a: 65~159 du genre Ananas. An. Cient. Paraguay (Ser. II) 4: HASSLER E. 1919. Bromeliacearum paraguariensium 250~322. conspectus. Annuaire Conserv. J. Bot. Geneve. BROWN G.K. AND A. J. GILMARTtN. 1989. Chromosome 268~341. numbers in Bromeliaceae. Am. J. Bot. 76: 657~ HUME H.H., AND H.K. MILLER. 1904. Pineapple cul 665. ture. II. Varieties. Florida Agr. Expl. Sta. Bul. 70: CAMARGO F. 1939. Ananas e abacaxf. Rev. Agr. Pira 37~62. cicaba 14: 321~338. LEAL F. 1989. On the history, origin and taxonomy of ---. 1943. Vida e utilidade das Bromeliaceas. the pineapple. Interciencia J 4: 235~241. Brasil Min. Agr. Nac. Ens. Pesq Agr. Nac. Pesq. ---. 1990. On the validity of Ananas monstruosus. lnst. Agron. Norte, Belem, Para, Boletin Tecnico J. Bromeliad Soc. 40: 246~249. 1: 31 pp. LEAL F. AND L. AMAYA. 1991. The curagua (Ananas ---. 1956. Nota previa. Ananas lyman-smithii n. lucidus: Bromeliaceae) crop in Venezuela. Econ. sp. Arquiv. Jard. Bot. Rio de Janeiro 14: 281~289. Bot. 45: 216~224. CAMARGO F. AND L.B. SMITH. 1968. A new species of LEAL F. AND G. COPPENS D'EECKENBRUGGE. 1996. Pine Ananas from Venezuela. Phytologia 16: 464~465. apple. Pp. 565~606 in J. JANICK AND J. N. MOORE, CANPtNPIN J.H. AND G.B. ROTOR. 1937. A cytological ed., Fruit breeding. Wiley, New York. and morphogenetic study of some pineapple va LEAL F. AND E. MEDINA. 1995. Some wild pineapples rieties and their mutant and hybrid derivatives. in Venezuela. J. Bromeliad Soc. 45: 152~158. Philippines Agr. 26: 139~158. LIN B.Y, P.S. RITSCHEL, AND F. R. FERREIRA. 1987. Nu CARRIERE E.A. 1870. Ananassa monstrosa. Revue mero cromoss6mico de exemplares da familia Hort. 42: 228~229. Bromeliaceae. Rev. Bras. Frutic. 9: 49~55. CARVAJAL G. DE. 1647. Relaci6n del nuevo descubri LINDEN I.J. 1879. Ananas mordilona Linden. Belgique miento del famoso Rio Grande de las Amazonas. Hort. 29: 302~303. Fondo de Cultura Econ6mica, Mexico. LINDMAN C.A.M. 1891. Kongl. Svenska Vetenskapsak- COLLINS J.L. 1933. Morphological and cytological ad. Hand!. 24(8): 39. characteristics of triploid pineapples. Cytologia 4: LINDLEY J. 1827. Billbergia. Bot. Reg. 13: 1068. 248~256. LINNAEUS C. 1753. Species plantarum, Stockholm. ---. 1960. The pineapple, botany, utilization, cul LOISON-CABOT C. 1992. Origin, phylogeny and evo- tivation. Leonard Hill, London. lution of pineapple species. Fruits 47: 25~32. COLLINS J.L. AND K.R. KERNS. 1931. Genetic studies MARCHANT C. J. 1967. Chromosome evolution in the of the pineapple. 1. A preliminary report upon the Bromeliaceae. Kew Bull. 21: 161~168. chromosome number and meiosis in seven pine MERRILL E.D. 1917. An interpretation of Rumphius' apple varieties (Ananas sativus Lind!.) and in 8ro Herbarium Amboinense. Bureau of Science, Ma melia pinguin L. J. Hered. 22: 139~142. nila. ---. 1946. Inheritance of three leaf types in the MEZ C. 1892. Bromeliaceae; Ananas. In: Martius, FI. pineapple. J. Hered. 37: 123~128. Bras. 3: 172~634. COPPENS D'EECKENBRUGGE G., M.-F. DUVAL AND F. ---. 1934. Das Pflanzenreich. Bromeliaceae. En- V AN MIEGROET. 1993. Fertility and self-incompat gler & Prantl. ibility in the genus Ananas. Acta Hort. 334: 45~ MILLER P. 1754. Gard. Dict., 4th ed. London. 51. ---. 1768. Gard. Dict., 8th ed. London. COPPENS D'EECKENBRUGGE G., F. LEAL AND M.-F. Du MORREN E. 1878. Description de I'Ananas macrodon VAt.. 1997. Germplasm resources of pineapple. tes. sp. nov. Ananas it fortes epines. Belgique Hort. Rev. 21: 133~175. Hort. 28: 140~144. CORREA P. 1952. Dicionario das plantas uteis do Brasil, PLUMIER C. 1755. Plantarum Americanarum. Amste Vol. III. Imprenta Naciona!. Rio de Janeiro. laedami. DUJARDIN M. 1991. Cytogenetique de l'ananas. Fruits REITZ P.R. 1968. A new species of pineapple from cen 46: 376~379. tral Brazil. Bromeliad Soc. Bull. 18(5): 109~1l1. DUVAL M.-F., G. COPPENS D'EECKENBRUGGE, F.R. FER REYES-ZUMETA H. 1967. Breve nota taxon6mica sobre REIRA, J.R.S. CABRAL, AND L. DE B. BlANCHETT!. piiias cultivadas Ananas comosus (L.) Merr. con 1997. First results from joint EMBRAPA-CIRAD menci6n de dos nuevas variedades silvestres. Rev. Ananas germplasm collecting in Brazil and Fac. Agron. L.U.Z. (Maracaibo) 39: 131~142. French Guyana. Acta Hort. 425: 137~144. RUMPHIUS G.E. 1747. Anassa, Nanas. Herbarium Am FERNANDEZ DE OVIEDO G. 1535. Historia general y nat boinense. 5: 227~231. ural de las Indias. Atlas, Madrid. SAINT-HILAIRE J. 1805. Expos Family 1. 122~125, t. 19. FERREIRA F.R. 1996. Expedi<;:ao para coleta de germo SCHULTES J.A. AND J.H. SCHULTES. 1830. Ananas sa plasma de abacaxi no Sui do Estado da Bahia. tivus. Syst. Veg. 7: 1283~1285. 1998] LEAL ET AL.: HISTORY OF ANANAS AND PSEUDANANAS TAXONOMY 235
SHARMA A.K. AND 1. GHOSH. 1971. Cytotaxonomy of SMITH L.B. AND R.J. DOWNS. 1979. Bromelioideae the family Bromeliaceae. Cytologia 36: 237-247. (Bromeliaceae). Fl. Neotr. Monogr. 14(3): 1493- SMITH L.B. 1934. Studies in the Bromeliaceae-V 2142. Contr. Gray Herb. 104: 71-83. THEVET A. 1557. Les singuJarites de Ja France Antarc ---. 1939. Notes on the taxonomy of Ananas and tique. Maspero, Paris. Pseudananas. Bot. Mus. Leafi. 7(5): 73-81. VELEZ 1. AND VM. BADILLO. 1946. Comentarios sobre ---. 1961. Notes on Bromeliaceae. Phytologia 8: 12. los pifiales silvestres y cultivados de la region del ---. 1962. A new look at the species of pineapple. Parguaza, Guayana Venezolana. Rev. Agric. Bromeliad Soc. Bull. 12(3): 54-55. Puerto Rico 37: 127-130. ---. 1971. Bromeliaceae. Flora de Venezuela. Min. VELLOZO J.M. C. 1825. Florae fiurninensis. Arch. Mus. Agric. & Cria, Caracas. 12(1): 1-361. Nac. 5: 122.