Taxonomic History Of The Lecythidaceae

The first known description and illustration of a member of the Lecythidaceae were made by Frei Christovao, a Portuguese Franciscan missionary, who lived in Maranhao, Brazil from 1624 to 1635. His Historia dos animais e arvores do Maranhao, not published until 1967 and then in manuscript facsimile accompanied by a modern translation, includes a description and illustration of Sapuqaiha, which without a doubt refers to Lecythis pisonis, as well as of Sequeriba, which may be the first reference and illustration of the brazil nut, Bertholletia excelsa. Frei Christovao's work is poorly known to systematists outside of Brazil but recently has been discussed in detail by Whitehead (1973).

The Dutch capture of the northeastern coast of Brazil prematurely ended the studies of Christovao and made possible those of the German George Marc-grave and the Dutch Willem Piso. Marcgrave (who lived 1610-1644) figured and described Japarandiba (=Gustavia), Jacapucaya (=Lecythis), and Ibiraba (=Lecythis) in his Historiae rerum naturalium brasiliae, a work edited by Johannes de Laet who had to decipher the secret code in which it was originally written by Marcgrave to prevent it from being pirated by Piso. It was published in 1648, four years after Marcgrave's death (Marcgrave, 1648). Another, somewhat altered edition was prepared by Piso (1658) under the title of De Jndiae utriusque re naturali et medica (generally referred to today incorrectly as authored by Marcgrave and Piso). Piso and Marcgrave were contemporaries in Brazil but apparently the relationship between America's first naturalists, like so many that followed, was strained (Whitehead, 1973). Gudger (1912) claims that Piso plagarized much of Marcgrave's work. Whitehead (1973) provides a detailed account of the Brazilian sojourn, publications, and drawings of Piso and of Marcgrave.

Another pre-Linnean reference to the family is that of Barrere (1741) who referred to Couroupita guianensis under the name Pekea, which he took from the earlier works of Marcgrave (1648) and Piso (1658), but it is extremely doubtful that Pekea of Marcgrave and Piso is the same as either Couroupita or Barr^re's Pekea.

In 1758, Linnaeus posthumously published Pehr Loefling's description of a new genus and species, Lecythis ollaria. Perhaps Linnaeus' most promising student, Loefling had died two years earlier in Venezuela at the age of 27. Lecythis was placed in the Linnean order Polyandria Monogynia. A type specimen for L. ollaria has never been found. However, Prance and Mori (1977) have matched modern collections with the original description and have selected a neotype. One year later, Linnaeus (1759) described Grias in the tenth edition of his Systema naturae, also placing it in his order Polyandria Monogynia, and basing it on Grias cauliflora from Jamaica.

Adanson (1763), in his Familles de plantes, treated Lecythis under the name Bergena in the Pistaciae and used Marcgrave's name Japarandiba for the genus now known as Gustavia. Adanson placed Japarandiba and Grias together in the Cisti. Gustavia was described by Linnaeus (1775) in his Plantae surinamenses where he proposed G. augusta, based on the collections of Allamand and Solander made between 1750 and 1770 in Dutch Guiana (Surinam). Gustavia was named in honor of Gustave III, king of Sweden, who presented Linnaeus with the collections of Allamand and Solander (Smith, 1812).

Also in 1775, three new genera, Couratari, Couroupita and Pirigara were proposed by Aublet in his Histoire des plantes de la Guiane Française. The genera Couratari and Couroupita have been maintained up to the present, but Pirigara was later placed in synonymy under Gustavia (Sprengel, 1825, Berg, 1856).

The large Asiatic genus Barringtonia was described by George Forster (1776). Foetidia was described by Commerson from Mauritius in 1788.

Following his custom of changing 'barbarous names,' Scopoli (1777) illegitimately substituted the name Pontopidana for Couroupita and Teich-meyeria for Gustavia, but neither was ever adopted by later authors.

It was Antoine Laurent de Jussieu (1789), in his pioneering Genera plan-tarum which forms a basis for all later systems of classification, who first brought most of the above genera together in his all-encompassing order ( = the Myrti). This order was quite a heterogeneous assemblage, containing not only genera now referred to as Myrtaceae and Lecythidaceae, but the genus Philadelphus (Hydrangeaceae) and other unrelated elements. Nevertheless, Jussieu did bring together most genera that are now considered part of the Lecythidaceae. He placed Barringtonia (under the name Butonica Rumphius), Couroupita, Gustavia (under the name Pirigara), and Lecythis (including Couratari as a synonym) in his second section of the Myrti. Foetidia was located in the first section of the same family, but Grias was aligned among the Guttiferae. Jussieu, apparently, had taken the generic characters mainly from Linnaeus' descriptions.

Necker (1790) added to the generic synonymy of the family by creating the illegitimate names Elsholtzia for Couroupita, and Spallanzia for Gustavia.

Gmelin (1791, 1792), in his edition of Linnaeus' Systema naturae, used the Linnaean sexual system of classification. He placed Lecythis (with 8 species) and Grias (with 1 species) in the Polyandria Monogynia and Barringtonia, Gustavia, Curupita (an orthographic variant of Couroupita), and Curatari (an orthographic variant of Couratari) in the Monadelphia Polyandria. Willdenow (1799), in his edition of Linnaeus' Species Plantarum, treated under Lecythis a mixture of 8 species now divided between Lecythis and Eschweilera.

In a generally overlooked publication, Palisot de Beauvois (1804)

described, in honor of the Emperor of" France, the African genus Napoleonaea. This genus was again discussed by Palisot de Beauvois (1810) under the orthographic variant Napoleona, a spelling adopted in most later works, until Liben (1971a) (see also Heine, 1967) discussed the original publication of Napoleonaea in his monograph of the genus.

Humboldt and Bonpland (1807) described the genus Bertholletia to accommodate the single species B. excelsa, the Brazil nut. As they had seen only leaves and fruit but no flowers, they placed it, with some doubt, in the Lin-naean order Monodelphia Polyandria, and did not suggest a position for it in the Jussieuan system as they did in the same work for other genera based on complete material.

Roxburgh (1814) published Careya from Malesia, with its single species, C. arborea.

Desfontaines (1820) proposed the new genus Asteranthos from Brazil which he related to Napoleonaea from Africa. He did a remarkable job of reconstructing the morphology of Asteranthos since he had only fragmentary material. One year later, Schrank & Schrank (1821) described another new genus from Brazil, Lecythopsis, which has since been regarded as synonymous with Couratari. Robert Brown (1821) proposed the new family Belvisieae to accommodate the genera Asteranthos and Belvisia Desv. ( = Napoleonaea).

In his edition of Linnaeus' Systema vegetabilium, Sprengel (1825) was the last major worker to classify the group by the Linnaean sexual system. Asteranthos, Grias, and Lecythis (including Couroupita) were placed in Polyandria Monogynia, Gustavia in Monadelphia Polyandria, with Pirigara cited in synonymy, and Couratari was left unmentioned. Also in 1825, Humboldt, Bonpland and Kunth treated three genera—Lecythis, Pirigara (with Gustavia cited in synonymy) and Bertholletia—as members of the family Myr-taceae, tribe Lecythideae. Poiteau (1825) made an important contribution to knowledge of the neotropical members of the family known up to that date. Referring to the second section of Jussieu's Myrtaceae, he said:

Ayant eu l'occasion d'examiner la plupart de ces plantes, pendant mon séjour à la Guyane français, j'ai reconnu qu'en effet elles n'ont pas les caractères des Myrtes, et qu'elles en possèdent d'autres qui leur sont particuliers. Je propose donc d'en former une nouvelle famille sous le nom de Lecythidées.

Because of his statement that the Lecythidaceae merit the rank of a family separate from the Myrtaceae the family name Lecythidaceae should be attributed to Poiteau rather than to Lindley. Poiteau included in the family the genera Lecythis, Couroupita, Bertholletia, Gustavia and Couratari, and made no reference to the extra-American genera.

De Candolle (1828) did not follow Poiteau, keeping the Lecythidaceae within the Myrtaceae and dividing the latter into 5 tribes. His fourth tribe, Barringtonieae, contained Barringtonia, Stravadium Juss. (now considered as part of Barringtonia) and Gustavia; and his fifth tribe, Lecythideae, held Lecythis, Eschweilera Mart, ex DC, Bertholletia, Couroupita and Couratari. Eschweilera was the only new genus proposed by de Candolle, and it contained 2 species. Lecythis included 19 species many of which were transferred to Eschweilera by later authors. Grias, Careya and Foetidia were placed at the end of the Myrtaceae in an assortment of genera labelled "Myrtaceae dubiae." De Candolle did not treat Napoleonaea and Asteranthos until a much later volume of his Prodromus (1839), where he placed the two genera together in the Napoleoneae between Collumeliaceae and Vaccinieae, two quite unrelated groups. The classification of de Candolle was followed with slight modifications by Reichenbach (1828), Dumortier (1829), Bartling (1830), Meisner (1839), Endlicher (1839), and Baillon (1877). Reichenbach, Meisner and Endlicher treated Asteranthos and Napoleonaea as a separate family.

Lindley (1830) kept the Barringtoniae in the Myrtaceae, noting that they probably did not belong to the order, and placed the Lecythidaceae near the Ternstroemiaceae apart from the Myrtaceae. However, in 1836, he again placed the Lecythidaceae near the Myrtaceae. In the 1836 edition of his system, the Barringtoniae contained Barringtonia, Stravadium, (= Barringtonia) Gustavia, Catinga Aubl ( = Eugenia of the Myrtaceae), Coupoui Aubl. (=Duroia of the Rubiaceae), Careya and Foetidia. The Lecythidaceae contained Lecythis, Eschweilera, Bertholletia, Couroupita, Couratari, and ? Touroulia ( = Quiinaceae). Napoleonaea and Asteranthos were retained in Robert Brown's Belvisiaceae, placed between Styracaceae and Sapotaceae in 1830 and in his alliance Campanales in 1836. However, in 1846, Lindley was the first worker to relate the Belvisiaceae to Myrtaceae and Lecythidaceae, placing them between the Rhizophoraceae and Melastomataceae in his alliance Myrtales. Lindley (1846) decided that the Lecythidaceae, in which he included Couratari, Cariniana, Lecythis, Eschweilera, Bertholletia, Couroupita and Crossostylis Forst. (Rhizophoraceae), were a family separate from Myrtaceae but retained them in the myrtle alliance. He supported the separation of the Lecythidaceae at the family level because of "the great almond-like seeds and alternate, often serrated, non-punctate leaves." Furthermore he removed the Barringtoniaceae (Barringtonia, Stravadium, Careya, Gustavia and Foetidia) from the myrtle alliance to the Grossales. Although he admitted the similarity of the Barringtoniads to the Lecythids, he concluded that the presence of stipules and the hooded plate of sterile or additional stamens in the latter warranted their separation. Stipules have since been shown to be present in both groups (Pit-tier, 1927; Weberling, 1957; Payens, 1967).

Don (1832) retained Barringtonieae as a tribe of the Myrtaceae and recognized the Lecythidaceae as a separate family. In Brogniart's 1850 enumeration of the plants in the Museum of Natural History in Paris the Lecythidaceae are also recognized as a separate family.

Casaretto (1842) described the new genus Cariniana from Brazil with a single species C. brasiliensis (=C. legalis (Mart.) Kuntze). One year later (1843) he published a second species C. excelsa ( = C. estrellensis (Raddi) Kuntze). Blume (1852) published Planchonia based on P. sumatrana.

Otto Carl Berg (1858), after a series of smaller publications (1856, 1857), presented a monograph of the Brazilian members of Lecythidaceae in Martius' Flora Brasiliensis. Berg, a specialist in the Myrtaceae, described over 1000 species of myrtles (McVaugh, 1958). He accepted de Candolle's classification without change and likewise did not include Asteranthos in the Myrtaceae. By that time a large number of Brazilian species were known: Gustavia with 5,

Couroupita with 2, Bertholletia with 1, Lecythis (including Eschweilera) with 37, Lecythopsis with 2 (one a Couratari, the other an Eschweilera) and Couratari with 12 (including some species now placed in Allantoma and Cariniana).

Bentham (1859) followed Lindley (1846) by suggesting that Asteranthos was related to the Myrtaceae (including Lecythidaceae).

Bentham and Hooker (1865) combined the Barringtonieae and Lecythideae into one tribe, the Lecythideae, which in turn they subdivided into three subtribes: the Barringtonieae (Barringtonia, Petersia Welwitsch ex Bentham & Hooker, Careya, Planchonia, Gustavia, and Grias), the Eulecythideae (Couratari, Couroupita, Lecythopsis (= Couratari), Lecythis and Bertholletia), and the Napoleoneae3 (Napoleonaea and Asteranthos). Foetidia was treated by them as an anomalous genus. The only new genus proposed by Bentham and Hooker was Petersia based on a Welwitsch manuscript name. The genus has generally been accepted but the name is illegitimate because of Petersia Klotsch (1861) of the Capparidaceae. Petersia Welw. ex Benth. & Hook, was renamed Petersianthus by Merrill (1916).

Le Maout and Decaisne (1873) followed essentially the same classification as Bentham and Hooker, but elevated two of the subtribes of the Myrtaceae to full tribes: the Barringtonieae and Lecythideae and raised the rank of the Napoleoneae to a family close to the Myrtaceae.

The first monographer of the Lecythidaceae as a separate family was John Miers (1874, 1875a, 1875b). He recognized three families, the Lecythidaceae, with Gustavia (21 species), Couroupita (9), Bertholletia (2), Lecythis (42), Chytroma Miers (25), Eschweilera (46), Jugastrum Miers (6), Couratari (8), Cariniana (7), Allantoma Miers (12), Grias (4), and Cercophora Miers (1) (= Strailia Th. Dur.); the Belvisiaceae, with a single genus Napoleonaea, and the Barringtoniaceae, with Barringtonia, Agasta Miers, Butonica Rumph., Stravadium, Planchonia, Careya, Doxomma Miers, Petersia, Megadendron Miers and Chydenanthus Miers. He also concluded that Asteranthos, which he mistakenly thought had a superior ovary, was related to the Rhododendreae (Ericaceae). As can be seen from the summary given above, Miers described eight new genera. He also described many new species, many of which have been placed in synonymy in the present work. The following comment on Miers* work by Steam (1971), in a paper on the Jamaican Boraginaceae, is useful in understanding Miers' taxonomic philosophy:

In 1869 John Miers, at the age of 79, published a survey of Bourreria which he divided into two genera, i.e. Bourreria proper and Crematomia, typified by B. exsucca. This appeared in Ann. Mag. Nat. Hist. IV. 3: 199-210, 300-313, (Mar.-Apr.) 1869, and was reprinted in his Contributions to Botany 2: 230-242. He was an acute and accurate observer, who retained into old age an intense curiosity about morphological details which he investigated with assiduity, and as a retired chemist and engineer, he brought to his botanical studies the precision of his former calling, together with a somewhat mechanical aproach. As William Carruthers remarked in his obituary of Miers (Jour.

The correct orthography for this subtribe is Napoleonaeae. Throughout this chapter the spelling used in each work cited is given.

Bot. (London) 18: 36. 1880), 'he had a very quick sense of differences but he sometimes failed to distinguish the real value of the differences he saw.' The editor, M.T. Masters, of the Gardeners Chronicle (cf. 1879. 2: 522 (Oct.) 1879) said much the same: 'as a botanist his tendency was to minute elaboration rather than judicial estimate of the relative importance of details. In practice, therefore, he multiplied species, and even genera and orders, to an extent opposed to the prevailing tendencies of his contemporaries. Miers never adopted the doctrine of the mutability of the species'; he was 70 when Darwin published The Origin of the Species. Seemingly he took as representing well-marked species any herbarium specimen or little group of herbarium specimens which differed in appearance from other specimens. Thus, on the evidence of sixteen specimens in the British Museum and Kew Herbaria, he recorded nine taxa from Jamaica under the names Bourreriasucculenla, B. ovata, B. rigida, . . . these being described accurately but defined vaguely. O. E. Schulz in 1911 reduced them all to three. . . .

The Lecythidaceae lent themselves well to the mechanical approach of Miers who placed great emphasis on fruit characters, especially minute differences in pyxidium shape and form. He described 14 species of Lecythis from material consisting of pyxidia alone, 7 in Chytroma, 2 in Jugastrum, one each in Couratari and Cariniana, and 5 in Allantoma, making a total of 28 species! He also recognized an additional six species already described by Berg from pyxidia alone. Along with scientific specimens, Miers used as taxonomic material a large number of pyxidia brought back as curiosities by travellers. The classic example of this is the type of Lecythis ampullaria Miers, which is a pyxidium carved and scraped to form a utensil. Field studies have shown that there is much variation in the pyxidia from individual trees (Dugand, 1947; Mori, 1970 and Allantoma in the present work), and thus, while Miers gave an accurate description of what he saw, his species are not based on the genetic species concept of modern systematists, Miers' monograph, however, provided a thoroughly accurate, descriptive and well-illustrated account of all the material available at that time. The illustrations are especially helpful for the placing and matching of his species.

Masters (1869) gave a detailed account of the flower of Napoleonaea and provided additional evidence of its relationship to Asteranthos, but he did not try to suggest relationships within the family.

Sagot (1885) reviewed the Lecythidaceae of French Guiana, using the de Candolle system of tribes, i e Barringtonieae and Lecythideae in the Myr-taceae, and added a few new species.

Asteranthos was not treated in Martius' Flora Brasiliensis until thirty years after Berg's account of the Lecythidaceae. In Eichler's (1889) account for Flora Brasiliensis, Asteranthos was placed in the Napoleonaeaceae, which he hesitantly related to the Cactaceae or Mesembryanthemum.

Costantin and Dufour (1885) recognized anatomical differences between the Myrtaceae and Lecythidaceae, the Myrtaceae having secretory cavities and internal phloem which are absent in the Lecythidaceae. Furthermore, the latter family has cortical bundles which are not found in the former. These authors also concluded that there is no anatomical evidence to warrant the separation of the Napoleonaeaceae and the Barringtoniaceae from the Lecythidaceae and that Foetidia belongs in the Lecythidaceae. On the basis of a very detailed anatomical study of Gustavia, Barringtonia and Napoleonaea, Lignier (1890) divided the family into the following three tribes: Barringtoniees, Lecythidees, and Napoleonees. He was able to separate those tribes in terms of stem and petiole anatomy, and reinforced the separation of the Lecythidaceae from the Myrtaceae on anatomical grounds.

Niedenzu (1892), in Engler's Natürlichen Pflanzenfamilien, divided the family into four subfamilies. In the Foetidioideae he placed Foetidia; in the Planchonioideae, Planchonia, Petersia, Careya, Barringtonia, and Chydenan-thus; in the Napoleonaeoideae, Asteranthos and Napoleonaea and in the Lecythidoideae, Japarandiba (—Gustavia), Grias, Couroupita, Lecythis, Eschweilera (including Chytroma and Jugastrum), Bertholletia, Cariniana, Cercophora, Couratari, and Allantoma.

During the first part of the present century several new genera were added to the family. Huber (1902) proposed Goeldinia from Brazil, which later proved to be a synonym of Allantoma, and Chevalier (1909) proposed Com-bretodendron from Africa. However, Liben (1968) has pointed out that Chevalier (1909) later concluded that Combretodendron was a synonym of Petersia and added this to his publication of the protologue (appended as a note on page 301). According to the Code of Nomenclature (Article 34) a name is not validly published if it is not accepted by its author in the original publication. This invalidated Combretodendron, making Merrill's (1916) Petersianthus the correct name for the genus. As noted earlier Petersia was already pre-empted for a genus of Capparidaceae. E. G. Baker (1913) described Crateranthus from Africa, and Ducke (1925) described Holopyxidium from Brazil. Five years later Ducke (1930), after further field observations on the dehiscence of the fruit, decided that Holopyxidium was not a good genus and accordingly placed it in synonymy under Eschweilera.

The present century has also seen several regional accounts of the Lecythidaceae in various Floras. The most notable in the New World are those of Pittier (1927) for Central America, Eyma (1932, 1934) for Surinam, Benoist (1933) and Lemee (1953) for French Guiana, Macbride (1941) for Peru, and Woodson (1958) for Panama. Perhaps the most significant of these is the work of Eyma who made a critical study of the Surinam species and cleared up many taxonomic and nomenclatural problems pertaining to extra-Surinam species, especially in the circumscription and differential characters of Allantoma and Couratari which had previously been confused. The other regional works cited above were mainly compilations from the literature and did not present much new critical taxonomic work, although most included descriptions of a few new species.

Thompson's (1927) classical study of the floral morphology of the Lecythidaceae entitled "A study in advancing gigantism" led him to divide what he called the Lecythideae, a tribe of Myrtaceous affinity, into four series; the Barringtonieae (Barringtonia, Careya, Planchonia, Gustavia, and Grias); the Napoleoneae (Asteranthos, Crateranthus, and Napoleonaea); the Couratarieae (Couratari and Lecythopsis); and the Bertholletieae (Couroupita, Lecythis, and Bertholletia). He gives a detailed analysis of the flowers of each tribe and genus. The author also concluded here, and in his earlier work on Couroupita (Thompson, 1921), that progressive staminal sterilization within the family is accompanied by cell gigantism, which in turn produces floral zygmorphy.

An anatomical study of the wood of Old and New World Lecythidaceae by Diehl (1935) supports the separation of the Lecythidaceae from the Myr-taceae, stressing the homogeneity of the former, with banded metatracheal parenchyma as its only distinctive feature. He also noted that the New World genera possess crystal strands, which are lacking in all but Foetidia of the Old World, and supported the inclusion of Asteranthos and Napoleonaea in the Lecythidaceae.

Despite his prestigious publications in Engler's Das Pflanzenreich Rheinhard Knuth (1939a, b, c), a botanist with the Berlin Botanic Garden who compiled an "Initia Florae Venezuelensis" in 1928, added little to the comprehension of the family. Knuth recognized the Barringtoniaceae, the Lecythidaceae (Gustavia, Grias, Cariniana, Allantoma, Couroupita, Corythophora Knuth, Lecythis, Holopyxidium, Sapucaya Knuth, Chytroma, Eschweilera, Jugastrum, Bertholletia, Cercophora and Couratari), and the Asteranthaceae (Asteranthos) as separate families. He made no further subdivisions of the Lecythidaceae and Asteranthaceae but divided the Barringtoniaceae into five tribes: the Barringtonieae (Barringtonia, Careya, Planchonia, and Chydenanthus); the Combretodendreae (Combretodendron), the Foetidieae (Foetidia); the Craterantheae (Crateranthus); and the Napoleoneae (Napoleonaea). As in other families monographed by Knuth, his species separation is not based on the biology of the group. Like Miers, he was concerned with morphological minutiae rather than a biological and evolutionary comprehension of the taxa. The keys do not work, and species are described from inadequate material. More species known only from pyxidia were added by Knuth, and most of Miers' similarly described species were uncritically accepted. With so many species recognized from inadequate material, the monograph is not very useful for the identification of specimens but does provide a good replication of the species descriptions of Lecythidaceae known to that time. All subsequent monographic work in Lecythidaceae points to the need for drastic reduction in the number of species recognized by Knuth, e g Payens (1967), Liben (1971a) and the present work. For example, Payens' (1967) monograph of Barringtonia reduced Knuth's 109 species to 39, and Liben's (1971a) monograph of Napoleonaea included 8 species as compared to the 18 recognized by Knuth (1939a). It is rather unfortunate that both major monographers of the Lecythidaceae, Miers and Knuth, had the same descriptive approach characteristic of "splitters," thus producing a classification independent of the group's biology.

Recently two further neotropical genera which cannot be maintained were added to the family by Ledoux. Neohuberia, described in 1963, is based on a species of Eschweilera; and Pachylecythis, proposed in 1964 from a pyxidium only, is a Lecythis.

Pichon (1945) discussed the position of the genus Combretodendron (=Petersianthus) and, at the same time, proposed a new classification of

Subfamily I:


Tribus I,:

Planchonieae (Planchonia)

Tribus I2:

Barringtonieae (Careya, Barringtonia, Chydenanthus)

Tribus I3:

Combretodendreae (Combretodendron)

Tribus I4:

Foetidieae (Foetidia)

Subfamily II:


Tribus II,:

Griadeae (Gustavia, Grias)

Tribus II2:

Couroupiteae (Couroupita, Corythophora)

Tribus II3:

Lecythideae (Lecythis)

Tribus II4:

Couratarieae (Cariniana, Couratari, Cercophora)

Tribus II5:

Bertholletieae (Allantoma, Sapucaya, Chytroma, Eschweil

Was this article helpful?

0 0

Post a comment