An anatomical survey of the leaflet margins of the genus Attalea (Arecaceae: Arecoideae: Cocoseae: Attaleinae)Abstract
An anatomical survey of the leaflet margins of 24 species of Attalea revealed insights into the taxonomy of the genus. Examination of cross-sections of Attalea leaflets revealed that the proximal and distal margins are not identical. Proximal margins nearly always contain a region of expansion cells, which are usually absent on the distal margin. The abaxial side of proximal margins is nearly always protruded, angling the margin upwards. Conversely, the adaxial side of the distal margin is protruded, angling the margin downwards. Thus we have upward facing proximal margins meeting downward facing distal ones. The leaflet margins were used to collaborate or question the accepted species of a recent taxonomic revision of Attalea, which synonymized several species. The anatomy of the leaflet margins lent support to uniting some species, but remained inconclusive in uniting others. Finally, although Attalea palms formerly belonged to five genera based on differences in floral morphology, evidence from the leaflet margins supports the notion that all Attalea palms belong to a single genus. Leaflet anatomy also revealed intermediate characters between hybrids and their parental species.
Key words: anatomy; distal margin; expansion tissues; proximal margin; species complexes
Resumo
O estudo da anatomia da margem dos folíolos de 24 espécies de Attalea revelou-se importante para a taxonomia do gênero. Transversalmente, a margens proximal e distal dos foliolos de Attalea não são idénticas. As margens proximais quase sempre contêm uma região de células de expansão, que geralmente estão ausentes na margem distal. A face abaxial das margens proximais é quase sempre saliente, inclinando a margem para cima. Por outro lado, o lado adaxial da margem distal é projetado, inclinando a margem para baixo. Assim, temos as margens proximais voltadas para cima encontrando as distais voltadas para baixo. A anatomia da margem dos folíolos foram usadas para colaborar ou questionar as espécies aceitas de uma recente revisão taxonômica de Attalea, que sinonimizou várias espécies. As margens dos folíolos deram suporte à união de algumas espécies, mas permaneceram inconclusivas na união de outras. Finalmente, embora as palmeiras Attalea pertencessem anteriormente a cinco gêneros com base em diferenças na morfologia floral, a anatomia da margem dos folíolos suporta a unidade do gênero Attalea. A anatomia dos folíolos também revelou caracteres intermediários entre os híbridos e suas espécies parentais.
Palavras-chave: anatomia; margem distal; tecidos de expansão; margem proximal; complexos de espécies
Introduction
The palm family, Arecaceae, is divided into five subfamilies: Calamoideae, Nypoideae, Coryphoideae, Ceroxyloideae and Arecoideae (Asmussen et al. 2006; Dransfield et al. 2008). The most diverse subfamily is the Arecoideae containing 14 tribes (Baker & Dransfield 2016), one of which is Cocoseae. Cocoseae is further divided into three subtribes: Attaleinae, Bactridinae and Elaeidinae. The non-spiny Attaleinae contains ten genera (Baker & Dransfield 2016) and is named for the genus Attalea, the American oil palm. Attalea taxonomy has undergone major changes in the last 25 years. In the late 90’s, Glassman (1999) published his revision in which he recognized five separate genera: Attalea, Orbignya, Scheelea, Maximiliana and his newly described Ynesa, based on the morphology of their male or staminate flowers. Attalea flowers had flat petals and straight anthers. Orbignya had curvy strap-shaped or spoon-shaped petals with coiled anthers. Scheelea had pine-needle shaped (terete) petals with short, straight anthers. Maximiliana was similar to Scheelea, but the stamens were much longer than the short terete petals. Finally, Glassman (1999) believed Ynesa to be of hybrid origin with Orbignya-like petals and straight or slightly twisted anthers. The overall habit of these five genera are remarkably similar to each other, often with long, strongly ascending pinnate leaves, suggesting that all are closely related. This similarity motivated Wessels Boer (1965, 1988) to transfer several species of the other genera directly into Attalea, including A. butyracea (Mutis ex L.f) Wess.Boer and A. osmantha (Barb.Rodr.) Wess.Boer. Henderson followed Wessels Boer (1988) and continued this rational in his field guide (Henderson et al. 1995), joining the five genera under a single umbrella, Attalea.Zona (2002) made all the name changes necessary to transfer every accepted species to the single genus. Pintaud (2008) later clarified some of the confusion concerning Attalea species by identifying which species names were uncontroversial, like A. maripa (Aubl.) Mart., A. colenda (O.F.Cook) Balslev & A.J.Hend., A. crassispatha (Mart.) Burret and A. insignis (Mart.) Drude, and organizing others into several species complexes, like A. phalerata Mart. ex Spreng., A. butyracea, A. cohune Mart., A. speciosa Mart., and A. oleifera Barb.Rodr. Henderson (2020) published the most current revision for the genus, where he reduced the number of accepted species from 65 (Glassman 1999) to 30 species (Henderson 2020) by reducing many of Pintaud’s species complexes to a single species.
In this paper, information gathered from leaflet sectioning (Fig. 1) was used to explore taxonomic questions surrounding Attalea. In Attalea, each pinnate leaf is composed of a series of leaf segments, leaflets or pinnae. The current study examines the leaflet proximal margin, closest to the base of the leaf (the petiole side) and the distal margin, the margin most distant from the base of the leaf (the apical or leaf tip side). Marginal anatomy helped answer several questions: (1) Do differences or consistencies in the leaflet margins support the separation of Attalea palms into four or five genera as suggested by Glassman (1999) or do they point to a single cohesive genus? (2) Does the leaflet anatomy corroborate the new accepted species that Henderson (2020) has proposed? And, (3) Can leaflet marginal anatomy be used to detect or to confirm suspected hybrids and their parents?
Leaf, leaflet and leaflet anatomy of the proximal and distal margins of an Attalea phalerata. A = a trapped air bubble, an artifact of making slides; C = cuticle; E = vascular bundle with an enlarged sheath; EP = epidermis; ET = expansion tissue; FS = fibrous strand; HY = hypodermis; P = primary vascular bundle; PL = palisade parenchyma; S = secondary vascular bundle; SL = spongy parenchyma; T = tertiary vascular bundle. Scale bars = 0.5 mm.
Materials and Methods
Plants examined
Mostly fresh and some rehydrated leaf material from 24 Attalea species was used to prepare slides of both proximal and distal leaflet margins for each species. The Attalea specimens for this study came from the living collections of the Montgomery Botanical Center (MBC), Miami, FL. Most were grown from seed collected from the wild. Dried material was obtained from the author’s dried collections that were preserved at the moment that he collected in-situ voucher specimens now deposited at F, FTG, CEPEC, ALCB, HUEFS, etc. Eighty-five accessions and 110 plants were sampled from 24 species (Tab. 1).
Anatomical preparation
Methods for hand sectioning are covered in Noblick (2013, 2017). For this study a hand microtome, a straight razor and a carrot cube were used for hand sectioning and the razor was maintained sharpened with a Diamond Stone. The hand microtome was purchased from homesciencetools.com, Billings, MT, U.S.A. A Dovo Straight Razor 3” Full Hollow Ground Carbon Steel Blade was used for sectioning. The Dia-sharp 3 micron 8000 mesh Diamond Stone maintained a razor-sharp edge on the straight razor (DMT D8EE 8” Extra Extra Fine Diamond Stone) (DMTsharp.com or Diamond Machining Technology, Marlborough, MA, U.S.A).
Preparing the cross section (cutting perpendicular across the veins)
A piece of carrot was cut into a small cube that fit into the hand microtome as described in Noblick (2017). A deep perpendicular vertical slit was cut in the top of the carrot cube with a double-sided razor blade. The slit in the carrot cube was used to secure the leaflet vertically. If the leaflet was stiff and coriaceous, it was mounted with a few millimeters showing above the top of the carrot cube. The stiffer, thicker Attalea leaflets mounted in this way allow one to cut thinner sections more quickly with the straight razor without having to sort through carrot debris. If the leaflet was too thin, then it was mounted within the carrot slit for firmer support while sectioning. The carrot cube was secured in the hand microtome and adjusted down until it was just below the microtome plate. The razor and specimen were lubricated with water and the straight razor was slid across the microtome in a slicing movement while pressing the side of the blade firmly against the microtome plate. The microtome was adjusted upwards by about a quarter of a turn after each resulting section. The sections were rinsed off the blade by dipping it into a watch glass of water or with a small brush. Occasionally, the blade was re-sharpened using the Diamond Stone and water.
Preparing the slide
The glass slide and glass cover slip were cleaned before use. A drop or two of the 1:1 glycerine and water solution was placed onto the labelled slide. While looking through the dissecting microscope, the best sections were selected from the watch glass with a narrow artist’s brush and transferred into the 1:1 glycerine droplet on the slide. After placing about six sections on the slide, the sections were covered with a glass cover slip. One edge of the cover slip was placed at the edge of the glycerine droplet on the slide and slowly lowered into place over the sections by placing the dissecting needle tip on its side under the other edge of the cover slip. While slowly pulling out the needle as the cover slip lowered into place, most of the air bubbles exited from under the cover slip on the side of the exiting needle.
Photography
Prepared slides were placed under a 40×-2000× Trinocular Biological Compound Microscope available from Amscope (model T490B) and photographed under the 10× objective (100× magnification). Images were taken with a 5 Mb AmScope digital camera. Images were cleaned of background spots, adjusted for brightness and sharpened, if necessary, using Adobe Photoshop. A stage micrometer was used to design a scale that was applied to each image.
Attalea species sampled. Collections: Noblick 100, 3733-4663. L = Lima; G = Glassman; RM = Robert Montgomery; MBC plants = Number with * Letter or Letters.
Characters defined
Leaflet sampling and the palm leaflet margin parts are illustrated in Figure 1:
C - The cuticle is thickest on the adaxial (upper) surface.
E - Vascular bundles with enlarged sheaths are basically a large bundle of sclerenchymous fiber strands with a small portion of vascular tissue (xylem and phloem). They have been referred to as somewhat enlarged marginal veins (Tomlinson et al. 2011), vascular bundles with exaggerated sheaths (Noblick 2013, 2017) or vascular bundles with reinforced sclerenchymous sheaths (Sant’Anna-Santos et al. 2018). These are found on the outer edge of the leaflet margins, usually with the largest one on the abaxial (lower) side.
EP - In Attalea, the epidermis is composed of a single layer of small square to round cells.
ET - Expansion tissue is a cluster of achlorophyllous cells often two to three layers thick and located on the adaxial side of the proximal leaflet margin. Rarely are they seen on the distal margin.
FS - Fibrous strands in Attalea are more frequently found on the adaxial surface, usually nestled among and between the hypodermal cells and offer addition support in the leaf.
HY - In Attalea, the hypodermis is a layer of cells just below the epidermis, which are larger than the epidermal cells and are one or two cell layers thick. The round cells lack chlorophyll and frequently include fibrous strands, especially on the adaxial surface.
P - Primary vascular bundles are the largest vessels in the leaflet and often connect to both the adaxial and abaxial hypodermis (HY) with their thick sclerenchymous sheath. They are slightly swollen with large adaxial xylem vessels and at least two to four phloem poles consisting of sieve elements and companion cells located abaxially.
PL - The palisade cell layer is located adaxially and is composed of one to two layers of seamlessly aligned, vertically elongated, chloroplast-rich cells. Not all Attalea species have a well-defined palisade layer.
S - Secondary vascular bundles are the second largest vessels in the leaf blade. Xylem and phloem vessels are less visible than in the primary vascular bundles.
SL - The spongy layer is located between the adaxial palisade layer (if there is one) and the abaxial hypodermis and is composed of round, more loosely packed chlorophyll-rich cells.
T - Tertiary vascular bundles are the smallest vessels in the leaf and are often attached to the abaxial hypodermis.
Results and Discussion
Summarizing the following important facts concerning Attalea leaflets (Figs. 2-9):
Protruded margins
The abaxial surface of proximal leaflet margins is usually protruded, appearing to bend or angle the margin upwards (Figs. 2a-c, 2e-f, 2i-j; 3a-b, 3d, 3f-h; 4e-g), while the adaxial surface of distal leaflet margins is usually protruded, appearing to bend or angle the marginal downwards (Figs. 2a-c, 2e-j; 3a-h; 4a-g). If neither surface is protruded or prominent, then the proximal leaflet margins will frequently bend upwards and the distal margins will bend downwards (Fig. 2d). Thus we have upward facing proximal margins meeting downward facing distal margins in Attalea leaflets. Sometimes and less commonly, the proximal margin does not bend upwards or downwards (Figs. 2h; 3c,e; 4a-d).
Expansion tissue
Proximal leaflet margins usually include a region of achlorophyllous cells (cells without chlorophyll) on the adaxial side of the leaflet called expansion tissue (Fig. 1). It is called expansion tissue because these cells are thought to expand when water is abundant and contract when water is scarce. The distal leaflet margins almost never contain expansion tissue (Fig. 1).
In Allagoptera arenaria (Gomes) Kuntze, Defaveri et al. (2015) first noted that the leaflet margin can be abaxially or adaxially prominent (protruded) or not at all. Pinedo et al. (2016) later pointed out that adjacent leaflet margins of Allagoptera complement each other with one protruded adaxially and the adjacent one protruded abaxially. The same phenomenon was found to occur in Attalea, but with a consistent pattern. The distal margins always protruded adaxially and the proximal margins nearly always protruded abaxially. This pattern was found to hold true in our MBC Allagoptera arenaria as well, reconfirming Pinedo’s (2016) observations. Attalea and Allagoptera both belong to the Attaleinae subtribe, so their similarity should be no surprise. From 121 palm genera sampled so far, this pattern seems to hold true for proximal and distal margins (personal observation) when they are dissimilar. If the margins are angled or bent downwards, they are distal margins and if they are angled or pointed upwards they are proximal margins (Noblick unpublished). The abaxially or adaxially prominent margins that Defaveri et al. (2015) noted most likely represented opposite leaflet margins. Based on this study’s findings, one can even guess which of his margins were proximal and which were distal.
Attalea leaflets nearly always have expansion tissue on the adaxial surface of the proximal margin. These cells are usually missing on the distal margin or are rarely seen as in our sample of A. funifera (Fig. 2e). Expansion tissues were found to be missing on the proximal margins of A. burretiana (Fig. 2d) and A. butyracea (Fig. 4e) and nearly absent in A. oleifera (Fig. 2g) and A. vitrivir (Fig. 3g). Expansion cells were also observed in our MBC Allagoptera arenaria, but not in A. campestris (Mart.) Kuntze. To date, expansion tissues in leaf margins have rarely been observed in palm genera with the exception of Acanthophoenix, Cyphophoenix and Nypa. One can only speculate as to the purpose of these expansion cells. Attalea are renowned for their characteristic strongly ascending leaves that tend to be held in an almost vertical position, rather than a spreading, pendulous one, and many Attalea species grow in areas of seasonal rainfall. When water is scarce during the dry season, the palm leaf desiccates, which shrinks the expansion tissues along the proximal leaflet margin pulling or bending the proximal margin upwards forming a “gutter” or groove (Figs. 1; 2c; 4a-b). One hypothesis is that when it rains, water will run down the surface of the leaflet from the distal to the proximal side where it will encounter the “gutter” along the proximal margin and drain down the leaflet to the midrib and then continue to stream down the stem to the ground, where it pools and is absorbed by the roots. Alternately, this area of expansion cells may be used to store water, sugars or essential minerals.
a-j. Leaflet anatomy of the proximal and distal margins of Attalea s.s. - a. A. allenii; b. A. barreirensis; c. A. brasiliensis; d. A. burretiana; e. A. funifera; f. A. geraensis; g. A. oleifera; h. A. humilis; i. A. pindobassu; j. A. seabrensis. Scale bars = 0.5 mm.
Phylogenetic patterns of the former genera
Attalea was formerly separated into Attalea, Orbignya, and Scheelea, and two smaller genera Maximiliana and Ynesa (Glassman 1999). Several studies have been conducted on Attalea phylogeny (Meerow et al. 2009; Rodríguez del Castillo et al. 2016; Freitas et al. 2016). They showed high support for a monophyletic Attalea, even though some of the studies used different markers (Rodríguez del Castillo et al. 2016). Within Attalea, Meerow et al. (2009) identified a well-supported Attalea clade (A. brasiliensis, A. humilis, A. oleifera, A. pindobassu, A. funifera, and A. seabrensis). They further found that the Caribbean A. crassispatha (with its unique, very narrow, straight to stiffly-curved trough-shaped petals and coiled stamens) was sister to two remaining clades, A. cohune and A. guacuyule with their Orbignya flowers as sister species and sister to a Scheelea clade and an Orbignya clade.
Freitas et al. (2016), using similar molecular markers as Meerow et al. (2009, 2015), included 37 species, adding 21 more species than Meerow (2009, 2015), and recovered three main clades. Their results were similar to Meerow (2009, 2015)in that the resulting clades were mixed, combining species from the previously separated genera. These phylogenetic studies firmly established that only one genus, Attalea, should be recognized.
Likewise, when the margins of species from these former genera were compared with each other (Figs. 2-4) no anatomically unique characters were found to maintain them as different genera. In fact, the anatomy of some species of Attalea and Scheelea were found to be more similar to each other than to the margins of species within their own former genera (Fig. 5).
a-h. Leaflet anatomy of the proximal and distal margins of formerly Orbignya - a. A. brejinhoensis; b. A. cohune; c. A. crassispatha; d. A. eichleri; e. A. guacuyule; f. A. microcarpa; g. A. vitrivir; h. A. speciosa. Scale bars = 0.5 mm.
Resolving Attalea species complexes using leaflet anatomy
Leaflet anatomy has been successfully used to key out palm species, as in Syagrus (Glassman 1972; Noblick 2013, 2017), Attaleinae (Pinheiro 1997), Ceroxylon (Sanin & Galeano 2011), Mauritinae (Guevara et al. 2011), Johannesteijsmannia (Noraini et al. 2012), Acrocomia (Vianna et al. 2017), and Butia (Sant’Anna-Santos et al. 2018). It has been used to distinguish new species of Butia (Sant’Anna-Santos 2021), new species of Syagrus (Sant’Anna-Santos et al. 2023a,b,c), to clearly define two controversial species of Butia (Sant’Anna-Santos et al. 2015), and to resolve species complexes, as in Syagrus glaucescens (Firmo et al. 2021). Leaflet anatomy has even been used to understand palm evolution (Horn et al. 2009). More recently, Mata et al. (2022) used leaf anatomy to understand the taxonomic similarities and identification of seven Attalea species of the ‘babassu’ complex. Botanists have found several challenging species complexes in Attalea. Pintaud (2008) suggested nine unresolved complexes within the genus. Some botanists split these into separate species (Glassman 1999), and others lump them into single polymorphic species (Henderson et al. 1995; Henderson 2020). Based on our results, leaflet anatomy can be used to shed light on some of these complexes and to clarify uncertainties.
Attalea butyracea and A. osmantha (Fig. 6a-f)
At MBC, our A. butyracea collections produce small one to two-seeded fruits with a bright yellow epicarp. MBC also grows A. osmantha from Trinidad, which has medium to large brown fruits when mature. The larger fruits of A. osmantha are never “buttery” colored, so it was decided to maintain this species separate from A. butyracea with the smaller, buttery yellow colored fruits. Nevertheless, at MBC the younger palms of A. osmantha from Trinidad look nearly identical to the A. butyracea from Costa Rica in the way their leaves are interwoven together basally around the developing stem. At this stage of development, one would find them difficult to tell apart. Likewise, Attalea butyracea (Fig. 4e) and A. osmantha (Fig. 4f) have very similarly shaped margins, although A. osmantha usually has a more developed region of expansion tissue (Figs. 4f; 6d-f). Additional specimens (Fig. 6b-c) of A. butyracea from Costa Rica appear to have larger fibrous strands than A. osmantha along the adaxial surface, which add additional support in the leaf, but since A. butyracea is so widely distributed, one should expect some local variation throughout its range. Perhaps differences in fruit size and color and size of fibrous strands on the adaxial margins are not enough to maintain these two species as distinct. Because the habit and growth of these two species are so morphologically similar, Henderson (2020) was probably justified in synonymizing A. osmantha into A. butyracea.
a-g. Leaflet anatomy of the proximal and distal margins of formerly Scheelea (a-b, e-g) and Maximiliana (c-d) - a. A. leandroana; b. A. phalerata; c-d. A. maripa; e. A. butyracea; f. A. osmantha; g. A. anisitsiana. Scale bars = 0.5 mm.
Attalea phalerata, A. leandroana and A. anisitsiana (Figs. 4; Fig. 6g-m)
Glassman (1999) separated A. anisitsiana from A. phalerata based on plant size differences. Attalea anisitsiana is nearly acaulescent, with a short stem, shorter rachillae and smaller fruits. While A. anisitsiana and A. phalerata both have male flowers that are unilaterally arranged on their rachillae, A. leandroana has spirally arranged male flowers. Henderson (1995) first placed A. leandroana (Scheelea leandroana) in synonomy with A. phalerata, but later, Henderson (2020) placed it in synonymy with A. butryacea, probably because of its spirally arranged flowers (A. Henderson 2023, personal communication). The original author of A. leandroana, Barbosa Rodrigues (1989), clearly illustrates A. leandroana as having an A. phalerata habit with persistent leaf bases rather than a self-cleaning habit typical of A. butyracea. Our MBC specimen of A. leandroana, which has spirally arranged flowers, has leaflet margins (Fig. 4a) that look to be nearly identical to A. phalerata (Fig. 4b), thereby justifying its synonomy with that species. However the anatomy of A. anisitsiana (Fig. 4g) differs from these two and may be distinct. After comparing additional specimens (Fig. 6g-m), a pattern emerged between the A. phalerata specimens from Brazil (Fig. 6k-m) and those from Paraguay (Fig 6h-i). The difference in shape of the proximal margins of several of the Brazilian specimens might be due to desiccation, which causes the expansion tissues to contract. Compare the “hydrated” specimen of A. phalerata from Tocantins, Brazil (Fig. 6j) with the “desiccated” A. phalerata from Pará (Fig. 6g-m). Perhaps the difference in the proximal margin may only be due to the fact that one specimen is more desiccated, since their distal margins match morphologically. Still, the Paraguayan palms’ proximal margins consistently had a different shape than the Brazilian ones. The Paraguayan A. phalerata are also closer anatomically to the Paraguayan A. anisitsiana than to the Brazilian specimens. Examination of more palms from different regions of Brazil and Paraguay is needed before we can draw any definite conclusions.
Attalea cohune and A. guacuyule (Fig. 7a-d)
Attalea cohune grows along the East coast of Mexico and Central America and A. guacuyule grows along the West coast. According to Glassman (1999), A. guacuyule has shorter staminate rachillae (10-18 vs. 20-30 cm), shorter pistillate rachillae (5-8 vs. 10-13 cm), and shorter staminate flowers (8-10 vs. 13-15 cm) with narrower petals (4-5 vs. 6-8 mm). Also, the fiber clusters of the endocarp are common and conspicuous in A. guacuyule, but inconspicuous in A. cohune.
Comparative leaflet anatomy of the proximal and distal margins of Attalea and formerly Scheelea showing more similarity to each other than to margins within their own former genera. Scheelea (a-c), Attalea (d-f) - a. A. butyracea; b. A. osmantha; c. A. anisitsiana; b. A. oleifera; e. A. burretiana; f. A. allenii. Scale bars = 0.5 mm.
The leaflet margins of A. cohune specimens varies from East Mexico and Honduras (Fig. 7b-d), but they are similar enough in their proximal margins to be recognized as belonging to the same species. However, the adaxial (upper) side of the distal margin of the Honduran specimen (Fig. 7d) differs from those of East Mexico (Fig. 7b-c), but closely resembles the distal margin of A. guacuyule of West Mexico (Fig. 7a). These margins are not that much different from our single specimen of A. guacuyule. If the distal margin of A. guacuyule is straightened out, it resembles the distal margin of the A. cohune specimens, especially the one from Honduras. Still, the square shape of the proximal leaflet margin of A. guacuyule differs from the abaxially protruded margin of the A. cohune specimens. It also lacks the large marginal vascular bundle with the enlarged sheath found in all of the A. cohune samples. More samples of A. guacuyule are needed to confirm that these differences are consistent and are not just representative of this isolated specimen. Evidence from leaflet margins is inconclusive as to whether A. guacuyule and A. cohune are the same species as proposed by Henderson (2020).
Leaflet anatomy of the proximal and distal margins of Attalea butyracea-A. osmantha complex and the Attalea phalerata-A. anisitsiana-A. leandroana complex. a. A. butyracea (MBC); b-c. A. butyracea (Costa Rica); d-f. A. osmantha (Trinidad); g. A. anisitsiana (Paraguay); h-i. A. phalerata (Paraguay); j. A. phalerata (Brazil: TO); k. A. leandroana (Brazil); l. A. phalerata (MBC); m. A. phalerata (Brazil: PA). Scale bars = 0.5 mm.
Attalea oleifera and A. burretiana (Fig. 7e-j)
There is no question that A. burretiana and A. oleifera are closely related as they both have similar fibers emanating from their leaf sheaths. However, our MBC A. oleifera from Areia, Paraíba, Brazil are robust palms with massively thick stems (dbh = 50-66 cm). On the other hand, our A. burretiana from Bahia have ordinary medium-sized stems (dbh = 31-42 cm). Attalea oleifera has a more highly branched infructescence and smaller fruits than the more sparsely branched A. burretiana with much larger fruits. Glassman (1999) separated A. burretiana from A. oleifera in his key based on the width of their middle leaflets (6-8 vs. 3.5-5.5 cm) and anther length (9-14 vs. 6-8 mm). Leaflet samples from Bahian Attalea burretiana have proximal margins with no trace of expansion cells, which makes it somewhat unique among other Attalea species (Figs. 2d; 5e; 7h-j). Our A. oleifera from Areia, Paraíba have a thin region of expansion cells on their proximal margins (Figs. 2g; 7f-g). This throws some doubt on Henderson’s lumping of these two species; however, further examination of the margins of a dried A. oleifera specimen collected from Bananeiras, Paraíba, ca. 25 km from Areia, also shows no expansion cells (Fig. 7e) and is nearly identical to the A. burretiana leaflet margins of Bahia. Therefore, the Bananeiras specimen lends support to Henderson (2020) synonymizing of A. burretiana into A. oleifera.
Leaflet anatomy of the proximal and distal margins of Attalea cohune-A. guacuyule complex and the A. oleifera-A. burretiana complex. a. A. guacuyule (West Mexico); b-c. A. cohune (East Mexico); d. A. cohune (Honduras); Brazil (e-j) - e. A. oleifera (PB - Bananeiras); f-g. A. oleifera (PB - Areia); h-j. A. burretiana (BA - Amélio Rodrigues); i. A. burretiana (BA - Ibiripitanga); j. A. burretiana (Salvador). Scale bars = 0.5 mm.
Attalea pindobassu and A. seabrensis (Fig. 8)
Attalea seabrensis and A. pindobassu are geographically separated. Attalea pindobassu occurs in the eastern Serra do Ouro and Serra do Tombador of Bahia, and A. seabrensis occurs in the more western Serra do Sincorá and the Chapada Diamantina region. These two sierra regions are separated by a wide valley of Caatinga vegetation unsuitable for Attalea.
These species look very similar and perhaps should be synonymized, even though A. pindobassu is overall a more robust palm. Glassman’s key states that the lower third to lower one half of A. seabrensis leaves have clustered leaflets, but none or less than one tenth of the lower leaflets are clustered in A. pindobassu, a character that has been confirmed in the field. Attalea pindobassu palms have a maximum of only 12 stamens and A. seabrensis have a maximum of 17. Furthermore, a distinct spike in the chemical composition of A. seabrensis waxes was observed during gas chromatography, but no such spike was detected in the waxes extracted from A. pindobassu samples (Noblick, not published). Although the chemical composition of this spike was never analyzed, it may still indicate a distinct genetic difference separating the two species.
The leaflet margins of A. pindobassu from Tapiramutá, Bahia (Figs. 2i; 8a-b) in the Serra do Tombador look suspiciously similar to the A. seabrensis sample from Seabra, Bahia (Figs. 2j; 8e). However the leaflet margins from the Serra do Ouro near Miguel Calmon in the northern end of A. pindobassu’s range have a slightly different shape (Fig. 8c-d). The Tapiramutá population of A. pindobassu appears to be more closely related to A. seabrensis than to the Miguel Calmon population. However, their similarity is not all that convincing and so evidence from the leaflet anatomy was inconclusive.
The question of whether they are two species or one has perhaps been resolved by MBC’s A. seabrensis, whose seed was collected from the type locality of Seabra, Bahia and has leaves with leaflets that are all evenly distributed along its midrib identical to A. pindobassu. The visible character of clustering leaflets that Glassman (1999) and Noblick (1991) used to separate these two species is apparently invalid. Clustering may be environmentally induced in the A. pindobassu-A. seabrensis complex. Henderson (2020) appears justified in synonymizing A. seabrensis into A. pindobassu based on the similarities between the Tapiramutá leaflet margin sample and the Seabra sample and the non-clustered leaflets of MBC’s A. seabrensis specimen.
Attalea speciosa, A. brejinhoensis and A. vitrivir (Fig. 9a-f)
Glassman (1999) separated A. brejinhoensis, A. vitrivir and A. speciosa by the size of their petioles and the size of the pistillate portion of their androgynous rachillae. According to Glassman (1999), Attalea brejinhoensis (Orbignya brejinhoensis), A. vitrivir (O. oleifera) and A. speciosa (O. phalerata) have petioles with the following respective lengths: 80-100 vs. 0-10 vs. 15-50 cm and the pistillate portion of their androgynous rachillae is up to 1 vs. 2-3 vs. 10-15 cm long, respectively. Neither of these key characters work with the specimens growing at MBC. Glassman (1999) never clarified how he determined the presence and absence of petioles and how he measured his petioles, nor has this author been able to determine what Glassman (1999) was measuring (e.g., sheath included?). In addition, his measurements of the pistillate portion of inflorescences does not match anything we have growing at MBC, some of which we grew from seed collected from the type locality (i.e., A. brejinhoensis). The leaflet margins of these three species of Attalea look suspiciously similar (Figs. 3a,g-h; 9a-d). In the field and in the garden, the palms closely resemble one another. Attalea brejinhoensis has unique fruits, where the persistent perianth (fruit cup) covers ca. 1/2-3/4 of the mature fruit, whereas it only covers about 1/4 of the mature fruits of A. speciosa (M. Balick 1991, personal communication). This character has been verified both in the field and in the living collections at MBC. Nevertheless the margins of A. brejinhoensis (Fig. 3a) and A. speciosa (Fig. 3h) are almost identical (Fig. 9a-d). Attalea vitrivir (Figs. 3g; 9b) differs the most from the other two in lacking expansion tissues on the proximal margin and according to Glassman (1999), it has larger fruit than the other two (up to 14 cm long vs. 10-12.5 cm). The presence or absence of the expansion tissues may not be a reliable character as we saw in the A. oleifera-A. burretiana and the A. butyracea-A. osmantha complexes. The marginal anatomy of A. vitrivir is close enough to A. brejinhoensis and A. speciosa to justify sinking these three into one species, A. speciosa, as Henderson (2020) proposed.
Leaflet anatomy of the proximal and distal margins of Attalea pindobassu-A. seabrensis complex. Brazil: BA (a-e) - a-b. A. pindobassu (Tapiramutá); c-d. A. pindobassu (Miguel Calmon); e. A. seabrensis (Seabra). Scale bars = 0.5 mm.
Using leaflet marginal anatomy to confirm hybrid parentage
Naturally occurring hybrids are not uncommon in Attalea (Fig. 10b,e). Among some of the currently recognized hybrids are A. × teixeirana (Bondar) Zona, A. × voeksii Noblick ex Glassman, A. × piassabossu Bondar. In Goiás, the author collected fruits from a palm that looked like a robust acaulescent A. speciosa, which was larger than the A. eichleri growing in the same area, but smaller than A. speciosa seen nearby. Bondar (1954) originally recognized this hybrid as Orbignya × teixeirana Bondar (= A. × teixeirana), the natural hybrid between A. speciosa and A. eichleri, but Glassman (1999) later wrote that it was instead a hybrid between A. phalerata and A. eichleri based on a paper by Medeiros-Costa et al. (1985). Henderson (2020) accepted Bondar (1954) and recognized A. speciosa and A. eichleri as the parental species of A. × teixeirana. Regardless of the proper name for our specimen, it is undoubtedly a hybrid between A. speciosa and A. eichleri. Palms grown from seed revealed that they were neither A. eichleri, nor A. speciosa. The palms have longer leaves and evenly spaced leaflets like A. speciosa, but they have remained acaulescent with smaller inflorescences like A. eichleri. Examination of the hybrid’s margins reveals the intermediate nature of this palm, pointing to A. eichleri as the likely mother plant (Fig. 10b-c). The hybrid’s lamina is thin like A. eichleri, but the shape of the margin somewhat resembles A. speciosa, even sporting fine hairs or tomentum on both the proximal and distal margins like A. speciosa.
Bondar (1942) first recognized and described the hybrid A. × piassabossu from Salvador, Bahia, Brazil, as the natural hybrid between A. burretiana (A. oleifera) and A. funifera. In the Salvador area, A. funifera assumes an acaulescent habit, but A. burretiana is a tall columnar palm. Close examination of the margin of this hybrid (Fig. 10e), confirms that the likely maternal parent is A. funifera (Fig. 10f), and less subtle characters, such as the thicker leaf blade, margin shape and numerous adaxial fiber strands recalls characters of A. burretiana. While the arrangement of the vascular bundles and shape of the primary vascular bundle in the distal margin resembles A. funifera, in the proximal margin they look more like A. burretiana, especially the presence, size and location of the vascular bundle with an enlarged sheath. The intermediate size of the expansion tissue upholds its hybrid status between these two parents.
Leaflet anatomy of the proximal and distal margins of A. speciosa-A. brejinhoensis-A. vitrivir complex. Brazil - a. A. brejinhoensis (BA); b. A. vitriver (BA); c. A. speciosa (PI); d-e. A. speciosa (TO); f. A. speciosa (MA). Scale bars = 0.5 mm.
Shedding light on the anatomy and taxonomic uncertainties
This survey of the leaflet anatomy of Attalea revealed stark differences between the proximal and distal margins. With few exceptions the proximal margin has expansion tissues at its tip and usually bends upward and the distal margin bends downward.
Leaflet anatomy of the proximal and distal margins of Attalea hybrids and their parental species. a. A. speciosa; b. A. × teixeirana; c. A. eichleri; d. A. burretiana; e. A. × piassabossu; d. A. funifera. Scale bars = 0.5 mm.
Attalea leaflet margin anatomy does not support Attalea being separated into four or five separate genera. Marginal similarities between former genera (i.e., Attalea, Scheelea) confirm a single genus, Attalea.
Attalea leandroana is A. phalerata in spite of its spirally arranged flowers. The Paraguyan A. anisitsiana is similar to the Paraguyan A. phalerata, but cannot be ruled out as a distinct species. Some Brazilian A. phalerata appear to be consistently different from the Paraguyan A. phalerata, although this difference may be due to shrinkage of the proximal margin expansion tissues.
In spite of differences in fruit size and coloration, A. butyracea from Costa Rica and A. osmantha from Trinidad appear identical as young developing palms. Their leaflet margins are similar, except for the larger fibrous strands in the Costa Rican A. butyracea and its inconsistent formation of expansion tissue. They probably belong to the same species as Henderson (1995, 2020) has proposed.
Comparisons between A. cohune (East Mexico) and A. guacuyule (West Mexico) are inconclusive. Examination of more material from A. guacuyule is needed.
At first, the proximal leaflet margins of A. burretiana appeared to be unique in their absence of expansion tissues, but a dried specimen of A. oleifera from Bananeiras, Paraíba also lacked expansion tissues. Therefore, leaflet margins of these two species are not distinct enough to maintain them as separate species, and Henderson (2020) seems justified in synonymizing them into one, A. oleifera.
The leaflet margins of the southern population of A. pindobassu from Tapiramutá resemble those of A. seabrensis from Seabra. MBC’s A. seabrensis grown from seed collected in Seabra lacks the clustered leaflets present in its native relatives and therefore appears to be identical to A. pindobassu. Therefore, Henderson (2020) was justified in synonymizing A. seabrensis into A. pindobassu.
The key differences proposed by Glassman (1999) between A. speciosa, A. brejinhoensis and A. vitrivir were found lacking. In spite of slight differences in the marginal anatomy of A. vitrivir, enough similarities between the leaflet marginal anatomies of A. speciosa, A. brejinhoensis and A. vitrivir justify Henderson’s proposal to sink them into a single species, A. speciosa.
Leaflet margin anatomy reflects intermediate characters between hybrid and their parental species. This was demonstrated clearly in the hybrid between A. speciosa and A. eichleri and the hybrid between A. burretiana (A. oleifera) and A. funifera.
Acknowledgements
Thanks is due to the Montgomery Botanical Center, for supporting this research both financially and with its onsite palm specimens. Thanks to Joanna Tucker Lima and to the reviewers who made valuable suggestions for the improvement of this paper.
Data availability statement
In accordance with Open Science communication practices, the authors inform that all data are available within the manuscript.
References
-
Asmussen CB, Dransfield J, Deickmann V, Barfod AS, Pintaud J-C & Baker WJ (2006) A new subfamily classification of the palm family (Arecaceae): evidence from plastid DNA phylogeny. Botanical Journal of the Linnean Society 151: 15-38. Available at <Available at https://academic.oup.com/botlinnean/article/151/1/15/2420456?login=true >. Access in July 2023.
» https://academic.oup.com/botlinnean/article/151/1/15/2420456?login=true - Barbosa Rodrigues J (1989) Sertum Palmarum Brasiliensium, ou, Relation des palmiers noveaux du Bresil: decouverts, descrits et dessines d’après nature. Jardim Botânico de Rio de Janeiro, Rio de Janeiro. 254p.
-
Baker W & Dransfield J (2016) Beyond genera Palmarum: progress and prospects in palm systematics. 182: 207-233. DOI: 10.1111/boj.12401
» https://doi.org/10.1111/boj.12401 - Bondar G (1942) A piassaveira e outra palmerias Attaleaineas na Bahia. Boletim Instituto Central Fomento Econômico da Bahia 13: 1-73.
- Bondar G (1954) Nova espécie de Orbignya, produtora do óleo de Babaçu. Archivos do Jardim Botânico Rio de Janeiro 13: 58.
-
Defaveri ACA, Barros CF, Arruda RCO, Simas NK & Sato A (2015) Allagoptera arenaria (Arecaceae): Leaf anatomy of a palm from the Brazilian shore. Brittonia 67: 336-349. DOI: 10.1007/s12228-015-9384-2
» https://doi.org/10.1007/s12228-015-9384-2 - Dransfield J, Uhl N, Asmussen C, Baker W, Harley M & Lewis C (2008) Genera Palmarum. The evolution and classification of palms. Kew Publishing, Kew. 732p.
- Freitas C, Meerow A, Pintaud J-C, Henderson A, Noblick L, Costa F, Barbosa C & Barrington D (2016) Phylogenetic analysis of Attalea (Arecaceae): insights into the historical biogeography of a recently diversified Neotropical plant group. Botanical Journal of the Linnaean Society: 287-302.
-
Firmo DHT, Santos SA, Perez MEMP, Soffiatti P & Sant’Anna-Santos BF (2021) Reassessing species boundaries in the Syagrus glaucescens complex (Arecaceae) using leaf anatomy. Botany 99: 379-387. DOI: 10.1139/cjb-2020-0165
» https://doi.org/10.1139/cjb-2020-0165 - Glassman SF (1972) Systematic studies in the leaf anatomy of palm genus Syagrus American Journal of 59: 775-788.
- Glassman SF (1999) A taxonomic treatment of the palm subtribe Attaleinae (Tribe Cocoeae). Illinois Biological Monographs 59: 1-414.
-
Guevara LI, Stauffer FW & Jáuregui DJ (2011) Anatomía comparativa de las lámina foliar y sistemática en las subtibu neotropical Mauritiinae (Arecaceae, Calamoideae). Brittonia 63: 379-395. DOI: 10.1007/s12228-010-9176-7
» https://doi.org/10.1007/s12228-010-9176-7 - Henderson A, Galeano G & Bernal R (1995) A field guide to the palms of the Americas. Princeton University Press, Princeton. 352p.
-
Henderson A (2020) A revision of Attalea (Arecaceae, Arecoideae, Cocoseae, Attaleinae). Phytotaxa 444: 001-076. DOI: 10.11646/phytotaxa.444.1.1
» https://doi.org/10.11646/phytotaxa.444.1.1 -
Horn JW, Fisher JB, Tomlinson PB, Lewis CE & Laubengayer K (2009) Evolution of lamina anatomy in the palm family (Arecaceae). American Journal of Botany 96: 1462-1486. DOI: 10.3732/ajb.0800396
» https://doi.org/10.3732/ajb.0800396 -
Mata LR, Moreira ALC, Pinedo AS, Gomes SM & Azevedo VCR (2022) Leaf anatomy as an aid to the taxonomy of the ‘babassu’ complex (Attaleaspecies). Flora 293: 152092. DOI: 10.1016/j.flora.2022.152092
» https://doi.org/10.1016/j.flora.2022.152092 - Medeiros-Costa JT, Mendes AMC & Brito JLC (1985) Morfodiagnose de Orbignya teixeirana Bondar (Palmae-Cocosoideae) nos estados de Piauí e Maranhão. EMBRAPA-UEPAE, Teresina. 5p.
-
Meerow A, Noblick L, Borrone J, Couvreur T & Mauro-Herrera M (2009) Phylogenetic analysis of seven WRKY genes across the palm subtribe Attaleinae (Arecaceae) identifies Syagrus as sister group of the coconut. PLoS ONE 4: e7353. DOI: 10.1371/journal.pone.0007353
» https://doi.org/10.1371/journal.pone.0007353 -
Meerow AW, Noblick L, Salas-Leiva DE, Sanchez V, Francisco-Ortega J, Jestrow B & Nakamura K (2015) Phylogeny and historical biogeography of the cocosoid palms (Areacaceae, Arecoideae, Cocoseae) inferred from sequences of six WRKY gene family loci. Cladistics 31: 509-534. DOI: 10.1111/cla.12100
» https://doi.org/10.1111/cla.12100 - Noblick L (1991) Indigenous palms of the state of Bahia, Brazil. PhD Dissertation. The University of Illinois, Chicago. 523p.
-
Noblick L (2013) Leaflet anatomy verifies relationships within Syagrus (Arecaceae) and aids in identification. PhytoKeys 26: 75-99. DOI: 10.3897/phytokeys.26.5436
» https://doi.org/10.3897/phytokeys.26.5436 -
Noblick L (2017) Key to Syagrus identification using leaflet margin anatomy: supplement to “a revision of Syagrus (Arecaceae)”. PhytoKeys 81: 19-43. DOI: 10.3897/phytokeys.81.12909
» https://doi.org/10.3897/phytokeys.81.12909 - Noraini T, Ruzi AR, Murnida MK & Hajar NR (2012) Systematic significance of leaf anatomy in Johannesteijsmannia H.E. Moore (Arecaceae). Pertanika Journal of Tropical Agricultural Science 35: 223-236.
-
Pinedo AS, Martins RC, Oliveira RC & Gomes SM (2016) Leaflet anatomy in Allagoptera (Arecaceae). Botanical Journal of the Linnean Society 182: 361-375. <https://doi.org/10.1111/boj.12439>.
» https://doi.org/10.1111/boj.12439 - Pinheiro C (1997) Systematic and agro-ecological studies in the Attaleinae (Palmae). PhD thesis. City University of New York, New York. 479p.
-
Pintaud J-C (2008) An overview of the taxonomy of Attalea (Arecaceae). Revista Peruana de Biología 15: 55-63. Available at <Available at https://dialnet.unirioja.es/servlet/articulo?codigo=2874767 >. Access in July 2023.
» https://dialnet.unirioja.es/servlet/articulo?codigo=2874767 - Rodríguez del Castillo A, García-Dávila C, Mejía K & Pintaud J-C (2016) Attalea: insights in to the diversity and phylogeny of an intriguing genus. Palms 60: 109-124.
-
Sanin MJ & Galeano G (2011) A revision of the Andean wax palms, Ceroxylon (Arecaceae). Phytotaxa 34: 1-64. Available at <Available at https://www.mapress.com/phytotaxa/content/2011/f/pt00034p064.pdf >. Access in July 2023.
» https://www.mapress.com/phytotaxa/content/2011/f/pt00034p064.pdf -
Sant’Anna-Santos BF (2021) A new endemic and critically endangered species of Butia (Arecaceae) with comments on morpho-anatomical novelties in the genus. Plant Systematics and Evolution 307: 4. DOI: 10.1007/s00606-020-01729-w
» https://doi.org/10.1007/s00606-020-01729-w -
Sant’Anna-Santos BF, Carvalho Junior WG & Amaral VB (2015) Butia capitata (Mart.) Becc. lamina anatomy as a tool for taxonomic distinction from B. odorata (Barb.Rodr.) Noblick comb. nov. (Arecaceae). Anais da Academia Brasileira de Ciências 87: 71-81. Available at <Available at https://www.scielo.br/scielo.php?script=sci_arttext&pid=S0001-37652015000100071&lng=en&nrm=iso >. Access in July 2023.
» https://www.scielo.br/scielo.php?script=sci_arttext&pid=S0001-37652015000100071&lng=en&nrm=iso -
Sant’Anna-Santos BF, Santos SA, Nunes ELP, Francino DMT & Carvalho Jr WGO (2018) Does leaf anatomy aid in species identification of Butia (Arecaceae)? Aob Plants 10: 1-15. DOI: 10.1093/aobpla/ply046
» https://doi.org/10.1093/aobpla/ply046 -
Sant’Anna-Santos BF, Azevedo IFP, Micheli R & Soffiatti P (2023a) Morpho-anatomical novelties of a dwarf Syagrus (Arecaceae) of canga: implications for ecology, conservation, and taxonomy. Plant Systematics and Evolution 309: 8. DOI: 10.1007/s00606-023-01843-5
» https://doi.org/10.1007/s00606-023-01843-5 -
Sant’Anna-Santos BF, Carvalho LFL & Soffiatti P (2023b) A new rupicolous palm from the campos rupestres, Minas Gerais, Brazil. Edinburgh Journal of Botany 80: 1-15. DOI: 10.24823/EJB.2023.1974
» https://doi.org/10.24823/EJB.2023.1974 -
Sant’Anna-Santos BF, Micheli R, Carvalho LFL & Soffiatti P (2023c) A new bluish-leavedSyagrus(Arecaceae) from an overlooked OCBIL in the Espinhaço Range (Brazil). Plant Ecology and Evolution 156: 129-145. DOI: 10.5091/plecevo.101027
» https://doi.org/10.5091/plecevo.101027 -
Vianna SA, Carmelo-Guerreiro SM, Noblick LR & Colombo CA (2017) Leaf anatomy of Acrocomia (Arecaceae): an additional contribution to the taxonomic resolution of the genus with great economic potential. Plant Systematic Evolution 303: 233-248. Available at <Available at https://www.researchgate.net/publication/312130663_Leaf_anatomy_of_Acrocomia_Arecaceae_an_additional_contribution_to_the_taxonomic_resolution_of_a_genus_with_great_economic_potential >. Access in July 2023.
» https://www.researchgate.net/publication/312130663_Leaf_anatomy_of_Acrocomia_Arecaceae_an_additional_contribution_to_the_taxonomic_resolution_of_a_genus_with_great_economic_potential - Tomlinson PB, Horn JW & Fisher JB (2011) The anatomy of palms. Oxford University Press, Oxford. 251p.
- Wessels Boer J (1965) Flora of Suriname. Vol. 5 - Palmae. E. J. Brill, Leiden. 172p.
- Wessels Boer J (1988) Palmas indígenas de Venezuela. Pittiera 17: 1-332.
- Zona S (2002) Name changes in Attalea Palms 46: 132-133.
Publication Dates
-
Publication in this collection
16 Aug 2024 -
Date of issue
2024
History
-
Received
28 July 2023 -
Accepted
16 Feb 2024
