Bertholletia excelsa, Bonpland, 1807

Huang, Ya-Yi, Mori, Scott A. & Kelly, Lawrence M., 2015, Toward a phylogenetic-based Generic Classification of Neotropical Lecythidaceae- I. Status of Bertholletia, Corythophora, Eschweilera and Lecythis, Phytotaxa 203 (2), pp. 85-121 : 99-102

publication ID

https://doi.org/ 10.11646/phytotaxa.203.2.1

persistent identifier

https://treatment.plazi.org/id/03B0445E-FFDC-FFD1-FF19-6ACA500F84A9

treatment provided by

Felipe

scientific name

Bertholletia excelsa
status

 

Bertholletia excelsa View in CoL clade (100% BS; Figs. 2A View FIGURE 2 , 7 View FIGURE 7 )

This clade includes only Bertholletia excelsa , which is distributed throughout Amazonia and parts of the Guianas (see Fig. 21 in Prance & Mori, 1990). In the present study, B. excelsa is sister to the Lecythis chartacea clade. It differs from species of that clade by having two calyx lobes (character 16; Fig. 7B View FIGURE 7 ), seeds without an aril (character 47; Fig. 7G View FIGURE 7 ), and a type of secondary indehiscence in which the seed is larger than the opercular opening (character 41; Fig. 7E, F View FIGURE 7 ). Bertholletia excelsa provides good examples of petals pressed against the androecium (character 33; Fig. 7A View FIGURE 7 ) and, as seen in the field or in color images, of the yellow color on the androecial hood at the entrance into the flower. Bertholletia is the only genus of the family with a boney seed testa and the complete absence of an aril ( Fig. 7G View FIGURE 7 ). It does, however, share features with species of the L. poiteaui and L. chartacea clades, including similar androecia with swept in appendages ( Fig. 7D View FIGURE 7 ), 4-locular ovaries ( Fig. 7C View FIGURE 7 ), and long, slender, oblique or geniculate styles ( Fig. 7B View FIGURE 7 ).

Mori & Prance (1990) hypothesized that B. excelsa is related to Lecythis lurida ( Miers 1874: 262) S. A. Mori (1981a: 362) . This hypothesis was based on the following shared characters of the two species ( Mori & Prance, 1990): the presence of cuticular papillae on the abaxial leaf blade surface (character 5; see Fig. 96 in Mori & Prance, 1990), hood appendages swept or curved inward without forming a complete coil (character 31; Fig. 7D View FIGURE 7 ), mature fruits that fall to the ground with the seeds remaining inside (character 40), a unique dehiscence (character 41; Fig. 7E, F View FIGURE 7 ) in Bertholletia , and fruits that do not open at all in L. lurida . The relationship of B. excelsa with L. lurida and related species of the L. poiteaui clade is not supported by this study ( Fig. 3A View FIGURE 3 ).

There are no other species of Lecythidaceae with fruits morphologically similar to those of B. excelsa . The fruits of B. excelsa have thicker and woodier pericarps and are, in fact, dehiscent but the opercular opening is smaller in diameter than that of the seeds, and the operculum falls into the fruit when it dehisces ( Tsou & Mori, 2002) ( Fig. 7E, F View FIGURE 7 ). It has been hypothesized that this type of dehiscence is related to selection for dispersal by rodents, especially agoutis ( Ducke, 1948; Prance & Mori, 1978). In neotropical Lecythidaceae , shifts to different dispersal agents and accompanying morphological changes have occurred a number of times ( Tsou & Mori, 2002). For example, in Allantoma there has been a shift from wind-dispersal facilitated by a unilateral seed wing in most terra firme species to the water-dispersed A. lineata with only a vestigial seed wing ( Huang et al., 2008). Another shift has been from the terra firme dehiscent-fruited, arillate seeded, animal-dispersed L. chartacea to the riverine, indehiscent-fruited, non arillate-seeded, water-dispersed L. rorida O. Berg (1858: 488) ( Kubitzki & Ziburski, 1994). Thus, species of neotropical Lecythidaceae may belong to the same genus even though the morphological adaptations for seed dispersal by different dispersal agents may be quite different.

of the hypocotyl. This monotypic clade is part of the larger Bertholletia clade. Drawings by B. Angell and photo by S.A. Mori.

Other morphological characters that would suggest relationships between B. excelsa and some species of Lecythis may be misleading and are homoplasious on our trees. For example many species of Amazonian Lecythidaceae have thick cuticles and papillae that arise from them, most likely to reduce water loss from the leaves—thus, the presence or absence of papillae should not be given much weight in predicting evolutionary relationships in this family. Even the unique two-lobed calyx ( Fig. 7B View FIGURE 7 ) of B. excelsa is not an absolute indicator of evolutionary relationships because nearly all zygomorphic-flowered neotropical Lecythidaceae (including B. excelsa ) have six calyx-lobe primordia in early floral development (see Fig. 78 in Tsou & Mori, 2007).

Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF