Estudos em Eugenia L. (Myrtaceae) na Amazônia Central: taxonomia com o uso de ferramentas morfoanatômica
Twenty-eight species of the genus Eugenia naturally occurring in the Adolpho Ducke Forest Reserve are described, including their anatomical characterization with a statistical analysis of their morphological and anatomical features. One species (E. illepida) was previously unrecorded for the Amazon...
| Autor principal: | Souza, Maria Anália Duarte de |
|---|---|
| Grau: | Tese |
| Idioma: | por |
| Publicado em: |
Instituto Nacional de Pesquisas da Amazônia - INPA
2020
|
| Assuntos: | |
| Acesso em linha: |
https://repositorio.inpa.gov.br/handle/1/12831 http://lattes.cnpq.br/6155969493322017 |
| Resumo: |
|---|
|
Twenty-eight species of the genus Eugenia naturally occurring in the Adolpho Ducke Forest Reserve are described, including their anatomical characterization with a statistical analysis of their morphological and anatomical features. One species (E. illepida) was previously unrecorded for the Amazon region, and one is a new occurrence in Brazil (E. adenocalyx). Six species (E. caducipetala, E. kerianthera, Eugenia sp. B, Eugenia sp. C, Eugenia sp. D and E. marleneae) are described as new to science. Three sections of the genus are represented: Eugenia (three), Racemosae (five) and Umbellatae (20). The leaf anatomy of 27 species were studied to provide identification tools, and to provide characters for supraspecific groupings. For this analysis, section Umbellatae was divided into six informal groups based on morphological characters. Characters related to the vascular bundles both the in petiole and in the lamina were most useful in the grouping of species and sections. Section Eugenia is characterized by the vascular bundle being cylindrical at the base of petiole and in the form of an attenuated arch at the apex, and bicollateral phloem in the central and apical regions; the abaxial surface has pleated cuticle and peristomatal ornamentation folds; in the midrib it is an attenuated in arch with bicollateral phloem and a surrounding perivascular sheath; in section Racemosae section the form of the vascular bundle is constant over the throughout the petiole, are is different in each species, and tends to have surrounding phloem; in section Umbellatae there are many varying characters that are useful in separating groups only when used in combination; these include the external cuticle and paraclinal wall being striated or smooth on the upper surface, and striated in the lower surface, and straight anticlinal walls. Characters mentioned in the literature as dependent on environmental factors were not corroborated here: including thickened cuticle, presence of trichomes, compressed mesophyll, palisade in more than one layer, which are described as xerophytic occurred in species that are mostly mesomorphic. An alternation function to explain these features is suggested, that these characters could be related to humidity control by obstruction stomata, and for defense against herbivores and pathogenic microorganisms, among others. Characters such as the palisade being in more than one layer, the hypodermis and presence of trichomes, which are extremely plastic, do seem to be related to xerophytic environments, but can not be related to taxonomy groupings. They may, however, be useful in local floras. The data resulting from the morphological and anatomical descriptions were tested using Principal Component Analysis
(PCA) and Hierarchical Cluster Analysis (HCA) to ascertain which characters are most related to the phylogenetic groupings (sections) and morphologically related groups. In the PCA, the presence of trichomes was important in the separation of species, independently of their relationships, while other characters segregated the formal and informal groups. In the cluster analysis, dendrograms of the branches well defined some informal groups (Feijoi and Ramiflora complexes), while species in other groups (sections Racemosae and Eugenia) were dispersed, often in distant branches. Of these, the morphological characters provided the best supported for the groups, especially when combined with anatomical ones. This result don´t agrees with other studies, where anatomical were more relevant. The statistical methods used were useful in selecting characters that strengthen, groupings, principally in morphologically similar groups. We conclude that, as in other studies to characterize groups in the generic and specific levels in Myrtaceae, the segregation of groups is only possible by a combination of |