Geometric morphometrics to assess leaf variation in populations of Larrea divaricata Cav. (Zygophyllaceae) from Argentina.

WASSERMAN, ELIANA; SASSONE, AGOSTINA B.; LOPEZ, A.; MORELLO, S.

Online

Congreso; Botany Conference 2020; 2020

Botanical Society of America

Larrea divaricata Cav. (Zygophyllaceae), one out of five species of the genera, is a woody shrub distributed in Perú, Bolivia, Chile, and Argentina. In Argentina, it is a dominant species in the xerophytic environments of Monte and dry Chaco. Anthropical actions led to salinization, soil degradation, and the consequent loss of biodiversity in these fragile environments. Also, in the past few years has been an increase in demands on L. divaricata as a resource for cosmetic applications and pharmaceutical uses, for both, western and traditional practices. These have resulted in a current threatening situation of the species. Taxonomic studies, including some morphological and cytogenetical approaches, had been made in the past, and leaf morphology had been pointed out as a diagnostic character to separate species and recognize hybrids. Here we explore the morphological variation within 11 populations of L. divaricata with a geometric morphometric approach. Leaves were scanned and the extraction of outline coordinates from leaves images led to a dataset of 497 outlines. There were nine harmonics selected to reconstruct the leaf shape and two landmarks were used to align samples. Elliptic Fourier Analysis (EFA) was performed on these outlines. Further methods were carried on R using Momocs package. A principal component analysis (PCA) was used to summarize the information. To study phylogenetic relationships among populations, we performed Bayesian inference from a plastidial marker using MrBayes and CIPRES Science Gateway. PCA of the data matrix revealed that more than 90 % of the shape variance was represented by PC1 and PC2. This was influenced by leaf width and deepness of the incision, respectively. There was an unequal morphospace occupation, as some populations showed little variation and a compact distribution along the morphospace, while others were distributed in a big area showing high intrapopulation morphological variation. To characterize the evolutionary patterns of leaf shape diversity, we projected the topology of the phylogenetic tree onto the morphospace constructed with the mean shape of each specimen. Phylogenetic branches in morphospace did not tend to radiate. Instead, they crossed one another repeatedly and closed related specimens were often found in different places in the morphospace. Our preliminary results suggested that variation of leaf shape might be higher within populations than among them. Due to leaf morphology can be influenced by environmental variables, our next steps are, broadening the sampling and testing if there is a correlation between morphology, phylogeny, and environmental variables.