Disentangling Species Delineation and Guiding Conservation of Endangered Magnolias in Veracruz, Mexico

Abstract

The Mexican state of Veracruz has suffered very high deforestation rates in the last few decades, and despite the establishment of protected areas and conservation projects, primary forest is now mainly persisting in mostly small, scattered, fragmented remnants. New species of Magnolia section Talauma in this state have been described with little to no reference to the already existing ones, potentially resulting in over-splitting, obscuring their taxonomic delineation and conservation status, and consequently conservation programs. To study the conservation units and their genetic diversity, we here employ 15 microsatellite markers on a highly representative sampling of 254 individuals of what are presumed to be five Magnolia species. The results support at least three species and maximum five main conservation units. We propose downgrading the latter to four, given morphological, ecological, demographical, and geographical considerations. Two out of the three sympatrically occurring species in the rainforest in the Los Tuxtlas volcanic area have weak genetic evidence to be considered separate species. Similarly, the individuals in the Sierra de Zongolica in central Veracruz, who bear a very high morphological and genetic similarity to Magnolia mexicana, have weak genetic evidence to be recognised as a separate species. Nonetheless, the individuals could be identified as Magnolia decastroi based on morphology, and further research including the full range of this species is recommended.

Keywords: IUCN Red List conservation status; Magnoliaceae; SSR; Talauma; conservation units; genetic diversity; microsatellite; neotropical trees.

Figure A1

Delta K plot for the Northern Zone.

Figure A2

Mean Ln(K) plot for the Northern Zone.

Figure A3

Delta K plot for the Southern Zone.

Figure A4

Mean Ln(K) plot for the Southern Zone.

Figure 1

Morphology of Magnolia sect. Talauma in Puebla and Veracruz, Mexico. (A) Magnolia decastroi: 1. Tree, 2. Voucher, 3. Fruit; 1–3 By F.A. Aldaba Núñez, 2019. (B) Magnolia lopezobradorii: 1. Tree, 2. Voucher, 3. Fruit; 1–2 By F.A. Aldaba Núñez, 2019, 3 By E.M. Martínez Salas, 2019. (C) Magnolia mexicana: 1. Tree, 2. Voucher, 3. Fruit, 4. Flower; 1–4 By F.A. Aldaba Núñez, 2019. (D) Magnolia sinacacolinii: 1. Tree, 2. Voucher, 3. Fruit, 4. Flower; 1–2 By F.A. Aldaba Núñez, 2019, 3–4 By E.M. Martínez Salas, 2019. (E) Magnolia zoquepopolucae: 1. Tree, 2. Voucher, 3. Fruit; 1–3 By F.A. Aldaba Núñez, 2019.

Figure 2

Map of study zones showing their natural regions; in Puebla and Veracruz states.

Figure 3

Structure bar plots of Magnolia sect. Talauma individuals in Puebla and Veracruz, Mexico. A–E: analyses on the complete dataset of 18 localities. F–H: analyses of the 15 wild localities only. (A) The delta K plot for the complete dataset. (B) The mean Ln(K) plot for the complete dataset. (C) Representative bar plot (out of ten replicates) for K = 2. (D) Representative bar plots for K = 4. The upper bar plot is found in 9/10 replicates, the lower bar plot in 1/10 replicates. (E) Representative bar plots for K = 5. The upper bar plot is found in 8/10 replicates, the lower bar plot in 2/10 replicates. (F) The delta K plot for the wild localities only. (G) The mean Ln(K) plot for the wild localities only. (H) Representative bar plot (out of ten replicates) for K = 5.

Figure 4

Discriminant Analysis of Principal Components (DAPC) of Magnolia sect. Talauma individuals in Puebla and Veracruz, Mexico. The axes represent the first two linear discriminants. The upper left graph (principal component analysis (PCA) eigenvalues) inset displays the variance explained by the principal component axes used for DAPC and the bottom-right inset (DA eigenvalues) displays in relative magnitude the variance explained by the two discriminant axes plotted. (A) DAPC graph of the complete dataset analysis, 150 principal components (PCs) retained. (B) DAPC graph of the M. decastroi and M. mexicana localities, 50 PCs retained. (C). DAPC graph of the M. lopezobradorii and M. zoquepopolucae localities, 5 PCs retained. (D) DAPC graph of the wild dataset, 150 PCs retained. (E) DAPC graph of the M. decastroi and M. mexicana wild localities, 60 PCs retained.

Figure 5

Pairwise F_ST and D_JOST values between the 18 localities of Magnolia species in Puebla and Veracruz, Mexico, visualized as a heatmap, C=cultivated. (A) Pairwise F_ST values. (B) Pairwise D_JOST values. Locality metadata can be found in Table 5.

Figure 6

Pairwise F_ST and D_JOST values between the five Magnolia species studied in Puebla and Veracruz, Mexico, visualized as a heatmap. W = wild, C = cultivated. (A) Below the diagonal pairwise F_ST values are tabulated. (B) Pairwise D_JOST values are tabulated. Locality metadata can be found in Table 5.