Collecting Venezuelan Stylosanthes species

A report on recent activities to collect representative germplasm samples of Stylosanthes species native to Venezuela is presented. The main objective was to obtain original seeds for phylogenetic studies and, at the same time, to contribute to safeguarding Stylosanthes diversity, in view of the increasing danger of genetic erosion. Seeds of 33 samples, comprising all 11 Venezuelan Stylosanthes species, were collected: S. angustifolia, S. capitata, S. gracilis, S. guianensis, S. hamata, S. humilis, S. scabra and S. viscosa plus the endemic and little-known S. falconensis, S. sericeiceps and S. venezuelensis. Populations of the latter 3 species were small and difficult to locate; these species must be considered endangered and require further attention by genetic resources specialists. Furthermore, assessment of their agronomic and forage potential, in particular, S. falconensis and S. sericeiceps, is warranted.


Introduction
Stylosanthes is considered one of the most important tropical forage legume genera (Burt et al. 1983;Chakraborty 2004).Primarily because most of its species are adapted to infertile soils and drought, it has received considerable attention by tropical pasture scientists and is well represented in the major forage genetic resources collections (Stace and Edye 1984;Cook et al. 2005).
Venezuela is an important center of diversification of Stylosanthes; 11 species have been reported, 3 of them being endemic to the country (Calles and Schultze-Kraft 2010a).During 1973During −1986, a significant collection of more than 600 accessions of Venezuelan Stylosanthes germplasm was assembled (Schultze-Kraft 1991).
A recent study of diversity of Venezuelan Stylosanthes (Calles 2012) has shown the convenience of molecular marker analyses to further describe genetic variation and species boundaries and relationships.For this, original seed samples (those collected at the original sites) should be used, rather than material that has undergone various cycles of seed multiplication in gene bank plant nurseries, with the concomitant risk that genetic integrity of the accession may be distorted (Chebotar et al. 2003).
This risk was recently pointed out by Garcia et al. (2013) for several Stylosanthes species, which usually are considered to be predominantly self-pollinating.They reported outcrossing rates of up to 31%.The Venezuelan diversity study also highlighted the increasing destruction of native habitats due to land use changes in the country, and subsequent endangerment of Stylosanthes populations.
This paper reports on recent (2006−2008) collecting activities in Venezuela to obtain representative germplasm samples of all 11 Venezuelan Stylosanthes species, with the objective of: (1) making original seeds available for phylogenetic analyses; and, at the same time, (2) safeguarding Stylosanthes diversity in view of the danger of genetic erosion.

Methodology
Three collection sites for each Venezuelan Stylosanthes species were predefined, based on exhaustive herbarium studies using specimens from 13 international and 22 Venezuelan herbaria (Calles and Schultze-Kraft 2010a).In order to cover as much intraspecific diversity as possible, the sites identified were in environments as contrasting as possible and/or with the largest possible geographic distance between them.
At each site, we collected as many mature seeds as possible from the target populations; if possible, 10 or more plants per population were sampled and the seeds bulked.Soil samples from the 0−10 cm horizon were taken and analyzed to determine texture, pH and phosphorus (P) content.
Geographical location and elevation were recorded with a GPS device and later refined using the geographical information program Google Earth (http://earth.google.com).Climatic information (i.e.mean annual rainfall, number of dry months and annual mean temperature) at collection sites was obtained from the software DIVA-GIS (Hijmans et al. 2005); a month with less than 60 mm of rainfall was considered dry.Vegetation/ecozone classification at collection sites was performed by locating the collection sites manually on the vegetation map of Venezuela (Huber and Alarcón 1988).
In compliance with national regulations governing access to Venezuelan plant genetic resources, a collection permit was obtained from the Ministry of Environmental Affairs (MINAMB, its Spanish acronym).

General
A total of 33 seed samples were collected.Except for S. capitata, S. scabra and S. sericeiceps (4 samples each), S. viscosa (2 samples) and S. venezuelensis (1 sample), 3 samples of each Venezuelan species were obtained as intended.The samples were incorporated in the collection of the Venezuelan Genebank for Forage Legumes (BGLFV, its Spanish acronym) at INIA-Anzoátegui, El Tigre; it is intended to deposit a duplicate at the Genetic Resources Program of CIAT, Cali, Colombia.Furthermore, at least 2 voucher herbarium specimens from plants with flowers and fruits were taken from each population and deposited at the Herbario Nacional de Venezuela (VEN).Passport data of the collected material are provided in Table 1.

Climate, soil and vegetation
Collections were made from an elevation range of 10−1,871 masl, with the only exception being a site at 2,481 masl, where a population of S. sericeiceps (CAL014) was collected in the Andes region, municipality of Mérida (Table 1).Most sampled populations were in areas with mean annual rainfall (MAR) ≥1,000 mm (Table 2) and with a dry season of up to 6 months.Interestingly, all populations of S. falconensis and one S. viscosa population (CAL005) were collected in areas with MAR as low as 781 and 312 mm, respectively.Moreover, at the collection site of sample CAL005, no month registered more than 60 mm rainfall.Most species were collected in areas with annual mean temperature (AMT) between 20 and 28 °C (Table 2), but it is noteworthy that some samples of S. guianensis (CAL030, CAL032), S. hamata (CAL012), S. scabra (CAL028, CAL031) and all samples of S. sericeiceps were collected in areas with AMT <20 °C.
Savanna was the original vegetation type, where the largest number of samples originated (12 samples), followed by: thorn forest and semi-deciduous forest (5 samples each); deciduous forest (4 samples); littoral vegetation and evergreen forest (3 and 2 samples, respectively); and gallery and cloud forests (1 sample each).In many cases, habitat destruction, as a consequence of land use changes, was observed.All species were growing on extremely acid to slightly acid soils (Table 3), except for S. falconensis and S. venezuelensis (slightly alkaline to neutral).It is noteworthy that S. angustifolia, S. capitata, S. gracilis [formerly known as S. guianensis var.gracilis (Kunth) Vogel; Calles and Schultze-Kraft 2010b] and S. humilis consistently came from very strongly acid soils (classification according to SSDS 1993).Soil P content was low in most cases, except for S. guianensis, S. hamata and S. sericeiceps (medium to high P content).It was significant that only S. scabra was growing on a heavier textured soil (clay loam).
In general, climatic and soil conditions at collection sites are in agreement with what is known about ecological conditions where Stylosanthes species occur (Stace and Edye 1984).We draw attention to the very high site elevation of sample CAL014 (the endemic S. sericeiceps; above 2,400 masl; Figures 1A, B & C) and the very low rainfall at the site of CAL005 (S. viscosa; GIS-DIVA calculated MAR of 312 mm).Both are quite unusual for Stylosanthes species and germplasm from those sites might be a particularly interesting resource for selection and/or breeding projects concerned with coldand drought-tolerance, respectively.1G & H), is particularly rare.In spite of our intensive search, we found only 1 small population in strongly intervened vegetation within the Caracas urban area.
Continued collecting of germplasm of these endemic species is required in order to safeguard their genetic diversity.In the case of S. venezuelensis, protection measures might be needed to avoid the extinction of the species.An initial step would be to multiply the collected material and disperse the seed into the species' native habitats.

Conclusion and Outlook
Original seed of all Venezuelan Stylosanthes species is now available for the intended phylogenetic studies.With the material collected, we might have made only a minor contribution to increased genetic diversity within the 8 non-endemic species, since, with the exception of S. angustifolia, they had been fairly well sampled in the 1970s and 1980s (Schultze-Kraft 1991).Rather, the significance of the collection reported here lies in the fact that germplasm of the 3 endemic species, S. falconensis, S. sericeiceps and S. venezuelensis, is available for the first time.These species are rare and must be considered endangered.Efforts to collect more material to safeguard genetic diversity are indicated, particularly in view of the destruction of the species' habitats.
Assessment of the agronomic and forage potential of the endemic species, particularly of S. falconensis and S. sericeiceps, is suggested.

Table 1 .
Passport data of Stylosanthes germplasm samples collected in Venezuela. www.tropicalgrasslands.info

Table 3 .
Some soil characteristics (mean values) at collection sites of Venezuelan Stylosanthes species.