SciELO - Scientific Electronic Library Online

vol.23 issue2Development of pollen grain in yellow passion-fruit (Passiflora edulis f. flavicarpa; Passifloraceae)Multivariate analysis as a tool for measuring the stability of morphometric traits in Lycopersicon plants from in vitro culture author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Genetics and Molecular Biology

Print version ISSN 1415-4757On-line version ISSN 1678-4685

Genet. Mol. Biol. vol.23 n.2 São Paulo June 2000 




Solange B. Tedesco1, Miguel Dall'Agnol1, Maria Teresa Schifino-Wittmann1 and José F.M. Valls2
1Departamento de Plantas Forrageiras e Agrometeorologia, Faculdade de Agronomia, Universidade Federal do Rio Grande do Sul, Av. Bento Gonçalves, 7712, Caixa Postal 776, 91501-970 Porto Alegre, RS, Brasil. Send correspondence to S.B.T.
Rua Conde de Porto Alegre, 1368, ap. 301, 97015-110 Santa Maria, RS, Brasil. E-mail:
2Embrapa Recursos Genéticos e Biotecnologia, 70770-900 Brasília, DF, Brasil.




Mode of reproduction was studied in 15 species of Adesmia DC. (Leguminosae). In six species, three treatments were used: mutual pollination, mechanical stimulation and control. Fifty-four plants of these six species were grown in a greenhouse, individually isolated in nylon screen boxes. Flowers were labelled and submitted to the different treatments. In addition, the frequency of spontaneous self-pollination in the absence of pollinators was studied in 200 plants of nine other species. These 200 plants were kept in a greenhouse, which avoided contact with any possible pollinator. Adesmia bicolor, A. muricata, A. punctata and A. riograndensis produced seed both by cross- and self-pollination. Adesmia punctata and A. riograndensis need mechanical stimulation for self-pollination. Adesmia incana reproduced by self-pollination; however, the possibility of cross-pollination cannot be totally ruled out. Adesmia tristis reproduced mainly by cross-pollination and a mechanism of self-incompatibility is suggested. Among the nine species that were not exposed to pollinators, A. securigerifolia produced a large amount of seed, indicating that it is a self-pollinating species. Adesmia arillata, A. araujoi, A. ciliata, A. psoraleoides, A. rocinhensis, A. reitziana, A. sulina and A. vallsii did not produce any seed under the experimental conditions, suggesting that they are cross-pollinated or that they need mechanical stimulation to reproduce.




The South American genus Adesmia DC. belongs to tribe Adesmieae (Benth.) Hutch. and comprises about 200 species. Seventeen species and one variety, annual or perennial, herbaceous or shrubby, are found in Brazil. All of them are endemic to the southern region, which includes the States of Rio Grande do Sul, Santa Catarina and Paraná (Miotto and Leitão-Filho, 1993).

Many of the Brazilian species are promising as forage crops and have attracted the interest of researchers. Several species of Adesmia from southern Brazil present good winter growth, when there is normally lack of cattle forage. They are well adapted to the regional environment, are widespread and have a high nutritional value. Crude protein percentages range from 6.9 to 17.5 in A. tristis, 18.6 in A. latifolia, 17.9 to 21.5 in A. ciliata, 19.7 in A. psoraleoides and 23.4 in A. punctata (Dall´Agnol and Gomes, 1994). High crude protein values, as well as good in vitro organic matter digestibility (IVOMD), were also found for A. latifolia, A. tristis and A. punctata, by Scheffer-Basso (1999).

Germplasm accessions of the Brazilian species of Adesmia have been stored at Embrapa Recursos Genéticos e Biotecnologia/CENARGEN, Brasília, DF, since 1984, when intensive efforts to collect and conserve this germplasm were initiated. Germplasm characterization, including knowledge about the mode of reproduction, is essential for any further research. The reproductive success of a given species depends on the reproductive strategy used to ensure perpetuation.

The determination of the mode of reproduction should be one of the first steps in germplasm characterization, since the process of collecting, multiplication, and further plant breeding procedure depends on how each species reproduces (Valls, 1988). Depending on the mode of reproduction, different plant breeding methods should be used (Poehlman, 1965; Borém, 1997). Native legumes are important components of many natural pastures in several countries and regions, such as in southern Brazil; nevertheless the reproductive processes are unknown for many species (Izaguirre et al., 1994). Very little is known about the mode of reproduction of Adesmia (Valls, 1984). Adesmia latifolia is considered a versatile species, allowing for self and cross fertilization (Tedesco et al., 1998), but there is no published information about the other Brazilian species. Visual observation has shown that the flowers are visited and pollinated by insects.

We determined the mode of reproduction of 15 Brazilian species of Adesmia. This is expected to be useful for plant breeding.



Fifty-four plants of six species (Adesmia bicolor, A. incana, A. muricata, A. punctata, A. riograndensis and A. tristis (Table I)) were grown in pots in a greenhouse, individually isolated in nylon screen boxes of 0.5 x 0.5 or 0.5 x 1.0 m, depending on plant height. The effects of three treatments were tested: T1: pairs of plants, which were mutually pollinated by reciprocally transferring pollen from one to the other. Flowers were labelled and manually pollinated with a small cuneiform piece of cardboard, suitable to pollinate without damaging the flowers; T2: flowers were labelled and mechanically stimulated by manually pressing the keel with a piece of cardboard as in T1, one per flower, thus liberating the anthers and releasing pollen. This artificial tripping simulates the action of the pollinating insect; T3: flowers were labelled and left undisturbed. The number of flowers in each treatment varied among species and among treatments, depending on the availability of exposed flowers. T1 and T2 were carried out from 10:00 a.m. to 1:00 p.m. for several days. The number of fruits produced was counted for each plant and treatment, and compared by a c2 test.



For nine other species, Adesmia araujoi, A. arillata, A. ciliata, A. psoraleoides, A. securigerifolia, A. sulina, A. reitziana, A. rocinhensis and A. vallsii, plants of 37 accessions (Table II) were kept in a greenhouse, totally isolated from pollinators. After flowering, fruiting and seed formation were observed. Adesmia securigerifolia, A. riograndensis and A. muricata are annual, while the other species are perennial (Miotto and Leitão-Filho, 1993).




Manual stimulation was the most efficient treatment for A. riograndensis (Table III), suggesting that this species is autogamous, but needs mechanical stimulation. Percentages of cross-pollination are only estimates, since there was no emasculation, but just mutual pollen transfer, so that some of the fruits may have been formed by self-pollination. As A. riograndensis is annual, a higher percentage of autogamy would be expected. A. bicolor needed mechanical stimulation to set fruit, suggesting that it is a preferentially allogamous species (Table III), even considering that some of these fruits could have been formed by self-pollination. Similar results were found for A. punctata and A. tristis.



Versatility in mode of reproduction has been described for other native and exotic legumes, as for example, Trifolium riograndense (Becker et al., 1987), Desmodium intortum and D. uncinatum (Hutton, 1977), and Adesmia latifolia (Tedesco et al., 1998).

A. muricata is probably mainly cross-pollinating (Table III). In A. incana, no fruits were formed by T1 and T3 but 6.77% of fruits were formed by T2 (mechanical stimulation). The number of flowers treated was small (Table III), but the hypothesis that this species is an inbreeder that needs mechanical stimulation cannot be ruled out.

Apparently A. tristis, A. bicolor, A. incana and A. punctata need mechanical stimulation for self-pollination, which in nature is done by pollinating insects. There may be a self-incompatibility mechanism in A. tristis, as the percentage of fruits formed was small (Table III).

Self-incompatibility may hinder genetic breeding, avoiding selfing, but at the same time assures maintenance of high heterozygosis. It is genetically controlled (Nettancourt, 1997) and common in several economically important legumes of the genera Trifolium, Lotus, Melilotus and Phaseolus (Brewbaker, 1982). Several perennial, winter-producing forage legumes are self-incompatible (Paim, 1980).

Observations with the other nine species suggest that A. securigerifolia is an inbreeder, since it produced a large amount of seed by self-pollination (Table IV). The other species did not produce any seed at all in the absence of pollinators, suggesting that they are obligate outbreeders or need mechanical stimulation for selfing.



Data on mode of reproduction should be taken into account when sampling new populations, to assure a representative portion of the existing genetic variability, as well as when plant breeding experiments are designed. Studies with molecular markers may help to elucidate the mode of reproduction of these Adesmia species; they are being planned by our research group.




Foram estudadas 15 espécies do gênero Adesmia DC. (Leguminosae), quanto ao modo de reprodução. Em seis espécies do gênero Adesmia, o modo de reprodução foi determinado através de três tratamentos: polinização mútua, estímulo mecânico e controle. As 54 plantas submetidas aos tratamentos foram cultivadas em casa de vegetação e mantidas isoladas individualmente, através de armações de tela de náilon. As flores foram marcadas e submetidas aos distintos tratamentos. Adicionalmente, foram observadas 200 plantas de outras 9 espécies do mesmo gênero quanto à ocorrência de autofecundação na ausência de polinizadores. As 200 plantas apenas isoladas e observadas foram cultivadas em vasos e mantidas em casa de vegetação telada, sem acesso de polinizadores. Os resultados mostraram que, das espécies investigadas pelos três tratamentos, A. bicolor, A. muricata, A. punctata e A. riograndensis são versáteis, pois permitiram a reprodução por fecundação cruzada e autofecundação. As duas últimas, para se autofecundarem, necessitaram de estímulo mecânico. Adesmia incana se reproduziu por autofecundação, mas não se descarta a possibilidade de ocorrer fecundação cruzada. Adesmia tristis se reproduziu quase que totalmente por fecundação cruzada e é possível a ocorrência de mecanismos de autoincompatibilidade. Das nove espécies apenas observadas, Adesmia securigerifolia se reproduziu por autofecundação, formando elevado número de sementes por planta. A. arillata, A. araujoi, A. ciliata, A. psoraleoides, A. rocinhensis, A. reitziana, A. sulina e A. vallsii não produziram sementes por autofecundação espontânea. Estas espécies são de fecundação cruzada ou necessitam do estímulo do polinizador.




Becker, L.C.M., Schifino-Wittmann, M.T., Paim, N.R. and Riboldi, J. (1987). Observations on the mode of reproduction of Trifolium riograndense Burkart (Leguminosae). Ciênc. Cult. 39: 304-306.         [ Links ]

Borém, A. (1997). Melhoramento de Plantas. Editora UFV, Viçosa.         [ Links ]

Brewbaker, J.L. (1982). Systematics, self-incompatibility, breeding systems, and genetic improvement of Leucaena species. In: Leucaena Researchs in the Asian Region. Proceedings of Workshop, Singapore, pp. 17-22.         [ Links ]

Dall'Agnol, M. and Gomes, K.E. (1994). Qualidade de acessos de leguminosas nativas do gênero Adesmia. Anais da XXXI Reunião da Sociedade Brasileira de Zootecnia, Maringá, PR, pp. 653.         [ Links ]

Hutton, E.M. (1977). Selection and breeding of tropical pasture legumes. In: Tropical Forage Legumes. FAO, Plant Prod. Prot. Series 2: 174-185.         [ Links ]

Izaguirre, P., Mérola, S. and Beyhaut, R. (1994). Seed ontogeny in Adesmia securigerifolia (Fabaceae- Adesmieae). Nord. J. Bot. 14: 547-556.         [ Links ]

Miotto, S.T.S. and Leitão-Filho, H.F. (1993). Leguminosae-Faboideae, gênero Adesmia DC. Bol. Inst. Biociênc. 52: 1-157.         [ Links ]

Nettancourt, D. (1997). Incompatibility in angiosperms. Sex. Plant Reprod. 10: 185-199.         [ Links ]

Paim, N.R. (1980). Reprodução e poliploidia no melhoramento de plantas forrageiras. In: Simpósio sobre Citogenética Vegetal e suas Aplicações no Melhoramento de Plantas,32a Reunião Anual da SBPC e XXVI Congresso Nacional de Genética, Rio de Janeiro, RJ.         [ Links ]

Poehlman, J.M. (1965). Mejoramiento Genético de las Cosechas. Limura-Willey, México.         [ Links ]

Scheffer-Basso, S.M. (1999). Caracterização morfofisiológica e fixação biológica de nitrogênio de espécies de Adesmia DC. e Lotus L. Doctoral thesis, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS.         [ Links ]

Tedesco, S.B., Dall'Agnol, M. and Schifino-Wittmann, M.T. (1998). Observações sobre o modo de reprodução em Adesmia latifolia (Spreng.) Vog. (Leguminosae). Ciênc. Rural 28: 141-142.         [ Links ]

Valls, J.F.M. (1984). Notas Sobre a Taxonomia, Disponibilidade de Germoplasma e Problemas para a Utilização Forrageira de Adesmia spp. no Sul do Brasil. EMBRAPA-CENARGEN, Brasília, DF (datilografado).         [ Links ]

Valls, J.F.M. (1988). Caracterização morfológica, reprodutiva e bioquímica de germo-plasma vegetal. In: Encontro sobre Recursos Genéticos, Jaboticabal, SP. Resumos, pp. 106-127.         [ Links ]


(Received January 12, 2000)

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License