SciELO - Scientific Electronic Library Online

vol.23 issue3Flora of seasonal deciduous montane forest at Serra das Almas, Ceará State, BrazilKnowledge and use of plants in a Caiçara community located on the southern coast of Rio de Janeiro State, Brazil author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Acta Botanica Brasilica

Print version ISSN 0102-3306On-line version ISSN 1677-941X

Acta Bot. Bras. vol.23 no.3 São Paulo July/Sept. 2009 



A new case of late-acting self-incompatibility in Capparis L. (Brassicaceae): C. jacobinae Moric. ex Eichler, an endemic andromonoecious species of the Caatinga, Pernambuco State, Brazil1


Novo registro de auto-incompatibilidade de ação tardia em Capparis (Brassicaceae): C. jacobinae Moric. ex Eichler, uma espécie andromonóica endêmica da Caatinga, PE, Brasil



Luis Miguel PrimoI,*; Isabel Cristina MachadoII

IUniversidade Federal de Pernambuco, Programa de Pós-Graduação em Biologia Vegetal, Recife, PE, Brasil
IIUniversidade Federal de Pernambuco, Departamento de Botânica, Recife, PE, Brasil (




We studied the reproductive system of Capparis jacobinae Moric ex Eichler (Brassicaceae), based on controlled hand-pollination and observation of pollen tube growth made in a fluorescence microscope. Of 105 self-pollinated flowers only one produced fruits (success = 0.95%), all the other flowers abscised at the same time, between the eighth and tenth day after anthesis. Nevertheless, self- and cross-pollinated pollen tubes reached the micropyle. The rate of penetrated ovules in self-pollinated flowers was lower during the first 24 h after pollination; thereafter this rate was similar between self- and cross-pollinated flowers for treatments of 48 h, 72 h and 96 h after pollination. In addition, we carried out two indirect estimates of the reproductive system, based on pollen/ovule and seed/ovule ratios, which resulted in typical features of xenogamous species. We concluded that C. jacobinae has a late-acting self-incompatibility system. This is the third record of this mechanism for the genus and the first for a species of Capparis endemic to the Caatinga. We suggest that this self-incompatibility system may occur in other species of the same genus and family.

Key words: andromonoecy, Capparaceae, pollen-ovule ratio, pollen tube growth, sexual system


O sistema reprodutivo de Capparis jacobinae Moric. ex Eichler (Brassicaceae), uma espécie endêmica da Caatinga no Brasil, foi analisado através de polinizações controladas e observações do desenvolvimento dos tubos polínicos por meio de microscopia de fluorescência. De 105 flores autopolinizadas, apenas uma formou fruto (sucesso= 0,95%), ocorrendo a abscisão das demais em um intervalo de tempo uniforme, entre o oitavo e o décimo dia após a antese. Entretanto, tanto tubos polínicos procedentes de autopolinização quanto de polinização cruzada penetraram na micrópila, sendo a taxa de óvulos penetrados menor em flores autopolinizadas durante as 24 horas posteriores à polinização, igualando-se entre os dois tratamentos para 48, 72 e 96 horas posteriores à polinização. Além disso, foram realizadas duas estimativas indiretas do sistema reprodutivo, baseadas na razão pólen/óvulo e semente/óvulo, cujos resultados foram característicos de espécies xenógamas. Conclui-se, portanto, que C. jacobinae é auto-incompatível, apresentando sistema de auto-incompatibilidade de ação tardia, sendo este o terceiro registro deste mecanismo encontrado para o gênero e o primeiro encontrado em uma espécie de Capparis endêmica da Caatinga. É sugerido que este sistema de auto-incompatibilidade seja encontrado em outros representantes do mesmo gênero e família.

Palavras-chave: andromonoicismo, Capparaceae, razão pólen/óvulo, sistema sexual, tubos polínicos




The Caatinga covers an area of about 1,000,000 km2 mainly located in northeastern Brazil. It is characterized by low rainfall (usually below 800 mm/year) and by the occurrence of xerophytic deciduous vegetation, in general thorny, with succulent plants, varying from shrubs to trees with a seasonal herbaceous stratum (Sampaio 1995; Leal et al. 2003). The genus Capparis L. has 250 species (Mabberley 1993) and is represented in Brazil by about fifteen species that occur mainly in the Caatinga and Restinga environments (Barroso 1978; Fuks & Costa e Silva 2000).

The available data on the reproductive system of the genus Capparis indicate that its members are mainly self-incompatible (about 85% of the studied species). Only C. flexuosa L. (Ruiz-Zapata & Arroyo 1978) and C. hastata L. (Primo 2004; Tab. 1) were considered to be self-compatible, though in the Restinga (Carvalho 2002) and in semi-deciduous forests (Primo 2008) C. flexuosa was considered to be self-incompatible.

Observations on pollen tube growth in this genus were carried out for only two species: C. pittieri Standl. (Seavey & Bawa 1986) and C. retusa Grisenb. (Bianchi & Gibbs 2000), showing that the pollen tubes formed through self-pollination penetrated the ovule's micropyle. However, most of these ovules did not produce seeds, what was interpreted as a recognition-rejection of pollen tubes coming from self-pollination only when they were already inside the ovules. Species with this mechanism have a reproductive system with late-acting self-incompatibility (LSI) (sensu Seavey & Bawa 1986). This system, like conventional ones, such as sporophytic and gametophytic (where inhibition occurs in the stigma and style, respectively), may be distributed among species from phylogenetically related groups, such as genera or families (Gibbs & Bianchi 1999).

Taking into account this feature discussed by Gibbs & Bianchi (1999) and the two preceding occurrences of LSI in the genus, the present study aims to describe the reproductive system of C. jacobinae Moric. ex Eichler and to determine the nature of the incompatibility system through analysis of pollen-tube growth.


Material and methods

Study areas and species - Field work was carried out during two flowering cycles of C. jacobinae, from September 2002 to April 2003 and from October to December 2003, in two natural populations in the state of Pernambuco, Brazil: one in Arcoverde municipality (8º25'S, 37º02W) and the other in Vale do Catimbau, Buíque municipality (8º67'S, 37º01W). Each population was composed of approximately 40 individuals, distributed over an area of 10,000 m2. The individuals of both populations formed groups of two or three, next to each other and joined by their canopies, or they grew in isolation. The distance between these groups or between isolated individuals was about 10 m.

Capparis jacobinae is widely distributed in the Caatinga and is endemic to this biome (Costa e Silva 2002). It is a shrubby species and has polystaminate brush-type flowers with an andromonoecious sexual system (Primo 2004; Primo & Machado unpublished data). The anthesis process is slow: it starts at 11:00 h and flowers are completely opened around 16:00 h. The flowering period lasts from August to March (Primo 2004; Primo & Machado unpublished data).

Voucher specimens of C. jacobinae were deposited in the Herbarium Geraldo Mariz at the Botany Department, Universidade Federal de Pernambuco (UFP 33764).

Reproductive system - To investigate the plant's reproductive system we carried out experiments using hermaphrodite flowers, which consisted of a control and four treatments: 1) Control: we marked hermaphrodite flowers (n = 99) of different individuals to test for the production of fruits under natural conditions; 2) Apomixis: we emasculated flowers of different individuals in pre-anthesis phase (n = 32), which were maintained bagged for one day; 3) Spontaneous self-pollination: we bagged flowers of different individuals in preanthesis phase (n= 33), and maintained them bagged for one day; 4) Manual self-pollination: we bagged flower buds of different individuals (n = 105) and, after flower opening and anther dehiscence, they were hand-pollinated with their own pollen and once more bagged and maintained this way for one day; 5) Manual cross-pollination: we emasculated and bagged flower buds of different individuals with indehiscent anthers (n = 72). Thereafter, these flowers received pollen from other individuals, and were bagged again. Each flower was pollinated with pollen grains from only one kind of donor flower, hermaphrodite or staminate, since there was no difference in the viability of pollen grains from these two flower types. The distance between individuals used in crossbreeding was > 15 meters. In all treatments, the hermaphrodite flowers were subsequently monitored to verify whether fruits were produced.

Manual pollination for the tests was carried out between 19:00 and 22:00 h. On non-pollinated staminate and hermaphrodite flowers, as well as on flowers submitted to self-pollination, we evaluated abscission time.

Aiming to test for the possibility of pollination leading to fruit production near the end of anthesis, we carried out manual cross-pollination between 04:30 and 05:00 h (n = 10), and thereafter we recorded whether fruits were produced.

In all treatments, the flowers were isolated from their visitors with semi-permeable paper bags closed with plastic-coated wires. We emasculated the flower buds with a sharp tip scissors. Anthers were cut immediately before dehiscence and filaments were left intact. Manual pollination was carried out by direct contact of the dehiscent anthers with the stigma.

We calculated the index of self-incompatibility (ISI) based on Bullock (1985) analyzing the percentage of fruit production from flowers submitted to manual self-pollination and cross-pollination. Fruits obtained in each treatment were also evaluated in terms of number of seeds.

A subsample of self-pollinated hermaphrodite flowers and another subsample of cross-pollinated flowers were fixed in FAA70 24, 48, 72 and 96 h after manual pollination (n = 4 to 6 pistils for each treatment and time). On these fixed flowers, we studied the percentage of ovules penetrated not only by pollen tubes originating from self-pollination but also by pollen tubes originating from cross-pollination. For this analysis we used fluorescent microscopy and staining with aniline blue (Martin 1959). We used χ2 (Sokal & Rohlf 1995) to compare simultaneous treatments.

We tested for the viability of pollen grains in 10 hermaphrodite flowers and in 10 staminate flowers of five individuals, all collected just before anthesis, by staining the cytoplasm with acetic carmine 2% (Radford et al. 1974).

We also carried out two indirect estimates of the reproductive system, based on pollen/ovule and seed/ovule ratios, as suggested by Cruden (1977) and Wiens (1984), respectively. For the first estimate, we collected flower buds (n = 10 hermaphrodite flowers and n = 10 staminate flowers) of different individuals and fixed them in ethanol 70%. We then determined the average number of ovules per ovary and the average number of pollen grains per anther and per flower. To estimate the number of pollen grains, we placed the contents of all six anthers from 20 flower buds in pre-anthesis in a solution of 1 ml lactic acid and glycerin (3:1), following Lloyd (1972). Using this solution, we counted pollen grains using a Neubauer chamber, following Maeda (1985). The number of pollen grains per anther was multiplied by the number of stamens of each flower (Primo 2004; Primo & Machado unpublished data). Finally, we calculated the pollen/ovule ratio with the following formula:


s = staminate flowers
h = hermaphrodite flowers
ps = total number of pollen grains per staminate flower
ns = total number of staminate flowers
R = ratio between staminate and hermaphrodite flowers [= 2.34 (Primo 2004; Primo & Machado unpublished data)]
ph = total number of pollen grains per hermaphrodite flower
nh = total number of hermaphrodite flowers
NO = number of ovules per flowers

The estimate of female fertility, i.e. the percentage of ovules that produced seeds (seed/ovule ratio), was calculated as the average number of seeds per fruit of the control group divided by the average number of ovules per flower (Wiens 1984).


Results and discussion

The results of controlled pollination experiments (Tab. 2) indicate that C. jacobinae is self-incompatible, with ISI = 0.013, since out of 105 self-pollinated flowers, only one produced fruits (success = 0.95%), which is consistent with the data of Ruiz-Zapata & Arroyo (1978). The lack of fruit production by apomixis and spontaneous self-pollination indicates that C. jacobinae depends on pollen vectors for reproduction. Moreover, the lack of fruit production through manual cross-pollination carried out at the end of anthesis (04:30-05:00 h.) suggests that these vectors have crepuscular/nocturnal activity, as observed by Primo (2004) and Primo & Machado (unpublished data).

Because Capparis jacobinae has staminate and hermaphrodite flowers on the same individual, this is the second record of andromonoecy for the genus (Primo 2004; Primo & Machado unpublished data); the first was C. pittieri (Bawa & Beach 1981). The occurrence of andromonoecy in the family has already been mentioned for Cleome (Brassicaceae), and there are records of trimonoecy in C. rosea (Carvalho 2002) and C. spinosa (Machado et al. 2006). Andromonoecy has been interpreted as a favorable sexual system when the pollination system requires huge amounts of pollen. This sexual system would have been developed in the plants' life history to increase the level of cross-pollination and recombination through higher resource allocation on paternal reproductive success (pollen donation) (Heithaus et al. 1974; Bawa & Beach 1981; Ramirez et al. 1984). Although the trimonoecious and andromonoecious individuals of Cleome spinosa are self-incompatible (Machado et al. 2006), Ruiz-Zapata & Arroyo (1978) recorded a strong association between andromonoecy and self-incompatibility in another tropical plant community.

The viability of pollen grains in Capparis jacobinae is high, not only in bisexual flowers (99.2 %) but also in staminate flowers (99.8 %), showing that both kinds of flower may be acting as pollen donors. Regarding the female function, there is an average of 30.5 ovules per hermaphrodite flower, with the P/O ratio = 108647.37 and the seed/ovule ratio = 0.44. According to Cruden (1977), this high P/O ratio is characteristic of xenogamous species. In the same way, according to the criterion of Wiens (1984), the seed/ovule ratio varies from 0 to 1, where the lowest values belong to allogamous species. A ratio of 0.44 would indicate that the seeds are produced mainly by cross-pollination. Bullock (1985) points out that a species is self-incompatible when it shows values of ISI < 0.25. Therefore, C. jacobinae has at least three characteristics that classify it as self-incompatible.

In Capparis jacobinae, both the ovules of flowers submitted to self-pollination and the ovules of flowers submitted to cross-pollination were reached by pollen tubes (Fig. 1-4). The abscission time in flowers submitted to self-pollination was similar; all fell between the eighth and the tenth day after anthesis, except for two developing fruits, from the same individual, that fell 23 and 36 days after self-pollination (Fig. 5). Otherwise, the abscission of staminate and hermaphrodite flowers that have not been pollinated occurred between the third and the fifth day after anthesis (Fig. 5). Different from endogamous depression, in which the accumulation of many lethal recessive alleles causes the abscission of self-pollinated flowers, late-acting self-incompatibility (LSI) is an active recognition-rejection mechanism that is genetically controlled by a few loci (Lipow & Wyatt 1999; 2000).



However, we frequently observed chaotic growth of pollen tubes in pistils of self-pollinated flowers, which go in several directions before penetrating the micropyle (Fig. 1). The percentage of penetrated ovules differs (χ2 = 4.53; p= 0.03) among treatments during the first 24 h after pollination, when there are a higher number of ovules penetrated by pollen tubes originating from cross-pollination (Tab. 3).

Therefore, we recorded three reproductive features in Capparis jacobinae that place it in a late-acting self-incompatibility system: first, the abscission of self-pollinated flowers occurs even if the pollen tubes reach the embryonic sacs (cf. Fig. 1 and 2), a phenomenon that does not exist in the sporophytic and gametophytic systems. Similar results were observed in Capparis pittieri and in C. retusa (Seavey & Bawa 1986; Bianchi & Gibbs 2000), two species that exhibit LSI.

Second, the difference between the passive mechanism of endogamous depression and the mechanism of active rejection of LSI is that the abscission of all self-pollinated flowers occurs over a short period and in all populations (Seavey & Bawa 1986; Lipow & Wyatt 1999). In Capparis jacobinae we observed in both populations, during the two flowering episodes, that the self-pollinated flowers fell in a well-defined period, between the eighth and the tenth day (Fig. 5).

The third evidence to be considered is the rhythm of ovule penetration by pollen tubes after self-pollination and cross-pollination (Tab. 3), which is similar to other species that have the LSI mechanism, such as Capparis retusa (Capparaceae) (Bianchi & Gibbs 2000), Chorisia chodatii Hassl., C. speciosa St. Hill. (Bombacaceae), Tabebuia caraiba (Mart.) Bur., T. ochracea (Cham.) Standl. (Bignoniaceae) (Gibbs & Bianchi 1992), Dolichandra cynanchoides Cham., Tabebuia nodosa (Grisenb.) Grisenb. (Bignoniaceae) (Gibbs & Bianchi 1999) and Spathodea campanulata Beauv. (Bignoniaceae) (Bittencourt et al. 2003).

Therefore, because Capparis jacobinae is the third record of LSI within the genus, it reinforces the hypothesis that the late-acting self-incompatibility mechanism could be distributed among species from phylogenetically close groups (Gibbs & Bianchi 1999). New studies with this approach on Capparis species and on the family Brassicaceae may reveal new occurrences of LSI.



Prof. Dr. Peter Gibbs (University of St. Andrews - Scotland) made valuable suggestions during the development of this study. Dr. Maria Bernadete Costa e Silva identified the plant vouchers. Dr. Ana Virgínia Leite (UFPE) and MSc. Tarcila Nadia (UFPE) helped in lab work. Genivaldo da Silva helped in fieldwork. Two anonym reviewers made valuable comments. CAPES and CNPq granted a scholarship (Program PEC/PG) to the first author and a productivity fellowship to the second author.


Bibliographic references

Aizen, M.A. & Feinsinger, P. 1994. Forest fragmentation, pollination and plant reproduction in a Chaco dry forest, Argentina. Ecology 75: 330-351.         [ Links ]

Barroso, G.M. 1978. Sistemática de angiospermas do Brasil. v. I LTC/Edusp.         [ Links ]

Bawa, K.S. & Beach, J.H. 1981. Evolution of sexual systems in flowering plants. Annals of the Missouri Botanical Garden 68: 254-274.         [ Links ]

Bianchi, M.B. & Gibbs, P.E. 2000. Late-acting self-incompatibility in Capparis retusa (Capparaceae), a species of Chaco woodland in NE Argentina. Revista Brasileira de Botânica 23: 393-398.         [ Links ]

Bittencourt, N.S.; Gibbs, P.E. & Semir, J. 2003. Histological study of post-pollination events in Spathodea campanulata (Bignoniaceae), a species with late-acting self-incompatibility. Annals of Botany 91: 827-834.         [ Links ]

Bullock, S.H. 1985. Breeding systems in the flora of a tropical deciduous forest in Mexico. Biotropica 17: 287-301.         [ Links ]

Carvalho, H.A.L. 2002. Capparaceae Juss. na Restinga de Maricá, RJ - Estudo sobre a biologia da reprodução de Capparis lineata Domb. ex Pers., C. flexuosa L. e Cleome rosea Vahl ex DC. Tese de Doutorado, Universidade Federal de Rio de Janeiro.         [ Links ]

Costa e Silva, B. 2002. Distribuição das espécies de Capparaceae na caatinga. Pp. 127-131. In: E.V.S.B. Sampaio; A.M. Giulietti; J. Virginio & C.F.L. Gamarra-Rojas (eds.). Vegetação & Flora da caatinga. Recife, Associação Plantas do Nordeste/Centro Nordestino de Informações sobre plantas.         [ Links ]

Cruden, R.W. 1977. Pollen-ovule ratios: A conservative indicator of breeding systems in flowering plants. Evolution 31: 32-46.         [ Links ]

Fuks, R. & Costa e Silva, M.B. 2000. Capparis L. (Brassicaceae Burnett) do estado do Rio de Janeiro. Albertoa 1(série Brassicales): 1-12         [ Links ]

Gibbs, P.E. & Bianchi, M.B. 1992. Post-pollination events in species of Chorisia (Bombacaceae) and Tabebuia (Bignoniaceae) with late-acting self-incompatibility. Botanica Acta 106: 64-71.         [ Links ]

Gibbs, P.E. & Bianchi, M.B. 1999. Does late-acting self-incompatibility (LSI) show family clustering? Two more species of Bignoniaceae with LSI: Dolichandra cynanchoides and Tabebuia nodosa. Annals of Botany 84: 449-457.         [ Links ]

Heithaus, E.R.; Opler, P.A. & Baker, H.G. 1974. Bat activity and the pollination of Bauhinia pauletia: plant-pollinator coevolution. Ecology 55: 412-419.         [ Links ]

Leal, I.R.; Tabarelli, M. & Silva, J.M.C. (orgs.). 2003. Ecologia e conservação da Caatinga. Recife, Editora Universitária UFPE.         [ Links ]

Lipow, S.R. & Wyatt, R. 1999. Floral morphology and late-acting self-incompatibility in Apocynum cannabium (Apocynaceae). Plant Systematics and Evolution 219: 99-109.         [ Links ]

Lipow, S.R. & Wyatt. R. 2000. Single gene control of postzygotic self-incompatibility in poke milkweed, Asclepias exaltata L. Genetics 154: 893-907.         [ Links ]

Lloyd, D.G. 1972. Breeding systems in Cotula L. (Compositae, Anthemidae) I. The array of monoclinous and diclinous systems. New Phytologist 71: 1181-1194.         [ Links ]

Mabberley, D.J. 1993. The plant-book. Cambridge, Cambridge University Press.         [ Links ]

Machado, I.C.; Lopes, A.V.; Leite, A.V. & Neves, C.B. 2006. Cleome spinosa (Capparaceae): polygamodioecy and pollination by bats in urban and Caatinga areas, northeastern Brazil. Botanischer Jahrbücher für Systematik 127: 69-82.         [ Links ]

Maêda, J.M. 1985. Manual para uso da câmara de Neubauer para contagem de pólen em espécies florestais. Rio de Janeiro, UFRRJ.         [ Links ]

Martin, F.W. 1959. Staining and observing pollen tubes by means of fluorescence. Stain Technology 34: 125-128.         [ Links ]

Primo, L.M. 2004. Biologia floral e reprodutiva de Capparis jacobinae Moric. ex Eichler e Capparis hastata L. (Capparaceae). Dissertação de Mestrado, Universidade Federal de Pernambuco, Brasil.         [ Links ]

Primo, L.M. 2008. Guilda de plantas esfingófilas e quiropterófilas em remanescente de floresta Atlântica estacional semidecidual em Pernambuco: sazonalidade de recursos e biologia reprodutiva. Tese de Doutorado, Universidade Federal de Pernambuco, Brasil.         [ Links ]

Ramirez, N.; Sobrevila, C.; Enrech, N.X. de & Ruiz-Zapata, T. 1984. Floral biology and breeding system of Bauhinia benthamiana Taub. (Leguminosae), a bat-pollinated tree in Venezuelan "Llanos". American Journal of Botany 71: 273-280.         [ Links ]

Ruiz-Zapata, T. & Arroyo, M.T.K. 1978. Plant reproductive ecology of a secundary deciduous tropical forest in Venezuela. Biotropica 10: 221-230.         [ Links ]

Sampaio, E.V.S.B. 1995. Overview of the Brazilian Caatinga. Pp. 35-63. In: S.H. Bullock; H.A. Mooney & E. Medina (eds.). Seasonally dry tropical forest. Cambridge, Cambridge University Press.         [ Links ]

Seavey, S.R. & Bawa, K.S. 1986. Late-acting self-incompatibility. Botanical Review 52: 196-217.         [ Links ]

Sokal, R.R. & Rohlf, F.J. 1995. Biometry. San Francisco, W.H. Freeman.         [ Links ]

Wiens, D. 1984. Ovule survivorship, brood size, life story, breeding systems and reproductive success in plants. Oecologia 64: 47-53.         [ Links ]



Received: May 15, 2008
Accepted: November 10, 2008



* Corresponding author:
1 Part of the Master Dissertation of the first author

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