Reproductive ecology of the endangered living rock ...

4 downloads 0 Views 330KB Size Report
2001) and Hatiora (Boyle 2003) display a gametophytic system controlled by a single multiallelic locus, and in Hylocereus polyrhizus. Britton & Rose a SI has ...
Reproductive ecology of the endangered living rock cactus, Ariocarpus fissuratus (Cactaceae) Author(s) :Concepción Martínez-Peralta and María C. Mandujano Source: The Journal of the Torrey Botanical Society, 138(2):145-155. 2011. Published By: Torrey Botanical Society DOI: 10.3159/TORREY-D-10-00010.1 URL: http://www.bioone.org/doi/full/10.3159/TORREY-D-10-00010.1

BioOne (www.bioone.org) is a a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/ terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

Journal of the Torrey Botanical Society 138(2), 2011, pp. 145–155

Reproductive ecology of the endangered living rock cactus, Ariocarpus fissuratus (Cactaceae)1 Concepcio´n Martı´nez-Peralta2 and Marı´a C. Mandujano3 Laboratorio de Gene´tica y Ecologı´a, Departamento Ecologı´a de la Biodiversidad, Instituto de Ecologı´a, Universidad Nacional Auto ´ noma de Me´xico MARTI´NEZ-PERALTA, C. AND M. C. MANDUJANO. (Laboratorio de Gene´tica y Evolucio´n, Departamento Ecologı´a de la Biodiversidad, Instituto de Ecologı´a, Universidad Nacional Auto ´ noma de Me´xico). Reproductive ecology of the endangered living rock cactus, Ariocarpus fissuratus (Cactaceae). J. Torrey Bot. Soc. 138: 145–155. 2011.—The genus Ariocarpus comprises seven endangered species endemic to the Chihuahuan Desert. We studied the reproductive ecology of Ariocarpus fissuratus at Cuatrocie´negas region, Mexico, in order to explore if reproductive traits influence population viability. The size of perianth, pistils and stamens, floral behavior and controlled pollination treatments were assessed to determine Cruden’s outcrossing index (OCI), the outcrossing rate and the mating system. The breeding system was estimated with pollen/ovule ratios (P/O) and outcrossing rate was estimated by comparing progeny from self- and outcross-pollination treatments. Frequency and behavior of floral visitors was determined, and nectar production recorded. Ariocarpus fissuratus blooms synchronously during three weeks in autumn. Flowers display herkogamy and homogamy, and OCI and P/O ratio suggest xenogamy. Generalized linear mixed models were fitted to evaluate the effect of pollination treatment on fruit set and number of viable seeds yielded per fruit. Fruit set was significantly higher in the outcrossing treatment than in the selfing treatments. The mating system indicates that A. fissuratus is mainly an outcrosser, showing a partial self-incompatibility with low fruit set for hand self-pollination. Floral visitors are native solitary and introduced bees, and beetles cause high levels of florivory damage. Reproductive ecology suggests that the need for pollinators, presence of exotic pollinators, partial self-incompatibility and florivory negatively affect sexual reproduction and thus the persistence of this endangered species. Key words: Bee pollination, endemic species, floral morphology, florivory, fruit set, melittophily, outcrossing, pollination ecology, self incompatibility.

Reproductive biology encompasses mating and breeding systems, as well as attributes related to pollination such as rewards for pollinators, floral display, flowering phenology, etc. (Dafni 1992). Reproductive traits are particularly important for endangered or rare species (those with small and isolated populations) as they may constrain the subsequent stages of germination and establishment

1 This work was supported by the Secretarı ´a de Medio Ambiente y Recursos Naturales and Consejo Nacional de Ciencia y Tecnologı´a [0350] to MCM. The collected material was according to the license from SEMARNAT number SGPA/DGVS/01975/ 06. 2 We thank the authorities of CONANP at Cuatrocie´negas who provided logistic support and provided access to the Natural Protected Area. Mr. Manuel Gonza´lez for access to his property ‘‘Rancho Orozco’’ and Promotora Turı´stica de Coahuila that provided housing at ‘‘La Becerra’’. Olivia Ya´n˜ez-Ordo ´ n˜ez and Santiago Zaragoza-Caballero for indentifying the insects. J Golubov and ME McIntosh read through the entire manuscript and provided comments on this study. 3 Author for correspondence: E-mail: [email protected] miranda.ecologia.unam.mx Received for publication October 20, 2010, and in revised form February 14, 2011.

(Routley et al. 1999, Esparza-Olguı´n et al. 2005, Camargo-Smidt et al. 2006, Saunders and Sipes 2006, Strong and Williamson 2007) and determine spatial and temporal patterns of genetic diversity within and between populations (Borba et al. 2001, Barrett 2003). For example, species with outcrossing mating systems require pollinator services to produce progeny (Cruden 1977, Dafni 1992), but in small populations few potential mates result in depressed fertility (Agren 1996), which may decrease population viability through low recruitment of new individuals. This situation is critical when species are either obligate outcrossers or self-incompatible (Johnson 1992; Bowers 2002, Clark-Tapia et al. 2006). In species with an obligate outcrossing mating system, disruption in pollination services can increase the risk of population extinction (Bond 1994, Arago´n and Escudero 2008, Mandujano et al. 2010). Although factors affecting pollination systems are well documented (habitat fragmentation, invasive species of plants and animals, pesticides and other threats, Bond 1994, Kearns et al. 1998), the quantitative effects on plant populations of losing pollinators and the possible buffer

145

146

JOURNAL OF THE TORREY BOTANICAL SOCIETY

effects of the presence of other pollinators are not well understood (Bond 1994). Additional threats to reproduction include infertile pollen grains, resource limitation, inbreeding depression, flower and fruit predation and self-incompatibility (Mandujano et al. 1996, Navarro and Guitia´n 2002, McIntosh 2002, Pin˜a et al. 2007, Hoebee et al. 2008). Inbreeding depression has been documented in several species of Cactaceae either with self or outcross-pollination mating systems (Mandujano et al. 1996, McIntosh 2002, Mandujano et al. 2010). Studies of florivory have shown that it can decrease fertility in two ways: a) by lowering floral attractiveness via flower damage and b) by direct impact on fruit formation via herbivory (Petterson 1991, Leavitt and Robertson 2006, Sa´nchez-Lafuente 2007). Florivory by beetles has also been reported in cacti (Pimienta and Del Castillo 2002, Pin˜a et al. 2010) but the degree of damage and effects on plant fertility have not been quantified. In the case of self-incompatible (SI) species, the requirements of genetically distinct mates and dependence on pollen vectors for outcrossing make the sexual reproduction of rare plants prone to failure. Simulations including demography and a SI system in a rare shrub have shown that the presence of a SI system decreases effective population size and thus increases risk of extinction (Hoebee et al. 2008). SI systems in Cactaceae have been barely studied (Mandujano et al. 2010), but their presence has been suggested in ca. 30% of the genera from the three more diverse subfamilies (Boyle 1997), but very few studies describe SI in detail (Mandujano et al. 2010). For example, the genera Schlumbergera (Boyle 1997, 2003), Echinopsis (Boyle and Idnurm 2001) and Hatiora (Boyle 2003) display a gametophytic system controlled by a single multiallelic locus, and in Hylocereus polyrhizus Britton & Rose a SI has been seen at the ovary level (Lichtenzveig et al. 2000). All of these studies were from cultivars or glasshouse plants; so there is no information about the performance of SI in natural populations of Cactaceae. Successful conservation programs for rare or threatened plant species often rely on knowledge of their population ecology (Schemske et al. 1994, Gigon et al. 2000, Saunders and Sipes 2006, Mandujano et al. 2007). However, few studies explore reproduc-

[VOL. 138

tive attributes even though they are central to understand population dynamics of rare plants (Kearns and Inouye 1997). For example, the review of Kwak and Bekker (2006) compares the reproductive biology among rare and common species including 30 studies of endangered species, and 105 of non-endangered or common species (22% of the studies were of rare plants). The review suggests that pollination specificity, life span, seed production, seed bank, dispersal, and clonality are fitness and dispersal traits that may constrain natural regeneration of endangered plant populations. They found that rare species were more vulnerable to extinction than common ones, and the cases with successful recovery plans were the species which included information about pollination and reproductive ecology (Kwak and Bekker 2006). For instance, around 125 genera and 1810 species belong to the cactus family (Anderson 2001), with 85% of species being endemic to some region of America and are considered rare species, 35% appearing in the Red List (Anderson 2001, Red List Category summary for all plant classes and families, IUNC 2009), and , 1% of the taxa are listed in CITES appendix 1 (Lu¨thy 2001, Arias et al. 2005), but only 2% of species have been subjects of reproductive ecology studies (Mandujano et al. 2010) and none with the aim to determine the role of reproductive system on conservation status of species. Several authors suggest different factors are responsible for the threat to cactus species. Biotic factors include long life cycles (Bravo-Hollis and Sa´nchez-Mejorada 1991, Jime´nez-Sierra et al. 2007), low growth rates and limited recruitment (Mandujano et al. 2001, Esparza-Olguı´n et al. 2005), small population sizes (Martı´nez-Avalos et al. 2007, Strong and Williamson 2007), predation (Martı´nez-Avalos et al. 2007) and geographical rarity (Esparza-Olguı´n et al. 2005, Mandujano et al. 2007). Among anthropogenic factors are habitat fragmentation, changes in land use, and plant collection (CarrilloAngeles et al. 2005, Martorell and Peters 2005). Design of conservation programs for endangered species should consider the factors that affect their population dynamics (Schemske et al. 1994, Contreras and Valverde 2002, Kwak and Bekker 2006, Mandujano et al. 2007). Demographic parameters such as fertility and seedling recruitment are largely

2011]

MARTI´NEZ-PERALTA AND MANDUJANO: REPRODUCTIVE ECOLOGY OF ROCK CACTUS

dependent on the reproductive biology of a species (e.g., Johnson 1992, Mandujano et al. 1996, Bowers 2002, McIntosh 2002, Navarro and Guitia´n 2002). Our aim in this study was to determine attributes of the reproductive biology of Ariocarpus fissuratus, a rare cactus endemic to the Chihuahuan Desert, and how these attributes could affect the endangered status of the species. Methods. SPECIES AND STUDY SITE. Ariocarpus fissuratus (Engelmann) Schumann (Cactaceae) belongs to the subfamily Cactoideae, tribe Cacteae. This species is solitary, geophytic and has large taproots. The stems, which are 15 cm in diameter, are small and compact, possessing tubercles, which can be laterally divergent, crowded, flattened or slightly convex above. The tubercles, which are 1–2 cm long, 1.5–2.5 cm broad, are usually apically rounded and often have numerous fissures on the upper surface. These plants possess woolly centers with furrowed areoles extending the length of the tubercle. Taproots are large, fleshy and with an extensive mucilage system (Bravo-Hollis and Sa´nchez-Mejorada 1991, Anderson 2001). The flowers are magenta in color, 2.5–4.5 cm in diameter and grow from the areoles of the woolly center of the plant during autumn (Bravo-Hollis and Sa´nchezMejorada 1991, Anderson 2001). The fruits are ovoid and rarely emerge from the central areoles of the plant, which causes the seeds to remain on the plant until they are eventually flushed away by water or wind (CMP, personal observations). This species is found in calcareous soils and is endemic to the Chihuahuan Desert from San Luis Potosı´, Mexico to the Big Bend and Pecos River in Texas, USA (Anderson 2001). This species is listed as endangered according to CITES (Lu¨thy 2001, IUNC 2009) and Mexican legislation NOM-059-ECOL2001 (Arias et al. 2005). The study site was located in the Cuatrocie´negas Valley in central Coahuila, Mexico (26u 549 43.140 N, 102u 79 21.540 W). The target population was in a desert scrub bajada with calcareous soil. Climate is arid, semi warm, median annual temperature ranges from 18–22 uC, and rainfall is scarce through the year but present mainly in summer, with total annual precipitation about 400 mm (Garcı´a 1981, Arriaga et al. 2000). Vegetation is dominated by Larrea tridentata Coville, Prosopis glandulosa var. torreyana Torr.,

147

Acacia greggii A. Gray, Fouquieria splendens Engelm., Suaeda mexicana Standl., Suaeda palmeri Standl., Suaeda suffruticosa Standl., Jatropha dioica var. graminea Sesse´, Dasylirion sp., Agave lechuguilla Torr., Euphorbia antisyphilitica Zucc., and other cactus species such as Cylindropuntia imbricata Haw. (DC.), Cylindropuntia leptocaulis (DC.) F.M. Knuth, Cylindropuntia kleiniae (DC.) F.M. Knuth, Mammillaria gigantea Hildm., Mammillaria pottsii Scheer, Grusonia bradtiana Britton & Rose, Grusonia schottii (Engelm.) H.Rob., Echinocereus engelmannii (Parry ex Engelm.) Ru ¨ mpler, Ferocactus hamatacanthus Britton & Rose, Epithelantha micromeris F.A.C. Weber, and Opuntia rufida Engelm. (Pinkava 1984). FLORAL PHENOLOGY AND FLORAL CYCLE. We studied the floral cycle by recording perianth aperture and characteristics of sexual structures at two hour intervals from 0700 to 1900 throughout the flower lifespan (which can be one, two or three days; seven intervals per day) in randomly selected first day flowers located on different plants (n 5 26 in 2005, and n 5 27 in 2006). In order to determine the flowering peak we tagged 172 plants in 2005 and 30 in 2006 and followed floral aperture. The diameter of each plant was also measured in order to determine the effect of plant size on reproductive success, measured as the number of flowers produced through the flowering peak observed. A non-parametric correlation (Spearman r) was carried out to test for a relationship between these two variables. BREEDING SYSTEM. We determined the outcrossing index (OCI) (Cruden 1977) by means of corolla size (i.e., perianth diameter), and the presence of herkogamy and dichogamy. Perianth diameter was considered the maximum aperture recorded from the two hour intervals described before. Pistil and stamen length were measured and compared with a t-test to determine the presence of herkogamy. Dichogamy was evaluated at each of the two hour intervals during the floral cycle by both registering pollen availability at the anthers (estimated visually from dehiscence) and stigma receptivity (determined by observing the stigma lobes for opening or presence of exudations, Mandujano et al. 1996, Negro´nOrtiz 1998). Homogamy occurs if the curves of stigma receptivity and pollen dehiscence through time overlap, otherwise there is dichogamy.

148

JOURNAL OF THE TORREY BOTANICAL SOCIETY

We collected 32 flowers, each from a different plant, to determine pollen/ovule ratios (Cruden 1977). Ovule number per flower was counted with the aid of a microscope (103). To determine the mean number of stamens per flower, flowers were sectioned longitudinally and the number of stamens counted. We also collected one anther per flower in a 1.5 ml microtube containing 1 ml of distilled water and homogenized the pollen using a vortex mixer. A 100 ml aliquot (dilution factor 5 10) of the suspension was sampled and pollen grains counted under a microscope (253). This number was multiplied by the dilution factor and by the number of anthers to obtain total number of pollen grains per flower (Cruden 1977; Mandujano et al. 1996). MATING SYSTEM. We carried out five pollination treatments (Kearns and Inouye 1993) in 2005 to determine the mating system: 1) control (C, n 5 182), where flowers were left undisturbed and available to pollinators; 2) cross-pollination (CROSS, n 5 44), where flowers were emasculated and hand pollinated with pollen from 10 flowers (each from a different plant), and covered with a mesh bag to prevent access by floral visitors; 3) hand self-pollination (SELF, n 5 67), where flowers were pollinated with self pollen and covered with mesh bags; 4) automatic self-pollination (ASELF, n 5 33), where flowers were left undisturbed and covered with a mesh bag to assess automatic self-pollination; and 5) supplemental cross pollination (SP, n 5 68), where flowers were available to pollinators and were hand pollinated using supplemental pollen from a mix of pollen from 10 different parents. After 4 months we collected fruits from all pollination treatments and counted the seeds. Fruit loss was not prevented because the development of the fruit occurs while buried in the central wool of the plant, naturally protected from predation. Fruit set and seed set were analyzed with non-parametric deviance analysis by means of a generalized linear model, with a binomial error for proportion data and Poisson distribution for seed counts, using the logit link function. Differences among treatments were evaluated using orthogonal contrasts for both fruit and seed set (JMP 2007). From pollination treatment results, we calculated the outcrossing rate (te) by comparing the fruit set of

[VOL. 138

outcrossed progeny (CROSS) and the fruit set of selfed plus outcrossed progeny (SELF+ CROSS), as proposed by Mandujano et al. (2010) te~wx =ðwx zws Þ, where wx and ws are the fruit set proportions resulted from CROSS and SELF, respectively. This outcrossing rate varies from 0–1, 1 for outcrossing species, 0 for selfing species and 0.5 for mixed mating systems. In addition, a pollen limitation index was calculated (Larson and Barrett 2000) L~1{ðPo =Ps Þ where Po is the percent fruit set of open pollinated controls and Ps is the fruit set by plants that received SP. POLLINATION SYNDROME AND NECTAR PROIn 2006 we tagged 42 flowers at two patches of plants, 300 m apart from each other for observation of floral visitors (20 and 22 flowers at each patch). Frequency and general behavior of floral visitors were observed on target flowers for 20 min every 2 h from 1000 to 1600 throughout a flower’s complete life span (lasting up to 3 days). Frequencies of morphospecies were pooled by interval to determine differences in pollinator frequencies and time of visit. Data were analyzed using a generalized linear model with a Poisson distribution (JMP 2007). Specimens of each morphospecies were collected in the field and identities of main floral visitors were determined by Olivia Ya´n˜ez-Ordo ´ n˜ez (Facultad de Ciencias, UNAM). Nectar production was determined using two sets of flowers. One set of flowers was covered with a mesh bag (BAGGED, n 5 28 flowers, n 5 11 opened a second day) and the other set was left uncovered (OPEN, n 5 27 flowers, n 5 8 flowers opened a second day). From these samples we measured nectar production every 2 h from 1030 to 1830 using 1 ml capillary tubes and converting nectar length in the capillary tubes to volume units (Mandujano et al. 1996). At each of these two hour intervals we obtained mean nectar volume and standard error for first and second day flowers for both treatments. We applied a two way analysis of variance (ANOVA) to evaluate differences in nectar production using day and treatment as factors (JMP 2007). The unequal number of flowers that opened two DUCTION.

2011]

MARTI´NEZ-PERALTA AND MANDUJANO: REPRODUCTIVE ECOLOGY OF ROCK CACTUS

days did not allow us to carry out an ANOVA for repeated measures. FLORIVORY. In 2005 we observed a beetle species that consumed floral parts mainly during anthesis and post-anthesis stages, and rarely in the bud stage. To estimate florivory damage we monitored all flowers on 30 plants (n 5 45 flowers) during the maximum flowering period in 2006 and assigned each flower to three damage categories: UN 5 undamaged flowers, PC 5 partially consumed flowers and CC 5 completely consumed flowers. We obtained proportions for each damage category per day of observation and used a generalized linear model to test for differences in the proportion of damaged flowers per day of observation (JMP 2007). Identity of the florivore beetle was determined by Santiago Zaragoza (Instituto de Biologı´a, UNAM). Results. FLORAL PHENOLOGY AND FLORAL CYCLE. The flowering period of Ariocarpus fissuratus is in October. Although flowers were observed over a three -week period, we registered a single peak flowering period of approximately five days in 2005 and three days in 2006. The 172 tagged plants in the first year produced 414 flowers, 2.6% of them opened on the 15 of October, and 70% opened the next two days. By the 21st of October, all flowers ended anthesis. On the second floral season, the 30 tagged plants produced 42 flowers, flowering started on October 24 were 40.85 and 47.9% of the flowers opened on the fist and second day, respectively, and by October 27 all plants had closed flowers. Herbarium specimens also corresponded to the same period (October; accession numbers in MEXU, 1026934, 1128229, 1142195). Flowers of Ariocarpus fissuratus open for one, two and rarely three days (CMP, personal observation; in our samples no flowers were open more than two days). In 2005, most of the flowers opened for one day (61.8%) while in 2006 most opened for two days (78.1%). Flowers in their first and second day of life opened roughly at 0900 and closed at 1800. Maximum length in perianth aperture (Table 1) was reached around 1300 during both days. Plants producing two flowers were more frequent than plants producing any other number of flowers (1 flower plant21 5 23.7%, 2 flowers 5 37.9%, 3 flowers 5

149

Table 1. Floral measurements used to calculate the outcrossing index (OCI; Cruden 1977, n 5 27 flowers) and to determine the breeding system in Ariocarpus fissuratus according to P/O ratio (Cruden 1977, n532 flowers). Floral attribute

Perianth aperture (cm) Pistil length (cm) Stamen length (cm) Ovules per flower Anthers per flower Pollen grains per flower P/O ratio

Mean 6 SE

378 1.94 1.63 117.57 351.07 155020 1371

6 6 6 6 6 6 6

0.11 0.05 0.05 3.75 16.17 6588.96 87.15

21.5%, 4 flowers 5 10.2%, 5 flowers 5 5.1%, 6 flowers 5 1.7%, n5177 plants, data from 2005), even though we found plants that could produce up to 9 flowers. We also found a positive correlation between plant size and number of flowers (r 5 0.5398, P , 0.01). BREEDING SYSTEM. Pistil and stamen lengths were measured at the time of maximum diameter of perianth aperture (Table 1). According to the t-test, sexual structures showed significant differences in length (t 5 4.37, d.f. 544, P , 0.0001) and therefore the flowers display homomorphic herkogamy. We did not find a complete temporal separation between sexual functions which indicates homogamy, but there was a tendency to protandry. During the fist two hours after anthesis 65% of the plants showed dehiscent stamens and 55% showed also mature stigmas, so 10% of the plants are functioning as males, but after 4 hours both genders are functioning simultaneously. With these measurements and perianth diameter the calculated OCI was 4, suggesting xenogamy requiring cross-pollination and a pollen vector. The other floral attributes measured (Table 1) suggested a P/O ratio closer to a facultative xenogamous species: an outcrosser that accepts self pollination in the absence of or in addition to cross pollination, but whose adaptations are clearly favoring xenogamy (Cruden’s index, Cruden 1977). MATING SYSTEM. Fruit and seed production differed among pollination treatments (fruit set: x2 5 87.89, d.f.53, P , 0.00001, seed set: x2 5 92.99, d.f. 5 3, P , 0.0001). The automatic self-pollination (ASELF) treatment did not set fruit so it was excluded from further analysis. Fruit set in self and crosspollination treatments were significantly dif-

150

JOURNAL OF THE TORREY BOTANICAL SOCIETY

[VOL. 138

ferent from the supplemental cross pollen and control treatments, but among the latter two there were no statistical differences (Fig. 1a). The lowest mean seed set was for the crosspollination treatment and was significantly different from the others (Fig. 1b). The outcrossing rate (te) obtained for fruits was 0.93, indicating a mating system that tends towards outcrossing (Mandujano et al. 2010). The calculated pollen limitation index L was 0.071, the tendency towards 0 indicates no pollen limitation in the population under study. POLLINATION SYNDROME AND NECTAR PROFloral visitors in Ariocarpus fissuratus included bees (Lasioglossum sp., Megachilidae sp., Apis mellifera Linnaeus and Diadasia sp.), one ant species, two wasp morphospecies, two species of butterflies and one species of florivore beetle. Of these, the bees and wasps were the more frequent visitors and both were considered for further analysis. However, only the bees are considered pollinators because of their behavior: they touched stigmas before reaching stamens whereas wasps tended to touch stamens or stigmas but rarely both. The bees Lasioglossum sp. and Megachilidae sp. were pooled in the analysis since their visits were recorded together in the field. Visits to flowers began around 1000 and reached the highest frequency at 1200 (56.4%, n 5 42 flowers). We found significant differences in visitation frequency per species and per hour (x2 5 24.6, d.f. 5 9, P 5 0.0035). The most abundant species was A. mellifera in all observations (91.3% of total visits, n 5 42 flowers) (Fig. 2). The mean number of visits per flower in a 160 min observation period was 15.2 6 1.1 (x¯6 SE), ranging from 3 to 32 visits per flower (n 5 42 flowers). Nectar production began 1.5 h after anthesis (1030) and there were no differences in nectar production among days or treatments, neither was there an interaction between days and treatments (F 5 1.9, P 5 0.18). Pooling the data, total nectar production was 1.2 ml 6 0.3 (x¯ 6 SE) for bagged flowers and 1.12 ml 6 0.31 (x¯ 6 SE) for uncovered flowers. DUCTION.

FLORIVORY. The florivore beetle was identified as belonging to the Tenebrionidae. Day 1 was excluded in the generalized linear model because none of the tagged flowers opened or was damaged (UN 5 100%). The analyses indicated that there were differences in the

FIG. 1. Fruit (a) and seed set (b) for each pollination treatment (2005). No fruits were produced from natural self-pollination. Different letters indicate statistical differences with orthogonal contrast among treatments in fruit or seed set (P , 0.001). SELF 5manual self-pollination (n 5 67), CROSS 5cross-pollination (n 5 44), SP5supplementary pollen (n 5 68), C5control (n 5 182).

proportion of damaged flowers between days during the flowering peak (x2 5 8.9, d.f. 5 2, P 5 0.0116). Days 3 and 4 did not differ from each other (38% and 47% of flowers in some category of damage, respectively; x2 5 1.0, P 5 0.32) but they were statistically higher than day two (18% of flowers in some category of damage; x2 5 7.8, P 5 0.01). Flowers were affected up to the second day of observation, and most of them were consumed during anthesis (66.7%). After four days UN was 53.33% and the total percentage of damage by the Tenebrionid beetle to flowers was 46.67% (PC 5 15.56% and CC 5 31.11%). However, we do not know if partially consumed flowers set any fruit, so only CC was considered as florivory.

2011]

MARTI´NEZ-PERALTA AND MANDUJANO: REPRODUCTIVE ECOLOGY OF ROCK CACTUS

151

FIG. 2. Percentage of visits per observational period of the main visitors to Ariocarpus fissuratus (n 5 42 flowers).

Discussion. Can floral biology provide useful information regarding the endangered status of Ariocarpus fissuratus? Our study has shown several factors (i.e., flowering period, floral morphology, breeding system, mating system, partial self-incompatibility, exotic floral visitors and florivory) that may be relevant in understanding the endangered status of A. fissuratus. The flowering period of A. fissuratus occurs in October with a flowering peak that may change slightly from year to year in onset and duration. In the few other cacti in which floral biology has been studied, flowering periods are generally longer, lasting several months with one or several flowering peaks (Mandujano et al. 2010). Of the cactus species with short flowering periods (e.g., Echinomastus erectocentrus Britton & Rose and Echinocereus engelmannii var. acicularis, Johnson 1992), A. fissuratus has the shortest which suggests a massive and synchronous flowering strategy (‘‘big bang flowering’’ sensu Gentry 1974). This would in theory lead to an increase in pollen flow and quality of progeny and an increase in successful pollination by avoiding predation of reproductive structures (Primack 1985, Trapnell and Hamrick 2006). Synchronous flowering in A. fissuratus could enhance the attraction of floral visitors, reduce pollen limitation and the high density of flowers could buffer high florivory damage. Florivory reduces the available number of flowers, decreases pollinator attraction when it is high (ca. 55–64%), and limits fruit and seed set

(Sa´nchez-Lafuente 2007). Florivory by beetles in A. fissuratus was as high (CC 5 31.11%) as that reported for other species (e.g., Linaria lilacina Lange, Plantaginaceae, Sa´nchez-Lafuente 2007, Lepidium papilliferum A.Nelson & J.F. Macbr., Brassicaceae, Leavitt and Robertson 2006, Opuntia microdasys (Lehm.) Pfeiff., Cactaceae, Pin ˜ a et al. 2010) which reduces 36% of plant fitness. Synchronous flowering is hypothesized to buffer the negative effects of florivory, but it has been rarely measured (e.g., Astralagus, Crone and Lesica 2004). However, in small isolated populations commonly found in endangered species, the benefits of attractiveness can be counterbalanced by other factors such as specific weather conditions that can determine visitor availability and floral longevity (Kwak and Bekker 2006, Oaxaca-Villa et al. 2006). During the second season of study (2006) the weather was rainy, so the long floral life span in this year could be related to both cooler temperature and delayed pollination. It has been observed in plants of the same genus that cold, rainy days, affect the quality of experimental hand pollinations and that non-pollinated flowers live longer, up to three days (CMP personal observations, Pimienta and Del Castillo 2002). In A. fissuratus this counterbalance is further enhanced by a xenogamous breeding system (given the OCI index), facultative xenogamy (given by the pollen:ovule ratios), outcrossing mating system (given by the outcrossing rate) and self-incompatibility. The breeding system

152

JOURNAL OF THE TORREY BOTANICAL SOCIETY

of A. fissuratus is similar to other species in the family (Pimienta and Del Castillo 2002, Mandujano et al. 2010) and agrees with observational reports of other species in the same genus (Sa´nchez-Mejorada et al. 1986). However a more detailed, quantitative method (for example, pollen grain counts for male function and pollinations at different times for female function) could help to resolve the tendency towards protandry that was found. The high production of fruits from treatments involving cross-pollination and the low production of fruits from self-pollination treatments indicate that A. fissuratus depends on pollen from other genetically distinct individuals. These observations jointly suggest that cross pollination in this species is needed for setting fruits and seeds and therefore depends on the services provided by pollinators, as reported for rare species (Kwak and Bekker 2006) and in other Cactaceae such as E. erectrocentrus (Johnson 1992), M. grahamii (Bowers 2002), Coryphanhta robbinsorum (W.H.Earle) A.D.Zimmerman (Schmalzel et al. 1995), Ferocactus histrix (DC.) G. E. Linds., Sclerocactus polyancistrus Britton & Rose, and Coryphantha scheeri var. robustispina (Kuntze) L.D.Benson (Anderson 2001, Pimienta and Del Castillo 2002). The low success of the self-pollination treatment could involve other processes in the population, such as partial self-incompatibility, where some individuals in the population are self-compatible (Ferrer and Good-Avila 2007, Colautti et al. 2010). In addition, the low production of fruits from hand cross-pollination compared with that of control treatment could be explained by a SI system with a small number of S-alleles, so that the number of pollen donors used were not enough to produce similar fruit set as found after natural pollination (Barrett 1988, Uyenoyama 1993, Boyle 1997). SI systems were regarded for a long time as static traits with no variation in the inhibiting response (de Nettancourt 1977). However, some self-incompatible species are in fact capable of setting fruits and seeds after selfing at very low proportions (Zapata and Arroyo 1978, Borba et al. 2001, Ferrer and GoodAvila 2007, Colautti et al. 2010). So, the threshold for defining a self-incompatible species based on the production of seeds is arbitrary for those cases. Species with variation in the response of the SI system (most of

[VOL. 138

the individuals are self sterile but some are not) are called partially or weakly selfincompatible, and they are being understood only recently. Self-incompatibility in rare plants represents another factor for decline of populations by means of diminishing the probability of successful reproduction (Hoebee et al. 2008). The partial SI of Ariocarpus fissuratus could affect the success of sexual reproduction because of the necessity of available mates and also adequate pollen transfer provided by bees. The most likely bee pollinators are Diadasia sp. as well as other native bee species (Megachilidae sp. and Lasioglossum sp.), based on their frequency and their behavior, since they touched the stigma when reaching a flower and then went down to collect pollen and nectar. These bees have been reported as pollinators in other cactus species (Beutelspacher 1971, Mandujano et al. 1996, McIntosh 2005). Even though this is the first report of Apis mellifera (an introduced species) on A. fissuratus, it is worrying because it was the most frequent floral visitor. It has been demonstrated for Prosopis velutina Wooton when considering the number of visits required by a particular insect to effect pod production equal to that of open-pollinated inflorescences, native bee species were more efficient than the introduced A. mellifera (Keys et al. 1995). Although progeny production from supplemental pollen was slightly higher than that of the control in A. fissuratus, we did not find a significant difference between supplemental pollen and control treatments. This result was supported by the L index which points that A. fissuratus did not display pollen limitation. The last observation contrasts with the fact that nectar could be found in covered and uncovered flowers treatments, suggesting that bees may be foraging for pollen and not for nectar and that a flower produces more nectar than needed for the actual number of foraging bees. In addition, it has been proposed that self-incompatible plants tend to display higher levels of pollen limitation (L 5 0.59 for 66 self-incompatible angiosperms, Larson and Barrett 2000). In species with life cycles that depend on sexual reproduction to recruit new individuals, establishment of new individuals will be constrained by failures in reproduction, especially in species which display outcrossing mating systems (Louda 1982, Weppler and

2011]

MARTI´NEZ-PERALTA AND MANDUJANO: REPRODUCTIVE ECOLOGY OF ROCK CACTUS

Sto¨cklin 2006, Arago´n and Escudero 2008). Our study with Ariocarpus fissuratus indicates several reproductive traits that may not buffer the negative effects of rarity: population isolation, the species dependency on seed production, and seedling establishment. Therefore the reproductive biology of A. fissuratus to an extent is determining subsequent developmental stages and population dynamics. We found that reproduction is limited by high levels of florivory, selfincompatibility, and low fertility, that coupled with anthropogenic disturbance (changes in land use and farming) and illegal collection increase the endangered status of the species. In addition, the pollinator requirement suggests that A. fissuratus may be vulnerable to changes in its pollination system by the displacement of native bees by the exotic Apis mellifera. Specific responses to these changes are difficult to assess (Johnson and Steiner 2000), but could be an important aim for the conservation of bee pollinated species. Literature Cited AGREN, J. 1996. Population size, pollinator limitation, and seed set in the self incompatible herb Lythrum salicaria. Ecology 77: 1779–1790. ANDERSON, E. F. 2001. The Cactus Family. Timber Press Inc., Portland, OR. 776 p. ARAGO´N, C. F. AND A. ESCUDERO. 2008. Mating system of Helianthemum squamatum (Cistaceae), a gypsophile specialist of semi-arid Mediterranean environments. Bot. Helv. 118: 129–137. ARIAS, S., U. GUZMA´N, M. C. MANDUJANO, M. SOTO, AND J. GOLUBOV. 2005. Las especies mexicanas de cacta´ceas en riesgo de extincio´n. I. Cac. Suc. Mex. 50: 100–125. ARRIAGA, L., J. M. ESPINOSA, C. AGUILAR, E. MARTI´NEZ, L. GO´MEZ, AND E. LOA. 2000. Regiones prioritarias de Me´xico. Comisio ´n Nacional para el Conocimiento y Uso de la Biodiversidad. Mexico. BARRETT, S. C. H. 1988. The evolution, maintenance, and loss of self-incompatibility systems. p. 98–124. In J. Lovett-Doust and L. LovettDoust [eds.], Plant Reproductive Ecology: Patterns and Strategies, Oxford University Press, New York, NY. BARRETT, S. C. H. 2003. Mating strategies in flowering plants: the outcrossing-selfing paradigm and beyond. Phil. Trans. Roy. Soc. B. 358: 991–1004. BEUTELSPACHER, C. R. 1971. Polinizacio´n en Opuntia tomentosa Salm-Dyck y O. robusta Wendland en ´ ngel. Cact. Suc. Mex. 16: el Pedregal de San A 84–86. BOND, W. J. 1994. Do mutualisms matter? Assessing the impact of pollinator and disperser disruption on plant extinction. Phil. Trans. Roy. Soc. B. 344: 83–90.

153

BORBA, E. L., J. SEMIR, AND G. J. SHEPHERD. 2001. Self-incompatibility, inbreeding depression and crossing potential in five brazilian Pleurothallis (Orchidaceae) species. Ann. Bot. 88: 89–99. BOWERS, J. E. 2002. Flowering patterns and reproductive ecology of Mammillaria grahamii (Cactaceae), a common, small cactus in the Sonoran Desert. Madron ˜ o 49: 201–206. BOYLE, T. H. 1997. The genetics of self-incompatibility in the genus Schlumbergera (Cactaceae). J. Hered. 88: 209–214. BOYLE, T. H. 2003. Identification of self-incompatibility groups in Hatiora and Schlumbergera (Cactaceae). Sex. Plant Reprod. 16: 151–155. BOYLE, T. H. AND A. IDNURM. 2001. Physiology and genetics of self-incompatibility in Echinopsis chamaecereus (Cactaceae). Sex. Plant Reprod. 13: 323–327. BRAVO-HOLLIS, H. AND R. H. SA´NCHEZ-MEJORADA. 1991. Las cacta´ceas de Me´xico. Universidad Nacional Auto ´ noma de Me´xico, Coordinacio´n de la Investigacio´n Cientı´fica, Mexico City, MX. CAMARGO-SMIDT, E., E. SILVA-PEREIRA, AND E. LEITE-BORBA. 2006. Reproductive biology of two Cattleya (Orchidaceae) species endemic to north-eastern Brazil. Plant Spec. Biol. 21: 85–91. CARRILLO-ANGELES, I. G., J. GOLUBOV, M. ROJASARE´CHIGA, AND M. C. MANDUJANO. 2005. Distribucio´n y estatus de conservacio´n de Ferocactus robustus (Pfeiff.) Britton et. Rose. Cact. Suc. Mex. 50: 36–55. CLARK-TAPIA, R., F. MOLINA-FREANER, A. CORRADO, AND M. C. MANDUJANO. 2006. Reproductive consequences of clonal growth in Stenocereus eruca, a rare clonal cactus of the Sonoran Desert. Evol. Ecol. 20: 131–142. COLAUTTI, R. I., N. A. WHITE, AND S. C. H. BARRETT. 2010. Variation of self-incompatibility within invasive populations of purple loosestrife (Lythrum salicaria L.) from eastern North America. Int. J. Plant Sci. 171: 158–166. CONTRERAS, C. AND T. VALVERDE. 2002. Evaluation of the conservation status of a rare cactus (Mammillaria crucigera) through the analysis of its population dynamics. J. Arid Environ. 51: 89–102. CRONE, E. E. AND P. LESICA. 2004. Causes of synchronous flowering in Astragalus scaphoides, an iteroparous perennial plant. Ecology 85: 1944–1954. CRUDEN, R. 1977. Pollen-ovule ratios: a conservative indicator of breeding systems in flowering plants. Evolution 31: 32–46. DAFNI, A. 1992. Pollination Ecology. Oxford University Press, Oxford, UK. 272 p. DE NETTANCOURT, D. 1977. Incompatibility in angiosperms. Springer-Verlag. New York, NY. 230 p. ESPARZA-OLGUI´N, L., T. VALVERDE, AND M. C. MANDUJANO. 2005. Comparative demographic analysis of three Neobuxbaumia species (Cactaceae) with differing degree of rarity. Pop. Ecol. 47: 229–245. FERRER, M. M. AND S. GOOD-AVILA. 2007. Macrophylogenetic analyses of the gain and loss of selfincompatibility in the Asteraceae. New Phytol. 173: 401–414.

154

JOURNAL OF THE TORREY BOTANICAL SOCIETY

GARCI´A, E. 1981. Modificaciones al sistema de clasificacio´n clima´tica de Ko ¨ ppen para adaptarlo a las condiciones de la Repu´blica Mexicana. Universidad Nacional Auto ´ noma de Me´xico, Mexico City, MX. GENTRY, A. H. 1974. Flowering phenology and diversity in a tropical Bignoniaceae. Biotropica 6: 64–68. GIGON, A., R. LANGENAUER, C. MEIER, AND B. NIEVERGELT. 2000. Blue Lists of threatened species with stabilized or increasing abundance: a new instrument for conservation. Conserv. Biol. 14: 402–413. HOEBBE, S. E., P. E. THRALL, AND E. A. G. YOUNG. 2008. Integrating population demography, genetics and self-incompatibility in a viability assessment of the Wee Jasper Grevillea (Grevillea iaspicula McGill., Proteaceae). Conserv. Genet. 9: 515–529. IUNC. INTERNATIONAL UNION FOR NATURE CONSERVATION. 2009. IUCN Red List of Threatened Species Version 2009.1. Accessed 20 October 2009. ,www.iucnredlist.org. JIME´NEZ-SIERRA, C., M. C. MANDUJANO, AND L. E. EGUIARTE. 2007. Are populations of the candy barrel cactus (Echinocactus platyacanthus) in the desert of Tehuaca´n, Mexico at risk? Population projection matrix and life table response analysis. Biol. Conserv. 135: 278–292. JMP. 2007. Version 7. SAS Institute Inc. Cary, NC. JOHNSON, R. 1992. Pollination and reproductive ecology of acun˜a cactus, Echinomastus erectrocentrus var. acunensis. Int. J. Plant Sci. 153: 400–408. JOHNSON, S. D. AND K. E. STEINER. 2000. Generalization versus specialization in plant-pollinator systems. Trends Ecol. Evol. 15: 140–143. KEARNS, C. A. AND D. W. INOUYE. 1993. Techniques for pollination biologists. University Press of Colorado, Niwot, CO. 583 p. KEARNS, C. A. AND D. W. INOUYE. 1997. Pollinators, flowering plants, and conservation biology. BioScience 47: 297–307. KEARNS, C. A., D. W. INOUYE, AND N. M. WASER. 1998. Endangered mutualisms: the conservation of plant-pollinator interactions. Annu. Rev. Ecol. Syst. 29: 83–112. KEYS, R. N., S. L. BUCHMANN, AND S. E SMITH. 1995. Pollination effectiveness and pollination efficiency of insects foraging Prosopis velutina in South–eastern Arizona. J. Appl. Ecol. 32: 519–527. KWAK, M. M. AND R. M. BEKKER. 2006. Ecology of plant reproduction: Extinction risks and restoration perspectives of rare plant species. p. 362–386. In N. M. Waser and J. Ollerton [eds.], Plant-pollinator Interactions: From Specialization to Generalization. University of Chicago, Chicago, IL. LARSON, B. M. H. AND S. C. H. BARRETT. 2000. A comparative analysis of pollen limitation on flowering plants. Biol. J. Linn. Soc. 69: 503–520. LEAVITT, H. AND I. C. ROBERTSON. 2006. Petal herbivory by chrysomelid beetles (Phyllotreta sp.) is detrimental to pollination and seed production in Lepidium papilliferum (Brassicaceae). Ecol. Entomol. 31: 657–660.

[VOL. 138

LICHTENZVEIG, J., S. ABBO, A. NERD, N. TEL-ZUR, AND Y. MIZRAHI. 2000. Cytology and mating systems in the climbing cacti Hylocereus and Selenicereus. Am. J. Bot. 87: 1058–1065. LOUDA, S. M. 1982. Limitation of the recruitment of the shrub Haplopappus squarrosus (Asteraceae) by flower- and seed-feeding insects. J. Ecol. 70: 43–53. LU¨THY, J. M. 2001. The cacti of CITES Appendix I. CITES identification manual. CITES Management Authority of Switzerland. Bern, Switzerland. MANDUJANO, M. C., I. G. CARRILLO-ANGELES, C. MARTI´NEZ-PERALTA, AND J. GOLUBOV. 2010. Reproductive biology of Cactaceae. p. 197–230. In K. G. Ramawat [ed.], Desert Plants - Biology and Biotechnology. Springer Berlin, Heidelberg, DE. MANDUJANO, M. C., C. MONTAN˜A, AND L. E. EGUIARTE. 1996. Reproductive ecology and inbreeding depression in Opuntia rastrera (Cactaceae) in the Chihuahuan desert: Why are sexually derived recruitments so rare? Am. J. Bot. 83: 63–70. MANDUJANO, M. C., C. MONTAN˜A, M. FRANCO, J. GOLUBOV, AND A. FLORES-MARTI´NEZ. 2001. Integration of demographic annual variability in a clonal desert cactus. Ecology 82: 344–359. MANDUJANO, M. C., J. A. M. VERHULST, I. G. CARRILLO-ANGELES, AND J. GOLUBOV. 2007. Population dynamics of Ariocarpus scaphirostris Bo ¨ deker (Cactaceae): evaluating the status of a threatened species. Int. J. Plant Sci. 168: 1035–1044. MARTI´NEZ-AVALOS, J. G., J. GOLUBOV, M. C. MANDUJANO, AND E. JURADO. 2007. Causes of individual mortality in the endangered star cactus Astrophytum asterias (Cactaceae): The effect of herbivores and disease in Mexican populations. J. Arid Environ. 71: 250–258. MARTORELL, C. AND E. PETERS. 2005. The measurement of chronic disturbance and its effects on the threatened cactus Mammillaria pectinifera. Biol. Conserv. 124: 119–207. MCINTOSH, M. E. 2002. Plant size, breeding system, and limits to reproductive success in two sister species of Ferocactus (Cactaceae). Plant Ecol. 162: 273–288. MCINTOSH, M. E. 2005. Pollination of two species of Ferocactus: interactions between cactus-specialist bees and their host plants. Funct. Ecol. 19: 727–734. NAVARRO, L. AND J. GUITIA´N. 2002. The role of floral biology and breeding system on the reproductive success of the narrow endemic Petrocoptis viscosa Rothm. (Caryophyllaceae). Biol. Conserv. 103: 125–132. NEGRO´N-ORTIZ, V. 1998. Reproductive biology of a rare cactus, Opuntia spinosissima (Cactaceae), in the Florida Keys: why is seed set very low? Sex. Plant Reprod. 11: 208–212. OAXACA-VILLA, B., A. CASAS, AND A. VALIENTEBANUET. 2006. Reproductive biology in wild and silvicultural populations of Escontria chiotilla (Cactaceae) in the Tehuaca´n Valley, Central Mexico. Genet. Resour. Crop Evol. 53: 227–287.

2011]

MARTI´NEZ-PERALTA AND MANDUJANO: REPRODUCTIVE ECOLOGY OF ROCK CACTUS

PETTERSON, M. W. 1991. Flower herbivory and seed predation in Silene vulgaris (Caryophyllaceae): effects of pollination and phenology. Holarctic Ecology 14: 45–50. PIMIENTA, E. AND R. F. DEL CASTILLO. 2002. Reproductive Biology. p. 75–90. In P. S. Nobel [ed.], Cacti: Biology and Uses. University of California Press, Berkeley, CA. PINKAVA, D. 1984. Vegetation and flora of the Bolso ´ n of Cuatro Cie´negas region, Coahuila, Me´xico: IV. Summary, endemism and corrected catalogue. J. Arizona & Nevada Acad. Sci. 19: 23–47. PIN˜A, H. H., C. MONTAN˜A, AND M. C. MANDUJANO. 2007. Fruit abortion in the Chihuahuan-Desert endemic cactus Opuntia microdasys. Plant Ecol. 193: 305–313. PIN˜A, H. H., C. MONTAN˜A, AND M. C. MANDUJANO. 2010. Olycella aff. junctolineella (Lepidoptera: Pyralidae) florivory on Opuntia microdasys, a Chihuahuan Desert endemic cactus. J. Arid Environ. 74: 918–923. PRIMACK, R. B. 1985. Patterns of flowering phenology in communities, populations, individuals, and single flowers. p. 571–593. In J. White [ed.], The Population Structure of Vegetation. Dr. W. Junk Publishers, Dordrecht, NL. ROUTLEY, M. B., K. MAVRAGANIS, AND C. G. ECKERT. 1999. Effect of population size on the mating system in a self-compatible, autogamous plant, Aquilegia canadensis (Ranunculaceae). Heredity 82: 518–528. SA´NCHEZ-LAFUENTE, A. M. 2007. Corolla herbivory, pollination success and fruit predation in complex flowers: an experimental study with Linaria lilacina (Scrophulariaceae). Ann. Bot.-London 99: 355–364. SA´NCHEZ-MEJORADA, R. H., E. F. ANDERSON, N. P. TAYLOR, AND N. P. R. TAYLOR. 1986. Succulent

155

plant conservation studies and training in Mexico. Stage 1, Part I. World Wildlife Fund U.S. Washington D.C. SAUNDERS, N. E. AND S. D. SIPES. 2006. Reproductive biology and pollination ecology of the rare Yellowstone Park endemic Abronia ammophila (Nyctaginaceae). Plant Spec. Biol. 21: 75–84. SCHEMSKE, D. W., B. C. HUSBAND, M. H. RUCKELSHAUS, C. GOODWILLIE, I. M. PARKER, AND J. G. BISHOP. 1994. Evaluating approaches to the conservation of rare and endangered species. Ecology 75: 584–606. SCHMALZEL, R. J., F. W. REICHENBACHER, AND S. RUTMAN. 1995. Demographic study of the rare Coryphanta robbinsorum (Cactaceae) in Southeastern Arizona. Madron˜o 42: 332–348. STRONG, A. AND P. S. WILLIAMSON. 2007. Breeding system of Astrophytum asterias: an endangered cactus. Southwest. Nat. 52: 341–346. TRAPNELL, D. W. AND J. L. HAMRICK. 2006. Floral display and mating patterns within populations of the neotropical epiphytic orchid, Laelia rubescens (Orchidaceae). Am. J. Bot. 93: 1010–1018. UYENOYAMA, M. K. 1993. Genetic incompatibility as a eugenic mechanism. p. 60–73. In: N. W. Thornhill [ed.], The Natural History of Inbreeding and Outbreeding: Theoretical and Empirical Perspectives. The University of Chicago Press, Chicago, IL. WEPPLER, T. AND J. STO¨CKLIN. 2006. Does predispersal seed predation limit reproduction and population growth in the alpine clonal plant Geum reptans? Plant Ecol. 187: 277–282. ZAPATA, T. R. AND M. T. K. ARROYO. 1978. Plant reproductive ecology of a secondarydeciduous tropical forest in Venezuela. Biotropica 10: 221–230.

Suggest Documents