Seed rain from forest fragments into tropical pastures in Los Tuxtlas, Mexico

Plant Ecology 145: 255–265, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands.

255

Seed rain from forest fragments into tropical pastures in Los Tuxtlas, Mexico C. Mart´inez-Garza1 & R. Gonz´alez-Montagut2,∗∗ ´ UNAM. A.P. 70-275 Mexico City 04510 Mexico; (present addresses: University of Illinois, de Ecologia, Department of Biological Sciences (M/C 066), Chicago, Illinois 60607-7060, USA; (present address: Mexican Nature Conservation Fund, Mexico City 03900, Mexico) 2 Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, MA 02138, USA 1 Instituto

Received 7 July 1998; accepted in revised form 3 July 1999

Key words: Germinability, Natural regeneration, Plant strategies, Riparian vegetation, Seed dispersal

Abstract Deforestation has resulted in the fragmentation of forests. Remnant fragments are widely assumed to be sources of seeds for forest regeneration in abandoned pastures. The seed rain in 12 pastures at Los Tuxtlas, Mexico, their relationship with riparian vegetation adjacent to them, and their closeness to the reserve is described. For all the species found in the seed rain, we classified them by typical habitat (pasture or rain forest), life form, dispersal syndrome, and plant strategy (pioneer or non-pioneer species). In addition, germinability was evaluated for all seeds. We also assessed the correlation in composition of the seed rain and the riparian vegetation. Only 11% of the total species that occur in the Los Tuxtlas region reach pastures via dispersal from forest fragments. However, nearly 80% of the species in the seed rain did not come from the fruiting individuals in the adjacent riparian vegetation. The proportion of dispersal syndromes, life forms and plant strategies of the species in the seed rain was similar to those observed in the rain forest. On average, the germinability of forest species was less than 30%. The forest species richness was similar in the seed rain in pastures and inside the forest, but the pasture seed rain contained fewer seeds per species. Pastures have a high potential for natural regeneration because of seed dispersal from adjacent forest. However, the forest that regenerates in pastures close to the reserve is expected to contain different species than the forest regenerating far away from it.

Introduction During the last ten years tropical rain forests have had an annual loss of 0.8% of its area, mostly occurring in the Americas (Whitmore 1997). As a result of this process, the landscape now consists of a mosaic of pastures, forest fragments, and agricultural areas. In Mexico, some pastures may be abandoned as a result of changes in land use and the halt on subsidies for cattle ranching. For re-establishment of forest vegetation in abandoned pastures, seeds which arrive via dispersal events from adjacent forest fragments play an important role (Gómez-Pompa & Vázquez-Yanes 1981; Nepstad et al. 1990). Consequently, the seed rain in pastures is a measure of the instant regenerative potential. To evaluate this potential, the dispersal

of forest species into pastures from the surrounding landscape needs to be assessed. Specifically, we must know what species arrive, which of them remain viable and their establishment requirements. In order to predict some of the resources required for tree species establishment, a plant strategy classification has been developed by Martínez-Ramos (1985) and Popma et al. (1992): Obligate-gap or pioneer species are those that start and complete their life cycle in gaps; non-pioneer species are divided in two groups: gap-dependent species require small gaps to pass through one or several stages in their life cycle, but they can survive in the understory for prolonged periods as juveniles, and gap-independent species that are able to complete their entire life cycle in the shade.

256 One of the potential sources of rain forest seeds to pastures in the Los Tuxtlas region, Mexico, is the remaining riparian vegetation. These forest fragments are left in areas adjacent to streams and rivers because they prevent river desiccation and provide shelter and fruits for people and cattle (Guevara et al. 1997). Furthermore, they create a microenviromental gradient of light, temperature and humidity adjacent to the pastures that could influence seed germination and establishment of forest species (González-Montagut 1996). In addition to abundant riparian fragments, the landscape contains a large tract of continuous forest, of which the riparian vegetation is a subsample. If natural regeneration is allowed, future forest may largely consist of the species from the closest riparian fragment and those species coming from the reserve. This study quantified the seed rain along a riparian vegetation-pasture gradient in the Los Tuxtlas region. The area is well-suited for analysis of pasture seed rain from riparian vegetation because riparian vegetation is represented by a web of forest remnants across the landscape. We sampled the seed rain in pastures at different distances from riparian vegetation to explore: (1) proportion of species belonging to different vegetation types (rain forest or pasture), dispersal syndromes, life forms and plant strategies; (2) germination capacity of all dispersed species; (3) the correlation between main seed source (closest riparian forest) and the proximity of the sites to the reserve, and (4) temporal variation of the seed rain.

Study site and methods Study site This study was conducted in pastures within a 3 km radius of Los Tuxtlas Biological Station (LTBS) in the state of Veracruz, southeast Mexico (18 ◦ 300 and 18◦400 N, 95◦030 and 95◦ 100 W). This is the northernmost rain forest in the neotropics. Soil is sandy loam, and classified as vitric andosols (FAU/UN 1975 in Soto-Esparza 1976). Mean annual temperature is 27 ◦ C and mean annual rainfall is 4900 mm. A dry season extends from March to May, and a rainy season from June to February. LTBS lies within a reserve of 640 ha of lowland tropical rain forest. This reserve is surrounded by forest remnants, agricultural lands and mostly pastures. Paspalum conjugatum, Axonopus compressus, Pan-

Table 1. Similarity index between the species in the riparian vegetation and those in the seed rain and their relation to the reserve in Los Tuxtlas, Mexico. Site/ Focal tree

Relation to the reserve Similarity index (%)

1 Cecropia 2 Cecropia 3 Cecropia 4 Stemmadenia 5 Stemmadenia 6 Stemmadenia 7 Heliocarpus 8 Heliocarpus 9 Heliocarpus 10 11 12

Far away Far away Far away Close Close Far away Close Close Close Far away Far away Far away

22 23 29 5 9 24 13 10 20 21 26 28

icum spp., and the introduced African stargrass, Cynodon plectostachyus are the most important species in the pastures (Guevara et al. 1992). We selected 12 sites of riparian vegetation adjacent to pastures in Los Tuxtlas (Figure 1). In each site, we chose a tree located on the edge of the riparian vegetation (named focal tree hereafter). Focal trees represent three common species with different dispersal syndromes. We selected two pioneer species with contrasting dispersal syndromes: Heliocarpus appendiculatus (wind-dispersed) and Cecropia obtusifolia (animal-dispersed) and one non-pioneer species Stemmadenia donnell-smithii (bird-dispersed). Three other focal locations adjacent to riparian vegetation were selected at random in terms of focal tree. The search of pasture of at least 50 m2 and riparian fragments suitable for the study results in five sites near (less than 50 m) and seven sites far (more than 1.7 km) from the forest reserve (Figure 1, Table 1). This array allows us to compare the seed rain at two spatial levels: at different distances from the riparian fragments and at two distances from the reserve. We placed seven seed traps every 7 m along a transect running perpendicular to the stream from the focal tree to the open pasture (Figure 2). In addition, three traps were placed along a diagonal between the stream and the perpendicular line. Finally, the sites with Heliocarpus appendiculatus as a focal tree had a total of six traps along two diagonals plus the 7 traps in the perpendicular. Diagonal traps were placed at 14, 28 and 42 m parallel to the riparian vegetation. This system increased the area sampled for seed rain.

257

Figure 1. Map of Los Tuxtlas region, Veracruz, Mexico and location of the sites near and far from the reserve. After Ricker (1998).

258 ples were, on average, two months in storage before they were planted. Each seed sample was allowed to germinate for three to five months. Germinability was registered monthly. Data analysis

Figure 2. Experimental design showing the arrangement of the sites with 13 seed traps ( ) and focal tree ( ) in the riparian forest into the pastures.

2

#

Seed traps were built with four 1.5 m poles stuck into the ground to support 1 m2 square cloth (0.5 mm aperture) that was sewn to a line of barbed wire and nailed at 0.4 m from the ground. Traps were protected from perching birds, cattle, and insect predation. We had a total of 120 m2 of capture area with these traps. We collected seeds fortnightly from February to August 1994, a period that included both the dry and rainy seasons. Seeds were identified in most cases to species level and only in a few cases to the genus level (we used reference collections from the Herbarium of the National University of Mexico; De Konrinch 1973; Hitchcock 1951; Díaz 1985). Nomenclature follows Ibarra-Manríquez & Sinaca (1995, 1996a, b). Identified species were classified according to vegetation type, life form, plant strategy (for forest trees), and dispersal syndrome, according to the available literature (A. Castillo & G. Ibarra-Manríquez unpublished data; Ibarra-Manríquez & Sinaca 1995, 1996a, b; Martínez-Ramos 1985; Orozco-Segovia et al. 1985; Popma et al. 1992; Soto-Castro 1992; Trejo 1976; Van Dorp 1985; Sarukhán 1980). To evaluate the relationship between the number of species in riparian vegetation and in the seed rain reaching pastures, we identified all reproductive woody species along 50 m of riparian vegetation. This included 25 m on each side of the focal tree and all the width of the fragment at one side of the river (approximately 10 m wide). Migrant species were those present in the seed rain, but that did not occur in the 50 m of riparian vegetation under study. We obtained height, crown radius, and diameter at breast height of each reproductive woody species. Collected seeds were planted in a greenhouse at LTBS in order to evaluate germinability. Seed sam-

The data from the total 120 m2 of capture area (traps in perpendicular and in diagonal) was used to quantify the seed rain in the 12 pasture-riparian vegetation gradient. To compare among sites we used the data from the 10 traps, 7 located in the perpendicular and 3 located in one diagonal of each site. Correspondence between the species in the riparian vegetation and those in the seed rain were compared using the Similarity Index SI= 2C/A + B where A is the number of woody species in riparian vegetation, B is the number of woody species represented in the seed rain and C equals the number of species shared between the two sites (Krebs 1985). To compare the similarity indices of sites near (n= 5) and far (n= 7) from the forest reserve we used a Mann–Whitney U test.

Results Description of the seed rain: the floristic assemblage In the total seed rain (94 212 seeds), we identified 115 species plus three other taxa identifiable only to genus in 49 families (Appendix 1). Poaceae, Asteraceae, Fabaceae, Euphorbiaceae, and Moraceae were the most important families in terms of the number of species represented (13, 9, 7, 7 and 5 species respectively). Among the pasture species, Poaceae, Cyperaceae, and Asteraceae were the most abundant families; among the rain forest species, Asteraceae, Fabaceae, and Araceae were the best represented families. Of the species in the seed rain, 67.8% (80 species) were rain forest species and 32.2% (38 species) were pasture species. The total number of seeds from forest species (54 751) was almost 1.5 times greater than the number of seeds from pasture species (37 475). Dispersal syndromes Most rain forest species in the seed rain were animaldispersed (61.25%) followed by wind-dispersed species (36.25% and 5899 seeds) and gravitydispersed species (2.5%; Figure 3).

259

2



Figure 3. Total number of rain forest ( ) and pasture species ( ) found in the seed rain between riparian forest and pastures according to dispersal syndrome.

Among the pasture species present in the seed rain, only a low percentage belonged to winddispersed species (23.7%) compared to the percentage of animal-dispersed species (42.1%). Nonetheless, the number of wind-dispersed seeds (27 155) was five times greater than animal-dispersed seeds (4759). There were only three gravity-dispersed pasture species but the number of these seeds (3254) was also higher than the number of animal-dispersed seeds (2232). Finally, 10.5% (3935) of the seeds of pasture species had an unknown dispersal syndrome. Plant strategies Seeds of 44 tree species and one palm were found in the seed rain of which 12 were identified as pioneer species (obligate-gap) and 33 were identified as non-pioneer species. Among the latter, 22 were gap-dependent and 11 were gap-independent species (sensu Martínez-Ramos 1985; Popma et al. 1992) (Figure 4). The total amount of tree seeds collected was 51 104 of which 65.5% belonged to pioneer species and the remaining 34.5% were from nonpioneer species (33.3% of gap-dependent and 1.1% of gap-independent). Fifty-five percent of the pioneer seeds were from C. obtusifolia and 17% of the gap-dependent seeds were from Ficus spp. For gap-dependent species, three groups of species are recognized on the basis of their seed size and dispersal syndrome (Martínez-Ramos 1985): animaldispersed species with large seeds (> 5 mm in diameter), animal-dispersed species with small seeds (< 5 mm), and wind-dispersed seeds of any size. We recorded nine large seeded animal-dispersed species which contributed less than 1% of total seeds obtained (764 seeds). Six small seeded animal-dispersed

2



Figure 4. Number of species ( ) and seeds ( ) of rain forest trees according to plant strategy found in the seed rain in a riparian forest-pasture gradient in Los Tuxtlas.

species contributed to the largest number of seed in this plant strategy category (16 015 seeds). This represents 94% of the total amount of seeds from gap-dependent species. Seven gap-dependent winddispersed species contributed fewer seeds than the other groups (237 seeds). Among the gap-intolerant species, one is wind-dispersed, nine are dispersed by birds and/or bats, and one species, Astrocaryum mexicanum, is dispersed by squirrels. Life form Forty-five species were trees and they had largest number of seeds (51 135). Lianas followed with 20 species (25% of the total number of species) with proportionally fewer seeds. Of the 20 species of lianas registered (2060 seeds), 14 were wind-dispersed (1933 seeds) and six were animal-dispersed (127 seeds, 107 from Cissus microcarpa). The remaining 17.5% of the species corresponded to herb (7.5%), vines (3.7%), epiphytes (5%), and palms (1.2%). Aechmea bracteata represented 97.7% of the epiphytes seeds. Among the pasture species, the most important life form was herb with 32 species (84.2%) and 37 934 seeds. Eleven of the herb species (34%) and 30 222 of the herb seeds (81.8%) were grasses. Finally, 10.5% of species were mainly from introduced tree species. Relationship between number of species in the riparian vegetation and in the seed rain We found 292 reproductive trees and lianas of 84 woody species in the riparian vegetation, belonging to 35 families. The most important species by number of individuals were Croton schiedeanus (n=66),

260 Sapium nitidum (n=25), Cupania glabra (n=20), and Tabernaemontana alba (n=13). Fifty-four percent of the species found in the riparian vegetation were represented only by one individual. The DBH of riparian vegetation trees was, on average, 21.86 cm ± 0.88 SE and the height was on average 10.61 cm ± 0.89 SE. Kendall’s coefficient comparing the number of woody species in the riparian vegetation and in the seed rain in the adjacent pasture was not significant (τ =0.112; P>0.6). The number of species in the riparian vegetation was not associated with the number of species in the seed rain adjacent to them. The similarity index between the woody species in the riparian vegetation and the species of the seed rain from the pasture next to it was low (≤ 29%, Table 1). Few woody species present in riparian vegetation arrived to the adjacent pasture. A test of the similarity index between sites close (n= 5) and distant to the reserve (n= 7) revealed major differences between these groups (Mann–Whitney U = −2.84; P< 0.01). The similarity index of the closer sites was lower (0.115 ± 0.03 SD) than the index of those sites far from the reserve (0.247 ± 0.01 SD). In the sites close to the reserve, seeds arriving to the pasture seem to come from forest fragments other than the closest 50 m of riparian vegetation whereas far from the reserve, seeds arriving to pasture came from the adjacent riparian vegetation.



Germinability of all collected seeds Germinability of pasture species was on average 32.47% ± 17.71 SD and that of forest species was on average 25.9% ± 22 SD. Seeds of gap-dependent species achieved the highest germinability (29.8% ± 23.26 SD) which was very similar to those of obligate-gap species (29.05% ± 19.92 SD), followed by gap-independent species, which had the lowest germinability (18.73% ± 9.97 SD). Seed rain at different distances from the riparian vegetation and temporal variation A decrease of forest seeds in the seed rain occurred as distance from the riparian vegetation increased while the number of forest species, pasture species and pasture seeds remain similar through the entire gradient (Figure 5). The greatest number of species was found in the June seed rain, while the largest number of seeds was obtained in July (Figure 6). A great number of forest seeds (19 915), which corresponded to 35 species

2



Figure 5. Total number of rain forest ( ) and pasture species ( ), and total number of rain forest ( ) and pasture seeds ( ) of all traps of the riparian forest-pasture gradient. The final values are averages ± standard deviation.

2





Figure 6. Total number of seeds ( ) and species ( ) of all traps combined for rain forest and pasture species for each month of collection. Seed rain in a riparian system-pasture gradient in Los Tuxtlas.

was recorded in August. In contrast, 11 002 seeds of 43 species were collected in July. With respect to pasture species, a dispersal peak of dry fruits occurred in the dry season (19 species) whereas fleshy fruits showed an increase in number of species (6 species) during the rainy season.

Discussion The large number of forest species in the seed rain indicates that there is a good potential for natural regeneration of forest. The number of species and seeds arriving in pastures is related primarily to the distance from the riparian vegetation: there are more species and seeds close to the riparian vegetation (Figure 5).

261 Furthermore, the results strongly suggest that vicinity to the reserve modify the pasture seed rain in terms of species identity. The species arriving to pastures located far from the reserve seems to come from the adjacent riparian systems while nearer the reserve, species composition in the seed rain does not resemble the composition of the adjacent riparian forest. The floristic assemblage We obtained only 10.8% of the total species from the seed rain adjacent to riparian systems in the reference inventory for Los Tuxtlas forest (see Ibarra-Manríquez & Sinaca 1995, 1996a, b). This great reduction of species could be caused by the differential contribution of the distinct sources of rain forest seeds, the efficiency of the dispersal vectors, the phenology of species in fruit at the time of our study and the species represented in the forest fragments. Relationship between number of species in the riparian vegetation and in the seed rain The sources of forest seeds in this study can be from the closest riparian fragment (local species) and other forest fragments (migrant species). Similarity indices showed a strong relationship with the proximity of each site to the largest forest patch (the reserve). The similarity index of the sites close to the reserve was low, meaning that many species in that seed rain were probably coming from the reserve which explains the low similarity (Table 1). The closeness of the reserve indicates a different input of species: these species reach pastures via animals and wind. In pastures far away from the reserve, every riparian fragment seems to be composed of the same set of species and this results in higher similarity indices for these sites. The percentage of animal dispersed species in the seed rain reaching pastures (71.1%) is lower than the seed rain inside the forest (88.4%; Soto-Castro 1992). Frugivore diversity in pastures is restricted to those species that can forage in agricultural lands and remnant forests such as riparian vegetation (Estrada et al. 1993). The percentage of animal-dispersed species arriving to pastures suggests that only a subset of frugivorous animals which forage and disperse seeds inside the forest go out to pastures or riparian vegetation. Germination and plant strategies Plant strategy traits of the species that reach pastures allow us to make inferences about the regenerative po-

tential of these sites. The low percentage of pioneer species (sensu Martínez-Ramos 1985) in pasture seed rain agrees with the low percentage of pioneer species found in other forests (Vázquez-Yanes & OrozcoSegovia 1984; Howe 1989). However, pioneer species were well represented in terms of the number of seeds (Figure 4). The great vagility of pioneer species due to their small seed size could affect their capacity to establish in sites with poor resources (Harper 1977; Willson 1993). Pioneer seedlings have a low probability to survive predation and drought conditions that exist in these sites (Careaga 1989; González-Montagut 1996; Guevara & Laborde 1993; Nepstad et al. 1990). For non-pioneer species, gap-dependent species dispersed by animals were the most abundant in terms of number of seeds (Figure 4). Gap-dependent species with large seeds are likely to germinate in open sites and they could produce seedlings which resist the adverse conditions of pastures (Nepstad et al. 1990). Nonetheless, they have poor dispersal to pastures (Sanchez-Garduño 1995) and when dispersed their seeds lose water rapidly and die in open sites (Vázquez-Yanes & Orozco-Segovia 1993). Those species reaching pastures may have different regeneration strategies, but they seem to have similar germinability under ideal conditions. Given this seed rain, the resulting forest could resemble the original one in terms of plant strategy group proportions if pastures could function as big gaps. Pasture species are a barrier for establishment of rain forest species (Nepstad et al. 1990; GonzálezMontagut 1996). Pasture weedy species were well represented in the seed rain because they proliferate with cattle activities, overgrazing and lack of appropriate management (Guevara et al. 1997). Introduced tree species, for example: Psidium guajaba and Byrsonima crassifolia are found as remnant trees in the pastures. They could be sources of food and perches in the middle of pastures for frugivorous birds, and thus catalyze regeneration even when native species are more important in this regard (Guevara and Laborde 1993). This study shows that in the pasture seed rain there are rain forest species with the same proportion of dispersal syndromes and plant strategies as inside the rain forest. Many of the species that arrive to pastures might not establish in the early stages of natural regeneration but their presence in the seed rain enhances the regenerative potential of those sites. A great proportion of the species that occur in the pasture seed rain does not come from the adjacent

262 riparian vegetation. This emphasizes the importance of all rain forest fragments. The forest that regenerates in pastures from riparian vegetation is expected to contain different species than the forest regenerating from the reserve because riparian vegetation only contains a subset of the species that occur in the larger reserve. Of those species arriving to the pasture, few will successfully establish due to the hard conditions in pastures. We have shown that there is good potential for natural regeneration in pastures as seen in the high number of seeds in the seed rain. Pasture seed rain also showed temporal variation similar to that inside the forest. However, the conditions within pastures present barriers for establishment of rain forest species. Our current research involves the investigation of the factors affecting the establishment of rain forest species in pastures.

Acknowledgements The NSF-Doctoral Dissertation Improvement Grant DEB. given to Renee Gonzàlez-Montagut provided the financial support for this work; The Institute of Ecology, National University of Mexico (UNAM) provided the facilities during the writing of the paper. An early draft of this paper was enriched with the help of Rodolfo Dirzo and Juan Fornoni. CM-G thanks her colleagues from University of Illinois for repetitive revisions: Nobby Cordeiro, Ruthie Harari-Kremer, Hank Howe, Maria Miriti, Manolo Pacheco and Sonali Saha.

References Careaga, S.1989. Efecto de la variación en el tamaño de las semillas sobre el desempeño de las plántulas de especies tropicales. Bachelor Thesis, School of Sciences, UNAM. Mexico. De Konrinch, M. E. 1973. Gramíneas. Colección Aula. Volumen 3. Editorial Universitaria Guatemala. Díaz, C. 1985. Manual de gramíneas. UAM-Xochimilco. Mexico. Estrada, A., Coates-Estrada, R., Meritt, D. Jr., Montiel, S. & Curiel, D. 1993. Patterns of frugivore species richness and abundance in forest islands and agricultural habitats at Los Tuxtlas, Mexico. Vegetatio 107/108: 245–257. Gómez-Pompa, A., & Vázquez-Yanes, C. 1981. Successional studies of a rain forest in México. Pp. 246–266. In: West, D. C., Shugart, H. H. & Butkin, D. B. (eds), Forest Succession Concepts and Applications. Springer-Verlag, New York. González-Montagut, R. 1996. Establishment of three rain forest species along the riparian corridor- pasture gradient in Los Tuxtlas, Mexico. Ph. D. Dissertation. Harvard University, Cambridge, Massachusetts, USA.

Guevara, S., Meave, J., Moreno-Casasola, P. & Laborde, J. 1992. Floristic composition and structure of vegetation under isolated trees in neotropical pastures. J. Veg. Sci. 3: 655–664. Guevara, S. & Laborde, J. 1993. Monitoring seed dispersal at isolated standing trees in tropical pastures: consequences for local species availability. Vegetatio 107/108: 319–338. Guevara, S., Laborde, J., Liesenfeld, D. & Barrera, O. 1997. Historia natural de los potreros de Los Tuxtlas, Veracruz. Pp. 43–57. In: González, F., Dirzo, R. & Vogt, R. (eds), Historia Natural de Los Tuxtlas. CONABIO, Instituto de Biología, Centro de Ecología, UNAM Mexico. Harper, J. L. 1977. Population biology of plants. Academic Press, London. Hitchcock, A. S. 1951. Manual of the grasses of the United States. United States Department of Agriculture. Miscellaneous Publication. No. 200, USA. Howe, H. F.1989. Scatter -and clump- dispersal and seedling demography: hypothesis and implications. Oecologia 79: 417–426. Ibarra-Manríquez, G. & Sinaca, S. 1995. Lista florística comentada de la Estación de Biología Tropical ‘Los Tuxtlas’, Veracruz, México. Rev. Biologia Trop. 43: 75–115. Ibarra-Manríquez, G. & Sinaca, S. 1996a. Lista florística comentada de la Estación de Biología Tropical ‘Los Tuxtlas’, Veracruz, México (Mimosaceae a Verbenaceae). Rev. Biologia Trop.44: 41–60. Ibarra-Manríquez, G. & Sinaca, S. 1996b. Lista florística comentada de la Estación de Biología Tropical ‘Los Tuxtlas’, Veracruz, México (Violaceae a Zingiberaceae). Rev. Biologia Trop. 44: 427–447. Krebs, C. J. 1985. Ecología: Estudio de la distribución y la abundancia. Editorial Harla, Mexico. Martínez-Ramos, M.1985. Claros, ciclos vitales de los árboles tropicales y regeneración natural de las selvas altas perennifolias. Pp. 191–239. In: A.Gómez-Pompa, & Del Amo, S. (eds), Investigaciones sobre regeneración de selvas altas en Veracruz. Vol II. Alhambra mexicana. Nepstand, D., Uhl, Ch. & Serrao, A. E. 1990. Surmounting barriers to forest regeneration in abandoned highly degraded pastures: a case study from Paragominas, Pará, Brasil. Pp. 15–229. In: Anderson, A. B.(eds), Alternatives to deforestation steps toward sustainable use of the Amazonas. Columbia University Press, New York. Orozco-Segovia, A., Vázquez-Yañes, C., Armella, M. A. & Correa, N. 1985. Interacciones entre una población de murciélagos de la especie Artibeus jamaicensis y la vegetación del área circundante, en la región de Los Tuxtlas, Veracruz. Pp. 365–378. In: Gómez-Pompa, A. & Del Amo, S. (eds), Investigaciones sobre regeneración de selvas altas en Veracruz. Vol. II. Alhambra mexicana, Mexico. Popma, J., Bongers, F. & Werger, M. J. A. 1992. Gap-dependence and leaf characteristics of trees in a tropical lowland rain forest in Mexico. Oikos 63: 207–214. Sánchez-Garduño, C.1995. Ecología de la depredación predispersión de semillas de Nectandra ambigens en Los Tuxtlas. Bachelor Thesis, School of Sciences, UNAM. Mexico, 96 pp. Sarukhán, J. 1980. Demographic problems in tropical systems. Pp. 161–188. In: Solbrig, O. T. (eds), Demography and evolution in plant populations. Blackwell Scientific Publications, Oxford. Soto-Castro, A. 1992. Patrones espaciales y temporales de la lluvia de semillas de grupos de árboles en una selva húmeda de México. Bachelor Thesis, School of Sciences, UNAM. Mexico. Soto-Esparza, M. 1976. Algunos aspectos climáticos de la región de ‘Los Tuxtlas’. Pp. 70–111. In: Gómez-Pompa, A., Vázquez-

263 Yanes, C., Del Amo, S. & Butana A. (eds), Regeneración de selvas. CECSA. Mexico. Trejo, P. L. 1976. Diseminación de semillas por aves en Los Tuxtlas, Veracruz. Pp. 447–470. In: Gómez-Pompa, A., Vázquez-Yanes, C., Del Amo, S. & Butana, A. (eds), Regeneración de selvas. CECSA. Mexico. Van Dorp, D. 1985. Frugivoría y dispersión de semillas por aves. Pp. 333–364. In: Gómez-Pompa, A. & Del Amo, S. (eds), Investigaciones sobre regeneración de selvas altas en Veracruz. Vol. II. Alhambra mexicana, S. A. Mexico. Vázquez-Yanes, C., & Orozco-Segovia, A. 1984. Fisiología ecológica de las semillas de árboles de la selva trópical: Un reflejo de su ambiente. Ciencia 35: 191–210.

Vázquez-Yanes, C. & Orozco-Segovia, A. 1993. Patterns of seed longevity and germination in the tropical rain forest. Ann. Rev. Ecol. Syst. 24: 69–87. Whitmore, T. C. 1997. Tropical forest disturbance, dissappearance and species loss. Pp. 3–12. In: Laurance, W. F. & Bierregaard, R. O. Jr. (eds), Tropical Forest Remnants. University of Chicago Press, USA. Willson, M. F. 1993. Dispersal mode, seed shadow and colonization patterns. Vegetatio 107/108: 261–280.

Appendix 1. Identified seeds obtained in traps in a riparian system-pasture gradient in Los Tuxtlas, Ver. It includes species, family, life form (h: herb; l: liana; t: tree; e: epiphyte; et: epiphyte tree; he: hemiepiphyte p: palm); the vegetation to which each species belongs (Veg. p: pastures and roads; rf: rain forest); life history (only for rain forest trees, og: obligate-gap; gd: gap-dependent species; gi: gap-independent); the number of collected seeds (n) per species, its percentage (%) respect to the total (94,212) and dispersal syndrome (D.S. a: animal dispersed; w: wind dispersed; g: gravity dispersed). ∗ This species is considered to be dispersed by squirrels or by gravity. Species

Family

Lifeform Type veg Plant strategy Seeds D.S. n %

Abuta panamensis (Standl.) Krukoff et Barneby Acacia cornigera (L.) Willd. Acacia sp. Acalypha arvensis Poepp. & Endl. Achyranthes aspera L. Achyranthes sp. Aechmea bracteata (Sw.) Griseb. Albizia purpusii Britton et Rose Alchornea latifolia Sw. Ampelocera hottlei (Standl.) Standl. Andropogon bicornis L. Anthurium schlechtendalii Kunth in H.B.K. Aphelandra aurantica (Scheidw.) Lundell Aristolochia ovalifolia Duch. Arrabidaea verrucosa (Standley) A. Gentry Asclepias curassavica L. Astrocaryum mexicanum Liebm. ex Mart. Axonopus compressus (Swartz) Beauv. Blechum brownei (L.) Ant. Juss. Borreria laevis (Lam.) Griseb. Bursera simaruba (L.) Sarg. Byrsonima crassifolia (L.) Kunth in H. B. K,. Calathea lutea (Aubl.) G. Mey. Capsicum annum L. Cecropia obtusifolia Bertol. Cedrela odorata L. Cissus microcarpa Vahl Citharexylon affine D. Don. Citrus sp. Clusia flava Jacq. Coccoloba montana (L.) Britt. et Rose Conostegia xalapensis (Kunth) G. Don ex DC.

Menispermaceae Mimosaceae Mimosaceae Euphorbiaceae Amaranthaceae Amaranthaceae Bromeliaceae Mimosaceae Euphorbiaceae Ulmaceae Poaceae Araceae Acanthaceae Aristolochiaceae Bignoniaceae Asclepiaceae Arecaceae Poaceae Acanthaceae Rubiaceae Burseraceae Malpighiaceae Marantaceae Solanaceae Cecropiaceae Meliaceae Vitaceae Verbenaceae Rutaceae Clusiaceae Polygonaceae Melastomatacea

l t t h h h e t t t h e h l l h p h h h t t h h t t l t t et t t

rf rf rf p p p rf rf rf rf p rf rf rf rf p rf p rf p rf p rf p rf rf rf rf p rf rf rf

– og og – – – – gd gd gd – – – – – – gi – – – gd – – – og gd – gd – – st p

2 0.002 10 0.011 78 0.083 856 0.909 55 0.058 10 0.011 384 0.408 7 0.007 193 0.205 4 0.004 477 0.506 1 0.001 4 0.004 5 0.005 1 0.001 34 0.036 22 0.023 1270 1.348 994 1.06 25 0.027 195 0.207 11 0.012 1 0.001 379 0.402 28426 30.17 44 0.047 107 0.114 9 0.01 2 0.002 53 0.056 3 0.003 2064 2.191

a a a a a a a w a a w a g w w w a-g w w g a a a a a w a a a a a a

264 Appendix 1. Continued. Species

Family

Lifeform Type veg Plant strategy Seeds n

Coussapoa purpusii Standl. Crotalaria sp. Croton schiedeanus Schldl. Cupania glabra Sw. Cymbopetalum bailloniiR. E. Fries. Cyperus laxus Lam. Chaptalia nutans (L.) Polak. Dalbergia glomerata Hemsl. Desmodium adscendens (Sw.) DC. Desmodium macrodesmum S.F. Blake Dioscorea composita Hemsl. Erythroxylum tabascense Britton. Eugenia capuli (Schltdl. et Cham.) O. Berg. Eupatorium galeottii Robins. Ficus eugeniaefolia (Liebm.) Hemsl. Ficus insipida Willd. Ficus sp. Fimbristylis dichotoma (L.) Vahl. Gouania lupuloides (L.) Urban. Heliocarpus appendiculatus Turcz. Homolepsis aturensis (Kunth) Chase Ipomoea phillomega (Vull.) House Ipomoea sp. Iresine arbuscula Uline & Bray. Lantana hirta Graham. Lasiacis divaricata (L.) Hitch. Lasianthaea fruticosa Leptocoryphium lunatum (H.B.K.) Nees. Lonchocarpus guatemalensis Benth. Lunania mexicana Brandege Lycianthes heteroclita (Sendtner) Bitter Lycianthes nitida Bitter Machaerium floribundum Benth. Mascagnia vacciniifolia Niedz. Melothria pendula L. Mikania leiostachya Benth. Mimosa pudica L. Monstera acuminata G..Koch. Muhlenbergia rigida (H.B.K.) Kunth Oreopanax liebmanii Marchal Panicum havardii Vasey Parathesis psychotrioides Lundell. Paspalum conjugatum Bergius. Paspalum virgatum L. Pavonia schiedeana Steud. Pennisetum villosum Philodendron sp. Phyllanthus niruri L. Phytolacca rivinoides Kunth & Bouch´e Piper amalago L.

Cecropiaceae Fabaceae Euphorbiaceae Sapindaceae Annonaceae Cyperaceae Asteraceae Fabaceae Fabaceae Fabaceae Dioscoreaceae Erythroxylaceae Myrtaceae Asteraceae Moraceae Moraceae Moraceae Cyperaceae Rhamnaceae Tiliaceae Poaceae Convolvulaceae Convolvulaceae Amaranthaceae Verbenaceae Poaceae Asteraceae Poaceae Fabaceae Flacourtiaceae Solanaceae Solanaceae Fabaceae Malpighiaaceae Cuccurbitaceae Asteraceae Mimosaceae Araceae Poaceae Araliaceae Poaceae Myrsinaceae Poaceae Poaceae Malvaceae Poaceae Araceae Euphorbiaceae Phytolaccaceae Piperaceae

et h t t t h h t t t t t t t t t t h l t h l l t h h h h t t t et l l t l h he h et h t h h h h e h h t

rf p rf rf rf p p rf p rf rf rf rf rf rf rf rf p rf rf p rf rf rf p p rf p rf rf rf rf rf rf p rf p rf p rf p rf p p p p rf p rf rf

n – p n n – – n – – – st st p n n n – – p – – – t – – – – n n p – – – – – – – – n – st – – – – – – – p

D.S. %

1178 1.25 1 0.001 58 0.062 114 0.121 5 0.005 8 0.008 10 0.011 1 0.001 279 0.296 2 0.002 5 0.005 5 0.005 47 0.05 494 0.524 7055 7.489 38801 4.118 1681 1.784 65 0.069 63 0.067 21762 2.31 371 0.394 27 0.029 3 0.003 45 0.048 4 0.004 17 0.018 2 0.002 47 0.05 1 0.001 1238 1.314 20 0.021 85 0.09 33 (.035) 795 0.844 30 0.032 201 0.213 78 0.083 6 0.006 11 0.012 983 1.043 4 0.004 347 0.368 24929 26.46 2595 2.754 376 0.399 2 0.002 2 0.002 3228 3.426 2 0.002 3 0.003

a a g a a – w w a w w a a w a a a – w w w w w w a a w w w a a a w w a w a a w a a a w a a w a g a a

265

Appendix 1. Continued. Species

Family

Lifeform Type veg Plant strategy Seeds n

D.S. %

Piptocarpha chontalensis Baker in Mart. Pithecoctenium crucigerum (L.) A. Gentry Platymiscium pinnatum (Jacq.) Dugand,. Poulsenia armata (Miq.) Standl. Prestonia mexicana C.DC. Pseudelephantopus spicatus (Aubl.) Rohr. Psidium guajaba L. Psychotria limonensis Krause. Paullinia venosa Radlk. Randia xalapensis Martius & Galeotti Rhynchospora radicans ( Schldl. & Cham.) Pfeiffer ssp. radicans. Robinsonella mirandae G´omez-Pompa. Rollinia jimenezii Saff. Rourea glabra Kunth in H.B.K. Sapium nitidum (Monach.) Lundell. Scleria melaleuca Reichb. ex Schltdl. et Cham. et Cham. Serjania mexicana (L.) Willd. Setaria lutescens Hubb. Sida rhombifolia L. Smilax aristolochiifolia Mill. Solanum rudepannum Dunal Spigelia palmeri Rose Stemmadenia donnell-smithii (Rose ex J. D. Smith) Woodson. Stigmaphyllon lindenianum A.Juss. Synedrella nodiflora (L.) Gaertn. Syngonium podophyllum Schott Tabernaemontana alba Mill. Tetrorchidium rotundatum (Hook. et Arn.) Trema micrantha (L.) Blume Trichospermum galeottii (Turcz.) Kosterm. Trichostigma octandrum (L.) H. Walt. in Engl. Tripogandra floribunda (Hook. et Arn.) Trophis mexicana (Liebm.) Bureau in DC. Tuxtla pittieri (Greenm.) Villaseñor et Strother Ulmus mexicana (Liebm.) Planchon in DC. Verbesina crocata (Cav.) Less.

Asteraceae Bignoniaceae Fabaceae Moraceae Apocynaceae Asteraceae Myrtaceae Rubiaceae Sapindaceae Rubiaceae Cyperaceae

l l t t l h t t l t h

rf rf rf rf rf p p rf rf rf p

– – n n – – – st – st –

736 1 82 15 1 28 96 31 1 1 137

0.781 0.001 0.087 0.016 0.001 0.03 0.102 0.033 0.001 0.001 0.145

w w w a w a a a a a a

Malvaceae Annonaceae Connaraceae Euphorbiaceae Cyperaceae

t t l t h

rf rf rf rf rf

n n – n –

96 2 10 227 17

0.102 0.002 0.011 0.241 0.018

w a a a a

Sapindaceae Poaceae Malvaceae Smilacaceae Solanaceae Loganiaceae Apocynaceae

l h h l t h t

rf p p rf p p rf

– – – – – – p

6 499 14163 1 117 2 70

0.006 0.53 1.501 0.001 0.124 0.002 0.074

w a a a a – a

Malpighiaceae Asteraceae Araceae Apocynaceae Euphorbiaceae Ulmaceae Tiliaceae Phytolaccaceae Commelinaceae Moraceae Asteraceae

l h he t t t t l h t l

rf p rf rf rf rf rf rf p rf rf

– – – st st p p – – st –

4 1 29 37 14 118 11 6 1 8 57

0.004 0.001 0.031 0.039 0.015 0.125 0.012 0.006 0.001 0.008 0.061

w a a a a a w a g a w

Ulmaceae Asteraceae

t h

rf p

n –

6 4

0.006 w 0.004 w

1 The total number of seeds for this species is the sum of the unripen seeds and the ripen seeds. 2 For this species the number of seeds shown is obtained by the relationship of 1.5 seeds per fruit (Mart´ınez-Ramos 1985), and

the quantity was rounded off to an integer. 3 For this species the total number of seeds is the sum of the two varities which are present for this specie, seeds with two

prolongations and seeds with one prolongation.