American Journal of Botany 101(6): 970–978. 2014.
SEED SOURCE, SEED TRAITS, AND FRUGIVORE HABITS: IMPLICATIONS FOR DISPERSAL QUALITY OF TWO SYMPATRIC PRIMATES1
JULIETA BENÍTEZ-MALVIDO2,5, ANA MA. GONZÁLEZ-DI PIERRO3, RAFAEL LOMBERA3, SUSANA GUILLÉN,2 AND ALEJANDRO ESTRADA4 2Centro
de Investigaciones en Ecosistemas (CIEco), Universidad Nacional Autónoma de México, Antigua Carretera a Pátzcuaro No. 8701, Ex-Hacienda de San José de la Huerta, CP 58090 Morelia, Michoacán, México; 3Universidad Intercultural de Chiapas, Unidad Académica Multidisciplinaria Las Margaritas, Chiapas, México; and 4Estación de Biología “Los Tuxtlas”, Instituto de Biología, Universidad Nacional Autónoma de México, San Andrés Tuxtla, Veracruz, México
• Premise of the study: Frugivore selection of fruits and treatment of seeds together with seed deposition site are crucial for the population dynamics of vertebrate-dispersed plants. However, frugivore species may influence dispersal quality differently even when feeding on the same fruit species and, while animals disperse some seeds, others simply fall beneath the parent plant. • Methods: In southern Mexico, we investigated to see if within-species seed traits (i.e., length, width, weight, and volume) and germination success differed according to seed source. For five tropical tree species we obtained ingested seeds from two sources, howler monkey (Alouatta pigra) and spider monkey (Ateles geoffroyi) feces; and noningested seeds from two sources, the ground and tree crowns (with predispersed seeds used as control). • Key results: A principal components’ analysis showed that traits of seeds ingested by howler monkeys differed from other sources while seeds ingested by spider monkeys were similar to noningested seeds. Howlers consumed on average the larger seeds in Ampelocera hottlei, Brosimum lactescens, and Dialium guianense. Both primate species consumed the smaller seeds in Spondias mombin, while no seed trait differences among seed sources were found in Spondias radlkoferi. For all five tree species, germination rate was greatest for seeds ingested by howler monkeys. • Conclusions: For the studied plant species, seed ingestion by howler monkeys confers higher dispersal quality than ingestion by spider monkeys or nondispersal. Dispersal services of both primate species, however, are not redundant and may contribute to germination heterogeneity within plant populations in tropical forests. Key words: Alouatta pigra; Ateles geoffroyi; endozoochory; germination; seed size.
Evolution in plant species has resulted in different physical and chemical traits in fruits and seeds that direct dispersal and maximize the number of successfully dispersed seeds (Howe and Smallwood, 1982; Traveset et al., 2007; Russo and Chapman, 2011). The quality of seed dispersal depends on the treatment given to a seed in the mouth and gut of frugivores, and on the quality of seed deposition as determined by the probability that a deposited seed will survive in a given site (Schupp, 1993). The same frugivore species may have different effects on germination depending upon the plant species consumed, whereas different frugivore species feeding on the same plant species often have different effects on germination (i.e., enhancement, inhibition, and/or neutral effects; Traveset, 1998; Traveset and Verdú, 2002; Martins, 2006; Robertson et al., 2006). Fruit treatment in
the digestive tract (quality) can influence germination success and, therefore, is one of the components of disperser effectiveness that may be crucial for the population dynamics of many plant species (Traveset et al., 2007). Primates play a critical role in enhancing the recruitment of several tropical tree species by dispersing their seeds, thus maintaining the diversity of plant communities (Howe and Smallwood, 1982; Julliot, 1997; Link and Di Fiore, 2006). In neotropical rain forest, two of the largest canopy primates are howler monkeys (Alouatta spp.), which are folivores-frugivores, and spider monkeys (Ateles spp.), primarily frugivores (Milton, 1981, 1984, 1998; Crissey et al., 1990; Lambert, 1998; Stevenson et al., 2002). Howler monkeys have an enlarged hindgut to process the large amounts of foliage ingested daily, with an average transit time for ingesta of 21 h (Milton, 1981, 1984, 1998; Crissey et al., 1990; Di Fiore et al., 2008, 2011). Because leaves eaten in each foraging bout take a long time to digest, howler monkeys are very selective in their diet (Milton, 1981, 1984, 1998; Julliot, 1996; Chapman et al., 2012). For a particular plant species, howler monkeys (e.g., Alouatta seniculus in Julliot, 1996) selectively feed on the largest fruits (and seeds) that have higher concentrations of carbohydrates or fats. This is not the case for spider monkeys that have a food passage rate of about 4-5 h (Milton, 1981; Stevenson et al., 2002; Dew, 2008; Di Fiore et al., 2008, 2011). In addition to differences in hindgut length (a morphological trait) and digestion time (a physiological trait), they also have differential demographic (population densities) and behavioral habits (Table 1). These divergences may result in different food
1 Manuscript received 1 April 2014; revision accepted 7 May 2014. This research was supported by grants from the Consejo Nacional de Ciencia y Tecnología, Mexico (CONACyT-79121) and Universidad Nacional Autónoma de México (PAPIIT IN206111). The authors thank the Comisión Nacional de Áreas Naturales Protegidas (CONANP) for granting permits to work in the Montes Azules Biosphere Reserve (MABR) and the Centro de Investigaciones en Ecosistemas, Universidad Nacional Autónoma de México (UNAM), for providing logistical support. We are grateful to G. Lombera for his valuable assistance in the field, and for the technical support provided by J. M. Lobato-García, J. Rodríguez-Velázquez, H. Ferreira and A. Valencia García. 5 Author for correspondence (e-mail: [email protected])
doi:10.3732/ajb.1400147
American Journal of Botany 101(6): 970–978, 2014; http://www.amjbot.org/ © 2014 Botanical Society of America
970
June 2014]
BENÍTEZ-MALVIDO ET AL.—PLANT AND FRUGIVORE TRAITS
971
TABLE 1.
Habits of two sympatric primate species in the Montes Azules Biosphere Reserve (MABR), Mexico; see Cháves (2010) and González-Di Pierro (2011).
Habit Physiological Diet (Crissey et al., 1990) Food passage rate (average ± SD) (Crissey et al., 1990) Vision (Surridge et al., 2003) Morphological Average colon length and width (Milton, 1981) Adult body mass (Reid, 1997) Behavioral Seed dispersal distance in MABR Habitat types in MABR Defecation pattern (Andresen and Levey, 2004) Home ranges in MABR Population density in MABR (Estrada et al., 2004) Group size in MABR a
Howler monkeys (Alouatta pigra)
Spider monkeys (Ateles geoffroyi)
Folivore/frugivore 20.4 ± 3.5 h Routine trichromacy (both sexes)
Frugivore 4.4 ± 1.5 h Allelic trichromacy (females only)
43 and 3.5 cm a 6–8 kg
33 and 1.5 cm 5–9 kg
50–375 m Primary & secondary forests, fragments Clumped (group defecation) < 25 ha 0.14 individuals/ha 4–9 individuals
0.5–345 m Mostly primary forests Scattered (individual defecation) 30–90 ha 0.029 individuals/ha 36–44 individuals
Data from Alouatta palliata.
choices (e.g., fruit size and color) and treatment of ingested seeds, even when feeding on the same plant species (Lambert, 1998; Martins, 2006). While it has been shown that seed traits such as weight, size, coat permeability, coat thickness, texture, and hardness can affect germination rate and germinability (percentage) (Traveset et al., 2007; Gross-Camp and Kaplin, 2011), the effect of seed size after passage through different primates’ guts on seed germination is unclear (Traveset et al., 2007). In the tropical rain forest of the Montes Azules Biosphere Reserve (MABR), Chiapas, Mexico, the black howler monkey, Alouatta pigra, and the black-handed spider monkey, Ateles geoffroyi, coexist in the forest and share at least 35 fruit species in their diets (Estrada et al., 2004; Cháves, 2010; González-Di Pierro, 2011; Table 2). Both primate species are considered important seed dispersers for numerous tree species in neotropical forests (Link and Di Fiore, 2006; Felton et al., 2009). Previous studies from this area (MABR) have found that howler and spider monkeys have positive effects of gut passage on seed germination (Cháves, 2010; González-Di Pierro, 2011; González-Di Pierro et al., 2011). In the current study, we investigated to see if germination rate (speed) and germinability (percentage of cumulative germination) differed depending on the seed source (i.e., predispersed seeds attached to the parent trees, seeds on the ground below the parent tree, and seeds from spider and howler monkey fecal
clumps). We refer to “seed source” as the setting from which seeds were obtained. For the experiment we used seeds from five native tree species all of which are found in howler and spider monkey feces. We also tested to see if within-species seed traits differed among ingested and noningested seeds. We examined whether within-species differences in seed responses varied according to seed source using germination trials. And we tested whether sympatric primate species sharing the same resources provide similar seed dispersal quality. We predicted that because of differences in fruit selection and average food passage times, germination success (i.e., rate and germinability) would differ according to primate species. In addition, we predicted that ingested seeds by either primate species would have greater germination success than seeds collected from the ground and tree crowns. Differences in seed traits within and among tree species may contribute to dispersal and germination heterogeneity with important consequences for forest regeneration.
MATERIALS AND METHODS Study area—Fieldwork was conducted in the southern portion of the Montes Azules Biosphere Reserve (MABR), Chiapas, Mexico, which is 330 000 ha (Gómez-Pompa and Dirzo, 1995) and is part of the Lacandon rain forest region
TABLE 2.
Fruit morphology, tree density, and consumption habits by two primate species for the five focal tree species in the Montes Azules Biosphere Reserve (Pennington and Sarukhán, 1998). Seed size includes average (length/width) in cm (after Ibarra-Manríquez and Cornejo, 2010). Tree density in 7 ha is indicated within parentheses (M. Martínez-Ramos, UNAM, unpublished data). Feeding behavior data and the number of seeds in feces for Aletes geoffroyi is from Cháves (2010) and for Alouatta pigra, is from González-Di Pierro (2011). Average no. of seeds per species in feces and % of total feeding time eating fruits
Species
Fruit morphology
Ampelocera hottlei
Drupe, green-yellow, 1 seed (1.25/1.25)
Brosimum lactescens
Berry, yellow-orange, 1 seed (2.0/2.0)
Dialium guianense Spondias mombin
Legume, indehiscent, globose, drupe-like, greenish-brownish, 1-2 seeds (1.5/1.5) Drupe, yellow-orange, 3-4 seeds (3.25/1.75)
Spondias radlkoferi
Drupe, green, 3-4 seeds (3.25/1.55)
Trees/ha
A. pigra
A. geoffroyi
14.3 (100)
8.3 99.0% 8.2 47.2% 9 100% 5 100% 4.4 100%
4.4 92.8% — — 5.2 16.2% 2.9 100% 3.1 96.7%
0.6 (4) 28.14 (197) 1 (7) 8.6 (60)
972
[Vol. 101
AMERICAN JOURNAL OF BOTANY
(16°05′58′′N, 90°52′36′′W with an elevation 10-50 m a.s.l.), and in an adjacent area (Marqués de Comillas region or MCR) modified by human activity and separated from the MABR by the Río Lacantún. Human colonization of MCR began about 30-40 yr ago, and cattle ranching resulted in rapid deforestation and forest fragmentation (Mariaca-Méndez, 2002). The original vegetation in the area is lowland tropical rain forest and semideciduous rain forest with trees reaching heights of 45 m. The climate is hot and humid with mean annual precipitation and temperature of 2874 mm and 25°C, respectively (Estrada et al., 2004). Tree species and seed collection—Five tree species that are common in the diet of howler and spider monkeys and that display synchrony in fruiting at some stage (June-August 2011) were selected for the germination tests. These tree species were as follows: Ampelocera hottlei (Standl.) Standl. (Ulmaceae); Brosimum lactescens (S. Moore) C.C. Berg (Moraceae); Dialium guianense (Aubl.) Sandw. (Fabaceae-Caesalpinioideae); Spondias mombin L. (Anacardiaceae); and Spondias radlkoferi Donn. Sm. (Anacardiaceae). Fruits of all five species have fleshy and/or pulpy structures that offer rewards to frugivores and contain one to four large seeds (1.5-4.0 cm in size, Table 2). For the two Spondias species the diaspore is the hard endocarp containing the seeds. These tree species, except for B. lactescens, are common in the study area (i.e., more than 1 tree/ha) and considered top food species in the diets of spider and howler monkeys, representing > 80% of their feeding time (González-Zamora et al., 2009; Cháves, 2010; González-Di Pierro, 2011). Seeds were collected from four different sources: (1) predispersed seeds from ripe fruits attached to the parent trees; (2) seeds from ripe fruits on the ground below the parent tree; (3) seeds in howler monkey feces; and (4) seeds in spider monkey feces. Seeds collected from tree crowns were considered as experimental controls as they were not yet “dispersed” (e.g., by gravity or fauna) and were used to compare with fruit selection preferences by arboreal primates (i.e., seed size). Seeds from tree crowns were collected from five Ampelocera hottlei, six Brosimum lactescens, five Dialium guianense, eight Spondias mombin, and seven S. radlkoferi trees. Depending on the size and location of the individual trees, mature fruits were either cut manually from branches at reachable heights or a branch full of mature fruits was detached from the tree. For large trees, it was sometimes necessary to climb a neighboring tree to reach for the target fruits. When possible, fruits were detached from different locations in the tree crowns. Seeds from the ground were collected from mature fruits that fell naturally beneath the parent trees; seeds were collected beneath three A. hottlei, four B. lactescens, six D. guianense, five S. mombin, and four S. radlkoferi trees. Seeds from tree crowns and from the ground were not necessarily collected from the same individual trees. A total of 110 seeds/seed source and tree species were collected. To collect seeds defecated by the primates, during the same time period we followed four groups of spider monkeys and five groups of howler monkeys from 0700 to 1700. Groups were followed until 110 seeds per tree species and seed source were collected. Feces were collected immediately after defecation in forest areas where our primate groups fed. Germination trials—Since fruit pulp often contains germination inhibitors that can decrease and even prevent germination (Traveset et al., 2007), immediately after collection all seeds were removed from the fruits and rinsed with water to manually remove the fleshy pulp and/or aril. Seeds in feces were thoroughly washed to remove fecal matter and attached plant material. Seed cleaning was applied so that all seed species and sources had the same initial germination conditions and to measure seed metrics accurately. Seeds that were visibly damaged by insects or other organisms were not used in the germination trials. Germination experiments were conducted outdoors in the field as this has been shown to produce more reliable results than running germination trials in a greenhouse or under laboratory conditions (Rodríguez-Pérez et al., 2005; Traveset et al., 2007). For each seed collected we measured its length, width, and weight. Because seeds of the tree species of interest are fairly ellipsoid, we calculated seed volume as 4/3 × Π × b × a2 where a = seed width (thickness) and b = seed length. Overall, 2200 intact seeds were used for the germination trials. For each tree species we used 110 seeds per seed source (N = 440 seeds per tree species). Seeds were pooled within each seed source and species. According to tree species, seeds were placed in germination trays (10 seeds per tray) filled with river sand and watered every two days. River sand was used as a substrate instead of forest soil as it was readily available and relatively free of animals and decomposing organic material. We protected the trays against seed predators (e.g., large ants and rodents) with a transparent plastic top with small perforations. In addition to preventing seed predation, the experimental trays also provided an adequate microenvironment (i.e., constant moisture and temperature) for seed germination. For 60 d we counted the number of germinated seeds daily. Seeds
were considered to have germinated when the radicle extruded from the seed coat. Maximum germination (or germinability) was defined as the cumulative percentage of seeds that had germinated 60 d from the sowing date, whereas germination rate refers to the proportion of seeds germinating per day. All seeds that germinated survived as seedlings for more than 60 d but seedling establishment was not considered in this study. Larger seeds may have advantages for seed germination and seedling establishment compared to small seeds of the same species (Paz and Martínez-Ramos, 2003). Although different traits were measured on every seed before sowing, seeds were not individually monitored, however, and therefore, seed traits were not considered as covariates in the analysis. Statistical analyses—Differences in seed length, width, volume, and weight among seed sources were analyzed with one-way ANOVA for each plant species separately using R, version 2.13.0 (Ripley, 2001). In the case of significant ANOVA results differences were compared with Tukey tests for multiple comparisons (95% confidence level). In addition, to detect general trends in seed traits in each seed source category and to separate seed sources according to their traits we performed a principal components’ analysis (PCA) using R. For this analysis we used the four seed-source categories and the four seed-traits described above. We used a PCA instead of other analytical methods (i.e., hierarchical cluster analysis) because seed metrics were highly correlated, and this method allows for a better assessment of the differences of seed traits within seed source. We evaluated the effect of seed source on maximum cumulative germination (germinability) and germination rate using analysis of deviance based on generalized linear models (GLIM, Crawley, 1993; Green and Payne, 1994). In our model, the cumulative proportion of germinating seeds was the dependent variable, time (days after sowing date) was a continuous independent variable, and the seed source (factor with four levels) was the categorical independent variable. We used a binomial error and a logistic link function as indicated for proportional dependent variables; in this error type, the deviance (equivalent of variance in a model with a normal type error) explained by independent variables can be considered to be an approximated chi-square (χ2) value (Crawley, 1993). The cumulative proportion of germinating seeds (y) after t days was then calculated as follows:
y=
2 ¯ ¡ a+ bt ct ° ±
e¢
2 ¯ ¡ a + bt ct ° ±
1+ e ¢
Here, coefficient a is the y intercept (starting germination), and coefficient b the initial germination rate. Furthermore, c is a coefficient indicating whether the initial germination rate increases (in such a case, c adopts a negative value) or decreases (in such a case, c adopts a positive value) with time (after Guillén et al., 2009). The effect of seed source on germination rate was evaluated by the deviance explained by the interaction of each or both of these factors with the linear or the quadratic time (t 2) term. Differences among seed sources were examined using t tests, following Crawley (1993). Statistical analyses were conducted using GLIM software version 3.77 (Royal Statistical Society, 1985).
RESULTS Seed traits— All species, except Spondias radlkoferi, differed significantly in seed traits (i.e., length, width, weight, and volume) according to seed source (Table 3). Seeds of S. radlkoferi were very similar among seed sources for all the metrics considered. On average, howler monkeys ingested the larger, wider, heavier, and more voluminous seeds of Ampelocera hottlei, Brosimum Lactescens, and Dialium guianense; whereas both primate species ingested the smaller seeds in S. mombin as compared to noningested seeds (Table 3). Moreover, seeds of D. guianense ingested by spider monkeys were significantly larger than noningested seeds but smaller than those ingested by howler monkeys. In general, no significant differences were found among seed traits for the noningested seeds (i.e., those collected from tree crowns and from the ground). Noningested seeds differed significantly according to seed source in
June 2014]
BENÍTEZ-MALVIDO ET AL.—PLANT AND FRUGIVORE TRAITS
973
Traits of fresh seeds, mean ± SD, from five native tree species collected from different sources in Montes Azules Biosphere Reserve, Mexico; N = 110 seeds per tree species for each seed source. Different letters and numbers in bold indicate significant differences among seed sources (oneway ANOVA). P < 0.05.
TABLE 3.
Seed Traits Ampelocera hottlei Length (cm) Width (cm) Volume (cm3) Weight (g) Brosimum lactescens Length (cm) Width (cm) Volume (cm3) Weight (g) Dialium guianense Length (cm) Width (cm) Volume (cm3) Weight (g) Spondias mombin Length (cm) Width (cm) Volume (cm3) Weight (g) Spondias radlkoferi Length (cm) Width (cm) Volume (cm3) Weight (g)
Alouatta pigra feces
Ateles geoffroyi feces
Tree crown
Ground
F 3, 436
P
1.18 ± 0.07b 0.87 ± 0.07b 3.78 ± 0.78b 0.44 ± 0.08b
1.13 ± 0.05a 0.81 ± 0.07a 3.21 ± 0.61a 0.33 ± 0.08a
1.13 ± 0.06a 0.84 ± 0.05a 3.35 ± 0.56a 0.42 ± 0.07b
1.13 ± 0.06a 0.83 ± 0.05a 3.33 ± 0.55a 0.42 ± 0.073b
14.8 14.1 16.7 45.3
SEED SOURCE, SEED TRAITS, AND FRUGIVORE HABITS: IMPLICATIONS FOR DISPERSAL QUALITY OF TWO SYMPATRIC PRIMATES1
JULIETA BENÍTEZ-MALVIDO2,5, ANA MA. GONZÁLEZ-DI PIERRO3, RAFAEL LOMBERA3, SUSANA GUILLÉN,2 AND ALEJANDRO ESTRADA4 2Centro
de Investigaciones en Ecosistemas (CIEco), Universidad Nacional Autónoma de México, Antigua Carretera a Pátzcuaro No. 8701, Ex-Hacienda de San José de la Huerta, CP 58090 Morelia, Michoacán, México; 3Universidad Intercultural de Chiapas, Unidad Académica Multidisciplinaria Las Margaritas, Chiapas, México; and 4Estación de Biología “Los Tuxtlas”, Instituto de Biología, Universidad Nacional Autónoma de México, San Andrés Tuxtla, Veracruz, México
• Premise of the study: Frugivore selection of fruits and treatment of seeds together with seed deposition site are crucial for the population dynamics of vertebrate-dispersed plants. However, frugivore species may influence dispersal quality differently even when feeding on the same fruit species and, while animals disperse some seeds, others simply fall beneath the parent plant. • Methods: In southern Mexico, we investigated to see if within-species seed traits (i.e., length, width, weight, and volume) and germination success differed according to seed source. For five tropical tree species we obtained ingested seeds from two sources, howler monkey (Alouatta pigra) and spider monkey (Ateles geoffroyi) feces; and noningested seeds from two sources, the ground and tree crowns (with predispersed seeds used as control). • Key results: A principal components’ analysis showed that traits of seeds ingested by howler monkeys differed from other sources while seeds ingested by spider monkeys were similar to noningested seeds. Howlers consumed on average the larger seeds in Ampelocera hottlei, Brosimum lactescens, and Dialium guianense. Both primate species consumed the smaller seeds in Spondias mombin, while no seed trait differences among seed sources were found in Spondias radlkoferi. For all five tree species, germination rate was greatest for seeds ingested by howler monkeys. • Conclusions: For the studied plant species, seed ingestion by howler monkeys confers higher dispersal quality than ingestion by spider monkeys or nondispersal. Dispersal services of both primate species, however, are not redundant and may contribute to germination heterogeneity within plant populations in tropical forests. Key words: Alouatta pigra; Ateles geoffroyi; endozoochory; germination; seed size.
Evolution in plant species has resulted in different physical and chemical traits in fruits and seeds that direct dispersal and maximize the number of successfully dispersed seeds (Howe and Smallwood, 1982; Traveset et al., 2007; Russo and Chapman, 2011). The quality of seed dispersal depends on the treatment given to a seed in the mouth and gut of frugivores, and on the quality of seed deposition as determined by the probability that a deposited seed will survive in a given site (Schupp, 1993). The same frugivore species may have different effects on germination depending upon the plant species consumed, whereas different frugivore species feeding on the same plant species often have different effects on germination (i.e., enhancement, inhibition, and/or neutral effects; Traveset, 1998; Traveset and Verdú, 2002; Martins, 2006; Robertson et al., 2006). Fruit treatment in
the digestive tract (quality) can influence germination success and, therefore, is one of the components of disperser effectiveness that may be crucial for the population dynamics of many plant species (Traveset et al., 2007). Primates play a critical role in enhancing the recruitment of several tropical tree species by dispersing their seeds, thus maintaining the diversity of plant communities (Howe and Smallwood, 1982; Julliot, 1997; Link and Di Fiore, 2006). In neotropical rain forest, two of the largest canopy primates are howler monkeys (Alouatta spp.), which are folivores-frugivores, and spider monkeys (Ateles spp.), primarily frugivores (Milton, 1981, 1984, 1998; Crissey et al., 1990; Lambert, 1998; Stevenson et al., 2002). Howler monkeys have an enlarged hindgut to process the large amounts of foliage ingested daily, with an average transit time for ingesta of 21 h (Milton, 1981, 1984, 1998; Crissey et al., 1990; Di Fiore et al., 2008, 2011). Because leaves eaten in each foraging bout take a long time to digest, howler monkeys are very selective in their diet (Milton, 1981, 1984, 1998; Julliot, 1996; Chapman et al., 2012). For a particular plant species, howler monkeys (e.g., Alouatta seniculus in Julliot, 1996) selectively feed on the largest fruits (and seeds) that have higher concentrations of carbohydrates or fats. This is not the case for spider monkeys that have a food passage rate of about 4-5 h (Milton, 1981; Stevenson et al., 2002; Dew, 2008; Di Fiore et al., 2008, 2011). In addition to differences in hindgut length (a morphological trait) and digestion time (a physiological trait), they also have differential demographic (population densities) and behavioral habits (Table 1). These divergences may result in different food
1 Manuscript received 1 April 2014; revision accepted 7 May 2014. This research was supported by grants from the Consejo Nacional de Ciencia y Tecnología, Mexico (CONACyT-79121) and Universidad Nacional Autónoma de México (PAPIIT IN206111). The authors thank the Comisión Nacional de Áreas Naturales Protegidas (CONANP) for granting permits to work in the Montes Azules Biosphere Reserve (MABR) and the Centro de Investigaciones en Ecosistemas, Universidad Nacional Autónoma de México (UNAM), for providing logistical support. We are grateful to G. Lombera for his valuable assistance in the field, and for the technical support provided by J. M. Lobato-García, J. Rodríguez-Velázquez, H. Ferreira and A. Valencia García. 5 Author for correspondence (e-mail: [email protected])
doi:10.3732/ajb.1400147
American Journal of Botany 101(6): 970–978, 2014; http://www.amjbot.org/ © 2014 Botanical Society of America
970
June 2014]
BENÍTEZ-MALVIDO ET AL.—PLANT AND FRUGIVORE TRAITS
971
TABLE 1.
Habits of two sympatric primate species in the Montes Azules Biosphere Reserve (MABR), Mexico; see Cháves (2010) and González-Di Pierro (2011).
Habit Physiological Diet (Crissey et al., 1990) Food passage rate (average ± SD) (Crissey et al., 1990) Vision (Surridge et al., 2003) Morphological Average colon length and width (Milton, 1981) Adult body mass (Reid, 1997) Behavioral Seed dispersal distance in MABR Habitat types in MABR Defecation pattern (Andresen and Levey, 2004) Home ranges in MABR Population density in MABR (Estrada et al., 2004) Group size in MABR a
Howler monkeys (Alouatta pigra)
Spider monkeys (Ateles geoffroyi)
Folivore/frugivore 20.4 ± 3.5 h Routine trichromacy (both sexes)
Frugivore 4.4 ± 1.5 h Allelic trichromacy (females only)
43 and 3.5 cm a 6–8 kg
33 and 1.5 cm 5–9 kg
50–375 m Primary & secondary forests, fragments Clumped (group defecation) < 25 ha 0.14 individuals/ha 4–9 individuals
0.5–345 m Mostly primary forests Scattered (individual defecation) 30–90 ha 0.029 individuals/ha 36–44 individuals
Data from Alouatta palliata.
choices (e.g., fruit size and color) and treatment of ingested seeds, even when feeding on the same plant species (Lambert, 1998; Martins, 2006). While it has been shown that seed traits such as weight, size, coat permeability, coat thickness, texture, and hardness can affect germination rate and germinability (percentage) (Traveset et al., 2007; Gross-Camp and Kaplin, 2011), the effect of seed size after passage through different primates’ guts on seed germination is unclear (Traveset et al., 2007). In the tropical rain forest of the Montes Azules Biosphere Reserve (MABR), Chiapas, Mexico, the black howler monkey, Alouatta pigra, and the black-handed spider monkey, Ateles geoffroyi, coexist in the forest and share at least 35 fruit species in their diets (Estrada et al., 2004; Cháves, 2010; González-Di Pierro, 2011; Table 2). Both primate species are considered important seed dispersers for numerous tree species in neotropical forests (Link and Di Fiore, 2006; Felton et al., 2009). Previous studies from this area (MABR) have found that howler and spider monkeys have positive effects of gut passage on seed germination (Cháves, 2010; González-Di Pierro, 2011; González-Di Pierro et al., 2011). In the current study, we investigated to see if germination rate (speed) and germinability (percentage of cumulative germination) differed depending on the seed source (i.e., predispersed seeds attached to the parent trees, seeds on the ground below the parent tree, and seeds from spider and howler monkey fecal
clumps). We refer to “seed source” as the setting from which seeds were obtained. For the experiment we used seeds from five native tree species all of which are found in howler and spider monkey feces. We also tested to see if within-species seed traits differed among ingested and noningested seeds. We examined whether within-species differences in seed responses varied according to seed source using germination trials. And we tested whether sympatric primate species sharing the same resources provide similar seed dispersal quality. We predicted that because of differences in fruit selection and average food passage times, germination success (i.e., rate and germinability) would differ according to primate species. In addition, we predicted that ingested seeds by either primate species would have greater germination success than seeds collected from the ground and tree crowns. Differences in seed traits within and among tree species may contribute to dispersal and germination heterogeneity with important consequences for forest regeneration.
MATERIALS AND METHODS Study area—Fieldwork was conducted in the southern portion of the Montes Azules Biosphere Reserve (MABR), Chiapas, Mexico, which is 330 000 ha (Gómez-Pompa and Dirzo, 1995) and is part of the Lacandon rain forest region
TABLE 2.
Fruit morphology, tree density, and consumption habits by two primate species for the five focal tree species in the Montes Azules Biosphere Reserve (Pennington and Sarukhán, 1998). Seed size includes average (length/width) in cm (after Ibarra-Manríquez and Cornejo, 2010). Tree density in 7 ha is indicated within parentheses (M. Martínez-Ramos, UNAM, unpublished data). Feeding behavior data and the number of seeds in feces for Aletes geoffroyi is from Cháves (2010) and for Alouatta pigra, is from González-Di Pierro (2011). Average no. of seeds per species in feces and % of total feeding time eating fruits
Species
Fruit morphology
Ampelocera hottlei
Drupe, green-yellow, 1 seed (1.25/1.25)
Brosimum lactescens
Berry, yellow-orange, 1 seed (2.0/2.0)
Dialium guianense Spondias mombin
Legume, indehiscent, globose, drupe-like, greenish-brownish, 1-2 seeds (1.5/1.5) Drupe, yellow-orange, 3-4 seeds (3.25/1.75)
Spondias radlkoferi
Drupe, green, 3-4 seeds (3.25/1.55)
Trees/ha
A. pigra
A. geoffroyi
14.3 (100)
8.3 99.0% 8.2 47.2% 9 100% 5 100% 4.4 100%
4.4 92.8% — — 5.2 16.2% 2.9 100% 3.1 96.7%
0.6 (4) 28.14 (197) 1 (7) 8.6 (60)
972
[Vol. 101
AMERICAN JOURNAL OF BOTANY
(16°05′58′′N, 90°52′36′′W with an elevation 10-50 m a.s.l.), and in an adjacent area (Marqués de Comillas region or MCR) modified by human activity and separated from the MABR by the Río Lacantún. Human colonization of MCR began about 30-40 yr ago, and cattle ranching resulted in rapid deforestation and forest fragmentation (Mariaca-Méndez, 2002). The original vegetation in the area is lowland tropical rain forest and semideciduous rain forest with trees reaching heights of 45 m. The climate is hot and humid with mean annual precipitation and temperature of 2874 mm and 25°C, respectively (Estrada et al., 2004). Tree species and seed collection—Five tree species that are common in the diet of howler and spider monkeys and that display synchrony in fruiting at some stage (June-August 2011) were selected for the germination tests. These tree species were as follows: Ampelocera hottlei (Standl.) Standl. (Ulmaceae); Brosimum lactescens (S. Moore) C.C. Berg (Moraceae); Dialium guianense (Aubl.) Sandw. (Fabaceae-Caesalpinioideae); Spondias mombin L. (Anacardiaceae); and Spondias radlkoferi Donn. Sm. (Anacardiaceae). Fruits of all five species have fleshy and/or pulpy structures that offer rewards to frugivores and contain one to four large seeds (1.5-4.0 cm in size, Table 2). For the two Spondias species the diaspore is the hard endocarp containing the seeds. These tree species, except for B. lactescens, are common in the study area (i.e., more than 1 tree/ha) and considered top food species in the diets of spider and howler monkeys, representing > 80% of their feeding time (González-Zamora et al., 2009; Cháves, 2010; González-Di Pierro, 2011). Seeds were collected from four different sources: (1) predispersed seeds from ripe fruits attached to the parent trees; (2) seeds from ripe fruits on the ground below the parent tree; (3) seeds in howler monkey feces; and (4) seeds in spider monkey feces. Seeds collected from tree crowns were considered as experimental controls as they were not yet “dispersed” (e.g., by gravity or fauna) and were used to compare with fruit selection preferences by arboreal primates (i.e., seed size). Seeds from tree crowns were collected from five Ampelocera hottlei, six Brosimum lactescens, five Dialium guianense, eight Spondias mombin, and seven S. radlkoferi trees. Depending on the size and location of the individual trees, mature fruits were either cut manually from branches at reachable heights or a branch full of mature fruits was detached from the tree. For large trees, it was sometimes necessary to climb a neighboring tree to reach for the target fruits. When possible, fruits were detached from different locations in the tree crowns. Seeds from the ground were collected from mature fruits that fell naturally beneath the parent trees; seeds were collected beneath three A. hottlei, four B. lactescens, six D. guianense, five S. mombin, and four S. radlkoferi trees. Seeds from tree crowns and from the ground were not necessarily collected from the same individual trees. A total of 110 seeds/seed source and tree species were collected. To collect seeds defecated by the primates, during the same time period we followed four groups of spider monkeys and five groups of howler monkeys from 0700 to 1700. Groups were followed until 110 seeds per tree species and seed source were collected. Feces were collected immediately after defecation in forest areas where our primate groups fed. Germination trials—Since fruit pulp often contains germination inhibitors that can decrease and even prevent germination (Traveset et al., 2007), immediately after collection all seeds were removed from the fruits and rinsed with water to manually remove the fleshy pulp and/or aril. Seeds in feces were thoroughly washed to remove fecal matter and attached plant material. Seed cleaning was applied so that all seed species and sources had the same initial germination conditions and to measure seed metrics accurately. Seeds that were visibly damaged by insects or other organisms were not used in the germination trials. Germination experiments were conducted outdoors in the field as this has been shown to produce more reliable results than running germination trials in a greenhouse or under laboratory conditions (Rodríguez-Pérez et al., 2005; Traveset et al., 2007). For each seed collected we measured its length, width, and weight. Because seeds of the tree species of interest are fairly ellipsoid, we calculated seed volume as 4/3 × Π × b × a2 where a = seed width (thickness) and b = seed length. Overall, 2200 intact seeds were used for the germination trials. For each tree species we used 110 seeds per seed source (N = 440 seeds per tree species). Seeds were pooled within each seed source and species. According to tree species, seeds were placed in germination trays (10 seeds per tray) filled with river sand and watered every two days. River sand was used as a substrate instead of forest soil as it was readily available and relatively free of animals and decomposing organic material. We protected the trays against seed predators (e.g., large ants and rodents) with a transparent plastic top with small perforations. In addition to preventing seed predation, the experimental trays also provided an adequate microenvironment (i.e., constant moisture and temperature) for seed germination. For 60 d we counted the number of germinated seeds daily. Seeds
were considered to have germinated when the radicle extruded from the seed coat. Maximum germination (or germinability) was defined as the cumulative percentage of seeds that had germinated 60 d from the sowing date, whereas germination rate refers to the proportion of seeds germinating per day. All seeds that germinated survived as seedlings for more than 60 d but seedling establishment was not considered in this study. Larger seeds may have advantages for seed germination and seedling establishment compared to small seeds of the same species (Paz and Martínez-Ramos, 2003). Although different traits were measured on every seed before sowing, seeds were not individually monitored, however, and therefore, seed traits were not considered as covariates in the analysis. Statistical analyses—Differences in seed length, width, volume, and weight among seed sources were analyzed with one-way ANOVA for each plant species separately using R, version 2.13.0 (Ripley, 2001). In the case of significant ANOVA results differences were compared with Tukey tests for multiple comparisons (95% confidence level). In addition, to detect general trends in seed traits in each seed source category and to separate seed sources according to their traits we performed a principal components’ analysis (PCA) using R. For this analysis we used the four seed-source categories and the four seed-traits described above. We used a PCA instead of other analytical methods (i.e., hierarchical cluster analysis) because seed metrics were highly correlated, and this method allows for a better assessment of the differences of seed traits within seed source. We evaluated the effect of seed source on maximum cumulative germination (germinability) and germination rate using analysis of deviance based on generalized linear models (GLIM, Crawley, 1993; Green and Payne, 1994). In our model, the cumulative proportion of germinating seeds was the dependent variable, time (days after sowing date) was a continuous independent variable, and the seed source (factor with four levels) was the categorical independent variable. We used a binomial error and a logistic link function as indicated for proportional dependent variables; in this error type, the deviance (equivalent of variance in a model with a normal type error) explained by independent variables can be considered to be an approximated chi-square (χ2) value (Crawley, 1993). The cumulative proportion of germinating seeds (y) after t days was then calculated as follows:
y=
2 ¯ ¡ a+ bt ct ° ±
e¢
2 ¯ ¡ a + bt ct ° ±
1+ e ¢
Here, coefficient a is the y intercept (starting germination), and coefficient b the initial germination rate. Furthermore, c is a coefficient indicating whether the initial germination rate increases (in such a case, c adopts a negative value) or decreases (in such a case, c adopts a positive value) with time (after Guillén et al., 2009). The effect of seed source on germination rate was evaluated by the deviance explained by the interaction of each or both of these factors with the linear or the quadratic time (t 2) term. Differences among seed sources were examined using t tests, following Crawley (1993). Statistical analyses were conducted using GLIM software version 3.77 (Royal Statistical Society, 1985).
RESULTS Seed traits— All species, except Spondias radlkoferi, differed significantly in seed traits (i.e., length, width, weight, and volume) according to seed source (Table 3). Seeds of S. radlkoferi were very similar among seed sources for all the metrics considered. On average, howler monkeys ingested the larger, wider, heavier, and more voluminous seeds of Ampelocera hottlei, Brosimum Lactescens, and Dialium guianense; whereas both primate species ingested the smaller seeds in S. mombin as compared to noningested seeds (Table 3). Moreover, seeds of D. guianense ingested by spider monkeys were significantly larger than noningested seeds but smaller than those ingested by howler monkeys. In general, no significant differences were found among seed traits for the noningested seeds (i.e., those collected from tree crowns and from the ground). Noningested seeds differed significantly according to seed source in
June 2014]
BENÍTEZ-MALVIDO ET AL.—PLANT AND FRUGIVORE TRAITS
973
Traits of fresh seeds, mean ± SD, from five native tree species collected from different sources in Montes Azules Biosphere Reserve, Mexico; N = 110 seeds per tree species for each seed source. Different letters and numbers in bold indicate significant differences among seed sources (oneway ANOVA). P < 0.05.
TABLE 3.
Seed Traits Ampelocera hottlei Length (cm) Width (cm) Volume (cm3) Weight (g) Brosimum lactescens Length (cm) Width (cm) Volume (cm3) Weight (g) Dialium guianense Length (cm) Width (cm) Volume (cm3) Weight (g) Spondias mombin Length (cm) Width (cm) Volume (cm3) Weight (g) Spondias radlkoferi Length (cm) Width (cm) Volume (cm3) Weight (g)
Alouatta pigra feces
Ateles geoffroyi feces
Tree crown
Ground
F 3, 436
P
1.18 ± 0.07b 0.87 ± 0.07b 3.78 ± 0.78b 0.44 ± 0.08b
1.13 ± 0.05a 0.81 ± 0.07a 3.21 ± 0.61a 0.33 ± 0.08a
1.13 ± 0.06a 0.84 ± 0.05a 3.35 ± 0.56a 0.42 ± 0.07b
1.13 ± 0.06a 0.83 ± 0.05a 3.33 ± 0.55a 0.42 ± 0.073b
14.8 14.1 16.7 45.3
