SEARCH

SEARCH BY CITATION

Keywords:

  • frugivorous bats;
  • habitat disturbance;
  • seed dispersal

ABSTRACT

  1. Top of page
  2. ABSTRACTRÉSUME
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. LITERATURE CITED
  8. Supporting Information

To assess the impact of bats on seed dispersal in a tropical mature forest (Nouragues, French Guiana), we conducted a bat exclusion experiment and tested the hypotheses that an artificial reduction in the abundance of bats would result in: (1) a decrease in seed species diversity, at the community level; and (2) an increase in seed limitation (a failure of seeds to reach all suitable sites for germination) at the species level. Seed rain was sampled in two contiguous forest plots for a total of 120 d, using 49 seed traps (1 m2) arranged in 7 × 7 grids and spaced at 5-m intervals. Using mist nets, bat activity was reduced in one forest plot for a total of 60 nights. Thirty-nine plant species, or species groups, likely to be consumed and dispersed by bats, were collected within a total sample of 50,063 seeds. The overall seed rain was dominated by epiphytic Araceae and Cyclanthaceae (83.3%) and tree species within the genera Cecropia and Ficus (16.0%). Seeds from bat-dispersed shrubs and treelets (Piper, Solanum, and Vismia) were relatively rare (0.7%). The bat exclusion resulted in a 30.5 percent reduction in seed species richness and increased seed limitation for most of the species sampled. Seed limitation was caused mainly by a reduced seed rain (seed source limitation) rather than a decrease in seed dispersal uniformity (seed dispersal limitation). Therefore, bat-dispersed plants with low seed production are likely to be particularly affected by a decline in bat abundance, as a result of anthropogenic change.

RÉSUME

La dispersion des graines par les chauves-souris frugivores semble déterminante pour la restauration de la diversité végétale dans les forêts tropicales perturbées. Afin d'évaluer l'impact des chauves-souris sur la dispersion des graines dans une forêt mature (Nouragues, Guyane Française), nous avons entrepris une exclusion expérimentale des chauves-souris permettant de tester l'hypothèse qu'une réduction artificielle de leur abondance résulterait en: 1) une diminution de la diversité spécifique des graines à l'échelle de la communauté, et 2) une augmentation de la limitation des graines (mesure de l'échec des graines à atteindre tous les microsites favorables) à l'échelle de l'espèce. La pluie de graines a étééchantillonnée dans deux parcelles forestières contiguës pendant 120 jours, au moyen de 49 collecteurs disposés en grilles de 7 × 7 à intervalles de 5 m. l'activité des chauves-souris a été réduite dans l'une des parcelles au moyen de captures au filet durant 60 nuits. Trente-neuf espèces ou morpho-groupes de plantes susceptibles d'être consommées et dispersées par les chauves-souris ont été identifiés parmi les 50 063 graines collectées. Globalement, la pluie de graines était dominée par les Aracées et Cyclanthacées épiphytes (83,3%) et les arbres des genres Cecropia et Ficus (16,0%), tandis que les graines des arbustes et arbrisseaux chiroptérophiles (Piper, Solanum, Vismia) étaient relativement rares (0,7%). En moyenne, 30,5 pourcent moins d'espèces étaient collectées pendant l'exclusion des chauves-souris, et les valeurs de limitation des graines ont augmenté pour la plupart des espèces étudiées. La limitation des graines résultait davantage d'une restriction de la quantité de graines dispersées (limitation de la source) plutôt que d'une moins bonne uniformité de leur dissémination (limitation de la dispersion). Ainsi, les plantes chiroptérophiles produisant peu de graines pourraient être plus affectées par le déclin des chauves-souris frugivores dans les milieux modifiés par l'homme.

Seed rain, the flow of seeds dispersed and deposited by wind, animals, and other dispersal agents, is an essential component of natural forest succession and plant diversity restoration in tropical human-disturbed lands. Seed availability ranks among the most limiting factors for plant regeneration in abandoned pastures and agricultural areas (Wunderle 1997, Benitez-Malvido 1998, Duncan & Chapman 1999, Martínez-Garza & González-Montagut 1999, 2002, Holl 1999, Wijdeven & Kuzee 2000, Cubina & Aide 2001). For instance, the density and diversity of both seed rain and seed bank decrease dramatically beyond a few meters from the edge of the forest (Holl 1999, Cubina & Aide 2001). Furthermore, depending on the intensity and duration of the disturbance, the soil seed bank may be substantially depleted, tree seeds being progressively replaced by an abundance of less-diverse seeds of second-growth plants (Uhl et al. 1988, Nepstad et al. 1996). All these mechanisms hamper forest regeneration. An influx of seeds from off-site forest habitats is thus a prerequisite for the reestablishment of the native flora in disturbed areas.

Given their high mobility and relative ubiquity, frugivorous birds and bats are thought to be keystone contributors to seed rain in disturbed areas. The seed rain provided by these aerial dispersers has been documented in open habitats such as pastures and agricultural lands. In these areas, the seed rain tends to concentrate around perches used by birds and bats (McClanahan & Wolfe 1993, Nepstad et al. 1996, Holl 1998, Duncan & Chapman 1999, Galindo-Gonzales et al. 2000), such as shrubs, treelets, or trees, and particularly those that produce fleshy fruits (Nepstad et al. 1996). Moreover, it has been argued that frugivorous bats disperse more seeds than birds in open areas, partly because birds mostly defecate seeds from a perched position whereas bats also defecate while flying (e.g., Foresta et al. 1984, Charles-Dominique 1986, Thomas et al. 1988, Gorchov et al. 1993, Medellín & Gaona 1999).

Although seed rain surveys have been conducted in open areas, no attempt has so far been made to assess the seed rain pattern produced by bats inside disturbed or undisturbed forests. Yet frugivorous bats are strongly affected by forest disturbances and bat populations may suffer a dramatic decrease in abundance in remote fragments of formerly pristine forests (Fenton et al. 1992, Estrada et al. 1993, Brosset et al. 1996, Cosson et al. 1999, Schulze et al. 2000, Estrada & Coates-Estrada 2001, Henry 2005; J.-M. Pons and J.-F. Cosson, pers. comm.). This reduction in frugivorous bats may, in turn, have detrimental repercussions on both seed rain diversity and seed dispersal efficiency. We investigate this problem by focusing on a small group of Neotropical plant species belonging to seven taxonomical families that are likely to be dispersed by bats. The model used was that of bat-dispersed plants whose seeds are reported to frequently occur in bat fecal samples and are commonly cited in diet studies. Once ingested, these small seeds (< 5 mm in length) may travel several hundred meters before being released through defecation, thereby increasing the chances of long-range (re)colonization events.

In this study, the relationship between variations in seed rain and the local availability of seed-dispersing bats was assessed by means of a bat exclusion experiment. We describe the seed rain pattern at the community level (seed species diversity) and the species level (seed limitation) using seed traps. Seed limitation parameters measure the failure of seeds to reach all microsites suitable for germination and establishment (Eriksson & Ehrlen 1992, Turnbull et al. 2000, Dalling et al. 2002, Muller-Landau et al. 2002). Our specific objectives were: (1) to provide a brief description of the seed rain pattern of the target plant species; (2) to test the hypothesis that a decreased bat abundance will reduce seed rain diversity and increase seed limitation; and (3) to determine whether increased seed limitation results from a restriction of the quantity of dispersed seeds over the study site (“seed source limitation”) and/or from a restriction of the spatial scattering of the dispersed seed (“seed dispersal limitation”).

METHODS

  1. Top of page
  2. ABSTRACTRÉSUME
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. LITERATURE CITED
  8. Supporting Information

Study area.— The seed rain sampling protocol was carried out at the Nouragues research station (4°50′ N, 52°42′ W) in the core area of the Nouragues Natural Reserve in northern French Guiana. This region is covered by continuous undisturbed rain forest, characterized by trees 30 to 35 m high, mostly belonging to the families Caesalpinaceae, Sapotaceae, and Lecythidaceae, and a fairly open understory (Poncy et al. 2001). Total annual rainfall ranges from 2500 to 3200 mm with a marked dry season from August to November. Bat surveys conducted within the study area reported the occurrence of 76 species, among which seven are understory frugivores and 12 are canopy frugivores (Brosset et al. 2001, Delaval et al. 2005). Altogether, frugivorous bats accounts for 43 percent of mist net captures. The numerically dominant frugivorous species are Artibeus jamaicensis, A. obscurus, and Rhinophylla pumilio, each accounting for 16 percent to 19 percent of all frugivorous bats.

Study species.— After studying the current literature on bat–plant interactions at Nouragues and elsewhere in French Guiana, we identified seven plant families known as keystone resources for bats in our study area, namely Cecropiaceae (with genus Cecropia), Moraceae (Ficus), Piperaceae (Piper), Solanaceae (Solanum), Clusiaceae (Vismia), and epiphytic Cyclanthaceae and Araceae (Philodendron) (Charles-Dominique 1986, 1995, Cosson 1994, Cockle 1997, 2001, Charles-Dominique & Cockle 2001, Geiselman et al. 2002, Lobova et al. 2003; Delaval 2004, Delaval et al. 2005). All these plants produce small seeds (0.5 to 3.5 mm in length) typically ingested and defecated by dispersers. They also provide fruit resources available year-round, although in some species fruiting tends to peak at certain periods of the year (e.g., Piper species, Thies & Kalko 2004). Life-forms are diversified, encompassing pioneer trees (Cecropia species), epiphytic or strangler trees becoming large free-standing trees (Ficus species), epiphytes growing on trunks from understory to subcanopy levels (Araceae and Cyclanthaceae), and pioneer or understory shrubs and treelets (Solanum, Piper and Vismia species).

Experimental design.— To study the effects of a reduction in bat abundance on the seed rain pattern, we introduced a disturbance of the bat population on one seed rain sampling site by the means of an exclusion experiment. Seed rain was measured in the exclusion site and a nearby control site, both before and during the experimental exclusion. The intersite differences in seed rain pattern observed during the exclusion were compared to expected differences recorded before the exclusion. The experiment was undertaken from early March to early May of 2003 and was repeated during the same period in 2004. Sites were located close to a ca 5-m wide stream in a relatively flat area.

There was a risk of the experiment being biased by possible changes in the availability of seeds due to changing phenology of species in one of the two sampling sites. To keep such risks as low as possible, we worked on a relatively short spatial scale, with small (0.12 ha, see below) contiguous sampling sites unlikely to differ much in plant composition and phenology. Moreover, to reduce the risks of high intersite discrepancies in seed flows, we inspected the tree database computed by Poncy et al. (2001) to make sure that no adult individuals (dbh > 30 cm) of Cecropia and Ficus species were present within the sampling sites. Finally, we conducted a survey of the target shrubs and epiphytes within the experimental and control plots to ensure that intersite differences in parent plant abundance and composition was reasonably low. We found 196 epiphytes from 12 target species in the exclusion plot and 237 epiphytes from 13 target species in the control plot. Values of relative species abundance were significantly correlated between the two sites (Spearman rank correlation = 0.869, P < 0.001). Overall, only five individuals of Piper spp. and one of Solanum sp. were found, all juveniles (0.5 to 1.5 m tall). We thus felt comfortable about treating the two sites as nonindependent plots regarding variations in plant phenology.

We sampled the seed rain during sampling sessions of 30 consecutive days, with the “before exclusion” session immediately preceding the “during exclusion” session (total sampling duration: 30 days × 2 sampling sessions × 2 experiments = 120 days). Longer sampling sessions could include major phenological variations in seed production likely to mask the effects of bat exclusion. Earlier pilot studies (M. Henry, pers. obs.) revealed that the coefficient of variation of mean number of collected seeds per trap varies little beyond 15–20 d for the most common species. Furthermore, this time scale fits latency periods separating seed deposition from germination and from seedling stage (e.g., 20–30 d and 1.5–2 mo, respectively, for target epiphyte species; Cockle 1997) and during which competitive exclusion constrains processes of plant recruitment (Nathan & Muller-Landau 2000).

Seed rain sampling.— The seed rain was sampled by means of 1-m2 seed traps. We used square plastic sheets stretched between four trees and/or aluminum poles ca 1 m above the ground to avoid disturbance by terrestrial animals. To drain the water, each trap was fitted with a central cylinder-shaped filter 4 cm in diameter and 14 cm long, made of stainless steel wire mesh wrapped in a light permeable polyester fabric. This filter was placed vertically through the trap, secured with putty resin (Rubson©, Henkel, Boulogne Billancourt, France).

Each sampling site was a 35 × 35 m (0.1225 ha) quadrat made up of 49 subquadrats 5 × 5 m in size. A seed trap was set as close as possible to the center of each of the 49 subquadrats. The surface sampled by the 49 traps (49 m2) represented four percent of the total surface of each sampling site. Exclusion and control sites were separated by a 35-m wide narrow buffer zone. Traps were emptied in the middle and at the end of each sampling session (day 15 and 30, respectively). This was done by collecting the entire contents (including leaves, branches, dust, etc.) of each trap in a waterproof bag. All material removed from the traps was thoroughly rinsed over a series of three sieves. The first sieve retained particles from 0.125–0.625 mm, the second sieve retained particles from 0.626–5.0 mm, and the third sieve retained particles grater than 5.0 mm. The resulting samples with particles < 5 mm (maximal seed length that can be swallowed by the bats) was stored in small opaque plastic bags to maintain the seeds dormant during several (2–12) weeks until examination under a binocular microscope (6× to 16× magnification) to count and collect the seeds. Turgescent seeds are easier to identify than dried seeds, and if necessary can be germinated for taxonomical confirmation. In about 0.04 percent of the samples, this moist environment favored the development of numerous mycelium filaments, but the seeds were still easily identifiable. Intact seeds were assigned to family, genus, species, or morphospecies on the basis of their external characteristics with the help of a seed reference collection (Muséum National d'Histoire Naturelle, Dept. Ecologie et Gestion de la Biodiversité, Brunoy, France).

Bat exclusion.— A phenomenon well known to bat biologists is the sharp decline of the capture rate when netting bats over several days at the same capture site (e.g., Thomas & LaVal 1988). This capture depletion may result from a combination of mist net habituation by bats and a general avoidance of the area. Given the small surface area of the seed rain sampling sites, we assumed that setting many nets inside and around a site would locally impede bat movements and increase the cost of flying over the area at the understory level, eventually leading to a substantial reduction of bat activity during the experimental session.

The exclusion was achieved by using 15 12 × 2.5 m nets (mesh 16 mm) simultaneously, left open all night (1830 h to 0630 h), weather permitting. Accidental bird captures were fairly rare and concerned frugivorous individuals on two occasions only. Since the nets were closed during periods of abundant rainfall, they were only open for 22 and 23 nights of the 2003 and 2004 exclusion sessions respectively, i.e., 73 and 77 percent of the 30 nights. Nets were checked every 30 min during the first and the last 2 h of the night, and every 90 min the rest of the time. Three to four nets were moved every 2 d to reduce a possible effect of mist net habituation by bats. This exclusion technique is likely to affect bats foraging in the understory more than other species, although bats eating canopy fruits also often commute in lower forest strata. Captured frugivorous bats were kept in cloth bags and later identified (Simmons & Voss 1998, Charles-Dominique et al. 2001), ringed with numbered plastic wing-bands (A. C. HUGUES, England) and released at the research station, 350 m away from the sampling site (except lactating females that were rapidly released close to the capture site).

Measures of seed species diversity.— To compare the patterns of seed species diversity during the sessions of the study, we plotted the curves of species richness accumulation in relation to sampling effort (number of seed traps) for each sampling site and for each session. Curves were smoothed by means of 100 random reorganizations of the trap orders using the software EstimateS (version 5, R. K. Colwell, URL: http://viceroy.eeb.uconn.edu/estimates). We used the accumulation rate of new species with increasing sampling effort (i.e., the slope of the accumulation curves) as a diversity indicator. Species richness accumulation curves are typically composed of two parts, an initial steeply ascending phase followed by a near-linear, less steep, phase corresponding to a drastic slowdown of species accumulation. In this analysis, we focused on the quasi-linear second part of the curves, and used Generalized Linear Models (GLMs) to compare the accumulation slopes obtained before and during bat exclusion at each session. Using preliminary exploratory linear regressions, we found that the N= 20 final data points of each curve were sufficient to provide the tightest linear fits and to permit reliable GLMs (although GLMs virtually delivered identical results over the range N= 10 to 35).

Measures of seed limitation.— In order to quantify the failure of seeds to reach all suitable microsites (here materialized by the traps), we used the fundamental seed limitation index “FL” (Nathan & Muller-Landau 2000, Muller-Landau et al. 2002) that measures the proportion of traps that seeds do not reach

  • image

FL is a combination of the limitation of seed quantity (or source limitation, “SL”) and limitation of seed dispersal (or dispersal limitation “DL”). SL refers to seed limitation due solely to insufficient seed numbers deposited within the study site, whereas DL refers to seed limitation due to nonuniform deposition of seeds among traps (Muller-Landau et al. 2002). SL was calculated as the probability that no seed reaches a given trap during the sampling session, i.e., the probability for zero dispersal events. It is assumed that the occurrence of a seed in a given trap is a rare and random event and that consequently the corresponding probability of occurrence may follow a Poisson distribution (Muller-Landau et al. 2002). Source limitation SL is then the Poisson probability of a zero event given an expectation of [number of collected seeds/number of traps] events

  • image

Finally, DL measures the ratio between the proportion of traps effectively reached by seeds and the proportion of traps that would be reached by seeds given a perfectly uniform seed distribution

  • image

To test the hypothesis that the experimental bat exclusion increases seed limitation (FL, SL, and DL), we began by calculating the “intersite differences” for each species and each sampling session, as: intersite difference =[seed limitation in exclusion site]−[seed limitation in control site]. We then used paired t-tests to compare the mean “before-exclusion” and “during-exclusion” intersite differences in order to visualize the general trends regarding the relation between bat exclusion and seed limitation values. This indirect method to analyze the effect of bat exclusion has the advantages: (1) to take into account both intersite and intersession variations in a single test; and (2) to avoid large batteries of analogous pairwise tests with increased risks of type I errors.

Seed limitation values are likely to be spurious for the rarest seed species with very few dispersal events. We thus fixed at 0.95 the maximum acceptable FL (averaged on the two sites) for inclusion of a species in the analyses. With the mean FL > 0.95 during a given sampling session, the reliability of SL and DL values may become questionable. Nevertheless, in order to include the rarer seed species in the analyses, we grouped together closely related species or species producing fruits harvested and consumed in a similar way by frugivores: the typical bat shrubs and small trees (species of Piper, Solanum and Vismia), the three Ficus species known to be consumed by bats at Nouragues (F. amazonica, F. insipida, F. nymphaefolia; Delaval et al.[2004]), the rarest Ficus species (F. trigona and undetermined species of Ficus) and the rarest Philodendron species (P. deflexum, P. melinonii and “oval-seeded species of Philodendron”). The analyses were thus conducted on “functional units” representing either individual species or groups of rarer species

The significance level was set at α= 0.05 and all statistical tests were computed using the software Systat 9.0.

RESULTS

  1. Top of page
  2. ABSTRACTRÉSUME
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. LITERATURE CITED
  8. Supporting Information

General seed rain description.— A total of 381 trap samples yielded 50,063 seeds < 5 mm in length belonging to 39 species or morpho-groups (Table S1) of the seven target plant families. Eleven samples out of 392 (98 traps × 4 study sessions) were lost due to windfalls collapsing the trap. Overall seed rain density averaged 131 seeds per m2 and per month (s/m2/mo) over the duration of the experiment, ranging from 57 to 219 s/m2/mo during individual sessions. The seed rain was largely dominated by epiphytic Cyclanthaceae and Araceae species (56.1% and 27.2%, respectively), followed by Cecropia species and, to a lesser extent, by Ficus species (13.0% and 3.0%, respectively). The typical shrub and treelet bat plants Piper, Solanum, and Vismia remained rare or uncommon in seed samples (< 0.7%).

We considered the seed sampling protocol to be little affected by seeds falling directly into the traps without the mediation of an animal disperser because we never found target seeds embedded in unconsumed fruits or pulp. However, it was impossible to ascertain whether seeds were effectively dispersed by bats or other tree-dwelling dispersers (birds, primates, opossums, kinkajous).

Bat exclusion.— The experimental bat exclusion led to the capture of 129 individual frugivorous bats belonging to 10 species (Table 1). Bat assemblages differed from one year to the other. In 2003, 84.5 percent of individuals belonged to the species of the guild of understory frugivorous bats (i.e., mainly feeding on fruits produced by understory plants), the other 15.5 percent being indexed as canopy frugivorous bats (i.e., mainly feeding on fruits produced by canopy trees). The ratio was reversed in 2004, with 21.1 and 78.9 percent of understory and canopy frugivorous bats, respectively (Table 1). The difference in the proportions of the two guilds from one year to the next was highly significant (χ2= 31.22, P < 0.001). Bat captures were dominated by the understory species R. pumilio (46.5%) in 2003 and by the canopy species A. jamaicensis (50.7%) in 2004. Several observations support our expectation that most bats avoided the area during the exclusion experiment. First, capture rates significantly decreased during the 2003 session (linear regression, N= 22, P < 0.001, R2= 0.48) and nearly so in 2004 (N= 23, P= 0.06, R2= 0.15), ranging from about 5 captures/10 net-nights during the first 10 d, down to < 1 capture/10 net-nights during the last 10 d. Second, only two individuals were recaptured on the experimental plot, leading to a local recapture probability of 1.9 percent, while the recapture probability over the 1-km2 study area usually averages 19.7 percent (±13.1 SD, range: 1.9–35.7) for the same bat species and with a similar time span (Delaval et al. 2005).

Table 1. List of captured frugivorous bats during the experimental mist netting exclusion (U = guild of understory frugivorous bats; C = guild of canopy frugivorous bats) and main diet items (see methods for literature review). Netting effort is of 22 and 23 whole nights with 15 nets for, respectively, 2003 and 2004.
Species (average body mass) Guild20032004Main diet
Carolliinae subfamily
 Carollia brevicauda(12 g)U 5 1Piper spp., Solanum spp.
 Carollia perspicillata(17 g)U11 1Piper spp., Solanum spp.
 Rhinophylla pumilio(8.8 g)U2710Araceae, Cyclanthaceae, Cecropia sciadophylla
Stenodermatinae subfamily
 Artibeus gnomus(9.9 g)C  2Ficus spp.
 Artibeus jamaicensis(56 g)C 336Cecropia obtusa, Ficus spp.
 Artibeus lituratus(66 g)C 3 4Cecropia obtusa, Ficus spp.
 Artibeus obscurus(36 g)C 312Cecropia obtusa, Ficus spp.
 Ectophylla macconnelli(8 g)C  1Ficus spp.
 Sturnira tildae(23 g)U 6 3Solanum spp., Araceae, Cecropia obtusa
 Uroderma bilobatum(17 g)C  1(Not available)
Total understory frugivorous bats U4915 
Total canopy frugivorous bats C956 

The effect of bat exclusion on seed diversity.— The seed species richness (number of species or morpho-groups) per site and per sampling session ranged from 13 to 25 (Table S1). Despite their relative scarcity, Piperaceae and Solanaceae accounted for more than half of the seed species richness (26% and 28%, respectively). The total number of species and morpho-groups decreased in the exclusion site (−33% and −28% for 2003 and 2004, respectively), while it remained stable or even increased in the control site (+32% and +12%, respectively). The decrease observed in the exclusion site compared to the control site is mainly due to the absence of some rare seed species (mostly Piperaceae, Solanaceae and Araceae).

All curves of seed species richness accumulation clearly include a transition zone between an initial steeply ascending phase and an almost linear second phase with a more gentle slope (Fig. 1). The gradient of the second section of the slope was lower during the exclusion sessions, when bat abundance was reduced (−77% and −72% for 2003 and 2004, respectively), whereas it increased in the control site (+24% and +73% for 2003 and 2004, respectively) (Fig. 1). All these changes were highly significant (GLM: N= 40; df = 1; F= 73.10 to 538.67; R2= 0.658 to 0.934; P < 0.001 in every case).

image

Figure 1. Seed species accumulation (number of collected species and morpho-groups) in relation to the seed rain sampling effort (number of seed traps set and examined) in exclusion and control sites, before (left) and during (right) the experimental bat exclusion, in 2003 and 2004. The curves were smoothed by 100 random reorganizations of the trap order. Bars materialize the standard deviation. The numbers refer to the slope of the linear section beyond the shoulder (see Methods).

Download figure to PowerPoint

The effect of bat exclusion on seed limitation.— Seed limitation values were calculated for nine functional units (i.e., species or species morpho-groups) in 2003 and 11 functional units in 2004 (see Table S2). We found that species significantly experienced an increase in fundamental limitation (FL) during exclusion (Fig. 2A). In both 2003 and 2004, FL increased significantly more in the exclusion site than in the control site—this also expressing consistent results from one year to the next. In a second step, we split FL into seed source limitation (SL) and seed dispersal limitation (DL). These two seed limitation components responded differently to the bat exclusion experiment, the same trend being observed in 2003 and 2004. SL, like FL, displayed a significant increasing trend during exclusion as compared with the control site (Fig. 2B). In contrast, DL appeared to remain unaffected overall by the bat exclusion (Fig. 2C).

image

Figure 2. Comparison among study sessions of mean intersite differences in seed limitation of the studied plant groups. The error bars equal 1 SEM (standard error of the mean). FL and SL values were lower on the exclusion site than on the control site before the exclusion (site difference < 0), but became higher during the bat exclusion (site difference > 0). These significant variations (as given by paired t-tests) indicate that seed species tend to display greater seed limitation values overall (as a result of bat exclusion) than expected considering control site results. (*P < 0.05, **P < 0.01, ns = not significant; N= 9 and N= 11 in 2003 and 2004, respectively).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. ABSTRACTRÉSUME
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. LITERATURE CITED
  8. Supporting Information

General seed rain description.— Detailed data on parent plant density, productivity, and phenology are needed to adequately compare the seed limitation values of different species and to reach meaningful conclusions on the relative importance of source limitation and dispersal limitation. However, judging from the quantity and ubiquity of their seeds found in the traps, there is little doubt that the epiphytic Araceae and Cyclanthaceae (all together totaling over 83% of the collected seeds) rank among the least seed-limited plants of the studied species. The low FL of epiphytes may result from their great abundance around the sampling site and from the huge number of minute seeds that a single large infructescence may contain (from 5 × 103 to 200 × 103; Cockle 1997). Our epiphyte survey revealed that over 80 percent of the trees supported adult individuals within the sampling sites. Moreover, the dispersal mode provided by the common understory frugivorous bat R. pumilio, specialized on epiphytic Araceae and Cyclanthaceae, may optimize the spatial scattering of seeds. Since the infructescences of these plants are too large to be removed by bats, they are directly eaten in situ by R. pumilio, during feeding bouts of an average of 11.6 min according to radiotracking surveys (Henry and Kalko, 2007). Given that food transits very quickly through the digestive tract (5–30 min; Charles-Dominique 1986, Cockle 1997), defecation may often occur over the foraging area as the bat visits widely scattered epiphytes, thus enhancing seed dispersal uniformity.

In contrast, the typical bat-dispersed shrub and treelet species of Piper, Solanum, and Vismia produce fewer seeds per fruiting individual (only several tens to several hundreds at a time) and are visibly less abundant in the study area. This may partly explain their high FL values compared to epiphytes. Their dispersers, mainly Carollia species at Nouragues (Delaval et al. 2005), may also contribute to seed limitation through the repetitive use of feeding roosts since they pick up the berries of Solanum and the fruiting spikes of Piper and consume them away from the parent plant. In his study of a Costa Rican bat-dispersed species of Piper, Fleming (1981) estimated that over 90 percent of seeds were deposited beneath feeding roosts, with 40 percent left on the discarded part of the spike. High fidelity to particular roosts may thus result in clumped or “contagious” seed dispersal (sensuSchupp et al. 2002), as seen in plants dispersed by spider monkeys (Russo & Augspurger 2004), and lead to an underestimation of the seed rain and dispersal limitation as measured with seed traps.

The effects of bat exclusion.— In agreement with our hypothesis, the experimental bat exclusion induced a significant drop in seed diversity (as indicated by both species richness and accumulation rate of new species in the samples). We should therefore expect a substantial reduction of the diversity of the bat-generated seed rain in disturbed forests with a lower bat activity. The simultaneous rise of seed diversity recorded on the control site suggests that the lower bat activity in the exclusion site was compensated by higher activity in the nearby control site. This does not invalidate our experiment, which aimed at comparing seed rain characteristics under two different levels of bat activity.

The fall in diversity recorded at community level is to be linked to a general and significant increase of the fundamental limitation FL among the studied functional units, that is, a higher failure of seeds to reach all traps. The general trend of FL values during bat exclusion confirms the role of bats as efficient dispersers for the studied plants. A rise of FL may curtail plant diversity restoration in isolated forest remnants or other disturbed areas avoided by bats. A notable exception was Ficus guianensis, a species producing small reddish figs consistent with the ornithochorous syndrome and probably less dispersed by bats than by other frugivores (Neotropical bat-figs tend to be greenish in color; Kalko et al. 1996, Korine et al. 2000).

There are two possible explanations for the failure of seeds to reach all traps: the seeds may not be numerous enough (seed source limitation SL), or may not be dispersed evenly enough (seed dispersal limitation DL). Our experiment revealed that reducing bat activity tended to significantly increase SL, while DL remained mostly unchanged. An exception to this is Cecropia sciadophylla, a mostly ornithochorous pioneer tree, for which DL increased substantially during bat exclusion. Since birds usually defecate from a perched position whereas bats also defecate on the wing (e.g., Charles-Dominique 1986, Gorchov et al. 1993), the small proportion of C. sciadophylla fruits dispersed by bats (Lobova et al. 2003) may have a greater effect on DL than on SL.

The repetition of the experiment in 2004 produced fairly similar results overall, but with noticeable variations at the level of the individual species. In some species/functional units, the exclusion effect was quite a lot lower in one year than in the other. These interannual variations could be partly explained by the very dissimilar bat assemblages observed in 2003 and 2004 (dominated by understory and canopy frugivorous bats, respectively). The effect of exclusion on the limitation values of the commonest Araceae and Cyclanthaceae, for example, appeared greater in 2003, when their dispersers were more abundant (understory bats), whereas the exclusion effect on bat-consumed Ficus appeared greater in 2004 when the canopy fig-eating bats were more abundant. The greater abundance of canopy frugivorous bats in 2004 may also result from the presence of fruiting Ficus trees attracting many Artibeus in the vicinity (Morrison 1978). Contrasting weather conditions may also generate interannual variability. Although the 2003 and 2004 study periods occurred at the same time of the year, the weather was much drier in 2003 than in 2004, with 38 percent less total rainfall and 46 percent more days with less than 1 mm rainfall. Rainfall is liable to have an impact on the timing of fruit production, which could indirectly affect bat activity. Our data sets did not, however, allow us to test this hypothesis.

In spite of the overall conclusive results, the scope of our experimental approach remains limited in space and time and forced us to subjectively pool some plant species. Furthermore, hypothesis testing on a species-per-species basis was problematic due to the scarcity of seed limitation values per treatment. More extensive surveys monitoring natural seed rain variation across larger temporal and geographical scales in relation to disperser availability could side-step these difficulties and might give more accurate seed rain profiles for these small ingested seeds.

Bats and plant recruitment in tropical forests.— In this study, we expose a link between the activity of frugivorous bats and the seed limitation of some of their resource plants. Frugivorous bats can reasonably be considered efficient dispersers in this respect, but there is a conceptual distinction between dispersal efficiency and dispersal effectiveness (Bustamante & Canals 1995). While dispersal efficiency refers to the propensity of dispersers to disseminate seeds in suitable places for germination, dispersal effectiveness measures the contribution of dispersers to seedling establishment and recruitment success. Studies linking the activity of dispersers to plant recruitment patterns (e.g., Herrera et al. 1994) are scarce because “closing the seed dispersal loop” is a tedious task for complex systems involving several to many potential dispersers and plant competitors (Schupp & Fuentes 1995, Wang & Smith 2002).

Post-dispersal seed fate was out of the scope of this experiment and further studies are required to bridge bat activity and plant recruitment patterns. Seed limitation in itself is an intermediary stage that does not necessarily mirror effective recruitment limitation (Nathan & Muller-Landau 2000). Small-seeded plants, in particular, are known to suffer from much lower seed-to-seedling transition probabilities than large-seeded plants (Harms et al. 2000). In particular, exclusion by superior competitors (Turnbull et al. 1999) or a limitation of suitable microsites for germination and development (Eriksson & Ehrlén 1992, Harms et al. 2000, Dalling & Hubbell 2002) might be the main factors responsible for recruitment limitation. By definition, traps are regarded as suitable microsites, but effective suitable microsites may be much rarer, especially for light-demanding pioneer plants like Cecropia (Dalling & Hubbell 2002, Dalling et al. 2002). Further studies should specifically aim at quantifying the microsite limitation in relation with fundamental seed limitation.

Conclusions on seed dispersal in disturbed forests.— As we hypothesized, we found evidence that a disturbance of bat activity has negative repercussions on seed rain diversity and increases the fundamental seed limitation FL of their resource plants. We found that this failure of seeds to reach all suitable microsites as a result of bat activity depletion tends to proceed from an increase in source limitation SL (lower seed numbers in the sampling site) rather than an increase in dispersal limitation DL (less even seed deposition). In other words, intrinsically source-limited bat plants might be particularly likely to suffer from bat activity depletion in disturbed areas. Accordingly, in our study, disappearance of the rarest seed species from seed rain samples was what contributed most to the loss in diversity during bat exclusion. This heralds what we would expect in disturbed forests in which frugivorous bat activity is reduced.

It thus appears that dispersal limitation DL was not the main contributor to fundamental seed limitation FL. Frugivorous bats can still produce a spatially uniform seed deposition pattern even in case of a substantial reduction of their activity (and at least up to a certain threshold that cannot be determined in this study). This underlines the crucial role of bats, which even in disturbed forests are capable of delivering an efficient minimum service of seed dispersal.

ACKNOWLEDGMENTS

  1. Top of page
  2. ABSTRACTRÉSUME
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. LITERATURE CITED
  8. Supporting Information

We wish to thank Pierre Charles-Dominique (CNRS) for providing valuable criticism and suggestions, as well as Jean-Marc Pons (MNHN), who contributed to the design of the seed sampling protocol and assisted in the first trials. We are also specially thankful to Natalia Norden and Patrick Châtelet for their help in the field. Adeline Caubert and Pierre Belbenoit (MNHN-CNRS, UMR 5176) provided useful information for seed identification. Tatyana A. Lobova and Scott Mori (New York Botanical Garden), Anya Cockle-Betian, Kim McConkey, as well as two anonymous referees made helpful comments and language corrections to improve the manuscript. We are grateful to Elisabeth Kalko, Pierre-Michel Forget, and Jean-François Cosson for their help and suggestions in earlier stages of the work. The fieldwork was supported by the Department Ecologie et Gestion de la Biodiversité (MNHN-CNRS, UMR 5176, Brunoy, France). M. H. received a Ph.D. grant from La Fondation des Treilles. Our experiments comply with the current French environmental legislation.

LITERATURE CITED

  1. Top of page
  2. ABSTRACTRÉSUME
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. LITERATURE CITED
  8. Supporting Information
  • Benitez-Malvido, J. 1998. Impact of forest fragmentation on seedling abundance in a tropical rain forest. Conserv. Biol. 12: 380389.
  • Boggan, J., V. Funk, C. Kelloff, M. Hoff, G. Cremers, and C. Feuillet. 1997. Checklist of the plants of the Guianas (Guyana, Surinam, French Guiana), 2nd edition, Biological Diversity of the Guianas Program. Smithsonian Institution , Washington , DC .
  • Brosset, A., P. Charles-Dominique, and A. Cockle. 2001. The bat community. In F.Bongers, P.Charles-Dominique, P.-M.Forget, and M.Théry (Eds.). Nouragues: Dynamics and plant-animal interactions in a Neotropical rainforest, pp. 115120. Kluwer Academic Publishers, Dordrecht , The Netherlands .
  • Brosset, A., P. Charles-Dominique, A. Cockle J.-F., Cosson, and D. Masson. 1996. Bat communities and deforestation in French Guiana. Can. J. Zool. 74: 19741982.
  • Bustamante, R., and M. Canals. 1995. Dispersal quality in plants: How to measure efficiency and effectiveness of a seed disperser. Oikos 73: 133136.
  • Charles-Dominique, P. 1986. Inter-relation between frugivorous vertebrates and pioneer plants: Cecropia, birds and bats in French Guiana. In A.Estrada and T. H.Fleming (Eds.). Frugivores and seed dispersal, pp. 119135. Dr W. Junk Publishers, Dordrecht , The Netherlands .
  • Charles-Dominique, P. 1995. Interactions plantes-animaux frugivores, conséquences sur la dissémination des graines et la régénération forestière. Rev. Ecol. (Terre Vie) 50: 223235.
  • Charles-Dominique, P., A. Brosset, and S. Jouard. 2001. Atlas des chauves-souris de Guyane, Patrimoine Naturel, No 49. Muséum National d'Histoire Naturelle, Paris , France .
  • Charles-Dominique, P., and A. Cockle. 2001. Frugivory and dispersal by bats. In F.Bongers, P.Charles-Dominique, P.-M.Forget, and M.Théry (Eds.). Nouragues: Dynamics and plant-animal interactions in a Neotropical rainforest, pp. 207216. Kluwer Academic Publishers, Dordrecht , The Netherlands .
  • Cockle, A. 1997. Modalité de dissémination et d'établissement de lianes de sous-bois (Cyclanthaceae et Philodendron) en forêt guyanaise. Ph.D. Thesis, Université Paris VI, Paris , France .
  • Cockle, A. 2001. The dispersal and recruitment of Cyclanthaceae and Philodendron (Araceae) understorey root-climbing vines. In F.Bongers, P.Charles-Dominique, P.-M.Forget, and M.Théry (Eds.). Nouragues: Dynamics and plant-animal interactions in a Neotropical rainforest, pp. 251264. Kluwer Academic Publishers, Dordrecht , The Netherlands .
  • Cosson, J.-F. 1994. Dynamique de population et dispersion de la chauve-souris frugivore Carollia perspicillata en Guyane française. PhD Dissertation. Université Paris-Sud XI, Paris , France .
  • Cosson, J.-F., J.-M. Pons, and D. Masson. 1999. Effects of forest fragmentation on frugivorous and nectarivorous bats in French Guiana. J. Trop. Ecol. 15: 515534.
  • Cubina, A., and T. M. Aide. 2001. The effect of distance from forest edge on seed rain and soil seed bank in a tropical pasture. Biotropica 33: 260267.
  • Dalling, J. W., and S. P. Hubbell. 2002. Seed size, growth rate and gap microsite conditions as determinants of recruitment success for pioneer species. J. Ecol. 90: 557568.
  • Dalling, J. W., H. C. Muller-Landau, S. J. Wright, and S. P. Hubbell. 2002. Role of dispersal in the recruitment limitation of Neotropical trees. J. Ecol. 90: 714727.
  • DeLaval, M. 2004. Impact des perturbations d'origine anthropique sur les peuplements de chauves-souris en Guyane française. Ph.D. Dissertation. Université Paris VI, Paris , France .
  • DeLaval, M., M. Henry, and P. Charles-Dominique. 2005. Interspecific competition and niche partitioning: Example of a Neotropical rainforest bat community. Rev. Ecol. (Terre Vie) 60: 149166.
  • Duncan, R. S., and C. A. Chapman. 1999. Seed dispersal and potential forest succession in abandoned agriculture in tropical Africa. Ecol. Appl. 9: 9981008.
  • Eriksson, O., and J. Ehrlen. 1992. Seed and microsite limitation of recruitment in plant populations. Oecologia 91: 360364.
  • Estrada, A., and R. Coates-Estrada. 2001. Bat species richness in live fences and in corridors of residual rain forest vegetation at Los Tuxtlas, Mexico. Ecography 24: 94102.
  • Estrada, A., R. Coates-Estrada, and D. Meritt, Jr. 1993. Bat species richness and abundance in tropical rain forest fragments and in agricultural habitats at Los Tuxtlas, Mexico. Ecography 16: 309318.
  • Fenton, M. B., L. Acharaya, D. Audet, M. B. C. Hickey, C. Merriman, M. K. Obrist, and D. M. Syme. 1992. Phyllostomid bats (Chiroptera: Phyllostomidae) as indicators of habitat disruption in the Neotropics. Biotropica 24: 440446.
  • Fleming, T. H. 1981. Fecundity, fruiting pattern, and seed dispersal in Piper amalago (Piperaceae), a bat-dispersed tropical shrub. Oecologia 51: 4246.
  • Foresta, H. De, P. Charles-Dominique, C. Erard, and M. F. Prévost. 1984. Zoochorie et premiers stades de la régénération naturelle après coupe en forêt guyanaise. Rev. Ecol. (Terre Vie) 39: 369400.
  • Galindo-González, J., S. Guevara, and V. J. Sosa. 2000. Bat- and bird-generated seed rains at isolated trees in pastures in a tropical rainforest. Conserv. Biol. 14: 16931703.
  • Geiselman, C. K., S. A. Mori, and F. Blanchard. 2002. Database of Neotropical bat/plant interactions. http://www.nybg.org/botany/tlobova/mori/batsplants/database/dbase_frameset.htm.
  • Gorchov, D. L., F. Cornejo, C. F. Ascorra, and M. Jaramillo. 1993. The role of seed dispersal in the natural regeneration of rain forest after strip-cutting in the Peruvian Amazon. Vegetatio 107/108: 339349.
  • Harms, K. E., S. J. Wirght, O. Calderón, A. Hernández, and E. A. Herre. 2000. Pervasive density-dependent recruitment enhances seedling diversity in a tropical forest. Nature 404: 493495.
  • Henry, M. H. 2005. Le déclin des populations de chauves-souris frugivores en forêt néotropicale fragmentée—conséquences sur la dispersion des graines. Ph.D. Thesis. Université Paris VI, Paris , France .
  • Henry, M. H., and E. K. V. Kalko., 2007. Foraging strategy and breeding constraints of Rhinophylla pumilio (Phyllostomidae) in the Amazon lowlands. J. Mamm. 88: 8193.
  • Herrera, C., P. Jordano, L. López-Soria, and J. Amat. 1994. Recruitment of a mast-fruiting, bird-dispersed tree: Bridging frugivore activity and seedling establishment. Ecol. Monogr. 64: 315344.
  • Holl, K. D. 1998. Do bird perching structures elevate seed rain and seedling establishment in abandoned tropical pasture? Restor. Ecol. 6: 253261.
  • Holl, K. D. 1999. Factors limiting tropical rain forest regeneration in abandoned pasture: Seed rain, seed germination, microclimate, and soil. Biotropica 31: 229242.
  • Kalko, E. K. V., E. A. Herre, and C. O. Handley, Jr . 1996. Relation of fig fruit characteristics to fruit-eating bats in the new and old world tropics. J. Biogeog. 23: 565576.
  • Korine, C., E. K. V. Kalko, and E. A. Herre. 2000. Fruit characteristics and factors affecting fruit removal in a Panamanian community of strangler figs. Oecologia 123: 560568.
  • Lobova, T. A., S. A. Mori, F. Blanchard, H. Peckham, and P. Charles-Dominique. 2003. Cecropia as a food resource for bats in French Guiana and the significance of fruit structure in seed dispersal and longevity. Am. J. Bot. 90: 388404.
  • McClanahan, T. R., and R. W. Wolfe. 1993. Accelerating forest succession in a fragmented landscape—the role of birds and perches. Conserv. Biol. 7: 279288.
  • Martínez-Garza, C., and R. González-Montagut. 1999. Seed rain from forest fragments into tropical pastures in Los Tuxtlas, Mexico. Plant Ecol. 145: 255264.
  • Martínez-Garza, C., and R. González-Montagut. 2002. Seed rain of fleshy—fruited species in tropical pastures in Los Tuxtlas, Mexico. J. Trop. Ecol. 18: 457462.
  • Medellín, R. A., and O. Gaona. 1999. Seed dispersal by bats and birds in forest and disturbed habitats of Chiapas, México. Biotropica 31: 478486.
  • Morrison, D. W. 1978. Foraging ecology and energetics of the frugivorous bat Artibeus jamaicensis. Ecology 59: 716723.
  • Muller-Landau, H. C., S. J. Wright, O. Calderón, S. P. Hubbell, and R. B. Foster. 2002. Assessing recruitment limitation: Concepts, methods and case-studies from a tropical forest. In D. J.Levey, W. R.Silva, and M.Galletti (Eds.). Seed dispersal and frugivory: Ecology, evolution and conservation, pp. 3553. CAB International, New York , New York .
  • Nathan, R., and H. C. Muller-Landau. 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Trends Ecol. Evol. 15: 278285.
  • Nepstad, D. C., C. Uhl, C. A. Pereira, and J. M. C. DaSilva. 1996. A comparative study of tree establishment in abandoned pasture and mature forest of eastern Amazonia. Oikos 76: 2539.
  • Poncy, O., D. Sabatier, M.-F. Prévost, and I. Hardy. 2001. The lowland high rainforest: Structure and tree species diversity. In F.Bongers, P.Charles-Dominique, P.-M.Forget, and M.Théry (Eds.). Nouragues: Dynamics and plant-animal interactions in a Neotropical rainforest, pp. 3146. Kluwer Academic Publishers, Dordrecht , The Netherlands .
  • Russo, S. E., and C. K. Augsburger. 2004. Aggregated seed dispersal by spider monkeys limits recruitment to clumped patterns in Virola calophylla. Ecol. Lett. 7: 10581067.
  • Schulze, M. D., N. E. Seavy, and D. F. Whitacre. 2000. A comparison of the Phyllostomid bat assemblages in undisturbed Neotropical forest and in forest fragments of a slash-and-burn farming mosaic in Petén, Guatemala. Biotropica 32: 174184.
  • Schupp, E. W., and M. Fuentes. 1995. Spatial patterns of seed dispersal and the unification of plant population ecology. Ecoscience 2: 267275.
  • Schupp, E. W., T. Milleron, and S. E. Russo. 2002. Dissemination limitation and the origin and maintenance of species-rich tropical forests. In D. J.Levey, W. R.Silva, and M.Galletti (Eds.). Seed dispersal and frugivory: Ecology, evolution and conservation, pp. 1933. CAB International, New York , New York .
  • Simmons, N. B., and R. Voss. 1998. The mammals of Paracou, French Guyana: A Neotropical lowland rainforest fauna. Part I: Bats. Bull. Am. Mus. Nat. Hist. 237: 0219.
  • Thies, W., and E. K. V. Kalko. 2004. Phenology of Neotropical pepper plants (Piperaceae) and their association with their main dispersers, two short-tailed fruit bats, Carollia perspiscillata and C. castanea (Phyllostomidae). Oikos 104: 362376.
  • Thomas, D. W., and R. K. Laval. 1988. Survey and census methods. In T. H.Kunz (Ed.). Ecological and behavioral methods for the study of bats, pp. 7789. Smithsonian Institution Press, Washington , DC .
  • Thomas, D. W., D. Cloutier, M. Provencher, and C. Houlé. 1988. The shape of bird- and bat-generated seed shadows around a tropical tree. Biotropica 20: 347348.
  • Turnbull, L. A., M. J. Crawley, and M. Rees. 2000. Are plant populations seed limited? A review of seed sowing experiments. Oikos 88: 225238.
  • Turnbull, L. A., M. Rees, and M. J. Crawley. 1999. Seed mass and the competition/colonization trade-off: A sowing experiment. J. Ecol. 87: 899912.
  • Uhl, C., R. Buschbacher, and E. A. S. Serrão. 1988. Abandoned pastures in eastern Amazonia. I. Patterns of plant succession. J. Ecol. 76: 663681.
  • Wang, B. C., and B. T. Smith. 2002. Closing the seed dispersal loop. Trends Ecol. Evol. 17: 379385.
  • Wijdeven, S. M. J., and M. E. Kuzee. 2000. Seed availability as a limiting factor in forest recovery processes in Costa Rica. Restor. Ecol. 8: 414424.
  • Wunderle, J. M. Jr . 1997. The role of animal seed dispersal in accelerating native forest regeneration on degraded tropical lands. For. Ecol. Manage. 99: 223236.

Supporting Information

  1. Top of page
  2. ABSTRACTRÉSUME
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. ACKNOWLEDGMENTS
  7. LITERATURE CITED
  8. Supporting Information
FilenameFormatSizeDescription
Suppmat_BTP_286.doc182KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.