| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Departamento de Ecología Evolutiva, Instituto de Ecología, Universidad Nacional Autónoma de México (UNAM), Apartado Postal 70-275, México D.F. CP 04510, México
Received for publication September 24, 2004. Accepted for publication May 13, 2005.
ABSTRACT
We studied a community of Agave species that coexist in the Metztitlán Canyon in Central Mexico. During 2001, 2002, and 2003, we analyzed floral traits and rosette allometry in five species belonging to the subgenus Littaea: A. celsii albicans, A. xylonacantha, A. difformis, A. striata, and Agave sp.; and observed floral visitors for each species. We report the first evidence of bat visitation in the subgenus Littaea and find that bats (Leptonycteris curasoae, Choeronycteris mexicana, and Glossophaga sp.) are the primary pollinators in four species. Honeybees, hummingbirds, bumblebees and hawkmoths were also common visitors. We propose that the presence of diurnal pollinators may increase the reproductive success of the plant when offering pollinator services additional to the coadapted pollinator. We also found evidence of selection pressures toward semelparity because pollinators are selecting for taller inflorescences in three of the five species. There is phenological complementarity in this community because the flowering periods of the five species span the entire year, although there are some periods when pairs of species overlap. Additionally, we found evidence for character displacement in rosette sizes and separation of spatial and temporal resource use in pollinator composition among species.
Key Words: Agavaceae • Agave • bat pollination syndrome • coexistence • Metztitlán • phenological complementarity • pollination biology • semelparity
Agaves show one of the most spectacular floral displays in nature: most of them reproduce only once in their life, and they die after producing tremendous amounts of floral resources in a very large inflorescence. Even though agaves are considered typical examples of semelparity (Schaffer and Schaffer, 1977a
; Young, 1990
), there are some species that show an iteroparous life history (Gentry, 1982
). According to the reproductive effort model, semelparity may evolve when inflorescence height affects attractiveness to pollinators and they select for larger inflorescences (Schaffer and Schaffer, 1977a
; Aarssen, 1995
; Donnelly et al., 1998
); subsequently, plants may progressively increase their reproductive effort until the resources allocated to floral displays reach the maximum, causing the death of the plant (Schaffer and Rosenzweig, 1977
). Schaffer and Schaffer (1977a)
suggested that pollinators have reinforced this reproductive strategy in agaves by selecting only the largest floral displays when attracted to the taller plants, which as a result may receive greater pollen deposition or greater outcrossing rates (Aarssen, 1995
).
This singular and massive event of reproduction in agaves attracts a vast and diverse group of animals, ranging from true pollinators to nectar and pollen robbers or to animals that use inflorescences as shelter. Thus, agaves may be considered as keystone species in arid environments by offering abundant resources to pollinators, which in turn have a substantial effect on the reproductive success of the plant. Even though diurnal visitors can act as pollinators, agaves usually depend on nocturnal pollinators to set fruits (Arizaga et al., 2000b
; Eguiarte et al., 2000
; Slauson, 2001
; Molina-Freaner and Eguiarte, 2003
).
In general, floral traits of Agave suggest adaptation to bat pollination or "chiropterophily" (Faegri and van der Pijl, 1966
); they present robust and pale flowers (usually yellow or white-green), which produce protein-rich pollen and abundant diluted nectar mainly at night, and they smell like ripening fruit (Howell, 1972
; Eguiarte et al., 2000
; Slauson, 2000
).
The genus Agave is grouped according to the inflorescence type in two subgenera: Littaea species have rosettes with spicate or racemose inflorescences, while members of the subgenus Agave possess paniculate inflorescences (Gentry, 1982
). Classical pollination studies of Agave suggested that paniculate species were predominantly pollinated by bats (Howell, 1972
; Gentry, 1982
), while spicate agaves were mainly insect-pollinated (Schaffer and Schaffer, 1977b
). Interestingly, floral traits associated with insect pollination syndromes are indeed present in some spicate species; for example, floral tubes are smaller, nectar production is less abundant, sugar concentration is high, floral color is attractive to insects, and flowers are sweet-smelling (Slauson, 2000
, 2001
). Furthermore, there are no reports of bat visitation to Littaea species thus far.
However, the relationship between Agave and the pollinators is not so straightforward. Several studies have shown that chiropterophilous agaves (from the subgenus Agave), occupying habitats at the edge or outside the distribution of nectarivorous bats are pollinated by other animals such as insects or birds (Sutherland, 1987
; Kuban, 1989
; Molina-Freaner and Eguiarte, 2003
). Secondly, the relative importance of bat pollination varies geographically: paniculate agaves show more specialized pollination within the tropics and moderate generalization outside the tropics where they are pollinated by a variety of diurnal and nocturnal pollinators (Arizaga et al., 2000b
; Slauson, 2000
).
The highest diversity of Agave occurs in Central Mexico (García-Mendoza, 2002
; Tambutti, 2002
), but few studies on the pollination biology of Agave have been carried out in this area (Arizaga et al., 2000a
, b
; Ornelas et al., 2002
; Silva-Montellano and Eguiarte, 2003
). The present study was carried out in the Metztitlán Canyon, which is considered the area of highest Littaea species diversity in Mexico. Thus, this site was ideal to analyze basic aspects of the reproductive ecology of Agave Littaea species and to compare the patterns among species. We used this community of agaves as a model to test both the evolution of semelparity and the evolution of plant– pollinator interactions in the subgenus Littaea.
In particular, we were interested in comparing the characteristics of four semelparous and one iteroparous (A. striata) species. Another important goal of this study was to establish the degree of pollinator specialization that occurs in this group of coexisting Littaea species. Because of the geographic location of the Metztitlán Canyon in central Mexico, we can also test whether the geographic trend toward greater bat pollination in the tropics for the Agave subgenus is true even for the subgenus Littaea.
MATERIALS AND METHODS
Study area
The Reserva de la Biosfera Barranca de Metztitlán is a national protected area located in the state of Hidalgo in Central Mexico. It comprises a large depression called Metztitlán Canyon that is located between the parallels 98°23'00'' and 98°57'08'' W and 20°14'15'' and 20°45'26'' N. Elevation ranges from 1000 to 2000 m, and the total area of the reserve is 96 042 ha. It is an arid region, which represents the southernmost limit of the Chihuahuan desert in Mexico, and it is recognized as a high endemism zone, mainly for cacti and succulents (Sánchez-Mejorada, 1978
).
Study species
Several species of Agave have been reported in the region (Sánchez-Mejorada, 1978
; Galván and Hernández-Sandoval, 2002
). After extensive exploration of the area, we have identified 11 species of Agave, making the Barranca de Metztitlán the area of second highest diversity of species in the country, after the Tehuacán-Cuicatlán valley (also in Central Mexico), and the first place in number of Littaea species (eight species, representing ca. 11% of the total in the subgenus). We chose the five most abundant species and sampled the largest populations in order to observe a large number of inflorescences each season. The species selected and their groups according to Gentry (1982)
were: A. striata Zucc. (Group Striatae), the endemic A. celsii Hook. var. albicans (Jacobi) Gentry (Group Polycephalae), A. xylonacantha Salm., A. difformis Berger, and a new species Agave sp. (which is currently being described, A. García-Mendoza, Institute of Biology-National Autonomous University of Mexico, personal communication) (the last three belonging to the group Marginatae). From these five species only A. striata is iteroparous, and the rest are semelparous. One population per species was selected and their locations are shown in Table 1.
|
Rosette sizes and allometry
For each of the populations in Table 1, we marked at least 15 flowering individuals that were accessible to be measured and manipulated. In 2001, we worked with A. difformis, A. striata, and Agave sp.; in 2002 with A. xylonacantha, A. striata, and Agave sp.; and in 2003 with A. celsii albicans, A. xylonacantha, and A. difformis. The total number of individuals per species is shown in Table 2. We measured rosette width and height, height of the inflorescence (measured from the top of the rosette), and the number of active flowers (i.e., open flowers in staminate or pistillate phase that produce nectar).
|
Floral traits
Nectar production
To determine the amount and concentration of nectar, we monitored nectar production over a 16-h period for each species. Twenty-five flowers from five individuals (5 flowers/individual) were covered at noon with a soft mesh fabric bag to exclude pollinators from consuming the nectar. Beginning at 1600 hours we visited each flower every hour and recorded the onset of the nectar production until production finished. Once the flower began to produce nectar, we removed, measured, and discarded the nectar with a micropipette every 2 h and determined the concentration (as sucrose equivalents) with a handheld field refractometer. We called this measure "extracted-nectar production" or "E"). Additionally, for each individual, three flowers were bagged at noon and remained bagged overnight; the next morning, when nectar production had finished, we determined total volume and concentration of nectar (and we named this measure "non-extracted nectar production" or "N-E"). A one-way ANOVA was used to compare nectar volume and concentration among species.
Floral morphometry
In each population, we randomly selected three plants and 10 floral buds (that were expected to open the next day) in each plant. The next day, we measured pistil, stamen, anther, corolla, and ovary length and ovary and corolla diameter on each flower using calipers. Over the next 7 d, we followed the development of each flower and recorded the changes in the structures over time by repeating the measurements every day.
Natural fruit and seed production
We measured natural fruit set and seed set as estimators of reproductive success. In general, most plants produced fewer fruits than flowers in a given season, indicating that several factors may limit fruit formation (Sutherland, 1986
). On a sample of 15 individuals per species (that had flowered the preceding year), we measured inflorescence height and counted the number of mature fruits and floral scars. The sum of the latter two equals the total number of flowers in the inflorescence in the previous year, and the division of the number of mature fruits by this total is an estimate of fruit set. Seed set was calculated as the proportion of ovules that develop into viable seeds; this was calculated by counting the viable seeds (black ones) and the unviable seeds (white ones) in a total of 25 fruits (five fruits from each plant in five different plants per species).
Pollination experiments
To determine the relative importance of diurnal vs. nocturnal pollinators, we conducted exclusion experiments on five flowers per plant (5–10 plants/species depending on availability) by bagging flowers with immature pistils. We performed four pollination treatments: (1) control, in which flowers were available at all times to both nocturnal and diurnal pollinators, (2) flowers bagged prior to anthesis and covered throughout the experiment, (3) flowers available to nocturnal visitors all night but bagged before sunrise, and (4) flowers available to diurnal pollinators but bagged at night and unbagged before sunrise. In addition, to determine the relative success of self- vs. cross-pollination, we performed two additional treatments. (5) For self-pollination, we manually deposited pollen from other flowers of the same plant onto the stigma of the focal flower (geitonogamy) when the stigma was receptive (i.e., stigmatic exudates were present). (6) For cross-pollination,a mixture of pollen from four different plants at least 100 m from the experimental plant was deposited using the method in treatment (5).
Floral visitors
The number of hours that an inflorescence was observed varied according to accessibility of the inflorescence, the number of days of observation, and the number of observers. An average of 20 inflorescences per species per season was selected (Table 2). Each focal individual was observed at 10-min intervals, and all animal visits were recorded. Everyday we began the observations 2 h before sunset, continued during the night, and stopped when visitor activity decreased. We started again 30 min before sunrise and continued until pollinator activity decreased. Additionally, for each species we stayed one complete night (3–5 plants/night) to record the activity of visitors. Night observations were performed without artificial light, staying as close to the plant as possible to use natural sources (moonlight) for illumination and to prevent bat perturbation. We carefully observed the behavior of visitors, noting the presence of nectar and pollen robbers, which were subsequently excluded from the analyses. For all observations, we calculated the frequency of visitors per minute per flower and an average at every 30 min of the day (i.e., the average frequency at 0530 hours, 0600 hours, 0630 hours, and so on, but only for observation hours) and for each group of pollinators. Using this estimate, we determined the time when visitors were most frequent in order for comparison with peaks in nectar production. Finally, using the same estimate of frequency per minute per flower, we calculated the proportion of all visits for each group of visitors. All calculations included day and night observations.
Niche overlap
To estimate the degree of pollinator specificity per species, we calculated the proportional similarity (PS) index for pollinator assemblages between species pairs. This index takes into account both the identity of pollinators and their relative visitation rates. It is calculated as
). Additionally, we estimate the niche overlap index suggested by Pianka (1975)
by assuming that pollinators are a resource that can be shared between Agave species. The index is calculated as 
).
Statistical analyses
All statistical analyses were conducted using the JMP package (SAS Institute, 1997
). Percentages and proportions were arcsine transformed prior to analysis.
RESULTS
Floral phenology
The flowering periods of the Agave species in the Metztitlán Canyon are distributed throughout the entire year. There are three periods when two species overlap in flowering time (Fig. 1). Early in the spring, flowering occurs at the same time in A. celsii albicans and A. xylonacantha, and later A. celsii albicans and A. difformis flower in May and June. In the summer, we find a third temporal overlap between A. difformis and A. striata. Only Agave sp. flowers in winter. Nevertheless, there are still some periods when A. xylonacantha (February), A. celsii albicans (May), and A. striata (September) do not overlap with the others.
|
|
|
|
E = 11.57%; N-E = 13.66%), while A. striata (E = 18.25%; N-E = 19.32%) and A. xylonacantha (E = 17.74%; N-E = 24.67%) produce the more-concentrated nectar (no significant differences were found between them; t = 1.99, P = 0.04). Nectar of intermediate concentration was found in A. difformis (E = 14.29%; N-E = 13.1%) and A. celsii albicans (E = 13.77%; N-E = 14.76%) (for this pair t = 2.02, P = 0.006).
|
; Slauson, 2000
; Molina-Freaner and Eguiarte, 2003
): they are protandrous, and the staminate phase typically begins on the second day after anthesis followed by the pistillate phase. In Fig. 5 we show how the measurements change over time. The smallest rosettes have a shorter duration of flowering (i.e., A. celsii albicans), while the largest species have longer phases (i.e., Agave sp. with total duration of flowering of 7 d in contrast to 4 d for A. celsii albicans).
|
Natural fruit and seed production
Measures of fecundity are shown in Table 4. Among species, fruit set is not correlated with plant size, i.e., some large species have lower fruit set than smaller plants (Table 2). On the contrary, within species fruit set is related to inflorescence height; positive slopes are found for all the species, although for A. xylonacantha the slope is almost zero (Fig. 6). The slope was highest for A. celsii albicans, second highest for Agave sp., and the third steepest slope was observed for A. difformis. In these three species, pollinators prefer larger inflorescences according to Schaffer and Schaffer's model of pollinator selectivity (Schaffer and Schaffer, 1977a
). The same pattern is observed if we analyze the total number of viable seeds, which is also correlated to plant size: smaller inflorescences yielded fewer seeds and larger inflorescences yielded more seeds (Table 4). Moreover, we observed high variation in fruit set among species and years (Fig. 7); in our estimates of fruit set for 3 yr in A. difformis, the differences among years were significant (F = 11.89, df = 2, 44, P = 0.0001). For three species, data were obtained for two years: A. xylonacantha (t = 2.08, P = 0.049), A. striata (t = 2.71, P = 0.107), and Agave sp. (t = 2.38, P = 0.031), and significant differences between years were only observed in Agave sp. If we analyze the three years together, A. xylonacantha has the highest fruit set and A. celsii albicans and A. difformis the lowest. Agave sp. also showed high fruit set in 2001, although this value decreased in 2002. On the contrary, A. striata, which has the lowest bat visitation frequency, did not show a significantly lower value of fruit set.
|
|
|
|
|
|
We also calculated the periods in the observation hours (including day and night observations) when pollinator visitations rates were highest for each species (Table 5). For A. celsii albicans, A. xylonacantha, and A. difformis, the maximum peak of visitors occurred after the maximum production of nectar (Fig. 3), and bats were the main visitors with a high frequency of visits. The maximum peak in A. striata coincides with its peak of nectar production. Only in the case of Agave sp. did the maximum visitation frequency occur at the beginning of the production of nectar.
Niche overlap
Comparisons of the pollinator assemblages among species are shown in Table 6. Values of niche overlap scales from 0 (no overlap) to 1 (identical composition of pollinators); the similarity index also ranges from 0 to 1 in the same way and takes into account both the identity of pollinators and their relative visitation rates. The lowest niche overlap and similarity indexes were found in A. striata compared to the rest of the species (the iteroparous vs. the semelparous species), while the highest similarity and overlap are between Agave sp. and A. xylonacantha (both belonging to the group Marginatae). Lower values indicate less overlap in pollinator use and, therefore, a larger potential contribution of pollination system to reproductive isolation in these sympatric species.
|
Analyzing the reproductive ecology in Agave species in a community context provided new insights into the understanding of the Littaea subgenus evolution. First of all, there was phenological complementarity in the Metztitlán community because the flowering periods of the five species span the entire year. Nevertheless, we found temporal overlap in flowering times for some pairs of species, which may be explained by some other ecological reasons. We found evidence for character displacement in rosette sizes and separation of spatial and temporal resource use in pollinator composition among species. We also found evidence of selection pressures towards semelparity because pollinators are selecting for taller inflorescences in three of the five species. In relation to pollinator composition, we reported the first evidence of bat visitation in the subgenus Littaea and observed that bats (Leptonycteris curasoae, Choeronycteris mexicana, and Glossophaga sp.) were the0 primary pollinators in four species. Honeybees, hummingbirds, bumblebees and hawkmoths were also common visitors. We propose that the presence of diurnal pollinators may increase the reproductive success of these species when offering pollinators services additional to the co-adapted pollinator.
Phenological complementarity and character displacement
When analyzing assemblages of species with similar requirements (in this case, species from the same genus and subgenus), we may expect either resource competition or some mechanisms to avoid it, such as character displacement (Morin, 1999
; Tokeshi, 1999
). As niche theory predicts, the more closely related the species, the more niche axes will be expected to overlap (Schoener, 1989
; Silvertown et al., 2001
). In this work, we explored three niche axes related to the reproductive ecology: flowering timing, plant size, and pollinator use. We found some extent of overlap in the flowering periods for three species, but for the five species, the reproductive seasons are distributed throughout the entire year. In this community, differentiation in the flowering time and, hence, phenological complementarity could allow the coexistence of competitors (Rathcke, 1983
; Rathcke and Lacey, 1985
). Nevertheless, we are aware that our data are qualitative, and quantitative information would be needed to perform formal statistical analysis of the null model distributions of flowering times (Poole and Rathcke, 1979
; Rathcke, 1984
, 1988
). For the species overlapping at the same flowering time, we may find other explanations for their overlap. On one hand, species that flower on the same dates may produce nectar at different times in the night, minimizing possible competition for pollinators. Furthermore, for the pair A. celsii albicans and A. difformis, that flower at the same time during May and June, there is also spatial separation; A. celsii albicans grows in steeper hillsides with calcareous rocks as a substrate and A. difformis grows in drier zones. A similar situation exists between A. celsii albicans and A. xylonacantha: the latter grows in thin and calcareous soils in the lower parts of the canyon. The evolutionary relevance of the temporal separation in flowering time or phenological asynchrony is that the separation may act as a prezygotic barrier in sympatric species (Husband and Sabara, 2004
).
Owing to the importance of "body size" to reproduction in agaves, this niche axis may be as useful in this group of plants as it is in animals to study character displacement (Reich, 2001
). Allometric relationships have been considered a surrogate of the relationship between the organism and the environment, and plant size can be used as the orthogonal axis that explains this relationship (Niklas, 1994
). We have found differences in plant sizes and allometric relationships that may suggest character displacement.
The parameters evaluated so far illustrated the segregation of habitats and resource distribution that enable the coexistence of these species. Future research on patterns of spatial distribution and fine-scale use of resources, however, is still required for us to fully understand the coexistence of these species.
Floral traits and the bat pollination syndrome
Even though there were no previous reports of bats visiting Littaea species, and shifts towards other pollination syndromes were described before for the subgenus, the floral traits analyzed in this study seem to be in agreement with the bat pollination syndrome. For all the species, nectar production was nocturnal, the production peaks are between 2000 hours (in A. striata) and 0200 hours (in A. difformis and Agave sp.), and the sucrose equivalents also fall in the interval preferred by bats (10–19%) (Cruden et al., 1983
; Proctor et al., 1996
; Tschapka and Dressler, 2002
). In the case of A. striata, the time of nectar production overlaps with the visitation times of diurnal pollinators, suggesting a slight shift to diurnal pollination.
Additionally, our successful pollination experiments showed that nocturnal pollinators are responsible for the production of more viable fruits, but diurnal pollinators still have an effect when producing fruits and viable seeds in Agave sp. Nevertheless, we have limited results only for A. celsii albicans and A. striata and none for A. xylonacantha and A. difformis. We need further analyses to be able to demonstrate that other visitors, such as bees, hummingbirds, or hawkmoths, are active pollinators, although previous works have shown that these animals are indeed true pollinators in Agave Littaea species (Slauson, 2001
; Silva-Montellano and Eguiarte, 2003
).
Reproductive success and the evolution of life history strategies
We found a positive correlation in inflorescence height vs. fruit set in A. celsii albicans, A. difformis, and Agave sp., while there is no significant correlation in A. striata and A. xylonacantha. The rosettes of A. striata did not die after reproduction, and so they adjust to Schaffer and Schaffer's (1977a)
prediction that in iteroparous species pollinators do not select for larger inflorescences. But on the other hand, in A. xylonacantha there is no correlation, and an alternative explanation is that this species is more resource-limited, although detailed experiments are needed to corroborate this (see for instance Parra-Tabla et al., 1998
).
Comparing patterns of reproductive success may provide an important contribution to the understanding of semelparous and iteroparous reproductive strategies and size/fecundity patterns in closely related plants. In semelparous species, fruit set is usually positively correlated with inflorescence height, while in contrast, both pollinator preference and fruit set are independent of inflorescence size in congeneric iteroparous species (Young, 1990
). Reproductive output in semelparous organisms should be more sensitive to increases in resources (Young and Augspurger, 1991
), and there is evidence of an increasing reproductive output as a function of inflorescence size in several rosette-bearing species including Lobelia telekii (Schaffer and Schaffer, 1977a
, 1979
; Young, 1990
; Young and Augspurger, 1991
). Moreover, individuals of the semelparous Yucca whipplei var. whipplei produced larger inflorescences, with more flowers and more seed-bearing fruits than did individuals of the iteroparous Y. whipplei var. caespitosa (Huxman and Loik, 1997
).
In general terms, optimal foraging theory predicts that pollinators should associate a taller plant with increased reward, and an increase in pollinator visitation frequency to individuals has been generally assumed to increase the proportion of flowers visited (Schaffer and Schaffer, 1977a
, 1979
). Larger inflorescences are thus produced as a result of differential pollinator preference (Schaffer and Schaffer, 1979
). However, experimental studies in yuccas and agaves showed that on Yucca whipplei and A. chrysantha fruit set was usually limited by resources and not by pollinators (Acker, 1982
; Sutherland, 1982
). Resource limitation may explain the lack of correlation in fruit set and inflorescence size in A. xylonacantha, because we observed that this species tolerates the harshest and driest conditions in the canyon and reproduces in the drier season.
Visitation patterns and competition for pollinators
This is the first report of bats visiting Littaea species, and bats were the most important visitor in four of the five species. As stated before, the floral traits are clearly associated with a bat pollination syndrome, and visiting patterns confirm that bats are the co-adapted pollinators in this community.
We compared the pollinator fidelity using niche overlap and similarity indexes, both showing that the most similar composition exists between A. xylonacantha and Agave sp., which are separated temporally, the former flowering in spring and the latter in autumn. For species flowering at the same time, values are low in the pair A. celsii albicans vs. A. xylonacantha, although values are high in the pair A. celsii albicans and A. difformis, which may avoid possible competition among pollinators by growth in different microhabitats and perhaps by size differentiation of the inflorescence. The lowest overlap/ similarity values are for the pair A. striata and A. difformis, in which we found a slight shift to diurnal pollinators in the former.
A generalist or a specialist system?
Extreme specialization in pollination is rare and risky, and those systems in which a plant depends on only one pollinator are more susceptible to fail in reproduction due to the lack of pollinators (Howell and Roth, 1981
; Kearns et al., 1998
). The more generalist plants can resist the disappearance or the temporal absence of one of their pollinators (Gomez, 2002
).
In the case of Agave Littaea pollination, generalization may contribute to reproductive success with an additive effect supplementary to that produced by visits of the co-adapted pollinators; this response to new pollinators may be due to the high variability in bat visits. In this study we observed days and weeks when bats were absent in parts of the populations (data not shown), threatening the success of some individuals. Moreover, after nights when bats were absent, inflorescences offered a very abundant resource for diurnal pollinators. The shift to diurnal pollinators is notorious in A. striata, for which diurnal pollinators accounted for almost 60% of visitations, while in the rest of the species (except A. difformis) bats represent almost half of all visits. Maintenance of bat specialization in some species may have several explanations. First of all, new pollinators are not likely either to stop or to destroy the genetic integration that conforms a floral type once it has evolved in response to the co-adapted pollinator. Bond (1994)
suggested that many plants with specialized pollination systems have alternative reproductive mechanisms such as clonality, longevity, and facultative self-pollination, and all of these traits are present in the Agave genus.
The dichotomy Agave-Littaea and the geographic trend
Species belonging to the subgenus Littaea were previously thought to be insect- or bird-pollinated (Schaffer and Schaffer, 1977b
; Eguiarte et al., 2000
; Slauson, 2001
). In particular, our results contrast with the most recent study on floral visitors of another Agave Littaea: A. lechuguilla, which is a member of the group Marginatae (A. difformis and A. xylonacantha and Agave sp. belong to the same group). Floral morphology of A. lechuguilla is very similar to the species studied in this work, but no bat visits were reported (Silva-Montellano and Eguiarte, 2003
). Another issue in analyzing patterns of pollination biology is the notion that tropical pollination systems are more specialized than temperate systems (Feinsinger, 1983
; Johnson and Steiner, 2000
), although recent reviews of specialization across latitudinal gradients present conflicting conclusions (Olesen and Jordano, 2002
; Ollerton and Cranmer, 2002
). There is evidence of this geographical trend in some Agave (subgenus Agave) species; tropical species show more specialization than species outside the tropics (Arizaga et al., 2000a
, b
; Slauson, 2000
, 2001
). A similar pattern is observed in a plant system similar to that of Agave, the case of bat-pollinated columnar cacti (Valiente-Banuet et al., 1996
; Fleming et al., 1998
, 2001
; Fleming and Valiente-Banuet, 2002
). Valiente-Banuet and colleagues (1996)
suggested that generalized systems are favored in the northern distributional limits of columnar cacti because of the variation in the abundance of migratory nectar-feeding bats. When the effective pollinator of a plant is unreliable in space and time, natural selection should favor traits that increase its spectrum of pollinators or that favor a switch from an unreliable to a reliable pollinator (Fleming, 2000
). The species richness of nectar-feeding bats in Mexico reaches maximum values along the Pacific versant (the Balsas region) and decreases with latitude (Arita and Santos-del-Prado, 1999
; Rojas-Martinez et al., 1999
). In northern columnar cacti, bat unpredictability has been suggested as the major ecological force behind the evolution of generalized pollination systems (Valiente-Banuet et al., 1996
), and we might expect the same model in Agave. This may be the explanation for generalized systems found in the Littaea species reported previously because those studies were carried out in regions beyond the nectarivorous bat's distribution. The high visitation rates of bats in Metztitlán may be due to the high abundance of bats in Central Mexico, which in some cases (particularly for A. difformis, A. xylonacantha, and Agave sp.) may contribute to their reliability as pollinators. The results of this work confirm the co-adaptation (in a very broad sense) of bats and agaves, not only for the subgenus Agave, but for the subgenus Littaea as well.
FOOTNOTES
1 The authors thank the directives and staff of the Reserva de la Biosfera Barranca de Metztitlán who kindly provided facilities to develop this work; Mr. Erasto Badillo for invaluable hours of field assistance; Andrea González, Jaime Gasca, Ricardo Colín, Xitlali Aguirre, Roberto Trejo, Arturo Silva, Antonio Cruz, Eugenio Mancera, Yani Monges, Andrés Ocampo, Cristian Torres, Sergio Montiel, and Sara Good-Ávila, for enthusiastically looking for hours at inflorescences and helping with experiments; Dr. Francisco Molina-Freaner, Dr. Sara Good-Ávila, and Dr. Juan Nuñez-Farfán for reading the manuscript; Dr. Alberto Rojas-Martínez for helping with bat capture and identification; Dr. Abisaí García-Mendoza for identifying Agave species; M.Sc. Olivia Yañez-Ordoñez for identifying insects. M. R. received a Ph.D. scholarship from CONACyT and DGEP-UNAM and financial help from PAEP-UNAM. This work was supported by the CONACyT-SEMARNAT grant number COI-0246/A-1. 
2 Author for correspondence (e-mail: mrocha{at}miranda.ecologia.unam.mx
)
LITERATURE CITED
Aarssen L. W. 1995 Hypotheses for the evolution of apical dominance in plants: implications for the interpretation of overcompensation. Oikos 74: 149-156[CrossRef][ISI]
Acker C. L. 1982 Regulation of flower, fruit and seed production by a monocarpic perennial, Yucca whipplei. Journal of Ecology 70: 357-372[CrossRef]
Arita H. T. K. Santos-del-Prado 1999 Conservation biology of nectar-feeding bats in Mexico. Journal of Mammalogy 80: 31-41[CrossRef]
Arizaga S. E. Ezcurra E. Peters F. R. de Arellano E. Vega 2000a Pollination ecology of Agave macroacantha (Agavaceae) in a Mexican tropical desert. I. Floral biology and pollination mechanisms. American Journal of Botany 87: 1004-1010
Arizaga S. E. Ezcurra E. Peters F. R. de Arellano E. Vega 2000b Pollination ecology of Agave macroacantha (Agavaceae) in a Mexican tropical desert. II. The role of pollinators. American Journal of Botany 87: 1011-1017
Bond W. J. 1994 Do mutualisms matter? Assessing the impact of pollinator and disperser disruption on plant extinction. Philosophical Transactions of the Royal Society of London, Series B 344: 83-90[CrossRef]
Cruden R. W. S. M. Hermann S. Peterson 1983 Patterns of nectar production and plant–pollinator coevolution. In B. Bentley and T. Elias [eds.], The biology of nectaries, 80–125. Columbia University Press, New York, New York, USA
Donnelly S. E. C. J. Lortie L. W. Aarssen 1998 Pollination in Verbascum thapsus (Scrophulariaceae): the advantage of being tall. American Journal of Botany 85: 1618-1625
Eguiarte L. E. V. Souza A. Silva-Montellano 2000 Evolución de la familia Agavaceae: filogenia, biología reproductiva y genética de poblaciones. Boletin de la Sociedad Botanica de Mexico 66: 131-150
Faegri K. L. van der Pijl 1966 The principles of pollination ecology. Pergamon, Oxford, UK
Feinsinger P. 1983 Coevolution and pollination. In D. J. Futuyma and M. Slatkin [eds.], Coevolution, 282–331. Sinauer, Sunderland, Massachusetts, USA
Fleming T. H. 2000 Pollination of cacti in the Sonoran Desert. American Scientist 88: 432-439[CrossRef]
Fleming T. H. S. Maurice J. L. Hamrick 1998 Geographic variation in the breeding system and the evolutionary stability of trioecy in Pachycereus pringlei (Cactaceae). Evolutionary Ecology 12: 279-289
Fleming T. H. C. T. Sahley J. N. Holland J. D. Nason J. L. Hamrick 2001 Sonoran Desert columnar cacti and the evolution of generalized pollination systems. Ecological Monographs 71: 511-530
Fleming T. H. A. Valiente-Banuet 2002 Columnar cacti and their mutualists. The University of Arizona Press, Tucson, Arizona, USA
Galván R. L. Hernández-Sandoval 2002 Agave garciae-mendozae, a new species from central Mexico. Cactus and Succulent Journal 74: 188-191
García-Mendoza A. 2002 Distribution of Agave (Agavaceae) in Mexico. Cactus and Succulent Journal 74: 177-186
Gentry H. S. 1982 Agaves of continental North America. The University of Arizona Press, Tucson, Arizona, USA
Gomez J. M. 2002 Generalizations in the interactions between plants and pollinators. Revista Chilena de Historia Natural 75: 105-116[ISI]
Howell D. 1972 Physiological adaptations in the syndrome of chiropterophily with emphasis on the bat Leptonycteris lydekker. Ph.D. dissertation, University of Arizona, Tucson, Arizona, USA
Howell D. B. S. Roth 1981 Sexual reproduction in agaves: the benefits of bats, the cost of semelparous advertising. Ecology 62: 1-7[CrossRef][ISI]
Husband B. C. H. A. Sabara 2004 Reproductive isolation between autotetraploids and their diploid progenitors in fireweed, Chamerion angustifolium (Onagraceae). New Phytologist 161: 703-713[CrossRef][ISI]
Huxman T. E. M. E. Loik 1997 Reproductive patterns of two varieties of Yucca whipplei (Liliaceae) with different life histories. International Journal of Plant Sciences 158: 778-784[CrossRef][ISI]
Johnson S. D. K. E. Steiner 2000 Generalization versus specialization in plant pollination systems. Trends in Ecology and Evolution 15: 140-143
Kay K. M. D. W. Schemske 2003 Pollinator assemblages and visitation rates for 11 species of neotropical Costus (Costaceae). Biotropica 35: 198-207[ISI]
Kearns C. A. D. W. Inouye N. M. Waser 1998 Endangered mutualisms: the conservation of plant–animal interactions. Annual Review of Ecology and Systematics 29: 83-112[CrossRef][ISI]
Kuban J. F. 1989 The pollination biology of two populations of the big bang century plant, Agave havardiana Trel.: a multiple pollinator syndrome with floral specialization for vertebrate pollinators. Ph.D. dissertation, Syracuse University, Syracuse, New York, USA
Molina-Freaner F. L. E. Eguiarte 2003 The pollination biology of two paniculate agaves (Agavaceae) from northwestern Mexico: contrasting roles of bats as pollinators. American Journal of Botany 90: 1016-1024
Morin P. J. 1999 Community ecology. Blackwell Science, Malden, Massachusetts, USA
Niklas K. 1994 Plant allometry. University of Chicago Press, Chicago, Illinois, USA
Olesen J. M. P. Jordano 2002 Geographic patterns in plant–pollinator mutualistic networks. Ecology 83: 2416-2424[ISI]
Ollerton J. L. Cranmer 2002 Latitudinal trends in plant–pollinator interactions: are tropical plants more specialized?. Oikos 98: 340-345[CrossRef][ISI]
Ornelas J. F. M. Ordano A. Hernandez J. C. Lopez L. Mendoza Y. Perroni 2002 Nectar oasis produced by Agave marmorata Roezl. (Agavaceae) lead to spatial and temporal segregation among nectarivores in the Tehuacan Valley, Mexico. Journal of Arid Environments 52: 37-51
Parra-Tabla V. C. F. Vargas L. E. Eguiarte 1998 Is Echeveria gibbiflora (Crassulaceae) fecundity limited by pollen availability? An experimental study. Functional Ecology 12: 591-595[CrossRef][ISI]
Pianka E. R. 1975 Niche relations of desert lizards. In M. L. Cody and J. M. Diamond [eds.], Ecology and evolution of communities, 292–314. Belknap Press of Harvard University Press, Boston, Massachusetts, USA
Poole R. W. B. J. Rathcke 1979 Regulatory randomness, and aggregation in flowering phenologies. Science 203: 470-471
Proctor M. P. Yeo A. Lack 1996 The natural history of pollination. Timber Press, Portland, Oregon, USA
Rathcke B. J. 1983 Competition and facilitation among plants for pollination. In L. A. Real [ed.], Pollination biology, 305–329. Academic Press, New York, New York, USA
Rathcke B. J. 1984 Patterns of flowering phenologies testability and causal inference using a random model. In D. R. Strong, D. Simberloff, L. G. Abele, and A. B. Thistle [eds.], Ecological communities: conceptual issues and the evidence, 383–393. Princeton University Press, Princeton, New Jersey, USA
Rathcke B. J. 1988 Flowering phenologies in a shrub community competition and constraints. Journal of Ecology 76: 975-994[CrossRef]
Rathcke B. J. E. P. Lacey 1985 Phenological patterns of terrestrial plants. Annual Review of Ecology and Systematics 16: 179-214[CrossRef][ISI]
Reich P. B. 2001 Body size, geometry, longevity and metabolism: do plant leaves behave like animal bodies?. Trends in Ecology and Evolution 16: 674-680[CrossRef]
Rojas-Martinez A. A. Valiente-Banuet M. C. Arizmendi A. Alcantara-Eguren H. T. Arita 1999 Seasonal distribution of the long-nosed bat (Leptonycteris curasoae) in North America: does a generalized migration pattern really exist?. Journal of Biogeography 26: 1065-1077[CrossRef][ISI]
Sánchez-Mejorada H. 1978 Manual de campo de las cactáceas y suculentas de la Barranca de Metztitlán. Sociedad Mexicana de Cactología, Mexico City, Mexico
SAS Institute. 1997 JMP statistical software package, version 3.1. SAS Institute, Cary, North Carolina, USA
Schaffer W. M. M. L. Rosenzweig 1977 Selection for optimal life histories: II. Multiple equilibria and the evolution of alternative reproductive strategies. Ecology 58: 60-72[CrossRef][ISI]
Schaffer W. M. M. V. Schaffer 1977a The adaptive significance of variations in reproductive habit in the Agavaceae. In B. Stonehouse and C. M. Perrins [eds.], Evolutionary ecology, 261–276. Macmillan, London, UK
Schaffer W. M. M. V. Schaffer 1977b The reproductive biology of Agavaceae. I. Pollen and nectar production in four Arizona agaves. Southwestern Naturalist 22: 157-168
Schaffer W. M. M. V. Schaffer 1979 The adaptive significance of variations in reproductive habitat in the Agavaceae. II. Pollinator foraging behavior and selection for increased reproductive expediture. Ecology 60: 1051-1069[CrossRef][ISI]
Schoener T. W. 1989 The ecological niche. In J. M. Cherrett [ed.], Ecological concepts, 79–113. Blackwell, Oxford, UK
Silva-Montellano A. L. E. Eguiarte 2003 Geographic patterns in the reproductive ecology of Agave lechuguilla (Agavaceae) in the Chihuahuan desert. I. Floral characteristics, visitors, and fecundity. American Journal of Botany 90: 377-387
Silvertown J. M. Dodd D. Gowing 2001 Phylogeny and the niche structure of meadow plant communities. Journal of Ecology 89: 428-435[CrossRef]
Slauson L. A. 2000 Pollination biology of two chiropterophilous agaves in Arizona. American Journal of Botany 87: 825-836
Slauson L. A. 2001 Insights on the pollination biology of Agave (Agavaceae). Haseltonia 8: 10-23
Sutherland S. D. 1982 The pollination biology of paniculate agaves: documenting the importance of male fitness in plants. Ph.D. dissertation, University of Arizona, Tucson, Arizona, USA
Sutherland S. D. 1986 Patterns of fruit-set: what controls fruit-flower ratios in plants?. Evolution 40: 117-128[CrossRef][ISI]
Sutherland S. D. 1987 Why hermaphroditic plants produce many more flowers than fruits: experimental tests with Agave mckelveyana. Evolution 41: 750-759[CrossRef][ISI]
Tambutti M. 2002 Diversidad del género Agave en México: una síntesis para su conservación. Undergraduate dissertation, National Autonomous University of Mexico, Mexico City, Mexico
Tokeshi M. 1999 Species coexistence. Ecological and evolutionary perspectives. Blackwell Science, Oxford, UK
Tschapka M. S. Dressler 2002 Chiropterophily: on bat-flowers and flower bats. Curtis's Botanical Magazine (series 6) 19: 114-125
Valiente-Banuet A. M. C. Arizmendi A. Rojas-Martinez L. Domínguez-Canseco 1996 Ecological relationships between columnar cacti and nectar feeding bats in Mexico. Journal of Tropical Ecology 12: 103-119
Young T. P. 1990 Evolution of semelparity in Mount Kenya lobelias. Evolutionary Ecology 4: 157-171[CrossRef][ISI]
Young T. P. C. K. Augspurger 1991 Ecology and evolution of long-lived semelparous plants. Trends in Ecology and Evolution 6: 285-289
| ||||||||||||||||||
