| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ecology |
2Departamento de Ecología de los Recursos Naturales. Instituto de Ecología, Universidad Nacional Autónoma de México, Apartado Postal 27-3 (Xangari), 58089, Morelia, Michoacán, México; 3Instituto Nacional de Ecología, Av. Revolución 1425, Col. Campestre, C.P. 01040, México, D.F. Mexico
Received for publication July 26, 2001. Accepted for publication October 30, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Seedlings showed low survival rates. Cohorts placed outside nurse plants died in less than 1 yr, while 1–10% of cohorts under nurse plants survived for more than 2 yr. Herbivores negatively affected seedling survival in non-nursed plots. In rainy years, survival rates increased. Bulbils showed higher survival rates than seedlings.
The excavation of rosettes showed that most are derived from vegetative shoots, as indicated by remains of rhizomes in their base. Most rosettes had ground-level vegetative offspring totaling almost three shoots per rosette.
In A. macroacantha, the establishment of seedlings and bulbils is a rare event that possibly only occurs under nurse plants in rainy years, while ground-level cloning is highly effective as a propagation mechanism. These results are consistent with the aggregated spatial pattern of the species.
Key Words: Agavaceae • Agave macroacantha • bulbil • seedling establishment • sexual reproduction • shoot • survivorship curves • vegetative propagation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Both processes may give rise to new individuals, but in reproduction the new plant is originated from a zygote, while in vegetative propagation the new plant is originated from a meristem (Harper, 1977
). Plants showing both processes, normally known as clonal plants, are very common in nature (Abrahamson, 1980
; Cook, 1985
).
From a morphological point of view, the development of a plant may be interpreted as a process of modular construction in which growth is determined by the addition of new morphological units or modules. In clonal plants, each independent unit of modular growth is called a ramet and corresponds to clonally originated individuals. Similarly, a genetic individual, or genet, is defined as the array of ramets derived from a common zygote, i.e., the set of clonal individuals that originated from a common parental stock and thus have the same genotype (Harper, 1977, 1985
). In clonal plants, a single genet may occupy large areas extending hundreds of meters (Cook, 1985
).
The propagules and diaspores (both seeds and vegetative offspring) produced by clonal plants seem to play different roles in the life cycle of the plant. In highly unpredictable or extreme environments, clonal multiplication operates as a safe strategy to perpetuate the genotype in a given habitat, while sexual reproduction seems to constitute a strategy that involves greater risks and mostly serves as a mechanism to explore new environments (Abrahamson, 1980
; Silander, 1985
; Eriksson, 1989, 1992
). Although normally the successful establishment of sexually derived seedlings involves a very low proportion of all the seeds produced, it nonetheless plays an important role in genetic recombination and in maintaining the genetic variation of the population (Silander, 1985
; Ellstrand and Roose, 1987
). Furthermore, the clonal propagation of a genet may also increase the chance of successful sexual reproduction of a single genet, which may reproduce repeatedly, at different times and in different places, through its constituent ramets (Cook, 1985
).
The amount of sexual diaspores produced by a clonal plant is regulated by environmental factors such as the availability of resources and the trade-off between sexual reproduction and vegetative growth (Cook, 1985
; Silander, 1985
; Eriksson, 1992
). The theory of resource allocation predicts that the amount of resources available for a plant is limited and that the onset of reproduction necessarily implies a competitive reduction in resources available for vegetative growth (Watson, 1984
). In some plants, this trade-off between clonal growth and reproduction is related to direct competition for the same reserves in such a way that resource allocation for either one of these two processes develops at the expense of the other (Harper, 1977
; Willson, 1983
; Arizaga and Ezcurra, 1995
).
Although the dispersal ability of sexual diaspores is an extremely important element in the demography of plants (Harper, 1977
; Grime, 1979
), the probability of successful establishment of seedlings is often extremely low and frequently constrained to narrow windows of opportunity in space or time (Thomas and Dale, 1975
; Cook, 1985
). In contrast, clonal propagules have the advantage of being linked to the parental plant during their early stages through their vascular systems, a fact that allows them a much higher probability of survival in extreme environments (Harper, 1985
; Pitelka and Ashmun, 1985
).
Within the group of succulent rosette plants that form the genus Agave, numerous species present both propagation mechanisms, i.e., sexual reproduction and cloning. The latter can be achieved in different species through the development of clonal shoots in different parts of the rosette, including: (1) aerial bulbils produced in the flowering scape, (2) lateral shoots produced in leaf axils, (3) short basal shoots produced below the rosette, or (4) long rhizomatous shoots that emerge from the ground at a distance from the parental plant (Gentry, 1972, 1982
; Fig. 1). As is the case with other xerophilous succulents such as cacti (Nobel, 1984
; Elheringer, 1985
; Valiente-Banuet and Ezcurra, 1991
), the establishment and seedling stages represent the most vulnerable phases of the Agave life cycle. Seedlings have a limited amount of reserves, a low capacity to absorb water, and are exposed to the high variations in surface ground temperatures that characterize arid environments. Thus, their successful establishment is usually confined to rainy seasons (Turner et al., 1966
; Nobel, 1977, 1984
; Jordan and Nobel, 1979
; Raphael and Nobel, 1986
). In terms of survival probabilities, cloning seems to be a more efficient mechanism of propagation (Gentry, 1972, 1982
).
|
; Álvarez, 1989
; García-Mendoza, 1995
). They are native to North America, Central America, the Caribbean, and South America and range in latitude from 40° N to 6° N. The highest species diversity of the genus is found in Mexico (Gómez-Pompa, 1963
; Álvarez, 1989
; Rzedowski, 1991
; García-Mendoza, 1995
). Agaves have been used by humans since pre-Columbian times and are still widely used as sources of food, beverages, fibers, construction materials, and natural medicines (Gonçalves y Lima, 1956
; Callen, 1965
; Felger and Moser, 1970
; Gentry, 1982
). Although most of the economic uses of agaves depend on plants from wild populations, some species are commercially cultivated. Although many Agave species are characteristic elements of arid lands, little information exists on the production of different types of propagules and diaspores in natural conditions. This is a study on the reproductive and propagation strategies of A. macroacantha, a clonal species with monocarpic rosettes (i.e., semelparous ramets).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Mean annual rainfall is 380 mm, mean annual temperature is 21.4°C, and annual evaporative demand is on the order of 2000 mm (García, 1987
; Peters, 1993
).
|
), and the dominant species is the giant columnar cactus Neobuxbaumia tetetzo, growing in association with other cacti, succulent rosettes, xerophytic shrubs, and annuals (Zavala-Hurtado, 1982
). Zapotitlán Salinas is one of the many small valleys that converge into the large Tehuacán Valley, which in turn forms part of the Tehuacán-Cuicatlán Biosphere Reserve in the border region between Oaxaca and Puebla. The ranges of the Eastern Sierra Madre, known as Sierra de Zongolica and Sierra de Juárez, create a rain shadow that constitutes the main cause of the aridity of the Tehuacán Valley (Smith, 1965
; Villaseñor, Dávila, and Chiang, 1990
). In contrast with other arid zones, the short rainy season in the Tehuacán Valley is relatively predictable. The rainy period normally starts in late May and continues until late September. Hurricanes may extend the rains into October and, rarely, into November. In August, a midsummer drought frequently occurs (Valiente, 1991
).
The soils of the Tehuacán Valley are shallow and rocky, derived from calcareous sedimentary rocks, and are frequently halomorphic, with varying degrees of salinity and alkalinity. In some hills, the soils form typical limestone rendzinas. In general, soils have loamy textures, light colors, alkaline pH, low organic matter, and high carbonate contents (Aguilera, 1970
).
The economy of Zapotitlán Salinas is based on extracting and processing onyx, extracting salt from briny aquifers, and grazing the natural scrub with goats. Agriculture is limited to the valley bottoms. The collection of wild fruits and wild plants is common, but mostly oriented towards self-supply and not aimed at external markets.
Study species
Agave macroacantha is a midsize paniculate agave, endemic to the Tehuacán Valley (Gentry, 1982
; García-Mendoza, 1995
). It is locally known as "maguey espadita" (little sword), and its rosettes are occasionally planted to delimit land plots with hedge-like rows of different plants. It grows mostly on low-grade slopes forming small colonial aggregations, possibly as a result of its cloning ability. It flowers in early summer, between April and June, coinciding with the end of the dry season. As in all other paniculate agaves, the flowers in the scape of A. macroacantha are developed in lateral "branches," which are lateral umbellate clusters. For the sake of brevity, we shall refer to these structures as "umbels." The flowers are protandrous and the crossing system is mainly allogamous. Its reproductive success is extremely dependent on the presence of nocturnal pollinators, chiefly nectar-feeding bats and, to a lesser extent, moths (Arizaga and Ezcurra, 1995
; Arizaga et al., 2000a, b
). The fertile capsules are formed in late summer, between June and October (Arizaga et al., 2000a)
. Most of the seedlings and many solitary rosettes are observed growing in association with higher shrubs, a fact that suggests that these agaves may profit in the early stages of their life cycle from the shade of nurse plants. Another interesting aspect of the ecology of A. macroacantha is its capacity to produce a large amount of bulbils in the scape if pollinators fail to arrive or if grazing damages the flowering stem (Arizaga and Ezcurra, 1995
).
Evaluation of rosette morphology and spatial distribution
During the reproductive period (April to November) of 1992, we selected 188 clumps or colonies of A. macroacantha localized within the study area, all located in a gentle (14° or 23%) southwestern slope. A clump is normally formed by a group of rosettes growing less than 1 m apart. Previous excavations of plants showed that rhizomatous shoots are normally generated less than 1 m away from the parental rosette. Hence, we assumed that rosette clumps, which are clearly distinguishable in the field, are formed by a population of ramets belonging to a single genet. In each clump, we counted the number of rosettes (ramets) and measured the height and diameter of each one. In reproductive rosettes, we also measured the size of the scape.
Quantification of propagules
Seeds (reproductive diaspores)
1. Seed production—Between April and November 1992, we counted the number of floral buds and the number of fruits that developed from them (i.e., the fruit set) in all the flowering rosettes located within the study area, totaling 53 scapes. During fruiting season (July to November) we took a random sample of 219 capsules that had not yet reached dehiscence from 38 randomly chosen individuals (we collected 5–6 unopened capsules per scape). In each capsule we counted the number of fertile seeds, which are black in color, and the number of aborted seeds, which are light brown. The estimate of total ovules is determined by summing the numbers of the fertile and aborted seeds. With both the fruit- and seed-set data, we estimated the mean fecundity of the ramets as the product of the mean number of capsules formed in a scape and the mean number of viable seeds formed within each capsule.
2. Seed germination—Once the fruits mature, the flat seeds are dispersed by wind. Some fall directly on the ground, where they either germinate rapidly or are subject to predation, while others fall in rock cracks, which act as safe sites where radiation conditions are less intense and where escape from predation is more likely. Additionally, a significant proportion of the seeds remain in the dry capsules up in the scape. This aerial seed bank is released gradually during the next 1–2 yr by the slow, progressive opening of the capsules. Taking into account the variation in seed age, we quantified the germination capacity of seeds by applying the following treatments: (1) Fifteen lots of 100 seeds each (N = 1500) were germinated in 1% agar plates in the laboratory at 23°C and under 12 h light, 12 h darkness conditions. These seeds were produced in 1992 and were approximately 1 mo old when germinated. (2) Five lots of 100, 1-mo-old seeds (N = 500) were germinated as in (1), but under total darkness by covering the tray with aluminum foil (24 h darkness at 23°C). (3) Eight lots of 100 seeds (N = 800) collected in 1991, i.e., older than 20 mo and stored under laboratory conditions, were germinated under the same conditions of treatment (1).
Vegetative aerial propagules
1. Bulbil production—To evaluate the effect of goat grazing on bulbil production, we quantified bulbil production in two areas with varying degrees of protection: (1) In the botanical reserve (i.e., the "protected" treatment), we evaluated herbivore damage during four reproductive seasons (1992–1994 and 1996) and counted bulbil production in all flowering rosettes found within the reserve, totaling 254 rosettes. The reserve is intended to be protected from grazing, but goats often break in and so only partial exclusion from herbivores was accomplished in this treatment. (2) For the "unprotected" treatment, we evaluated herbivore damage in a grazing commons, 1.3 km south of the reserve, that belongs to the community of Zapotitlán. During two reproductive seasons (1995–1996), we counted bulbil production in all flowering rosettes found in the grazing area, totaling 129 rosettes.
2. Fate of bulbils—For the two areas described above (protected and unprotected for the period between 1992 and 1996), we marked at the end of each reproductive period all the bulbils produced in 86 bulbiferous rosettes (48 rosettes in the protected area and 38 in the unprotected one), totaling 518 bulbils. At monthly intervals during the 5-yr period we returned to both sites and followed the fate of the recorded bulbils, registering the mortality of bulbils in the scape, the abscission of bulbils from the scape, and their final fate on the ground.
3. Axillary shoot production—We conducted periodic observations in the same rosettes (N = 383) in the protected and unprotected areas to determine the number of lateral shoots produced above ground, in the axils of the leaves.
4. Basal and rhizomatous shoots (vegetative ground-level propagules)—We excavated ten randomly selected clumps (genets) by carefully removing the top layers of the soil (see Böhm, 1979
) until the underground stems, rhizomes, and lateral root system were exposed. In each rosette (ramet) we registered: (a) the diameter and height of the rosette, (b) the vertical buried depth of the main stem, (c) the number of basal shoots and their respective size, and (d) the number and size of rhizomatous shoots.
Propagule establishment dynamics
Seedling establishment
To evaluate the establishment dynamics of A. macroacantha seedlings, we performed a factorial experiment with two levels of radiation (open space and under a nurse plant), two levels of predation exposure (exposed to predators and excluded), two planting seasons (dry and rainy season), and three replicates within each treatment. Three cohorts of nursed seedlings from seeds collected in 1991 and germinated in 1% agar gel at 25°C were planted under one of three randomly selected Mimosa luisana shrubs with canopy cover of about 2 m2. Flores-Martínez (1994)
estimated a minimum shrub cover of 0.235 m² for the nurse plant to effectively have some protective effect on cactus establishment. The open space cohorts were planted 30 cm north of three randomly selected A. macrocantha rosettes, with no other canopy shading them. Each cohort was planted in a 30 x 30 cm plot. Vertebrates and larger insects were excluded by covering the plots with 12-mm wire mesh; smaller insects were excluded by applying a contact insecticide (Clordane) in the perimeter of the plot each week for the first 70 d. The seedlings were planted 1 wk after germinating. Planting was done in the following two contrasting periods: (1) Dry season establishment was simulated by planting the seedlings in December 1991. Seedlings were hardened off after planting by watering them daily for the first 2 wk and every 4 d thereafter for three more occasions. The survivorship analysis was done on the seedlings that survived the acclimatization period. (2) Rainy season establishment dynamics were evaluated by planting the seedlings in early June 1992, at the beginning of the rainy season. Because the soil was wetter at this time, the acclimatization of the seedlings was done by watering only for the first week.
We planted 100 seedlings in each plot. Because of the high seedling mortality during acclimatization, the plots in the dry period treatment were uniformly thinned to an initial number of 75 seedlings at the beginning of the survival counts, while the wet season treatment retained the original density of 100 seedlings per plot when counts started. Surviving seedlings in both treatments were then periodically counted after 1 wk, thereafter at increasing intervals of 2, 5, and 10 wk. Both experiments were ended in April 1994. To characterize the environmental conditions of the site, we placed two rain gauges (Peters, 1993
) and a maximum-minimum thermometer on the southwest-facing slope where the experiments were done.
Bulbil establishment
To evaluate the establishment dynamics of A. macroacantha bulbils, we used 102 bulbils collected in 1991 from flowering scapes. Due to the low number of bulbils, this experiment was reduced to analyzing the effect of bulbil size as a factor leading to successful establishment. We divided the bulbils into three categories: small (<4.0 cm height, N = 48), intermediate (4.0–5.9 cm, N = 30), and large bulbils (
6 cm, N = 24). The bulbils were planted under three nurse shrubs (Acacia coulteri) of similar size, excluding in all cases larger predators with 12-mm wire mesh. At the time of planting (November 1991), the bulbils had not yet developed roots. The bulbils were planted in a 40 x 40 cm plot, 30 cm north of their nurse shrub. Rows of bulbils of different sizes were mixed within the plot in order to avoid the systematic placement of a given bulbil size in a fixed position under the canopies. Survival counts started in mid-November 1991 and were performed at increasing intervals of 1, 2, 5, and 10 wk for 30 mo.
Analysis of survivorship curves
To evaluate the characteristics of the survivorship process we used a nonlinear model developed for this study. We started with the assumption that the mortality of a cohort, which defines the survivorship curve, may belong to one of three basic types: (1) type I mortality, in which the initial mortality rate (k0) increases with time (i.e., the risk of mortality increases with the age of the cohort); (2) type II, in which the mortality rate is constant (the mortality risk is the same for any age); and (3) type III, in which the risk of mortality decreases with time (i.e., the initial mortality rate k0 decreases with time). Mathematically, this can be written as a system of two simultaneous differential equations: 
n/t = –kn, and k/t = bk, where t = time; n is the number of survivors, k is the instantaneous mortality rate, and b is a shape parameter that describes the type of survivorship process. If b > 0, then k will increase with time, and the resulting curve will be of type I. If, on the other hand, b < 0, then k will decrease with time, and the resulting curve will be of type III. Finally, if b = 0, then k will remain constant with time, and the survival dynamics will be of type II.
Integrating the two previous equations, we get the nonlinear survivorship function: nt = n0 exp[–k0/b (exp(bt) – 1)], where nt is the number of survivors at time t; n0 is the initial size of the cohort (at time t = 0); k0 is the initial mortality rate (at t = 0); and b is a parameter, defined above, that describes how k0 changes with time.
This model was fitted to both the seedling and the bulbil survival counts by means of nonlinear regression. Once the parameters of the function (k0 and b) were estimated, we compared them by means of t tests. These results were corroborated with the test of Peto and Peto for survivorship curves (Pyke and Thompson, 1986
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
2 = 435.0; df = 12; P << 0.0001) from the predictions of the Poisson distribution (Fig. 3). Clumps with only one rosette, and also clumps with more than 12 rosettes, are significantly more abundant than could be expected by chance alone, a fact that highlights the highly aggregated nature of ramet development.
|
2. Seed germination—Four days after initiation of the experiment, recently collected seeds germinating under alternating light/darkness conditions had the highest germination rate (76%; ±2%); seeds germinated in total darkness had 70% (±2%) germination, and the seeds that had been stored for 20 mo had 59% (±2%) germination. These differences are statistically significant (2 = 18.4, df = 2, P = 0.0001) and suggest that older seeds lose some of their germination ability. Biologically, however, the differences seem to be relatively minor: 8 d after initiation of the experiment, the stored seeds had increased their germination rate to about 85%, a value not significantly different from that of the recently collected seeds.
Bulbils
1. Bulbil production—Highly significant differences (2 = 57.4, df = 5, P < 0.0001) were found between the unprotected and the protected areas in terms of the number of grazed scapes. Only 38% of all scapes (N = 254) were grazed in the reserve, whereas 85% were grazed in the unprotected area (N = 129). Although these differences were obtained by comparing different years for both sites, and hence represent an invalid replication, a similar trend was observed when bulbil production was measured at both sites simultaneously in 1996. Additionally, significant variation in the proportion of grazed scapes (2 = 32.4, df = 3, P < 0.0001) was found within the reserve, ranging from 5% damage in 1996 (a year in which protection was successfully enforced) to 64% damage in 1994 (a year in which enforcement was very low).
In both sites, the grazing commons and the botanical reserve, a similar proportion (37% for the pooled data set) of the grazed scapes formed bulbils, and 26% of the bulbiferous scapes shed some of their propagules to the ground. In contrast, only 5% (±1%) of scapes that were not grazed (N = 156 in the reserve and N = 19 in the unprotected area) formed bulbils, possibly as a result of deficient pollination (Arizaga and Ezcurra, 1995
). The proportion of ungrazed scapes forming bulbils was similar in the reserve and the grazing unprotected area. In both sites, 26% of all scapes that formed bulbils shed some of these to the ground (±1 SE) (3.9 ± 1.0 bulbils per plant were shed). No significant differences were found between years or between sites.
2. Fate of bulbils—Under field conditions we observed that bulbils may have the following final fates: (1) Some bulbils are predated in the scape by insects (mostly by caterpillars and secondarily by grasshoppers) and to a lesser degree by goats and cattle. (2) A certain proportion, which varies significantly according to the site and the year (2 = 63.8, df = 4, P < 0.0001), are shed and fall to the ground. In 1992, 1994, and 1996, 32%, 34%, and 0%, respectively, of all bulbils were shed in the reserve and only 4% were shed in the grazed commons in 1995 and 1996. Shedding may be related to the density of neighboring canopies that brush against the drying scape. The plants that do shed bulbils dropped a mean of 3.9 bulbils. The shedding scapes in the reserve dropped a mean of 5.6 (±0.8) bulbils, while the plants in the grazing commons shed 2.2 (±0.8) bulbils. (3) Of those bulbils that fall to the ground, a small proportion may eventually become established. However, this seems to be a rare event in the arid environments of Zapotitlán. In our 5 yr of observations, we only counted four successfully established bulbils from a total of 78 bulbils that were followed after shedding. These four bulbils were shed in 1992, a relatively wet year (Fig. 4), and fell under the nurse plants M. luisana and Jatropha neopauciflora. In all other years, bulbils that were shed failed to take root on the ground. (4) Sixty-six to 96% of the annual crop of bulbils that are produced remain in the scapes and slowly dry up with them, remaining viable for a year or more until they die (Fig. 1).
|
Basal and rhizomatous shoots
Agave colonies (N = 10) contained 1–7 well-defined ramets each, totaling 44 adult rosettes with a mean diameter of 28.0 (±1.5) cm, a mean height of 22.9 (±2.7) cm, and a vertical underground depth of the main stem of 6.6 (±0.5) cm. We found 80 live rhizomes of variable length, the longest one being 78 cm, mostly unbranched, with a mean diameter of 2.9 cm (±0.6; N = 64). Eighteen of the 80 rhizomes (23%) were starting to develop small terminal rosettes (i.e., they were transitioning from undetermined to determined growth; Bell, 1991
), while the rest were still in the exploratory, undetermined growth phase. We also counted 47 basal shoots, with a mean diameter of 8.9 (±2.4) cm and a mean height of 5.8 (±1.0) cm. In short, we found a mean of 2.9 offspring ground-level shoots for each adult rosette, including 1.1 basal shoots, 0.4 determined (i.e., established) rhizomes, and 1.4 undetermined (i.e., exploratory) rhizomes (Fig. 1).
Propagule establishment dynamics
Seedling establishment
1. Dry season establishment dynamics (1991 cohorts)—Seedlings placed in open, non-nursed plots showed rapid mortality, and none survived after 245 d (Fig. 5a, b). Initial mortality was higher in the plots that were exposed to predators, and in both treatments (exposed to predators and excluding predators) the shape of the curve was type I (i.e., the initial mortality tended to accelerate with time; see parameters in Table 1). In contrast, the initial mortalities in seedlings planted under the shade of the nurse plant M. luisana tended to decrease with time resulting in a type III curve (Fig. 5c, d and Table 1). No significant differences were found between shaded plots exposed to predators and those excluding them. At the end of the 30-mo experiment, 2.4% (±1.1) of the initial seedlings had successfully established and survived in these plots.
|
|
|
Bulbil establishment
No significant differences were found in bulbil survivorship between the three size classes, either with the survivorship curve analysis or with Peto and Peto's test. Hence, we pooled the different size-specific bulbil counts under each nurse tree into a single data series and fitted one survivorship curve for all bulbil sizes. Once established, bulbils had lower initial mortality than seedlings, and these initial mortalities tended to decrease rapidly with time, in a markedly concave type III curve (Fig. 7 and Table 1). Thirty months after being placed under a nurse tree, 56% (±4.04) of the initial bulbils had successfully rooted and survived in the plots.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Herrera, 1987
). On average, a rosette of A. macroacantha produces some 2800 seeds (24 fertile capsules per scape x 117 fertile seeds per capsule). This is only 14% of the approximately 20 000 seeds that could be potentially produced if the average 170 flowers present in a scape developed. This fecundity seems low when compared to other agaves (Table 2), although many of these other estimates of seed set lack an explicit explanation of the methodology used and may possibly be only the result of extrapolations. Three main factors seem to playing an important role in decreasing the fecundity of A. macroacantha: (1) Predation of reproductive structures by insects and principally by goats was an important factor throughout our study. Goats consumed 85% of all scapes in the grazed unprotected area and predated 38% of all scapes in the protected reserve. Other authors (e.g., Smith, 1965
; Meyrán, 1980
; Gentry, 1982
) have long reported the damaging effect of goats on these natural systems. We have also identified evident goat damage in many other agave species in Tehuacán, such as A. kerchovei, A. marmorata, A. peacockii, A. potatorum, A. salmiana var. ferox, A. stricta, and A. triangularis. (2) The local winds may also have a strong effect on flower and bulbil retention. This seems to be especially important in dense scrubs, where under windy conditions the spiny canopy of neighboring plants often detaches the reproductive structures from the scape. (3) Lastly, A. macroacantha is pollinized chiefly by bats and secondarily by moths, and its fertilization is strongly pollinator limited (Arizaga et al., 2000b)
. The failure of bat pollinators to arrive in sufficient numbers in some years may decrease significantly the number of fertile capsules produced in a given scape (Arizaga and Ezcurra, 1995
).
|
). Freeman (1973, 1975)
, however, has reported germination values for A. lechuguilla and A. parryi that are similar to ours, while Jordan and Nobel (1979)
found 92% germination rate in recently collected A. deserti seeds.
Seedling establishment
All seedlings planted in open, non-nursed space died within 3–8 mo. Seedlings planted under nurse shrubs survived for 2.5 yr, but the demographic toll was considerable; only some 2.4% of the initial cohorts survived. Similar high mortalities have been reported for the Sonoran Desert where planted seedlings died within the next 63 d (Nobel, 1977
; Jordan and Nobel, 1979
).
Seed predation, principally by ants, showed detectable but quantitatively low effects in the 1991 cohort and in open spaces. Survival behavior was very similar in the 1991 and the 1992 cohorts. In all cases, the main factor controlling seedling establishment was the protective shade of a nurse plant. These results suggest that A. macroacantha depends critically on nurse plants for the survival of seedlings during the establishment phase in the hot tropical desert of Zapotitlán, in the same manner as cacti and other succulents depend on nurse plants in most deserts (Turner et al., 1966
; Steenbergh and Lowe, 1969
; Jordan and Nobel, 1979
; Valiente-Banuet and Ezcurra, 1991
).
There was a surprising lack of differences between the 1991 "dry" cohort and the 1992 "rainy" cohort. In reality, the 1991 cohort started during an abnormally wet winter and spring, while the 1992 cohort had to start its survival process during an exceptionally dry summer and fall (Fig. 4). It is likely that in a more normal year the cohorts planted in midsummer would survive better, but this remains to be tested. In any case, our results highlight the potentially critical importance of atmospheric anomalies in the establishment of this desert species and possibly of many others.
In follow-up sampling done in 1998, we found that the established seedlings of the 1991 cohort had decreased from eight to four, while the 1992 cohort still maintained the four survivors it had in 1994. This new value brings the 7-yr survival probability of nursed seedlings to 1.5%. This value is possibly overestimated as the experiment started with germinated seedlings that were watered for a few days to allow them an initial advantage. It is difficult to estimate at this point what the probability is that the seeds in a scape will produce successful offspring. Our establishment experiment showed that none of the seeds that fell in the open survived and that only a fraction of less than 1.5% of seeds that fell under a nurse plant survived. However, no data is available regarding the number of seeds predated, the proportion of seeds that fall under nurse plants, the chances of successful germination in the field, or the probability of establishment if no acclimatization is given to the seedlings.
In any case, these very low survival numbers suggest that seedling establishment is an extremely rare event, a fact that seems to be confirmed by the very small number of seedlings that have been observed in the field. Between 1991 and 1998, we failed to detect any new seedlings that established naturally in the study site. Sampling done in 1997 in the reserve revealed the presence of 16 isolated juvenile plants with some evidence of having established from seed: they had a rounded underground stem quite different from the funnel-like base that results from rhizomes and basal shoots. Furthermore, these plants were grown in isolation, without any other colony nearby from which they could have been derived as bulbils. These juveniles showed a mean diameter (±1 SE) of 7.9 (±0.8) cm, a mean height of 5.3 (±0.6) cm, and 8.4 (±1.9) photosynthetic leaves. Based on a size-age model we developed (see Arizaga, 1999
), we estimated the age of these plants to be about 14 yr. This would set the age of establishment around 1983, a year that brought an unusually high rainfall to central Mexico.
Various authors (Gentry, 1972
; Nobel, 1977
; Raphael and Nobel, 1986
) have also pointed out that recruitment from seed in agaves is an extremely rare event. Additionally, it has been reported that atmospheric anomalies may play a role in releasing harsh environmental restrictions: Jordan and Nobel (1979)
found a strong correlation between unusually rainy years and the establishment of A. deserti seedlings under nurse plants.
The importance of vegetative shoots
Bulbils
A direct consequence of the abatement of fecundity or reproductive failure in A. macroacantha is the formation of bulbils from bracteal meristems in the main scape, lateral umbels, or even in the floral pedicels (Arizaga and Ezcurra, 1995
). As was the case with seeds, most authors report a much higher productivity of bulbils for other agave species (Table 2), but the exact methodology of these calculations is unclear. Bulbils are commonly used to clonally propagate commercial species such as A. fourcroydes, A. sisalana, A. angustifolia, and many others (Benítez, 1965
; Gentry, 1982
; Álvarez, 1987
). Although it is clear that reproductive failure in A. macroacantha induces bulbil formation, the effectiveness of these propagules as a means of cloning is less certain. Only 37% of damaged plants produced bulbils, and of these, only 26% shed one or more bulbils to the ground. Plants that shed bulbils dropped a mean of 3.9 bulbils and only 5% of these eventually rooted and survived (in the field we counted four successfully rooted bulbils out of 78 shed ones; our establishment experiment showed that if bulbils fall under a nurse plant their survival is much higher, but this does not always occur under natural field conditions). In short, the expectation that a damaged plant will produce bulbils, shed them in time, and that bulbils will root is around 2% (0.37 x 0.26 x 3.88 x 0.05). That is, 98% of all bulbiferous scapes will not produce offspring. Even so, any chance, however small, of clonally recovering from reproductive failure may be adaptive.
Basal shoots and rhizomes
In contrast with the questionable importance of bulbils as reliable mechanisms for clonal propagation, basal and rhizome shoot production was undeniably very effective as a cloning mechanism. Most (98%) of the rosettes excavated had one or more ground-level vegetative offspring, either consisting of basal shoots, developed rhizome propagules, or developing rhizomes, totaling almost three offspring shoots per rosette. Most of the independent rosettes were formed from similar vegetative shoots, as indicated by the remains of dead rhizomes in the base of the stem and by their funnel-like form. Of the 44 randomly selected plants, only one small individual, 8 cm in diameter, had a rounded stem base and no traces of decomposed rhizomes, a fact that suggests that it was originated from seed. The idea that ground-level cloning is the main propagation mechanism in A. macroacantha is also consistent with the highly aggregated spatial pattern of the clumps.
The formation of basal and underground shoots may vary significantly in different Agave species from less than 5 to 15 or more clonal offspring per rosette (Table 2). Other authors have also reported basal shoots and rhizomes as a main mechanism of colonization that has a demographic quantitative importance that surpasses that of sexual reproduction (Gentry, 1972, 1982
; Freeman and Reid, 1985
; Raphael and Nobel, 1986
). A consequence of this natural trait is the widespread use of cloning to propagate cultivated agaves, with the consequence of decreased genetic variation in those crops (Colunga-GarcíaMarín and May-Pat, 1993, 1997
; Colunga-GarcíaMarín, Estrada-Loera, and May-Pat, 1996
). When asked, agave farmers in the Tehuacán Valley failed consistently to recognize the seeds of their own plants.
|
FOOTNOTES |
|---|
4 Author for correspondence (sarizaga{at}oikos.unam.mx
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Aguilera N. 1970 Suelos de las zonas áridas de Tehuacán, Puebla y sus relaciones con las cactáceas. Cactáceas y Suculentas Mexicanas 15: 51-63
Álvarez A. 1987 Sistemática y filogenia de la familia Agavácea Endlicher. Ph.D. dissertation, Facultad de Biología, Jardín Botánico Nacional, Universidad de la Habana, Habana, Cuba
Álvarez A. 1989 Distribución geográfica y posible origen de las Agaváceas. Revista del Jardín Botánico Nacional de la Habana X: 25-36
Arizaga S. 1999 Biología reproductiva de Agave macroacantha Zucc. en Tehuacán, Puebla. Ph.D. dissertation, Facultad de Ciencias, Universidad Nacional Autónoma de México, México D.F., México
Arizaga S. E. Ezcurra 1995 Insurance against reproductive failure in a semelparous plant: bulbil formation in Agave macroacantha flowering stalks. Oecologia 101: 329-334[CrossRef][ISI]
Arizaga S. E. Ezcurra E. Peters F. Ramírez de Arellano E. Vega-Peña 2000a Pollination ecology of Agave macroacantha Zucc. 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. Ramírez de Arellano E. Vega-Peña 2000b Pollination ecology of Agave macroacantha Zucc. in a Mexican tropical desert. II. The role of pollinators. American Journal of Botany 87: 1011-1017
Baskin J. M. C. C. Baskin 1971 The ecological life history of Agave virginica L. in Tennessee Cedar Glades. American Midland Naturalist 86: 449-462[CrossRef][ISI]
Bell A. D. 1991 Plant form: an illustrated guide to flowering plant morphology. Oxford University Press, Oxford, UK
Benítez F. 1965 El drama de un pueblo y una planta: Ki, vida y pensamiento de México. Fondo de Cultura Económica, México D.F., México
Böhm W. 1979 Methods of studying root systems. Springer-Verlag, Berlin, Germany
Callen E. O. 1965 Food habits of some precolumbian Mexican Indians. Economic Botany 19: 335-343
Colunga-GarcíaMarín P. E. Estrada-Loera F. May-Pat 1996 Patterns of morphological variation, diversity and domestication of wild and cultivated populations of Agave in Yucatan, Mexico. American Journal of Botany 83: 1069-1082[CrossRef][ISI]
Colunga-GarcíaMarín P. F. May-Pat 1993 Agave studies in Yucatan, Mexico. I. Past and present germplasm diversity and uses. Economic Botany 47: 312-327[ISI]
Colunga-GarcíaMarín P. F. May-Pat 1997 Morphological variation of Henequén germplasm and its wild ancestor under uniform growth conditions: diversity and domestication. American Journal of Botany 84: 1449-1465[Abstract]
Cook R. E. 1985 Growth and development in clonal plant populations. In J. B. C. Jackson, L. W. Buss, and R. E. Cook [eds.], Population biology and evolution of clonal organisms, 259–300. Yale University Press, New Haven, Connecticut, USA
Elheringer J. R. 1985 Annuals and perennials of warm deserts. In B. F. Chabot and D. A. Mooney [eds.], Physiological ecology of north plant communities, 162–180. Chapman & Hall, New York, New York, USA
Ellstrand N. C. M. L. Roose 1987 Patterns of genotypic diversity in clonal plant. American Journal of Botany 74: 123-131[CrossRef][ISI]
Eriksson O. 1989 Seedling dynamic and life history in clonal plants. Oikos 55: 231-238[CrossRef][ISI]
Eriksson O. 1992 Evolution of seed dispersal and recluitment in clonal plants. Oikos 63: 439-444[CrossRef][ISI]
Felger R. S. M. B. Moser 1970 Seri use of Agave. Kiva 35: 159-167
Flores-Martínez A. 1994 Papel de Mimosa luisana en la estructura de la comunidad y su relación con Neobuxbaumia tetetzo en el Valle de Zapotitlán de las Salinas, Puebla. Ph.D. dissertation, Centro de Ecología, Universidad Nacional Autónoma de México, México D.F., México
Freeman C. E. 1973 Some germination responses of lechuguilla (Agave lechuguilla Torr). Southwestern Naturalist 18: 125-134[CrossRef]
Freeman C. E. 1975 Germination responses of a New Mexico population of parry agave (Agave parryi Engelm. var. parryi) to constant temperature, water stress, and pH. Southwestern Naturalist 20: 69-74
Freeman C. E. W. H. Reid 1985 Aspects of the reproductive biology of Agave lechuguilla Torr. Desert Plants 7: 75-80
García E. 1987 Modificaciones al sistema de clasificación climática de Köppen (para adaptarlo a las condiciones climáticas de la República Mexicana), 4th ed. Instituto de Geografía, Universidad Nacional Autónoma de México, México D.F., México
García-Mendoza A. 1995 Riqueza y endemismos de la familia Agavaceae en México. In E. Linares, P. Dávila, F. Chiang, R. Bye, and T. Elias [eds.], Conservación de plantas en peligro de extinción: diferentes enfoques, 51–75. Universidad Nacional Autónoma de México, México D.F., México
Gentry H. S. 1972 The agave family in Sonora. U.S. Department of Agriculture, Handbook No. 399, Washington D.C., USA
Gentry H. S. 1982 Agaves of continental North America. University of Arizona Press, Tucson, Arizona, USA
Gómez-Pompa A. 1963 El género Agave. Cactáceas y Suculentas Mexicanas 8: 3-25
Gonçalves y Lima O. 1956 El maguey y el pulque en los códices mexicanos. Fondo de Cultura Económica, México D.F., México
Granados D. 1993 Los agaves en México. Universidad Autónoma Chapingo, Texcoco, México
Grime J. P. 1979 Plant strategies and vegetation processes. John Wiley & Sons, Chichester, UK
Harper J. L. 1977 Population biology of plants. Academic Press, London, UK
Harper J. L. 1985 Module, branches, and the capture of resources. In J. B. C. Jackson, L. W. Buss, and R. E. Cook [eds.], Population biology and evolution of clonal organisms, 1–33. Yale University Press, New Haven, Connecticut, USA
Herrera J. 1987 Flower and fruit biology in southern spanish mediterranean shrub lands. Annals Missouri Botanical Garden 74: 69-78
Howell D. J. B. S. Roth 1981 Sexual reproduction in Agave: the benefits of bats; the cost of semelparous advertising. Ecology 62: 1-7[CrossRef][ISI]
Jordan P. W. P. S. Nobel 1979 Infrequent establishment of seedlings of Agave deserti (Agavaceae) in the northwestern Sonora desert. American Journal of Botany 66: 1079-1084[CrossRef][ISI]
Martínez-Morales R. S. E. Meyer 1985 A demographic study of maguey verde (Agave salmiana spp. crassispina) under conditions of intense utilization. Desert Plants 7: 61-64, 101–103
Meyrán G. J. 1980 Guía botánica de cactáceas y otras suculentas del Valle de Tehuacán. Sociedad Mexicana de Cactología, México D.F., México
Nobel P. S. 1977 Water relations of flowering of Agave deserti. Botanical Gazette 138: 1-6
Nobel P. S. 1984 Extreme temperatures and thermal tolerances for seedlings of desert succulents. Oecologia 62: 310-317[CrossRef][ISI]
Peters E. M. 1993 Variaciones microclimáticas de un desierto intertropical en el Centro de México: algunas implicaciones sobre la vegetación. B.Sc. thesis, Facultad de Ciencias, Universidad Nacional Autónoma de México, México D.F., México
Pitelka L. F. J. W. Ashmun 1985 Physiology and integration of ramets in clonal plants. In J. B. C. Jackson, L. W. Buss, and R. E. Cook [eds.], Population biology and evolution of clonal organisms, 399–435. Yale University Press, New Haven, Connecticut, USA
Pyke D. A. J. N. Thompson 1986 Statistical analysis of survival and removal rate experiments. Ecology 67: 240-245[CrossRef][ISI]
Raphael D. O. P. S. Nobel 1986 Growth and survivorship of ramets and seedling of Agave deserti: influences of parent-ramet connections. Botanical Gazette 147: 78-83[CrossRef]
Rzedowski J. 1978 Vegetación de México. Editorial Limusa, México D.F., México
Rzedowski J. 1991 Diversidad y origenes de la flora fanerogámica de México. Acta Botanica Mexicana 14: 3-21
Silander J. A. Jr. 1985 Microevolution in clonal plants. In J. B. C. Jackson, L. W. Buss, and R. E. Cook [eds.], Population biology and evolution of clonal organisms, 107–152. Yale University Press, New Haven, Connecticut, USA
Smith E. C. 1965 Flora, Tehuacán Valley. Fieldiana Botany 31: 107-143
Steenbergh W. F. C. H. Lowe 1969 Critical factors during the first years of life of the saguaro (Cereus giganteus) at Saguaro National Monument. Ecology 50: 825-834[CrossRef]
Stephenson G. A. 1981 Flowers and fruit abortions: proximate causes and ultimate functions. Annual Review of Ecology and Systematics 12: 253-279
Szarek S. R. G. E. Holmesley 1996 Physiological activity in persistent bulbils of Agave vilmoriniana (Agavaceae). American Journal of Botany 83: 903-909[CrossRef][ISI]
Thomas A. G. H. M. Dale 1975 The role of seed reproduction in the dynamics of established populations of Hieracium floribundum and comparison with that of vegetative reproduction. Canadian Journal of Botany 53: 3022-3031
Turner R. M. S. M. Alcorn G. Olin J. Booth 1966 The influence of shade, soil, and water on saguaro seedling establishment. Botanical Gazette 127: 95-102[CrossRef]
Valenzuela A. G. 1994 El agave tequilero: su cultivo e industrialización. Editorial Agata, Guadalajara, Mexico
Valiente L. 1991 Patrones de precipitación en el Valle semiárido de Tehuacán, Puebla, México. B.Sc. thesis, Facultad de Ciencias, Universidad Nacional Autónoma de México, México D.F., México
Valiente-Banuet A. E. Ezcurra 1991 Shade as a cause of the association between the cactus Neobuxbaumia tetetzo and the nurse plant Mimosa luisana in the Tehuacán valley, Mexico. Journal of Ecology 79: 961-970[CrossRef][ISI]
Villaseñor J. L. P. Dávila F. Chiang 1990 Fitogeografía del Valle de Tehuacán-Cuicatlán. Boletín de la Sociedad Botánica de México 50: 135-149
Watson M. A. 1984 Developmental constraints: effect on population growth and patterns of resource allocation in clonal plants. American Naturalist 123: 411-426[CrossRef][ISI]
Willson M. F. 1983 Plant reproductive ecology. John Wiley, New York, New York, USA
Zavala-Hurtado J. A. 1982 Estudios ecológicos en el Valle semiárido de Zapotitlán, Puebla. I. Clasificación numérica de la vegetación basado en atributos binarios de presencia o ausencia de las especies. Biótica 7: 99-120
This article has been cited by other articles:
|
|
 |
|
  K. C. Parker, J. L. Hamrick, W. C. Hodgson, D. W. Trapnell, A. J. Parker, and R. K. Kuzoff Genetic consequences of pre-Columbian cultivation for Agave murpheyi and A. delamateri (Agavaceae) Am. J. Botany, September 1, 2007; 94(9): 1479 - 1490. [Abstract] [Full Text] [PDF]
|
|
