Annals of Botany 89: 427-434, 2002
© 2002 Annals of Botany Company
Temporal and Spatial Patterns of Seed Dispersal in Two Cistus Species (Cistaceae)
1Departamento de Ciencias Agroforestales, Escuela Politécnica Superior, Universidad de Huelva, Ctra. Palos s/n, 21819 Palos de la Frontera, Huelva and 2Departamento de Biología Vegetal y Ecología, Universidad de Sevilla, Apdo. 1095, 41080 Sevilla, Spain
* For correspondence. Fax: +34 959 017304, e-mail bastida{at}uhu.es
Received: 23 August 2001; Returned for revision: 26 October 2001; Accepted: 2 January 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Cistus species are obligate seeding, early colonizers that follow disturbance, particularly fire, in Mediterranean ecosystems. We studied seed release, seed dispersal and soil seed populations in stands of Cistus ladanifer and C. libanotis. Seed release started in mid- to late summer (C. ladanifer) or in early autumn (C. libanotis), and continued for a very extended period: 8–10 months in C. ladanifer, and for a mean of 16 months in C. libanotis. The xerochastic capsules of both species released seeds by successive dehiscence of the locules. All capsules begin to dehisce simultaneously at the start of the seed release period, but in C. libanotis capsule fragmentation replaced dehiscence early in the seed release period. In plants of both species, seed shadows were characterized by a peak of density beneath the plant canopy and a very short tail of much lower densities, indicating that seeds are concentrated beneath mother plants when dispersed. Nevertheless, in late May, at the onset of the fire season, soil seed densities beneath plant canopies were low compared with densities expected from seed shadows, but were apparently high enough to allow recovery of the stands if a disturbance, such as fire, had taken place. Seed-eating Bruchidae in summer and granivorous ants during the seed release period were apparently the main causes of seed losses. Results suggest that in both Cistus species, the staggered seed release could constitute an efficient risk-reducing trait. The plant pool of seeds existing throughout most of the year could be a relevant component of Cistus seed banks.
Key words: Cistus ladanifer, Cistus libanotis, seed dispersal, seed bank, seed release, seed shadow.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
When seed is shed from mother plants, most seeds fall under the parent canopies (Janzen, 1970; Howe and Smallwood, 1982; Willson, 1993), where survival is likely to be low because of distance- or density-dependent mortality factors, such as sib and seedling competition, predation or pathogen attack (Janzen, 1969, 1971; Augspurger, 1983; Levin et al., 1984; Harms et al., 2000). Thus, the pattern of seed distribution further from mother plants has been considered the most important part of the seed shadow for plant fitness and population ecology (Schupp and Fuentes, 1995; Wenny, 2000). Nevertheless, dispersal can interact with other traits in determining plant fitness. For example, seed dormancy, leading to the production of soil seed banks, allows escape from unfavourable conditions in time rather than in space. Because of the risk-reducing advantages of dormancy and dispersal, a trade-off may be established, such as decreasing dispersal and increasing dormancy (Levin et al., 1984; Venable and Brown, 1988; Rees, 1993; Willson, 1993). Another trait that may reduce risks is the timing of seed release, because it may influence dispersal distances (Nathan et al., 1999), predation (Harper, 1977; Willson, 1992) and the environmental conditions during germination and seedling establishment (Whelan et al., 1998).
The genus Cistus L. (Cistaceae) comprises obligatory-seeder shrubs mainly distributed around the Mediterranean basin (Le Houérou, 1974; Arianoutsou and Margaris, 1981). Cistus populations constitute early successional stages adapted to disturbances operating in Mediterranean ecosystems (Trabaud, 1995), particularly fire (Arianoutsou and Margaris, 1981; Thanos and Georghiou, 1988). Cistus seeds are characterized by the presence of physical dormancy, which, in addition to high seed longevity (Troumbis and Trabaud, 1986) and small size and mass, allows the generation of persistent soil seed banks (Grime, 1989). Dormancy is broken down by the high temperatures generated in the top layers of soil by fire (Thanos and Georghiou, 1988; Thanos et al., 1992; Pérez-García, 1997; Izhaki et al., 2000), and this process leads to high densities of Cistus seedlings in the rainy season following fire (Naveh, 1974; Thanos et al., 1989; Pugnaire and Lozano, 1997; Ferrandis et al., 1999, 2001). Moreover, each year adult Cistus individuals release a vast amount of seeds with no special adaptations for distant dispersal. Thus, it has been suggested that they concentrate beneath mother plants after release (Martín and López-Guinea, 1949; Troumbis and Trabaud, 1986; Trabaud and Oustric, 1989), although no specific quantitative data have been published. Therefore, distance- or density-dependent mortality factors could operate on Cistus seeds and seedlings. In fact, over several years we have recorded high levels of seed predation by diverse granivorous ants (Pheidole pallidula, Messor bouvieri, Messor sp., Goniomma hispanicum and G. kugleri) in Cistus stands in the SW Iberian Peninsula, and some of these ants collected seeds both on the plants and on the soil (Wilcock and de Almeida, 1988; Bastida, 1999).
The aim of this study was to explore seed release and dispersal characteristics in Cistus populations and their influence on seed and seedling mortality. We studied two Cistus species, C. ladanifer L., a widespread species with small seeds, and C. libanotis L., a species with larger seeds and a more restricted distribution. We addressed the following subjects: (1) the temporal pattern of seed release; (2) the spatial pattern of the dispersed seeds; (3) the individual seed crop; and (4) the density of the soil seed bank.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Study site and species
The study was conducted from 1993 to 1995 in two stands of Cistus ladanifer L. (hereafter C. ladanifer-PZ and C. ladanifer-LP) and one stand of Cistus libanotis L., growing on a peniplane at 80–90 m altitude approx. 30 km from the sea (37°18' to 37°20'N and 6°30' to 6°22'W) in Huelva province, SW Spain. The climate is typically Mediterranean, with a mean annual temperature of 16·2 °C and mean annual rainfall of 563 mm. The vegetation in the study area consists of a mixed woodland of stone pine (Pinus pinea L.) and cork oak (Quercus suber L.). The shrub layer is composed mainly of Cistaceae [Cistus salviifolius L. and Halimium halimifolium (L.) Willk.], Lamiaceae (Rosmarinus officinalis L.), Leguminosae (Genista triacanthos Brot. and Cytisus grandiflorus Brot.), Myrtaceae (Myrtus communis L.) and Arecaceae (Chamaerops humilis L.).
Cistus ladanifer is a shrub of 100–250 cm in height, found throughout southern France, the Iberian Peninsula and in northern Africa (Morocco and Algeria). It inhabits nutrient-poor dry soils in warm open areas, forming both extensive dense populations and sparse populations. Fruits are woody capsules with eight to ten locules (carpels) containing seeds approx. 0·8 x 0·6 mm in size. C. libanotis is a shrub 60–150 cm tall, endemic to the south-west of the Iberian Peninsula. It inhabits sandy substrates of stabilized dunes, forming both dense and sparse populations. Fruits are woody capsules with five locules (carpels) and seed size is approx. 1·6 x 1·2 mm. Both species flower in spring and their capsules mature in early summer (May–June) (Herrera, 1987; Talavera et al., 1993).
Temporal pattern of seed release
In late July 1993, at the onset of seed release, three plants were selected at random in each stand and 30–35 mature capsules were tagged per plant. The stands were visited throughout the seed release period, generally at intervals of 5–20 d. On each visit we recorded the number of open locules of each capsule and we estimated visually the fraction of the seed content still remaining inside each open locule. Abscission of the whole capsule or capsule fragments and signs of predation were also recorded. For each plant, the rate of seed release between consecutive visits was estimated as the percentage of the seed content in the tagged capsules released per week (excluding capsules destroyed by pre-dispersal predation), and cumulative seed release was estimated by the percentage of the seed content of the capsules released up to each visit.
Spatial pattern of seed dispersal
In late September 1993, 108 and 84 seed-traps were established in stands of C. ladanifer-LP and C. libanotis, respectively. Four individual plants were selected in each stand. Maximum canopy radius ranged between 0·9 and 1·1 m (mean 1·0 m) in C. ladanifer plants, and between 0·7 and 0·9 m (mean 0·8 m) in C. libanotis plants. To ensure that the capsules closest to a seed-trap were those of a selected plant, we chose plants at least 1·8 m away from the nearest conspecific, i.e. selected plants were at local densities below the mean stand density. Moreover, capsules of adjacent plants located less than 2 m away from a seed-trap, as observed in the C. ladanifer-LP stand, were removed. Three transects, 180 cm long (C. ladanifer) or 140 cm long (C. libanotis), were arranged radially from each plant centre in the three main wind directions at the study site (SW, N and NE). Each transect consisted of contiguous 20 cm (radial) x 8 cm (tangent) wooden, white-painted quadrats (seed-traps), positioned 20 cm above soil level to restrict deposition of wind-driven soil particles and accessibility to predators (ants). To collect fallen seeds, upper surfaces of quadrats were covered with a 1 mm thick layer biodegradable grease (Brugarolas S.A.; Rubí, Barcelona, Spain). This grease was chosen because it maintained its stickiness for about 1 week, after which we counted the number of seeds and removed the grease. This procedure was repeated eight times in the C. ladanifer-LP stand and 12 times in the C. libanotis stand, between October and March; thus the total sampling period was about 2 and 3 months, respectively. The total number of seeds collected at each 20 cm interval (aggregate of three seed-traps), relative to the total number of seeds collected around each plant, gives an estimate of the distance distribution, i.e. the frequency distribution of distances travelled by seeds. When adjusted for 1 m2, these values estimate the relative seed densities at different distances from the source plants, i.e. the shape of the seed shadow (Clark, 1998; Clark et al., 1999; Nathan and Muller-Landau, 2000). Relative seed densities, as defined, are not restricted to the 0–1 interval.
Seed crop and soil seed bank
In September 1994, at the start of seed release, we estimated the fruit crop of individual plants. Ten plants were selected at random in both the C. ladanifer-LP and C. libanotis stands, and the number of capsules per plant was counted. Moreover, the seed content of fruits was estimated by counting the number of fully developed seeds in 30–55 capsules from 12–15 plants, selected at random in each stand. The individual seed crop was then estimated as the product of the number of capsules and the mean seed content of capsules.
In late May 1995, before the onset of seed release, we estimated the viable seed content in the soil beneath individual plant canopies, where soil seed densities were expected to be maximal. In both the C. ladanifer-LP and C. libanotis stands, soil samples (0·50 x 0·50 x 0·05 m) were obtained from beneath the canopy of ten individual plants selected at random. Samples were stored in the laboratory in open plastic boxes for 6 months. Because of the high fraction of dormant seeds, we applied an extraction method to determine the viable seed content in soil samples. Each soil sample was first homogenized and weighed, then we retained a quarter of the sample weight. This sub-sample was passed through a graded series of sieves. If necessary, tap water was added to break up soil aggregates. Two size fractions were retained: those containing the fruits and the seeds. In fruit-containing fractions, whole capsules and capsule fragments were sorted and the seeds they contained were extracted. Seed-containing fractions were first homogenized and weighed, and then ten samples of 0·5 g each were obtained. Seeds in the samples were sorted under a binocular microscope. All the seeds obtained were tested for viability by tetrazolium staining, and the number of viable seeds was adjusted for 1 m2 soil surface.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Temporal pattern of seed release
In both stands of C. ladanifer, a high proportion of the tagged mature capsules abscised early in the season (August–mid-September), generally before starting to dehisce. The major cause of fruit loss was predation. Predation was mainly due to a capsule-perforating Coleopteran of the family Bruchidae, whose larvae developed inside capsules and fed on seeds. Pre-dispersal predation of capsules was notably higher in stand C. ladanifer-PZ (43 % of the tagged fruits) than in C. ladanifer-LP (12 %). In contrast, in the C. libanotis stand only one capsule (1 % of the tagged fruits) was lost during the summer.
In both C. ladanifer stands, the plants showed a very extended period of seed release (Fig. 1). Seed release started in the summer (late July–September) and 50 % of the seeds had been released by early November (C. ladanifer-LP stand) and mid-December (C. ladanifer-PZ stand; Fig. 1). Seed release was completed (at least 90 %) between March and May. Therefore, in both stands, plants required 8–10 months for full seed release. In C. libanotis, the period of seed release was markedly longer than in C. ladanifer. Seed release started in late September or early October, and 50 % seed release was not achieved until January or early February (Fig. 1). From October, the plants required a mean of 16 months (range 9–20 months) to achieve 90 % seed release (Fig. 1). In October 1994, at the start of the next seed release period, 12·5 % (range 7·8–17·8 %) of the previous year’s seed crop still remained on the plants (Fig. 1). In all the stands, the plants showed a very variable rate of seed release throughout the season (Fig. 1).
|
In C. ladanifer capsules, seed release took place by successive dehiscence of the locules. Once a locule was open, a rise in humidity (e.g. due to rain) induced its partial closure, with subsequent reopening occurring when the humidity decreased again (xerochasy; Weberling, 1989). In fact, the seed dispersal rate was notably higher when capsules dried after a rainy period. In contrast, under continuous dry weather, particularly in the absence of dew formation (east winds blowing), no seed release was observed. In both C. ladanifer stands, all the capsules had already started to dehisce in October, but needed between 4 and 10 months to accomplish seed release. Of the capsules, 16·7 (C. ladanifer-PZ stand) or 39·4 % (C. ladanifer-LP stand) accomplished seed release in 4–6 months; another 54·2 or 32·4 %, respectively, required 6–8 months; and the remaining 29·1 or 28·2 % required 8–10 months. The different durations of seed release were probably due to the location of the fruit, with capsules in shaded locations being slower to release seeds than capsules in exposed, outer parts of the canopy. Throughout the seed release period, we observed an intense activity of granivorous ants (Goniomma hispanicum and G. kugleri). These ants collected seeds directly from open locules of capsules and carried them to their nests.
In C. libanotis, a significant fraction of seeds were dispersed inside whole capsules or in capsule fragments that abscised. Release of free seeds via locule dehiscence accounted for only 38·1 % (range 32·1–45·9 %) of the indi vidual seed crop. In each plant, 17·1 % (range 14·8–18·2 %) of the capsules dropped as whole units. Nevertheless, most of the capsules in each plant (78·5 %, range 77·7–78·8 %) initially released the seed content of two or three locules via dehiscence, followed by capsule breakdown and dispersal of one or two abscised, seed-filled locules. Less frequent dispersal features were dehiscence of only one locule (1·0 % of the capsules in each plant, range 0–3·0 %), four locules (2·5 %, range 0–7·4 %) or capsules breaking down in two fragments (1·0 %, range 0–3·0 %). As in both C. ladanifer stands, activity of granivorous ants was observed (Messor bouvieri and Messor sp.) during the seed release period. These ants took seeds from the plants and the soil, and the largest foragers collected seed-filled capsule fragments from the soil surface.
Spatial pattern of seed dispersal
During the sampling period, a total of 793 and 123 seeds was collected by the seed-traps in the C. ladanifer-PZ and C. libanotis stands, respectively. The frequency distribution of seeds beneath and outside the plant canopies did not differ between seed-traps positioned in the three different directions, either in the C. ladanifer-LP stand (test for independence: 
2 = 3·40, P = 0·182, d.f. = 2) or in the C. libanotis stand (2 = 2·57, P = 0·276, d.f. = 2). The distance distribution (Fig. 2A) indicated that the vast majority of seeds landed beneath plant canopies. In C. ladanifer plants, the mean proportion of the seed crop landing beneath the plant canopy was 79·6 % (range 69·4–89·9 %) and in C. libanotis plants it was 83·4 % (range 62·0–100 %). Moreover, only a mean of 1·6 % of the seed crop landed beyond 40 cm from the canopy edges in both C. ladanifer (range 0–4·4 %) and C. libanotis (range 0–6·4 %) plants. Figure 2B shows the proportion of the seed crop landing per square metre (a relative measure of seed density) as a function of the distance from the plant centre. This figure thus reflects the shape of the seed shadow of C. ladanifer and C. libanotis plants. The maximum value in C. ladanifer plants was 26 % and was recorded at a distance of 20–60 cm (Fig. 2A). In C. libanotis plants, the maximum value was 71 % and was recorded at a distance of 20–40 cm (Fig. 2B). From these points, mean seed densities quickly decreased in both directions.
|
Seed crop and soil seed bank
Cistus ladanifer capsules contained 838 ± 27·0 seeds (mean ± s.e.), whereas the mean seed content in C. libanotis capsules was 18 ± 1·1. The mean fruit crop remaining in September was 189 ± 32·7 and 789 ± 200·0 capsules per plant in the C. ladanifer-LP and C. libanotis stands, respectively. Thus, the calculated mean seed crop at the onset of seed release was 158 382 seeds per plant in the C. ladanifer-LP stand and 14 202 seeds per plant in the C. libanotis stand. Taking into account the mean proportion of seeds landing per square metre beneath plant canopies (Fig. 2B), we estimate that the potential mean seed rain beneath plant canopies resulting from dispersal of the available seed in September was 35 496 and 6152 seeds per m2 in the C. ladanifer-LP and C. libanotis stands, respectively.
The viability of soil borne seeds was similar for the two species: 62·5 % (n = 232) and 66·9 % (n = 372) for C. ladanifer and C. libanotis, respectively. As expected from the data on seed release, a major fraction of the C. libanotis seeds found in the soil samples was contained in capsule fragments, whereas none of the C. ladanifer seeds were found in capsule fragments (Table 1). Mean soil seed densities beneath the individual canopies were 1928 and 1 406 viable seeds m–2 in the C. ladanifer-LP and C. libanotis stands, respectively (Table 1). Thus, soil seed density was only a small fraction of the expected annual seed rain (5·4 % in the C. ladanifer-LP stand and 22·9 % in the C. libanotis stand).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
In plants of C. ladanifer and C. libanotis about 80 % of the seed crop landed beneath the canopies, and the remaining fraction landed in the immediate vicinity of the mother plants. Thus, our results are in agreement with those in the literature that assume barochory to be the primary dispersal strategy of Cistus species based on seed traits (small size and mass, and an aerodynamically inefficient shape) (Martín and López-Guinea, 1949; Barry, 1960; Troumbis and Trabaud, 1986; Trabaud and Oustric, 1989). Seed dispersal is not only determined by seed traits but also by architectural traits of the mother plants (Westoby and Rice, 1982; McCanny and Cavers, 1989; Donohue, 1998). In C. ladanifer, fruits tend to be placed in the outer half and in the upper part of the plant canopy (Talavera et al., 1993), as is generally the case in Cistus species. This positioning of fruits may explain why the highest seed densities were found at distances between 20 and 60 cm from the centre of the mother plant, with the density decreasing further away from, or towards, the plant centre. Seed shadows are also affected by plant density because it influences plant fecundity (Harper, 1977; Silander, 1978), plant architecture (Silander, 1978; Donohue, 1998) and the level of interference of neighbours on dispersing seeds (Theide and Augspurger, 1996). Thus, dispersal distances found in this study could represent upper estimates for individuals in denser clumps, but they reinforce the very short dispersal range in C. libanotis and C. ladanifer plants.
In the two stands of C. ladanifer, moderate to high levels of fruit and seed predation were recorded during the summer. Summer predation was apparently due to the activity of seed-eating Bruchidae. Although pre-dispersal predation in C. libanotis was not relevant in mid- to late summer in the study year, we have noted that substantial fruit loss usually takes place by late spring to late summer. Despite pre-dispersal predation, the number of seeds remaining in the plants in September was notably high. Nevertheless, in May, when seed release had already finished in C. ladanifer, or when approx. 80 % of the seeds had been released in C. libanotis, mean soil densities of viable seeds beneath plant canopies were only a small fraction of the potential densities resulting from the dispersal of the current seed crop in September. Small soil seed populations relative to the annual seed crop have also been reported in stands of other Cistus species (Troumbis and Trabaud, 1987; Troumbis, 1996; Ne’eman and Izhaki, 1999; Trabaud and Renard, 1999) and in ecologically similar plant species of the Californian chaparral (Keeley, 1977; Kelly and Parker, 1990). Granivorous ants that collected seeds directly from open capsules in the plants throughout the seed release period were apparently the main cause of the observed seed losses. Granivorous ants are known to cause substantial seed losses in a variety of ecosystems (e.g. Reichman, 1979; Wellington and Noble, 1985; Kerley, 1991; McMahon et al., 2000).
Seed release in C. ladanifer and C. libanotis capsules takes place by successive dehiscence of the locules, in response to dry weather conditions following wet periods. In C. libanotis, capsule fragmentation replaced dehiscence early in the seed release period. In both C. ladanifer and C. libanotis plants, all the capsules begin to dehisce at the onset of seed release rather than in a staggered fashion, and the length of the plant and stand seed release period is established by the slow-dispersing capsules. The result is a very extended period of seed release, from mid-summer/early autumn to spring (C. ladanifer) or even longer (C. libanotis). To our knowledge, this is the first time that such extended seed release periods have been recorded in C. ladanifer and C. libanotis, or in any Cistus species. However, we monitored seed release for one season only and thus the length of the seed release period should only be considered indicative; between-year fluctuations could occur depending on climatic conditions (Primack, 1980).
Some factors can be postulated to be the selective pressures determining the extended seed release pattern. First, because of the unpredictable distribution of the rainy periods, some dry weeks may occur in the rainy season in Mediterranean climates. Thus, the gradual seed release can be interpreted as a favourable, opportunistic strategy for spreading the risk of germination and seedling establishment. In fact, Cistus seeds germinate over a wide range of mild to moderate low temperatures (Vuillemin and Bulard, 1981; Thanos and Georghiou, 1988), allowing germination to be dependent upon water availability. Moreover, seedling emergence in several Cistus species has been recorded in nature throughout the rainy season (Bastida, 1999).
Secondly, because of successive waves of locule dehiscence, an iteration of events of sudden exposition of great amounts of seeds, easily accessible to ants, takes place in each Cistus plant throughout the seed release period. Thus a predator-satiation strategy (Janzen, 1969) could allow some seeds to escape predation in each event. These seeds can germinate if suitable conditions occur immediately or, because a high fraction of the annual seed crop of Cistus species is dormant (e.g. Thanos et al., 1992; Trabaud, 1995), they can be incorporated into the soil seed bank. In the non-sprouting genus Cistus, recovery of stands after a disturbance, such as fire, is dependent on seeds surviving in the soil (Le Houérou, 1974; Troumbis and Trabaud, 1986). Thus, this soil seed bank replenishing process could be a favourable strategy giving suitable soil seed densities in late spring and early summer, at the onset of the fire season. In fact, soil seed densities beneath plant canopies in May were high enough to account for the densities of Cistus seedlings reported in the rainy season following fire (Thanos et al., 1989; Pugnaire and Lozano, 1997). Nevertheless, it is not only the temporal components of the above-mentioned processes that are influenced by the extended seed release period; spatial components are also affected. Granivorous ant colonies disperse a fraction of all seeds collected and returned to the nest by foragers (Höldobler and Willson, 1990; Wolff and Debussche, 1999; McMahon et al., 2000). Thus, diszoochory could result from the Cistus–granivorous ant interaction. Seed dispersal by ants increases the probability of seedling survival because of diverse causes (e.g. Beattie, 1985), some of which could operate in Cistus. For example, seedlings are unlikely to survive beneath the closed canopy that characterizes the dense stands of Cistus species (Barry, 1960; Roy and Sonié, 1992; Robles et al., 1999), and secondary dispersal by ants could reduce or eliminate the influence of adult plants. In addition, longer ranging seed dispersal has been found in C. ladanifer due to ingestion of mature closed capsules during summer by red deer (Malo and Suárez, 1996).
The staggered seed release pattern implies that C. ladanifer and C. libanotis plants retain seeds most of the year, and that for many months each year seed numbers on the plants are higher, probably by several orders of magnitude, than seed numbers in the soil. Thus, we argue that the plant pool of seeds could be a relevant component of the Cistus seed bank. Each year, in the absence of disturbance, the plant pool is transferred to the soil pool in such a staggered manner that plant fitness is maximized under the selective pressures of environmental variability and during- and post-dispersal predation. Therefore, the seed release pattern in C. ladanifer and C. libanotis, together with the long-recognized traits of dormancy and seed size and numbers, should be considered as a risk-reducing factor.
|
ACKNOWLEDGEMENTS |
|---|
We thank Drs X. Espadaler, A. Tinaut and L. Domínguez for insect identification, and Drs P. E. Gibbs and J. Herrera for their helpful comments.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
-
Arianoutsou M, Margaris MS. 1981. Early stages of regeneration after fire in a phryganic ecosystem (East Mediterranean) I. Regeneration by seed germination. Biologie-Écologie Méditerranéenne 8: 119–128.
Augspurger CK. 1983. Seed dispersal of the tropical tree Platypodium elegans, and the escape of its seedlings from fungal pathogens. Journal of Ecology 71: 759–771.[CrossRef]
Barry JP. 1960. Contribution à l’étude de la végétation de la région de Nîmes. Année Biologique 36: 311–550.
Bastida F. 1999. Aspectos reproductivos del matorral de Cistaceae en el litoral onubense. PhD Thesis, Universidad de Sevilla, Seville, Spain.
Beattie AJ. 1985. The evolutionary ecology of ant–plant mutualisms. Cambridge: Cambridge University Press.
Clark JS. 1998. Why trees migrate so fast: confronting theory with dispersal biology and the paleorecord. The American Naturalist 152: 204–224.
Clark JS, Silman M, Kern R, Macklin E, HilleRisLambers J. 1999. Seed dispersal near and far: patterns across temperate and tropical forests. Ecology 80: 1475–1494.[CrossRef]
Donohue K. 1998. Maternal determinants of seed dispersal in Cakile edentula: fruit, plant and site traits. Ecology 79: 2771–2788.
Ferrandis P, Herranz JM, Martínez-Sánchez JJ. 1999. Effect of fire on hard-coated Cistaceae seed banks and its influence on techniques for quantifying seed banks. Plant Ecology 144: 103–114.
Ferrandis P, de las Heras J, Martínez-Sánchez JJ, Herranz, JM. 2001. Influence of a low-intensity fire on a Pinus halepensis Mill. forest seed bank and its consequences on the early stages of plant succession. Israel Journal of Plant Sciences 49: 105–114.
Grime JP. 1989. Seed banks in ecological perspectives. In: Leck MA, Parker VT, Simpson RL, eds. Ecology of soil seed banks. New York: Academic Press.
Harms KE, Wright SJ, Calderón O, Hernández A, Herre EA. 2000. Pervasive density-dependent recruitment enhances seedling diversity in a tropical forest. Nature 404: 493–494.[CrossRef][Medline]
Harper JL. 1977. Population biology of plants. New York: Academic Press.
Herrera J. 1987. Biología reproductiva de algunas especies del matorral de Doñana. Anales del Jardín Botánico de Madrid 44: 483–497.
Höldobler B, Willson EO. 1990. The ants. Cambridge: Harvard University Press.
Howe HF, Smallwood J. 1982. Ecology of seed dispersal. Annual Review of Ecology and Systematics 7: 469–495.[CrossRef][Web of Science]
Izhaki I, Henig-Sever N, Ne’eman G. 2000. Soil seed banks in Mediterranean Aleppo pine forests: the effect of heat, cover and ash on seedling emergence. Journal of Ecology 88: 667–675.[CrossRef]
Janzen DH. 1969. Seed-eaters versus seed size, number, toxicity and dispersal. Evolution 23: 1–27.
Janzen DH. 1970. Herbivores and the number of tree species in tropical forests. The American Naturalist 104: 501–528.[CrossRef][Web of Science]
Janzen DH. 1971. Seed predation by animals. Annual Review of Ecology and Systematics 2: 465–492.
Keeley JE. 1977. Seed production, seed populations in the soil, and seedling production after fire for two congeneric pairs of sprouting and non-sprouting chaparral shrubs. Ecology 58: 820–829.
Kelly VR, Parker VT. 1990. Seed bank survival and dynamics in sprouting and nonsprouting Arctostaphylos species. American Midland Naturalist 124: 114–123.
Kerley GIH. 1991. Seed removal by rodents, birds and ants in the semi-arid Karoo. South African Journal of Arid Environments 20: 63–69.
Le Houerou HN. 1974. Fire and vegetation in the Mediterranean basin. In: Kormarck R, ed. Proceedings of the 13th Annual Tall Timber Fire Ecology Conference. Tallahassee: Tall Timbers Research Station.
Levin S, Cohen D, Hastings A. 1984. Dispersal strategies in patchy environments. Theoretical and Population Biology 26: 165–191.
McCanny SJ, Cavers PB. 1989. Parental effects on spatial patterns of plants: a contingency table approach. Ecology 70: 368–378.
McMahon JA, Mull JF, Crist TO. 2000. Harvester ants (Pogonomyrmex spp.): their community and ecosystem influences. Annual Review of Ecology and Systematics 31: 265–291.
Malo JE, Suárez F. 1996. Cistus ladanifer recruitment – not only fire, but also deer. Acta Oecologica 17: 55–60.
Martín M, López-Guinea E. 1949. Jarales y jaras (Cistografía hispánica). Madrid: Instituto Forestal de Investigaciones y Experiencias.
Nathan R, Muller-Landau HC. 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Trends in Ecology and Evolution 15: 278–285.
Nathan R, Safriel UN, Noy-Meir I, Schiller G. 1999. Seed release without fire in Pinus halepensis, a Mediterranean serotinous wind-dispersed tree. Journal of Ecology 87: 659–669.[CrossRef]
Naveh Z. 1974. Effects of fire in the Mediterranean region. In: Kozlowsky TT, Ahlgren CE, eds. Fire and ecosystems. New York: Academic Press.
Ne’eman G, Izhaki I. 1999. The effect of stand age and microhabitat on soil seed banks in Mediterranean Aleppo pine forests after fire. Plant Ecology 144: 115–125.[CrossRef]
Pérez-García F. 1997. Germination of C. ladanifer seeds in relation to parent material. Plant Ecology 133: 57–62.[CrossRef]
Primack RB. 1980. Variation in the phenology of natural populations of montane shrubs in New Zealand. Journal of Ecology 68: 849–862.[CrossRef]
Pugnaire FI, Lozano J. 1997. Effects of soil disturbance, fire and litter accumulation on the establishment of Cistus clusii seedlings. Plant Ecology 131: 207–213.[CrossRef]
Rees M. 1993. Trade-offs among dispersal strategies in British plants. Nature 336: 150–152.
Reichman OJ. 1979. Desert granivore foraging and its impact on seed densities and distributions. Ecology 60: 1085–1092.
Robles C, Bonin G, Garzino S. 1999. Potentialités autotoxiques et allélopathiques de Cistus albidus L. Comptes Rendus Academie Sciences Paris, Sciences de la vie 322: 677–685.
Roy J, Sonié L. 1992. Germination and population dynamics of Cistus species in relation to fire. Journal of Applied Ecology 29: 647–655.
Schupp EW, Fuentes M. 1995. Spatial patterns of seed dispersal and the unification of plant population ecology. Écoscience 2: 267–275.
Silander JA. 1978. Density-dependent control of reproductive success in Cassia biflora. Biotropica 10: 292–296.
Talavera S, Gibbs PE, Herrera J. 1993. Reproductive biology of Cistus ladanifer (Cistaceae). Plant Systematics and Evolution 186: 123–134.[CrossRef]
Thanos CA, Georghiou K. 1988. Ecophysiology of fire-stimulated seed germination in Cistus incanus spp. creticus (L.) Heywood and C. salvifolius L. Plant, Cell and Environment 11: 841–849.
Thanos CA, Georghiou K, Kadis C, Pantazi C. 1992. Cistaceae: a plant family with hard seeds. Israel Journal of Botany 41: 251–263.
Thanos CA, Marcou S, Christodoulakis D, Yannitsaros A. 1989. Early post-fire regeneration in Pinus brutia forest ecosystems of Samos island (Greece). Acta Oecologica, Oecologia Plantarum 10: 79–94.
Theide DA, Augspurger CK. 1996. Intraspecific variation in seed dispersion of Lepidium campestre (Brassicaceae). American Journal of Botany 83: 856–866.
Trabaud L. 1995. Modalités de germination des cistes et des pins méditerranéens et colonisation des sites perturbés. Revue Ecologie (Terre Vie) 50: 3–14.
Trabaud L, Oustric J. 1989. Heat requirement for seed germination of three Cistus species in the garrigues of Southern France. Flora 183: 321–325.
Trabaud L, Renard P. 1999. Do light and litter influence the recruitment of Cistus spp. stands? Israel Journal of Plant Sciences 47: 1–9.
Troumbis A. 1996. Seed persistence versus soil seed persistence: the case of the post-fire seeder Cistus incanus L. Écoscience 3: 461–468.
Troumbis A, Trabaud L. 1986. Comparison of reproductive biological attributes of two Cistus species. Acta Oecologica, Oecologia Plantarum 7: 235–250.
Troumbis A, Trabaud L. 1987. Dynamique de la banque de graines de deux espèces de Cistes dans le maquis grecs. Acta Oecologica, Oecologia Plantarum 8: 167–179.
Venable DL, Brown JS. 1988. The selective interactions of dispersal, dormancy, and seed size as adaptations for reducing risk in variable environments. The American Naturalist 131: 360–384.[CrossRef]
Vuillemin J, Bulard C. 1981. Ecophysiologie de la germination de Cistus albidus L. et Cistus monspeliensis L. Naturalia Monspeliensia 46: 1–11.
Weberling F. 1989. Morphology of flowers and inflorescences. Cambridge: Cambridge University Press.
Wellington AB, Noble IR. 1985. Seed dynamics and factors limiting recruitment of the Mallee Eucalyptus incrassata in semi-arid south-eastern Australia. Journal of Ecology 73: 657–666.
Wenny DG. 2000. Seed dispersal, seed predation, and seedling recruitment of a neotropical montane tree. Ecological Monographs 70: 331–351.[CrossRef]
Westoby M, Rice B. 1982. Evolution of the seed plants and inclusive fitness of plant tissues. Evolution 36: 713–724.[CrossRef][Web of Science]
Whelan RJ, de Jong RH, von der Burg S. 1998. Variation in bradyspory and seedling recruitment without fire among populations of Banksia serrata (Proteaceae). Australian Journal of Ecology 23: 121–128.
Wilcock C, de Almeida CJ. 1988. The flora of the Algarve. Plants Today 1: 151–157.
Willson MF. 1993. Dispersal mode, seed shadows, and colonization patterns. Vegetation 107/108: 261–280.
Wolff A, Debussche M. 1999. Ants as seed dispersers in a Mediterranean old-field succession. Oikos 84: 443–452.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||