| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Biology, Fairmont State College, Fairmont, West Virginia 26554-2470
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
An assemblage of myxomycetes associated with inflorescences of large Neotropical herbs, a microhabitat not previously known to support these organisms, is described and characterized ecologically from a number of study sites in Costa Rica, Ecuador, and Puerto Rico. Thirty-one different taxa were found among 652 specimens of myxomycetes recorded in the field or obtained from 358 moist chamber cultures prepared with decaying floral parts. A comparison with the results of 696 moist chamber cultures prepared with various other litter substrates showed that thirteen myxomycete taxa occurred more often on inflorescences. Six taxa had a strong preference for this microhabitat, and three of those seem to be new for the Neotropics. Correspondence analysis of the data set compiled for inflorescences indicated that the assemblage of myxomycetes was relatively consistent across all of the various study sites. The actual myxomycete substrates were the rapidly decaying floral parts enclosed by the massive, still living bracts. Richest in myxomycetes were species of Heliconia and Costus. Here, nectar residuals probably promoted a rapidly developing community of yeasts and bacteria. A high density of these organisms was indicated by the frequent occurrence of myxobacteria in the moist chamber cultures prepared with floral parts. Results from canonical correspondence analysis suggested that a substrate pH between 8 and 9 and the presence of massive, compact inflorescences on plants occurring at lower elevations in localities with moderate annual rainfall provide optimal conditions for inflorescence-inhabiting myxomycetes. An incidental dispersal of myxomycete spores by birds that pollinate the flowers or feed upon the fruits seems possible and may have accounted for the high degree of preference exhibited by some of the inflorescence-inhabiting myxomycetes, for which the term "floricolous" is proposed.
Key words: dispersal, ecology, inflorescence, Neotropics, slime molds
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Well-known substrates for myxomycetes are decaying wood, forest floor litter, or the dung of herbivorous animals. However, these substrate types, which have been the focus of almost all published studies of myxomycete diversity, were found to be relatively poor in terms of both species abundance and diversity in tropical forests (Schnittler and Stephenson 2000
). On the other hand, several types of aerial litter, such as fallen leaves trapped in tree branches or dead but still attached leaves of Heliconia, proved to be productive for myxomycetes in moist chamber cultures. Moreover, during field surveys in Costa Rica and Ecuador, myxomycetes were observed rather frequently on the living inflorescences of these large tropical herbs. The primary objective of the study reported herein was to investigate the assemblage of species of myxomycetes occurring in this newly discovered microhabitat, including field observations as well as results obtained from moist chamber cultures. |
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
|
with a few exceptions, when a reference to a protolog is given. Nomenclature for myxobacteria follows Reichenbach (1993)
. Element analysis – By pooling a small fraction (about one gram dry weight) of the material later used for each of the ca 20 moist chamber cultures constituting one series, samples for analysis of element concentrations were obtained. Eleven samples from inflorescences and nine from leafy litter were sent to Brookside Laboratories (New Knoxville, Ohio). Samples were treated with standard nitric acid-perchloric acid digestion methods, modified slightly from the standard method in that a smaller sample weight and reduced acid amounts were utilized. Trace elements were analyzed on a thermo jarell ash inductively coupled plasma spectrometer (ICP).
Data analysis –
To estimate the completeness of the survey in terms of the species recorded, a bootstrap analysis, as described in detail by Schnittler and Stephenson (2000)
, was applied to the data set obtained from the moist chamber component of the survey. The mean of 100 cumulative plots of species vs moist chamber cultures (sequence permutated randomly for each plot) was subjected to a regression analysis, using a saturation formula y = ax/(b + x), with the parameter a giving an estimate for the total number of species to be expected. In addition, a first-order jackknife estimation of species richness according to Heltshe and Forrester (1983)
was carried out. The respective formula for the number of species S to be expected is S = Sobs + ((n - 1)/n) * Sun, with Sun as the number of species observed in only one sample, Sobs the total number of species observed, and n as the number of samples.
For the correspondence analysis (CA) of the same data set, values of myxomycete frequency in a series of moist chambers prepared from one plant species at a given locality were used. An initial detrended correspondence analysis showed a length of gradient larger than 2, indicating an unimodal distribution of species scores. Accordingly, CA was applied for the final analysis. Since a downweighting of rare species resulted in a similar ordination diagram, the analysis was performed with the original values.
For the canonical correspondence analysis (CCA), each moist chamber culture was regarded as one sample, and weighted abundance values were calculated for each species of myxomycete by dividing the absolute number of sporocarps recorded in a particular culture by the mean value for all cultures yielding the same species. Consequently, the sum of all weighted abundances for all cultures with the species in question is equal to the number of records for this species. This equalizes the often very different sporocarp numbers of the species (compare Table III
). Each myxomycete record was coded for nine environmental parameters. Site parameters were average annual rainfall and elevation (Table I
), whereas microhabitat parameters (recorded for each individual inflorescence) included the sampling height above the ground, light intensity, wind exposure, substrate pH as well as bract arrangement, length and diameter of the inflorescences. Table II
gives average values for a series of moist chambers. Flowers of species of Heliconia and Calathea were shaded by the large leaves overtopping the inflorescences, those of the other plants were terminal. The arrangement of the bracts was quantified with a four-part scale (compare Table II
). Light intensity over the course of a day was estimated using five categories ranging from complete darkness over various degrees of shade to full sunshine. Wind exposure was described using a scale consisting of four categories ranging from fully sheltered to strongly exposed. Both ordinations were carried out with the program Canoco (Ter Braak 1988
). For CA, the resulting eigenvalues, ranging between 0 and 1, are a measure of the extent to which species distribution can be explained by the respective ordination axis (Ter Braak 1987
). CCA was used to assess the relative importance of the recorded abiotic and biotic parameters. For the biplot presented as Fig. 3a , species scores and these of the environmental variables on the canonical axes were symmetrically scaled to mean 1 and sd 1. The centroids of the environmental variables were associated with species by the use of an Euclidian distance matrix.
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. Arcyria cinerea (Bull.) Pers. (ARCcin, 13 records, fc 0, mc 13, pH 7.1 ± 0.5, range 6.5–7.9)
Our specimens represent the typical form of the species, having cream-white to gray, short to long-cylindrical sporothecae on stalks 1 to 1.5 times longer than the sporotheca. In almost all cases, the spores are less than 8 µm in diameter and ornamented with scattered, blunt warts. This species was recorded from all countries and in four different study sites; it was found most often on Heliconia latifolia (CR3, 7 records). Arcyria cinerea is common on aerial and ground litter in most Neotropical forests.
Arcyria cf. cinerea (Bull.) Pers., dwarf form (ARCcin, dwarf, 30 records, fc 1, mc 29, pH 7.9 ± 0.7, range 6.5–9.2)
A detailed description of this taxon, which constitutes at least a distinct biotype, is provided by Schnittler (2001)
. It can be distinguished from typical specimens of Arcyria cinerea on the basis of the much smaller size, a pure white color, and a tiny, globose to elongated sporotheca on a stalk (3–)4–10 times longer than the sporotheca. Usually, the spore diameter exceeds 8 µm, and spores are more densely covered with warts. Since this taxon may intergrade with A. afroalpina Rammeloo, a thorough statistical analysis of morphological characters is necessary before it can be described formally. The taxon was observed on inflorescences of various species of plants from all of the sites investigated, but it was more common on other types of aerial litter (e.g., decaying leaves and palm fronds) as well as on covers of epiphyllous liverworts on living leaves.
Didymium iridis (Ditmar) Fr. (DDYiri, 60 records, fc 7, mc 53, pH 8.2 ± 0.7, range 6.5–9.8)
Our specimens represent the typical form of the species. In contrast to the concept proposed by Clark and Mires (1999)
, specimens possessing a discoid pseudocolumella are separated under the name D. bahiense Gottsb. to maintain comparability with previous studies by other workers. The species was found on various species of plants in all study sites except Ec2 and Pr2. It is very common on all types of aerial litter.
Didymium squamulosum (Alb. & Schwein.) Fr. (DDYsqu, 28 records, fc 8, mc 20, pH 8.2 ± 0.6, range 6.7–9.2)
This species was found on various species of plants from all study sites except Ec2 and Pr2. It is common on leafy litter, both aerial and ground.
Lamproderma arcyrionema Rostaf. (LAManm, 28 records, fc 0, mc 28, pH 7.7 ± 0.4, range 6.9–8.5)
Our specimens have very long, slender stalks. The species was found regularly only on the inflorescences of Hedychium coronarium (PR2, 25 of the 28 records). It is fairly common on aerial litter.
Perichaena cf. dictyonema Rammeloo (PERdic, 38 records, fc 0, mc 38, pH 8.5 ± 0.4, range 7.2–9.3)
The assignment of our specimens to this species is tentative. Specimens are very uniform in habit and clearly distinct from all other species of Perichaena encountered by us in the Neotropics. It seems to fruit on very moist substrates covered with a film of water. The small, globose sporothecae blend in well with the color of decaying inflorescence parts. Both factors may explain why this species is known exclusively from moist chamber cultures. Perichaena cf. dictyonema was recorded from all study sites except those in cloud forests (Ec2, PR2, and CR2) and occurred on all plants with larger, more compact inflorescences that can accumulate an appreciable amount of water. Our records would be the first for the species from the western hemisphere. Other than from inflorescences, it was recorded only twice on the decaying, fleshy sheaths of the under story palm Chamaedorea tepejiote Liebm. and on decaying banana leaves, in both cases above the ground.
Perichaena vermicularis (Schwein.) Rostaf. (PERver, 17 records, fc 4, mc 13, pH 7.8 ± 0.6, range 6.9–9.1)
As recorded in this study, Perichaena vermicularis occurs with very small plasmodiocarps with a thin, membranous peridium. The species was recorded from all countries except Puerto Rico and was most common on inflorescences of Calathea ischnosiphonoides (Ec2, 9 records). It is fairly common on various kinds of (mostly aerial) litter.
Physarum compressum Alb. & Schwein. (PHYcom, 174 records, fc 48, mc 126, pH 8.5 ± 0.6, range 6.4–9.7)
In accordance with the concept proposed by Irawan et al (2000)
, forms resembling P. nicaraguense T. Macbr. are included here. These forms characterized by a cluster of sporothecae on a common stalk appeared only rarely and seemed to intergrade with typical forms. This species was recorded from all countries and plants except Hedychium coronarium. It is fairly common on aerial litter with a basic pH value.
Physarum didermoides (Pers.) Rostaf. (PHYdio, 117 records, fc 66, mc 51, pH 8.6 ± 0.6, range 6.8–9.8)
This species was recorded from all lowland sites included in our study. It was most common on Costus guanaiensis (19 records) and Heliconia mariae (63 records). Both plants have very compact, rather large inflorescences and their decaying floral parts are characterized by a very basic pH. Although represented by a few collections in herbaria, the species was not encountered by us on any substrates other than inflorescences.
Physarum cf. limonium Nann.-Brem. Proc. Kon. Ned. Akad. Wetensch., C 69: 357. 1966 (PHYlim, 6 records, fc 2, mc 4, pH 8.2 ± 0.5, range 7.7–8.9)
The specimens assigned to this species in the present study have yellow sporocarps with long, limeless stalks. Except for the color, they are very similar to the typical form of Physarum pusillum. All six of our specimens are fairly uniform in habit. Physarum cf. limonium was recorded from all countries and four study sites. In addition to inflorescences, it was recorded once from aerial litter (decaying banana leaves).
Physarum melleum (Berk. & Broome) Massee (PHYmel, 24 records, fc 23, mc 1, pH 7.3 ± 0.8, 6.0–8.8)
Except one record from Calathea plurispicata, this species was recorded from only two host plants: Hedychium coronarium (PR2, 2 records, but at least ten orange-yellow plasmodia in moist chambers that did fruit could be assigned with some certainty to Physarum melleum, based upon the experience of the authors) and Psychotria poeppigiana (21 records, along with 6 plasmodia probably representing the species). Based upon the pH values recorded, this species seems unable to inhabit inflorescences of Costus and Heliconia characterized by a very basic pH. It is more common on leafy litter, both aerial and ground.
Physarum cf. pusillum (Berk. & M.A. Curtis) G. Lister, badhamioid form (PHYpus, 57 records, fc 23, mc 34, pH 8.1 ± 0.7, range 6.2–9.8)
Our specimens possess a short, rather stout stalk and a capillitium composed of a solid network of lime tubes. In accordance with the comments of Martin and Alexopoulos (1969
: 326), we tentatively include this form, although it is clearly separable from the typical form that was found more rarely, in P. pusillum. This species is known from most of the study sites, including those in cloud forests, and was recorded from most of the plants sampled. In addition to inflorescences, Physarum pusillum was recorded several times from decaying banana leaves still attached to the plants.
Physarum superbum Hagelst. (PHYsup, 26 records, fc 13, mc 13, pH 8.1 ± 0.7, range 7.1–9.2)
This species, not previously known from the Neotropics, was recorded from all countries and all study sites except CR2 and Ec2. It was represented by four additional records from aerial litter, with two of these from decaying banana leaves.
Another 18 species were encountered on decaying floral parts of living inflorescences, none of them represented by more than five records. These additional species are an apparently undescribed taxon belonging to the genus Arcyria, Comatricha lurida Lister, C. tenerrima (M.A. Curtis) G. Lister (both represented by very scanty specimens leaving the determination doubtful), Didymium difforme (Pers.) Gray, D. bahiense Gottsb., D. nigripes Fr., D. ochroideum G. Lister, Diderma effusum (Schwein.) Morgan, D. hemisphaericum (Bull.) Hornem., Lamproderma scintillans (Berk. & Broome) Morgan, Perichaena chrysosperma (Currey) Lister, P. depressa Libert, P. pedata (Lister & G. Lister) G. Lister, Physarum cinereum (Batsch) Pers., Ph. javanicum Racib., Ph. pusillum (Berk. & M.A. Curtis) G. Lister, (typical form), Ph. serpula Morgan, and Stemonitis fusca Roth. All of these species were recorded more often on other substrate types, especially aerial litter, from the same study sites.
Figure 1 shows the rank-abundance plot developed for all myxomycete records from inflorescences. Collectively, the 13 more common species in the annotated list (represented by more than five records) constitute 613 records (94% of the total number for all records).
|
All of the more common species observed in the field also were recorded from moist chamber cultures. Three species (Lamproderma arcyrionema, Perichaena cf. dictyonema and the typical form of Arcyria cinerea) were recorded only from moist chamber cultures. Of the 13 more common species, one (the dwarf form of Arcyria cf. cinerea) is apparently new to science, and three others (Perichaena cf. dictyonema, Physarum cf. limonium and Ph. superbum) would be new records for the Neotropics. Eleven of the 13 more common species were recorded from all three of the countries investigated. Accordingly, CA of the data from moist chamber cultures shows no clustering of series of samples from a particular country (Fig. 2 ).
|
Members of the Physarales (represented by 8 of the 13 more common species and 5 of the 6 species displaying a strong preference for inflorescences) constitute the majority of the myxomycetes associated with the inflorescence microhabitat, followed by members of the genera Arcyria and Perichaena, both belonging to the Trichiales (4 of the 13 more common species). Lamproderma arcyrionema is the only member of the Stemonitales represented among the 13 more common species.
Compared with other litter substrates, decaying floral parts from inflorescences have about the same levels of nitrogen but are richer in phosphorus, potassium, and several trace elements, such as magnesium, manganese, zinc, and copper (Table IV ). In contrast, inflorescences have lower levels of iron and aluminum. In addition, decaying floral parts of inflorescences exhibit very basic pH values (Table II ). The mean pH for the substrates of the 358 moist chamber cultures prepared with inflorescences was 8.08 ± 0.04 as opposed to 7.00 ± 0.02 for the 696 cultures prepared with samples of leafy litter from the same study sites. The very frequent occurrence of myxobacteria, which are known to be predatory on other bacteria, suggests the presence of a rich bacterial flora. Various myxobacteria were recorded 401 times in the 358 moist chambers prepared with inflorescences (75% of all cultures positive), whereas the 696 moist chambers prepared with other kinds of litter yielded 456 records of myxobacteria (51% positive cultures). Three species of myxobacteria with relatively large fructifications were especially more productive in moist chambers prepared with inflorescences. Chondromyces crocatus produced an average of 128 ± 13 fructifications (from 153 records) in cultures with inflorescences, compared with 93 ± 20 fructifications (44 records) on other kinds of litter. The differences for C. lanuginosus were 80 ± 23 fructifications (10 records) on inflorescences as compared to 29 ± 8 fructifications (32 records) on other litter, whereas Stigmatella aurantiaca produced a mean of 161 ± 50 fructifications (41 records) on inflorescence as compared to 99 ± 12 fructifications (122 records) on other litter.
|
A second biplot combining the means of all sample scores belonging to one series of moist chamber cultures (i.e., prepared with material from the same plant species at the same locality) with the environmental parameters is provided in Fig. 3b . Most of the plant species investigated cluster together, with Heliconia velligera (Ec4, series 12), Psychotria poeppigiana (Ec4, series 13), Calathea ischnosiphonoides (Ec2, series 17) and Hedychium coronarium (PR2, series 19) as the exceptions. Comparing the standard errors obtained by calculating the means of the sample scores of all cultures belonging to one series, some species of Heliconia with large inflorescences (e.g., H. mariae, series 2 and 7; H. pogonantha, series 3) tend to have the most uniform myxomycete assemblages.
Physarum compressum, as the most common species encountered in the entire survey, has a clear preference for very basic substrates, as indicated by the comparison of moist chamber cultures positive for this species with the total for cultures prepared with all kinds of litter substrates (Fig. 4a ). In cultures of material from inflorescences, this species is significantly more productive than in cultures prepared with other litter substrates (Fig. 4b , Table III ). Psychotria poeppigiana, the only dicot investigated in this study, had a clearly lower pH (mean 6.6) than all of the monocots. Physarum melleum was the most common species on this plant, and except for Ph. compressum (one record) and Ph. pusillum (two records), none of the other inflorescence-preferring myxomycetes was found on this substrate.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). The information reported herein adds yet another aspect to these "microcosms"—the existence of a stable and specific assemblage of "floricolous" myxomycetes on decaying floral parts of inflorescences, especially those of giant herbaceous plants belonging to the order Zingiberales. Several lines of evidence support this, since (i) this assemblage was found on several plant species and in eight localities throughout the Neotropics, with 11 of the 13 more common myxomycetes occurring in all three investigated countries, (ii) no evidence of clustering of series of moist chamber cultures from a certain locality was found in the CA shown in Fig. 2
, (iii) a few myxomycete species, especially Physarum compressum and Ph. didermoides, are exceedingly common in the inflorescence microhabitat, and (iv) six myxomycete taxa showed a clear preference for this microhabitat. This assemblage seems to be stable over years, as suggested by repeated field surveys of stands of Heliconia mariae in La Selva (March 1999 and 2000) and Cahuita National Park (July 1999, April 2001). As indicated by the data presented in the rank-abundance plot (Fig. 1 ), the results from the field survey and moist chamber culture components of this study complement each other, and the high number of rare "tail" species in comparison to the few very common species suggests a rather exhaustive survey for the plant species investigated. For the moist chamber component of the study, the bootstrap method estimates the survey to be complete to 90% with a simple saturation model (31 species to be expected, 28 found in reality). The jackknife estimate as a nonparametric model results in a degree of completeness of 78% (36 species to be expected). However, as to be expected for organisms capable of long-distance dispersal via airborne spores, a rather high number of species should be very rare, since they may develop occasionally from spores carried into the "microcosm". Accordingly, a saturation model extended by a linear term for these species provided a significantly better fit to the data and gave an estimate of 19 species to occur regularly, with 22 species recorded more than once in reality. This would equal a survey complete to 86% for the species found more than once. Therefore, it can be assumed that the present study revealed at least four fifths of all myxomycete species to be expected for the surveyed plant species. However, as shown by the examples of Hedychium coronarium and Psychotria poeppigiana, which do not cluster with the other plant species in the CA, new plant species, especially those not belonging to the order Zingiberales, may carry additional myxomycete species.
The assemblage of myxomycetes associated with inflorescences is rather poor in species. A comparison with the myxomycetes recorded from other litter substrates shows that more than half of the 13 more common species are ubiquists, occurring with about the same frequency or even more often on other litter substrates (Table III ). Prominent examples are Didymium squamulosum and D. iridis, both very common on all kinds of aerial litter. Both species are several times more common on other litter substrates than on inflorescences. On the other hand, six species of myxomycetes, including two taxa (Physarum compressum and Ph. didermoides) that are very common in this microhabitat, show a clear preference for inflorescences. The preference values derived from both frequency and productivity in moist chamber cultures (Table III ) reveal two clear-cut groups of species. These are ubiquists, with preference values well below 5, and specialists, with values of 10 and more. This pattern is confirmed by the experience of the authors from several years of rather intense field surveys in the Neotropics (see comments in the annotated species list).
One obvious feature of the assemblage of myxomycetes associated with inflorescences is the prevalence of a few genera, especially Perichaena and Physarum, among the 13 more common species, but especially among the six species that seem to display the strongest preference for this microhabitat. These genera seem to have a rather high proportion of species specialized to utilize basic substrates. A study from a desert in Kazakhstan (Schnittler and Novozhilov 2000
), a totally different habitat but also with a high proportion of basic substrates, revealed a similar high proportion of these two genera among the assemblage of species recorded (among the 27 species recorded where 4 members of the genus Perichaena and 5 species of Physarum).
The results from CCA (limited to the moist chamber component of this study), when considered along with the data obtained from the field survey, help to provide an answer to the question of why inflorescences constitute a special microhabitat for myxomycetes in the Neotropics. The most obvious factor is the very basic pH of the decaying floral parts of almost all of the inflorescences investigated (Table II
). Except for Psychotria poeppigiana (mean pH 6.6) all other plants had mean pH values above 7.5. In the compact inflorescences of Heliconia mariae, values between 8 and 9 were recorded. As such, inflorescences appear to represent some of the most basic substrates available for myxomycete growth. Only the fleshy, herbaceous stems of species of Heliconia or Musa, essentially other portions of the plants producing the inflorescences that were investigated, display similar pH values (e.g., a series of moist chambers prepared from aerial banana leaf litter collected in study site Ec4 had pH values between 7.4–8.0). Not unexpectedly, the few records of the six taxa of inflorescence-preferring myxomycetes came from other substrates exhibiting similar rather high pH values. These included the fleshy sheaths of the palm Chamaedorea tepejiote and, even more often, aerial litter of Heliconia or Musa. In accordance with this, pH is the environmental parameter explaining most of the variance in species distribution in the CCA (Fig. 3a
), and the two most common myxomycetes clearly exhibit a preference for high pH and cluster with the environmental parameter pH. From the 13 more common myxomycete taxa, only those with a mean pH > 8.0 for all records had preference factors for inflorescences higher than 2.5 (Table III
). The only other known ecological group of myxomycetes with an apparent preference for higher pH values are succulenticolous species, found on remnants of cacti in arid regions (Lado et al 1999
), where pH values in a similar range as given for inflorescences are recorded (Blackwell and Gilbertson 1984
).
In tropical forests, Physarum compressum seems to be a good indicator species for very basic microhabitats. Perhaps due to the higher contents of some nutrients in the inflorescences, the colonies of this species are larger on this substrate and indicate a pH optimum between 8.5 and 9.0 for the species (Figs. 4a, b ).
The largest and most compact inflorescences have the highest pH values and the most consistent assemblages of myxomycetes, as indicated by the smaller standard errors of their sample scores in the second CCA biplot (Fig. 3b ). Accordingly, Costus guanaiensis and Heliconia mariae have the highest values of inflorescences positive for myxomycetes in the field (Table II ). Physarum didermoides, the inflorescence-preferring species with the highest productivity, was found most often on these two plants (16 and 43 of the 66 field records, respectively).
A description of flower features, pollination ecology, and seed dispersal of Costus guanaiensis, as observed at the Maquipucuna reserve in Ecuador (site Ec1), may illustrate the function of inflorescences as a special microhabitat for myxomycetes. The plant is an herb up to 4 m tall and produces a single flower head on the end of a shoot. The inflorescence is a cylindrical, cone-like spike with 50–120 densely imbricate, persistent, sturdy, green bracts with a scarlet-red margin. Each bract is 3–4 cm long and supports one flower. The corolla is pale orange, up to 6 cm long, and each flower blooms only one day. A hummingbird (Phaethornis yaruqui, white-whiskered hermit) was seen every morning visiting the 1–3 flowers in bloom at each inflorescence. By the following day, the decay of the corolla had begun. Its remnants were enclosed by the tightly appressed bracts, thus creating a "natural moist chamber." A nectarial "callus" (extrafloral nectary) at the base of the bract provides a sugar-containing solution (to support ants that protect the inflorescence), perhaps ensuring optimal conditions for a community of yeasts and bacteria degrading the corolla remnants. Myxobacteria, indicating a community rich in other bacteria as well, developed quickly in moist chamber cultures prepared with corolla remnants (Chondromyces crocatus as the most common species, followed by C. lanuginosus, Stigmatella aurantiaca, and Myxococcus sp.). The flowering sequence in the inflorescences is ascending, with the terminal flowers the last to bloom. With often more than 75 flowers in one inflorescence, a flowering period of 2–3 mo can be assumed. Judged from the flowering sequence, 3–5 wk after bloom the myxomycetes begin to form fructifications on the outer surface of the still living bracts. Here, the fructifications can easily dry out, which allows the spores to become airborne. At this time, the uppermost bracts still produce flowers, whereas the lowermost bracts are already open, exposing the white seed capsules with black seeds for bird dispersal. The yellow-rumped tanager (Ramphocelus icteronotus) was seen regularly plundering these capsules. With a dry weight of 105 ± 9 mg (n = 7) per corolla and 50–120 flowers per inflorescence, 5–12 g of substrate are available for myxomycete growth. On the 16 inflorescences with colonies of Physarum didermoides observed in the field, numbers ranging from 15 to 1100 sporocarps were recorded (mean 219 ± 87). With an estimated number of 105 spores per sporocarp, on average more than two millions spores per inflorescence were available for dispersal.
Based upon our field observations, as well as results from CCA, inflorescences with the following features provide the best conditions for myxomycetes: (i) rather massive and compact, (ii) possessing large, slightly fleshy corollas, as is often the case for flowers pollinated by birds or bats, (iii) with sturdy, tightly appressed bracts that appear to create "natural moist chambers" for the corolla parts, (iv) having an extended flowering period that provides enough decaying material to support large colonies of myxomycetes, and (v) with submersed flowers in water-filled or saturated bracts that leak remaining nectar, or the presence of extrafloral nectaries that secrete a sugar solution to provide initial support for a community of yeasts and bacteria associated with the decomposition of the floral parts. Extrafloral nectaries can be found in many species of Costus and in all species of Heliconia (Kubitzki 1998
). For some species of Heliconia pollinated by birds, one flower has already accumulated by the time of anthesis (usually the early morning) more than 100 µL of nectar with a concentration of 15–20% sucrose-equivalents (Pedersen and Kress 1999). Especially in species with erect, massive bracts collecting rainwater, the nectar leftovers not used by the pollinator leak out and provide a medium rich in nutrients for yeasts, which are likely to grow in the watery solution enriched with nectar and the decaying floral parts.
The general pattern of decreasing species richness of myxomycetes with increasing elevation in the Neotropics, reported by Stephenson et al (1999)
and Schnittler and Stephenson (2000)
, also seems to apply to the assemblages of myxomycetes associated with inflorescences. Most of the plant species producing inflorescences suitable as potential substrates can be found only at lower elevations. In Ecuador, the highest site (Ec3) had no species of Heliconia or Calathea, and the mid-elevation site (Ec2) had only a single species (Calathea ischnosiphonoides, with a very small inflorescence). Only the low elevation site (Ec1) had a variety of large species of Heliconia. The three species of myxomycetes (Physarum didermoides, Ph. superbum, and Perichaena cf. dictyonema) with the highest preference values for inflorescences were not observed in the field or seen in moist chamber cultures from the high elevation sites (CR2 and Ec2). The highest elevation site in Puerto Rico (PR2) produced only P. superbum, which was recorded from a population of the introduced Hedychium coronarium growing in an open habitat (the margin of a road). In the CCA (Fig. 3
), these myxomycete species cluster in the opposite direction of the environmental parameters elevation and mean annual rainfall.
The high preference values for the six species of inflorescence-inhabiting myxomycetes (Physarum didermoides was never recorded by us from any other microhabitat) raise the question of whether they are dispersed by the pollinating or seed-consuming animals that visit these same inflorescences. More evidence, especially direct evidence of viable spores on insects or feathers or bills of pollinating birds, would be necessary to answer this question. Ph. didermoides seems to be restricted largely to the species of Costus and Heliconia with orange or red bracts that are bird-pollinated (Kubitzki 1998
). In the present study, the only exceptions were a few records from moist chamber cultures prepared with flowers from a population of Calathea plurispicata in close vicinity to a stand of Costus guanaiensis in Ecuador (site Ec1). The investigated species of Calathea have green or brown bracts and cream to yellow flowers and are usually pollinated by euglossine bees (Kennedy 1978
). However, the species of hummingbird mentioned above was also observed to make occasional visits to the flowers of Calathea. Although hummingbirds were observed repeatedly on the flower heads of Psychotria poeppigiana with its conspicuous, scarlet-red bracts, Ph. didermoides was not recorded on this plant. The probable reason is the low pH value for the inflorescences of this plant (Table II
). Another open question is the occurrence and dispersal way of yeasts and bacteria in the inflorescences. Although the genera Costus and Heliconia have not been investigated, flowers in general seem to be a habitat for specific yeasts (Lachance et al 1998
, Rosa et al 1999
). Probably attracted by leaking nectar, ants were observed at inflorescences of most of the plant species investigated in this study and could function as a vector for micro-organisms. It seems quite possible that more detailed studies of this micro-ecosystem, being a well-circumscribed "island" with its own distinctive community of microorganisms, including myxomycetes, bacteria, and perhaps yeasts, will add a new story of interspecies relationships to the ecological web of Neotropical forests.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Accepted for publication May 10, 2001.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Blackwell M, Gilbertson RL., 1984 Distribution and sporulation phenology of myxomycetes in the Sonoran desert of Arizona Microbial Ecology 10:369-377
Brown S, Lugo AE, Silander S, Liegel L., 1983 Research history and opportunities in the Luquillo Experimental Forest New Orleans, Southern Forest Exp. Station: USDA Forest Service, Tech Rep SO-44. 128 p
Clark J, Mires A., 1999 Biosystematics of Didymium: the noncalcareous, long-stalked species Mycotaxon 71:369-382
Heltshe JF, Forrester NE., 1983 Estimating species richness using the jackknife procedure Biometrics 39:1-11[Medline]
Holdridge LR, Grenke WC, Hatheway WH, Liang T, Tosi JA Jr., 1971 Forest environments in tropical life zones: a pilot study Oxford, New York: Pergamon Press. 747 p
Irawan B, Clark J, Stephenson SL., 2000 Biosystematics of the Physarum compressum morphospecies Mycologia 92:884-893
Jørgensen PM, León-Yánez J, eds 1999 Catalogue of the vascular plants of Ecuador St. Louis: Missouri Botanical Garden Press. 1181 p
Kennedy H., 1978 Systematics and pollination of the "closedflowered" species of Calathea (Marantaceae) Univ Calif Publ Bot 71:1-90
Kubitzki K, ed 1998 The families and genera of flowering plants. Vol. IV Alismatanae and and Commelinanae (except Gramineae). Berlin-Heidelberg-New York: Springer. 511 p
Lachance MA, Rosa CA, Starmer WT, Schlag-Edler B, Barker JSF, Bowles JM., 1998 Wickerhamiella australiensis, Wickerhamiella cacticola, Wickerhamiella occidentalis, Candida drosophilae and Candida lipophila, five new related yeast species from flowers and associated insects Int J Syst Bacteriol 48:1431-1443
Lado C, Mosquera J, Beltran Tejera E., 1999 Cribraria zonatispora, development of a myxomycete with unique spores Mycologia 91:157-165
Lado C, Pando F., 1997 Flora Mycologia Iberica, Vol. 2 Myxomycetes, I. Ceratiomyxales, Echinosteliales, Liceales, Trichales. Madrid-Berlin-Stuttgart: Kramer. 323 p
Martin GW, Alexopoulos CJ., 1969 The Myxomycetes Iowa City: University of Iowa Press. 560 p
McDade LA, Bawa KS, Hespenheide HA., 1994 La Selva: ecology and natural history of a Neotropical rain forest Chicago: University of Chicago Press. 486 p
Mitchell DW., 1999 World inventory of myxomycete species Descriptions and taxonomic references. CD-ROM, distributed at the Third International Congress for Systematics and Ecology of Myxomycetes, held in Beltsville, Maryland
Nadkarni NM, Weelwright NT., 2000 Monteverde—ecology and conservation of a tropical cloud forest New York-Oxford: Oxford Univ. Press. 573 p
Pedersen LB, Kress WJ., 1999 Honeyater (Meliphagidae) pollination and the floral biology of Polynesian Heliconia (Heliconiaceae) Pl Syst Evol 216:1-21
Reichenbach H., 1993 Biology of the Myxobacteria: Ecology and Taxonomy In: Dworkin M, Kaiser D, eds. Myxobacteria II. Washington, DC: American Soc Microbiol. p 13–62
Richardson BA., 1999 The Bromeliad microcosm and the assessment of faunal diversity in a Neotropical forest Biotropica 31:321-336
Rosa CA, Lachance MA, Starmer WT, Barker JSF, Bowles JM, Schlag-Edler B., 1999 Kodamaea nitidulidarum, Candida restingae and Kodamaea anthophila, three new related yeast species from ephemeral flowers Int J Syst Bacteriol 49:309-318
Schnittler M., 2001 Foliicolous liverworts as a microhabitat for Neotropical Myxomycetes Nova Hedwigia 72:259-270
Schnittler M., Novozhilov YK., 2000 Myxomycetes of the winter-cold desert in western Kazakhstan Mycotaxon 74:267-285
Schnittler M., Stephenson SL., 2000 Myxomycete biodiversity in four different forest types in Costa Rica Mycologia 92:626-637
Stephenson SL, Landolt JC, Moore DL., 1999 Protostelids, dictyostelids, and myxomycetes in the litter microhabitat of the Luquillo Experimental Forest, Puerto Rico Mycol Res 103:209-214
Ter Braak CJF., 1987 Ordination In: Jongman RHG, Ter Braak CJF, Van Tongeren OFR, eds. Data analysis in community and landscape ecology. Wageningen, Netherlands: Pudoc. p 91–173
Ter Braak CJF., 1988 Canoco: a Fortran program version 2.1 for (partial) (detrended) (canonical) correspondence analysis, principal components analysis and redundancy analysis Report LWA-88–02. Wageningen, Netherlands: Agricultural, Mathematics Group. 95 p
This article has been cited by other articles:
 |
|
 |
|
  C. Rojas and S. L. Stephenson Distribution and ecology of myxomycetes in the high-elevation oak forests of Cerro Bellavista, Costa Rica Mycologia, July 1, 2007; 99(4): 534 - 543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. C. Ruivo, M.-A. Lachance, C. A. Rosa, M. Bacci Jr, and F. C. Pagnocca Candida heliconiae sp. nov., Candida picinguabensis sp. nov. and Candida saopaulonensis sp. nov., three ascomycetous yeasts from Heliconia velloziana (Heliconiaceae). Int J Syst Evol Microbiol, May 1, 2006; 56(Pt 5): 1147 - 1151. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Schnittler, M. Unterseher, and J. Tesmer Species richness and ecological characterization of myxomycetes and myxomycete-like organisms in the canopy of a temperate deciduous forest. Mycologia, March 1, 2006; 98(2): 223 - 232. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Stephenson, M. Schnittler, and C. Lado Ecological characterization of a tropical myxomycete assemblage--Maquipucuna Cloud Forest Reserve, Ecuador Mycologia, May 1, 2004; 96(3): 488 - 497. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Snell and H. W. Keller Vertical distribution and assemblages of corticolous myxomycetes on five tree species in the Great Smoky Mountains National Park Mycologia, July 1, 2003; 95(4): 565 - 576. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
