Mycologia, 98(2), 2006, pp. 218–222. # 2006 by The Mycological Society of America, Lawrence, KS 66044-8897
Protostelids from tropical forests, woodlands and deserts in Australia Donna Moore Powers1
INTRODUCTION
Biology/Chemistry Division, Corning Community College, Corning, New York 14830
Protostelids are a group of simple eumycetozoans (slime molds) with amoeboid trophic cells and simple fruiting bodies in their life cycles. Together with two other groups of slime molds, the myxomycetes and dictyostelids, they make up the taxonomic group Eumycetozoa (Olive 1975, Spiegel et al 1995). Slime molds are eukaryotic, phagotrophic bacteriovores that probably have an important role in the regulation of the populations of bacteria present in soils and other microhabitats in terrestrial ecosystems (Feest 1987). Our understanding of protostelid ecology has been enhanced greatly by recent sampling in a variety of ecosystems, including boreal forests and tundra of Alaska, temperate forests and grasslands of Arkansas and tropical forests of Puerto Rico, Costa Rica and Hawaii. As biogeographical data accumulate, patterns of distribution are becoming evident. In temperate regions of the world, species tend to be partitioned among aerial litter and ground litter microhabitats, with a particular taxon being relatively more abundant in one type of microhabitat than the other (Moore and Spiegel 2000a, b). However, in tropical forests of Puerto Rico and Costa Rica, species of protostelids commonly associated with the aerial litter microhabitat in temperate regions were rarely encountered (Moore and Spiegel 2000c, Moore and Stephenson 2003). In contrast species associated with ground litter microhabitats in temperate regions were present and abundant on both aerial litter and ground litter in Puerto Rico, but this was not the case in Costa Rica, where one temperate ground litter species still preferred ground litter. Species diversity and richness of protostelids vary among different types of ecosystems. Temperate habitats in Arkansas exhibit the greatest species diversity and richness reported to date (Moore and Spiegel 2000a), with tropical forests (Subtropical Wet, Subtropical Rain and Lower Montane Rain Forests sensu Holdridge [1967]) in the mountains of Puerto Rico (Moore and Spiegel 2000c) next, and boreal forests and tundra of Alaska (Moore et al 2000) characterized by the lowest values. Research in Tropical Wet and Tropical Moist Forests (Holdridge 1967) of Costa Rica (Moore and Stephenson 2003) indicated a level of species richness (10 species) similar to that found in Puerto Rico (10 species). Whether these trends are the result of differences in latitude, vegetation, particular microhabitats, or a combination of all these factors, is not yet known. The objectives of
Steven L. Stephenson Department of Biological Sciences, University of Arkansas, Fayetteville, Arkansas 72701
Abstract: Most of what is known about the distribution of protostelids is limited to results from surveys carried out in North and Central America. To increase our knowledge about protostelid diversity and distribution we surveyed protostelids from 12 study sites in northern Queensland and the Northern Territory of Australia during May–Jun 2003. Aerial litter and ground litter samples were randomly collected along a 200 m transect at each site. Study sites ranged from tropical forests to deserts. We recovered 10 species and two apparently undescribed species from samples of aerial (dead but still attached plant parts) and ground litter. Samples from a woodland site characterized by intermediate moisture conditions had the greatest species richness, followed by samples from dry woodland, tropical forest and desert sites. When species richness for a particular microhabitat was considered, samples of aerial litter yielded more species than samples of ground litter. Percentages of samples colonized with protostelids were similar for the aerial and ground litter microhabitats within a given habitat type except for dry woodlands, in which aerial litter samples were characterized by higher numbers of species than ground litter samples. Two species (Protostelium mycophaga and Soliformovum irregularis) that in temperate North America are associated with aerial litter microhabitats also were recovered from aerial litter in dry habitats in Australia. Schizoplasmodiopsis pseudoendospora, a North American temperate ground litter species, was equally abundant in aerial litter and ground litter in Australia. This study is the first of its kind for protostelid ecology in Australia and the most extensive study of protostelids in the southern hemisphere. These data complement ongoing research on protostelid distribution from around the world. Key words: desert, ecology, mycetozoans, protoctists, slime molds, tropics Accepted for publication 14 Feb 2006. 1 Corresponding author. E-mail:
[email protected]
218
POWERS AND STEPHENSON: PROTOSTELIDS FROM AUSTRALIA
219
TABLE I. Study sites from which samples for isolation of protostelids were collected in Australia. A 5 aerial litter, G 5 ground litter and IMC 5 woodland characterized by intermediate moisture conditions Precipitation Vegetation type
Locality TROPICAL FOREST Butcher’s Creek Lacey Creek Licuala Forest Mt. Lewis Davies Creek
Elevation (mm)
(m)
Complex mesophyll vine forest Complex mesophyll vine forest Mesophyll vine forest
3200
700
3000
100
3000
50
Simple notophyll vine/fern forest Simple notophyll vine forest
2000
1000
1000
1000
1000
800
2000
50
WOODLAND Davies Creek (IMC) Eucalyptus/ Allocasuarina forest Eucalyptus woodland Ellis Beach Big Mitchell Creek
Eucalyptus woodland
1000
380
Davies Creek
Tall open Eucalyptus forest Eucalyptus woodland
1000
600
1300
700
Desert shrub
,300
600
Desert shrub
,300
600
Lake Tinaroo DESERT Ross Highway Stuart Highway
the present study were (i) to determine whether the species of protostelids found in tropical forests, woodlands and deserts of Australia tend to be associated primarily with either aerial or ground litter microhabitats and (ii) to document the assemblage of protostelids for a region in the southern hemisphere not yet surveyed for these organisms. MATERIALS AND METHODS
During May–Jun 2003 10 paired aerial litter and ground litter samples were randomly collected along a 200 m transect in each of five tropical forest study sites, four dry woodland study sites, and one woodland study site characterized by intermediate moisture conditions in northern Queensland, along with two desert study sites near Alice Springs in the Northern Territory (TABLE I). The tropical study sites all were examples of species-rich, closed canopy, evergreen forests with no one species dominant. Woodland study sites were characterized by an open canopy usually consisting of various species of Eucalyptus but sometimes with an admixture of Allocasuarina and a ground layer of various grasses. The desert study
No. of species A
G
Tot. species
17u249S, 145u439E 17u519S, 146u039E 17u549S, 146u049E 16u339S, 145u169E 17u019S, 145u319E
3
3
4
Latitude/ Longitude
% plates colonized A
G
4
100
80
3
4
80
100
4
2
4
100
70
4
2
4
80
70
7
5
9
80
100
17u009S, 145u329E 16u429S, 145u389E 16u489S, 145u229E 17u009S, 145u439E 17u059S, 145u339E
7
5
9
90
100
4
3
5
80
60
7
5
9
100
80
6
3
6
100
60
5
5
7
100
80
23u459S, 133u559E 23u259S, 133u499E
2
3
4
20
60
3
3
5
50
30
sites were shrub deserts, with scattered individuals of several different species of Acacia and a ground cover consisting mostly of grasses. Names of vegetation types used herein are based on those described by Tracey (1982). Although woodlands tend to occur in areas of lower annual precipitation than is the case for tropical forests, this is not always the case, especially at the elevation extremes. For example tropical forests occurring at the highest elevations sampled in the present study were actually characterized by relatively lower precipitation totals than woodlands located near sea level. The one woodland study site (at Davies Creek) characterized by intermediate moisture conditions was identified on the basis of vegetation type and elevation rather than absolute precipitation totals. Aerial litter was defined as dead but still attached plant parts located above the ground, whereas ground litter was defined as the layer of fallen twigs, leaves, and other plant debris on the soil surface. All samples were returned to the laboratory, where five subsamples were selected from each sample and one ca 1.27 cm2 piece was cut from each. These pieces were soaked in sterile distilled water in a plastic Petri dish for 20 min and all five were plated onto a single weak malt-yeast agar plate (0.02 g malt extract, 0.02 g yeast extract, 0.75 g K2HPO4, 15–20 g agar/L of distilled water)
220
MYCOLOGIA
TABLE II. Percent occurrence of protostelids from Australian sites by habitat type and microhabitat. A 5 aerial litter, G 5 ground litter, IMC 5 intermediate moisture conditions, n 5 sample size Woodland
n
Tropical forest
Desert
Woodland (IMC)
Relative
A
G
A
G
A
G
A
G
% of 240 plates
40
40
50
50
20
20
10
10
100
33
43
74
82
0
40
80
100
56
88 48 3 23 20 13 3 0 3 5 0
23 10 3 3 3 3 5 0 3 0 0
8 16 24 6 2 4 0 4 0 0 0
2 2 18 0 2 0 2 0 2 0 0
25 10 0 5 0 0 0 0 0 0 5
0 20 5 0 5 0 0 0 0 0 0
40 70 10 0 10 20 0 20 0 0 0
0 0 10 0 10 0 10 0 10 0 0
24 19 11 6 6 4 2 2 2 1 ,1
Species Schizoplasmodiopsis pseudoendospora Protostelium mycophaga Echinosteliopsis oligospora Nematostelium gracile Cavostelium apophysatum Soliformovum irregularis Protostelium pyriformis Undescribed species A Echinostelium bisporum Nematostelium ovatum Undescribed species B Tychosporium acutostipes
(Spiegel 1990). The sample represented the replicate and the five individual pieces of plant material represented the subsamples. Observations (made with the use of the 103 objective lens of a compound light microscope) of plates began approximately 5 d after they were prepared and were continued twice a wk for 2 wk. We identified and recorded species for each piece per plate and summarized these data to generate values for frequency of occurrence of species per sample (plate) from each study site in relation to microhabitat and habitat type (TABLE II). Species richness also was compiled by study site, habitat and microhabitat type (TABLE I).
RESULTS
We recovered 10 described species and two apparently undescribed species from the samples (TABLE II). Nine species were recorded from the single woodland study site characterized by intermediate moisture conditions, the dry woodland sites yielded an average of 7.0 species, tropical forests an average of 5.8 species, and an average of 4.5 species was recorded from desert sites. For the dry woodland sites, Big Mitchell Creek yielded the highest number of species (nine) and Ellis Beach the lowest (five). For the tropical forest sites, Davies Creek had the highest species richness (nine), whereas five species were recorded from each of the other sites. The two desert sites, Ross Highway (four species) and Stuart Highway (five species) had similar species richness. For dry woodlands species richness of aerial litter samples (mean 5 5.5) was greater than that for
ground litter samples (4.0) (TABLE I). The same was true for aerial litter samples (4.4) and ground litter samples (3.0) from tropical forests. The woodland site characterized by intermediate moisture conditions (Davies Creek) yielded seven species from samples of aerial litter, the same number as recorded for Big Mitchell Creek, and sets of ground litter samples exhibited higher levels of species richness than most of the ground litter samples from tropical forests. Desert aerial litter and ground litter samples exhibited similar levels of species richness (2.5 and 3.0 species, respectively). Average species richness recorded for aerial samples (4.7) was slightly higher than the corresponding value for ground litter samples (3.5), regardless of whether they were from tropical forest, dry woodland or desert sites. If desert aerial and ground litter samples are removed from this comparison, aerial litter samples had a mean of 5.1 species and ground litter samples 3.6 species. The percentage of colonized plates from dry woodlands (mean 5 95%) and tropical forests (mean 5 88%) ranged from 80% to100% whereas the percentage of colonized plates from desert sites ranged from 60% to 70% (mean 5 65%) (TABLE I). We identified protostelids from 95% of plates containing samples collected from the one woodland with intermediate moisture conditions (Davies Creek). Dry aerial samples had higher colonization (mean 5 95%) than dry ground litter samples (mean 5 70%). For desert habitats protostelids colonized 35% of aerial samples and 45% of ground litter samples.
POWERS AND STEPHENSON: PROTOSTELIDS FROM AUSTRALIA Percentages of aerial samples colonized from tropical forests (88%) and the woodland with intermediate moisture conditions (90%) sites were similar; however, ground litter samples from the latter site had higher colonization (100%) than tropical forest ground litter samples (84%). Schizoplasmodiopsis pseudoendospora was the most abundant species, colonizing 55.8% of all samples (TABLE II). Protostelium mycophaga was the second most abundant species (24.2% of all samples), followed by Echinosteliopsis oligospora (18.8%). Examination of the most abundant species by habitat type suggests that P. mycophaga and E. oligospora prefer dry and intermediate aerial litter habitats. Schizoplasmodiopsis pseudoendospora prefers moist and intermediate habitats despite microhabitat. One species, Nematostelium gracile, was more abundant in both the aerial and ground litter microhabitats of moist habitats. Soliformovum irregularis and Cavostelium apophysatum were uncommon but seemed to prefer dry aerial microhabitats. DISCUSSION
Several trends are apparent from our results. Intermediate and dry sites had higher species richness and total colonization than moist and desert sites. Within a site aerial samples had higher species richness and colonization than ground litter samples, except for the desert site. Desert aerial samples had the lowest species richness and colonization of any microhabitat and site. The trends in distribution of three species are somewhat similar to previous collections from grassland and aerial sites in Arkansas (Moore and Spiegel 2000a). Protostelium mycophaga was more abundant in dry and aerial microhabitats in Australia and in dry (grassland) sites in Arkansas. Echinostelium oligospora was uncommon in Arkansas, but when it appeared was present in mesic forest aerial microhabitats; in Australia it was most abundant from the aerial microhabitats and dry and intermediate habitats. Soliformovum irregularis is more abundant from grassland aerial and forest aerial samples than ground litter samples in Arkansas and in dry aerial samples from Australia. Schizoplasmodium pseudoendospora is an abundant species in Arkansas and prefers forest and grassland ground litter to forest and grassland aerial microhabitats. In Australia, however, it preferred moist sites to dry sites and was only slightly more abundant in the ground litter than the aerial microhabitat. Nematostelium gracile, a ground litter preferring species in Arkansas, was present in both microhabitats and most abundant in moist sites. Results from Australia contrast with trends for
221
protostelid distribution in Costa Rica and Puerto Rico. Protostelium mycophaga and Soliformovum irregularis, both rare or absent in the aerial litter and ground litter microhabitats in tropical forests of Costa Rica (Moore and Stephenson 2003) and Puerto Rico (Moore and Spiegel 2000c), were present on samples from Australia; P. mycophaga was particularly abundant on dry aerial samples. Although different sampling techniques were used in Australia than in Costa Rica and Puerto Rico, Nematostelium gracile was rare in Costa Rica, comprising just 7% of the total colonies (Moore and Stephenson 2003) whereas in Puerto Rico this species made up 19% of the total colonies (Moore and Spiegel 2000c). In Australia N. gracile was present on 11% of all samples and was more abundant on moist aerial and ground litter samples than on samples from other sites. When only the data from tropical forest sites in Australia are compared to results obtained for tropical forests in Costa Rica (Moore and Stephenson 2003) and Puerto Rico (Moore and Spiegel 2000c), a different trend is apparent. Schizoplasmodiopsis pseudoendospora was the most abundant species in all three study areas. In three of the four tropical forest study sites in Puerto Rico this species was equally abundant in the aerial litter and ground litter microhabitats. Schizoplasmodiopsis pseudoendospora was appreciably more abundant in the ground litter microhabitat of the tabonuco forest study site in Puerto Rico, which was the forest located at the lowest elevation and the driest of all the forest types examined. In the Tropical Wet Forest at La Selva in Costa Rica, this species was much more abundant on ground litter than aerial litter. In the moist forests of Australia S. pseudoendospora was equally as abundant in the aerial litter and ground litter microhabitats. Comparison of species occurrence from samples collected in Puerto Rico, Costa Rica and Australia also can be made. At lower elevations in Puerto Rico more ground litter samples were colonized than aerial samples. In a study site (a palo colorado forest) at an intermediate elevation in Puerto Rico, colonization was similar, but at higher elevations more aerial samples were colonized than ground litter samples. Work by Spiegel (Spiegel personal communication) suggests that species richness is higher in dry and submesic forests than in truly mesic forests in both Puerto Rico and Hawaii. In Australia colonization was higher for aerial samples than ground litter samples in moist, dry and intermediate moisture sites; however colonization was higher for ground litter samples in the desert site. Our data and comparisons with published information indicate that abiotic and biotic characterizations of the habitat and microhabitat are important to
222
MYCOLOGIA
understanding the biology of protostelids. Although we classify habitats as ‘‘ground litter’’ or ‘‘aerial,’’ not all ground litter or aerial microhabitats are similar to each other. Distribution data for protostelids from continents and ecosystems have contributed to our understanding of the ecology of protostelids but do not define which environmental and biotic factors affect the distribution of these organisms. Island biogeography effects may further complicate distribution trends. In Puerto Rico, an island in the Caribbean, species richness (10) was lower than in Costa Rica (17), on the mainland. In Australia, an island continent, 12 species were identified. Certainly collection effort also must be taken into consideration. In Puerto Rico four forest types, along an elevation gradient in the Caribbean National Forest, were sampled intensively over a 9 wk period with the Moore technique (Moore and Spiegel 1995). This technique uses the introduction of sterile wheat straws to collect native protostelid communities. In Costa Rica this technique was used in one forest type at the La Selva Biological Research Station for a 6 wk period. In Australia, 12 sites were sampled (four dry sites, five moist sites, one site with intermediate moisture conditions and two desert sites) using native samples that were collected, plated and counted with a standardized method. Whether species richness was affected by these differences in sampling cannot be determined. Consequently inferences relating to island biogeography and trends in protostelid species richness remain problematic. The study reported herein is the most extensive of its kind for any region of the southern hemisphere. We now know that protostelids are present and particularly abundant in the sites sampled in northern Queensland and the Northern Territory. Some Australian protostelids have microhabitat preferences similar to those in temperate mesic habitats in North America. Other species that are rare or absent in other tropical habitats are present and can be abundant in Australia. At least two apparently unknown species were identified from samples from northern Queensland. With greater sampling of habitats across Australia more species probably would
be identified. This work increases our understanding of protostelid ecology, which heretofore was based almost solely on results of surveys in the northern hemisphere.
ACKNOWLEDGMENTS
This research was supported by a grant from the National Science Foundation and was carried out at Australian Tropical Mycology Research Centre. We thank Ceridwen A. Pearce and Lindsay A. Dillon for assisting in the field and laboratory.
LITERATURE CITED
Feest A. 1987. The quantitative ecology of soil mycetozoa. Progr Protistol 2:331–361. Holdridge L. 1967. Life zone ecology. San Jose, Costa Rica: Tropical Science Center. 206 p. Moore DL, Spiegel FW. 1995. A new technique for sampling protostelids. Mycologia 87:414–418. ———, ———. 2000a. Microhabitat distribution of protostelids in temperate habitats in northwestern Arkansas. Can J Bot 78:985–994. ———, ———. 2000b. The effect of season on protostelid communities. Mycologia 92:599–608. ———, ———. 2000c. Microhabitat distribution of protostelids in tropical forests of the Caribbean National Forest, Puerto Rico. Mycologia 92:616–625. ———, Stephenson SL. 2003. Microhabitat distribution of protostelids in a Tropical Wet Forest in Costa Rica. Mycologia 95:11–18. ———, ———, Laursen GL, Woodgate WA. 2000. Protostelids from boreal forest and tundra ecosystems in Alaska. Mycologia 92:390–393. Olive LS. 1975. The mycetozoans. New York: Academic Press. 293 p. Spiegel FW. 1990. Phylum plasmodial slime molds: class protostelids. In: Margulis L, Corliss JO, Melkonian M, Chapman DJ, eds. Handbook of Protoctista. Boston, Massachusetts: Jones and Bartlett. ———, Lee SB, Rusk SA. 1995. Eumycetozoans and molecular systematics. Can J Bot 73:S738–S746. Tracey JG. 1982. The vegetation of the Humid Tropical region of north Queensland. Melbourne: CSIRO.
