Aquat Ecol (2009) 43:235–248 DOI 10.1007/s10452-008-9202-y
Life at the extreme: meiofauna from three unexplored lakes in the caldera of the Cerro Azul volcano, Gala´pagos Islands, Ecuador Daniel Muschiol Æ Walter Traunspurger
Received: 17 September 2007 / Accepted: 13 June 2008 / Published online: 9 July 2008 Ó Springer Science+Business Media B.V. 2008
Abstract On Isla Isabela, Gala´pagos Archipelago, three so far unexplored lakes were investigated in the caldera of Cerro Azul, one of the most active volcanoes in the world. The lakes face recurrent desiccation and eruption events and showed distinct differences in their water chemistry. Thirty cores from the upper 15 cm of sediment indicate distinct differences in the composition of meiobenthic communities between the lakes. In total, 27 different aquatic metazoan species could be distinguished. Numerically, rotifers dominated in two of the lakes, with mean densities up to 4.56 9 106 individuals m-2 while the third lake was dominated by a gastrotrich of the genus Chaetonotus (0.67 9 106 individuals m-2). The largest lake harboured up to 14.4 9 106 nematodes m-2, which is the highest nematode density thus far reported for a freshwater habitat. The lakes yielded few nematode species (S = 7, N = 887) and calculation of the Shannon– Wiener index (H0 ) indicated an exceptionally low nematode diversity. The nematode community of one lake was clearly dominated by an undescribed suction-feeding Mesodorylaimus (59.6%), the community of the other lake by the epistrate feeder Achromadora pseudomicoletzkyi (89.3%), whereas the third lake surprisingly contained no nematodes. D. Muschiol (&) W. Traunspurger Animal Ecology, University Bielefeld, Morgenbreede 45, 33615 Bielefeld, Germany e-mail:
[email protected]
The benthic nematode biomasses for the two nematode-containing lakes differed by a factor 50. The food webs of the three lakes are presumed to have an exceptionable simply structure. Keywords Abundance Biomass Meiobenthos Nematodes Species diversity Volcanic lakes
Introduction Since Charles Darwin’s famous visit in 1835, the Gala´pagos Archipelago has been the target of many scientific investigations. Consequently, the flora and vertebrate fauna of these islands are presently among the best-studied in the world. In the last two decades, studies on the terrestrial arthropod fauna have described more than 1,700 species (e.g. Schatz 1998; Peck 2001). However, overall it has also become apparent that the invertebrate fauna of the Gala´pagos is species-poor. Because of the long distances from the South American continent and other Pacific islands combined with rather desert-like climatic conditions, many taxonomic groups have failed to establish themselves on the islands. The aridity of the Gala´pagos Islands may preclude them as a promising field for limnological investigations, but the islands have numerous sources of surface waters and it is surprising how few studies have been carried out, especially on meiobenthic organisms. At
123
236
present, there are few published works on the inland waters of Gala´pagos; these include reports by Colinvaux (1968a, 1968b, 1976a, 1976b), Ilow and Weber (1974); Peck (1992; Peck and Balke 1993, 1994) and Ferrington and Pehofer (1996). There were no records for the whole archipelago of some meiobenthic invertebrates such as gastrotrichs, nematodes, rotifers and oligochaets (Gerecke et al. 1995) until the study of Eyualem and Coomans (1995) on nematodes. The present study focuses on the meiofauna of three lakes in the caldera of the volcano Cerro Azul, which is located on Isla Isabela, the largest island of the Gala´pagos Archipelago. After its violent eruption in 1998, several publications (Mouginis-Mark et al. 2000; Naumann et al. 2002; Rowland et al. 2003) supplemented the classic geological data (e.g. Banfield et al. 1956; McBirney and Williams 1969) on this highly active volcano. Nonetheless, no limnological survey has ever been carried out within the lakes of the caldera of Cerro Azul, probably due to the difficulty in accessing the lakes. For this reason and because essentially nothing is known about many meiobenthic freshwater taxa on the archipelago, our goal was to examine the meiofaunal community of the caldera’s three lakes. Because of their high abundance and diversity, meiofaunal organisms play an important role in many freshwater ecosystems (Schmid and Schmid-Araya 2002; Hakenkamp et al. 2002). In lakes, meiofauna can account for half of the zoobenthic carbon assimilation (Strayer and Likens 1986), yet our knowledge of the taxonomy, ecology and particularly global distribution of most freshwater meiofauna is still limited. While small organisms (bacteria, protists) tend to have a cosmopolitan distribution, endemism is largely responsible for the global species richness of large organisms (Fenchel and Finlay 2004). Thus, an interesting question is how many of the organisms with a size range in between these two extremes, namely, meiofaunal taxa, are endemic and how many are cosmopolitan. Several lines of evidence have suggested that certain meiofaunal taxa are good longdistance dispersers (Rundle et al. 2002, Robertson 2002), but our knowledge is still rather limited. For example, while free-living aquatic nematodes possess a surprising variety of dispersal strategies, few of them are fully understood (Jacobs 1984). On the remote Gala´pagos Archipelago (960 km to the next
123
Aquat Ecol (2009) 43:235–248
continental landmass), the caldera of Cerro Azul is one of the most inaccessible places. Its steep walls and extended areas with desert-like climatic conditions may restrict mechanisms of dispersal/migration of aquatic organisms to birds and wind. The remoteness of these lakes and the fact that limnic habitats on the Gala´pagos Islands were probably completely absent during the last ice age (Colinvaux 1984) provide a unique opportunity to study the dispersal of meiofaunal taxa and the intensity of the gene flow between populations of presumably cosmopolitan species. Furthermore, because the lakes in the caldera of Cerro Azul are subject to regular volcanic eruptions and droughts (see ‘‘Study site’’), they provide a possibility to examine the effects of disturbance on meiofaunal communities and the rate of recolonization after extinction events. Recent eruptions of the two neighbouring volcanoes, Cumbre (May 2005) and Sierra Negra (October 2005), suggest that a major eruption of Cerro Azul within a few years is very likely.1 This event would make Cerro Azul an ideal natural laboratory to study the mechanisms, velocity and dynamics of (re-)colonization. In this study, we focused particularly on nematodes, as they belong to the most important groups (in terms of densities and species richness) among the meiofauna of both freshwater and marine habitats (Traunspurger 2002). We aimed to answer the following questions: (1) Which meiobenthic groups occur on these little-explored lakes and in what densities? (2) Are there differences between the three lakes in terms of community patterns of whole meiobenthos and nematode species? (3) Which nematode species dominate the lakes? and (4) Is there endemism?
Materials and methods Study site Cerro Azul is an active basaltic shield volcano forming the southwestern end of Isla Isabela in the western part of the Gala´pagos Archipelago (Fig. 1). 1
Indeed, on 29 May 2008, park rangers reported signs of a new eruptive process at Cerro Azul volcano (http://www. darwinfoundation.org/en/newsroom/newsreleases/2008-05-30_ vol_fcd_png).
Aquat Ecol (2009) 43:235–248
237
Fig. 1 The three lakes at the caldera of volcano Cerro Azul, Gala´pagos Archipelago on 24 February 2003. The photograph direction is westward. Side Lake and Main Lake are surrounded
by white desiccation residues. The diameter of the tuff cone, to the right, is approximately 800 m. The inset shows the central part of the archipelago and the location of Cerro Azul
At least ten eruptions occurred between 1932 and 1998 (Naumann and Geist 2000) with an average of one eruption every 6.6 years. Above the water surface, Cerro Azul is 34 km (NW–SE) 9 22 km (SW–NE) across and rises to 1,640 m, being the second highest after Volca´n Wolf (1,710 m) in the Gala´pagos Islands. The volcano’s flanks are up to 30° steep and surmounted by a wide flat summit rim and a 450-m-deep caldera (4.2 km 9 2.2 km wide). On 24 February 2003, the date the samples were collected, the caldera floor, which covers an area of 9.5 km2 (Naumann and Geist 2000), was occupied by three unconnected lakes. As they are not yet officially named, they are referred to herein as Main, Side and Cone Lake for reasons of simplicity (Fig. 1). Their exact age is unknown, but the first reported sight of a lake within the caldera appeared in a 1988 SPOT image (Munro and Rowland 1996). The presence of three lakes was reported in 1991, 1993 and 1998. However, the few records prior to 1988 documented a completely dry caldera floor (McBirney and Williams 1969, T. Simkin, Smithsonian Institute, pers. comm.) and in 1995 Main Lake had largely evaporated while Cone Lake was still present (Naumann and Geist
2000). The 1998 eruption severely impacted on the lakes and led to enormous steam plumes, which were even documented by satellite observation (MouginisMark et al. 2000). Main Lake This was the largest lake, with a maximum diameter of *2 km. Extensive areas of white desiccation residues on the lake’s shoreline evidenced pronounced changes in surface area depending on precipitation and/or the tectonic activity of the lake floor. The maximum depth of the lake was less than a few meters. The water surface level was at 1,178 m above mean sea level, as measured by GPS device. However, Main Lake does periodically dry out entirely (see photo in Naumann and Geist 2000). Cone Lake The lake was situated in a sizable tuff cone in the northeastern corner of the caldera (Fig. 1). The conspicuously turquoise lake was deeper than Main Lake; it had a diameter of 300–400 m with a small
123
238
Aquat Ecol (2009) 43:235–248
island in the middle. Because of its steeper profile and depth, changes in precipitation that affected Main Lake had no visible effect on Cone Lake (see Naumann and Geist 2000). On 24 February 2003, the water surface level of Cone Lake was several meters above that of Main Lake. Side Lake This oval-shaped lake, the smallest of the three in the caldera, was situated in a detonation crater in the flank of the tuff cone (Fig. 1). The maximum diameter of the lake on the day of sampling was *60 m. Desiccation residues on the shoreline suggested a temporary connection with Main Lake during periods of elevated water levels. Physicochemical analyses revealed distinct differences between the three lakes. In Main Lake, conductivity, sulphate, phosphate and nitrate values were higher than in the other three lakes (Table 1). However, Side Lake had a remarkably high pH of 10.0 (measured at noon on 24 February 2003) when compared with the pH values of 8.3 in Main Lake and 8.0 in Cone Lake. Based on the 18 samples collected from Main Lake, the sediment consisted of medium-scale (up to 8 mm) cinder covered by a thick microbial mat that acquired a rubber-like consistency through the fixation process. In several locations, the lake sediment was very fluffy and rich in organic material, and in some places the sediment became hot with increasing depth, occasionally reaching [40°C at only 15 cm below the sediment surface. Another result of the recent volcanic activity was a shallow pool (diameter 1.5 m) on the Main Lake shore, in which the water temperature was *50°C and that was covered by a thick film of cyanobacteria. In contrast, the water Table 1 Physicochemical parameters and altitude of the three lakes in the caldera of Cerro Azul, Gala´pagos Archipelago Main Lake
Side Lake
Cone Lake *1,190
Altitude (m)
1,178
1,184
Conductivity (lS cm-1)
3,560
670
1,600
pH
8.3
10.0
8.0
Sulphate (mg l-1SO42-)
[1,200 300–400 [1,200
Orthophosphate (mg l-1PO43-) 0.25
0.046
0.046
Nitrate (mg l-1NO3-)
*3
*2
123
*5
temperature of Main Lake was *22°C. Below a depth of 0.5 m, patches covered by submerged macrophytes (probably Potamogeton pectinatus) occurred regularly. Within the littoral zones of the three lakes, we observed dense populations of water boatmen Trichocorixa reticulata (Corixidae) and two distinct species of dragonflies. As potential dispersal vectors of meiobenthic organisms, the presence of birds might be relevant: dozens of White-Cheeked Pintails (Anas bahamensis galapagensis), many Semi-Palmated Plover (Charadrius semipalmatus) and a Cattle Egret (Bubulcus ibis) were observed while sampling took place. Sampling and analyses On 24 February 2003, a total of 30 sediment samples from the three lakes in the caldera of Cerro Azul were collected using PVC corers. Each sample covered 5.3 cm2 of sediment surface, reaching a depth of 15 cm. Eighteen samples (six locations, three replicates each) originated from Main Lake (Fig. 1), while from each of the two smaller lakes single samples were taken at six different locations at each lake. Coordinates were recorded using a Garmin GPS device. The collected material was fixed on the same day in 4% formaldehyde. Water samples from each lake were transported to the vessel in sealed polyethylene bottles and analysed on the same day for phosphate, nitrate, sulphate and pH using portable limnological kits (Merck & Co., Inc.). Due to a breakdown of our field conductivity electrode, additional water samples were stored in a refrigerator and conductivity was determined 2 weeks later by using a YSI 3500 water-quality meter (Yellow Springs Instrument Co., Yellow Springs, Ohio). Meiofauna from the sediment was extracted according to standard methods following the protocols for density centrifugation using LudoxÒ TM-50 colloidal silica (1.14 g ml-1; mesh size 35 lm) modified from Pfannkuche and Thiel (1988). The metazoan taxa from each sample were counted under a dissecting microscope (359 magnification) after staining with Rose Bengal (300 lg ml-1) and determined to the lowest possible taxonomic level. Specimens of each taxonomic group were sent to specialists for exact determination. Ciliates retained on 35-lm mesh were counted but not classified.
Aquat Ecol (2009) 43:235–248
Permanent slides of the first 100 nematodes from each sample were prepared by embedding the specimens in glycerine (Seinhorst 1962). If the number of nematodes in the sample was below 100, all nematodes were slide-mounted. In total, 887 nematodes were identified to the species level under 1,0009 magnification (Zeiss Axioplan 2). Morphometric values of taxonomic relevance (e.g. indices of de Man) were measured under the microscope for at least ten individual males and females of each species. If fewer than ten individuals of a certain species were found in the samples, as many individuals as possible were measured. Nematode biomasses were estimated from the morphometric measurements following Andra´ssy (1956) separately for males and females of each species. Juvenile individuals were assumed to have half the length and width of an average female of the particular species. Then, total nematode assemblage biomass (in mg wet weight m-2) was calculated with respect to the individual community structure (proportion of males, females, juveniles) of each species in the particular lake. Based on morphological characteristics of the buccal cavity (presence/absence of tooth/teeth/denticles/stylet), nematode species were classified into feeding types following Traunspurger (1997). Statistical analyses Differences in total meiofauna abundance and abundances of specific meiofaunal groups between the lakes were tested by means of either t-test (in the case of nematodes) or one-way analysis of variance (ANOVA) followed by Tukey honestly significant difference (HSD) post hoc tests. Interlocation variability of samples collected from Main Lake was tested by means of Kruskal–Wallis nonparametric test because homoscedasticity of abundance data was not given in several meiofaunal groups even after loge(x + 1) transformation. These analyses were carried out using the Statistica software package (v7.0, StatSoft Inc. 2004). The composition of meiofaunal groups and specifically of the Nematoda was compared by using multivariate techniques based on Bray–Curtis similarity. The Bray–Curtis coefficient fitted best because it is not influenced by the joint absences of taxa in two samples (Clarke and Green 1988). Similarity in
239
the composition of meiofaunal groups between samples was based on matrices consisting of square-root-transformed abundance data. Similarity in the nematode community composition was based on relative abundances, as differences in nematode community structure tended to be hidden by high variations of total abundance. Nonmetric multidimensional scaling (nMDS) was applied to compare differences in community structure between the lakes. Distances in an nMDS plot are relative due to the use of ranks between samples for ordination. The relative dissimilarity between samples is reflected in the relative distances in the plot. Therefore, nMDS plots can be arbitrarily rotated, scaled, etc., and do not possess defined axes. An analysis of similarities (ANOSIM) test was applied to assess whether significant differences in the compositions of the meiofaunal groups/nematode species were present between the lakes. The ANOSIM procedure compares the ranked similarities for differences within and between groups. The resulting R-value usually lies between 0 and 1, but can range from -1 to +1. An R-value of approximately 0 suggests similar assemblages while R-values close to 1 indicate distinct assemblages (Clarke and Warwick 2001). As with standard tests, R can be significantly different from 0 with a difference too small to be important, if there are enough replicates. According to our definition, R-values larger than 0.5 indicated different assemblages between lakes. These analyses were carried out using the Primer software package (PRIMER v6, PRIMER-E Ltd (2006), Plymouth, UK). Nematode diversity was estimated using the Shannon–Wiener index (H0 ) for each lake. We used base loge for H0 calculations as suggested by Yeates and Bongers (1999). Nematode assemblage evenness (J) was estimated by dividing H0 by H0 max = log S, where S is the number of species found in each lake (Krebs 1994).
Results The three replicate samples collected in the shallow hot (*50°C) pool on the Main Lake shore contained no meiofauna apart from acarids, nematodes and rotifers at very low densities. A major part of these meiofauna organisms had apparently drifted into the
123
240
Aquat Ecol (2009) 43:235–248
pool accidentally as they showed signs of disintegration and probably had been heat-killed prior to sampling. Thus, these samples were excluded from further analysis. Accordingly, data presented for Main Lake are based on five locations with three replicates each. Meiofaunal abundance Total meiofaunal density was highest in Main Lake with 11.20 (range 0.73–42.38) 9 106 ind. m-2 because of the contribution of meiofaunal-sized ciliates (35.9%). In Side Lake, the total density reached 7.27 (range 1.04–20.98) 9 106 ind. m-2 whereas the lowest densities were found in Cone Lake with only 1.65 (range 0.47–2.53) 9 106 ind. m-2. These differences in total meiofauna abundance were statistically significant (ANOVA: P \ 0.05; Table 2). A pairwise post hoc comparison revealed no significant
Table 2 Results of one-way ANOVA and Tukey HSD post hoc tests on the differences in meiofaunal abundance between Main (Ma), Cone (Co) and Side (Si) Lakes ANOVA F(2,23)
Post hoc P
Ma Co Si
Total meiofauna
3.727
*
a
b
Ciliata
1.812
0.186 a
a
a
Rotifera
6.723
**
b
a
– a
– b
a
Nematoda Gastrotricha
-1.651a 0.115 – 3.860 * ab
Acari
23.999
ab
***
a
b
a
Annelida
5.353
*
a
a
b
Diptera: Chironomidae
0.902
0.420 a
a
a
Platyhelminthes
1.114
0.345 a
a
a
Insecta excl. Chironomidaeb
0.673
0.520 a
a
a
Copepoda
0.971
0.394 a
a
a
Cladocera
0.958
0.398 a
a
a
Ostracodab
7.558
**
b
a
a
Asterisks indicate significance levels (* \0.05; ** \0.01; *** \0.001). Lower-case letters indicate homogenous groups a
t-value of t-test between Main and Cone Lake, as Side Lake contained no nematodes
b
loge(x + 1)-transformed abundances (5.3 cm-2) were homoscedastic in all groups except insects excl. chironomids and ostracods. We used ANOVA in these groups too, because ‘‘the validity of the test and the probabilities associated with the F-ratio distribution are not affected much by violations of the assumption’’ (Underwood 1997, pp. 193–195)
123
Fig. 2 Mean abundances (ind. m-2 ± 1 standard deviation [SD]) of the meiofaunal groups found in the three lakes of Cerro Azul, Gala´pagos Archipelago on 24 February 2003. (a) Main Lake, (b) Cone Lake and (c) Side Lake (Main Lake: N = 15; three samples from a hot pool of Main Lake were excluded (see text); Cone Lake: N = 6; Side Lake: N = 6)
differences between Main Lake and Side Lake and between Side Lake and Cone Lake, and a significant difference between Main Lake and Cone Lake (Table 2). On the level of specific meiofaunal organism groups, the three lakes also showed distinct differences (Fig. 2, Table 3): among metazoans, rotifers were abundant in Side Lake (4.56; range 0.60– 9.94 9 106 ind. m-2) and in Main Lake (3.82; range
Aquat Ecol (2009) 43:235–248
241
Table 3 Mean relative abundance (%) and standard deviation (SD) of metazoan meiobenthos in the three lakes in the caldera of Cerro Azul, Gala´pagos Archipelago Main Lake
Side Lake
Cone Lake
Rotifera
37.2 (±27.5) 82.9 (±15.0) 18.3 (±26.3)
Acari
20.7 (±24.8) 14.8 (±13.1)
Nematoda
19.5 (±20.2)
0.2 (±0.6)
11.9 (±16.3)
Gastrotricha
9.0 (±23.8)
0.0 (±0.0)
47.7 (±42.3)
Diptera: Chironomidae
6.1 (±9.9)
1.5 (±1.7)
2.4 (±1.7)
Annelida Platyhelminthes
4.7 (±8.5) 1.8 (±2.6)
0.0 (±0.0) 0.2 (±0.4)
10.6 (±11.5) 2.5 (±3.7)
Insecta excl. Chironomidae
0.8 (±1.6)
0.4 (±0.9)
0.2 (±0.3)
Cladocera
0.2 (±0.3)
0.0 (±0.0)
2.1 (±5.2)
Ostracoda
0.1 (±0.4)
0.0 (±0.0)
2.2 (±2.9)
Copepoda
0.1 (±0.3)
0.0 (±0.0)
0.4 (±0.5)
1.8 (±1.4)
Values represent averaged relative compositions of unpooled samples
0.01–19.14 9 106 ind. m-2) but showed significantly lower densities in Cone Lake (0.16; range 0.01–0.75 9 106 ind. m-2; Table 2). Nematodes were the next dominant group in Main Lake, reaching mean abundances of 2.44 (range 0–14.43) 9 106 ind. m-2 (Fig. 2). In Cone Lake, a much lower average density of 62,150 (range 5,650–139,380) ind. m-2 was found, whereas nematodes were completely missing in five of six samples from Side Lake. Within the sixth sample we found a single juvenile specimen of Monhystrella lepidura Andra´ssy, 1968. Because nematodes were virtually missing in Side Lake, we restricted statistical analysis in this group to Main Lake and Cone Lake (t-test). The difference was statistically nonsignificant (P = 0.155; Table 2) due to the extreme variability in Main Lake’s nematode abundances (range 0–14.43 9 106 ind. m-2). However, the exclusion of the single sample without nematodes in Main Lake from the analysis resulted in a significant P-level (t = -2.228; P \ 0.05). Another dominant group was represented by the gastrotrichs in Cone Lake (0.67; range 0–2.06 9 106 ind. m-2; Fig. 2). In Main Lake, they reached a mean abundance of 0.36 (range 0–3.35) 9 106 ind. m-2, while they were completely missing in Side Lake. Statistically, Main Lake’s gastrotrich abundance did
not differ from that of the two other lakes, whereas a significant difference was found between Side Lake and Cone Lake (Table 2). In contrast, acarids showed high abundances in Side Lake (0.35; range 0.17–0.60 9 106 ind. m-2) and Main Lake (0.25; range 0.01–0.90 9 106 ind. m-2) but not in Cone Lake (15,068; range 0–35,790 ind. m-2). Unfortunately, on closer examination, it became apparent that an unknown part of the acarids was accounted for by exoskeletons of the surfacedwelling mite Hydrozetes lemnae Coggi, 1899, so that the actual significance of this group in the three lakes remains unclear. While the abundances of many meiofaunal taxa differed significantly between the three lakes (Table 2), differences between sampling locations within Main Lake (five locations, three replicates each) were nonsignificant (Kruskal–Wallis: P [ 0.1) for all meiofaunal taxa apart from rotifers (Kruskal– Wallis: P \ 0.05). In total 27 different aquatic metazoan species (excluding rotifers, Table 4) could be distinguished with the help of specialists (see ‘‘Acknowledgements’’). Do meiofaunal and nematode assemblages differ between the lakes? The results of the multivariate analysis nMDS and ANOSIM showed few differences between the meiofaunal communities of the lakes. The nMDS plot showed only weak groupings of the replicates (Fig. 3), and the global ANOSIM test indicated little or no separation of the three lakes’ meiofaunal communities (R = 0.35, P \ 0.01). However, a pairwise ANOSIM test between Cone Lake and Side Lake revealed distinct differences between them (R = 0.79, P \ 0.01). A species-poor nematode assemblage was found in all three lakes (Main Lake: six species, Side Lake: one species, Cone Lake: four species). The 887 identified nematodes were comprised of not more than seven species, four of which did not exceed relative abundances of 0.7% each. Estimation of nematode diversity revealed values of H0 = 0.83 for Main Lake (evenness J = 0.52) and H0 = 0.34 (J = 0.31) for Cone Lake. Main Lake was clearly dominated by a new, undescribed suction-feeding species of the genus Mesodorylaimus (59.6%; Table 5), followed by the
123
242
Aquat Ecol (2009) 43:235–248
Table 4 List of identified aquatic taxa in the caldera of the volcano Cerro Azul, Gala´pagos Archipelago in 2003 Nematoda
Mesodorylaimus sp.; Udonchus cf. merhatibebi EYUALEM, 1996; Achromadora pseudomicoletzkyi VAN DER LINDE, 1938; Monhystrella lepidura ANDRA´SSY, 1968; Eumonhystera sp.; Rhabdolaimus cf. aquaticus DE MAN, 1880; Tobrilus sp.
Gastrotricha
Chaetonotus EHRENBERG, 1830 (Euchaetonotus SCHWANK, 1990) section simrothi SCHWANK, 1990
Coleoptera
Thermonectus sp.; Tropisternum sp.; Ochthebius sp.
Heteroptera
Trichocorixa reticulata GUERIN-MEMEVILLE, 1857
Odonata
Aeshnidae; Libellulidae
Diptera
Limonia sp.; Tanytarsus sp.; Chironomus sp.; Goeldichironomus holoprasinus GOELDI, 1905
Acari Platyhelm.
Hydrozetes lemnae COGGI, 1899 Gyratrix hermaphroditus EHRENBERG, 1831
Copepoda
Metacyclops sp.
Ostracoda
Cypridopsis vidua MU¨LLER, 1776
Cladocera
Alona sp.; Latona sp.
Annelida
Pristina aequiseta BOURNE, 1891; Aulophorus furcatus MU¨LLER, 1774; Aeolosoma sp.; Tubificidae
Table 5 The benthic nematode biomass (BM) in mg wet weight m-2 sediment surface, the relative contribution to biomass by nematode species (% BM) and the relative number of individuals (% ind.) Main Lake BM
Cone Lake
% BM % Ind. BM
% BM % Ind.
Mesodorylaimus 835.9 88.0
59.6
3.2 16.7
4.5
Achromadora Monhystrella
5.7 5.8
3.9 35.3
15.6 81.3 0.3 1.6
89.3 5.7
0.5
1.2
Othersa
54.3 55.2 \5
\0.1
0.5
0.5
a
‘Others’ consists of Eumonhystera, Udonchus and Tobrilus in Main Lake and Rhabdolaimus in Cone Lake. Full scientific designations of the species are given in Table 4
Fig. 3 nMDS plot (stress: 0.16) of the meiofaunal community composition. The Bray–Curtis similarity was calculated from the square-root-transformed abundance data (taxonomic resolution as in Fig. 2; Main Lake: N = 15; Cone Lake: N = 6; Side Lake: N = 6)
deposit feeder Monhystrella lepidura Andra´ssy, 1968 (35.3%). Achromadora pseudomicoletzkyi Van Der Linde, 1938 accounted for 3.9% of the nematode
123
fauna of Main Lake (Table 5) whereas three more species (Udonchus cf. merhatibebi Eyualem, 1996; Eumonhystera sp. and Tobrilus sp.) combined contributed \1.2% of total nematodes. In Cone Lake, the species composition was completely different. Here, the epistrate feeder Achromadora pseudomicoletzkyi clearly outnumbered all other nematode species. However, its dominance (89.3%; Table 5) was merely due to the fact that total nematode density in Cone Lake was much lower when compared to that of Main Lake; absolute abundances of the species in the two lakes were on the same order of magnitude. In comparison, Monhystrella lepidura (5.7%) and the new member of the genus Mesodorylaimus (4.5%) were much rarer in Cone Lake. The fourth species, Rhabdolaimus cf. aquaticus, was found at densities of 0.5%.
Aquat Ecol (2009) 43:235–248
243
Discussion
Fig. 4 nMDS plot (stress 0.02) of nematode species composition. The Bray–Curtis similarity was calculated from the untransformed relative abundance data (Main Lake: 14 replicates; Cone Lake: 6 replicates). Side Lake and one sample from Main Lake were excluded from the calculation (see text)
Main Lake sustained 50 times more total nematode biomass than Cone Lake (950 versus 19 mg m-2 wet weight; Table 5). The comparison between relative contribution to biomass and densities also showed differences (Table 5); for instance, small-sized M. lepidura accounted for 35.3% of the individuals of Main Lake’s nematode population but only accounted for 5.8% of the lake’s nematode biomass. The nematode assemblages of Main Lake and Cone Lake were compared by nMDS and ANOSIM on untransformed relative nematode species abundance data. Side Lake was excluded from the analysis as the only specimen found in the six samples was probably the result of contamination (see ‘‘Discussion’’). Additionally, one of the 15 samples from Main Lake that contained no nematodes was excluded. The nMDS plot showed an obvious grouping of the samples (Fig. 4) and ANOSIM indicated a strong separation of the nematode assemblages between these two lakes (global R = 0.91, P \ 0.001).
The three so far unexplored lakes in the caldera of Cerro Azul showed very distinct differences in the abundance and composition of their meiobenthic communities. However, uncovering the main drivers of the observed community patterns is a challenging task because of the wide variety of biotic and abiotic factors they depend on. Especially trophic interactions can be very complex and we know that food webs are more reticulate than previously assumed, particularly if micro- and meiofaunal-sized members are considered. For example, groups such as nematodes or rotifers comprise very different feeding types, ranging from deposit feeders to epistrate feeders, predators and omnivores (Traunspurger 1997; Wallace and Ricci 2002). Additionally, it is clear that the number of meiofaunal species consumed by most large macroinvertebrate predators is substantially greater than that of macrofauna prey (Schmid and Schmid-Araya 2002; Beier et al. 2004). In our samples from Cerro Azul, several potential predators of meiofauna were present (acarids, chironomids, beetles, annelids, turbellarians, ostracods, copepods and dragonflies), but our knowledge of their diets is very restricted. Thus, the functioning of an entire benthic community is quite a complex matter, dependent on numerous direct and indirect interactions that make the interpretation of observed community patterns a serious challenge. Nevertheless, from what is known about their ecology, some general conclusions and hypotheses can be formulated regarding the occurrence and density of some of the meiofauna taxa found in the caldera of Cerro Azul. Rotifers The metazoan meiobenthos of both Side and Main Lakes was dominated by extraordinary high rotifer densities (4.56 9 106 and 3.82 9 106 m-2, respectively). These densities considerably exceed any reports in the (rare) limnological studies that estimated densities per sediment surface area (SchmidAraya 1998). For example, Strayer (1985) reported 57,000 rotifers m-2 for Mirror Lake, USA, which is 80- and 67-fold lower than the respective values from Side and Main Lakes. In a 3-year study of Lake Obersee, Germany, rotifer densities ranged between
123
244
0.21 9 106 and 0.59 9 106 rotifers m-2 (Michiels and Traunspurger 2004), what is still 6- to 22-fold lower. As a result of their small size and high turnover rates, rotifers are known to transfer energy from detritus, bacteria, yeasts, algae and small protozoans to higher trophic levels (Wallace and Ricci 2002). Thus, it is very likely that these organisms play an important role in the benthic food web of the caldera lakes and would clearly deserve a closer examination than we were able to accomplish in this first survey. An intriguing task for future investigations in this or other habitats remains the elucidation of the factors that permit such high rotifer standing stocks. Since nutrients such as phosphate and nitrate were not limited (Table 1), a high benthic primary production of the lakes probably plays a role. However, productivity cannot be the sole factor that determines rotifer densities because other eutrophic habitats such as Lake Obersee should then show comparable rotifer densities. Other factors such as water conductivity or pH may be excluded because of the large differences between the two lakes (670 versus 3,560 lS cm-1, pH 10.0 versus 8.3; Table 1). What in our opinion remains the most plausible explanation for the extraordinary high rotifer densities in Side and Main Lakes is habitat disturbance. Although documentation is poor, we can assume that the lakes in the caldera of Cerro Azul regularly face heavy perturbation events. It is known that, at least in 1995, Main Lake had largely evaporated, while a volcanic eruption in 1998 literally boiled the lakes (see ‘‘Study site’’). Following such a—from the meiobenthos’ point of view—cataclysmic event, rotifers are certainly among the first organisms that recolonize the lakes. They are perfectly preadapted to this task: because many rotifers possess highly resistant dormant stages, they regularly occur in temporal habitats where they overcome periods of desiccation as dormant eggs or in the state of anhydrobiosis (Wallace and Ricci 2002). Together with their short development times, short life spans and high parthenogenetic reproduction rates, they can be classified as r-strategists and are assumed to play significant roles in habitats recovering from natural perturbation (Wallace and Ricci 2002). Thus, it appears plausible that the extraordinary high rotifer densities observed in Main and Side Lakes reflect a succession following a recent (some months) disturbance event, probably a
123
Aquat Ecol (2009) 43:235–248
complete or partial desiccation due to changes in precipitation or tectonic activity of the lake floor. During such a phase of habitat recovery, rotifers would face little inter- and intraspecific competition, as well as little predation pressure from slowerreproducing predators. These are typical conditions under which fast-reproducing r-strategists may reach extreme population densities (MacArthur and Wilson 1967). Within this framework, the moderate rotifer densities in Cone Lake also make sense: among the lakes in the caldera of Cerro Azul, Cone Lake represents the most stable habitat. Due to its steeper profile and depth, changes in precipitation that left Main Lake largely evaporated had no visible effect on Cone Lake (see Naumann and Geist 2000). Gastrotrichs In contrast to the two other lakes, Cone Lake was dominated by gastrotrichs, with mean densities of 0.67 9 106 ind. m-2 and maximum densities of 2.1 9 106 ind. m-2. These densities are higher than the 0.13 9 106 gastrotrichs m-2 reported by Strayer (1985) for Mirror Lake, but lie within the range reported by Nesteruk (1996) from Polish lakes and ponds (0.5–2.6 9 106 m-2). The overwhelming majority if not all gastrotrichs in Main and Cone Lakes belong to one species of the genus Chaetonotus that is known as microphagous. Hence, these may be an important link between the microbial loop and the higher trophic levels of the lakes. The absence of gastrotrichs in Side Lake might be explained in partly by the basic pH (Table 1). However, Balsamo and Todaro (2002) stated that freshwater gastrotrichs appear to be very tolerant to a variety of abiotic environmental factors, including pH. But these organisms are sensitive to variations in water regime and their ability during drought to maintain populations, presumably from resting eggs, varies among species. Side Lake is by far the smallest lake in the caldera and was surrounded by sizable white desiccation residues on the day of sampling (Fig. 1). Thus, it can be assumed to be sensitive to changes in precipitation and most probably shows rapid fluctuations in its water level, including regular complete desiccations. Under such highly disturbed conditions, the lack of gastrotrichs (combined with high densities of rotifers) is exactly what one would expect from what is known about their biology.
Aquat Ecol (2009) 43:235–248
Nematodes Nematodes showed very high abundances in Main Lake (up to 14.4 9 106 ind. m-2), while no other freshwater system has shown comparable values. Until now, the highest maximum densities reported for freshwater nematodes were 11.4 9 106 (individuals m-2) in Lake Vilslache, 5.4 9 106 in Lake Soiernsee (Traunspurger et al. 2006) and 3.5 9 106 in Lake Starnberger (Traunspurger 2002). The high nematode densities of Main Lake are similar to the highest values reported for terrestrial (8.3–20.0 million ind. m-2: Overgaard Nielsen 1949), marine (11.8 million ind. m-2: Rudnick et al. 1985) and estuarine habitats (0.8–22.9 million ind. m-2: Warwick and Price 1979, after Heip et al. 1985). Almost 60% of Main Lake’s nematode population consisted of Mesodorylaimus, which is a comparatively large member of the taxon. In our samples, adult females possessed a mean wet weight of 2.01 lg, which is about 20 times heavier than, for example, Monhystrella females (0.10 lg). Accordingly, it is no surprise that Main Lake’s benthic nematode biomass (950 mg m-2) ranks among the highest biomass values reported in the literature. Based on the bulk of published biomass estimates, only Lake Vilslache (2,829 mg m-2) and Traunsalpsee (1,486 mg m-2) had higher mean biomass values, whereas in Lake Pa¨a¨ja¨rvi (928–968 mg m-2) and Soiernsee (863 mg m-2) the values were similar (Holopainen and Paasivirta 1977, Traunspurger et al. 2006). What could be the reason for these extraordinary high abundance and biomass values? The new member of the genus Mesodorylaimus, which accounts for 59.6% of the individuals and 88.0% of the biomass of Main Lake’s nematode population, belongs to a cosmopolitan, mainly terricolous genus that comprises around 100 species (Loof 1999). Equipped with a roughly 15-lm-long stylet (odontostyl), it pierces its food and sucks the contents. In lakes, Mesodorylaimus is mostly found within the root system of aquatic plants (Prejs 1977), suggesting a mostly phytophagous nutrition. Yet, although Main Lake evinced patches of submerged macrophytes (probably Potamogeton pectinatus), we found the worm quite common in samples taken at places free of macrophytes. Instead, we observed numerous worms interwoven in the microbial mat that covered the sediment in many places. From what exactly it
245
nourishes there remains speculative. Occasionally we found the gut densely filled with amorphous, distinctly green material. However, this occurred only in juveniles, while adults contained amorphous brownish to colourless material. Russell (1986) reported on a species of the genus Mesodorylaimus that in vitro fed omnivorously on other nematodes, encysted amoebae, fungal hyphae, algae and even bacterial colonies. So, whatever exactly the worm’s food might be, the thick microbial mats and/or the associated fauna seem to be the key factor to the extraordinary high nematode abundances observed in Main Lake. In Cone Lake, where we observed no microbial mats, Mesodorylaimus occurred more than 500 times more rarely. Accordingly, this lake evinced a much lower benthic nematode biomass (19 mg wet weight m-2), which lies within the range commonly found in other lakes around the world (10–100 mg m-2: Traunspurger 2002). The nematode community of Cone Lake was overwhelmingly dominated by Achromadora pseudomicoletzkyi (89.3%), an epistrate feeder that probably relies on aufwuchs (e.g. diatoms, green algae) which it scrapes off light-exposed surfaces. Such a predominance of algae-feeding species in a freshwater habitat has never been found before, even in lacustrine periphyton communities (Peters and Traunspurger 2005). However, as a word of caution, the predominance of an epistrate feeder in Cone Lake is merely the result of the fact that other trophic groups that usually dominate benthic nematode communities, especially bacterial feeders, evinced low abundances—in Main Lake, Achromadora reached roughly the same population density as in Cone Lake. Within six investigated samples from Side Lake, only one sample contained a single juvenile Monhystrella lepidura. It appears probable that this one specimen was the result of sampling equipment contamination and that nematodes may have not (yet) been able to establish successful populations in Side Lake. A very recent complete desiccation of this very small lake could be the reason for this. However, many nematode species possess very short life cycles in the range of days to weeks (Muschiol and Traunspurger 2007) and thus can be assumed to be quick colonizers. Thus, the complete lack of nematodes in Side Lake is more probably the result of its high pH (10.0), as a nematicidal effect of alkaline pH levels has repeatedly been reported in the literature (e.g. Morgan and MacLean 1968; Kung et al. 1990).
123
246
Diversity The Gala´pagos Islands are famous for their large number of endemic species; for example, 83% of the islands’ terrestrial vertebrates (including birds; Jackson 1991) and between 60% and 70% of the terrestrial insects (Peck 1991) are endemic. In contrast, the macroinvertebrates of the inland waters of the archipelago show a relatively low (29%) endemism (Gerecke et al. 1995). Accordingly, only 2 of a total of 18 freshwater nematode species described by Eyualem and Coomans (1995) were endemic, which suggests that the general trend of low endemism occurs also within the meiofauna of the Gala´pagos. From our survey only one nematode species (Mesodorylaimus) proved to be new; moreover its endemism is questionable because the South American nematode fauna is insufficiently known. On the other hand, we cannot exclude the existence of cryptic species which can be difficult to detect solely on morphological characters (e.g. Derycke et al. 2005). Thus, an interesting approach for future investigations on Gala´pagos’ meiofauna will be the inclusion of genetic data to uncover possibly hidden diversity/speciation. The three lakes in the caldera of Cerro Azul had a conspicuously low species number. From the 887 identified nematodes we found not more than seven species, but effectively only three of them (A. pseudomicoletzkyi, Mesodorylaimus sp. and M. lepidura) accounted for [98% of the total nematode community in Main and Cone Lakes, in different proportions. In comparison, in a palaearctic lake surveyed with identical methods and with a comparable sample size, we expect [50 nematode species (Traunspurger 2002; Michiels and Traunspurger 2005), moreover 152 nematode species were reported from the extensively studied Lake Obersee (Michiels and Traunspurger 2004). Considering the low nematode species number observed in the caldera of Cerro Azul, it is not surprising that the values of nematode diversity (H0 = 0.34–0.83) and evenness (J = 0.31–0.52) were low. As a comparison, Michiels and Trauspurger (2005) investigated 11 alpine lakes with identical methods and similar sample sizes and found considerably higher nematode diversities H0 between 1.69 and 3.61. We also observed low species numbers for all other meiofaunal groups, which suggests that the
123
Aquat Ecol (2009) 43:235–248
assemblage structure is simple and consequently their food webs may have an exceptionally simple structure. However, the explanation of overall assemblage structure remains a serious challenge because of the immense number of factors that might play a role and which are difficult to separate from each other. Each of three factors, or even a combination of them might exclude a certain species from the lakes: (1) stage of colonization (following the 1998 volcanic eruption; desiccation events), (2) extreme environmental conditions (pH, solar irradiation) and (3) remoteness (insufficient long-distance dispersal abilities). For example, are there so few species of nematodes found because the lakes are in an early stage of colonization and more species will soon arrive (by birds, wind) from other inland waters of Gala´pagos? Or are some species outcompeted because other meiofaunal groups reach extraordinary densities due to missing predation pressure from potential predators that did not make it to the remote Gala´pagos? Or do the lakes represent a suitable habitat for many more nematode species which simply never arrived on the archipelago? It is impossible to separate these factors from each other as long as Gala´pagos’ inland water meiofauna is insufficiently known. Additionally, we urgently need more information on trophic relationships within meiofauna and between macro- and meiofauna (see Schmid and Schmid-Araya 2002 for a review on this topic). Yet, it is the comparably simple assemblage structure of the caldera’s lakes that offers the opportunity to understand connections that in other habitats would be hidden by the mass of existing interactions. Acknowledgements Work in the field was accomplished with the kind permission of the Gala´pagos National Park. Preparation for the collecting trip and the later work in the laboratory were kindly supported by the CDRS, with special thanks to Ana Mireya Guerrero, Jorge Luis Renterı´a and Dr. Alan Tye. Many thanks to the wonderful Gala´pagos crew, especially Thomas Bartolomaeus. The following persons kindly helped with the taxonomic determination of the various meiobenthic groups: Anton Brancelj (Cladocera), Ralf Deichsel (Acari), Anno Faubel (Platyhelminthes), Diana Galassi (Copepoda), Reinhard Gerecke (Acari), Alexander Kieneke (Gastrotricha), Peter Martin (Acari), Claude Meisch (Ostracoda), Nicola Reiff (Diptera), Helmut Rogg (Coleoptera, Heteroptera and Odonata), Heinrich Schatz (Acari), Tarmo Timm (Annelida) and Aldo Zullini (Nematoda). Lars Peters, Jenny M. Schmidt-Araya and three anonymous reviewers made valuable comments on earlier versions of the manuscript.
Aquat Ecol (2009) 43:235–248
References Andra´ssy I (1956) Die Rauminhalts- und Gewichtsbestimmung der Fadenwu¨rmer (Nematoda). Acta Zool Hung 2:1–15 Balsamo M, Todaro MA (2002) Gastrotricha. In: Rundle SD, Robertson A, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys Publishers, Leiden, pp 45–61 Banfield AF, Behre C, Clair D (1956) Geology of Isabela (Albemarle) Island, Archipelago de Colo´n (Gala´pagos). Geol Soc Am Bull 67:215–234. doi:10.1130/0016-7606 (1956)68[215:GOIAIA]2.0.CO;2 Beier S, Bolley M, Traunspurger W (2004) Predator-prey interactions between Dugesia gonocephala and free-living nematodes. Freshw Biol 49(1):77–86. doi:10.1046/j.13652426.2003.01168.x Clarke KR, Green RH (1988) Statistical design and analysis for a ‘‘biological effects’’ study. Mar Ecol Prog Ser 46:213– 226. doi:10.3354/meps046213 Clarke KR, Warwick RM (2001) Changes in marine communities: an approach to statistical analysis and interpretation. PRIMER-E, Plymouth Colinvaux PA (1968a) Paleolimnological investigations in the Gala´pagos Archipelago. Not Galapagos 11:13–18 Colinvaux PA (1968b) Reconnaissance and chemistry of the lakes and bogs of the Gala´pagos Islands. Nature 219:590– 594. doi:10.1038/219590a0 Colinvaux PA (1984) The Gala´pagos climate: present and past. In: Perry R (ed) Galapagos. Pergamon Press, Oxford, pp 55–69 Colinvaux PA, Schofield EK (1976a) Historical ecology in the Gala´pagos Islands. I. A holocene pollen record from El Junco Lake, Isla San Cristobal. J Ecol 64:989–1012. doi: 10.2307/2258820 Colinvaux PA, Schofield EK (1976b) Historical ecology in the Galapagos Islands. II. A holocene spore record from El Junco Lake, Isla San Cristobal. J Ecol 64:1013–1028. doi: 10.2307/2258821 Derycke S, Remerie T, Vierstraete A et al (2005) Mitochondrial DNA variation and cryptic speciation within the free-living marine nematode Pellioditis marina. Mar Ecol Prog Ser 300:91–103. doi:10.3354/meps300091 Eyualem A, Coomans A (1995) Freshwater nematodes of the Gala´pagos. Hydrobiologia 299(1):1–51. doi:10.1007/BF 00016885 Fenchel T, Finlay BJ (2004) The ubiquity of small species: patterns of local and global diversity. Bioscience 54(8):777–784. doi:10.1641/0006-3568(2004)054[0777: TUOSSP]2.0.CO;2 Ferrington LC, Pehofer HE (1996) In star distribution and biomass of Chironomidae larvae in Lago El Junco, Isla San Cristobal, the Gala´pagos. Hydrobiologia 318(1/2): 123–133. doi:10.1007/BF00014138 Gerecke R, Peck SB, Pehofer HE (1995) The invertebrate fauna of the inland waters of the Gala´pagos Archipelago (Ecuador)—a limnological and zoogeographical summary. Arch Hydrobiol Suppl 107(2):101–135 Hakenkamp CC, Morin A, Strayer DL (2002) The functional importance of freshwater meiofauna. In: Rundle SD, Robertson A, Schmid-Araya JM (eds) Freshwater meiofauna:
247 biology and ecology. Backhuys Publishers, Leiden, pp 321– 335 Heip C, Vincx M, Vranken G (1985) The ecology of marine nematodes. Oceanogr Mar Biol 23:399–489 Holopainen IJ, Paasivirta L (1977) Abundance and biomass of the meiozoobenthos in the oligotrophic and mesotrophic Lake Pa¨a¨jyarvi, southern Finnland. Ann Zool Fenn 14:124–134 Ilow S, Weber D (1974) Fens and bogs in the Gala´pagos Islands. Hikobia 7:39–52 Jackson MH (1991) Galapagos: a natural history guide. University of Calgary Press, Calgary Jacobs LJ (1984) The free-living inland aquatic nematodes of Africa: A review. Hydrobiologia 113:259–291. doi: 10.1007/BF00026614 Krebs CJ (1994) Ecology: the experimental analysis of distribution and abundance. HarperCollins College Publishers, New York Kung SP, Gaugler R, Kaya HK (1990) Influence of soil pH and oxygen on persistence of Steinernema spp. J Nematol 22(4):440–445 Loof PAA (1999) Nematoda: Adenophorea (Dorylaimida). Spektrum Akad. Verl., Heidelberg MacArthur RH, Wilson EO (1967) The theory of island biogeography. Princeton University Press, Princeton McBirney AR, Williams H (1969) Geology and petrology of the Gala´pagos Islands. The Geological Society of America, Boulder Michiels I, Traunspurger W (2004) A three year study of seasonal dynamics of a zoobenthos community in a eutrophic lake. Nematology 6(5):655–669. doi:10.1163/1568541 042843568 Michiels I, Traunspurger W (2005) Benthic community patterns and the composition of feeding types and reproductive modes in freshwater nematodes. Nematology 7(1):21–36. doi:10.1163/1568541054192234 Morgan GT, MacLean AA (1968) Influence of soil pH on an introduced population of Pratylenchus penetrans. Nematologica 14(2):311 Mouginis-Mark PJ, Snell H, Ellisor R (2000) GOES satellite and field observations of the 1998 eruption of Volcan Cerro Azul, Galapagos Islands. Bull Volc 62(3):188–198. doi:10.1007/s004450000078 Munro DC, Rowland SK (1996) Caldera morphology in the western Gala´pagos and implications for volcano eruptive behavior and mechanisms of caldera formation. J Volcanol Geotherm Res 72:85–100. doi:10.1016/0377-0273 (95)00076-3 Muschiol D, Traunspurger W (2007) Life cycle and calculation of the intrinsic rate of natural increase of two bacterivorous nematodes, Panagrolaimus sp. and Poikilolaimus sp. from chemoautotrophic Movile Cave, Romania. Nematology 9(2):271–284. doi:10.1163/156854107780739117 Naumann T, Geist D (2000) Physical volcanology and structural development of Cerro Azul Volcano, Isabela Island, Galapagos: Implications for the development of Galapagos-type shield volcanoes. Bull Volc 61:497–514 Naumann T, Geist D, Kurz M (2002) Petrology and geochemistry of Volcan Cerro Azul: Petrologic diversity among the western Galapagos volcanoes. J Petrol 43(5):859–884. doi: 10.1093/petrology/43.5.859
123
248 Nesteruk T (1996) Density and biomass of Gastrotricha in sediments of different types of standing waters. Hydrobiologia 324(3):205–208 Overgaard Nielsen C (1949) Studies on soil microfauna, II. The soil inhabiting nematodes. Nat Jutl 2:1–131 Peck SB (1991) The Galapagos Archipelago, Ecuador: with an emphasis on terrestrial invertebrates, especially insects; and an outline for research. In: Dudley EC (ed) The unity of evolutionary biology. Dioscorides Press, Portland, pp 319–336 Peck SB (1992) The dragonflies and damselflies of the Gala´pagos Islands, Ecuador. Psyche (Stuttg) 99:309–321 Peck SB (1994) Diversity and zoogeography of the non-oceanic Crustacea of the Gala´pagos Islands, Ecuador (excluding terrestrial Isopoda). Can J Zool 72:54–69 Peck SB (2001) Smaller orders of insects of the Gala´pagos Islands, Ecuador: evolution, ecology, and diversity. National Research Council of Canada Research Press, Ottawa Peck SB, Balke M (1993) A synopsis of the Dytiscidae of the Galapagos Islands, Ecuador, with description of Rhantus galapagoensis sp. nov. (Coleoptera: Dytiscidae). Can Entomol 125(2):259–266 Peters L, Traunspurger W (2005) Species distribution of freeliving nematodes and other meiofauna in littoral periphyton communities of lakes. Nematology 7(2):267–280. doi:10.1163/1568541054879520 Pfannkuche O, Thiel H (1988) Sample processing. In: Higgins RP, Thiel H (eds) Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, DC, pp 134– 145 Prejs K (1977) The nematodes of the root region of aquatic macrophytes, with special consideration of nematode groupings penetrating the tissues of roots and rhizomes. Ekol Polska 25(1):5–20 Robertson AL (2002) Changing times: the temporal dynamics of freshwater benthic microcrustacea. In: Rundle SD, Robertson A, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys Publishers, Leiden, pp 261–278 Rowland SK, Harris AJ, Wooster MJ et al (2003) Volumetric characteristics of lava flows from interferometric radar and multispectral satellite data: The 1995 Fernandina and 1998 Cerro Azul eruptions in the western Galapagos. Bull Volc 65(5):311–330. doi:10.1007/s00445-002-0262-x Rudnick DT, Elmgren R, Frithsen JB (1985) Meiofaunal prominence and benthic seasonality in a coastal marine ecosystem. Oecologia 67(2):157–168. doi:10.1007/BF003 84279
123
Aquat Ecol (2009) 43:235–248 Rundle SD, Bilton DT, Galassi D et al (2002) The geographical ecology of freshwater meiofauna. In: Rundle SD, Robertson A, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys Publishers, Leiden, pp 279–294 Russell CC (1986) The feeding habits of a species of Mesodorylaimus. J Nematol 18(4):641 Schatz H (1998) Oribatid mites of the Galapagos Islands – faunistics, ecology and speciation. Exp Appl Acarol 22(7):373–409. doi:10.1023/A:1006097928124 Schmid PE, Schmid-Araya JM (2002) Trophic relationships in temporary and permanent freshwater meiofauna. In: Rundle SD, Robertson A, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys Publishers, Leiden, pp 295–319 Schmid-Araya JM (1998) Rotifers in interstitial sediments. Hydrobiologia 387–388:231–240. doi:10.1023/A:10170 33525199 Seinhorst JW (1962) On the killing, fixation and transferring to glycerin of nematodes. Nematologica 8:29–32 Strayer D (1985) The benthic micrometazoans of Mirror Lake, New Hampshire. Arch Hydrobiol Suppl 72:287–426 Strayer D, Likens GE (1986) An energy budget for the zoobenthos of Mirror Lake, New Hampshire. Ecology 67(2): 303–313. doi:10.2307/1938574 Traunspurger W (1997) Bathymetric, seasonal and vertical distribution of the feeding-types of nematodes in an oligotrophic lake. Vie Milieu 47(1):1–7 Traunspurger W (2002) Nematoda. In: Rundle SD, Robertson A, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys Publishers, Leiden, pp 63–104 Traunspurger W, Michiels I, Eyualem A (2006) Composition and distribution of free-living freshwater nematodes: Global and local perspectives. In: Eyualem A, Traunspurger W, Andra´ssy I (eds) Freshwater nematodes: ecology and taxonomy. CABI Publishing, Oxfordshire, pp 46–76 Underwood AJ (1997) Experiments in ecology. Cambridge University Press, Cambridge Wallace RL, Ricci C (2002) Rotifera. In: Rundle SD, Robertson A, Schmid-Araya JM (eds) Freshwater meiofauna: biology and ecology. Backhuys Publishers, Leiden, pp 15–44 Warwick RM, Price R (1979) Ecological and metabolic studies from free-living nematodes from an estuarine mud-flat. Estuar Coast Mar Sci 9:257–271. doi:10.1016/03023524(79)90039-2 Yeates GW, Bongers T (1999) Nematode diversity in agroecosystems. Agric Ecosyst Environ 74(1–3):113–135. doi: 10.1016/S0167-8809(99)00033-X