Protistology Protistology 5 (4), 268–280 (2008)

Biodiversity patterns in protozoan communities: linking processes and scales 1

Yuri Mazei

Department of and , V.G. Belinsky State Pedagogical University, Penza, Russia

Summary

One of the general questions in ecology is how patterns of species diversity change across spa- tial scales. In this study additive partitioning methodology, which allows estimating relative contributions of alpha and beta diversity components to total diversity, was applied to data on protozoan (heterotrophic flagellates and ) communities from sphagnum bogs collected from a nested design consisting of five hierarchical levels. It allowed evaluating addi- tive diversity partitioning on four spatial scales: 1) the Russian plain vs. ecoregions, 2) ecore- gions vs. , 3) ecosystems vs. sites, and 4) sites vs. samples. A significant percentage of total species richness was attributed to beta diversity between ecoregions and among ecosys- tems (different bogs) within ecoregions. Protozoan communities seem to be alpha-dominant at the broadest spatial scale and beta-dominant at finer scales. A switch in relative dominance from beta to alpha diversity with increasing spatial scale suggests scale transitions in ecological processes. This pattern is likely to be a result of different processes operating at different scales. At fine scales protozoan species interact directly, and niche partitioning is the strongest deter- minant of diversity, which results in differences between local communities. At broader spatial scales, where processes such as dispersal and colonization–extinction dynamics structure the communities, these interactions are probably not evident.

Keywords: biodiversity patterns, additive biodiversity partitioning, alpha-diversity, beta-di- versity, gamma-diversity, scale, , , heterotrophic flagellates, testate amoebae

Introduction and evolutionary change in shaping the structure of ecological communities (Green et al., 2004). A central goal of ecology is to understand how Although spatial patterns have been documented biodiversity is generated and maintained. Spatial in many studies of and diversity, such patterns of species diversity provide information patterns are not as well documented in microbial about the mechanisms that regulate biodiversity at species, i.e. those of , , and micro- different scales (Levin, 1992; Gaston and Blackburn, scopic Eukarya (Green and Bohannan, 2006). This 2000; Hillebrand and Blenckner, 2002; Brown et al., is a serious omission given that 2002). For instance, these patterns can offer valuable could comprise much of the biodiversity on Earth clues to the relative influence of dispersal limitation, (Foissner, 1999; Torsvik et al., 2002) and have crucial environmental heterogeneity, and environmental roles in biogeochemical cycling and func-

1 Materials presented on the V European Congress of Protistology (July 23–27, 2007, St. Petersburg, Russia).

© 2008 by Russia, Protistology Protistology • 269 tioning (Gilbert et al., 1998; Morin and McGrady- scales of analysis it is often expressed as indices that Steed, 2004). weigh both the richness and equitability (evenness There are some reasons for our lack of understand- of abundance across species) of a sample. Moreover, ing of the scaling of microbial diversity. Conceptually, some authors, in order to distinguish and underline it is assumed that microbes are different biologically certain spatial scale, have adopted the term species from other forms of so that their biodiversity density for the number of species sampled in a stan- scales in a fundamentally different way (Azovsky, dardized sample unit, e.g. per unit area (Whittaker, 1996, 2000, 2002; Finlay et al., 1996, 1996a, 1999, 1975; Lomolino, 2001). Others have used this proto- 2004; Fenchel et al., 1997; Finlay, 1998, 2002; Finlay col, i.e. holding area constant, but retain the terms and Esteban, 1998; Finlay and Clarke, 1999; Finlay diversity or richness rather than density (O'Brien, and Fenchel, 1999, 2004; Hillebrandt and Azovsky, 1993; Fraser, 1998). 2001; Fenchel and Finlay, 2004, 2005). On the other Species richness is the simplest and the most hand, some of the recent research has challenged this frequently used diversity measure. However, spe- conception, providing evidence of microbial ende- cies-richness assessments are notoriously sensitive mism (Whittaker et al., 2003; Foissner, 2004, 2006; to scale, due to the species–area relationship (Palmer Mitchell and Meisterfeld, 2005), and also of a spatial and White, 1994; Veech, 2000), and to sampling ef- patterning of microbial biodiversity (Chernov, 1993; fort, due to the difficulty of obtaining complete spe- Wilkinson, 1994, 2001; Green et al., 2004; Noguez et cies lists (Palmer, 1995). The two problems are closely al., 2005; Bell et al., 2005; Smith et al., 2005) that is related: the number of the species observed generally similar qualitatively to that of and . increases with the number of individuals sampled, So, the question remains open. A reasoning that and the number of individuals increases with the could reconcile different views is that, since process- size of the sampling unit (Lu et al., 2007). Thus, an es that produce biological diversities operate differ- important starting point in analyzing spatial pat- ently and at different rates according to the position terns in richness is to control the area: a step that is of the biological phenomena along the scales of space very often ignored or fudged in analysis, especially at and time, many theories and paradigms are prob- coarser scales (Whittaker et al., 2001). ably more complementary than conflicting (Blondel, In order to consider scale in assessment of species 1987). diversity, Whittaker (1960, 1975) proposed scale-de- Here, I aimed to reveal spatial scaling of biodi- pendent species diversity terms. First, he designat- versity patterns in different types of protozoan com- ed inventory diversity, or simply richness, assessed munities from different biotopes using original data at four scales: (1) point scale (2) alpha (3) gamma collected within the European part of Russia. To dis- (landscape), and (4) epsilon (regional). Secondly, he cover the way in which total species diversity is parti- described a separate phenomenon, compositional tioned into the alpha and beta components on differ- turnover. This he termed differentiation diversity, ent spatial scales, the additive partitioning methodol- identifying three scales: (1) internal beta or pattern ogy was applied (Lande, 1996; Loreau, 2000; Wagner diversity, lying between the inventory scales of point et al., 2000; Gering, Crist, 2002; Veech et al., 2002; Lu and alpha; (2) beta diversity, between alpha and et al., 2007). This approach allows linking biodiver- gamma scales, and (3) geographical differentiation sity patterns with scales and processes operating at or delta diversity, between gamma and epsilon scales. different scales (Whittaker et al., 2001). He thus subdivided diversity into seven categories in total. However, this scheme has not been widely ad- Background opted because of a limited number of scales. Most of the authors currently referring to the framework fol- Species diversity and scale low, in practice, the version proposed by Cody (1975). There is a general agreement over the terms alpha and Measuring species diversity is critical for eco- beta in the two schemes. However, Cody’s gamma logical research and biodiversity conservation. In scale was intended to apply to the inventory diversity the ecological literature, many measures have been (species richness) of a whole landscape. Others ad- proposed to assess species diversity based on data opted his scheme, but generally took gamma diversi- on presence or abundance of species (Pielou, 1975; ty to refer to areas of different scales, perhaps because Magurran, 1988). Accordingly, a lot of terms have Whittaker initially set no upper bound to the term. arose. The term species richness is used for the num- What actual spatial scales do the terms alpha, ber of species in a sample. Species diversity is com- beta, gamma translate to? Given that different taxa monly used interchangeably for richness, but at local of terrestrial and aquatic creatures differ by many or- 270 • Yuri Mazei ders of magnitude in body size, it is evident that the spatial scales at which alpha, beta and gamma should be operationalized can vary between taxa (Burkovsky et al., 1994; Azovsky, 2000, 2002; Whittaker et al., 2001). The precise scale chosen is often a matter of convenience relating to the scale at which species have been mapped (Linder, 1991). I favor (in the same way as O'Brien et al., 2000 and Whittaker et al., 2001) the use of the more intuitive (and intentionally im- precise) terms local-scale (micro-scale), landscape- scale (meso-scale), regional-scale (macro-scale), and geographical-scale (mega-scale). However, the dis- tinction made by Whittaker (1977) between inven- tory and differentiation diversity is an important and useful one, as is the recognition that each of these concepts can be applied at different scales of analy- sis.

Additive diversity partitioning

Although the hierarchical concept of diversity has a strong conceptual meaning for ecologists, it has lacked until recently the mathematical proper- Fig. 1. Theadditivepartitioningoftotaldiversityintoalphaand ties to make it useful in empirical or experimental beta components at five nested spatial scales (modified from settings. Whittaker (1960) originally developed a Veech et al., 2002). Mean diversity within samples at each scale multiplicative formula to explain how alpha and (alpha1, alpha2, alpha3, alpha4 and alpha5) can be obtained beta contributed to gamma diversity (i.e. gamma = on the basis of species richness in each sample. From these alpha x beta). The disadvantage of this relationship values, beta diversity at any scale is determined by subtract- is that diversity components are not weighed equally ing the alpha diversity at that scale from the alpha diversity at when they are applied to more than one spatial scale the next highest scale (e.g. beta1= alpha2−alpha1). When there (Gering and Crist, 2002). However, the additive rela- are three sampling scales, total = alpha1+beta1+beta2+beta3+ tionship between the total diversity and its alpha and beta4; in a similar way, additive diversity partitioning can be beta components (i.e. gamma = alpha + beta) modi- extended to any number of scales. Converting each diversity fied Whittaker’s (1960) original formula. The addi- component into a percentage is a convenient way of expressing tive approach was originally adopted 40 years ago its relative contribution to total diversity. (MacArthur et al., 1966; Levins, 1968; Pielou, 1969; Lewontin, 1972; Allan, 1975), but has only recently been evaluated (Lande, 1996; Veech et al., 2002) and applied to ecological phenomena (Wagner et al. 2000; Species richness, scales and processes Loreau, 2000; Fournier and Loreau, 2001; Gering and Crist, 2002; Ricotta, 2003; Crist et al., 2003; Gering et There are numerous theories and hypotheses al., 2003; Martin et al, 2005; Stendera and Johnson, concerning spatial patterns of richness (Whittaker 2005; Chen et al., 2006; Lu et al., 2007). The additive et al., 2001). It is proposed that at larger scales they approach treats alpha-diversity as the average with- collapse to dynamic hypotheses (based on climate, in-unit diversity. Among-unit diversity (beta) is thus e.g. glaciation effect, dispersal, speciation rates), his- the average amount of diversity not found in a single, torical contingency, and available energy (partition- randomly chosen unit, and reflects the distinctive- ing of energy among species limits richness). Other ness of all units. Therefore, alpha- and beta-diversity hypotheses are largely operated in local-to-landscape are commensurate and can be compared directly scales of analysis. They are: (1) environmental stress (Veech et al., 2002). Recently, Loreau (2000), Wagner (fewer species are physiologically equipped to toler- et al. (2000), Fournier and Loreau (2001), and Veech ate harsh environments); (2) environmental stability et al. (2002) explicitly demonstrated how gamma-di- (fewer species are physiologically equipped to toler- versity is partitioned into alpha- and beta-diversities ate varying environments); (3) disturbance (distur- at multiple spatial scales (Fig. 1). bance prevents competitive exclusion); (4) biological/ Protistology • 271 ecological interactions (competition and predation insects and host-specific like parasites. affect niche partitioning). However, there is no information about diversity Thus, the species richness in local assemblages patterns in protozoan communities in this context, can be regulated by local factors (such as competi- although these tiny creatures could provide new in- tion, disturbance, abiotic conditions) and by regional sights into the understanding of mechanisms shap- factors (such as history of climate, and mi- ing communities of living things. gration). Conceptually, the assembly of a local com- munity can be visualized as species passing through a Aim and hypothesis series of filters, which represent historical (e.g. disper- sal, speciation) and ecological (e.g. competition, pre- The aim of this study is to investigate how the dation, disturbance, abiotic environmental factors) contributions of alpha and beta to regional diversity constraints on the arrival and survival of organisms change as a function of spatial scale. Documenting at a certain locality (Zobel, 1997; Lawton, 1999). In the scale dependence (if any) of alpha and beta to this concept, the local diversity is related to the diver- gamma diversity would be helpful in determining sity of the regional pool if processes associated with the processes that produce a pattern of species rich- the dispersal of organisms are mainly responsible for ness at a given spatial scale (Loreau, 2000; Scheiner the assembly of local communities. A dominant im- et al., 2000). pact of the local environment (abiotic and biotic) was Different hypotheses could be developed rep- supposed to lead to independence between local and resenting a broad range of possible scenarios. Scale regional species richness. Several contributions tried independence in alpha and beta could occur only if to disentangle the regional and the local constraints the relationship of alpha and beta to regional diver- of local species richness (Cornell and Lawton, 1992; sity remained unchanged across spatial scales. This Ricklefs and Schluter, 1993; Cornell and Karlson scenario is analogous to a null model (i.e. no change 1996, 1997; Cornell, 1999; Srivastava, 1999; Shurin, in alpha and beta across scales), but is the least like- 2001). Generally, these studies agree on an impor- ly, because processes that determine community tant influence of both regional and local factors, but structure change across spatial scales (Burkovsky et the relative importance of these factors according al., 1994; Peterson and Parker, 1998; Huston, 1999; to different scales and organisms’ body size is still Azovsky, 2000, 2002) and subsequently affect the uncertain (Hillebrand and Blencker, 2002). Some of balance between alpha and beta (Loreau, 2000). these scales are difficult to manipulate or are not at Alternatively, alpha and beta could exhibit constant all tractable, reducing the possibility to experimen- scale dependence, under which there would be a con- tally the predictions on regional and local influ- stant decrease (or increase) in the contribution of al- ence. Therefore, the importance of regional and local pha or beta to regional diversity as the spatial scale processes has been derived from the analysis of pat- is decreased (or increased). However, it is unclear terns. whether these changes occur in a constant manner A central method used in this discussion is the or in an irregular manner, so I also considered a regression of local species richness on regional one situation where alpha and beta diversity would ex- (Lawton, 1999; Srivastava, 1999). Significant linear hibit irregular scale dependence. This could occur if regressions are interpreted as an indication of the abrupt transition zones were encountered across the high impact of regional factors on local diversity, range of spatial scales. Transition zones represent whereas saturating or nonlinear functions would in- boundaries between scale domains, or ranges of spa- dicate an upper limit of local species richness set by tial scales that are dominated by particular ecologi- ecological interactions (Cornell and Lawton, 1992). cal processes (Wiens, 1989; King et al., 1991; Levin, However, this approach has some limitation in terms 1992). Finally, it is unclear whether alpha or beta di- of scaling (Hillebrand and Blencker, 2002). Another versity will contribute more to the regional diversity possibility of revealing processes operating at differ- across the range of spatial scales, although Huston ent scales is to examine the way in which gamma di- (1999) predicts that alpha diversity should contrib- versity is partitioned into alpha and beta components ute less to regional diversity as spatial scale decreases (Loreau, 2000; Gering and Crist, 2002). A switch in because direct interactions are more common at fine relative dominance from beta to alpha diversity with spatial scales. In any case, we have included both al- changing spatial scale suggests scale transitions in pha-dominant and beta-dominant scenarios. ecological processes (local vs. regional). I tested these hypotheses using two types of proto- Such analyses have been done in terrestrial, ma- zoan communities (testate amoebae and heterotrophic rine and freshwater systems, mainly for vertebrates, flagellates from sphagnum bogs) from a hierarchical- 272 • Yuri Mazei ly nested design that encompassed four spatial scales: Chernaya river basin; surroundings of Chernaya ecoregions (different continental native zones; mega- river village) regions (Fig. 2). The following hierar- scale), ecosystems (different bogs within one region; chical levels (corresponding to spatial scales) were macro-scale), parts of ecosystems (different sites [ e.g. represented in the design: the Russian plain, ecore- hummocks, lawns and hollows in bogs] within one gions, ecosystems, sites and samples. The highest ecosystem; meso-scale), and samples (within macro- level (broadest spatial scale) was represented by three scopically homogeneous microbiotopes) within the ecoregions (forest-steppe, southern taiga and north- European part of Russia (the Russian plain). ern taiga; the distance between them is measured as thousands of kilometers) situated within the Russian Material and Methods plain and differing in their present-day climate and vegetation. Four bogs (ecosystems) were nested with- Sampling design and study sites in each ecoregion (the distance between bogs is mea- sured as tens of kilometers; size of bogs varies from I used a hierarchically nested design to sample 2000 to 5000 m2). Within each bog, five sites were protozoans (testate amoebae and heterotrophic flag- selected that represented typical xeric, mesic and ellates) from sphagnum bogs of forest-steppe (Middle humidic parts of ecosystems (the distance between Volga; Sura river basin; Penza region), southern taiga them is measured as tens of meters; size of sites var- (Upper Volga; Latka river basin; surroundings of ies from 1 to 4 m2). From those sites three samples Borok village) and northern taiga (Nothern Karelia; were taken (their size was the size of one stem of

Fig. 2. Geographic position of the study sites (black circles). Protistology • 273 sphagnum moss (2–3 cm2) and the distance between REICHERT (Austria) with interference- them is measured as tens of centimeters) to represent contrast nozzles and oil-immersion lenses (x1000) the lowest hierarchical level (i.e., finest spatial scale) were used for light microscopic examination. The in the study. Thus, this sampling design consisted of were compounded with an AVT HORN five hierarchical levels, which allowed me to evaluate MC-1009/S analog video camera. To increase dis- additive diversity partitioning on four spatial scales: tinctness of identification of flagellates, the images 1) the Russian plain vs. ecoregions, 2) ecoregions vs. were recorded on a Panasonic NV-HS 850 recorder ecosystems, 3) ecosystems vs. sites, and 4) sites vs. in VHS and S-VHS formats with subsequent digiti- samples. zation and saving of video-film fragments AVI files. Heterotrophic flagellates were identified by means Protozoan sampling and processing of observations of living cells, with the exception of scale-bearing species. Drops of suspended scale- I sampled the protozoan communities from three bearing cells were placed on copper grids coated with ecoregions during the period from 15 June to 15 July, Formvar film and prepared as whole mounts by the 2004. method described by Moestrup and Thomsen (1980). During sampling of testate amoebae a part of Grids were shadowed with tungsten oxide, and were sphagnum was taken from moss carpet; one plant observed with a JEM-100C transmission electron were picked out, placed in 10-ml plastic vessels and microscope. fixed by 4-% formalin. To extract testate amoebae from the moss, samples were thoroughly shaken and Data analysis stirred for 10 min in distilled water. The suspension without sphagnum stems was poured off to a Petri I use additive partitioning methodology to un- dish; live amoebae and empty tests were distin- derstand how components of species diversity (in guished and counted separately in one-tenth field this study I estimate diversity as species richness, i.e. of vision of stereomicroscope MBS–9 (Russia) at a number of species at different scales) and richness magnification of x60. The amounts of cells obtained contribute to different scales and then to hypothesize were evaluated to 1 gram of absolute dry sphagnum about the processes that produce a pattern of species weight. If necessary, the tests were transferred, with richness at a given spatial scale (Allan, 1975; Lande, the help of a thin pipette, to an object-plate, placed in 1996; Loreau, 2000; Scheiner et al., 2000; Gering et a drop of glycerin and investigated at a magnification al., 2003). Within the context of this study, alpha and of x150 or x300 with the use of BIOMED–2 micro- beta diversity components maintain their traditional scope (Russia). interpretations (Whittaker, 1960, 1977) as within- For sampling heterotrophic flagellates, the sam- unit diversity (alpha-component) and between-unit ples containing peat water with organic debris and diversity (beta-component) on a given scale. Since live sphagnum stems, were placed in 10-ml plastic alpha diversity at a given scale is the sum of the al- vessels and conserved at a temperature of 3ºC dur- pha and beta diversity at the next lowest scale ( e.g., ing transportation to the laboratory. In the labo- alpha2 = alpha1 + beta1; Allan, 1975; Lande, 1996), ratory, samples 5 cm3 in volume were split into two the overall protozoan diversity in this study can be equal parts and put into Petri dishes (that is, each described by the following formula: alpha1 + beta1 + sample was analyzed in two replicates). To each dish beta2 + beta3 + beta4 (Fig. 3). 0.15 ml of a suspension of the bacteria Pseudomonas fluorescens containing approximately 25 mln bacte- Results rial cells was added. First, natural (non-enriched) samples were examined; then, after feeding them General community patterns with bacteria, enriched samples were analyzed three, six, and nine days later. This allowed us to more -ad During this study 103 heterotrophic flagellate and equately estimate the species richness. In order to 130 testate amoebae species were identified. The spe- reduce the number of photosynthesizing species and cies diversity of cercomonads, , and kineto- to enchance the development of the heterotrophic plastids is the highest. Spumella sp., Paraphysomonas organisms, the Petri dishes containing the samples vestita, Bodo saltans, B. designis, Goniomonas trun- were kept in the dark in a thermostat at the tempera- cata, Heteromita minima, H. reniformis, Cercomonas ture of 20ºC. BIOLAM-I microscope (Russia) with radiatus, C. longicauda, Dimastigella mimosa, KF-5 transmission light-contrast devices and water- Helkesimastix faecicola, and Spongomonas uvella are immersion lenses (total magnification x700), and the most common heterotrophic flagellate species. 274 • Yuri Mazei

Fig. 3. Relationships among hierarchical levels in additive partitioning mode, applied in this study, Whittaker’s (1960) terminology (in parentheses), and MacArthur’s (1965) designations (in brackets) of diversity (slightly modified after Gering et al., 2003). Because of the additive relationship between levels (e.g., point + pattern = alpha, alpha + beta = gamma), we can use substitution among levels to ar- rive at the following equation (illustrated by arrows and mathematical operators) to describe the total (i.e., regional or alpha5) diversity: alpha1 + beta1 + beta2 + beta3 + beta4.

Fig. 4. Percentage of total protozoan species richness explained by alpha and beta components of diversity on different spatial scales. The contributions to the total richness for each scale were determined by the additive partitioning of diversity. Protistology • 275

Among testate amoebae the most species-rich nant at lower spatial scales and distinctly alpha- are families Arcellidae, Centropyxidae, Difflugiidae, dominant at the broadest (ecoregional) spatial scale Nebelidae, and Euglyphidae. The most common (Fig. 5). species are Assulina muscorum, Archerella flavum, Nebela tincta, N. t. major, Phryganella hemisphaeri- Discussion ca, Hyalosphenia papilio, H. elegans, Euglypha laevis, Arcella arenaria, and A. catinus. The more detailed Additive partitioning methodology is simply a data on species composition and protozoan com- mathematical approach to describing the relative munity structure in sphagnum bogs investigated are contributions of components to a sum total. It can given in previous publications (Mazei and Bubnova, be used on a variety of metrics and can quantify spa- 2007; Mazei and Tsyganov, 2007, 2007a; Mazei et al., tial and temporal patterns of diversity as well as oth- 2007; Tikhonenkov and Mazei, 2007; Tsyganov and er ecological data (Gering et al., 2003). In practice, Mazei, 2007). however, this approach has not been applied to many ecological phenomena. The exceptions include study Additive biodiversity partitioning of benthic insect diversity in an alpine stream (Allan, 1975), plant species richness in agricultural land- The most noticeable result from the additive par- scapes (Wagner et al., 2000), temporal and spatial titioning is the highest contribution of beta3 and beta4 patterns of butterfly diversity in rainforests (DeVries components into total regional species richness both and Walla, 2001), arboreal beetle diversity in east- for heterotrophic flagellates and for testate amoebae ern deciduous forests in the USA (Gering and Crist, (Fig. 4). 2002; Gering et al., 2003). Even so, it has consider- Another point is that protozoan communities able potential because it allows one to understand from sphagnum bogs seem to be rather beta-domi- the contributions of alpha and beta diversity to the

Fig. 5. Percentage into which total species richness is partitioned by alpha and beta components on different spatial scales. The percentages of alpha and beta were determined by applying additive partitioning to the total species richness within an individual spatial scale. 276 • Yuri Mazei total diversity (and, thus, the processes that produce decrease at fine spatial scales because the number of a pattern of species richness at a given spatial scale; individuals is reduced and strong direct interactions Loreau, 2000; Scheiner et al., 2000; Gering et al., could dominate the community, thereby increasing 2003) over a range of user-defined spatial scales. beta richness. The reverse is also true: the impor- The result of this study – that broad-scale beta tance of alpha richness to overall regional richness components of diversity make a greater contribution should be more important at broader scales because to the regional diversity – indicates the role of the local interactions are less important or undetectable ecoregions structure’ as well as that of the structure (Huston, 1999; Loreau, 2000). of different bog ecosystems in forming species rich- These explanations are realistic for protozoan ness and composition of testate amoebae and hetero- communities because there is evidence that inter spe- trophic flagellate communities of boreal sphagnum cific interactions (e.g. competition, facilitation, and biotopes. This pattern is similar to those obtained resource sharing) among protozoan species occur from arboreal beetle communities in North America within rather small sites in the ecosystems (Fenchel, (Gering et al., 2003). 1969; Burkovsky, 1984, 1987, 1992; Azovsky, 1989, The major objective of this paper was to investigate 1989a). It has also been pointed out (Shmida and how the contributions of alpha and beta to the total Wilson, 1985) that niche relations are the strongest diversity of protozoan communities from sphagnum determinant of diversity at fine spatial scales (<10m2). bogs change as a function of spatial scale. There are However, these interactions are probably not evi- few predictions about how alpha and beta diversity dent at broader spatial scales, where processes such change across spatial scales (Gering and Crist, 2002). as dispersal and colonization–extinction dynamics Scale independence of alpha and beta is unlikely be- structure the communities. In fact, dispersal of spe- cause ecological processes are scale-dependent and cies into sites where they cannot be self-maintaining have transitions which could in turn affect the bal- almost always results in an increased alpha diversity ance between alpha and beta diversity on a given scale and is one mechanism, among others, that operates (Wiens, 1989; Burkovsky et al., 1994; Peterson and at broad spatial scales (>103 m2). Parker, 1998; Azovsky, 2000, 2002). Constant scale- In summary, I have documented empirical evi- dependence is also unlikely unless there were gradual dence of irregular scale dependence in alpha richness transitions in ecological processes that could generate (and therefore beta richness) and found that diversity constant and predictable changes in the contribution components could switch dominance over the range of of alpha and beta to regional richness (Gering and spatial scales. There is considerable indirect evidence Crist, 2002). Irregular scale dependence of alpha di- to suggest that this pattern may be related to changes versity is the most likely of the three possibilities and in dominant ecological processes such as interspecif- has already been alluded to by other authors (Wiens, ic interactions (at finer spatial scales) and coloniza- 1989; Gering and Crist, 2002). Wiens (1989), for ex- tion–extinction dynamics (at broader spatial scales). ample, conceptualized scale domains, or spatial scales However, as it was noted in some papers (Gering and over which ecological patterns and processes do not Crist, 2002; Hillebrand and Blenckner, 2002), the scale change or change monotonically. Scale domains are dependence of diversity components has not been well separated by abrupt scale transitions that occur when explored, so I cannot eliminate the possibility that a set of ecological patterns and processes are replaced sampling phenomena and/or statistical properties of by another set of patterns and processes. It is at these hierarchical data could also generate the patterns ob- transition points where non-monotonic changes are served. Further studies of scale dependence of diver- evident. Across the range of scales in this study, these sity components will strengthen our understanding transitions could result in a pattern similar to the ir- about the additivity and scale dependence of species regular scale dependence (Fig. 5). diversity in biological communities. I found that alpha richness accounted for a sig- nificantly larger portion (60.8–64.1 %) of the re- Acknowledgements gional richness at the broadest scale. Moreover, my empirical data indicate a clear shift in dominance I would like to thank D.V. Tikhonenkov, A.N. between alpha and beta components across the range Tsyganov and O.A. Bubnova for their help in col- of spatial scales (Fig.5). The analogous results were lecting and processing protozoan samples. The obtained during the study of arboreal beetle commu- work was supported by the Russian Foundation for nities (Gering and Crist, 2002). This switch in domi- Basic Research (grant no. 07-04-00185) and grant nance has been theorized by other authors. Loreau from the President of the Russian Federation (MK- (2000), for instance, stated that alpha richness should 7388.2006.04). Protistology • 277

References Mediterranean habitat gradients. In: Ecology and evo- lution of communities (Eds Cody M.L. and Diamond Allan J.D. 1975. Components of diversity. J.M.). Harvard University Press, Cambridge. pp. Oecologia. 18, 359–367. 214–257. Azovsky A.I. 1989. A niche structure if commu- Cornell H.V. 1999. Unsaturation and regional nity of marine psammophilous I. Location of influences on species richness in ecological com- the niches in a space of resources. J. General Biol. 50, munities: a review of the evidence. Ecoscience. 6, 3, 329–342 (in Russian with English summary). 303–315. Azovsky A.I. 1989a. A niche structure if commu- Cornell H.V. and Karlson R.H. 1996. Species rich- nity of marine psammophilous ciliates II. Parameters ness of reef-building corals determined by local and of ecological niche of species and its quantitative de- regional processes. J. Anim. Ecol. 65, 233–241. velopment. J. General Biol. 50, 6, 752–763 (in Russian Cornell H.V. and Karlson R.H. 1997. Local and with English summary). regional processes as controls of species richness. Azovsky A.I. 1996. The effect of scale on conge- In: Spatial ecology (Eds. Tilman D. and Kareiva P.). ners coexistence: can mollusks and polychaetes rec- Princeton University Press, Princeton. pp 250–268. oncile beetles to ciliates? Oikos. 77, 117–126. Cornell H.V. and Lawton J.H. 1992. Species inter- Azovsky A. I. 2000. Concept of scale in marine actions, local and regional processes, and limits to ecology: linking the words or the worlds. Web ecol- the richness of ecological communities: a theoretical ogy. 1, 28–34. perspective. J. Anim. Ecol. 61, 1–12. Azovsky A. I. 2002. Size-dependent species-area Crist T.O., Veech J.A., Gering J.C. and Summerville relationships in benthos: is the world more diverse K.S. 2003. Partitioning species diversity across land- for microbes? Ecography. 25, 273–282. scapes and regions: a hierarchical analysis of alpha, Bell T., Ager D., Song J.I., Newman J.A., Thompson beta, and gamma diversity. Amer. Nat. 162, 734–743. I.P., Lilley A.K. and van der Gast C.J. 2005. Larger is- DeVries P.J., Walla T.R. and Greenley H.F. 1999. lands house more bacterial taxa. Science. 308, 1884. Species diversity in spatial and temporal dimensions Blondel J. 1987. From biogeography to life history of fruit-feeding butterflies from two Ecuadorian theory: a multithematic approach illustrated by the rainforests // Biol. J. Linn. Soc. 68, 333–353. biogeography of vertebrates. Journal of Biogeography. Fenchel T. 1969. The ecology of marine micro- 14, 405–422. benthos. IV. Structure and function of the benthic Brown J.H., Gupta V.K., Li B.-L., Milne B.T., ecosystem, its chemical and physical factors and the Restrepo C. and West G.B. 2002. The fractal nature microfauna communities with the special reference of nature: power-laws, ecological complexity and bio- to the ciliated protozoa. Ophelia. 6, 1–182. diversity. Phil. Trans. R. Soc. Lond. B 357, 619–626. Finlay B.J. 1998. The global diversity of protozoa Burkovsky I.V. 1984. The ecology of free-living and other small species. Int. J. Parasitol. 28, 29–48. ciliates. Moscow Univ. Press, Moscow (in Russian). Finlay B.J. 2002. Global dispersal of free-living Burkovsky I.V. 1987. The partitioning of ecologi- microbial species. Science. 296, 1061– cal resources and the species interrelations in the 1063. community of marine psammophilous ciliates. Zool. Finlay B.J. and Clarke K.J. 1999. Ubiquitous dis- Zhurn. 66, 5, 645–654. persal of microbial species. Nature. 400, 828. Burkovsky I.V. Structral and functional organiza- Finlay B.J., Corliss, J.O., Esteban G. and Fenchel tion, and stability of marine benthic communities. T. 1996. Biodiversity at the microbial level: the num- Moscow Univ. Press, Moscow (in Russian). ber of free-living ciliates in the biosphere. Quat. Rev. Burkovsky I. V., Azovsky A. I. and Mokiyevsky Biol. 71, 221–237. V. O. 1994. Scaling in benthos: from macrofauna to Finlay B.J. and Esteban G.F. 1998. Freshwater pro- microfauna. Arch. Hydrobiol. Suppl. 99, 517–535. tozoa: biodiversity and ecological function. Biodiv. Chen X.Y., Lu H.P., Ying X.Y. and Song Y.C. Conserv. 7, 1163–1186. 2006. Additive partitioning of tree species diversity Finlay B.J., Esteban G.F. and Fenchel T. 1996. in Tiantong National Forest Park. Chinese J. Appl. Global diversity and body size. Nature. 383, 132– Ecol., 17, 567–571. 133. Chernov I.Yu. 1993. Geography of microorgan- Finlay B.J., Esteban G.F. and Fenchel T., 2004. isms and structure of ecosystems. Geography Bull. diversity is different? Protist. 155, 15–22. RAS. 6, 49–58 (in Russian with English summary). Finlay B.J., Esteban G.F., Olmo J.L. and Tyler P.A. Cody M.L. 1975. Towards a theory of conti- 1999. Global distribution of free-living microbial nental species diversities: bird distributions over species. Ecography. 22, 138–144. 278 • Yuri Mazei

Finlay B. J. and Fenchel T. 1999. Divergent per- Hillebrand H. and BlencknerT. 2002. Regional spectives on protist species richness. Protist. 150, and local impact on species diversity – from pattern 229–233. to processes. Oecologia. 132, 479–491 Finlay B.J. and Fenchel T. 2004. Cosmopolitan Huston M.A. 1999. Local processes and regional metapopulations of free-living microbial . patterns: appropriate scales for understanding varia- Protist. 155, 237–244. tion in the diversity of plants and animals. Oikos. 86, Fenchel T., Esteban G.F. and Finlay B.J. 1997. Local 393–401. versus global diversity of microorganisms: cryptic King A.W., Johnson A.R. and O’Neill R.V. 1991. diversity of ciliated protozoa. Oikos. 80, 220–225. Transmutation and functional representation of Fenchel T. and Finlay B.J. 2004. The ubiquity of heterogeneous landscapes. Landscape Ecol. 5, 239– small species: patterns of local and global diversity. 253. BioScience. 54, 777–784. Lande R. 1996. Statistics and partitioning of spe- Fenchel T. and Finlay B.J., 2005. Bacteria and is- cies diversity, and similarity among multiple com- land biogeography. Science. 309, 1997. munities. Oikos. 76, 5–13. Foissner W. 1999. Protist diversity: estimates of Lawton J.H. 1999. Are there general laws in ecol- the near-imponderable. Protist. 150, 363–368. ogy? Oikos. 84,177–192. Foissner W., 2004. Ubiquity and cosmopolitism Levin, S. The problem of pattern and scale in ecol- of questioned. SILnews. 43, 6–7. ogy. Ecology 73, 1943–1967 (1992). Foissner W. 2006. Biogeography and dispersal Levins R. 1968. Evolution in changing environ- of micro-organisms: a review emphasizing protists. ments, some theoretical explorations. Princeton Acta Protozool. 45, 111–136. Univ. Press, New Jersey. Fournier E., and Loreau M. .2001. Respective Linder H.P. 1991. Environmental correlates of pat- roles of recent hedges and forest patch remnants terns of species richness in the south-western Cape in the maintenance of ground-beetle (Coleoptera: Province of South Africa. J. Biogeogr. 18. 509–518. Carabidae) diversity in an agricultural landscape. Lewontin R.C. 1972. The apportionment of human Landscape Ecol. 16, 17–32. diversity. In: Evolutionary (Ed. Dobzhansky Fraser R.H. 1998. Vertebrate species richness at T.) Vol. 6. Appleton-Century-Crofts Publishing. pp. the mesoscale: relative roles of energy and heteroge- 381–398. neity. Global Ecol. Biogeogr. Lett. 7, 215–220. Lomolino M.V. 2001. Elevation gradients of spe- Gaston K. and Blackburn T. 2000. Pattern and cies-density: historical and prospective views. Global process in macroecology. Blackwell Science, Oxford. Ecol. Biogeogr. 10, 3–13. Gering J.C. and Crist T.O. 2002. The alpha–be- Loreau M. 2000. Are communities saturated? On ta–regional relationship: providing new insights into the relationship between α, β and γ diversity. Ecol. local–regional patterns of species richness and scale Lett. 3, 73–76. dependence of diversity components. Ecol. Lett. 5, Lu H.-P., Wagner H. and Chen X.-Y. 2007. A con- 433–444. tribution diversity approach to evaluate species di- Gering J.C., Crist T.O. and Veech J.A. 2003. versity. Basic Appl. Ecol. 8, 1–12. Additive partitioning of species diversity across mul- MacArthur R. H. 1965. Patterns of species diver- tiple spatial scales: implications for regional conser- sity. Biol. Rev. 40, 510–533. vation of biodiversity. Conserv. Biol. 17, 488–499. MacArthur R., Recher H. and Cody M. 1966. On Gilbert D., Amblard C., Bourdier G. and Francez the relation between habitat selection and species di- A.-J. 1998. The microbial loop at the surface of a peat- versity. Am. Nat. 100, 319–332. land: structure, function, and impact of nutrient in- Magurran A. E. 1988. Ecological diversity and its put. Microb. Ecol. 38, 83–93. measurement. Princeton Univ. Press, New Jersey. Green J. and Bohannan B. 2006. Spatial scaling Martin L.M., Moloney K.A. and Wilsey B.J. 2005. of microbial biodiversity. Trends Ecol. Evol. 21, 501– An assessment of grassland restoration success us- 507. ing species diversity components // J. Appl. Ecol. 42, Green J., Holmes A., Westoby M., Oliver I., 327–336. Briscoe D., Dangerfield M., Gillings M. and Beattie Mazei Yu.A. and Bubnova O.A. 2007. Species A. 2004. Spatial scaling of microbial eukaryote diver- composition and structure of testate amoebae com- sity. Nature. 432, 747–750. munity in a sphagnum bog at the initial stage of its Hillebrand H. and Azovsky A.I. 2001. Body size formation. Biol. Bull. 34, 619–628. determines the strength of the latitudinal diversity Mazei Yu. A. and Tsyganov A.N. 2007. Species gradient. Ecography. 24, 251–256. composition, spatial distribution and seasonal dy- Protistology • 279 namics of testate amoebae community in sphagnum sity: regional and historical influences. In: Species bog (Middle Volga region, Russia). Protistology. 5, diversity in ecological communities (Eds. Ricklefs 2/3, 156–206. R.E., Schluter D.) Univ. Chicago Press, Chicago. pp. Mazei Yu.A. and Tsyganov A.N. 2007a. Changes 350–363. of testate amoebae community structure along envi- Scheiner S.M., Cox S.B., Willig M., Mittlebach ronmental gradients in sphagnum dominated bog, G.G., Osenberg C. and Kaspari M. (2000). Species restorated after peat excavation. Povolzhsky J. Ecol. richness, species-area curves and Simpson’s paradox. 1, 24–33. Evol. Ecol. Res. 2, 791–802. Mazei Yu.A., Tsyganov A.N. and Bubnova O.A. Shurin J.B. 2001. Interactive effects of predation 2007. Structure of a community of testate amoebae and dispersal on zooplankton communities. Ecology. in a sphagnum dominated bog in upper Sura flow 82, 3404–3416. (Middle Volga Territory). Biol. Bull. 34, 4, 382–394. Smith V.H., Foster B.L., Grover J.P., Holt Mitchell E.A.D. and Meisterfeld R. 2005. R.D., Leibold M.A. and de Noyelles F.Jr. 2005. Taxonomic confusion blurs the debate on cosmo- Phytoplankton species richness scales consistently politan versus local endemism of free-living protists. from laboratory microcosms to the world’s oceans. Protist. 156, 263–267. Proc. Natl. Acad. Sci. U. S. A. 102, 4393–4396. Moestrup Ø. and Thomsen H.A. 1980. Srivastava D.S. 1999. Using local-regional rich- Preparations of shadow-cast whole mounts. In: ness plots to test for species saturation: pitfalls and Handbook of phycological methods: developmental potentials. J. Anim. Ecol. 68, 1–16. and cytological methods (Ed. Gantt E.). Cambridge Stendera S.E.S. and Johnson R.K. 2005. Additive Univ. Press, Cambridge. pp. 385–390. partitioning of aquatic invertebrate species diversity Morin P.J. and McGrady-Steed J. 2004. Biodiversity across multiple spatial scales. Freshwater Biol. 50, 13- and ecosystem functioning in aquatic microbial sys- 60–1375. tems: a new analysis of temporal variation and spe- Tikhonenkov D.V. and Mazei Yu.A. 2007. Species cies richness-predictability relations. Oikos. 104, composition and distribution of heterotrophic flagel- 458–466. lates in wetlands of Middle Volga Region. Povolzhsky Noguez A.M., Arita H.T., Escalante A.E., Forney J. Ecol. 3, 227–234. L.J., García-Oliva F. and Souza V. 2005. Microbial Torsvik V., Øvreas L. and Thingstad T.F. 2002. macroecology: highly structured prokaryotic soil as- Prokaryotic diversity – magnitude, dynamics, and semblages in a tropical deciduous forest. Glob. Ecol. controlling factors. Science. 296, 1064–1066. Biogeogr. 14, 241–248. Tsyganov A.N. and Mazei Yu.A. 2007. Species O'Brien E.M. 1993. Climatic gradients in woody composition and structure of testate amopebae plant species richness: towards an explanation based community in boggy lake in Middle Volga Region. on an analysis of Southern Africa's woody flora. J. Developments in current biology. 127, 3, 305–315. Biogeogr. 20, 181–198. Veech J.A. 2000. Choice of species–area function O'Brien E.M., Field R. and Whittaker R.J. 2000. affects identification of hotspots. Conserv. Biol. 14, Climatic gradients in woody plant (tree and shrub) 140–147. diversity: water energy dynamics, residual variation, Veech J.A., Summerville K.S., Crist T.O. and and topography. Oikos. 89, 588–600. Gering J.C. 2002. The additive partitioning of spe- Palmer M.W. 1995. How should one count spe- cies diversity: recent revival of an old idea. Oikos. 99, cies? Natural Areas Journ. 15, 124–135. 3–9. Palmer M.W. and White P.S. 1994. Scale depen- Wagner H.H., Wildi O. and Ewald K.C. 2000. dence and the species–area relationship. Amer. Nat. Additive partitioning of plant species diversity in an 144, 717–740. agricultural mosaic landscape. Landscape Ecol. 15, Peterson D.L. and Parker T. 1998. Ecological 219–227. scale: theory and applications. Columbia University Whittaker R.H. 1960. Vegetation of the Siskiyou Press, New York. Mountains, Oregon and California. Ecol. Monogr. Pielou E.C. 1969. An introduction to mathemati- 30, 279–338. cal ecology. Wiley, NY. Whittaker R.H. 1975. Communities and ecosys- Pielou E.C. 1975. Ecological diversity. Wiley, NY. tems. 2nd edn. Macmillan, NY. Ricotta C. 2003. Additive partition of parametric Whittaker R.H. 1977. Evolution of species diver- information and its associated beta-diversity mea- sity in land communities. . Vol. sure. Acta Biotheor. 51, 91–100. 10 (Eds. Hecht M.K., Steere W.C. and B. Wallace). Ricklefs R.E. and Schluter D. 1993. Species diver- Plenum Press, NY. pp. 250–268. 280 • Yuri Mazei

Whittaker R.H., Willis K.J. and Field R. 2001. Scale raphy of the protozoan genus Nebela in the southern and species richness: towards a general, hierarchical temperate and Antarctic zones. Area. 26, 150–157. theory of species diversity. J. Biogeogr. 28, 453–470. Wilkinson D.M. 2001. What is the upper size Whitaker R.J., Grogan D.W. and Taylor J.W. 2003. limit for cosmopolitan distribution in free-living mi- Geographic barriers isolate endemic populations of croorganisms? J. Biogeogr. 28, 285–291. hyperthermophilic archaea. Science. 301, 976–978. Zobel M. 1997. The relative role of species pools Wiens J.A. 1989. Spatial scaling in ecology. Funct. in determining plant species richness: an alterna- Ecol. 3, 385–397. tive explanation of species coexistence? Trends Ecol. Wilkinson D.M. 1994. A review of the biogeog- Evol. 12, 266–269.

Address for correspondence. Yuri Mazei. Department of Zoology and Ecology, V.G. Belinsky State Pedagogical University, Penza, 440026 Russia. E-mail: [email protected]