OUP user menu

Distribution and ecophysiology of the nitrifying bacteria emphasizing cultured species

Hans-Peter Koops, Andreas Pommerening-Röser
DOI: http://dx.doi.org/10.1111/j.1574-6941.2001.tb00847.x 1-9 First published online: 1 August 2001


Nitrification is an important factor in the global nitrogen cycle. Therefore, an increasing number of publications deal with in situ studies of natural bacterial populations participating in this process. However, some crucial points complicate suchlike investigations. At the time being, a total of 25 species of ammonia-oxidizers and eight species of nitrite-oxidizers are cultured but the existence of many more species has been indicated by molecular in situ investigations. With that, only a part of the existing nitrifiers has been defined via isolation and subsequent physiological and molecular characterization. Furthermore, the distribution patterns of the distinct species of nitrifiers depend on various environmental parameters. Hence the composition of nitrifying bacterial communities is complex and divers in heterogeneous habitats. In consequence of the above-mentioned problems, the representation of nitrifying community structures obtained from in situ investigations often has been incomplete and unbalanced in many respects. Polyphasic approaches, applying a combination of classical as well as molecular methods in parallel, could help to find the way for overcoming these problems in the future. Isolation and characterization of as many as possible new species seems to be one of the most important missing steps to advance at this way.

  • Nitrifying bacterium
  • Distribution
  • Ecology

1 Introduction

The nitrification process is defined as the biological transformation of reduced forms of nitrogen to nitrate. The most important groups of organisms, involved in this process, are the lithoautotrophic ammonia-oxidizing bacteria and the lithoautotrophic nitrite-oxidizing bacteria. For these organisms the oxidation of inorganic nitrogen compounds serves as their characteristic energy source. Together they catalyze two distinct stages of the total process, the two-step oxidation of ammonia to nitrite (NH3+2H++2e→NH2OH+H2O and NH2OH+H2O→HNO2+4 H++4e) and the oxidation of nitrite to nitrate (HNO2+H2O→HNO3+2H++2e), respectively.

Recently it has been indicated that representatives of the order Planctomyces are involved in the lithotrophic nitrification if anaerobic conditions are prevalent [1]. These organisms seem to combine ammonia and nitrite directly into dinitrogen gas. Unfortunately, pure cultures of the involved bacteria are missing as yet.

The so-called heterotrophic nitrification is catalyzed by a lot of different organisms, comprising fungi as well as heterotrophic bacteria. Inorganic pathways (NH4+→NH2OH→NOH→NO2→NO3) and organic pathways (RNH2→RNHOH→RNO→RNO2→NO3) have been postulated, but biochemical reactions are not well known [2]. The organisms cannot use the oxidation reactions as energy source.

The lithoautotrophic nitrifying bacteria exclusively will be in focus of the present review. Included are unpublished experiences of the authors obtained in isolation and characterization of several hundred nitrifying bacteria from a great variety of environments.

2 The organisms

Originally, the lithoautotrophic nitrifying bacteria altogether were comprised within one family, named Nitrobacteraceae. However, phylogenetic investigations made evident that a lot of distinct groups of organisms exist [3,5 which are not closely related to each other (Figs. 1 2).


Dendrogram based on 16S rDNA sequences showing the phylogenetic interrelationships among the cultured ammonia-oxidizing bacteria. The tree was constructed by using the neighbor-joining method. Included are information on ecophysiological parameters and preferred habitats.


Dendrogram based on 16S rDNA sequences showing the phylogenetic interrelationships among the cultured nitrite-oxidizing bacteria. The tree was constructed by using the neighbor-joining method. Included are information on ecophysiological parameters and preferred habitats.

2.1 The lithotrophic ammonia-oxidizers

At present two phylogenetically distinct groups are defined [6,7. One group is located within the γ subclass of the Proteobacteria. The only genus, Nitrosococcus, is represented by two described species, Nitrosococcus oceani and Nitrosococcus halophilus [8]. The second group belongs to the β subclass of the Proteobacteria. Two clusters exist within this assemblage, the Nitrosospira cluster and the Nitrosomonas cluster. All members of the three genera of the Nitrosospira cluster are very closely related to each other [5,6, whereas the Nitrosomonas cluster reveals at least five distinct lineages of descent [9]. A total of 14 species are described: Nitrosospira briensis, Nitrosovibrio tenuis, Nitrosolobus multiformis, the Nitrosomonas species Nitrosomonas europaea, Nitrosomonas eutropha, Nitrosomonas halophila, Nitrosomonas ureae, Nitrosomonas oligotropha, Nitrosomonas marina, Nitrosomonas aestuarii, Nitrosomonas communis, Nitrosomonas nitrosa, Nitrosomonas cryotolerans and Nitrosococcus mobilis10,11. The latter species phylogenetically belongs to the genus Nitrosomonas and thus should be reclassified as Nitrosomonas mobilis[6].

2.2 The lithotrophic nitrite-oxidizers

Four phylogenetically distinct groups of nitrite-oxidizing bacteria have been described. The major group, which belongs to the α subclass of the Proteobacteria, is represented by a single genus, Nitrobacter. Four species, Nitrobacter winogradskyi, Nitrobacter hamburgensis, Nitrobacter vulgaris and Nitrobacter alkalicus, have been described [12,13. Two marine species, Nitrococcus mobilis and Nitrospina gracilis[14], were assigned to the γ and the δ subclass of the Proteobacteria, respectively. The two species of the genus Nitrospira, Nitrospira marina and Nitrospira moscoviensis15,16, are members of a distinct phylum close to the δ subclass of the Proteobacteria.

3 Ecophysiologically relevant properties of the organisms

Although the basic metabolism is more or less uniform within the physiologically defined groups of lithoautotrophic ammonia- or nitrite-oxidizing bacteria, ecophysiological differences exist between the distinct representatives. This has led to different distribution patterns of the species or groups of species.

3.1 The lithotrophic ammonia-oxidizers

All species use ammonia as the sole energy source. However, the substrate affinity (Ks value of the ammonia-oxidizing system) differs significantly among the species. Within the genus Nitrosomonas, different Ks values reflect well the phylogenetically definable groups [9,17.

Many species, but not all, are able to use urea as ammonia source. Within the group of ammonia-oxidizers located in the β subclass of the Proteobacteria, there is an interesting difference between the two main clusters. With all species of the Nitrosospira cluster, urease-positive as well as urease-negative strains have been observed at a similar frequency (unpublished data). Within the Nitrosomonas cluster, urease activity is not observed with all species and, in contrast to the Nitrosospira cluster, nearly all isolates of a given urease-positive species revealed this property in laboratory experiments (unpublished data). Within the Nitrosomonas cluster, the presence or absence of urease activity is strongly correlated with the phylogenetic discrimination patterns [9].

Salt requirement is another ecophysiologically relevant discrimination factor. All isolates of the two species of Nitrosococcus (γ subclass of the Proteobacteria), N. oceani and N. halophilus, are obligately halophilic. The group located in the β subclass of the Proteobacteria, comprises obligately halophilic species and moderately halophilic or halotolerant species, respectively, together with species missing salt requirement or being salt sensitive. Within the genus Nitrosomonas, these differences are well reflected by the pronounced formation of phylogenetic lineages [9].

Especially in aquatic environments, some species of ammonia-oxidizing bacteria have been observed to occur almost exclusively attached to flocs or biofilms. These species appeared to be self-flocculating in pure cultures (Fig. 3). Oligotrophic conditions seem to stimulate the excretion of exopolymeric substances [18]. Other species were observed to occur predominantly as individual planktonic cells [18] but to be able to colonize already existing flocs or biofilms (Fig. 4).


Phase contrast photomicrograph of an aggregate of the self-flocculating ammonia-oxidizing bacterium Nitrosomonas sp. Nm84, being isolated from the River Elbe. Bar=10 μm (reproduced with permission from [18]).


Transmission electron micrograph of an ultrathin section of a cell aggregate of the ammonia-oxidizing bacterium Nitrosococcus mobilis embedded in an activated sludge flock sampled from the rendering plant Kraftesried (Germany) flock. Bar=1 μm (with permission from [23]).

3.2 The lithotrophic nitrite-oxidizers

Members of the Nitrobacter group are all able to use organic energy sources beside the major source nitrite [12]. The representatives of the other phylogenetic groups of nitrite-oxidizers altogether seem to be obligately lithotrophic. Ks values of the nitrite-oxidizing system seem to be different among the groups as indicated by physiological investigations [15,16,19.

Nitrococcus mobilis and Nitrospina gracilis, the single representatives of two distinct phylogenetic lineages of the lithotrophic nitrite-oxidizers, are both obligately halophilic. All isolates originate from marine habitats [14]. The genus Nitrospira comprises obligately halophilic species (N. marina) together with nonhalophilic species (N. moscoviensis). With Nitrobacter-isolates, obligate salt requirement has not been observed though some strains were isolated from marine environments or from soda lakes.

While members of the genus Nitrobacter seem generally to occur as free-living cells, representatives of Nitrospira repeatedly have been observed attached to flocs or biofilms in their natural environments. Analysing flocs of activated sludge via fluorescent in situ hybridization, only cells of Nitrospira-like organisms were detected to be attached. However, the only isolates of nitrite-oxidizers obtained from the highest MPN dilution of water samples taken from the same environment were representatives of the genus Nitrobacter[20].

4 Environmental distribution patterns of the organisms

Lithotrophic nitrifying bacteria are present in a great variety of habitats including soils, rocks, fresh- and seawaters and sediments. In the following chapter, the distribution patterns of distinct species or groups of species will be described. This chapter is mainly based on experiences obtained from enrichments and isolations of nitrifiers from diverse environments carried out in our or other laboratories. However, information also has been obtained from molecular in situ investigations carried out by several authors. The information from most probable number (MPN) enrichment and subsequent isolation of nitrifiers from the highest positive dilution is most trustworthy in one sense: the species positively detected or isolated at least occur in a significant high number compared to other culturable species in that particular environment. The information from molecular studies is more uncertain for reasons dealt with in Section 5. However, the presence of unculturable species can be indicated exclusively via application of molecular techniques used in parallel to MPN enrichment and isolation.

Generally, the distribution patterns of the distinct species turned out to reflect the physiological properties observed in laboratory experiments. However, environments like soils or waters are very heterogeneous and thus difficult to define as a habitat in detail. Second, the numerical dominance of distinct species or groups of species at distinct environments will be a transient result of the sum of gradual adaptations to the respective environmental conditions. Hence, the distribution patterns may overlap, even if clear differences in ecophysiological characteristics are recognizable among the species. Thus, for instance, the differentiation among authochtonic and allochtonic species sometimes is problematic.

4.1 The lithotrophic ammonia-oxidizers

The group of ammonia-oxidizers belonging to the β subclass of the Proteobacteria is divers. Within the Nitrosomonas cluster, a large number of phylogenetically definable subgroups exist, revealing more or less different ecophysiological characteristics. Consequently, these lineages are relatively well reflected by different ecological distribution patterns [9]. Comparable clear ecophysiologically based differentiations are missing within the Nitrosospira cluster though phylogenetically definable groups exist.

All cultured strains of the three species within the Nitrosomonas marina lineage and the only isolate of the N. cryotolerans lineage, originate from marine habitats (Atlantic and Pacific Ocean, Mediterranean Sea, North Sea). In accordance with their distribution in nature they are all obligately halophilic revealing NaCl optima between 300 and 400 mM. Most of the isolates were obtained from free water samples, but the type strain of Nitrosomonas marina was enriched from a sediment sample. All 18 isolates of our collection possess urease activity. This might be important due to the ammonia concentration at their natural habitats, which is extremely low with regard to the Ks values around 50 μM NH3 of their ammonia-oxidizing system (unpublished data). The existence of a novel marine species (or group of species) of Nitrosomonas has been indicated by molecular analyses [21,22, but this must be proved by isolation of a representative strain.

The 27 isolates of the three species of the N. europaea lineage as well as the seven isolates of the closely related species Nitrosococcus mobilis, existing in our collection, are halotolerant to moderately halophilic. 23 of these isolates were obtained from sewage disposal plants, where these species also could be detected via molecular investigations [23,24. Two and one isolates, respectively, of N. europaea and N. eutropha, were obtained from freshwater habitats (rivers and lakes) and two N. europaea strains originate from an estuary [25], sometimes obviously traceable to nearby located outlets from sewage disposal plants. Two other N. europaea strains were isolated from fertilized agricultural soils. The only existing laboratory strain of N. halophila as well as four of the seven strains of Nitrosococcus mobilis have been isolated from the North Sea [8,11. Recent molecular investigations have indicated alkaline salt lakes also to be natural habitats for members of the N. europaea lineage [26]. In accordance with their close relationship all these species are uniform in that urease activity is missing and the Ks values of their ammonia-oxidizing systems are above 30 μM NH3. These ecophysiological characteristics are in agreement with the fact that all species of this lineage seem to prefer eutrophic habitats where an alternative substrate besides free ammonia is not needed. In sewage disposal plants, representatives of this grouping generally were found being embedded in flocs or biofilms, though they are not typically self-flocculating in pure cultures.

Another phylogenetic lineage within Nitrosomonas is represented by N. ureae and N. oligotropha together with two not yet described species [25]. Most of the isolates were obtained from oligotrophic freshwater environments (rivers and lakes), some originate from natural and often moderately acid (pH around 6) soils. Recently representatives of this group could be detected in the river Schelde via molecular analysis [27]. Among the 36 strains of these four species, being investigated in our laboratory, only one was found to be urease-negative. All revealed very low Ks values of the ammonia-oxidizing system between 1.9 and 4.2 μM NH317,25. Pure cultures were observed all to be more or less self-flocculating and this is in accordance with the observation that they were often found floc- or biofilm-attached in their natural habitats [25]. All strains were sensitive to increasing salt concentrations.

The sixth lineage of Nitrosomonas comprises two subgroups. One of these subgroups is represented by N. communis and two not yet described species. These species seem not to occur in acid soils (below pH 6). Without exception all the 18 strains existing in our collection were enriched from neutral and often agricultural soils. The Ks value of the ammonia-oxidizing system was high for all three species, ranging from 14 to 43 μM NH3. They are also uniform in that all strains were found to be urease-negative. The second subgroup of this lineage is represented by a single species, N. nitrosa. The Ks value of ammonia-oxidation is at the same level as estimated with the first subgroup, but the isolates of N. nitrosa are different from the strains of the other subgroup in that all 14 isolates of our collection possess urease and, with only one exception, were isolated from eutrophic freshwater environments. Recently, using the molecular amoA approach representatives of this lineage could be detected in waste water treatment plants. In most other molecular investigations insufficient coverage of the primers used could have led to negative results [28].

Distribution patterns of the representatives of the Nitrosospira cluster are less clearly definable. The existence of at least nine species could be established and a taxonomic structure within the cluster was indicated by different morphology of the cells as well as by different G+C contents of the DNA [29]. Ecophysiologically sufficiently discriminating characteristics are missing as yet and differentiation among species via molecular approaches is not easy due to the very close relationship between the species of this cluster. Obviously the members of this assemblage represent a phylogenetically very young branching not sufficiently reflected in 16S rRNA gene sequences. Nevertheless, the distribution patterns are different between the two species of the genus Nitrosolobus on the one side and the species of the other two genera, Nitrosospira and Nitrosovibrio, on the other side. 16 of a total of 20 isolates of Nitrosolobus existing in our collection originate from neutral agricultural soils. In contrast, most of our 63 strains of Nitrosospira and Nitrosovibrio were obtained from natural and often acid soils (heathlands, grasslands, forest soils), from rocks (natural as well as building stones) and, occasionally, from oligotrophic freshwater. If strains have been isolated from agricultural soils, these soils without exception have been acid (generally the pH was between 4 and 6). Another distinguishing characteristic is the ability of urease-positive strains of Nitrosospira and Nitrosovibrio to exist in acid soils at pH values around 4 [30,31. This property seems to be missing with urease-positive strains of Nitrosolobus. Members of the Nitrosospira cluster were also found to be ubiquitous in natural soils via molecular methods [32]. However, it seems not to be possible at present to distinguish among Nitrosolobus and the other two genera of this cluster, Nitrosospira and Nitrosovibrio, via PCR-assisted methods. Several molecular ecological investigations have led to the suggestion that members of the Nitrosospira cluster generally might be the most ubiquitously distributed ammonia-oxidizers in nature [3234. However, in many of these PCR-assisted analyses an underestimation of the actual abundance of representatives of Nitrosomonas obviously has been caused by insufficient coverage of the primers used [28]. From other molecular investigations, the existence of a marine Nitrosospira species (or group of species) is indicated [35,36. However, these observations have not yet been proved by the isolation of a representative.

The two representatives of the genus Nitrosococcus (γ subclass of the Proteobacteria) have a relatively strong salt requirement. The NaCl optimum was estimated to be 500 and 700 mM with N. oceani and N. halophilus isolates, respectively. The distribution of N. oceani seems to be restricted to marine environments. Strains of this species were repeatedly enriched and isolated from offshore waters of the North and South Atlantic Ocean. The distribution of N. oceani in marine environments has been confirmed by molecular analyses [37]. The cell density at these habitats was calculated to be less than one cell per liter [38]. As the Ks value for ammonia-oxidation is relatively high compared with the substrate concentrations in the open oceans [39], it might be of importance that urea can serve as an alternative energy source. Strains of the second species of this grouping, N. halophilus, were isolated from sediment samples from a salt lake in Saudi Arabia and a salt lagoon in the Mediterranean Sea. Only small amounts of inoculum were needed for successful enrichments. This indicates the higher cell density of ammonia-oxidizers in sediments most probably caused by higher levels of ammonia concentrations compared with the overlaying water column where the abundance of ammonia-oxidizers was calculated to be less than one cell per liter [38]. Urease activity is missing with both cultured N. halophilus isolates. In enrichment cultures, but not in pure cultures, the production of cell aggregates (cysts) was observed with N. oceani[38].

4.2 The lithotrophic nitrite-oxidizers

Among the lithotrophic nitrite-oxidizers the distribution of the single species lineages represented by Nitrococcus mobilis and Nitrospina gracilis, respectively, seem to be restricted to marine environments. The only strains kept in culture were isolated from the South Pacific and from the South Atlantic Ocean, respectively. In laboratory experiments neither of these strains were able to grow in freshwater mineral salts water even if NaCl was added [14]. Optimum growth was obtained in 70–100% seawater-based media. Evidence of the presence of these species in the environment via molecular techniques is missing as yet.

Strains of the three Nitrobacter species, N. winogradskyi, N. hamburgensis and N. vulgaris are widely distributed in freshwater and soils. Isolates were obtained from rivers, lakes, sewage disposal plants, divers soils, building stones and even from brackish and marine water samples. There is no difference discernible of distribution patterns of the above-mentioned three closely related species. However, a fourth species of this genus, N. alkalicus that is very closely related to the other three species seems to differ with regard to its ecophysiological characteristics. The only available strains have been isolated from sediments of soda lakes. These strains were especially characterized by their ability to grow at pH values up to around 10 [13]. However, in our culture collection, N. alkalicus grows well in standard media revealing pH values around 8.

The Nitrospira group appears to be highly divers. The first species described, Nitrospira marina, was found to be obligately halophilic. This is in accordance with its obligately requirement of 70–100% seawater in the growth media [15]. The species appeared to be ubiquitous in ocean environments [15]. Another species, N. moscoviensis, was isolated from a corroded iron pipe of a heating water system [16]. Salt requirement was not observed and so this species is well adapted to freshwater environments. An ubiquitous distribution of representatives of Nitrospira has already been supposed by Watson et al. [40], since spiral-shaped cells have been observed in enrichments of nitrite-oxidizers from water and sediment samples from the New York and the Woods Hole Harbor, in water samples from the Black Sea, from beach sands and salt marshes on Cape Cod, from brackish waters in the Savannah and Mississippi Rivers and from various other samples. Nitrospira generally seems to be obligately lithotrophic and seems to prefer relatively low nitrite concentrations. Representatives of the genus Nitrospira have also been detected in sewage disposal plants [20] and nitrifying fluidized bed reactors [41] via in situ hybridization. In sewage disposal plants Nitrospira occurs predominantly embedded in cell aggregates such as flocs or biofilms.

5 Methods used with in situ investigations of nitrifying bacterial populations

In situ analyses of nitrifying bacterial communities are limited by the fact that most obviously species existing in nature are not necessarily available in culture.

A common classical method to enumerate culturable nitrifiers is the MPN technique [42]. However, this method often leads to a more or less significant underestimation of cell counts, since nitrifiers often occur as cell aggregates. Sufficient homogenization of natural samples seems to be nearly impossible. The presence of unculturable species would also lead to underestimation of the total count. Furthermore, MPN counts will not say which species are present at the sample site. A subsequent isolation of species present in high positive MPN dilutions will allow a much more clear statement about the nitrifying community. From our experience the enrichment conditions can be defined suchlike that distinct species will not be favored significantly above others if modified media are used in parallel reflecting variable environmental parameters as, for example, substrate concentration or salinity [11,29. This suggestion is corroborated by the fact that in our culture collection all the described species are represented at a similar frequency. However, it cannot definitely be excluded that species of nitrifying bacteria exist, that are unculturable at the employed growth conditions or that some slow growing species even could be outnumbered in positive MPN dilutions by other species being present at the same frequency but growing significantly faster. Nevertheless, only those strains that have been isolated can be defined with certainty at the species level via sequencing of the 16S rDNA and subsequent DNA–DNA hybridization [43].

Immunological analysis of nitrifying populations using the immunofluorescence technique is another tool [44,45. However, with many species of the nitrifying bacteria antigenic variances are common within natural populations. Therefore, application of immunofluorescence requires prior isolation of target species from the site under investigation, to produce adequate antibodies. More recently, the application of another immunological method could be demonstrated. Nitrite-oxidizing bacteria could be detected on the genus level if using monoclonal antibodies targeting different subunits of the nitrite oxidoreductase [46,47]. This method was employed successfully directly via immunofluorescence microscopic technique as well as indirectly via immunoblotting.

Several molecular methods are useful for in situ analysis of nitrifying bacterial populations. However, complete molecular data are available only with those species that are in culture. If the required molecular data are available [48], fluorescence in situ hybridization with rRNA-targeted oligonucleotide probes will allow direct detection of distinct species at the single cell level [49]. In combination with confocal laser scanning microscopy this technique has successfully been employed to analyze the spatial distribution of distinct species of nitrifying bacteria within flocs and trickling filter biofilms in sewage disposal plants [49]. If using primer-combinations of known specificity 16S rRNA-encoding genes of nitrifying bacteria can be amplified via PCR from DNA obtained directly from natural samples. Cloning and subsequent sequencing of the amplificates allows sufficient analysis of a nitrifying bacterial community [21,35. Separation of 16S rDNA amplificates via denaturing gradient gel electrophoresis and subsequent sequencing of the separated DNA fragments is an alternative way [49,27. However, by these molecular approaches, the nitrifiers often cannot be identified with certainty at the species level. Especially within groups comprising closely related species, the degree of 16S rDNA sequence diversity generally is too low for a certain identification. Another problem has become evident by a recent study. It could be demonstrated that none of the primers and probes targeting the 16S rRNA or the 16S rRNA-encoding gene of ammonia-oxidizers which are available at the time being does cover all the cultured species along with being 100% specific for these nitrifiers [28]. Blot hybridization techniques allow the quantification of specific DNA signature sequences from natural samples [32,50. Genes encoding rRNA, as well as functional genes representative of defined physiological activities, can be quantified. With ammonia-oxidizing bacteria, the gene encoding the ammonia monooxygenase may serve as suchlike functional gene [24,51. Primers of the amoA gene exist that have successfully been used for cultivation-independent analyses of ammonia-oxidizer diversity in the environment [28]. Theoretically, if using a suchlike approach not only the abundance but also the activity of the target cells become measurable. But in reality, it seems not to be possible to extrapolate from obtained relative quantities of definable DNA fractions to absolute cell counts in the sample, since cellular DNA fractions generally are not stable in quantity due to changing metabolic activities of the cells. Another problem could be a different number of copies per genome of the target genes.

6 Conclusions

To explore community structures of nitrifying bacteria in their natural habitats remains complicated. The classical way of enrichment and isolation of the organisms is relatively time consuming. However, this way seems to be unavoidable to reflect the actual diversity of species involved in the process and, especially, to define their ecophysiological properties. Generally, molecular ecological methods can be used as excellent alternative techniques. The existence of not yet cultured or even unculturable species in nature could be proved if using molecular techniques and indeed some of such indications exist in the literature [21,35. However, in part there is only a limited degree of 16S rDNA sequence diversity within environmental nitrifier sequence clusters due to the close relationships among the involved species. Therefore, in many cases it cannot be stated with certainty how many distinct species are represented by the obtained sequences or whether the presence of an as yet uncultured species is indicated [28].

In the future, polyphasic approaches combining both, classical and molecular methods, should be applied for in situ analyses of natural nitrifying community structures.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
View Abstract