Scale-dependence and mechanisms of dispersal in freshwater zooplankton

Communities of organisms form as a result of both interspecific and abiotic interactions within local habitat patches and dispersal among patches in a region. Local processes are expected to play a dominant role when dispersal occurs much more often than extinction. We performed two field experiments to examine rates and mechanisms of dispersal in freshwater pond zooplankton communities. First, we tested the effect of distance from a source on the rate of colonization of artificial habitat by placing wading pools at 5, 10, 30 and 60 m from two natural fishless ponds and observing the succession of zooplankton. Seventy-eight percent of the species in the source ponds that were capable of living in the pools colonized at least once during the experiment. A new species was found in the pools on average once every four days, suggesting that colonization events occur on the order of days to weeks for many species. Colonization rates declined further from the source at one pond but not the other, and the effect of distance was relatively weak at both ponds. This suggests that many species disperse broadly over short distances. The second experiment tested the role of animal vectors for zooplankton dispersal by restricting access to the pools. Eight treatments were imposed that excluded potential animal vectors along a body size gradient from large mammals to small insects. While the treatments affected zooplankton colonization, many species invaded even when all animals larger than 1 mm were excluded. Animal vectors may therefore be less important for dispersal than wind. Our results suggest that zooplankton are highly effective dispersers over short distances, and can disperse via several mechanisms. Local interactions should therefore play a dominant role in structuring these communities at small regional scales.     

Are distribution patterns linked to dispersal mechanism? An investigation using pond invertebrate assemblages

1. Dispersal ability influences the distribution and abundance of organisms, but empirical investigations of the relationship between dispersal ability and the composition of ecological assemblages are scarce. Here, we compare between-site variation in the species richness and community composition of actively and passively dispersing pond invertebrates.
2. Coleoptera (active dispersers) and microcrustacea (passive dispersers) were sampled over a season from 16 ponds within a 4-km radius in south-west England. Species richness and community composition were related to environmental variables using regression and Canonical Correspondence Analysis (CCA), respectively.
3. Coleopteran species richness was significantly and positively correlated with pond permanence and maximum area, whereas microcrustacean species richness was relatively equal across sites and did not correlate with environmental variables. The frequency of species' occurrence between sites was the same for both groups, which suggests that active and passive dispersers exhibited the same degree of dispersal.
4. Between-site variation in community composition was non-random for both groups, with pond permanence and area, together, explaining similar proportions of between-site variation for Coleoptera. Permanence was correlated most strongly with microcrustacean community composition and a high proportion (25%) of microcrustacean species were more numerous in smaller, more ephemeral ponds.
5. These data suggest that, at small spatial scales, Coleoptera which can undertake multiple dispersal events, are more likely to colonise large, more permanent ponds than passively dispersing microcrustacea. For microcrustacea, other traits (in this case those permitting existence in ephemeral habitats) may over-ride the influences of dispersal in driving between-site variation in species composition.     

Connectivity and cladoceran species richness in a metacommunity of shallow lakes

SUMMARY 1. The shallow ponds of the nature reserve 'De Maten' form a metacommunity, in which individual ponds are highly interconnected via a system of overflows and rivulets. This study reports on the relations between cladoceran species richness and (a) connectivity patterns and (b) local environmental variables.
2. No relation was found between local species richness and three connectivity variables or dispersal pathways.
3. Spatial configuration was related to richness, but was confounded by environmental variables for 2 of 3 years. In those 2 years, there was a significant linear relation between Secchi disc depth and species richness, suggesting an important impact of the clearwater/turbid state alternative equilibria in shallow lakes in determining cladoceran richness. Only in the year in which environmental variables were unimportant did connectivity between the ponds influence species richness.
4. These results suggest that local environmental variables related to the clearwater/turbid state alternative equilibria in shallow lakes are important in determining cladoceran species richness. Connectivity and dispersal of individuals between the different ponds only act secondarily by increasing the general species richness within a pond through dispersal from ponds with different environmental conditions.     

Strong spatial influence on colonization rates in a pioneer zooplankton metacommunity

The magnitude of community-wide dispersal is central to metacommunity models, yet dispersal is notoriously difficult to quantify in passive and cryptic dispersers such as many freshwater invertebrates. By overcoming the problem of quantifying dispersal rates, colonization rates into new habitats can provide a useful estimate of the magnitude of effective dispersal. Here we study the influence of spatial and local processes on colonization rates into new ponds that indicate differential dispersal limitation of major zooplankton taxa, with important implications for metacommunity dynamics. We identify regional and local factors that affect zooplankton colonization rates and spatial patterns in a large-scale experimental system. Our study differs from others in the unique setup of the experimental pond area by which we were able to test spatial and environmental variables at a large spatial scale. We quantified colonization rates separately for the Copepoda, Cladocera and Rotifera from samples collected over a period of 21 months in 48 newly constructed temporary ponds of 0.18–2.95 ha distributed in a restored wetland area of 2,700 ha in Doñana National Park, Southern Spain. Species richness upon initial sampling of new ponds was about one third of that in reference ponds, although the rate of detection of new species from thereon were not significantly different, probably owing to high turnover in the dynamic, temporary reference ponds. Environmental heterogeneity had no detectable effect on colonization rates in new ponds. In contrast, connectivity, space (based on latitude and longitude) and surface area were key determinants of colonization rates for copepods and cladocerans. This suggests dispersal limitation in cladocerans and copepods, but not in rotifers, possibly due to differences in propagule size and abundance.     

Multiple effects of land‐use changes impede the colonization of open water in fen ponds

Question: Dutch fen areas have become embedded in intensively used landscapes, resulting in biodiversity loss. Hence, plant species that colonize open water inducing the formation of species‐rich floating peat mats have disappeared. Despite many restoration efforts, they have not returned. Is natural succession towards floating mats impeded by site conditions, dispersal limitations or changed biotic interactions? Location: Six Dutch fen reserves: De Deelen, De Weerribben, De Wieden, Westbroek, Molenpolder and Terra Nova. Methods: In 62 fen ponds we determined plant species richness and expansion into open water. We related these to habitat quality (chemical composition of soil and surface water, pond morphology), dispersal potential (distance to remnant populations, likelihood of dispersal) and biotic interactions (presence of muskrats [Ondatra zibethicus L.] and the keystone species Stratiotes aloides). Results: Factor analysis showed that plants expanded further into open water and bank vegetation had higher species richness in areas with older ponds and lower muskrat densities. Locally, high turbidity hampered colonization. Whenever the water was clear, colonization was higher in shallow ponds, and in deep ponds only if Stratiotes was present. Species richness was negatively correlated to nutrient availability in soil and positively correlated to hydrological isolation (decreased sulphate concentrations). We also found that species richness was higher in sheltered banks. Conclusions: Multiple habitat characteristics (turbidity, water depth, nutrient and sulphate concentrations) and the influence of muskrats and Stratiotes all play a role in the lack of restoration success in Dutch fen ponds. Dispersal limitations seem to be overruled by habitat limitations, as colonization often fails even when sufficient propagule sources are present, or when connectivity is high.     

Invertebrate communities and turnover in wetland ponds affected by drought

1. Changes to the macroinvertebrate fauna found in small ponds on a freshwater marsh (Aberlady Bay, Scotland) in 1986, 1987 and 1992 were used to assess the impact of the 1988–92 drought on taxon turnover in pond communities. 2. Permanent ponds accumulated taxa over the study period. Ponds that were wet throughout 1986–87 but dried in 1992 lost some of the many taxa associated with permanent water but acquired a reduced fauna typical of temporary ponds. Ponds that were temporary in 1986–87 were dry during most of 1992 and lost almost all aquatic taxa. 3. Faunal turnover was considerable even in permanent ponds. Extinction rates were high for taxa typical of permanent or temporary ponds. Colonization rates were poor for the taxa from permanent water, but high for the taxa from temporary ponds. 4. The importance of colonization and extinction rates as main predictors of the distribution of species between the ponds was looked at using metapopulation incidence functions, where observed incidence of a taxon can be used to predict colonization and extinction rates. Predicted rates were then compared with observed rates. Incidence functions gave reasonable predictions of observed colonization rates but were poor predictors of extinction, even for taxa that appeared likely to be true metapopulations. 5. For the pond fauna, including fugitive species adapted to temporary ponds, whilst colonization may well depend on environmental stochasricity (how long a pond holds water), subsequent survival depends on other, demographic, processes (e.g. finding a mate, predation) rather than the pond drying out.     

Spatio-temporal nested patterns in macroinvertebrate assemblages across a pond network with a wide hydroperiod range

Nestedness has been widely used to measure the structure of biological communities and occurs when species-poor sites contain subsets of species-rich ones. Here, we examine nested patterns across the macroinvertebrate assemblages of 91 ponds in Doñana National Park, Spain, and explore temporal variation of nestedness and species richness in 19 temporary ponds over 2 years with differing rainfall. Macroinvertebrate assemblages were significantly nested; both pond spatial arrangement and environmental variation being important in driving nested patterns. Despite the nested structure observed, a number of taxa and ponds deviate from this pattern (termed idiosyncratic), by occurring more frequently than expected in species-poor sites, or having assemblages dominated by species largely absent from species-rich sites. Aquatic adults of winged insects, capable of dispersal, were more highly nested than non-dispersing taxa and life-history stages. Idiosyncratic taxa were found in ponds spanning a wide range of hydroperiods, although nestedness was higher in more permanent waterbodies. Monthly sampling demonstrated a gradual increase of species richness and nestedness from pond filling to April–May, when the most temporary ponds started to dry. Although the degree of nestedness of individual pond assemblages varied from month to month, the overall degree of nestedness in the two study years was practically identical despite marked differences in hydroperiod. Our results suggest that differential colonization and environmental variation are key processes driving the nested structure of Doñana ponds, that macroinvertebrate assemblages change in a predictable manner each year in response to cycles of pond wetting and drying, and that connectivity and environmental variability maintain biodiversity in pond networks.     

Abundance, distribution and life histories of grassland plants: a comparative study of 81 species

1 This study examines the abundance and distribution of grassland plant species in particular relation to features affecting colonization. Seed production (inversely related to seed size) and recruitment success (positively related) affect colonization ability, suggesting that seed size can be used as a key trait.
2 Data on seed size, dispersal mode, life form, geographical range size and abundance were gathered for 81 grassland plant species in a field study area in Sweden. Seed production and plant size were estimated for 69 of these species. Analyses were performed both across species, with species treated as independent data points, and for 43 'phylogenetically independent contrasts'.
3 The cross-species analyses suggested that local abundance was related to life forms but not dispersal or plant size. Perennials were generally most abundant, as were clonal species. If abundance reflects colonization we predicted that species with intermediately sized seeds (or intermediate seed production) would be most abundant, and this was supported by the phylogenetic contrast but not by cross-species analyses. In the former analysis, a high abundance of species was significantly associated with a small seed size deviation (and seed number deviation) from the median values of these traits in the community.
4 Local abundance, seed production and seed size deviation from the community median value were positively related to geographical range size in the cross-species analysis, but no relationships were seen in the phylogenetic contrast analysis.
5 We conclude that colonization processes do have a significant influence on abundance patterns in grasslands. Seed size is a key trait for colonizing ability, and the effects of the trade-off of seed size vs. seed number must be considered. No single mechanism can be identified that influences both abundance and geographical distribution range.     

Dispersal characteristics of seven odonate species in an agricultural landscape

Dispersal is an ecological phenomenon which is of fundamental importance to population biology. While dispersal behaviour of many orders of winged insects has received a great deal of attention, the dispersal characteristics of odonates have been poorly documented. We used capture-mark-recapture techniques to study dispersal behaviour of seven species of odonates breeding on a network of 11 small ponds in Cheshire, U.K. The ponds ranged in size from 615 to 1300 m²- and varied from 30 to 860 m apart. We found surprisingly high rates of dispersal between ponds, with 10–47% per species of recaptured individuals moving from their natal pond. The mean probability of dispersal differed significantly among species but the relationship between the probability of dispersal and distance moved consistently followed a simple negative exponential curve for all species. Most individuals stayed at their natal pond, but a few moved long distances. Neither the age at which an individual was marked (teneral vs sexually mature) nor its sex significantly affected its tendency to disperse. The negative exponential relationship suggests that dispersal should be relatively easy to incorporate in more complex models of odonate spatio-temporal dynamics. To our knowledge, this is the first large-scale, multi-species study to assess dispersal behaviour of odonates by direct observation.     

Temporal dynamics of aquatic communities and implications for pond conservation

Conservation through the protection of particular habitats is predicated on the assumption that the conservation value of those habitats is stable. We test this assumption for ponds by investigating temporal variation in macroinvertebrate and macrophyte communities over a 10-year period in northwest England. We surveyed 51 ponds in northern England in 1995/6 and again in 2006, identifying all macrophytes (167 species) and all macroinvertebrates (221 species, excluding Diptera) to species. The alpha-diversity, beta-diversity and conservation value of these ponds were compared between surveys. We find that invertebrate species richness increased from an average of 29.5 species to 39.8 species between surveys. Invertebrate gamma-diversity also increased between the two surveys from 181 species to 201 species. However, this increase in diversity was accompanied by a decrease in beta-diversity. Plant alpha-, beta- and gamma-diversity remained approximately constant between the two periods. However, increased proportions of grass species and a complete loss of charophytes suggests that the communities are undergoing succession. Conservation value was not correlated between sampling periods in either plants or invertebrates. This was confirmed by comparing ponds that had been disturbed with those that had no history of disturbance to demonstrate that levels of correlation between surveys were approximately equal in each group of ponds. This study has three important conservation implications: (i) a pond with high diversity or high conservation value may not remain that way and so it is unwise to base pond conservation measures upon protecting currently-speciose habitats; (ii) maximising pond gamma-diversity requires a combination of late and early succession ponds, especially for invertebrates; and (iii) invertebrate and plant communities in ponds may require different management strategies if succession occurs at varying rates in the two groups.     

Measuring dispersal in a metapopulation using stable isotope enrichment: high rates of sex-biased dispersal between patches in a mayfly metapopulation

Dispersal affects a wide array of ecological and evolutionary processes, but has been difficult to estimate empirically. A ¹⁵Nitrogen stable isotope enrichment technique was used to passively mark all developing Callibaetis ferrugineus hageni (Eaton) mayfly larvae in a beaver pond that had previously been shown to be a patch in a source-sink metapopulation. After enrichment during the larval stages, dispersal among ponds by adult females was demonstrated by the presence of unmarked females ovipositing in the labeled pond, and marked females in an unlabeled pond. Observed frequencies of marked females suggested incomplete mixing between ponds. In contrast, males rarely dispersed from their natal pond, which was consistent with the unusual mating system in this species – adult Callibaetis are short-lived, do not feed, and females are sexually receptive immediately after emerging from the larval habitat. The frequent dispersal demonstrated using the stable isotope technique was a critical component of the source-sink dynamic observed in this metapopulation, and further use of this technique will provide insights into patterns of dispersal in spatially structured habitats.     

Freshwater snail diversity: effects of pond area,habitat heterogeneity and isolation

Summary A large number of eutrophic ponds were surveyed for the presence of freshwater gastropods. Factors thought to influence the distribution of the snails were evaluated. As the investigated area has a homogeneous geological background physicochemical factors probably have a low effect on the local distribution of snails. There was a significant, positive regression between pond area and the number of gastropod species in the pond, but the regression only explained a minor part of the variation in species numbers. Multiple regression of an extended number of variables, associated with habitat complexity and dispersal, indicated that, in addition to area, macrophyte diversity and the mean number of gastropods in the 5 closest ponds (S₅) were important in explaining the distribution of gastropods. An increase in the number of macrophyte species increases the number of available microhabitats and refuges from predators. S₅ probably influences the dispersal rates between ponds. The gastropods in this area are thought to have additional dispersal modes, besides aerial dispersal with birds, and this probably increases the immigration rates and/or decreases the extinction rate.     

Benthos and zooplankton of coal strip mine ponds in the mountains of northwestern Colorado,U.S.A.

Zooplankton, benthos and associated physico-chemical conditions were sampled in coal strip-mine ponds in northwestern Colorado from June 1977 to May 1978. Two spoils ponds derived all of their drainage from the coal mine, but differed in age; one pond received only partial drainage from mine spoils; a control pond was located in an adjacent drainage basin. Acid mine drainage was not observed, although total dissolved solids (TDS), nitrate and sulfate values were higher in the spoils ponds than in the control pond. Net zooplankton were less abundant in the spoils ponds, although standing stock of benthos exhibited a progressive decrease from the youngest spoils pond to the control pond. Zooplankton and benthos species diversity values were lower in the ponds affected by mine drainage. Certain groups of zooplankters (Cladocera) and benthos (Trichoptera, Amphipoda, Hydracarina, and Sphaeriidae) were rare or absent in the youngest spoils pond. Percentage Similarity Coefficients indicate that colonization phenomena (age and distance from a source of colonizers) may be responsible, in part, for the faunal differences between ponds, since the higher levels of nitrate, sulfate, and TDS in the spoils ponds were generally within the tolerance range of aquatic organisms.     

The diversity of macro invertebrate and macrophyte communities in ponds

SUMMARY 1. The range of factors which may influence the wide variation in the structure of pond communities is reviewed.
2. Factors explaining the variation in numbers of macro invertebrate and macrophyte taxa are investigated, using multiple regression analysis, for a group of ponds in which biogeographical effects are likely to be minimal.
3. Numbers of macro invertebrate taxa in these ponds are highly correlated with pH and probably, therefore, with other associated aspects of water chemistry,
4. Many invertebrate taxa, from species to entire phyla, were not found below pH 5.5, but few characteristically'acidophilic'taxa were apparent.
5. Details of species composition of the macro invertebrate fauna differ widely even between ponds of similar pH.
6. The diversity and composition of macrophyte communities are not adequately predicted by any of the factors investigated.     

Lotic dispersal of lentic macroinvertebrates

Little is known on dispersal of lentic macroinvertebrates. We quantified dispersal of lentic macroinvertebrates through pond connections in a highly connected pond system, and investigated how dispersal rates were affected by connection properties and time of the day (day, night). Furthermore, by comparing the composition of assemblages of dispersing macroinvertebrates with the macroinvertebrate assemblages of source ponds, we tested whether dispersal was neutral or a taxon-specific process. We found that many taxa dispersed through the pond connections. Taxa richness of the dispersing macroinvertebrate assemblage was proportional to taxa richness in the source ponds. The number of individuals that dispersed, however, was not related to source pond densities, possibly because of the highly patchy distribution of lentic macroinvertebrates within ponds. Elevated dispersal rates were recorded for Baetidae, Chironomidae and Physidae, indicating a taxon-specific use of pond connections as dispersal pathway. None of the physical properties of the connections affected dispersal. Macroinvertebrates dispersed more during the night than during daytime. Of seven tested families, Chaoboridae and Chironomidae showed higher dispersal during the night, probably resulting from diel vertical migration, since pond connections mainly transported near-surface water. Flying and non-flying invertebrates dispersed equally frequent through pond connections. For non-flying invertebrates that disperse infrequently over land, dispersal through pond connections may provide an important additional means of dispersal.     

Environmental factors affecting the distribution of aquatic invertebrates in temporary ponds in Mammoth Cave National Park,Kentucky, USA

Despite a recent surge of interest in temporary lentic systems, a strong theory linking the biota to its environment has not emerged. Using data from 10 temporary ponds at Mammoth Cave National Park, Kentucky, USA, we investigated how invertebrate communities were structured along environmental gradients, both between and within ponds. Samples were collected with a benthic corer in winter and spring, and a sweep net in spring. Six between-pond and two within-pond datasets were created. Between-pond analyses yielded significant CCA’s with only one of the six data sets. The ranges of environmental variables (EV’s) within ponds were often similar to the ranges of EV’s when averaged and compared between ponds. Some taxa were aggregated in a single pond, and richness increased with pond area. The theory that richness increases with hydroperiod did not apply to these systems. Within-pond analyses yielded more consistent relationships, with both CCA’s being significant. Sample depth was the best predictor of invertebrate richness and abundance, with most taxa preferring shallow habitats. Richness and abundance were higher in both shallow ponds and shallow areas of deep ponds than in deep areas of deep ponds. Standardizing sample depth may be an effective way to remove this gradient as a confounding variable in future research. The presence of within-pond gradients, possibly coupled with the limited dispersal and random colonization of tolerant taxa, makes between-pond comparisons difficult. Electronic supplementary material Supplementary material is available in the online version of this article at and is accessible for authorised users. Handling editor: S. Declerck     

Species richness and succession of aquatic beetles in some kettle-hole ponds in northern Sweden

Ten closely-located kettle-hole ponds in northern Sweden were ordered along a vegetational, successional gradient from the present-day representation of the six vegetational units distinguished. Different rates of vegetational development caused by differences in depths of basins have made possible the long-term existence of early successional stages. A fact that has offered the possibility of studying most of the vegetational succession of these habitats.
The number of water beetle species inhabiting each pond were correlated to successional stage and area of pond. Mean species numbers decreased from 34 to 14 per pond with increasing maturity of the vegetational succession. This decrease in species richness of water beetles is caused by a decrease in number of habitats or vegetational units present. The increased floral conformity is mainly due to a more stable water-level caused by lower rates of bottom seepage from the accumulation of organic deposits. Besides the extended zonation of littoral vegetation, a high water-amplitude is also connected with recurrent nutrient pulses caused by the faster detritus composition during drought periods. Thus the changing water-levels contribute to a high water beetle diversity through the maintenance of both a high habitat complexity and a high production.
A positive correlation between pond size and number of water beetle species were only found if the fen ponds were excluded. From this it is concluded that the area effect is here a good predictor of species richness only within ponds of a similar successional stage.     

Connectivity, scale-dependence, and the productivity–diversity relationship

We surveyed freshwater ponds (localities) nested within watersheds (regions) to evaluate the relationship between productivity and animal species richness at different spatial scales. In watersheds where the ponds were relatively distant from one another (likely reducing the level of interpond dispersal of many organisms), we found a scale‐dependent productivity–diversity relationship; at local scales (among ponds), diversity was a hump‐shaped function of productivity, whereas at regional scales (among watersheds), diversity monotonically increased with productivity. Furthermore, this relationship emerged because there was a strong relationship between productivity and pond‐to‐pond species compositional differences. Alternatively, in watersheds where ponds were relatively close together (likely leading to higher rates of dispersal of many organisms), we found no scale‐dependence; diversity was a hump‐shaped function of productivity at both local and regional scales. Here, the relationship between species compositional dissimilarity and productivity was much weaker. We conclude that whether or not scale‐dependence is observed in productivity–diversity relationships will depend, at least in part, on the degree of connectivity among localities within regions.     

Assembly of zooplankton communities in newly created ponds

1. The creation or severe disturbance of habitat patches is generally followed by a phase of community (re)assembly. After such an event, the trajectory of community assembly in habitat patches may be highly variable because of stochasticity during the dispersal and colonization process. Conversely, assembly patterns may also be deterministic if communities are shaped by prevalent environmental conditions in the habitat patches (species sorting), or by systematic differences in the dispersal capacities of species. 2. In this study, we investigated the pattern of community assembly of zooplankton species in 25 newly created ponds at 13 different sites in Flanders (Belgium). Over a period of three consecutive years, we assessed at what rate and with what frequency species of the regional species pool colonized the newly created ponds. We also studied the development of community structure over time and tested whether the dynamics were consistent across different ponds at the different locations. In addition, we characterized the dynamics of metacommunity features, such as alpha, beta and gamma diversity in clusters of ponds. 3. Even within the first year after their creation, the new ponds were rapidly colonized by a small subset of species from the regional species pool (Daphnia obtusa, Chydorus sphaericus and Simocephalus vetulus). These species dominated the cladoceran assemblages during the subsequent years. Other species in the regional species pools were only sporadically able to colonize ponds. 4. During the entire study period, we observed no significant shifts in species lists or in the occurrences of species among years. The low incidence of the majority of species may be the result of dispersal limitation or the failure of immigrants to establish due to priority effects exerted by the first colonizers. There was, nevertheless, a consistent change in the relative abundance of species, which was most probably mediated by differences in the hatching time among species influencing species composition in the first year. 5. In contrast to expectations, we observed no increase in average alpha diversity (local species richness) and gamma diversity (total richness of entire pond clusters) during the course of the study period. Beta diversity was relatively low from the beginning and remained constant throughout the study period. These deterministic patterns can mainly be attributed to the dominance of the three first colonizing species and the low success rate of other species in colonizing the ponds.     

The relative importance of dispersal and the local environment for species richness in two aquatic plant growth forms

Local species richness can be affected by both the dispersal process and by environmental conditions (species sorting process). The evaluation of the relative roles of these two processes contributes not only to further understanding of the mechanisms determining species richness but also to biodiversity conservation. We studied the relative importance of hydrological dispersal and water chemistry for species richness of submerged and floating‐leaved macrophytes using 31 sets of interconnected ponds with different numbers of component ponds (defined as connection class). Connection class was slightly more important than, or equally important to, water chemistry in determining species richness of floating‐leaved macrophytes. In contrast, submerged macrophyte richness was much more influenced by water chemistry than by connection class, although increasing connection class had some positive effect. Similarly, the occurrence of a particular species of submerged macrophyte was better explained by pond water chemistry than by the occurrence of the same species in the pond immediately upstream. The reverse was true for floating‐leaved macrophytes; the presence of a given species was better explained by its presence in the pond immediately upstream than by water chemistry. These results indicated that the relative importance of the two processes that shape the species richness of aquatic plants is a consequence of the growth form of the plants. However, both the dispersal process via hydrologic connection and species sorting by water chemistry play some role in determining the species richness of both floating‐leaved and submerged macrophytes.