Stream macroinvertebrate response to catchment urbanisation (Georgia, U.S.A.)

SUMMARY 1. The effects of catchment urbanisation on water quality were examined for 30 streams (stratified into 15, 50 and 100 km² ± 25% catchments) in the Etowah River basin, Georgia, U.S.A. We examined relationships between land cover (implying cover and use) in these catchments (e.g. urban, forest and agriculture) and macroinvertebrate assemblage attributes using several previously published indices to summarise macroinvertebrate response. Based on a priori predictions as to mechanisms of biotic impairment under changing land cover, additional measurements were made to assess geomorphology, hydrology and chemistry in each stream. 2. We found strong relationships between catchment land cover and stream biota. Taxon richness and other biotic indices that reflected good water quality were negatively related to urban land cover and positively related to forest land cover. Urban land cover alone explained 29–38% of the variation in some macroinvertebrate indices. Reduced water quality was detectable at c. >15% urban land cover. 3. Urban land cover correlated with a number of geomorphic variables such as stream bed sediment size (–) and total suspended solids (+) as well as a number of water chemistry variables including nitrogen and phosphorus concentrations (+), specific conductance (+) and turbidity (+). Biotic indices were better predicted by these reach scale variables than single, catchment scale land cover variables. Multiple regression models explained 69% of variation in total taxon richness and 78% of the variation in the Invertebrate Community Index (ICI) using phi variability, specific conductance and depth, and riffle phi, specific conductance and phi variability, respectively. 4. Indirect ordination analysis was used to describe assemblage and functional group changes among sites and corroborate which environmental variables were most important in driving differences in macroinvertebrate assemblages. The first axis in a non‐metric multidimensional scaling ordination was highly related to environmental variables (slope, specific conductance, phi variability; adj. R²=0.83) that were also important in our multiple regression models. 5. Catchment urbanisation resulted in less diverse and more tolerant stream macroinvertebrate assemblages via increased sediment transport, reduced stream bed sediment size and increased solutes. The biotic indices that were most sensitive to environmental variation were taxon richness, EPT richness and the ICI. Our results were largely consistent over the range in basin size we tested.     

Correlates of habitat favourability for benthic macroinvertebrates at five stream sites in an Appalachian Mountain drainage basin, U.S.A.

SUMMARY.

• 1 Based on monthly samples taken over a 1-year period, average density (individuals m^-2). average standing biomass and annual production of benthic macroinvertebrates were estimated at five sites within an Appalachian Mountain drainage basin. Two sites were on first order streams and differed from the three second order sites: they were smaller and more shallow and they were depressed in pH and chemical richness.
• 2 Patterns of abundance of individual taxa, of higher taxonomic groups and of functional (feeding) groups differed according to whether abundance was measured as density, as standing biomass or as annual production. Standing biomass was chosen as the measure of macroinvertebrate abundance because available evidence indicates that only standing biomass is consistently, positively correlated with survivorship, and thus with habitat favourability.
• 3 Two non-insect taxa (the crayfish Cambarus and the snail Leptoxis carinata) dominated standing biomass at each site. Consequently, differences among sites in total macroinvertebrate standing biomass and differences within and among sites in standing biomass of functional groups were determined by differences in estimated standing biomass of these two taxa. Differences in estimates of crayfish standing biomass were consistent with an explanation based on the availability of refuges created by large substrate particles. The abundance of L. carinata appeared to be controlled primarily by water chemistry and possibly secondarily by predators. A number of insect taxa exhibited patterns of standing biomass consistent with hypotheses based on effects of annual depth-flow regimes. Hypotheses based on differences in food resource and on competition appeared, in general, to be inconsistent with observed patterns of macroinvertebrate abundance.
• 4 Contrary to predictions of the River Continuum Concept, the shredder functional group in the Guys Run drainage and in other temperate woodland streams was found to be a minor part of total macroinvertebrate standing biomass. Further, in a majority of small forested stream sites studied to date, standing biomass of grazers has been determined to be greater than that of shredders.

     

A comparison of modern and presettlement pollen from southern Michigan (U.S.A.)

Use of samples of modern pollen to interpret assemblages of fossil pollen requires that the effects of man's impact be identified in the modern pollen. A comparison of the modern and presettlement pollen in 23 short cores taken within the lower peninsula of Michigan reveals a varied effect of man's activities on the vegetation. When obvious indicators of human activity (such as ragweed pollen) are removed from the pollen sum, many differences in the composition of the arboreal pollen are still evident between presettlement and modern records. When compared to presettlement data, birch, elm, and willow pollen in the modern data appear in proportionately higher quantities throughout the state, whereas beech, maple, and hemlock are in lower quantities. Within the region of primeaval pine forests, white pine values are much lower and oak values show an increase. Despite these differences in the pollen record, the modern pollen reflect the same basic patterns in the vegetation as the presettlement pollen. Principal component analysis shows the same patterns to exist in the modern data as are in the presettlement pollen.     

Quantifying the habitat structure and spatial pattern of New Jersey (U.S.A.) salt marshes under different management regimes

Mosquito control measures have resulted in majorstructural alterations of many coastal marshes, withrelatively unknown consequences to larger ecosystemfunctioning. Parallel grid ditching and open marshwater management (OMWM) techniques are purposefullydesigned to alter the hydrological regime and therelative availability and/or characteristics ofstanding water on the marsh surface. With the growingrecognition of the important influence that habitatstructure and the configuration of the marsh-edgeboundary has on nekton utilization of salt marshsystems, the impact of mosquito control managementtechniques on the availability and quality of saltmarsh habitat deserves increased scrutiny. Usingdigital image processing techniques, we completed adetailed mapping effort for a 1800 ha study area inTuckerton, New Jersey to provide a picture ofsubtidal and intertidal habitat availability. Spatialanalysis techniques were used to quantify the relativeamount of habitat types and spatial pattern of theland-water interface under different managementregimes: parallel grid-ditched, OMWM and an unalteredreference. The parallel grid-ditched site had a muchlower incidence of marsh ponds which serve asimportant low tide and over-wintering refuge forcertain species of fish. In comparison, the OMWM sitehad a much greater amount of ponded water habitats.The grid-ditched site had a higher density of marshsurface-to-tidal water interface resulting in a lowermedian distance between the marsh interior and theadjacent tidal channel network. This study serves tobenchmark the habitat structure and spatial pattern ofa highly functioning natural marsh for use as areference site in regional wetland creation orrestoration efforts.     

Light absorption by phytoplankton and chromophoric dissolved organic matter in the drainage basin and estuary of the Neuse River, North Carolina (U.S.A.)

1. We examined the absorption of solar radiation by phytoplankton and chromophoric dissolved organic matter (CDOM) taking into account riparian shading in the rivers, reservoirs, swamps of the Neuse River Estuary and its drainage basin. 2. In the streams, CDOM typically absorbed 55 and 64% of photons in the spectral range of 400–700 nm (photosynthetically active radiation, PAR) and 500–600 nm, respectively. The large proportion of photons absorbed by CDOM indicates high potential for abiotic photochemial reactions in the 500–600 nm region. 3. Despite the high concentration of nutrients, phytoplankton contributed little (2%) to the total absorption of PAR in the streams. Small (<30 m wide) streams typically received only 7% of incident PAR that impinged onto the more exposed reservoirs and estuary. Riparian shading and the low contribution of phytoplankton to the total absorption resulted in conditions where phytoplankton absorbed nearly two orders of magnitude less PAR in the streams than in the estuary and reservoirs. 4. The results indicated that riparian shading and non‐algal absorbing components can significantly restrict phytoplankton production in nutrient‐rich streams with a high concentration of CDOM flowing throughout forested catchments.     

Longitudinal changes in the benthic bacterial community of the Mahoning River (Ohio, U.S.A.)

To examine spatial differences in bacterial communities along the Mahoning River in Northeast Ohio (USA), sediment samples were collected on two dates from three sites. Downstream portions (sites 2 and 3, in this study) of the Mahoning River have been highly impacted by human activities. Two approaches were used to characterize the bacterial community: fluorescent in situ hybridization (FISH) with taxon specific probes and denaturing gradient gel electrophoresis (DGGE). Bacterial numbers (per g ash free dry mass), based on staining with DAPI (4′, 6-diamidino-2-phenylindole) or FISH with Domain Bacteria, Acinetobacter calcoaceticus or Pseudomonas putida-specific probes, were higher at the most upstream site (site 1) compared to the more disturbed downstream sites (sites 2 and 3). In addition, the number of bands based on DGGE was higher at the upstream site (site 1) compared to the two downstream sites (2 and 3) during spring. However, in summer, the number of bands was similar among sites and the most upstream (1) and the middle site (2) had the same average number of bands. In spring, the percent similarity (based on the presence/absence of bands) among the three sites was relatively low compared to summer. In general, differences in the bacterial community were found among the sites with differing levels of anthropogenic disturbance but varied between the two dates examined.     

Pollen frequencies reflect vegetation patterns in a great basin (U.S.A.) mountain range

Five indices are used to quantify the relationship between vegetation and pollen in a mountain range in the arid Great Basin. Computations are based on vegetation coverage and pollen percentages from 63 stands. Association is a measure of whether the presence of the pollen type in a surface sample is an indication of the presence of the parent plant in the local vegetation. Over-representation and under-representation measure tendencies for pollen to occur where the parent plants are absent and vice versa. The correlation coefficient measures the relationship between plant and pollen in stands where both are present. Twenty-nine trees, shrubs, and herbs accurately reflect local vegetation conditions. A percentage diagram shows elevational trends in abundant pollen types. Regional pollen types are used to compute the accumulation rate of pollen in the surface samples. A diagram of pollen accumulation rates shows trends similar to those shown in the percentage diagram. The moss polsters used in this study may collect pollen over a fifteen-year interval.     

Holocene climate and vegetation in the Milford drainage basin,Maine, U.S.A., and their implications for human history

A 9200 ¹⁴C year fossil pollen record from a small kettle lake in central Maine, northeast U.S.A., records the development of nearby upland vegetation throughout the Archaic, Ceramic, and Historic periods of human history. The Early Archaic period (9000 to 8000 B.P.) began as open woodland dominated by Picea, Populus, and Larix, which was replaced by Pinus forest. During the Middle Archaic (8000-6000 B.P.) Tsuga-dominated forest, which developed ca. 7400 B.P., was followed by Pinus forest (ca. 6400 B.P.). The Late Archaic (6000-3000 B.P.) was a period of great transition; Tsuga forest developed again ca. 5700 B.P., but was abruptly replaced by northern hardwood forest ca. 4700 B.P. That Late Archaic expansion of hardwoods would have provided better forage for beaver. Coincidentally, boreal wetland mammals such as beaver (Castor canadensis) and muskrat (Ondatra zibethicus) increase in faunal assemblages of local archaeological sites, while remains of anadromous fish decrease. We postulate that the apparent increase in human populations throughout the region during the Late Archaic may be attributed to an increase in the resource base within both upland and wetland areas resulting from the development of hardwood forest in response to climatic cooling.     

Trophic interactions and benthic animal community structure in the Colorado River, Arizona, U.S.A.

1. Cladophora glomerata is the dominant filamentous green alga in the tailwaters of the Colorado River, U.S.A., below Glen Canyon Dam, but becomes co-dominant with filamentous cyanobacteria, Oscillatoria spp., below the confluence of the Paria River (26km below the dam) where suspended sediments are elevated. 2. Benthic algal assemblages played an important role in the distribution of the amphipod, Gammarus lacustris, in the dam-controlled Colorado River through Grand Canyon National Park, Arizona. Cladophara and G. lacustris showed a weak positive relationship at ten cobble-riffle habitats in the Colorado River from Lees Ferry (25km below the dam) to Diamond Creek (362km downstream), while no relationship was found between Oscillatoria and G. lacustris. 3. The relationship between algal substrata and G. lacustris was tested by a series of in situ habitat choice experiments. G. lacustris showed a significant preference for Cladophora (with epiphytes) over Oscillatoria spp., detritus and gravel in treatment pans at Lees Ferry. 4. Epiphytic diatoms (i.e. food) were the overriding determinant of subtratum choice by G. lacustris in laboratory experiments. Gammarus chose the Cladophora/epiphytic diatom community over sonicated Cladophora with few diatoms. The amphipods also chose string soaked in diatom extract over string without diatom extract. 5. Importance of mutualistic interactions in aquatic benthic community structure is discussed.     

Analysis and structure of four lentic communities on the Colorado (U.S.A.) shortgrass prairie

Four natural lentic habitats on the Central Plains Experimental Range in Colorado were intensively studied physically, chemically, and biologically for 2 yr (1968–1970) and intermittently for 5 (1971–1975). Lynn Lake and Lake George were chosen as examples of ephemeral or temporary sites, while Cottonwood Pond and Spring Pond were chosen as permanent habitats. Chemically, the waters of each site were analytically separated into dissolved and suspended (sestonic) components. The temporary sites were characterized by their exceptionally high seston loads, the presence of Eubranchiopoda, the prevalence of Ca as the major dissolved cation, ana the high Fe content of their sediments. The two permanent sites, on the other hand, had high total dissolved salt content, rich and varied flora and fauna, Na as the prevalent cation in the dissolved fraction, and high Ca content in the sediments. All four of the sites were “seston digesters” with large amounts of wind-blown organic material being entrapped, decomposed, and mineralized. By comparison with adjacent soils the sediments of all four basins were much richer in K, Zn, Fe, Cu, and Mn. Both temporary and permanent sites function as nutrient sinks. Rates of daily and annual phytoplankton primary productivity are reported for the two permanent sites.     

Estimating above-ground biomass and production in mangrove communities of Biscayne National Park,Florida (U.S.A.)

Total above-ground production isusually estimated by a combination of allometry andlitter collection. However, in coastal sites that aretidally influenced, or in juvenile or dwarf forestswhere the crown bases of dominant individuals maybegin within a few decimeters of ground level,estimates of community leaf production that depend onlitter collection may not be feasible. Thus, in thispaper, we present 1) allometric equations that allowaccurate estimation of total above-ground biomass ofthree mangrove species (Rhizophora mangle, Laguncularia racemosa, and Avicennia germinans)in very small to medium size classes, and 2) analternative method of estimating total above-groundproduction that overcomes the limitations of littercollection. The method we employ to estimate mangroveproductivity is an adaptation for woody plantcommunities of a procedure introduced by Dai andWeigert (1996) for grasslands. It incorporates adetailed census of all individuals within fixedsampling plots, along with periodic observations ofmarked leaf cohorts. The method allows the comparisonof biomass allocation patterns among forests thatdiffer widely in physiognomy and physiographicsetting.The method was applied to a South Florida fringemangrove forest in the early stages of recovery fromHurricane Andrew (August 1992), and an adjacent dwarfforest which was not substantially damaged by thestorm. Total above-ground production in the fringeforest from July 1996 through June 1997 was about 3times higher than dwarf forest production,26.1 Mg·ha^-1·yr^-1 vs.8.1 Mg·ha^-1·yr^-1, respectively. Furthermore, when compared to the dwarf forest, fringeproduction rates were approximately eight, six, six,and two times as high as dwarf forest rates forproproots, branches, stems, and leaves, respectively. Calculations of leaf production were based on mean redmangrove leaf longevities that ranged from about 189days to 281 days, depending on cohort and site.Repeated measures analysis of variance indicated thatleaf life spans did not differ significantly betweendwarf and fringe forests, but did differ among leafcohorts.Based on reported values for similar mangrove forests,the method provided reasonable estimates ofabove-ground biomass and production, while furnishingrelevant auxiliary information on spatial and temporalvariation in leaf demographic patterns. Furthermore,the partitioning of annual production between woodytissues and leaves followed the reported trend in mostforest ecosystems.     

The palaeoecology of the Ordovician Harding Sandstone (Colorado,U.S.A.)

The Middle Ordovician Harding Sandstone in Colorado, U.S.A. is the oldest formation known to have an environment rich in vertebrate fossils. In order to study the palaeoecology of these oldest vertebrates, a detailed analysis of both the sediments and the associated invertebrate fauna was made. Undulatory quartz, sandstone mineralogy, geochemistry, trace fossils, and ecological implications of the invertebrate fossils have been studied. The type of environment is not a typical one, but was probably a shallow-water marine (intertidal, or shallow subtidal), in a lagoon-like or estuarine—deltaic habitat, with changing salinity, and a generally warm climate.     

Response of stream‐breeding salamander larvae to sediment deposition in southern Appalachian (U.S.A.) headwater streams

1. Increased fine sediment deposition is a prevalent threat to stream biodiversity and has been shown to impact stream‐breeding salamanders negatively. However, their complex life histories make it difficult to determine which stage is affected. 2. We conducted field experiments from 26 August to 11 September 2010 and 11 October to 11 November 2010 in two southern Appalachian headwater streams (U.S.A.) to examine the response of larval salamanders to increased fine sediment deposition. Fine sediment was increased in artificial stream channels by 0, 33 and 67%. The number of larvae observed at the end of the experiments was used to determine whether larval microhabitat selection was influenced by fine sediment deposition. A concurrent survey of aquatic larvae in three nearby streams complemented this experiment. Stream substratum composition at survey sites was quantified to examine the effects of fine sediment on larval salamander abundance. 3. Increases in fine sediment deposition failed to explain the number of larval salamanders detected in stream channels. Similarly, a negligible effect of fine sediment was observed on abundance estimates. 4. These results suggest that fine sediment deposition has a minimal impact on aquatic salamander larvae. Therefore, the effects of increased fine sediment loads on stream‐breeding salamanders may not be the result of deleterious effects on the aquatic larvae but instead may be the result of effects on other stages. Management efforts that consider these other stages are therefore needed to protect stream‐breeding salamander communities.     

Factors influencing the longitudinal distribution of larval Hydropsychidae (Trichoptera) in a southern Appalachian stream system (U.S.A.)

The influence of physical habitat variables and suspended particulate organic matter (seston) on the distribution and production of eight species of larval Hydropsychidae was studied along a 6.4 km section of a southern Appalachian stream. Samples were collected at six stations encompassing stream orders 1–4 and an elevation range of 610 m. Multivariate analysis of covariance (using time as the covariable) and discriminant function analysis were used to examine habitat differences between the sampling stations due to the following variables: current velocity; coarse benthic detritus; substrate composition (by particle size); substrate heterogeneity; degree-days; and diel temperature fluctuation. The associations of these variables with the abundance (and production) of larval hydropsychids was also examined using the same statistical procedures. The six sampling stations represented three or four distinct habitats based on patterns of change in the variables along the stream continuum. Diel temperature fluctuation, median substrate particle size, the proportion of sand substrate, and substrate heterogeneity were most closely associated with the overall difference between the sampling stations. Hydropsychid species distribution along the stream system followed subfamily lines, i.e., Arctopsychinae and Diplectroninae were more abundant and productive in the upper 4.5 km of the stream, while Hydropsychinae were dominant in the lower 1.9 km. Diel temperature fluctuation was the habitat variable most highly correlated with patterns of hydropsychid abundance and production. The longitudinal pattern of species distribution, i.e., larger particle feeding Arctopsychinae being replaced downstream by smaller particle feeding Hydropsychinae, also coincided with the distribution of seston particle size classes along the stream. Mean seston particle size generally declined downstream, as particles <42 µm increased in relative abundance while those between 43 µm and 5 mm decreased. Temperature, seston, and substrate composition all undoubtedly exerted an important influence on the distribution and production of Hydropsychidae and other filter feeding insects in this stream. The validity and generality of these results depend upon the scope of the sampling effort. Conclusions drawn from data collected over the entire range (and limits) of a species' distribution are more sound than those based on data from a limited area.     

Geomorphology and fish assemblages in a Piedmont river basin, U.S.A.

1. We investigated linkages between fishes and fluvial geomorphology in 31 wadeable streams in the Etowah River basin in northern Georgia, U.S.A. Streams were stratified into three catchment sizes of approximately 15, 50 and 100 km², and fishes and geomorphology were sampled at the reach scale (i.e. 20–40 times stream width). 2. Non‐metric multidimensional scaling (NMDS) identified 85% of the among‐site variation in fish assemblage structure and identified strong patterns in species composition across sites. Assemblages shifted from domination by centrarchids, and other pool species that spawn in fine sediments and have generalised food preferences, to darter‐cyprinid‐redhorse sucker complexes that inhabit riffles and runs, feed primarily on invertebrates, and spawn on coarser stream beds. 3. Richness and density were correlated with basin area, a measure of stream size, but species composition was best predicted (i.e. |r| between 0.60–0.82) by reach‐level geomorphic variables (stream slope, bed texture, bed mobility and tractive force) that were unrelated to stream size. Stream slope was the dominant factor controlling stream habitat. Low slope streams had smaller bed particles, more fines in riffles, lower tractive force and greater bed mobility compared with high slope streams. 4. Our results contrast with the ‘River Continuum Concept’ which argues that stream assemblages vary predictably along stream size gradients. Our findings support the ‘Process Domains Concept’, which argues that local‐scale geomorphic processes determine the stream habitat and disturbance regimes that influence stream communities.     

Zoosporic marine fungi from the Pacific Northwest (U.S.A.)

Summary An investigation of the zoosporic fungi in the vicinity of the Friday Harbor Laboratory, San Juan Is., Washington, revealed the presence of great numbers of fungi. With one exception (Olpidium sp.) these were all biflagellate organisms. Predominating were species (11) of Thraustochytriaceae which abounded in water, in association with seaweeds, intertidal sands, and particularly on the surface of bottom samples down to depths of 298 m. A twelfth species of this group has several peculiarities and needs further investigation. Of the algal parasites, one on Polysiphonia and Pterosiphonia is considered new and termed Eurychasma joycei n. sp.The assistance of Drs. R. E. Norris and G. J. Hollenberg in calling my attention to certain algal parasites is gratefully acknowledged. I am also indebted to my colleague Dr. H. C. Whisler, and to Dr. C. Anastasiou for assistance in photographic work.     

Benthic microbial respiration in Appalachian Mountain, Piedmont, and Coastal Plains streams of the eastern U.S.A.

1. Benthic microbial respiration was measured in 214 streams in the Appalachian Mountain, Piedmont, and Coastal Plains regions of the eastern United States in summer 1997 and 1998. 2. Respiration was measured as both O₂ consumption in sealed microcosms and as dehydrogenase activity (DHA) of the sediments contained within the microcosms. 3. Benthic microbial respiration in streams of the eastern U.S., as O₂ consumption, was 0.37 ± 0.03 mg O₂ m^–2 day^–1. Respiration as DHA averaged 1.21 ± 0.08 mg O₂ m^–2 day^–1 4. No significant differences in O₂ consumption or DHA were found among geographical provinces or stream size classes, nor among catchment basins for O₂ consumption, but DHA was significantly higher in the other Atlantic (non‐Chesapeake Bay) catchment basins. 5. Canonical correlation analyses generated two environmental axes. The stronger canonical axis (W₁) represented a chemical disturbance gradient that was negatively correlated with signatures of anthropogenic impacts (ANC, Cl^–, pH, SO₄²), and positively correlated with riparian canopy cover and stream water dissolved organic carbon concentration (DOC). A weaker canonical axis (W₂) was postively correlated with pH, riparian zone agriculture, and stream depth, and negatively correlated with DOC and elevation of the stream. Oxygen consumption was significantly correlated with W₂ whereas DHA was significantly correlated with W₁. 6. The strengths of the correlations of DHA with environmental variables, particularly those that are proven indicators of catchment disturbances and with the canonical axis, suggest that DHA is a more responsive measure of benthic microbial activity than is O₂ consumption.     

Physical-chemical influences on vernal zooplankton community structure in small lakes and wetlands of Wisconsin,U.S.A.

We sampled zooplankton communities from 54 small water bodies distributed throughout Wisconsin to evaluate whether a snap-shot of zooplankton community structure during early spring could be used for the purpose of differentiating lakes from wetlands. We collected a single set of zooplankton and water chemistry data during a one-month time window (synchronized from south to north across the state) from an open water site in each basin as a means to minimize and standardize sampling effort and to minimize cascading effects arising from predator–prey interactions with resident and immigrant aquatic insect communities. We identified 53 taxa of zooplankton from 54 sites sampled across Wisconsin. There was an average of 6.83 taxa per site. The zooplankton species were distributed with a great deal of independence. We did not detect significant correlations between number of taxa and geographic region or waterbody size. There was a significant inverse correlation between number of taxa and the concentration of calcium ion, alkalinity and conductivity. One pair of taxa, Lynceus brachyurus and Chaoborus americanus, showed a significant difference in average duration of sites of their respective occurrence. All other pairs of taxa had no significant difference in average latitude, waterbody surface area, total phosphorus, total Kjeldahl nitrogen, alkalinity, conductivity, calcium ion, sulfate, nitrate, silicate or chloride. Taxa were distributed at random among the sites – there were no statistically significant pairs of taxa occurring together or avoiding each other. Multivariate analysis of zooplankton associations showed no evidence of distinct associations that could be used to distinguish lakes from wetlands. Zooplankton community structure appears to be a poor tool for distinguishing between lakes and wetlands, especially at the relatively large scale of Wisconsin (dimension of about 500 km). The data suggest that a small body of water in Wisconsin could be classified as a wetland if it persists in the spring and summer for only about 4 months, and if it is inhabited by Lynceus brachyurus, Eubranchipus bundyi, and if Chaoborus americanus and Chydorus brevilabris are absent.     

The relationship between pH and community structure of invertebrates in streams of the Shenandoah National Park, Virginia, U.S.A.

1. A replicated natural experiment was used to assess the influence of pH and low alkalinity on abundance and richness of invertebrate families in streams draining catchments that receive acid deposition. Individual streams were used as the unit of replication, allowing conclusions to refer to a class of streams rather than to particular streams. 2. We also controlled for several factors other than pH and alkalinity, including flow and temperature, that are recognized as determinants of invertebrate distribution. Samples were from indigenous rocks, standardized for size, surface area, and geology. 3. Repeated measures analyses of variance (ANOVA) revealed that invertebrate abundance and richness were significantly lower at pH 5.8 than at pH 7.1 for the total community. Thirty per cent of the forty-seven families exhibited significantly lower abundance at pH 5.8; thirteen families were absent at pH 5.8. Differences were greatest for Ephemeroptera: species richness was significantly lower at pH 5.8, and 71% of the twenty-four species were either absent (seven) or found in reduced abundance (ten) in the acidic streams. 4. Stream alkalinity is projected to continue to decrease with continued acid deposition in the Shenandoah National Park. Concurrent decreases in pH may lead to the absence or continued numerical decline of certain Ephemeroptera species in streams that acidify to less than pH 6.0 and/or 50μEql^?1 alkalinity.