sample number and colonization patterns of benthic macroinvertebrates and organic material on...

9
This article was downloaded by: [University of Oklahoma Libraries] On: 29 August 2013, At: 03:56 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Journal of Freshwater Ecology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tjfe20 Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata Richard J. Casey a & Sharon A. Kendall a a Alberta Research Council, Postal Bag 4000, Vegreville, Alberta, T9C 1T4, Canada Published online: 06 Jan 2011. To cite this article: Richard J. Casey & Sharon A. Kendall (1997) Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata, Journal of Freshwater Ecology, 12:4, 577-584, DOI: 10.1080/02705060.1997.9663572 To link to this article: http://dx.doi.org/10.1080/02705060.1997.9663572 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions

Upload: sharon-a

Post on 12-Dec-2016

218 views

Category:

Documents


3 download

TRANSCRIPT

Page 1: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

This article was downloaded by: [University of Oklahoma Libraries]On: 29 August 2013, At: 03:56Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Freshwater EcologyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/tjfe20

Sample Number and ColonizationPatterns of Benthic Macroinvertebratesand Organic Material on Artificial andNatural SubstrataRichard J. Casey a & Sharon A. Kendall aa Alberta Research Council, Postal Bag 4000, Vegreville, Alberta, T9C1T4, CanadaPublished online: 06 Jan 2011.

To cite this article: Richard J. Casey & Sharon A. Kendall (1997) Sample Number and ColonizationPatterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata,Journal of Freshwater Ecology, 12:4, 577-584, DOI: 10.1080/02705060.1997.9663572

To link to this article: http://dx.doi.org/10.1080/02705060.1997.9663572

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Page 2: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material

on Artificial and Natural Substrata Richard J. Casey and Sharon A. Kendall

Alberta Research Councila Postal Bag 4000, Vegreville

Alberta, T9C 7 T4 Canada

ABSTRACT

We determined differences in colonization by benthic macroinvertebrates among two types of artificial substratum (single rocks and bricks) and natural substratum after it was physically disturbed. We also determined the number of samples required for the same level of precision using the artificial substrata and direct sampling of the natural substratum. The study was conducted in a slow moving riffle-run in the Battle River in Alberta, Canada. All measures of the macroinvertebrate fauna (densities of each taxon, total number of organisms, and total number of taxa) and quantities of organic material were greater on the natural substratum than on artificial substrata. Physical disturbance of the natural substratum substantially reduced the macroinvertebrate community and quantity of organic material, and changed the composition and size of the natural substratum. But this disturbance to the substratum did not affect the colonization patterns of the macroinvertebrate fauna and organic material on the substratum types. Artificial substrata, similar those used in our study, are unlikely to give representative samples of macroinvertebrate fauna, especially for densities of zoobenthos taxa, or quantities of organic material on the natural substratum. The artificial substrata, in contrast to the natural substratum, required a larger number of samples for the same level of precision. Using the artificial substrata instead of direct sampling of the natural substratum therefore, does not appear to increase the precision of population estimates of benthic macroinvertebrates or quantities of organic material. Our results suggest that benthic macroinvertebrate communities and organic material in lotic systems might be sampled more efficiently using direct sampling of the natural substratum than the use of artificial substrata.

INTRODUCTION

Artificial substrata are used for sampling benthic macroinvertebrates in biomonitoring, impact assessment and toxicity studies in lotic systems. They are used to standardize the sampling method in the same type of habitat and to take samples in habitats where it is difficult or impossible to sample the natural substratum (Rosenberg and Resh 1982, Voshell et al. 1989, Buikema and Voshell 1993, Cairns and Pratt 1993, Resh and McElravy 1993). Research on the efficacy of artificial substrata relative to the natural substratum for sampling macroinvertebrates is necessary to determine appropriate uses of these samplers and the accuracy and precision of population measurements. Earlier studies on the efficacy of artificial substrata compared macroinvertebrates on artificial substrata and

" Formerly Alberta Environmental Centre

Journal of Freshwater Ecology, Volume 12, Number 4 - December 1997

Jour

nal o

f Fr

eshw

ater

Eco

logy

199

7.12

:577

-584

.

Page 3: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

undisturbed natural substratum (using Hess-type and Surber samplers) to determine if the fauna on the artificial substrata was representative of the natural substratum (Crossman and Cairns 1974, Rabeni and Gibbs 1978, Shaw and Minshall 1980, Canton and Chadwick 1983, Boothroyd and Dickie 1989, Clements 1991, Casey and Kendall 1996). But these comparisons were flawed because zoobenthos was already present on the natural substratum, in contrast to artificial substrata where organisms were new colonizers. A more appropriate comparison of the artificial and natural substratum types is to remove macroinvertebrates from the natural substratum before sampling. Completely removing the fauna from a natural substratum, however, is impossible without disturbing that substratum. Minshall and Minshall (1977) compared macroinvertebrates on artificial substrata to those on natural substratum after it was physically disturbed, but the artificial substrata were placed in the stream three months after the disturbance. Few studies concurrently compared the zoobenthos on more than one type of artificial substratum to the natural substratum (Crossman and Cairns 1974).

Using artificial substrata instead of direct sampling in quantitative studies of benthic macroinvertebrates is considered to decrease the variability of replicate samples or to increase the precision of population estimates, but results were conflicting (Rosenberg and Resh 1982, Morin 1985). Measures of variability, such as variance and coefficient of variation have been used to estimate the number of samples with a desired level of precision around the population mean (Elliott 1977, Allan 1984, Norris et al. 1992). The sampling effort and necessary resources for a study will be affected by the appropriate number of samples.

The objectives of this study were (1) to concurrently determine differences in the colonization of macroinvertebrates among different types of artificial substratum and the natural substratum after it was physically disturbed, and (2) to determine the number of samples required for the same level of precision using artificial substrata and direct sampling of the natural substratum.

METHODS AND MATERIALS

The study was conducted in a slow moving riffle-run in the Battle River (53' 01' N, 110" 50' W) in Alberta, Canada. The substratum was mostly cobble and pebble-gravel with an aquatic moss, Arnblvstegium riuarium (Hypnaceae), covering the upper surfaces of the substratum. Water velocities ranged from 0.23 to 0.63 rnls (measured at 0.4 x water depth with a Price AA meter) and depth was 22 to 34 cm. The riffle-run was marked into five 1-m transects, 3 m across the stream and 3 m apart. The substratum in each transect was physically disturbed using a hand-held hoe. Beginning at the most upstream transect, the operator stood downstream and uniformally moved the hoe in an upstream-downstream direction with as much force as possible for 10 minutes in each transect. Percent composition of the substratum was determined by visually estimating the proportion of four types of substratum (boulder >26 cm, cobble = 6-26 cm, pebble-gravel = 0.2-6 cm, sand ~ 0 . 2 cm) in five randomly selected 1 m2 quadrats in the transects and on the undisturbed natural substratum outside of the transects. After the substratum was disturbed, the macroinvertebrate fauna was sampled <2 h in five randomly selected areas on both the disturbed and undisturbed substratum. A modified Neill sampler (Neill 1938) was used to sample the zoobenthos (sample area = 0.091 m2, pore size = 0.210 rnm).

Jour

nal o

f Fr

eshw

ater

Eco

logy

199

7.12

:577

-584

.

Page 4: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

Organisms and epilithic material were removed by hand from the substratum to a depth of about 10 cm.

The artificial substrata were single clay bricks and natural rocks that were collected at the shore of the river. The clay bricks were rectangular (18.9 x 9.1 x 5.6 cm) with three circular holes (diameter = 3.5 cm) through the brick and a total surface area of 0.0785 m2. The natural rocks were oval-shaped cobble (maximum length = 5-15 cm) with a mean surface area of 0.0470 m2 (variance = 0.0077 m2, n = 15; based on the method in Casey and Kendall 1996). The bricks and rocks were cleaned in river water using brushes with rigid wire bristles and a knife to remove epilithic material. Fifteen bricks or rocks were randomly placed on the substratum 2 h after the substratum was disturbed in the transects. The substrata were installed from upstream to downstream sites and retrieved after 21 days, while holding a net (pore size = 0.210 mm) immediately downstream. Fifteen Neill samples were taken on the substratum in the transects, between the artificial substrata, immediately after they were removed. All samples were taken beginning at the most downstream sites to reduce disturbance. The samples were preserved in 85% ethanol.

In the laboratory, the samples were divided into coarse (>I. 18 mm) and fine ( 4 . 1 8 and >0.210 mm) fractions using sieves. Chironomidae larvae were subsampled in the fine fraction, based on the procedure of Wrona et al. (1982). The larvae were counted in a minimum of two 50-mL subsamples until >I00 organisms were obtained in the 1 L sample. The number of Chironomidae larvae per sample was estimated from the subsamples. The macroinvertebrates were identified and counted at 260 x magnification. Organic material (> 0.210 mm) in each sample was measured by separating it from the rest of the sample by elutrition and drying the material to a constant weight at 60 "C.

Numbers of macroinvertebrates per sample were converted to densities (organisms/m2) using the surface area of the substratum particles or the area sampled by the Neill cylinder sampler. Density data were log transformed (x+l) to obtain equal variances; plots of the mean and variance values demonstrated that the dependence of the mean on the variance was eliminated or reduced. The one-way analysis of variance and Student-Newman-Keuls multiple comparison tests were used to determine statistical differences among the densities of the most abundant taxa on the types of substratum. The Spearman rank correlation test was used to determine the similarity between the relative abundances of the 10 most abundant taxa on each of the artificial substrata and the natural substratum. Analyses were performed using the SAYSTAT statistical software package (SAS Institute Inc. 1988). The following formula from Elliott (1977) was used to calculate the number of samples (n) required to estimate the population density with a specific level of error or precision around the population mean: n = t2 . s2 ID2 . X2 where t = two-tailed critical value of the t distribution for P = 0.05 and n-1 degrees of freedom, s2 = sample variance, D = precision as a percentage of the mean (e.g., D = 0.2 is equivalent to the 95% confidence limits +20% of the mean), and X = arithmetic mean. This formula is applicable to benthic macroinvertebrates that typically have a negative binomial distribution (Elliott 1977, Noms et al. 1992). The number of samples required to detect differences between two population means was calculated using another formula requiring the power, or probability of making a type I1 error (Sokal and Rohlf 1981, p. 263). The power level of 30% was used

Jour

nal o

f Fr

eshw

ater

Eco

logy

199

7.12

:577

-584

.

Page 5: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

in the latter formula; this value was derived by Norris et al. (1992) using macroinvertebrate data collected in two streams. Sample numbers c 30 were solved by iteration using values of t for n-1 degrees of freedom (Elliott 1977).

RESULTS AND DISCUSSION

Effect of Physical Disturbance on Natural Substratum Densities of the total numbers of organisms were reduced by 56% on the natural substratum compared to the substratum that was physically disturbed. The densities of the total number of organisms and four of the six most common taxa (Chironomidae, Hyalella azteca, Stenonema and Leptophlebia) were statistically ( P c 0.05) reduced on the disturbed substratum (the remaining taxa were Optioservus and Acari). The six most common taxa made up 98% of the total number of organisms in the transects. Dry weights of organic material were statistically reduced by about 77% on the disturbed substratum (mean = 41 mg/m2) compared to undisturbed areas (mean = 18 1 mg/m2). Colonization by macroinvertebrates and organic material on the natural substratum that was physically disturbed, in contrast to undisturbed substratum, was therefore more directly comparable to new colonization on the artificial substrata. The physical disturbance reduced the composition and size of the natural substratum in the transects. Percent composition of the disturbed and undisturbed substratum, respectively, was 1% and 5% boulder, 69% and 74% cobble, 17% and 18% pebble-gravel, and 13% and 3% sand. Most of the substratum with the epilithic moss was displaced from the transects by the physical disturbance.

Colonization Patterns on ArtGcial and Natural Substrata Densities of the total numbers of organisms, the six most abundant taxa, and quantities of organic material were statistically greater on the natural substratum than on the two types of artificial substratum (Table 1). More taxa were found on the natural substratum (36) than on the single bricks (30) or rocks (25). These results were similar to those of another study we conducted at two sites in the McLeod River, Alberta (Casey and Kendall 1996). In the latter study, we compared macroinvertebrate colonization on a diverse range of artifical substrata (single rocks and bricks, and baskets of rocks) to the undisturbed natural substratum. The colonization patterns on the artificial and natural substrata were not affected by the physical disturbance to the natural substratum in the Battle River, even though the macroinvertebrate community was reduced by over half. The relative abundances of the ten most common taxa were not statistically correlated on the single rocks and natural substratum (r, = 0.43). but they were correlated on the single bricks and natural substratum (r, = 0.67) or the single rocks and bricks (r, = 0.84). Texture or roughness of the different substratum types might have affected our results; however, it is unlikely because both the artificial and natural substrata were relatively smooth to the touch. In previous studies there were no differences in the abundances of mayfly larvae on rocks with small changes in texture, i.e., limestone and sandstone (Casey and Clifford 1989), in contrast to clay tiles with greater textural differences of 1-2 mm (Clifford et al. 1992).

Sample Number using Artificial and Natural Substrata The smallest sample sizes were on the natural substratum compared to each of the

580

Jour

nal o

f Fr

eshw

ater

Eco

logy

199

7.12

:577

-584

.

Page 6: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

Table 1. Statistics and estimates of sample numbers for macroinvertebrates and organic material (n = 15) on single bricks (K) and rocks (R) and the natural substratum (N) in the Battle River. Substratum types with the lowest sample numbers for each group are shown in bold font.

Group Substratum No. Samples Mean Variance Sample Number Type With Zero r 40% Mean

Counts

Population Comparison Mean of Two Means

Chironomidae

Stenonema

Hyalella azteca

Hydropsyche

Cheumatopsyche

Acari

Total No. of Organisms

Organic Material (mg. dry weight)

artificial substratum types for the total number of organisms, up to five of the six most abundant taxa and organic material (Table 1). We used two different methods to calculate densities on the artificial and natural substrata; however, this did not affect the estimates of sample numbers on the substrata. For example, we re- calculated the sample numbers for the single rocks and bricks using half of the total surface area of each particle. Half of the surface of each brick or rock is approximately equal to the surface area used to calculate densities on the natural substratum (area enclosed by Neil1 cylinder). The sample numbers for the absolute sample area and half of this area were equal. Our findings showed a consistent pattern of the smallest numbers of samples for macroinvertebrates and organic material were almost exclusively on the natural substratum than on the artificial substrata. Using the artificial substrata instead of direct sampling of the natural substratum therefore, does not appear io increase the precision of population estimates of benthic macroinvertebrates or quantities of organic material.

In agreement with our results, Boothroyd and Dickie (1989) found less variability for the total number of organisms on undisturbed natural substratum than

Jour

nal o

f Fr

eshw

ater

Eco

logy

199

7.12

:577

-584

.

Page 7: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

on a perspex multiplate substratum. Reviews of other studies comparing macroinvertebrates on artificial and natural substrata, however, did not agree with our findings. In their extensive review of the advantages and disadvantages of artificial substrata, Rosenberg and Resh (1982) re-examined data for the total number of organisms from published lotic and lentic studies. They concluded that artificial substrata generally gave less sampling variability and, therefore, increased sampling precision compared to direct samples of the natural substratum. In another extensive review, Morin (1985) reanalysed data for functional feeding groups from lotic studies in cold temperate regions, including several of the same studies examined by Rosenberg and Resh (1982). Morin found that on average the variability of the feeding groups on artificial substrata was almost identical to the natural substratum.

The contrasting findings of the earlier studies compared to our results might be caused by important differences between the groups of studies. In the earlier studies only pooled data for the total number of organisms or feeding groups were used in the analyses (Rosenberg and Resh 1982, Morin 1985, Boothroyd and Diche 1989). We used lower taxonomic levels, often genus, which are more likely to show specific substratum preferences. In the earlier reviews, the precision data for the artificial or natural substratum types were often taken from independent studies (Rosenberg and Resh 1982, Modn 1985). The samples on the different substratum types in these studies were not concurrently collected at the same study site; therefore, in contrast to our study, the fauna were not always directly comparable. The previous studies used smaller numbers of samples than our 15 samples to calculate the precision of the sampling methods. For example, three to six samples were taken in these studies (Crossman and Cairns 1974, Rabeni and Gibbs 1978, Shaw and Minshall 1980, Canton and Chadwick 1983, Boothroyd and Diche 1989, Clements 1991). Increasing the number of samples generally will decrease the variance of a population mean and thus, reduce the estimated number of samples required for the same level of precision. Unlike the other studies, our samples often had zero counts, especially for the less common taxa, because we included lower taxonomic groups in the analysis of our data (Table 1). The zero values increased the variance and therefore, increased the number of samples required for the same level of precision. For most taxa the greatest number of zero counts were on the artificial substrata and thus, the lowest sample numbers were on the natural substratum. There were also differences in the type of sampler used to sample the natural substratum (Hess-type or Surber samplers) in the comparisons of previous studies. The Surber sampler, in contrast to the Neil1 cylinder, does not completely enclose the substratum and it is likely to be less efficient at sampling zoobenthos, especially in fast-flowing water. Finally, we attempted to compare only new colonizers on the artificial and natural substrata.

Our findings showed a large number of samples on artificial substrata were required for comparisons of two population means for most taxonomic groups and quantities of organic material. The precision level of +40% of the mean used to calculate the sample numbers was considered to be reasonable by other investigators sampling benthic macroinvertebrates in lotic studies (Elliott 1977, Allan 1984). Caution should be used when designing biomonitoring or studies requiring comparisons of population means. Sample numbers for the comparisons of two means on the natural substratum in the Battle River, for example, generally were much greater than the three to five samples often taken for biomonitoring in lotic systems using benthic macroinvertebrates (Resh and McElravy 1992). Sample

582

Jour

nal o

f Fr

eshw

ater

Eco

logy

199

7.12

:577

-584

.

Page 8: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

numbers for comparisons of means, however, can be reduced before beginning a study by making appropriate changes to the components of formulae used to calculate sample numbers (e.g., Norris et al. 1992). Methods of changing sample numbers include the selection of the difference in mean values to be detected and probability levels of type 1 and I1 errors, using transformed data (often required for fitting data to the assumptions of parametric tests), and choosing one-tailed instead of two-tailed statistical tests. Stratified sampling also might reduce the variability of macroinvertebrate samples. The final choice of the parameters used to estimate sample numbers will be based on the objectives of the study and an understanding of the variability of the macroinvertebrate populations being sampled. Our results suggest that benthic macroinvertebrate communities and organic material in lotic systems might be sampled more efficiently using direct sampling of the natural substratum than the use of artificial substrata.

ACKNOWLEDGEMENTS

We are grateful to Sandi Melenka for measuring the dry weights of organic material, and Hugh Clifford and Doug Craig for helpful comments on the manuscript.

LITERATURE CITED

Allan, J. D. 1984. Hypothesis testing in ecological studies of aquatic insects. Pages 484-507 in V.H. Resh and D.M. Rosenberg (editors). The ecology of aquatic insects. Praeger Scientific, New York, New York.

Boothroyd, I. K. G., and B. N. Dickie. 1989. Macroinvertebrate colonisation of perspex artificial substrates for use in biomonitoring studies. New Zealand Journal of Marine and Freshwater Research 23: 467-478.

Buikema Jr., A. L., and J. R. Voshell Jr. 1993. Toxicity studies using freshwater benthic macroinvertebrates. Pages 344-398 in D. M. Rosenberg and V. H. Resh (editors). Freshwater biomonitoring and benthic macroinvertebrates. Routledge, Chapman and Hall, Inc., New York, New York.

Cairns Jr., J . , and J. R. Pratt. 1993. A history of biological monitoring using benthic macroinvertebrates. Pages 10-27 in D.M. Rosenberg and V.H. Resh (editors). Freshwater biomonitoring and benthic macroinvertebratzs. Routledge, Chapman and Hall, Inc., New York, New York.

Canton, S. P., and J. W. Chadwick. 1983. Aquatic insect communities of natural and artificial substrates in a montane stream. Journal of Freshwater Ecology 2: 154-158.

Casey, R. J., and H. F. Clifford. 1989. Colonization of natural substrata of different roughness and colour by Ephemeroptera nymphs using retrieval and direct observation techniques. Hydrobiologia 173: 185- 192.

Casey, R. J., and S. A. Kendall. 1996. Comparisons among colonization of artificial substratum types and natural substratum by benthic macroinvertebrates. Hydrobiologia 341: 57-64.

Clements, W. H. 1991. Characterization of stream benthic communities using substrate-filled trays: colonization, variability and sampling selectivity. Journal of Freshwater Ecology 6: 209-221.

Jour

nal o

f Fr

eshw

ater

Eco

logy

199

7.12

:577

-584

.

Page 9: Sample Number and Colonization Patterns of Benthic Macroinvertebrates and Organic Material on Artificial and Natural Substrata

Clifford, H. F., R. J. Casey, and K. A. Saffran. 1992. Short-term colonization of rough and smooth tiles by benthic macroinvertebrates and algae (chlorophyll a) in two streams. Journal of the North American Benthological Society 11: 304-315.

Crossman, J. S., and J. Cairns Jr. 1974. A comparative study between two different artificial substrate samplers and regular sampling techniques. Hydrobiologia 44: 517-522.

Elliott, J. M. 1977. Some methods for the statistical analysis of samples of benthic invertebrates. Freshwater Biological Association Scientific Publication No. 25.

Minshall, G. W., and J. N. Minshall. 1977. Microdistribution of benthic invertebrates in a Rocky Mountain (U.S.A) stream. Hydrobiologia 55: 23 1-249.

Morin, A. 1985. Variability of density estimates and the optimization of sampling programs for stream benthos. Canadian Journal of Fisheries and Aquatic Sciences 42: 1530-1540.

Neill, R. M. 1937. The food and feeding of the brown trout (Salmo trutta L.) in relation to the organic environment. Transactions of the Royal Society of Edinburgh 59: 48 1-520.

Noms, R. H., E. P. McElravy, and V. H. Resh. 1992. The sampling problem. Pages 282-306 in P. Calow and G. E. Petts (editors). The rivers handbook. Volume one. Blackwell Scientific Publications, Oxford, United Kingdom.

Rabeni, C. F., and K. E. Gibbs. 1978. Comparison of two methods used by divers for sampling benthic invertebrates in deep rivers. Journal of the Fisheries Research Board of Canada 35: 332-336.

Resh, V. H., and E. P. McElravy. 1993. Contemporary quantitative approaches to biomonitoring using benthic macroinvertebrates. Pages 159-194 in D.M. Rosenberg and V.H. Resh (editors). Freshwater biomonitoring and benthic macroinvertebrates. Routledge, Chapman and Hall, Inc., New York, New York.

Rosenberg, D. M., and V. H. Resh. 1982. The use of artificial substrates in the study of freshwater benthic macroinvertebrates. Pages 175-235 in J. Cairns Jr. (editor). Artificial substrates. Ann Arbor Science Publishers Inc., Ann Arbor, Michigan.

SAS Institute Inc. 1988. SASISTAT user's guide, release 6.03 edition. SAS Institute Inc., Cary, North Carolina.

Shaw, D. W., and G. W. Minshall. 1980. Colonization of an introduced substrate by stream macroinvertebrates. Oikos 34: 259-27 1.

Sokal, R. R., and F. J. Rohlf. 1981. Biometry. 2nd. edition. W.H. Freeman and Company, San Francisco, California.

Voshell, Jr. J. R., R. J. Layton, and S. W. Hiner. 1989. Field techniques for determining the effects of toxic substances on benthic macroinvertebrates in rocky-bottomed streams. Pages 134-155 in U.M. Cowgill and L.R. Williams (editors). Aquatic toxicology and hazard assessment: 12th volume. American Society for Testing and Materials. Philadelphia, Pennsylvania.

Wrona, F. J., J. M. Culp, and R. W. Davies. 1982. Macroinvertebrate subsampling: a simplified apparatus and approach. Canadian Journal of Fisheries and Aquatic Sciences 39: 105 1-1054.

Received: 12 March 1997 Accepted: 13 July 1997

Jour

nal o

f Fr

eshw

ater

Eco

logy

199

7.12

:577

-584

.