Organism-substrate relationships in lowland streams (PDF)

Harry H. Tolkamp Organism-substrate relationships in lowland streams Proefschrift ter verkrijging van de graad van doctor in de landbouwwetenschappen, op gezag van de rector magnificus, dr. H.C. van der Plas, hoogleraar in de organische scheikunde, in het openbaar te verdedigen op vrijdag 6 februari 1981 des namiddags te vier uur in de aula van de Landbouwhogeschool te Wageningen. Centre for Agricultural Publishing and Documentation Wageningen - 1980  Abstract Tolkamp, H.H. (1980) Organism-substrate relationships in lowland streams. Agric. Res. Rep. (Versl. landbouwk. Onderz.) 907, ISBN 90 220 0759 6, (xi) + 211 p., 80 tables, 43 figs., 319 refs., Eng. and Dutch summaries, 14 appendices. Also: Doctoral thesis, Wageningen. A field and laboratory study on the microdistribution of bottom dwelling macro- invertebrates to investigate the role of the stream substrate in the development and preservation of the macroinvertebrate communities in natural, undisturbed low- land streams is described. Field data on bottom substrates and fauna were collected between 1975 and 1978 from two Dutch lowland streams. Substrates were characterized by the nature and the amount of organic detritus and the mineral particle sizes: in a field classification on the basis of the visually dominant particle sizes; in a grain-size classification on the basis of exact particle-size analysis in the labora- tory. Substrate preference for 84 macroinvertebrate species was demonstrated using the Index of Representation. Substrate-selection experiments were conducted in a laboratory stream for three Trichoptera species Qiiaropterna sequax, Chaetopteryx viVtosa and Serioostoma per- sonation) and one Ephemeroptera species (.Ephemera danica). An experiment on the colonization of artificial substrates in the field was also conducted. From the field data, several faunal groups could be distinguished, each group made up of species with similar substrate preferences. Detailed data on the micro- distribution in relation to substrate particle size are given for 26 species, which represent the various taxonomical units that compose the faunal groups: Trichoptera (5 species), Ephemeroptera (1 species), Plecoptera (1 species), Coleoptera (3 species), Amphipoda (1 species), Diptera (15 species, among which 13 species of Chironomidae). The microdistribution is discussed in relation to case-building behaviour, life cycle and food preferences. For several species substrate preferences may be different for different development stages or in different seasons. It is concluded that most species show distinct preferences for a specific sub- strate. The small scale spatial variation in substrate composition of the stream bed is essential for the existence of many lowland stream macroinvertebrate species. Free descriptors: substrate preference, microdistribution, benthos, macroinvertebrates, particle size, detritus, organic matter, habitat preference, environmental factors, lowland stream, laboratory stream Communication Nature Conservation Department 211. This thesis will also be published as Agricultural Research Reports 907. © Centre for Agricultural Publishing and Documentation, Wageningen, 1980. ü w ^ , 0 * t h l S b ° 0 k B a y b e r e P r o d u c e d °r Published in any form, by print photoprint, microfilm or any other means without written permission from the publishers  Woord vooraf Sinds ik in 1973 voor het eerst kennis maakte met de wetenschappelijke aspecten van de hydrobiologie heb ik geprobeerd me een deel van dit vakgebied eigen te maken onder de stimulerende leiding van Drs. J.J.P. Gardeniers. Jean, zonder jouw niet af- latende ideeënstroom, critische beschouwingen, enthousiasme en didactische talenten was dit proefschrift er niet gekomen. Niet alleen heb ik het vak van je geleerd, na mijn studie ben je zowel een gewaardeerd mentor als een goed vriend gebleven. Talloze- discussies liggen achter ons, waarvan vele tot diep in de nacht duurden terwijl onze wederhelften de avond alleen of samen doorbrachten. Mijn dank. Mijn promotor, Prof.Dr. M.F. Mörzer Bruyns ben ik dank verschuldigd voor de gelegenheid die hij mij bood om dit onderzoek bij de vakgroep Natuurbeheer te ver- richten. Mijn belangstelling voor het natuurbeheer is voor een belangrijk deel te danken aan Uw interessante colleges en verhalen. Tevens ben ik U zeer erkentelijk voor Uw waardevolle adviezen tijdens de bewerking van het manuscript. Prof. C.W. Stortenbeker wil ik bedanken voor zijn stimulerende commentaar op het manuscript, hetgeen zowel de stijl als.de opzet ten goede kwam. Bijzondere dank ben ik verschuldigd aan degenen die hebben bijgedragen aan de totstandkoming van dit proefschrift: - Ir. J.Chr. Both begon met de voorbereiding van dit onderzoek. Veel van jouw ideeën zijn in deze studie verwerkt. Tevens heb je het manuscript menig maal critisch door- genomen en de discussies die hierop volgden heb ik zeer gewaardeerd. - Ir. P.F.M. Verdonschot verrichtte twee deelonderzoeken van deze studie, welke zoveel interessante informatie opleverden dat we dit elders zullen publiceren. Dank- baar heb ik gebruik gemaakt van je noeste werklust en je ideeën om gegevens te ver- werken. - Ir. 0. Knol deed het onderzoek naar de afbraak van diverse bladsoorten. Ook dit werk zullen we elders publiceren, maar de vele discussies die we voerden hebben wezenlijk bijgedragen tot de huidige vorm van het manuscript. - Dr. H.K.M. Moller Pillot voor het determineren van vele muggelarven. - Dr. L.W.G. Higler voor het determineren van diverse kokerjuffers. - Drs. J.G.M. Cuppen voor het controleren van moeilijke kevers. - De heren I. Wolters en J.A.J. Beyer voor de verleende hulp bij het uitzoeken en analyseren van de monsters. - Ir. H. Bremer voor het schrijven van de meeste computerprogramma's, waarvan ik veel heb geleerd. - Drs. P. van Iersel voor zijn assistentie bij het clusteren.  - Ir. D. Davelaar voor het schrijven van enkele aanvullende programma's. - De heer I. Made Suwetja voor het schrijven van de plotterprogramma's. - Dr. L. van der Plas voor het beschikbaar stellen van analysefaciliteiten voor de korrelgrootteanalyse van de bodemmonsters op de vakgroep Bodemkunde en Geologie. - De heer A. Engelsma voor zijn assistentie bij deze analyses. - Mej. W. Hofstee voor haar hulp bij de korrelgrootteanalyses. - De heren J. van der Goor en H. Paardekoper en hun medewerkers van de werkplaats van het Biotechnion voor de bouw van de kunstbeek en kweekgoot. - Drs. J. Beunder voor het beschikbaar stellen van ruimte voor de laboratorium- experimenten in het Biotechnion. - Ir. J. Meuleman en Ir. J. den Duik voor hun hulp als er weer eens iets mis ging met een programma. - De heer M. Keuls voor zijn statistische adviezen. - Mevr. M. Gazenbeek-Dobbie voor de correctie van de engelse tekst. - Mej. A.M.N.G. Salden voor het typen van de literatuurlijsten. - Mej. E. Geurtsen van de afdeling Tekstverwerking voor het snelle en nauwkeurige typewerk van het definitieve manuscript. - De heren I.R.C. Cressie en R.J.P. Aalpol van Pudoc voor de correctie en redactionele vormgeving van het manuscript. - De medewerkers van de vakgroep Natuurbeheer voor de prettige werksfeer en de vele goede contacten. - Vele collega's in binnen- en buitenland voor het toezenden van overdrukken en de vele mondelinge en schriftelijke adviezen. Tenslotte maar niet in de laatste plaats, dank ik Ellen voor de talloze uren die eigenlijk voor ons samen waren, maar die ik aan dit proefschrift besteedde.  Curriculum vitae Harry Hendrik Tolkamp werd geboren op 3 september 1952 te Aalten te midden van de toen nog grotendeels natuurlijke beken. Hij volgde in Aalten het kleuter-, basis- en voortgezet onderwijs en bracht veel van zijn vrije tijd door in de bossen en de beken in de omgeving. Na het behalen van het einddiploma HBS-B, begon hij in 1969 met de studie in de Milieuhygiëne (N42) aan de Landbouwhogeschool te Wageningen, waar hij in 1972 het kandidaats- en in 1975 het doctoraalexamen aflegde. Voor het doctoraal examen werd het onderzoek vooral gericht op het waterkwaliteitsbeheer: Natuurbeheer- Hydrobiologie bij Prof.Dr. M.F. Mörzer Bruyns o.l.v. Drs.J.J.P. Gardeniers; Waterzui- vering bij Prof.P.G. Fohr o.l.v. Ir.J.G. den Blanken, T.H. Delft, Afd. Gezondheids- techniek; Microbiologie bij Prof.E.G. Mulder o.l.v. W.L. van Veen (f). Na het doctoraal examen was hij werkzaam als wetenschappelijk ambtenaar bij de Vakgroep Natuurbeheer van de Landbouwhogeschool tot december 1975, vervolgens als wetenschappelijk assistent bij dezelfde Vakgroep voor het verrichten van voorliggend onderzoek en van juni 1979 tot januari 1980 wederom als wetenschappelijk ambtenaar. Sindsdien is hij werkzaam als hydrobioloog bij het Waterschap Zuiveringschap Limburg te Roermond.  Contents 1 Introduction 1.1 General 1 1.2 Purpose of the investigation 4 1.3 Hypotheses 5 1.4 Research approach " •7 2 Description of the streams 2.1 The Snijdersveerbeek ' 2.2 The Ratumsebeek 11 16 Methods 1 g 3.1 Field procedure 3.1.1 Selection of sampling sites 3.1.2 Sampling dates 3.1.3 Field classification of substrates 3.1.4 Sampling method 3.1.5 Additional collections 3.2 Laboratory procedure 3.2.1 Sample sorting 3.2.2 Substrate analysis 3.2.3 Macrofauna analysis 3.3 Laboratory experiments 3.3.1 Artificial stream channel experiments 3.3.1.1 The stream 3.3.1.2 Substrates tested 3.3.1.3 Experimental design 3.3.1.4 Species tested 3.3.1.5 The influence of current velocity on substrate selection 3.3.2 Rearing channel 3.3.3 Experiments with Trichoptera 3.3.3.1 Case-building experiments 3.3.3.2 Grain size analysis of natural Trichoptera cases 3.4 Field experiments 3.4.1 Artificial substrate trays 3.4.2 Artificial substrates introduced without trays 16 16 17 18 20 20 20 20 23 25 25 25 27 27 29 29 30 30 30 31 32 32 33  3.4.3 Litter bags 34 •3.5 Data processing 34 3.5.1 Data storage 34 3.5.2 Statistical methods 35 Results 37 4.1 Data 37 40 40 4.1.1 Field classification of substrate types 37 4.1.2 Grain-size classification in the laboratory 4.1.2.1 Substrate composition 4.1.2.2 Substrate types 41 4.1.3 Discussion 4.1.3.1 The use of the Index of Representation compared to other statistical methods 4.1.3.2 The importance of the joint use of 11 and 10°s detritus classifications 4.1.3.3 The influence of animal abundance on the Index of Represen- tation 4.1.4 Grouping of substrate types based on species composition 47 4.1.5 Grouping of species based on substrate preferences 4.1.6 Numbers of species and specimens in different substrate types 4.1.7 Field experiments 4.1.7.1 Trays filled with artificial substrate 4.1.7.2 Particle sizes introduced without trays 4.2 Organism-substrate relationships 4.2.1 Outline of presentation 4.2.2 Autecological data for selected species 4.2.2.1 Lithax obscurus (Hagen) (Trichoptera: Goeridae) 4.2.2.2 Sericostoma personatum (Spence) (Trichoptera: Sericostomatidae) 4.2.2.3 Micropterna sequax (MacLachlan) (Trichoptera: Limnephilidae) 4.2.2.4 Chaetopteryx villosa (Fabricius) (Trichoptera: Limnephilidae) 4.2.2.5 Plectrocnemia conspersa (Curtis) (Trichoptera: Polycentropodidae) 4.2.2.6 Ephemera danica (Müller) (Ephemeroptera: Ephemeridae) 4.2.2.7 Nemoura cinerea (Retzius) (Plecoptera: Nemouridae) 4.2.2.8 Limnius volclonari (Panzer), Elmis aenea (Müller) and Oulimnius tuberculatus (Müller) (Coleoptera: Elminthidae) 116 4.2.2.9 Gammarus pulex (L.) (Malacostraca: Amphipoda, Gammaridae) 4.2.2.10 Dicranota Zetterstedt sp. (Diptera: Limoniidae) 4.2.2.11 Limnophila Marquart spp. (Diptera: Limoniidae) 4.2.2.12 Ptychoptera Meigen spp. (Diptera: Ptychopteridae) 46 49 56 58 58 59 60 60 62 62 68 75 91 100 103 114 119 128 132 134  4.2.2.13 Orthocladius van der Wulp spp. (Diptera: Chironoraidae, Orthocladiinae) 135 4.2.2.14 Micropsectra gr. praecox (sensu Tshernowskij) (Diptera: Chironomidae, Tanytarsini) 137 4.2.2.15 Paracladopelma Harnisch spp. (Diptera: Chironomidae, Chironominae) '40 4.2.2.16 Prodiamesa olivacea Meigen (Diptera: Oiironomidae, Orthocladiinae) 142 4.2.2.17 Epoicocladius flavens (Malloch) (Diptera: Chironomidae, Orthocladiinae) 144 4.2.2.18 Chironomidae o£ detritus substrates 146 Conchapelopia melanops (Wiedemann) (Diptera: Chironomidae, Tanypodinae) 14° Corynoneura Winnertz spp. (Diptera: Chironomidae, Orthocladiinae) 148 Brillia modesta (Meigen) (Diptera: Chironomidae, Orthocladiinae) ' 148 Diplocladius cultriger Kieffer (Diptera: Chironomidae, Orthocladiinae) 1 4 9 Rheocricotopus Thienemann spp. (Diptera: Chironomidae, Orthocladiinae) I 4 9 Eukiefferiella gr. discoloripes (sensu Moller Pillot, 1980) (Diptera: Chironomidae, Orthocladiinae) 1 5° Polypedilum laetum agg. (sensu Moller Pillot, 1979) (Diptera: Chironomidae, Chironominae) 1S1 5 General discussion 153 5.1 Particle size 154 5.2 Current velocity 156 5.3 Food conditions 158 6 Conclusions 165 6.1 Substrate composition 1 6 5 6.2 Substrate patterns 166 6.3 Effect of regulation 1 6 9 Surrmary 173 Samenvatting 176 Appendices 181 Identification references 196 References 200  Outline of substate classification Abbreviations used in the field classification of substrate types = Sand = Gravel CD = Coarse Detritus = Leaves = Detritus CD + L FD St Sh B CD/FD = = __ = Fine Detritus Stable Shifting Bare CD with or without FD Names, phi values and phi indices of grain-size fractions Fraction (mm) 128 64 32 16 8 4 2 1 0. 0. 0. 0. 0 t : _ _ _ _ _ _ _ _ 500 - 250 - 125 - 050 - - 0.050 256 128 64 32 16 8 4 2 1 0. 0. 0. 0. ,500 250 125. 050 Phi value (-log fraction) -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5-10 mm is used instead of 0. .0625 Phi index 8 7 6 5 4 3 2 1 1 2 3 4 0 mm Name of fraction cobbles cobbles large pebbles small pebbles coarse gravel medium gravel fine gravel very coarse sand coarse sand medium sand fine sand very fine sand silt and lutum Q.M.Q, index describes the grain-size composition of a substrate by giving the d 3 first (Q ), second (median) <Md> and third (Qg) quartiles expressed in integer phi values (phi indices).  1 Introduction 1.1 GENERAL Lowland streams, with a strongly fluctuating discharge, water level, current velocity, bottom composition and vegetation pattern, are a characteristic type of stream for the Netherlands. A large number of benthic macroinvertebrate species are restricted to this type of environment and form the characteristic animal community of the natural lowland stream. Since the 1930s most lowland streams in the Netherlands have been regulated in connection with land reclamation schemes or programmes aimed at the improvememt of stream hydrology to increase drainage and in the more recent decades also to lower the groundwater level in vast areas. This implies the reduction of inundation, peak flow, erosion and sediment transport. To this end meanders are cut off, bends are straightened, the stream bed is deepened and widened, the slope of the banks is stan- dardized to a 1:2 or 1:3 profile, bank vegetation is removed and paths are con- structed for grass mowers; weirs and barrages are placed at regular intervals to reduce sediment transport and retain a minimum water level in summer for irrigation purposes. Banks are often reinforced with concrete, nylon matting or wood and some- times the whole stream bed is reinforced with concrete blocks. Several terms are in use to describe these kind of physical perturbations. Hereafter the term regulation will be used, to save the term channelization for streams and rivers.that have been regulated with the purpose of facilitating transport by boat. Regulation strongly affects the physical characteristics of the stream system, and investigations in the Achterhoek (Gardeniers & Tolkamp, 1976) and elsewhere in the Netherlands (Heijdeman & van 't Oever, 1976; Peters & Leijten, 1977; de Graaff, 1979) demonstrate that the animal community living in regulated streams is quite different from that living in unperturbed, natural lowland streams, a feature fre- quently reported in other countries for other stream types as well (e.g. Wene & Wickliff, 1940; Engelhardt, 1951; Stuart, 1959; Einsele, 1960; Rouyer, 1975). The typical fauna present in natural lowland streams is often practically absent from streams that have been regulated. Only in streams regulated in the old fashioned way, where maintenance has been neglected since the major regulation works, may the fauna of the natural stream be able to stay or re-establish itself in places where conditions are returning to the original state. In such semi-regulated streams the fauna may consist of a mixture of pond- and stream-dwelling species. In most regu- lated streams, however, the species composition resembles that of canals,.ditches and 1  ponds. This can be attributed to the changes the physical environment has undergone. Levelling down the spatial and temporal variation in current velocity will result in more uniform substrate patterns, while the absence of bank vegetation (trees and shrubs) which would shade the water, results in higher water temperatures and the development of macrophytes or even microphytes when current velocity is low enough. Less allochthonous material will enter the stream system and algae and aquatic plants will replace leaves as the trophic basis of the community. The stream system changes from heterotrophic to autotrophic (Cummins et al., 1973). Consequently, major changes in community structure will result. For each animal species it is essential that environmental factors are present in the right extent within reach of the animal. Comparing natural and regulated lowland streams, and noting the differences in species composition of the communities present, the impression becomes stronger that the environmental demands of species characteristic for unperturbed lowland streams are so specific that they can only be met in very distinct places within the stream. Apparently these species can only find the optimal combination of physical parameters in a stream with much variation, with the right extent or range of each factor important for their existence (e.g. sub- strate particle size, current velocity, food, oxygen supply). Many investigations have been carried out in Dutch lowland streams during the last two decades, mainly concerning the description of the stream fauna and the effects of pollution (e.g. Mur Atzema, 1962; Gardeniers, 1966; Moller Pillot, 1971; Higler, 1972; Tolkamp, 1975a; Gardeniers & Tolkamp, 1976). These investigations on the physical and chemical quality of the streams as related to the macroinvertebrates have led to a reasonably complete picture of the geographical distribution of the benthic macroinvertebrates in Dutch lowland streams and of the differences between natural and regulated streams in the sense of the composition of the animal community. The differences in species composition have even been used as an instrument to mea- sure the degree of regulation of a lowland stream (Tolkamp & Gardeniers, 1977). However, current velocity is only one of the parameters that may change after regulation and no detailed information is available on the specific reasons for the differences in animal communities between natural and regulated streams. Current velocity is of course one of the reasons, but the distribution of freshwater animals is also determined by a whole complex of a large number of physical, chemical and biological factors (Wesenberg-Lund, 1943; Engelhardt, 1951; Albrecht, 1953; Macan, 1961, 1962; Cummins, 1966; Thorup, 1966; Cummins & Lauff, 1969; Hynes, 1970a, b; Meadows & Campbell, 1972; Macan, 1974; Decamps et al, 1975; Friberg et al, 1977). Schmitz (1954) followed Thienemann (1912) in the conception that current velocity and temperature are the primary factors influencing the development of running-water zoocoenoses. Whitehead (1935) concluded that the nature of the stream bed is of great importance as a factor determining the nature of the fauna, either directly (pro- viding shelter, suitable attachment sites) or indirectly (providing food, influencing the nature of plant growth, giving suitable oxygen conditions).  The stream bottom is the product of two sources that provide the basic com- ponents: mineral and organic allochthonous material supplied by water and air, and mineral and organic autochthonous material present in the stream bed, originating from geomorphological conditions. Under the influence of the discharge regime and the form of the stream channel, interacting in two major counter processes, erosion and sedimentation, the basic components are arranged in substrate patterns with characteristic properties concerning composition (particle size and nature of mineral and organic matter), structure (spatial variations-vertical and horizontal, stability, packing, porosity) and dynamics (temporal changes in structure). Because of the sorting activity of the current, the structure of the stream bottom is inseparably interwoven with the variation of the current (velocity, regime, turbulence) (Schmitz, 1961; Scherer, 1965), resulting in a constantly varying (in space and time) mosaic pattern of substrate types, each with different environmental conditions. Grain-size composition may vary at very short distances and influence, for example, the oxygen content of the interstitial water, the amount of trapped organic detritus, the growth of periphyton and the number of crevices. Cummins & Lauff (1969) presented a diagram showing the four major categories of environmental parameters (Fig. 1). They emphasized that current, temperature and con- centration of chemical factors may limit the range of habitat tolerances (macro- distribution) and that substrate particle size or food supply are probably the main microdistributional factors. Although several additional physical factors influencing distribution are often mentioned, these are mostly directly related to one of the main factors: flow regime, drought, spates, illumination, suspended solids, proximity of suitable habitats (Sprules, 1947; Hynes, 1970a). Biotic factors as competition and prédation (Macan, 1962), oviposition and drift habits (Macan, 1961; Hynes, 1970a) are also related to the main factors food, current and substrate. Ulfstrand (1967) stressed the point that the most important factor linked with substrate is the provision of food in the form of allochthonous and autochthonous matter and prey animals. This agrees with the view of Cummins (1975), who stated that food will undoubtedly be the ultimate determinant of macroinvertebrate distribution and abundance in non-perturbed running waters. But he added that when food conditions are favourable, other factors, as sediment particle size, current, competition for space or prédation will determine the microdistribution within a given section of a stream. MDreover, food is part of the substrate for algal feeders (grazers, scrapers) as well as detritus feeders (shredders, consumers) and the presence and abundance of food substances is influenced by, for example, the particle size (accumulation of detritus (Rabeni & Minshall, 1977)), sediment transport (scouring off attached algae), light conditions and current (distribution of detritus, growth of hydrophytes, bringing food to filter feeders). The preference of prey organisms for certain sub- strate or food types will influence the distribution of their predators (Ulfstand et al., 1971; Hildrew & Townsend, 1976). For the very reason that the substrate is the resultant of and strongly inter-  SUBSTRATE PARTICLE SIZE w o ss 3 g a MICRODISTRIBUTION ~> FOOD SUBSTANCES <- (Detritus,prey, microphytes,etc.) A V V CURRENT VELOCITY <- V V OTHER PHYSICAL - CHEMICAL PARAMETERS (e.g pH, CaC03, Fe203) O ,Temp., TURBULENCE and CURRENT -> : Interactions between environmental parameters to which a given benthic species interacts __> : Influence of the environment on a benthic species which results in recognition of the habitat .> : Influence of a given benthic species on the habitat Fig. 1. General relationship between environmental parameters and the micro- distribution of benthic stream macroinvertebrates (after Cummins & Lauff,1969) linked with a number of physical factors and is of great ecological importance for macroinvertebrates, it is an outstanding parameter for the study of macroinvertebrate microdistribition. Because the substrate composition, structure and dynamics are the first to change under the influence of alterations in current velocity and discharge regime following stream regulation, they may prove to be the major reasons for the differences in species composition between natural and1 regulated lowland streams. 1.2 PURPOSE OF THE INVESTIGATION The purpose of the investigation was to gain insight into the role the substrate plays in the development and preservation of the macroinvertebrate communities in natural, undisturbed lowland streams. Testing and quantification of the relationships between temporal and spatial environmental factors as measured by substrate charac- teristics and the composition of animal communities will increase the knowledge of  the autecology of some typical lowland stream species and will make it possible to indicate some of the critical conditions that distinguish natural lowland streams from other watercourses. As a basis for preservation and management of these streams this knowledge is indispensable and it will contribute to the quantification of the limiting conditions concerning stream dimensions, profile, meandering, bank vegetation, weir level, current speed, maintenance, effluent discharge, recreation and bank protection. There is an urgent need for this kind of information since the demands made from the hydrological point of view are often contradictory to the wish to preserve the Dutch lowland stream as a unique type of environment. 1.3 HYPOTHESES It has been shown by numerous authors that distinct differences occur in the species composition found in different substrate types (e.g. Pennak & van Gerpen, 1947; Thorup, 1966; Mackay, 1969; Mackay & Kalff, 1969; Ward, 1975; Cummins, 1975). Many authors used the substrate merely to describe the various habitats they encoun- tered and were not trying to establish differences in the faunal composition of the substrate types, but described the stream fauna itself and where to find it (e.g. Thienemann, 1912; Behning, 1928; Beyer, 1932; Geijskes, 1935; Sprules, 1947). More recently, studies directly relating species distribution and substrate composition have become numerous and many references were given by Cummins (1966), Thorup (1966) and Hynes (1970a, b ) . Although a species may show a distinct preference for a certain substrate type, other features prevailing in the same place or in the neighbourhood may cause the animals to tolerate less-preferred substrates. Cummins & Lauff (1969) demonstrated this for the stonefly Pevlesta plaaida, in substrate-selection experiments. Various authors have shown for several species that different instars may prefer or demand different substrate grain sizes or compositions (e.g. Scott, 1958; Hanna, 1961; Cummins, 1964; Schwoerbel, 1967; Mackay, 1969; Elliot, 1971; Harman, 1972; Otto, 1976). Rees (1972) found a significant relationship between the body length of Gammarua psevdolirnnaeus (Amphipoda) and the substrate particle size in laboratory experiments, and Wesenberg-Lund (1943) already gave examples of several caddis fly species that use different house-building materials during their development. Mackay (1977) demonstrated that Pyenopsyche scabripennis (Trichoptera) pupae burrow in other substrates than the larvae prefer. Thus many benthic-stream-dwelling macro- invertebrates may need several substrate types to complete their life cycle. These observations formed the basis for the following hypothesis: The demands of many benthic lowland stream macroinvertebrate species of the substrate are of such a nature that they can only be met in a non-regulated lowland stream with its typical substrate composition and pattern. Two working hypotheses were formulated to test this hypothesis:  - Many benthic lowland stream macroinvertebrate species show distinct preferences for a specific substrate composition. - For many of these species the small-scale spatial variation in substrate compo- sition of the stream bed is essential for their existence. 1.4 RESEARCH APPROACH Testing of the working hypotheses was performed in three steps: - Determination of the microdistributional patterns of the macroinvertebrates in two natural, undisturbed lowland streams (field investigation). - Determination of substrate preferences of a number of characteristic lowland streams species in substrate-selection experiments in a laboratory stream channel (laboratory experiments). - Determination of the macroinvertebrate colonization of artificial substrates in the stream bed (field experiments). This approach was chosen because the interpretation of data from field research only leads to correlations between animal distribution and certain environmental parameters. Experiments in the laboratory or the field, or both, are necessary to provide data that can be used to test the conclusions derived from the field research.