Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Soil fauna communities in cedar-hemlock and hemlock-amabalis fir forest types on northern Vancouver Island,… Battigelli, Jeffrey Paul 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_spring_battigelli_jeffrey_paul.pdf [ 4.38MB ]
JSON: 831-1.0086630.json
JSON-LD: 831-1.0086630-ld.json
RDF/XML (Pretty): 831-1.0086630-rdf.xml
RDF/JSON: 831-1.0086630-rdf.json
Turtle: 831-1.0086630-turtle.txt
N-Triples: 831-1.0086630-rdf-ntriples.txt
Original Record: 831-1.0086630-source.json
Full Text

Full Text

SOIL FAUNA COMMUNITIES IN CEDAR-HEMLOCK ANDHEMLOCK-AMABALIS FIR FOREST TYPES ONNORTHERN VANCOUVER ISLAND, BRITISH COLUMBIAbyJEFFREY PAUL BATTIGELLIB.Sc., The University of Victoria, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Soil Science)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAJanuary 1992© Jeffrey Paul Battigelli, 1992In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(SignatureDepartment of Soil ScienceThe University of British ColumbiaVancouver, CanadaDate DE-6 (2/88)ABSTRACTSoil fauna play an important role in regulating decomposition and nutrientcycling. They are able to process a portion of the annual litter input thus decreasingforest litter, increasing microbial activity and increasing mineralization of nutrients.The faunal composition of most Canadian soils has not been investigated. Thepurpose of this study is to determine and compare the composition of major soil faunagroups in Cedar-Hemlock and Hemlock-Amabalis fir forest types and to examineseasonal changes in vertical distribution within and between the forest types.Four methods were used to extract soil fauna: hand sorting, Lussenhop highgradient extractor, modified Baermann funnel and the Formalin method. Samplingoccurred five times between August 1989 and March 1991.Forty one groups were identified in both forest types. Results indicate that theHemlock-Amabalis fir forest type maintains a higher abundance and biomass of soilfauna than the Cedar-Hemlock forest type. Nematoda are numerically dominant inboth forest types followed by Acari (in toto), Collembola (in toto) and Copepoda.Diplopoda, Enchytraeidae, Diptera larvae and Acari (in toto) are the dominant groupsin both forest types when considering biomass.Similarity indices indicate that the forest types have similar group diversity.Biomass is slightly different between both forest types in October 1990. Analysis ofvariance performed on mean depth values for twenty groups indicate that mean depthwas not significantly different between the two forest types (p > 0.05). Variation inmean depth was significantly different between sampling times for Prostigmata,Isotomidae, Onychiuridae and Nematoda (p > 0.05). Soil moisture content did notsignificantly affect mean depths.Distribution through the organic horizons showed a decrease in percentabundance of microarthropods in the upper organic layer in the summer with higherabundances in the spring and fall. Enchytraeidae showed a similar pattern in bothforest types. Soil moisture content in the mat complex also exhibited this pattern withhigh moisture contents relating to higher abundances. Nematoda maintained a constantabundance in the upper organic layer throughout the study. This suggests that soilmoisture may not be solely responsible for the observed shifts in vertical distribution ofthe groups examined.Overall, the soil fauna communities are similar between the Cedar-Hemlock andHemlock-Amabalis fir forest types. Further work is required to increase the level oftaxonomy to genus or species, which may identify taxa unique to either forest type. Atthe very least, this could provide a higher resolution of the distributional trends thatmight also be observed in species within the organic horizons and possibly relateinformation on their biological activity within these horizons.iiiTABLE OF CONTENTSPageAbstract^ iiTable of contents^ ivList of tables viList of figures^ viiAcknowledgements viiiDedication^ ix1.0 INTRODUCTION 11.1 Site Description^ 52.0 METHODS^ 112.1 Sampling 112.2 Laboratory Analyses^ 122.3 Data Analyses 133.0 RESULTS^ 163.1 Abundance and Biomass^ 163.1.1 Nematoda 273.1.2 Acari^ 293.1.3 Collembola 293.1.4 Crustacea 303.1.5 Oligochaeta^ 303.1.6 Other Groups 313.1.7 Supplementary Abundance Data^ 313.2 Vertical Distribution^ 333.2.1 Mean Depth 333.2.2 Distribution through Horizons^ 353.3 Similarity Indices^ 42iv4.0 DISCUSSION^ 464.1 Abundance and Biomass^ 464.1.1 Comparison of Abundance Values^ 494.1.2 Comparison of Biomass Values 514.2 Similarity Indices^ 534.3 Vertical Distribution 544.3.1 Mean Depth^ 564.3.2 Distribution through Horizons^ 584.4 Limitations of the Research 624.4.1 Sampling^ 624.4.2 Extraction 644.4.3 Analysis 665.0 CONCLUSIONS^ 68LITERATURE CITED^ 71Appendix A: Equations for Similarity Indices^ 78Appendix B: Gravimetric soil moisture contents (%)for the sampled horizons^ 79Appendix C: Mean depth (cm) values + 1 standarddeviation of soil fauna for all sampling periods^80LIST OF TABLESPageTable I.^Median individual dry mass (mg) used tocalculate biomass values^ 14Table II.^Mean abundance (x 103 m-2), percent abundance(%) and biomass (mg m -2) of soil fauna for August 1989^17Table III.^Mean abundance (x 103 m-2), percent abundance(%) and biomass (mg m -2) of soil fauna for May 1990^19Table IV.^Mean abundance (x 103 m-2), percent abundance(%) and biomass (mg m -2) of soil fauna for July 1990^21Table V.^Mean abundance (x 103 m-2), percent abundance(%) and biomass (mg m -2) of soil fauna for October 1990^23Table VI.^Mean abundance (x 103 m-2), percent abundance(%) and biomass (mg m -2) of soil fauna for March 1991^25Table VII^Summary of mean relative abundance and biomass valuesfor the dominant groups from all sampling times^28Table VIII. Comparison of abundances (m-2) for groupsrecovered by different extraction methods forAugust 1989 (CH only)^ 32Table IX.^Results of ANOVA of mean depth of faunal groupsthrough time and forest type^ 34Table X.^Similarity indices calculated for both foresttypes for each sampling time^ 44viLIST OF FIGURESPageFigure 1.^Location of study site^ 6Figure 2.^Ecological climate diagram based on 30 yearmeans in the style of Walter (1979)^8Figure 3.^Seasonal distribution of Acari (in toto)abundance (%) through sampled horizons^36Figure 4.^Seasonal distribution of Cryptostigmataabundance (%) through sampled horizons^37Figure 5.^Seasonal distribution of Collembola (in toto)abundance (%) through sampled horizons^38Figure 6.^Seasonal distribution of Onychiuridaeabundance (%) through sampled horizons^39Figure 7.^Seasonal distribution of Enchytraeidaeabundance (%) through sampled horizons^40Figure 8.^Seasonal distribution of Nematode abundance(%) through sampled horizons^ 41Figure 9.^Gravimetric soil moisture content (%) of the matcomplex for both forest types^ 43viiACKNOWLEDGEMENTSThanks to:Shannon Berch, my supervisor, for her energy, enthusiasm and unique outlookon the data.Dr. V. G. Marshall, Pacific Forestry Centre, for starting me off on all of this andhelping in the field.Steve Joyce and Bill Dumont of Western Forest Products in Port McNeill forsupplying maps, forest tours and excellent accommodation.Paul and Buffy for their help in the field.Marilyn Clayton, Pacific Forestry Centre, for her help with the Baermann funnelextractions.Sharmin for her help with counting the critters and sampling in the snow.Rick for teaching me how to DRAW.My family for their support and encouragement throughout my academic career,thus far.Sandy for putting up with the writing process.viiiTo Guerino, Malcolm and Emilyix1.0 INTRODUCTIONThe forests in the Port McNeill area on northern Vancouver Island belong to theVery Wet Maritime Coastal Western Hemlock Subzone (CWHvm) (Klinka e„ 1991)[equivalent to CWHbi (Green et al. 1984) and CWHe (Klinka et al. 1979)]. Thisvariant is dominated by the "salal-moss" (Si) ecosystem association which can bedivided into two distinct forest types: cedar-hemlock (CH) and hemlock-amabalis fir(HA) (Lewis 1982). Both forest types occur on similar soils, but the physicalcharacteristics and amount of organic matter found in each type are different. Thesedifferences will be expanded upon later in the site description.During the late 1970's, it was discovered that large plantations of Sitka spruce[Picea sitchensis (Bong.) Carr.] on CH cutblocks were stagnating. After initial goodsurvival and height growth, the height increment decreased to about 10 cm yr-1 and thetrees became chlorotic within five years after planting. Lewis (1982) indicated that theCH maintained a lower total soil nitrogen content than the HA. Germain (1985)showed that the CH sites were very deficient in nitrogen and phosphorous and thatplantations responded well to fertilization. This suggested a nutrient cycling problemand/or lower decomposition rates in the CH. The HA, on the other hand, representssome of the most productive forest sites in Canada with unmanaged, 70 year old standsyielding timber volumes of 12 to 15 m3 ha-1 yr4 (Germain 1985). Furthermore,young plantations on these sites do well.Numerous reasons for the variation in productivity and differences in standdevelopment between the two forest types have been put forward. Preliminary researchby Western Forest Products (WFP), the University of British Columbia (UBC) and thePacific Forestry Centre (PFC) led to the establishment of the Salal-Cedar-HemlockIntegrated Research Program (SCHIRP) and the development of appropriate researchprojects, including one on soil fauna.1Soil fauna are responsible for less than 5 % of the total decomposer respirationand thus provide a negligible direct contribution to community metabolism (Petersenand Luxton 1982). However, through indirect effects, soil fauna have important,catalytic roles in forest litter decomposition, nutrient cycling, maintenance of soilstructure, and may also be excellent indicators of soil quality (Price 1973, Seastedt andCrossley 1981, Petersen and Luxton 1982, Paoletti et^1991). Persson (1980)suggested that soil invertebrates' impact on decomposition of litter can range from 1 to2 % to up to 30 %. Furthermore, soil invertebrates can be responsible for as much as68 % of the comminution of litter within some ecosystems (Seastedt 1984a, Anderson1988). These organisms are able to process 20-30 % of the annual litter input byconverting microbial biomass and detritus into smaller fragments and feces. Thisconversion results in additional surface area and modified substrates for furthermicrobial colonization and use (Kitazawa 1967 as cited by Petersen and Luxton 1982).Petersen and Luxton (1982) stated that this may be a conservative estimate for somesystems where the value may be as high as 40 %. Litter decreases, microbial activityincreases and more nutrients are mineralized and picked up by roots or mycorrhizalfungi as a result of soil fauna activity (Seastedt and Crossley 1981). The general rule isthat soil faunal biomass is positively correlated with the decomposition rate (Schaeferand Schauermann 1990).In forest ecosystems, soil fauna, in association with microbes, are alsoresponsible for regulating wood mineralization. As wood decays, it reaches a pointwhere fungi are dominant, resulting in a net immobilization of nutrients. Soil faunainvade the wood and initiate a net loss of nutrients through leaching and translocationas well as relocating microbial propagules within the woody material and regulatingmicrobial succession (Petersen and Luxton 1982). While soil fauna have some effecton wood decomposition, the presence of phenolic acids and the water holding capacitycan also affect the decomposition rate (Schaefer and Schauermann 1990). In the CH,2western red cedar (Thuja plicata Don.), which has a high phenolic acid content, is thedominant wood product being decomposed. Furthermore, the soil in the CH appears tobe wetter than that in the HA.Petersen and Luxton (1982) suggested that soil fauna may be able to selectivelyconcentrate limiting nutrients into their biomass. Isopods and Diplopods are able toaccumulate Ca and Mg (Krivolutsky and Pokarzhevsky 1977 as cited by Petersen andLuxton 1982) while other fauna have shown the ability to accumulate K and Na.Petersen and Luxton (1982) further suggest that this accumulation activity may beassociated with seasonal changes in nutrient composition of the food base, but othershave suggested that it is dependent upon the biochemical and physiologicalcharacteristics of the fauna involved (Carter and Cragg 1976 as cited by Petersen andLuxton 1982). Soil invertebrates utilize a small proportion of Ca, K, S and Mg in litterfall, but up to 70 % of N released during decomposition can be immobilized intoinvertebrate tissues (McBrayer 1977 as cited by Petersen and Luxton 1982). At thebeginning of the growing season much of the immobilized N and P are released as ions.Furthermore, Coleman et al. (1978) suggested that a build up in bacterial-grazingnematode populations could result in net immobilization of nutrients unless continualpredation pressures exists. This suggests that soil fauna could be important for theinternal retention and cycling of N and P as well as other nutrients within the forestecosystem (Petersen and Luxton 1982). This is an important factor since the CH wasconsidered by Germain (1985) to be deficient in N and P. The absence of certain soilfauna species or, perhaps, a depauperate soil fauna community in the CH would beunable to regulate nutrient flow. This could lead to the decreased productivityobserved in the CH. Soil fauna in the HA may not be adversely affected by loggingpractices. Perhaps the HA is more favourable to recolonization than the CH afterlogging which would enable it to establish better decomposition and nutrient cyclingsooner than the CH.3Generally, soil fauna populations are concentrated within five to ten centimetersof the soil surface depending upon soil and forest types (Price 1975, Petersen andLuxton 1982). In temperate coniferous forests, soil fauna are found concentrated in theuppermost soil layers (Price 1973, Price 1975). A variety of biotic and abiotic factorssuch as geographic location, climate, physical and chemical soil properties, type ofvegetative cover, nature and depth of the litter layer, seasonal changes in soil moistureand temperature, food supplies, interaction with other soil microflora and fauna andspecies' life cycles affect species abundance and the distribution of individuals withinthe soil community and soil profile. Berthet and Gerard (1965) stated that thedistribution of individuals is influenced by environmental factors. Abiotic factorswithin the habitat such as soil moisture, pH and temperature vary from place to placeand have varying affects on distribution while biotic factors can account for bothgregariousness of individuals as well as for dispersion of individuals as a result ofcompetition for resources. The distribution of species results from geneticcharacteristics unique to the species. As an environment changes along a gradient, thepopulations of most species also change. In most environments, species depend onlypartly on other species (e.g. one species of insect may feed upon one of several plantspecies while a predator may feed upon a variety of prey species). This suggests thatmany combinations of species will form associations that will vary continuously in timeand space (Krebs 1978).Soil fauna are integral to the development of humus forms (Klinka e 0. 1981)and the release of important plant nutrients through mineralization of organic matter.Humus forms typically exhibit different rates of nutrient turnover and contain differentnumbers and species of soil animals (Wallwork 1970). Since soil animals directlyaffect the development of humus forms, an understanding of their role in the soil isessential for the maintenance of high nutrient cycling and good plant growth within awell-managed forest ecosystem. Furthermore, by observing which humus forms or4horizons soil fauna prefer, we can increase our knowledge on the taxonomicclassification of humus forms.The soils in the CWHvin (= CWHbi) subzone occupy extensive areas onVancouver Island and on the adjacent mainland (Nuszdorfer et al. 1985). Their propermanagement could have enormous economic benefits for forestry in British Columbia.Future management of these soils could capitalize on silvicultural practices that wouldmeet main forestry objectives and at the same time favourably influence the soil faunato enhance soil fertility.However, in most Canadian soils, faunal composition has not yet beeninvestigated. Some exceptions include studies by Proctor (1977a, b), Ryan (1977) andAddison (1977) of Arctic Tundra soils on Devon Island in Northern Canada, Smith etL. (1990) also on Arctic Tundra soils in the northern Yukon, Dash and Cragg (1972)and Mitchell (1977) on sub-alpine aspen forest soil in Alberta, Willard (1972, 1973a,b, c, 1974a, b) and Willard et L . (1973) on grassland soil in Saskatchewan, andMarshall (1974) in a coastal forest soil in British Columbia. Other studies in temperateconiferous forests have been done in Alaska, Washington and Oregon as well asthroughout Europe. The purpose of this study is to determine and compare thecomposition of major soil fauna groups in virgin CH and HA forest types and,furthermore, to examine seasonal changes of soil fauna distribution in the soil horizonsof both forest types.1.1 Site DescriptionThe study site was located in the forest surrounding the SCHIRP site on Western ForestProducts' Tree Farm License (TFL) #25 (Block 4) northwest of Port McNeill(50° 60'/1270351) (Figure 1). This area is part of the Suquash Basin on northernVancouver Island and has a maximum elevation of 300 m. Biogeoclimaticclassification by Lewis (1982) placed the area in the Submontane North Island Coastal5LEGENDHA - HEMLOCK -AMABILISCH - CEDAR- HEMLOCKBK - DTERIMENTAL BLOCK(soil Fauna)Figure 1. Location of study,Western Hemlock subzone (CWHe). Germain (1985) classified the area as WetterNorthern Maritime Coastal Western Hemlock Subzone (CWHe). Further classificationby Messier (1991) placed the area in the submontane variant of the Very Wet CoastalWestern Hemlock subzone (CWHvm). The climate consists of cool, moist summersand mild, wet winters with minor snowfalls ( Figure 2). Rainfall is thought to besufficient through the growing season to prevent any soil moisture deficit (Messier1991).According to Lewis (1982), approximately 60 % of Block 4 in TFL 25 isdominated by a "salal-moss" ecosystem association (equivalent to the CWHvmi variantdescribed by Klinka et aL 1991). Within this ecosystem, two distinct forest types canbe identified: a cedar-hemlock (CH) phase (S1CH) and a hemlock-amabalis fir (HA)phase (siHA) (Lewis 1982). Both forest types are found growing side by side on siteswith similar parent material and underlying deposits (Lewis 1982). The transitionbetween the two forest types is often abrupt with the HA occurring on upper slopes andthe CH found below, usually in depressions.SlCH (CH) is considered to be a climatic climax community characterized bylarge, decadent old-growth western red cedar and western hemlock [Tsuga heterophylla(Ref.) Sargl. The canopy is relatively open and produces an understory of denseshrub dominated by salal (Gaultheria shallon  Pursh) with minimal herb [Blechnum spicant (L.)] and moss [Hylocomium splendens (Hedw.) B.S.C. and Rhytidiadelphusloreus (Hedw.) Warnst.] layers (Germain 1985, Messier 1991). The CH is equivalentto the C'WHbi(3) described by Green et al. (1984).siHA (HA) is characterized by dense, even-aged stands of western hemlock andamabalis fir [Abies amabalis (Dougl.) Forbes]. The closed canopy results in sparseshrub (Vaccinium alaskaense  Smith and V. parvifolium Howell) and herb [Blechnumspicant, Polystichum munitum (Kaulf.) Presl and Tiarella trifoliata L.] layers while theforest floor is dominated by a continuous moss covering, mainly Hylocomium 7mm200100Port Hardy Airport (22 m)^7.9 0^1782.8[30]Figure 2. Ecological climate diagram based on 30 year means in the style of Walter(1979). Left ordinate is temperature in intervals of 10 C . Right ordinate isprecipitation in intervals of 20 mm. Abscissa is months, beginning with January.Information at the top of the diagram includes station name, (elevation), mean annualtemperature in degrees Celsius, mean annual precipitation in millimeters and [number ofyears of observation]. Vertical line pattern indicates relatively humid periods. Blackshading indicates perhumid periods.8splendens and Stokesiella oregana (Su11.) Robins. (Germain 1985, Messier 1991). HAis more susceptible to periodic windthrow disturbances, the most recent occurringaround 1908, which results in friable soil conditions (Lewis 1982). The HA isequivalent to the CWHb (4) (Green et al. 1984).Lewis (1982) classified the soil of the S1 ecosystem as a moderately well (HA) toimperfectly (CH) drained Duric Humo-Ferric Podzol in coarse to medium texturedmaterials. Germain (1985) classified the soil as Duric Ferro-Humic Podzol whileMessier (1991) classified the soils as a Ferro-Humic Podzol. The parent material isdominated by sandy loam glacial tills with lesser areas of gravelly and sandyglaciofluvial and old sandy beaches (Lewis 1982).A deep, permeable, reddish-brown mineral Bf horizon is found in both the CHand HA phases. However, this horizon is friable in the HA phase and firm in the CHphase. This may influence air and water movements, soil temperature, root distributionand organic matter turnover rates. Lewis (1982) stated that the organic layer (LFH) ofthe HA phase was relatively thin (5-10 cm) and friable whereas in the CH phase, thelayer was thicker (10-25 cm) and more compact.Fox et al. (1987) also examined the soils in this area and classified them asLignic Folisols with a 70 cm accumulation of folic materials overlying a mineral soilwith a podzolic development. Germain (1985) identified an organic horizonapproximately 26 cm deep in the CH and 15 cm deep in the HA whereas Messier(1991) described an organic horizon 60 cm deep in the CH and 25 cm in the HA.Since the accumulation of organic material from both forest types ranged from less than15 cm to greater than 40 cm, the soil could be classified as a Podzol (folic) in order tocover the variation which occurs throughout both forest types (R. Trowbridge, Ministryof Forests, Smithers, B.C., pers. comm.).The organic horizons show the greatest variation between the two phases. In theHA, there is a LF (mat) complex sometimes held together by fungal hyphae and fine9roots followed by an F horizon, Hr (r = macroscopically recognizable plant residuesremain) and Hd (d = fine organic substances predominate with few plant residues).There is wood sometimes incorporated into the F, Hr and Hd horizons possibly as aresult of treefall. In the CH, the LF (mat) complex is more pronounced. Woodmaterial is also more abundant in the CH, which results in the formation of F X ,Hr X and Hd X horizons ( X = decaying wood horizon). Messier (1991) classified thehumus forms as Lignohumimor in the CH and Humimor in the HA (Klinka et al.1981).1 02.0 METHODS2.1 SamplingSampling occurred five times: August 9 and 10, 1989, May 24 and 25, 1990,July 22 and 23, 1990, October 14 and 15, 1990 and March 4 and 5, 1991. Samplingwas carried out in the four blocks identified in Figure 1 each time. Four 2.5 x 2.5 mplots were established in each forest type at the opposite ends of the black rectangles.Sampling plots were at least 50 m from the forest type boundary and not in the ecotonebetween the CH and HA. These plots were divided into 25 microplots (0.5 x 0.5 m).Each microplot was sampled only once during the entire study. At each samplingperiod, three microplots were selected at random from each of the eight plots. Two ofthe microplots were sampled using a 1 m long metal soil corer, which had a 5.1 cminternal diameter cutting edge. The main body of the corer was recessed to reducecompaction of the sample. These cores were used in the high gradient and modifiedBaermann funnel extractions. The penetration of the corer and length of the coreobtained was recorded. For August 1989 and May 1990 samples were taken at 3 cmintervals to 12 cm (i.e. 0-3; 3-6; 6-9, etc.) then every other 3 cm until 30 cm (i.e. 15-18; 21-24; 27-30) and then every other 6 cm until 100 cm or the mineral horizon wasreached, whichever came first. In the last three sampling periods (July 1990, October1990, and March 1991) samples were removed from the different, recognizablehorizons of the soil cores and their depths in the core recorded. A 0.3 x 0.3 m samplewas removed from the third microplot for hand-sorting of macrofauna in thelaboratory. The depth of these samples ranged from 5 to 10 cm to include the LFcomplex as most of the macrofauna would be expected to occur in these horizons.A 0.5 x 0.5 m plot was established at least 5 m away from each plot to samplefor Oligochaeta (Megascolecidae) using the Formalin method (Raw 1959). DuringAugust 1989, two washes of 9 L of 4.4 % formalin were used on the plots. For May1990 one wash of 9 L of 4.4 % formalin and one wash of 9 L of 8.8 % formalin were11used. For July 1990, one wash of 8.8 % formalin and one of 13.3 % formalin wereused. For October 1990 and March 1991 two washes of 13.3 % formalin (9 L each )were used. During the last four sampling periods the litter layer was removed fromeach plot to increase the absorption of formalin into the soil and to enhance observationof emerging individuals. Oligochaeta specimens were collected and preserved in 4 %formalin on site and returned to PFC in Victoria for identification.Samples were packed into coolers and transported to PFC in Victoria for theBaermann funnel extraction and to UBC for high gradient extraction and hand-sorting.Extractions were begun within 72 hours of sampling. The single exception was theMarch 1991 wet funnel samples which were stored in a cold room (4 0 C) for ten daysbefore extraction.2.2 Laboratory AnalysesA modified Lussenhop high gradient extractor (Lussenhop 1971) was used toextract microarthropods into a 1:1 (V:V) saturated aqueous picric acid solution (1.2 %w/v) : distilled water from one set of the soil cores. Extraction lasted for one week.Microarthropods were transferred to 70 % EtOH by washing out the collecting disheswith distilled water through a 50 Am screen until no picric acid remained. The contentsof the screen were then rinsed into glass shell vials (1 dram) using 70 % EtOH andstored for counting and identification later. The extracted soil samples were then oven-dried for at least 24 hours at 105 °C, weighed and the gravimetric moisture content foreach sample was calculated.The other set of soil samples was extracted for nematodes and enchytraeid wormsusing modified Baermann funnels (O'Connor 1962) at room temperature and humiditywith no additional light or heat. Soil samples were placed in a double-walled sampleholder with the bottom screen lined with facial tissue. Each sample was placed in a12glass funnel and partly submerged in distilled water. After 24 hours, the bottom 30 mlof H20 was drawn off. Five ml of 4 % formalin was added to preserve the specimensfor future counting and identification. After five days another 30 ml sample was drawnoff. Both samples were then transferred to 70 % Et0H using a fine porosity sinteredglass funnel.The 0.3 x 0.3 m soil samples were hand sorted for macrofauna (e.g. Coleoptera,Isopoda, Chilopoda, Diplopoda, etc ) in a white, enamel tray. Specimens werepreserved in 70 % Et0H.All samples were counted and identified using a dissecting microscope (8-50 X)with fibre optic lighting through a darkfield plate (manufactured by CHIU TechnicalCorp.) at UBC. Classification for all arthropods to class, order and/or family followsDanks (1979) except the Acari which follows Krantz (1978). Collembola andOligochaeta were identified to Family. Arachnida, Crustacea and Insecta wereidentified to Order. Mollusca, Myriapoda, Stelechopoda and the remaining Hexapoda(Protura and Diplura) were identified to Class. Nematodes and Rotifera were identifiedto Phylum.2.3 Data AnalysesThe counts of organisms were used to calculate percent abundance, abundanceper square meter and biomass per square meter. Biomass values were calculated forsixteen groups for which median individual dry mass (mg) were obtained from Petersenand Luxton (1982) (Table I).Morisita's Index of Similarity (MIS) (Morisita 1959, Wolda 1981) was calculatedfrom the actual counts of individuals. The counts were then transformed using [LN(x+1)], where x = the actual counts of individuals. These new values were used tocalculate the Simplified Morisita's Index of Similarity (SMIS)(Krebs 1989, Wolda13Table I. Median individual dry mass (mg) used to calculate biomass values (valuesfrom Petersen and Luxton 1982).Taxon^ median dry mass (mg)Nematoda 0.00005Enchytraeidae^0.032Megascolecidae *^17Collembola 0.0027Cryptostigmata^0.0053Mesostigmata 0.0077Prostigmata^ 0.001Diplopoda 14Chilopoda^ 1.9Araneae 0.8Gastropoda^ 8Diplura 0.069Pauropoda^ 0.006Symphyla 0.081Isopoda^ 0.2Diptera (L) 0.7* value given for large Oligochaeta (empty gut) found in temperate coniferous forests.141981). SMIS was also calculated for the biomass values. Renkonen Index wascalculated to compare the relative abundances (Krebs 1989, Wolda 1981) between thetwo phases (see Appendix A for equations).The mean depth of 23 groups was calculated according to Usher (1970). Thevertical distribution of these groups was then analyzed using the analysis of variance(ANOVA) program in SYSTAT (ver.5.0) (Wilkinson 1988). Box's small sample Fapproximation was used to test for homogeneity of variance since only four mean depthvalues were calculated for each group during each sampling time. Three groupsshowed heterogeneity of variance and the values were transformed by Logic, (X) whereX was the mean depth value. Three groups did not have enough values for the ANOVAto be performed, thus an independent samples T-test was done on mean depth groupedby forest type for the sampling periods which provided enough data. The remaining 17groups were analyzed using the ANOVA program without transformation. AnAnalysis of Covariance (ANCOVA) was also performed on the mean depth data forthel7 groups, using soil moisture as the covariate to determine if soil moisture had aneffect on vertical distribution.Distribution within organic horizons for six different faunal groups was alsoexamined. Percent abundance of each group in the mat complex (mat), F, H andmineral (min) horizons were plotted against sampling time. The horizon data for May1990 was split into two groups; the mat horizon and the remaining horizons combined(rest) due to poor horizon identification.153.0 RESULTS3.1 Abundance and BiomassForty-one major groups of soil organisms were extracted and identified during thestudy period. Tables II through VI show the mean abundance m -2 , relative abundance(%) and biomass (mg m -2), where available, for all the groups studied as well as thetotal fauna m -2 and total biomass (mg m-2) for each phase during each samplingperiod. None of these values were tested for significant differences due to the largevariances of the means which accompanied most of the values. Therefore the datafrom this study can only be used to suggest trends in the soil fauna communitiesbetween each forest type.The HA appears to maintain a higher abundance and biomass ( m -2) than theCH. In the HA, the highest mean abundance occurs in August 1989 (3.6 x 10 6 m-2)while the lowest abundance is found in March 1991 (1.2 x 10 6 m-2). In the CH,highest abundance also occurs in August 1989 (2.05 x 10 6 m-2) while the lowest valueoccurs in July 1990 (8 x 105 m-2). Both show the highest biomass in August 1989 (6.5g m-2 for the HA and 3.45 g m -2 for the CH). The lowest biomass value for the HAoccurs in October 1990 (1.44 g m -2). In the CH, the lowest biomass value appears inMarch 1991 (1.51 g m -2).Nematoda maintained the highest abundance throughout the sampling periods.Acari (in toto) were usually second followed by Collembola (in toto) and Copepoda.This order was maintained in the HA except during October 1990 where Collembola(in toto) were second followed by the Acari (in toto) and Copepoda. In the CH, therewas no clear order of abundance with Nematoda, Acari, Collembola and Copepodabeing the dominant groups.Biomass presents a very different picture of the faunal distribution. Diplopoda,Enchytraeidae, Diptera larvae and Acari tend to be the dominant organisms in bothforest types. A summary of the mean relative abundance and biomass values for these16Table II. Mean abundance (x 103 m-2), percent abundance (%) and mean biomass (mg m-2)of soil fauna for August 1989.AbundanceaCH% Biomass AbundanceHA% BiomassPhylum EntomaSubphylum ChelicerataClass ArachnidaSubclass AcariOrder Astigmata 0 0.1(0.2) 0.00Order Cryptostigmata 131(58) 6.39 694.3 342(165) 9.48 1812.6Order Mesostigmata 10(6) 0.49 77 16(7) 0.44 123.2Order Prostigmata 20(10) 0.98 20 110(79) 3.05 110Undetermined 15(30) 0.73 0Total 175(50) 8.54 752.9 468(222) 12.96 2010.7Subclass AraneidaOrder Aranaea 0.07(0.08) 0.00 56 0.06(0.02) 0.00 48Subclass ChelonethidaOrder Pseudoscorpionida 1.5^(1.1) 0.07 2.9(2.7) 0.08Subphylum CrustaceaOrder Copepoda 41(55) 2.00 14(18) 0.39Order Isopoda 0.03(0.05) 0.00 6 0Subphylum UniramiaSuperclass MyriapodaClass Pauropoda 0.2(0.3) 0.01 1.2 8(11) 0.22 48Class Diplopoda 0.09(0.1) 0.00 1260 0.1(0.08) 0.00 1400Class Chilopoda 0.09(0.09) 0.00 171 0.09(0.05) 0.00 171Class Symphyla 1.2 (1) 0.06 97.2 3.8(2.3) 0.11 307.8Superclass HexapodaClass Protura 0.6(0.7) 0.03 1(1.6) 0.03Class CollembolaOrder ArthropleonaFamily Anuridae 1.1(1.9) 0.05 0.7(0.6) 0.02Family Entomobryidae 4.3 (2.8) 0.21 18(6) 0.5Family Hypogasturidae 3.3(4.7) 0.16 12(8) 0.33Family Isotomidae 11(14) 0.54 38(42) 1.05Family Neanuridae 0.7(0.6) 0.03 3.5(2.8) 0.1Family Onychiuridae 79(12) 3.85 145(102) 4.02Family Tomoceridae 2.7(2.3) 0.13 2.6(2.4) 0.0717Table II. (continued)AbundanceCH% Biomass AbundanceHA% BiomassOrder SymphypleonaFamily Neelidae 1(0.9) 0.05 2.3(3) 0.06Family Sminthuridae 1.7(2.2) 0.08 3(3.7) 0.08Undetermined 8.2(11) 0.4 14(11) 0.39Total 113.4(40.4) 5.53 306.18 250(144.6) 6.93 675Class Diplura 0 0.2(0.5) 0.01 13.8Class InsectaOrder Homoptera 0.006(0.01) 0.00 0.006(0.006) 0.00Order Diptera (A) * 0.003(0.006) 0.00 0Order Diptera (L) ** 0.7(0.9) 0.03 490 1.7(1.5) 0.05 1190Order Coleoptera (A) O. 03(0.03) 0.00 0.04(0.02) 0.00Order Coleoptera (L) 0.006(0.006) 0.00 0.01(0.009) 0.00Order Hemiptera 0.006(0.006) 0.00 0Order Hymenoptera 0.006(0.006) 0.00 0Order Lepidoptera (L) 0 0.003(0.006) 0.00Order Thysanura 0.1(0.3) 0.00 0Phylum MolluscaClass Gastropoda 0.006(0.01) 0.00 48 0.003(0.006) 0.00 48Phylum Nematoda 1674(1442) 81.65 83.7 2822(1099) 78.21 141.1Phylum AnnelidaClass OligochaetaFamily Enchytraeidae 5.6(3.4) 0.27 179.2 15(8) 0.42 480Family Megascolecidae 0 0Phylum Rotifera 13(8) 0.63 23(14) 0.64Phylum StelechopodaClass Tardigrada 23(26) 1.12 9(9) 0.25TOTAL 2050.24 3451.41 3608.11 6509.38* (A) denotes adult**(L) denotes larvaea values in brackets are + 1 standard error.18Table III. Mean abundance (x 103 m-2), percent abundance (%) and biomass (mg m-2) of soilfauna for May, 1990.AbundanceaCH% Biomass AbundanceHA% BiomassPhylum EntomaSubphylum ChelicerataClass ArachnidaSubclass AcariOrder Astigmata 2.2(2.1) 0.13 18(11) 0.73Order Cryptostigmata 28(15) 1.62 148.4 101(56) 4.07 535.3Order Mesostigmata 6.7(3.9) 0.39 51.6 3(1.9) 0.12 23.1Order Prostigmata 6.1(2.3) 0.35 6.1 16(6.6) 0.65 16Undetermined 0.1(0.3) 0.01 1(1.1) 0.04Total 43(17.7) 2.49 184.9 139(72.8) 5.61 597.7Subclass AraneidaOrder Aranaea 0.05(0.05) 0.00 40 0.05(0.04) 0.00 40Subclass ChelonethidaOrder Pseudoscorpionida 0.4(0.5) 0.02 0.9(0.5) 0.04Subphylum CrustaceaOrder Copepoda 47(29) 2.72 66(77) 2.66Order Isopoda 0.1(0.1) 0.01 20 0Subphylum UniramiaSuperclass MyriapodaClass Pauropoda 1.5(2.3) 0.09 9 0.9(0.5) 0.04 5.4Class Diplopoda 0.03(0.02) 0.00 420 0.06(0.06) 0.00 840Class Chilopoda 0.06(0.05) 0.00 114 0.08(0.03) 0.00 152Class Symphyla 0.5(0.4) 0.03 40.5 0.5(0.4) 0.02 40.5Superclass HexapodaClass Protura 0.1(0.3) 0.01 0.3(0.5) 0.01Class CollembolaOrder ArthropleonaFamily Anuridae 0.7(0.8) 0.04 0.6(0.6) 0.02Family Entomobryidae 0.1 (0.3) 0.01 1.8(1.8) 0.07Family Hypogasturidae 6.9(4.6) 0.4 12(12) 0.48Family Isotomidae 20(12) 1.16 18(8.5) 0.73Family Neanuridae 0.3(0.5) 0.02 0Family Onychiuridae 29(17) 1.68 28(15) 1.13Family Tomoceridae 2.7(2) 0.16 0.9(0.8) 0.0419Table III. (continued).AbundanceCH% Biomass AbundanceHA% BiomassOrder SymphypleonaFamily Neelidae 3.2(2) 0.19 1.7(1.9) 0.07Family Sminthuridae 1.1(0.9) 0.06 0.1(0.3) 0.00Undetermined 4.8(4.5) 0.28 2.7(1.6) 0.11Total 68.4(22.1) 3.96 184.7 65.6(24) 2.65 177.12Class Diplura 0.1(0.3) 0.01 6.9 0.1(0.3) 0.00 6.9Class InsectaOrder Homoptera 0 0Order Diptera (A) * 0 0Order Diptera (L) ** 0.3(0.3) 0.02 210 0.1(0.3) 0.00 70Order Coleoptera (A) 0.02(0.02) 0.00 0.02(0.02) 0.00Order Coleoptera (L) 0.003(0.006) 0.00 0.02(0.02) 0.00Order Hemiptera 0.003(0.006) 0.00 0Order Hymenoptera 0.003(0.006) 0.00 0Order Lepidoptera (L) 0 0Order Thysanura 0 0Phylum MolluscaClass Gastropoda 0.01(0.02) 0.00 80 0.003(0.006) 0.00 24Phylum Nematoda 1539(1093) 89.2 79.95 2174(1203) 87.7 108.7Phylum AnnelidaClass OligochaetaFamily Enchytraeidae 17(4.6) 0.99 544 21(7.9) 0.85 672Family Megascolecidae 0.001(0.002)0.00 17 0Phylum Rotifera 2.2(0.3) 0.13 8.2(13) 0.33Phylum StelechopodaClass Tardigrada 5(5.1) 0.29 1.8(2.8) 0.07TOTAL 1725.28 1947.93 2478.83 2734.32*(A) denotes adult** (L) denotes larvaea values in brackets are + 1 standard error20Table IV. Mean abundance (x 103 m-2), percent abundance (%) and biomass (mg m-2) of soilfauna for July, 1990.AbundanceaCH% Biomass AbundanceHA% BiomassPhylum EntomaSubphylum ChelicerataClass ArachnidaSubclass AcariOrder Astigmata 7.2(14) 0.86 0.1(0.3) 0.01Order Cryptostigmata 31(36) 3.72 164.3 100(74) 6.76 530Order Mesostigmata 2.2(1.8) 0.26 16.94 4.8(1) 0.32 36.96Order Prostigmata 5(2.9) 0.6 5 6.7(6.2) 0.45 6.7Undetermined 4.7(5.4) 0.56 8.3(2.4) 0.56Total 50.3(53.2) 6.03 216.29 120(87.2) 8.13 516.86Subclass AraneidaOrder Aranaea 0.02(0.02) 0.00 16 0.02(0.02) 0.00 16Subclass ChelonethidaOrder Pseudoscorpionida 0.4(0.5) 0.05 0.6(0.5) 0.04Subphylum CrustaceaOrder Branchiopoda present absentOrder Copepoda 23(18) 2.76 36(49) 2.43Order Isopoda 0 0Subphylum UniramiaSuperclass MyriapodaClass Pauropoda 0 0Class Diplopoda 0.06(0.05) 0.01 840 0.07(0.03) 0.00 980Class Chilopoda 0.03(0.03) 0.00 57 0.04(0.009) 0.00 76Class Symphyla 1.3(1) 0.16 105.3 3.5(3.3) 0.24 283.5Superclass HexapodaClass Protura 0 0Class CollembolaOrder ArthropleonaFamily Anuridae 0.7(0.8) 0.08 1.8(1.9) 0.12Family Entomobryidae 0.3 (0.5) 0.04 0.6(0.7) 0.04Family Hypogasturidae 2.2(2.3) 0.26 5.8(3.9) 0.39Family Isotomidae 4.7(3.9) 0.56 23(20) 1.55Family Neanuridae 0.7(1.2) 0.08 1.5(1.7) 0.1Family Onychiuridae 41(50) 4.91 24(13) 1.62Family Tomoceridae 1.3(1.1) 0.16 0.3(0.5) 0.0221Table IV. (continued)AbundanceCH% Biomass AbundanceHA% BiomassOrder SymphypleonaFamily Neelidae 0.4(0.5) 0.05 1(1.4) 0.07Family Sminthuridae 0.3(0.5) 0.04 0.1(0.3) 0.01Undetermined 4.5(5) 0.54 3.2(1.7) 0.22Total 56.3 (61.3) 6.75 152.01 60.7(39.4) 4.1 163.89Class Diplura 0.1(0.3) 0.01 6.9 0Class InsectaOrder Homoptera 0 0Order Diptera (A) * 0.003(0.006) 0.00 0Order Diptera (L) ** 0 0.7(1.2) 0.05 490Order Coleoptera (A) 0.01(0.02) 0.00 O. 003(0. 006) 0.00Order Coleoptera (L) O. 003(0. 006) 0.00 O. 003(0. 006) 0.00Order Hemiptera 0.003(0.006) 0.00 0Order Hymenoptera 0 0.003(0. 006) 0.00Order Lepidoptera (L) 0 0.003(0.006) 0.00Order Thysanura 0 0Phylum MolluscaClass Gastropoda 0 0Phylum Nematoda 687(204) 82.33 34.35 1223(397) 82.67 61.15Phylum AnnelidaClass OligochaetaFamily Enchytraeidae 6.2(2.1) 0.74 198.4 19(11) 1.28 608Family Megascolecidae 0.008(0.01) 0.00 136 0.001(0.002) 0.00 17Phylum Rotifera 4.5(4.5) 0.54 9.3(11) 0.63Phylum StelechopodaClass Tardigrada 5.6(2.5) 0.67 5.9(5.7) 0.4TOTAL 834.44 1762.25 1479.34 3212.4*(A) denotes adult**(L) denotes larvaea values in brackets are + 1 standard error22Table V. Mean abundance (x 103 m-2), percent abundance (%) and biomass (mg m-2) of soilfauna for October, 1990.AbundanceaCH% Biomass AbundanceHA% BiomassPhylum EntomaSubphylum ChelicerataClass ArachnidaSubclass AcariOrder Astigmata 0 0Order Cryptostigmata 27(36) 2.73 143.1 25(23) 1.44 132.5Order Mesostigmata 2.6(2.6) 0.26 20.02 2.6(2.7) 0.15 20.02Order Prostigmata 0.6(1.2) 0.6 0.6 1(1.1) 0.06 1Undetermined 0.3(0.3) 0.03 1.8(3.7) 0.1Total 28.6(40.7) 2.9 122.98 30.2(27.1) 1.73 129.86Subclass AraneidaOrder Aranaea 0.06(0.02) 0.01 48 0.03(0.04) 0.00 24Subclass ChelonethidaOrder Pseudoscorpionida 0.1(0.3) 0.01 0.4(0.5) 0.02Subphylum CrustaceaOrder Copepoda 36(33) 3.65 8.1(12) 0.46Order Isopoda 0.04(0.06) 0.00 8 0.01(0.02) 0.00 2Subphylum UniramiaSuperclass MyriapodaClass Pauropoda 0 0Class Diplopoda 0.08(0.05) 0.01 1120 0.02(0.02) 0.00 280Class Chilopoda 0.04(0.006) 0.00 76 0.06(0.02) 0.00 114Class Symphyla 0 0Superclass HexapodaClass Protura 0.1(0.3) 0.01 0Class CollembolaOrder ArthropleonaFamily Anuridae 0.3(0.5) 0.03 0.1(0.3) 0.01Family Entomobryidae 0.1^(0.3) 0.01 0.1(0.3) 0.01Family Hypogasturidae 1.2(1.9) 0.12 0.4(0.5) 0.02Family Isotomidae 6(1.7) 0.61 7.5(1.8) 0.43Family Neanuridae 0.1(0.3) 0.01 0.6(0.7) 0.03Family Onychiuridae 4.2(4.4) 0.43 27(39) 1.55Family Tomoceridae 1(0.9) 0.1 0.3(0.2) 0.0223Table V. (continued).AbundanceCH%^Biomass AbundanceHA% BiomassOrder SymphypleonaFamily Neelidae 0 0Family Sminthuridae 0 0Undetermined 5.9(9.2) 0.6 3.8(6.3) 0.22Total 18.7(8.8) 1.89^50.49 40.1(36.6) 2.3 108.27Class Diplura 0 0.4(0.5) 0.02 27.6Class InsectaOrder Homoptera 0 0Order Diptera (A) * 0 0Order Diptera (L) ** 0.1(0.3) 0.01^70 0.6(0.6) 0.03 420Order Coleoptera (A) 0.02(0.02) 0.00 0.03(0.02) 0.00Order Coleoptera (L) 0.006(0.01) 0.00 0.01(0.02) 0.00Order Hemiptera 0.003(0.006) 0.00 0.003(0.006) 0.00Order Hymenoptera 0 0.006(0.01) 0.00Order Lepidoptera (L) 0 0Order Thysanura 0 0Phylum MolluscaClass Gastropoda 0 0Phylum Nematoda 888(445) 89.94^44.4 1646(732) 94.49 82.3Phylum AnnelidaClass OligochaetaFamily Enchytraeidae 9.3(5.7) 0.94^297.6 7.8(5.5) 0.45 249.6Family Megascolecidae 0 0Phylum Rotifera 1(1.6) 0.1 5.1(7.8) 0.29Phylum StelechopodaClass Tardigrada 3.2(1.5) 0.32 3.2(2.7) 0.18TOTAL 987.35 1837.47 1741.97 1437.63* (A) denotes adult** (L) denotes larvaea values in brackets are + 1 standard error24Table VI. Mean abundance (x 103 m-2), percent abundance (%) and biomass (mg m-2) of soilfauna for March, 1991.AbundanceaCH% Biomass AbundanceHA% BiomassPhylum EntomaSubphylum ChelicerataClass ArachnidaSubclass AcariOrder Astigmata 0 0Order Cryptostigmata 27.7(20.5) 1.95 146.8 92.9(63.7) 7.48 492.37Order Mesostigmata 3.4(5.9) 0.24 26.18 2.3(1.9) 0.19 17.71Order Prostigmata 7.3(5.3) 0.51 7.3 17.3(8.3) 1.39 17.3Undetermined 0.2(0.5) 0.01 0Total 38.7(28.9) 2.72 166.41 112.5(65) 9.06 483.75Subclass AraneidaOrder Aranaea 0.04(0.01) 0.00 32 0.03(0.02) 0.00 24Subclass ChelonethidaOrder Pseudoscorpionida 0.4(0.2) 0.03 0.6(0.6) 0.05Subphylum CrustaceaOrder Copepoda 27.3(20.2) 1.92 17.5(17.8) 1.41Order Isopoda 0.03(0.02) 0.00 6 0Subphylum UniramiaSuperclass MyriapodaClass Pauropoda 0.1(0.2) 0.01 0.6 0.2(0.5) 0.02 1.2Class Diplopoda 0.04(0.05) 0.00 560 0.06(0.04) 0.00 840Class Chilopoda 0.03(0.04) 0.00 57 0.04(0.03) 0.00 76Class Symphyla 0.003(0.006) 0.00 0.24 0Superclass HexapodaClass Protura 0 0Class CollembolaOrder ArthropleonaFamily Anuridae 8.4(16.6) 0.59 1.1(1) 0.09Family Entomobryidae 0 0Family Hypogasturidae 1.5(1) 0.11 2.1(2.6) 0.17Family Isotomidae 4.4(4.1) 0.31 14.2(3.9) 1.14Family Neanuridae 0.1(0.2) 0.01 0Family Onychiuridae 18.4(20) 1.29 18.4(12.2) 1.48Family Tomoceridae 0.7(0.3) 0.05 2.6(2) 0.2125Table VI. (continued).AbundanceCH%^Biomass AbundanceHA% BiomassOrder SymphypleonaFamily Neelidae 0.1(0.2) 0.01 0.2(0.3) 0.02Family Sminthuridae 0 0.5(0.4) 0.04Undetermined 1.3(1) 0.09 4.2(3) 0.34Total 34.9(27.1) 2.46^94.23 43.2(14.4) 3.48 116.64Class Diplura 0 0.5(0.4) 0.04 34.5Class InsectaOrder Homoptera 0 0Order Diptera (A) * 0 0Order Diptera (L) ** 0.2(0.5) 0.01^140 1.7(1.8) 0.14 1190Order Coleoptera (A) 0 0.008(0.17) 0.00Order Coleoptera (L) 0 0.01(0.01) 0.00Order Hemiptera 0.006(0.006) 0.00 0Order Hymenoptera 0.003(0.006) 0.00 0.003(0.006) 0.00Order Lepidoptera (L) 0.006(0.01) 0.00 0Order Thysanura 0 0Phylum MolluscaClass Gastropoda 0.003(0.006) 0.00^24 0Phylum Nematoda 1259(395) 88.57^62.95 1005(665) 80.93 50.25Phylum AnnelidaClass OligochaetaFamily Enchytraeidae 11.3(5.2) 0.79^361.6 9.9(5.7) 0.8 316.8Family Megascolecidae 0 0Phylum Rotifera 7(3.5) 0.49 3.8(1.9) 0.31Phylum StelechopodaClass Tardigrada 7.6(10.8) 0.53 3.9(2.1) 0.31TOTAL 1421.46 1505.03 1241.75 3133.14*(A) denotes adult**(L) denotes larvaea values in brackets are + 1 standard error26dominant groups for the entire sampling period is shown in Table VII. Nematodamaintain the highest abundance values overall in both forest types followed by theAcari, Collembola and Copepoda. The HA shows a greater relative abundance ofAcari than the CH; Collembola are equal in both forest types and Copepoda maintaingreater abundances in the CH than the HA. These four groups account forapproximately 98 % of the soil fauna in the CH and 99 % in the HA.The biomass values indicate that the Diplopoda maintain the highest values inboth forest types followed by Enchytraeidae, Acari and Diptera larvae in the CH andAcari, Diptera larvae and Enchytraeidae in the HA. These four groups represent about78 % of the biomass in the CH and 82 % in HA. It is interesting to note that whileNematoda have the greatest % abundance, they are responsible for a small portion ofthe total biomass in both forest types (about 3 %). Furthermore, the Diplopoda, whichaccount for the largest amount of the relative biomass (40 % in the CH and 26 % in theHA), are responsible for less than 0.01 % of the total faunal abundance in the soil.3.1.1 NematodaWhile numerically dominant, nematodes are responsible for little biomass.Highest abundance (and therefore biomass) occurred in August 1989 for both the CH2 and 83.7 mg m-2) and HA (2.8 x 106 m-(1.67 x 106 m-^ 2 and 141.1 mg m-2). Thelowest values in the CH occurred in July 1990 (6.87 x 105 m-2 and 34.4 mg m-2). Inthe HA, the lowest values occurred in March 1991 (1 x 10 6 m-During all sample periods the HA maintained a higher abundance and biomass ofnematodes than the CH.2 and 50.3 mg m-2).27Table VII. Summary of mean relative abundance and biomass values for the dominantgroups from all sampling times.% AbundanceCH^HA% BiomassCH^HANematoda 86 85 3 3Acari 5 8 14 22Collembola 4 4 7 7Copepoda 3 2Enchytraeidae 1 1 15 14Diptera larvae 0.01 0.05 9 20Diplopoda <0.01 <0.01 40 26TOTAL* 98 99 78 82* Total values for top four groups only.283.1.2 AcariThe Acari were usually the second most abundant group. Highest abundanceswere observed in both forest types in August 1989 (175 x 10 3 m-2 and 752.9 mg m-2in the CH and 468 x 103 m-2 and 2010.7 mg m-2 in the HA). Low values occurredfor both phases in October 1990 (28.6 x 10 3 m-2 and 122.98 mg m-2 for the CH and30.2 x 103 m-2 and 129.86 mg m2 for the HA). Cryptostigmata was the dominantgroup and followed the same pattern as Acari (in toto). The HA maintained a higherabundance and biomass of Acari throughout the sampling periods. Likewise,Cryptostigmata maintained a higher abundance and biomass except in the October 1990sample where the CH values appear slightly higher (27 x 10 3 m-2 and 143.1 mg m -2)than the HA (25 x 103 m2 and 132.5 mg m -2).3.1.3 CollembolaCollembola were generally third in abundance in both phases. High abundancevalues occurred in both phases in August 1989 (113.4 x 10 3 m-2 in the CH and 342 x103 m-2 in the HA). Like the Acari, the Collembola (in toto) have the lowestabundance in October 1990 (18.7 x 10 3 m2 for the CH and 40.1 x 10 3 m2 for theHA). The biomass follows the same pattern; however the Collembola are not amongthe groups with substantial biomass within either forest type.Family Onychiuridae is the dominant group of Collembola. High abundancevalues are found in August 1989 (79 x 10 3 m-2 for the CH and 145 x 103 m-2 for theHA). In the CH, low value is found in October 1990 (4.2 x 10 3 m-2) which is secondto the Isotomidae (6 x 103 m-2) at this time. The low value for the HA is found inMarch 1991 (18.4 x 10 3 m -2) which is the same as the value in the CH at this time.293.1.4 CrustaceaThree orders of Crustacea were identified during the study. The order Copepodawas found in both phases. This is the first record of soil copepods in Canada (C. Shih,Canadian Museum of Nature, Zoology Division, Ottawa, ON, pers. comm.). Theabundances in the CH ranged from a high of 47 x 103 m-2 in May 1990 to a low of23 x 103 m-2 in July 1990. The abundances in the HA had a wider range from66 x 103 m-2 in May 1990 to 8.1 x 103 m-2 in October 1990. This order was moreabundant in the CH during August 1989, October 1990 and March 1991 and in the HAduring the other two sampling periods.The order Isopoda were found predominantly in the CH. The abundance rangedfrom 100 m-2 in May 1990 to zero in July 1990. This order was only collected once inthe HA phase during October 1990 (10 m-2).The order Branchiopoda was identified only in July 1990 in the CH. It was neverobserved in the HA phase.3.1.5 OligochaetaThe Megascolecidae were observed in the CH (1 m-2 for May 1990 and 8 m-2for July 1990) and only in July 1990 of the HA (1 m-2). The specimens collected werelater identified as Arctiostrotus perrieri Benham by W. M. Fender ( Soil BiologyAssociates, McMinnville, Oregon).Enchytraeidae were the dominant Oligochaeta. Abundances ranged in CH from17 x 103 m-2 in May 1990 to 5.6 x 103 m-2 in August 1989. In the HA the highabundance occurred in May 1990 (21 x 103 m-2) and the low in October 1990(7.8 x 103 m-2). The abundance of enchytraeids was greater in the HA during August1989, May 1990 and July 1990 while being higher in the CH during the remaining twosampling periods. The Formalin method also caused large enchytraeids to surface.These organisms ranged in length from approximately 3 to 8 cm. While not30numerically important, these organisms could represent a substantial portion of thebiomass within both forest types due to their large size. Further taxonomicidentification is required with these specimens.3.1.6 Other GroupsEach remaining group usually represented less than 1 % of the total faunalabundance at any sampling time. The single exception is Tardigrada in August 1989(Table II) which represented 1.12 % of the total abundance. Several families ofColeoptera were collected during the study; however, their low numbers did not justifycounting them at the family level. Families identified include Carabidae,Staphylinidae, Pselaphidae, Phalacridae, Tenebrionidae and Curculionidae.Slugs were observed in both forest types but never collected in any of theextraction procedures. The frog, Rana aurora Baird and Gerard, was observed in thetransition zone between the two forest types during May 1990.3.1.7 Supplementary Abundance DataTable VIII shows a comparison of abundance estimates ( m -2) calculated fororganisms that were collected from the high gradient extractor and by hand sorting inAugust 1989 (CH only). Abundance estimates from the high gradient extractor areusually higher than those determined by hand sorting. In this example, there are twoexceptions, Diplopoda and Diplura, which show greater abundances based on the handsorting. Large Enchytraeidae were collected by hand sorting and the Formalin method.The Formalin method, in this instance, indicates an abundance estimate which is muchlower than the hand sorting value.31Table VIII. Abundances ( m-2) for groups recovered by two different extractionmethods for August 1989 (CH only).Hand sorting High gradient FormalinAranaea 70 360Pseudoscorpionida 3 1500Diplopoda 90 0Chilopoda 90 120Symphyla 20 1200Neanuridae 3 700Tomoceridae 20 2700Diplura 30 0Coleoptera (A) 30 100Coleoptera (L) 6 120Diptera (A) 3 120Diptera (L) 10 700Hemiptera 6 120Thysanura 0 100Enchytraeidae* 19 1* Large enchytraeids323.2 Vertical Distribution3.2.1 Mean DepthTable IX shows the results of the ANOVA performed on the mean depth data.Analysis of Covariance (ANCOVA) was attempted using the gravimetric soil moisturecontent as the covariate before the ANOVA procedure was used. No significantinteraction was observed between mean depth of these faunal groups and soil moisturecontent. Three groups (Astigmata, Entomobryidae and Neanuridae) did not haveenough values for the ANOVA to be performed, thus an independent samples T-testwas done on mean depth grouped by forest type for the sampling periods whichprovided enough data (a behind group name in Table IX). A fourth group,Sminthuridae, also did not have enough values for the ANOVA to be performed, thus aT-test was done on mean depth between August 1989 and May 1990 samples for theCH only (c behind group name on Table IX). Three groups showed heterogeneity ofvariance (Cryptostigmata, Mesostigmata and Copepoda) and the values weretransformed by Log10 (X) where X was the mean depth value before performing theANOVA (b behind group name in Table IX).Pseudoscorpionida is the only group which shows a significant forest type bysampling time interaction (p <0.05). Prostigmata, Isotomidae, Onychiuridae andNematoda show a significant difference between sampling times (p <0.05).Prostigmata, Acari (in toto), Onychiuridae, Collembola (in toto) and Symphyla show asignificant difference in vertical distribution between phases only when p <0.10.Cryptostigmata, Copepoda and Rotifera show a significant difference in verticaldistribution between sampling times only when p < 0.10.No trends in seasonal distribution were observed in the mean depth data, possiblyas a result of the large variances and/or level of taxonomy.33Table IX. Results of ANOVA for mean depth of faunal groups through time and foresttype.Fauna^Time^Forest Type^Type x TimeAstigmata a May 1990Cryptostigmata b^*Mesostigmata bProstigmata^**^*Acari (in total) *AnuridaeEntomobryidae a^Aug 1989HypogasturidaeIsotomidae **Neanuridae a^Aug 1989NeelidaeOnychiuridae^**^*Sminthuridae cTomoceridaeCollembola (in total)^ *Copepoda b^*EnchytraeidaeNematoda **PauropodaPseudoscorpionda^ **Rotifera^*Symphyla *Tardigrada** significant alpha = 0.05* significant alpha =-- 0.10a Independent samples T-test on mean depth grouped by type for sampling timeindicated in Time column.b Mean depth values transformed by LOGI() (X) before ANOVA.c Independent samples T-test on mean depth between August 1989 and May 1990 forCH phase only.see Appendix C for mean depth values and standard deviation343.2.2 Distribution through HorizonsFigures 3 through 8 show the percent abundance of individuals counted insampled horizons identified in each sampling period. Acari (in toto), Cryptostigmata,Collembola (in it), Onychiuridae, Nematoda and Enchytraeidae were selected fortabular presentation because of their high abundance and/or biomass. Diplopoda werenot included because they were collected from the hand-sorted samples and areconsidered to occupy the upper litter layers (Hoffman 1989). Missing horizons in thefigures indicates that the horizon was not sampled during that sampling time or was notrepresented in the sample taken. The columns may not equal 100 % due to theexclusion of minor horizons. Furthermore, the May samples were divided into twosections, the mat complex (mat) and the rest of the sample (rest).The percent abundance in the CH for Acari (in toto), Cryptostigmata, Collembola(in toto) and Onychiuridae (Figures 3a, 4a, 5a and 6a, respectively) decreases in themat from March to July and then increases through August and October. The Fhorizon shows a higher abundance in July than the mat complex. The Enchytraeidae inthe CH (Figure 7a) decrease in abundance in the mat from March to August with anincrease in abundance in the mat occurring in October. In August, the H horizonshows the highest abundance in the organic layer. Figure 8a shows the distribution ofthe Nematoda in the organic horizons of the CH. The abundance in the mat fluctuatesaround 20 % throughout the study period with a large abundance appearing in the Hhorizon in August.Figures 3b, 4b, 5b and 6b show the abundance distributions for the HA.Abundances peak in the mat complex during May and decrease through August. InAugust, the percent abundance is divided between the F and H horizons. TheEnchytraeidae (Figure 7b) show the same pattern in the HA as the CH with a largeabundance of the population occurring in the H horizon in August. In Figure 8b, theNematoda have very little fluctuation in the mat complex. A large abundance is35Jul OctMar AugMay(a) CH100 — n = 84 n = 6480 —60 —d40200n = 617n = 89 n = 176Jul OctMar AugMayrestF^Hmat mineral(b) HA100 _ n = 3328004844 60 —.2.40n = 282 n = 86Figure 3. Seasonal distribution of Acari (in taw) abundance (%) throughsampled horizons. n = mean total number of individualscounted. Columns do not equal 100 % due to the exclusionof minor horizons.36Mar^May^Jul^Aug^Octmat^ V. mineral^restFigure 4. Seasonal distribution of Cryptostigmata abundance (%) throughsampled horizons. n = mean total number of individualscounted. Columns do not equal 100 % due to the exclusionof minor horizons.37Mar May Jul Aug O ct80 -F restmineral• matvH^/(b) HA100 - n = 1328IS 60_40 -20 -n = 96Figure 5. Seasonal distribution of Collembola (in toto) abundance (%) throughsampled horizons. n = mean total number of individualscounted. Columns do not equal 100 % due to the exclusionof minor horizons.38Jul Aug Octn = 66Mar^May Jul^Aug^Oct(a) CH10040n = 59n = 56200n = 45rfil FA P'/A1 PA IF/AP218060n = 213n = 9IEl mat  ^F^iH^mineral^restFigure 6. Seasonal distribution of Onychiuridae abundance (%) throughsampled horizons. n = mean total number of individualscounted. Columns do not equal 100 % due to the exclusionof minor horizons.39100 - n = 28 n= 18 n = 22n = 24 ,,80 -60tiR40 -20 -00MarA8 80g 604020Mar Mayn = 43MayAugn = 40AugJul(b) HA100^n = 28JulOctn = 19O ct(a) CH n = 35mat^F^H^mineral^restFigure 7. Seasonal distribution of Enchytraeidae abundance (%) throughsampled horizons. n = mean total number of individualscounted. Columns do not equal 100 % due to the exclusionof minor horizons.40(a) CH100n = 31436040n = 2815Aug Oct(b) HA n = 4442n = 41208020Mar May Jul•Jul^Aug^OctMayMarn = 2963n = 7948n = 5484 An = 316610080604020mineral^restFigure 8. Seasonal distribution of Nematoda abundance (%) throughsampled horizons. n = mean total number of individualscounted. Columns do not equal 100 % due to the exclusionof minor horizons.41present in the F horizon during July; however in August, the population appears to beequally spread out in the organic horizons.The general trend indicated by these figures is that the fauna is found in the matcomplex during the spring and fall and in the F horizon during the summer monthsexcept for the Nematoda which tend to dominate the H horizon and maintain arelatively constant abundance in the mat complex throughout the year. Gravimetric soilmoisture contents for the mat complex plotted in Figure 9 show similar trends to thoseexhibited by these faunal groups. Low abundances for the faunal groups in the matcomplex coincide with low moisture contents (in July for the CH and in August for theHA). Even with this change in moisture content in the mat complex, Nematodamaintain a fairly constant relative abundance throughout the sampling period. Thisseems to indicate some seasonal variation in distribution between the organic horizonsin these six groups that is associated with soil moisture in some way.3.3 Similarity IndicesTable X shows the similarity indices calculated to compare the forest types.Morisita's Index (MIS) ranges from 1 (similar) to 0 (dissimilar). The MIS valuescalculated indicate that soil fauna communities, determined to high taxonomic levels(i.e. Family and above), are similar in composition between CH and HA ecosystems.Furthermore, the MIS value was not influenced by the inclusion of counts oforganisms collected by two different extraction methods. The MISa was calculatedusing the counts for all organisms from all extraction methods. The MISb and MIScvalues were calculated using counts obtained from the high gradient extractor and handsorting, respectively. Counts from the proper extraction method for each group oforganisms were used to calculate MISd. With the proper extraction methods, May1990 shows a lower similarity value (0.8626) than the other four sampling times (all42450 —Mar^May^Jul^Aug^OctIN CH^17 HAFigure 9. Gravimetric soil moisture content (%) for the mat complexfor both forest types.43Table X. Similarity Indices calculated for both forest types for each sampling timeIndex Aug-89 May-90 Jul-90 Oct-90 Mar-91MISa 0.9974 0.8399 0.9985 0.9985 0.9998MISb 0.9979 0.8623 0.9976 0.9987 0.9995MISc 0.998 0.8387 0.9984 0.9987 0.9999MISd 0.9979 0.8626 0.9985 0.9985 0.9999SMIS la 0.9469 0.9384 0.9599 0.9546 0.9606SMIS2a 0.7750 0.9512 0.8937 0.0752 0.7712SMIS2b 0.7746 0.9485 0.8488 0.0769 0.7718SMIS2c 0.9467 0.9227 0.9562 0.8451 0.9491SMIS2d 0.9174 0.9064 0.8395 0.5888 0.7519RI 93.16% 96.2% 94.33% 93.23% 91.37%MIS = Morisita's Similarity Index (Morisita 1959)SMIS = Simplified Morisita Similarity Index (Krebs 1989)RI = Renkonen Index (Krebs 1989, Wolda 1981)see Appendix A for equations of above indices1 SMIS calculated using transformed counts LN(x+1)2 SMIS calculated for faunal biomass (mg m -2)a index calculated using counts for fauna collected from all extraction methodsb index calculated using values obtained from high gradient extractorc index calculated using values obtained from hand-sortingd index calculated using values obtained from proper extraction method for fauna (highgradeint extractor for microarthropods, hand-sorting for macrofauna)44greater than 0.99); however this does not suggest a large difference between the twoforest types.The SMIS values calculated for the transformed values obtained using propermethods (SMIS1 in Table X) suggest, again, a similar community of soil fauna betweenthe two ecosystems (all values > 0.93). The SMIS values calculated for the biomass(SMIS2) show a range of results. In this case it is important to use values obtainedfrom the most efficient extraction method for the organisms being studied. Using allvalues obtained from all the extraction methods (SMIS2a), October 1990 showed twodistinct communities. This is a result of using values collected for the Diplopoda fromthe high gradient extractor which misrepresents the actual population in the forest types(Diplopoda are present in both ecosystems not just the CH). SMIS2b and SMIS2cshow values calculated using data from the high gradient extractor or from hand-sortingonly. While the differences between most of the sampling periods is slight, October1990 shows a large difference between the two values (0.0769 for SMIS2b and 0.8451for SMIS2c). Values calculated for SMIS2d give a better representation of thedifference in biomass between the two forest types. October 1990 still shows a lowsimilarity value (0.5888). The Renkonen Index (RI in Table X) compares the relativeabundance of organisms between the two forest types. This index suggests there islittle difference between the two forest types with values ranging from 91.37% inMarch 1991 to 96.2 % in May 1990.454.0 DISCUSSION4.1 Abundance and BiomassSpecies diversity can be defined as the total number of species that occur in aparticular community (Krebs 1978). The simplest way to examine species richness ordiversity then is to create a list of species found in the community being studied. Whenlooking at soil fauna, this task is more difficult than it appears since soil fauna areincorporated closely into the soil structure (Edwards 1991). Due to limited time andlabour, this study examined higher taxonomic groups (Family and above) of soil fauna.The group diversity of the two forest types is shown in Tables II through VI and, atthis taxonomic level, the soil fauna communities of the CH and HA are roughly thesame. All the groups listed (41 in total) appear in both forest types at some pointduring the sampling. However, presence/absence data can be misleading and result ina misinterpretation of the data. If one individual of the Acari is counted in one of theforest types, the group would be "present" in that forest type. If 1,000 individuals ofthe same group are collected in the other forest type, the group is also present. Theproblem here is the different effect 1 individual versus 1,000 individuals might have ineither forest type. Danks (1988) reported that there are 1,915 species of Acariidentified in Canada and estimated that there were also some 7,567 species stillundescribed or unrecorded. In this study, Acari have only been identified to theordinal level and, therefore, we have lost information that might have been importantwhen trying to compare differences in diversity of the soil fauna communities betweenthese forest types.Numerical abundance and biomass of the groups being studied provide us with abetter picture of the structure of the soil fauna communities in the CH and HA. Thegeneral trend observed in the study is that the HA maintains a higher abundance andbiomass overall throughout the study period except in March 1991 when the abundanceis slightly greater in the CH (1.42 x 10 6 individuals m-2) than the HA (1.24 x 10646individuals m-2). This trend is supported by the statement made by Schaefer andSchauermann (1990) that soil fauna biomass is positively correlated with decompositionrate. The HA is considered to be a more productive forest type than the CH, whenconsidering new plantations. This difference could be the result of a more rapiddecomposition rate caused by a larger soil fauna community present in the HA than theCH. Thus, nutrients are more readily available in the HA due to the higherdecomposition rate and trees, therefore, show better growth.The values in Table VII indicate that four groups compose 98 % of the faunalpopulation in the CH and 99 % in the HA. Another three groups along with one fromthe previous set are responsible for 78 % of the biomass in the CH and 82 % of thebiomass in the HA. Nematoda maintain the highest abundance in both forest types, butit is Diplopoda that are responsible for the greatest biomass in each forest type. In theCH, Diplopoda provide nearly 40 % of the biomass with the next group,Enchytraeidae, responsible for less than half of the Diplopoda portion. The HA showsa slightly more even pattern of biomass distribution among the dominant groups. Itmay then be suggested that a more even distribution of biomass among soil faunagroups creates a more balanced ecosystem which would allow for better decompositionand nutrient cycling. A single group responsible for a large portion of the biomass maycreate a "bottleneck" in decomposition and nutrient cycling thus limiting theproductivity of the system. However, this may change as populations fluctuatethroughout the year.Wallwork (1976) and Edwards (1991) suggest that soil fauna populations arehighest during the spring and fall. In this study, population peaks occur in May 1990and August 1989 for total fauna abundances. From the climate diagram we know thatMay shows spring-like conditions (increasing temperature and decreasing precipitation)while August maintains summer-like conditions with high temperatures and relativelylow precipitation. However, information on the specific years studied was not47collected. Sampling during September, which starts to show fall-like climateconditions (decreasing temperatures and increasing precipitation), could show an evenhigher faunal abundance than August. These spring and summer peaks could be relatedto increased activity in root and fungal growth during these times of year, whichprovides food sources for most soil animals (S. Berch, Soil Science Dept., UBC, pers.comm.). With an increase in food availability, populations would increase until foodbecame a limiting resource. At this point the fauna may migrate to find a better foodsource, mortality may occur thus reducing the abundances or the fauna may switch toanother, less preferable food source. Thus, knowledge of life cycles and feeding habitsof soil fauna species becomes very important for interpretation of these seasonal trends.The results do not indicate significant differences between the communities of thetwo forest types. The large variability encountered within each forest type duringsampling masks the possible differences between the forest types. Furthermore, thelevel of taxonomy limits further critical analysis. Anderson (1977) stated that mostclasses and orders of soil organisms are cosmopolitan in their distribution whilefamilies are more restricted and genera and species are highly endemic to isolatedhabitats or land masses. Thus, identification to genus or species level could provide theinformation required to show any differences in species diversity between the twoforest types. Identification to genus or species could also increase the variability ofmean numerical abundance estimates of the populations being studied, but thevariability could be alleviated by increasing the number of samples that are taken.The abundance and biomass calculations are dependent upon the efficiency of theextraction method being used to collect the group, as well as the efficiency of theoperator in collecting and identifying all the individuals. Table VIII shows howimportant it is to use the most suitable extraction method for the organisms beingcollected. The fact that an organism is not collected by one method does not mean thatit is not present in the area (e.g. Diplopoda, Diplura, Thysanura). Two problems arise48here. First, the multiplication factor used to calculate the abundance is different foreach method. One individual found in the hand-sorting would produce an abundance of11 individuals m-2 , while that same individual, found in the high gradient extractor,would result in an abundance of 490 individuals m -2 . This problem is transferred intothe biomass calculations and can result in a serious over- or under-estimate of thepopulation being studied within the ecosystem.An alternate way of looking at these values is to consider them as representingdifferent size classes of the same group of organism. The high gradient extractor isbiased towards smaller individuals and species. The screen in the bottom of the soilholder was 1 x 1 mm thus limiting the collection to individuals that can fit through thescreen. Collection by hand-sorting selects for the larger members of the population.This method allows for the collection of individuals which are readily observed by thecollector. Hand-sorting is dependent upon the person doing the sorting as well as theactivity level of the organisms. Most small individuals are overlooked during thisprocedure but can be collected by the high gradient extractor which is better suited forthis purpose. By combining methods it may be possible to obtain a better idea of thepopulation present or discern which method might be better for extraction of the group.Unfortunately, both extraction methods have drawbacks. The high gradient extractortends to collect only the mobile or active portion of the population while hand-sortingallows for the collection of both living and dead individuals. These disadvantages cancause problems in interpreting the populations being studied.4.1.1 Comparison of Abundance ValuesPetersen and Luxton (1982) present a number of abundance values for differentgroups of soil fauna in different environments. I have selected abundance values whichcome from similar environments in order to allow for a comparison. This narrowercomparison is complicated by differences in sampling and extraction methods,49extraction efficiencies, forest and soil types as well as the climate of the ecosystembeing studied.Within temperate coniferous forests, Petersen and Luxton (1982) suggested anematode abundance of 1 - 2.5 x 106 individuals m-2. The values from this study fallwithin this range except for the CH during July and October 1990 (6.9 x 105 m-2 and8.9 x 105 m-2, respectively).The values observed in this study for Copepoda abundance are above those valuesgiven for a Pseudotsuga plantation in Oregon (2 - 3 x 103 m-2) and are more in linewith the value given for the Canadian Arctic (23 x 103 m-2). It may be that the sitesampled in the Arctic had soil environmental conditions, especially moisture, whichwere similar to those found in the habitats of this study which provide a better habitatfor the existence of large populations of copepods.Collembola abundances for a Norwegian Spruce forest range from145 to 244 x 103 m-2. Values in this study are below this range except for the HAphase in August 1989 (250 x 103 m-2). When one accounts for the efficiency of theextraction method [41.3% based on core samples collected with compaction (Lussenhop1971)], values from August 1989 exceed this range, values from May and July 1990are within the range and October 1990 and March 1991 are still below the range.Abundances for Acari range from 274 x 103 m-2 in Finnish conifer forests to 206x 103 m-2 in Canadian Pseudotsuga plantations. Most abundances in this study arebelow this range except for the HA during August 1989 (468 x 103 m-2).Cryptostigmata are said to have a mean annual abundance ranging from100 to 300 x 103 m-2 in temperate coniferous forests. The CH maintains an abundancethat is lower than the given range (August 1989 being the only exception;131 x 103 m-2). Even when the abundance values are corrected by the efficiency value(93.1 % according to Lussenhop 1971), they are still below the range. The HA values50are within the abundance range even before the correction figure is applied with theexception of October 1990 (27 x 103 m-2 after efficiency correction).Density estimates for Prostigmata range from almost zero to 210 x 103 m-2 inboth woodland and non-forested ecosystems. This density covers the values obtained inthis study which range from 0.6 to 110 x 10 3 m-2 (uncorrected values: correctionfactor = 58.1 % from Lussenhop 1971).Astigmata are considered to have densities less than 1,000 m -2 . In this study,during May and July 1990, density estimates ranged from 2,200 m -2 in the CH to18,000 m-2 in the HA. However, during the rest of the sampling periods the densitywas 100 m-2 or zero. Krantz (1978) stated that among the Acari, populations of soilAstigmata tend to vary the most.Density estimates for the Oligochaeta, Araneae and Gastropoda were all wellunder those values given by Petersen and Luxton (1982). Generally, the CH maintainsabundance values that are lower the those given by Petersen and Luxton (1982) exceptfor the Copepoda which have a higher abundance in the CH than the HA. Thisdifference between abundances could be a result of the moisture levels being higher inthe CH and the more compact nature of the organic horizon. The HA may be moresimilar to those ecosystems for which Petersen and Luxton give abundance values.4.1.2 Comparison of Biomass ValuesPetersen and Luxton (1982) estimate the overall biomass in temperate coniferousforests to be 2.4 grams dry mass (g d.m.) m -2 . In this study, the CH maintains alower total biomass than the average except in August 1989 (3.45 g d.m. m -2)(Table II). Biomass in the CH ranges from 3.45 to 1.51 g d.m. m -2 . In the HA,biomass values range from 6.51 to 1.44 g d. m. m-2 , higher than average except duringthe October 1990 sample (1.44 g d.m. m -2) (Table V). The major assumption here is51that the biomass values calculated by Petersen and Luxton (1982) are based on the sameindividuals that are used to calculate the biomass values for this study (Table I).The nematode biomass is below average (120 mg d.m. m-2) on both phases at allsampling times except for August 1989 in the HA (141 mg d.m. m-2). This pattern issimilar in the Mesostigmata (average = 80 mg d.m. m-2), Prostigmata(average = 30 mg d.m. m-2) and Araneae (average = 50 mg d.m. m-2) while theOligochaeta maintain a biomass below the average (80 mg d.m. m-2) at all samplingtimes for both forest types. The Collembola (average = 80 mg d.m. m-2) maintain ahigher average biomass at all sampling times except during the October 1990 samplingin the CH (50.5 mg d.m. m-2). Diptera larvae (average = 260 mg d.m. m-2) andGastropoda (average -= 20 mg d.m. m-2) range above and below average with the HAvalues being larger than the CH for the Diptera and the opposite being true for theGastropoda. The Chilopoda biomass is above average (70 mg d.m. m-2) in both foresttypes at all sampling times except in the CH during the July 1990 sample (57 mg d.m.m-2) and March 1991 sample (57 mg d.m. m-2).The Diplopoda show the most interesting trend of being five to thirty timesgreater than the average value (50 mg d.m. m-2) given by Petersen and Luxton (1982).This could be the result of sampling "oddities". Hoffman (1989) suggested that theabundance of diplopods is correlated with presence of calcareous substrates. Furtherresearch to determine the level of calcium present in the habitats would be required tosupport this hypothesis.Again these comparisons are general, since differences in extraction methods andefficiencies vary with each study. Furthermore, habitats also vary in both aboveground and below ground characteristics and fluctuations can also occur as a result ofvariations at the microhabitat level which are rarely investigated. The groups looked athere also vary with respect to the number of different species which they contain.Danks (1988) states that about 33,000 species of terrestrial arthropods are known in52Canada and, possibly, another 33,000 species are still undescribed or unrecorded. Byconsidering only these large groups, we decrease the precision of the estimates becausevariations in abundance and biomass occur within each species as well as within eachgroup. Furthermore, variations in life cycles of different species results in differencesin abundance (e.g. natality and mortality) and biomass (e.g. egg, larval and adultstages).4.2 Similarity IndicesThere were no differences in similarity indices between the forest types duringeach sampling period with respect to the number of individuals counted in each sample,the relative abundance and biomass. Even the variation in the biomass calculations forthe Simplified Morisita's Index of Similarity (Table X) is the result of a single group,the Diplopoda, being present in great numbers in the CH and in low numbers in theHA in October 1990 (Table V). The abundance table for that sampling time (Table V)shows a high biomass of Diplopoda in the CH (1120 mg m -2) compared to the HA(280 mg m-2). This could be the result of a variation in food preference during thistime or, perhaps, represents a change in the life cycle of different species between thetwo forest types. Values calculated for March 1991 also suggests a lower similaritybetween the two phases (0.7519). This could be the result of the variation inCryptostigmata biomass between the CH (146.8 mg m - 2) and HA (492.37 mg m-2) orthe variation in Diptera larvae biomass (140 mg m -2 in the CH and 1190 mg m -2 in theHA). An important factor to keep in mind is the efficiency of the different extractionmethods being used. The use of a proper extraction method for organisms that arebeing studied is critical when comparing values. As well, extraction efficiency of theextractor being used also influences the results and must be considered wheninterpreting these values.53The indices suggest that the soil fauna communities are similar between bothforest types. However, the level of taxonomic identification is most likely the cause ofthis result. Most calculations for similarity indices are done using species levelidentification from the communities being examined (Krebs 1989). Species variationbetween the forest types may be more pronounced than class, order or family variation.Identification to species or genus could link specific organisms to either forest type.This could result in the calculation of similarity indices which could indicatedifferences between the forest types.4.3 Vertical DistributionThere are three zones in which soil fauna can be found; the epigeal (vegetation)zone, hemiedaphic (organic) zone or euedaphic (mineral soil) zone (Wallwork 1967).The general pattern associated with soil fauna is decreasing density with depth.Seasonal migration of Enchytraeidae, Collembola and Acari has been demonstrated byvarious authors (Springett et al. 1970, Usher 1970, 1971 and Marshall 1974). A largenumber of factors are responsible for the variations in vertical distribution betweenseasons and soil types. Price (1975) suggested that soil moisture has a greater influenceon vertical distribution patterns than soil temperature. Metz (1971) stated that soilmoisture content determined, to a large degree, the number of microarthropods presentin the soil. In the literature, examples of both positive and negative correlationsabound about soil fauna abundance and soil moisture. Usher (1970) stated that whilesoil moisture was related to vertical distribution, temperature was the most importantfactor in vertical distribution. Marshall (1974) also supports this idea. His resultssuggested that temperature had a greater influence on vertical migration than soilmoisture since no downward movement was observed in the fauna at the Shawnigansite during the dry summer.54While soil moisture may be an important factor influencing vertical distribution,other factors may also be important and perhaps the interaction of factors may be moreimportant than any single factor (Anderson 1977, Fjellberg 1985). Forest habitats,generally, maintain sufficiently high soil moisture contents throughout the year to meetthe needs of the soil fauna (Price 1975). Berthet and Gerard (1965) suggested that atnormal water content (25-90 %) of soil, there is no correlation between number ofAcari and water content of the soil sample. This probably holds true for mostmicroarthropods in the soil. The climate diagram and data in Appendix A suggests thatrelatively humid conditions persist throughout the year, even during the summermonths in the sampling area.Often what appears to be a single, uniform habitat to us with respect tovegetation, soil type, and various chemical and physical properties may, in fact, be agroup of unique, distinct microhabitats for soil arthropods. Bulk measurements such assoil moisture content or soil temperature may hide microhabitat variations which areimportant for soil fauna distribution. Gisin (1943 as cited by HAgvar 1983) defineddifferent "life forms" based on characteristic morphological changes in Collembolaspecies with increasing depth. For example, members of the family Tomoceridae havelong antennae, eye spots and a well-developed furca which are beneficial for speciesthat live in the litter layer. Onychiuridae have reduced antennae, eye spots are absentand the furca may be greatly reduced or absent. These characteristics are beneficial forspecies that live in the F or H horizons. Thus identification to species level isimportant to obtain a better understanding of vertical distribution. Furthermore, theidentification of species "lifeforms" could enhance information on the type of horizonin which the individuals are found, including certain soil chemical and physicalproperties.554.3.1 Mean DepthIn order to look at the spatial distribution of the soil population, we must assumethat all the individuals within the soil profile are uniformly distributed around thecentre of the soil core (Price 1975, Usher 1971). Analysis of mean depth values(Table IX) indicated that there was no significant difference in mean depth for themajority of the groups between the two forest types. Instead, significant differences inmean depth occurred as a result of differences in sampling time. This could indicate adifference in depth distribution as a result of seasonal changes (i.e. soil temperature,food availability, habitat, life cycles) within the forest types. Seastedt (1984b) studiedmicroarthropods in a tallgrass prairie ecosystem and found no significant changes inpopulations over time. Price (1975) stated that changes in vertical distribution couldalso be the result of a population decrease or migration into deeper soils. Murphy(1953) suggested that faunal distribution is dependent upon living space, oxygen, waterstatus and food supply. Hãgvar (1983) suggested that food was the limiting factor ofdepth distribution.Pseudoscorpionida was the only group to show a significant interaction betweenforest type and sampling time (p <0.05). This suggests some type of seasonaldifference between the forest types which may affect the habitat of the pseudoscorpionsor, perhaps, prey availability. However, another possibility is that this significantresult is a sampling phenomenon, more specifically insufficient sampling, not theresult of biological phenomena since abundance estimates of pseudoscorpions havelarge variances.Various authors have suggested that soil moisture content is an important factorwhich influences soil fauna distribution. The ANCOVA indicated that gravimetric soilmoisture content had no significant effect on mean depth. Lawrence (1986) found nosignificant correlation between soil fauna abundance and moisture content of each soilcore. Steinberger and Loboda (1991), studying nematode population dynamics in the56Negev Desert, found no correlation between nematode densities and precipitation, soilmoisture or root biomass. While soil moisture may be an important factor for soilfauna distribution, its interaction with some other factor such as gas diffusion or foodavailability may prove to be more meaningful. McKay et al. (1987) found soilmoisture to have little effect on Acari populations while the interaction between soilmoisture and temperature showed an obvious effect on both Acari and Collembolapopulations. Perhaps the level of taxonomy required to show the effect of soil moistureon faunal distribution needs to be substantially lower (i.e. to genus and species levels),since species may show a more restricted distribution in ecosystems based onenvironmental factors. Usher (1970) showed small variances in mean depth values forspecies of Collembola. However, combining all these species into one large group (i.efamily or class) increases the variation in the mean depth.Springett et al. (1970) found that vertical movement of Enchytraeidae was inresponse to changes in the Index of Humidity (IH). IH is not associated withtemperature or time of day, instead it fluctuates according to rainfall, windspeed andperhaps the occurrence of dew. IH is defined as the ratio of the mass of water to thedry mass of the soil sample. It is similar to calculating gravimetric moisture content,but this measurement is made over the span of a twenty-four hour period which reflectsthe diurnal changes in the soil water content. Hence, Springett et al.'s study may bedemonstrating diurnal migration in Enchytraediae. Andra et al. (1991) suggested thatsoil water potential is a more direct way of measuring the water availability thanvolumetric soil water content. Marshall (1974) converted his soil moisture readings topF values. The pF scale is similar to pH, but represents the logarithm of the negativepressure head in centimeters of water (Hillel 1980). Marshall's results indicated thatno downwards migration occurred during drier summer periods. Furthermore, themineral horizons maintained higher pF values than the organic layers, thus there wasno great advantage in migrating downward.57Most soil fauna are unable to actively burrow in the soil, therefore, they must useexisting pore spaces (Wallwork 1970). Hâgvar (1983) stated that only soil fauna of asmall body diameter could inhabit the soil below a certain depth since soil porositydecreases with depth. The HA maintains a more friable nature in its F horizons whilethe CH is more compact. Thus soil fauna may be able to penetrate more deeply intothe soil of the HA than the CH. In this study, microarthropods were found at depths of80 and 90 cm in the HA. Sampling in the CH only reached to 70 cm; usually nomicroarthropods were found at this depth. Nematodes were found in both forest typesat these depths. This vertical distribution suggests that the soil characteristics may limitthe distribution of microarthropods on the CH but not in the HA.4.3.2 Distribution through HorizonsThe distribution of individuals between the organic horizons (Figures 3 through8) indicates that most organisms are found in the upper horizons specifically the matcomplex and/or the F horizon. This supports the information given in other papers(Price 1973, Price 1975, Petersen and Luxton 1982). Kagvar (1983) stated that largepopulations of Collembola can be found where the most active decomposition oforganic material is occurring. He suggested that the F horizon is optimal for severalreasons: food is abundant, pore space is suitable and this layer is more resistant todesiccation. Anderson (1977) stated that maximum species diversity happens in the Fhorizon since the surface litter buffers diurnal temperature variation and the humidityremains relatively high.Yeates and Coleman (1982) related the distribution of nematodes in the soilprofile to the distribution of the organic matter produced in the ecosystem. Hdgvar(1983) suggested that the vertical distribution of Collembola species could indicate thestage of decomposition since, in podzolic soils, needles and above ground litter move58down the soil profile slowly during decomposition which could allow certain groups orspecies to be identified with certain levels of decomposition.The Nematoda are the exception as they tend to be found in greater abundance inthe H horizon. As mentioned previously, the soil structure may provide for a betterenvironment for the nematodes in this horizon. Perhaps the soil pores are too large inthe F horizon to allow for an adequate water film to be maintained around soilparticles. The H horizon may provide a preferred or more stable food source for thenematode population. Perhaps changes in soil moisture and temperature are minimizedfurther than in the F horizon thus making the H horizon an ideal habitat. Lawrence(1986) found a greater density of Nematoda higher in the soil profile (0-3 cm). This isto be expected because organic matter is restricted to the soil surface in recentlyreclaimed mine spoils.The distribution of nematodes in the soil profile is influenced not only by thedistribution of organic matter produced in the ecosystem, but by further interactionswith other soil organisms, seasons, depth and time of year (Yeates and Coleman 1982).Forests produce a variety of organic materials from the canopy, shrub and herb layersas well as root material in the soil itself. Both forest types have accumulated varyingamounts and types of organic matter on the forest floor. All these factors alsoinfluence the rest of the soil fauna community which makes a general survey like thisstudy that much more difficult to interpret.In this study, there is little change in the distribution of nematode populationsover time in either forest type (Figure 8). However, there are changes in theabundance of nematodes throughout the sampling period (Tables II through VI), whichindicate population peaks in May and August. These values suggest that variation indistribution occurs as a result of population changes. Wallwork (1970) stated thatpopulation fluctuations in nematodes do not appear to conform to any regular orcyclical pattern. Since nematodes require a water film around soil particles to move,59this could be interpreted as the moisture content of the soil being high enough for thepopulation to exist in the mat complex all year long. The fact that a constantpercentage of the population is found in the mat layer, while the greater abundance ofthe population is found in the H horizon, suggests that this distribution may be relatedto specific species or trophic groups occupying distinct horizons. Further identificationof nematodes to family, genus, species or, perhaps, trophic level could show avariation in distribution among these taxa or trophic groups. This could also increasethe variability of the abundance values of these groups if the number of samples beingtaken were not also increased.Marshall (1974) supports the hypothesis that critical levels in moisture may not bereached in forest ecosystems and, therefore, moisture content may not be the factorwhich directly causes migration or changes in distribution of faunal populations. Shiftsin vertical distribution observed in the other groups (Figures 3, 4, 5, 6, 7) could be theresult of partitioning by food, space, time or competition (Anderson 1977). However,moisture content could be the limiting factor for the presence and abundance of groupssuch as Copepoda, Branchiopoda and Isopoda. The CH appears to maintain a higheramount of water than the HA. Figure 9 shows the mat complex maintaining a highermoisture content in the CH than the HA. During our field trips, we observed standingwater in the CH on numerous occasions. Furthermore, pits dug by Dr. H. Takeda(Kyoto University, Japan) during my study often showed water at the bottom of the soilprofile of the CH but not in the HA.Price (1975) stated that soil fauna can extend to a depth equal to root penetrationin the soil. In this study fauna was found at depths of 80 and 90 cm in the HA. A fewroots were observed at these depths in both the CH and HA (Messier 1991).Furthermore, Price (1973) suggested that the presence of trees, logs, and woodymaterial in the soil modify the fauna that is present in the area. This is important dueto the presence of decaying woody debris found in the CH and sometimes found in the60HA. In the CH, the dominant decomposing wood product is cedar which takes longerto decompose than the hemlock or amabalis fir found in the HA (Harmon pt. al. 1986).The presence of phenolic acids in the cedar wood can retard the decomposition of thewood. Furthermore, wood debris has a large capacity for holding water. This couldinhibit decomposition by impeding gas diffusion into the wood which could slowmicrobial and faunal activity. As well, Lewis (1982) stated that the CH wasimperfectly drained. This could also have an impact on activity in the soil if it wasflooded for certain lengths of time. The greater time required to break down the cedarwood means nutrients are tied up longer in the CH phase resulting in a slower releaseof nutrients. J. Kethely (Division of Insects, Field Museum of Natural History,Chicago, Illinois, pers. comm.) suggested that a shift in organic matter input into anecosystem could also alter the vertical distribution of soil fauna. This could result fromevents such as windthrows, fires or logging activities which increase the amount ofcoarse woody debris entering the ecosystem. Depending upon the type of wood and thecondition it is in, soil fauna activity could be enhanced or impeded.Another trend is the apparent decrease in abundance during the summer monthsin the mat complex. This pattern is also apparent in the moisture content values shownin Figure 9. This decrease in moisture in this layer could result in a decrease in rootand fungal activity in this layer. Since a large number of soil organisms feed uponvarious fungi and bacteria, decreased activity would result in a lower food supplywhich would inhibit large populations.614.4 Limitations of the Research4.4.1 SamplingThe first critical observation to make is the large standard error values in thebrackets beside the mean abundances. This shows the high variability between sampleswhich is inherent with most soil communities (McSorley and Walter 1991). Thedistribution of soil animals is considered to be more or less aggregated (Huhta et al.1967). Abrahamson and Strand (1970) stated that most soil fauna fitted the NegativeBinomial Distribution (e.g. Collembola and Cryptostigmata) while the distribution ofrare species in the soil fitted the Poisson distribution. Increasing the sample size (i.e.the number of samples taken) is considered to be the way to increase precision instatistics (i.e. decrease the confidence interval for the sample means) (Krebs 1989).Price (1973) excluded trees, logs and other features from his sampling plots that hethought would have a modifying effect on the soil fauna. By doing this, you attempt todecrease the variation caused by these surface features. Variation is also present withinthe soil. Due to limited knowledge on the biology of most soil fauna, the exclusion ofthese features makes numerical analysis easier but results in a loss of information on thepopulation being studied and how it is affected by its environment.Furthermore, the amount of time required to process, count and identify theorganisms in each sample prevents one person from being overly ambitious. Marshall(1974) stated that by taking six samples from the Shawnigan study site you would needa 2.27 fold difference between the populations to establish the fact that the populationsare significantly different numerically. By increasing the number of samples to 10, youwould still need a 1.83 fold difference between the populations to prove they weresignificantly different. In this study, we would have to take 52 samples in the CH and83 samples in the HA to obtain + 10 % precision in estimating the total soil faunapopulation throughout the length of the study. With regards to the Acari we wouldrequire a minimum of 328 samples in the HA to achieve + 10 % precision.62Considering only the counts from August 1989, a minimum of 80 samples from the HAwould be required to achieve ± 10 % precision. Calculations indicate that by takingfour samples, as in this study, I achieved a ± 50 % precision in estimating the totalsoil fauna population for both the CH and HA. Robson and Regier (1964) state thatthis level of accuracy (± 50 %) is adequate for preliminary surveys. Needless to say,unless you either have much time or lots of people (or, preferably, both), it would takea long time to sort through this many samples.Previous studies have established that soil arthropods maintain a greaterabundance in the surface soil layers, population size varies with season and abundancesdecrease with depth. Brand (1979) suggested that the time of day may also beimportant when sampling soil fauna populations. He showed that the highest numberof Acari were found in surface samples at 14:00 while deeper soil samples showed nodifference between 14:00 and 22:00. Collembola numbers increased from 06:00 to14:00 to 22:00 in surface samples while numbers were greater at 06:00 or 14:00 than at22:00 for deeper soil samples. Usher (1970) suggested that diurnal variation could beeliminated by sampling at a standardized time of day. However, this creates a problemwhen sampling numerous sites. We would have required eight days for samplingfollowing Usher's suggestion. This would have required samples to be stored for morethan a week in a cooler before extraction. This would influence the extractionefficiency and result in a lower abundance estimate for the faunal populations.Distribution of soil fauna is usually assumed to vary vertically and decrease withdepth. However, Schenker (1984) observed different spatial and seasonal distributionalpatterns for a population of Cryptostigmata. He showed that the highly aggregateddistribution patterns of this population were centered around tree trunks. Furthermore,seasonal changes in distribution patterns were small, but suggested that Cryptostigmatamoved towards the tree base during the winter instead of deep into the soil. Anotherfactor which may influence the effect of temperature on vertical distribution is the63thickness of the organic horizons. Schaefer and Schauermann (1990) stated that a thickorganic layer can act as an insulating layer during the winter. The presence of aninsulating layer could decrease the need for soil fauna to move deeper into the soilprofile during the winter. This insulating layer may also decrease dessication withinthe organic horizons. At the extreme, a thicker organic layer could retain enoughwater which could inhibit faunal activity and, possibly, decomposition. These effectswould be more noticeable in the CH since its organic layer is much thicker than the HAand the mat complex in the CH appears to maintain a higher moisture content than theHA (Figure 9).Compaction of the soil cores during sampling resulted in inaccurate depthmeasurements. Therefore, in order to look at vertical distribution, identification of thedifferent organic horizons in the soil profile, which may affect the distribution of soilfauna, may be better for examining vertical distribution. By placing the samples inmonthly order, I examined the vertical distribution in the soil horizons and observed apattern which suggests changes in populations according to seasons (Figures 3 through8). This pattern indicates that monthly sampling is vital if seasonal trends in populationchanges are to be observed.4.4.2 ExtractionNo single extraction method can remove 100 % of all, or any, group or speciesfrom a soil sample (Edwards 1991, McSorley and Walter 1991). In order to examinethe overall population of soil invertebrates, several methods are usually required. Inthis study four methods were used to evaluate the soil fauna communities. Hand-sorting is limited by the accuracy and efficiency of the person doing the sorting. Eventhe same person will obtain varying results from day to day and sample to sample.The Formalin method is limited by the strength of the solution and the ability ofthe soil to absorb it. If the solution is weak, not all of the individuals will rise to the64surface. If the solution is too strong, it could kill the organisms before they have achance to get to the surface. The nature of the soil also plays an important role. If it ishydrophobic, the solution will not move into the soil thus the estimate will not includeindividuals in unwetted areas or from deep in the soil profile. In this study I graduallyincreased the strength of the formalin solution from the beginning. However it may bereasonable to increase the concentration further still. Raw (1959) used a 0.22 %formalin solution for sampling earthworm populations in agricultural soils. Due to thelarge amount of organic material on these forest sites however, it may be advisable touse that amount or more to achieve an efficient level of extraction. Perhaps anotherway is to increase the number of washes with formalin, while keeping the concentrationlow. However, this is limited by the amount of water you are able to carry into yourstudy site.The efficiency of the modified Baermann funnel extractor used for the semi-aquatic soil fauna depends upon a variety of factors including recovery temperature,soil type, soil depth in the sample holder, seasonal activity and previous storage of thesample. Furthermore, the method is not equally efficient for recovering all life stagesand inactive fauna. A 24 to 48 hour incubation period can result in egg hatching andemergence from cryptobiotic state which alters the population structure from the timeof the collection (McSorley and Walter 1991). Marshall (1974) stated that theextraction efficiency for the Baermann funnel method ranges from 65 to 70 % fornematodes and from 41 to 54 % for enchytraeids. Proctor and Marks (1974) suggestedthat by subsampling from a 20 core sample mixed together you could obtain a moreprecise estimate of nematode population densities than with a single core sample.The high gradient extractor is the most economical method when extractingmicroarthropods from organic soil layers (McSorley and Walter 1991). This method,however, is unable to extract inactive stages thus producing a slightly biased populationstructure. Lussenhop (1971) found no uniform extraction efficiency for all groups.65Collembola and Cryptostigmata responded differently to coring disturbances whileAstigmata, Prostigmata and Collembola extraction efficiency varied with depth.Furthermore, compaction as a result of coring reduced the number of Collembolaextracted from the top three cm of soil (41.3 % efficiency). Lussenhop found that bycollecting pyramid samples no compaction of the sample was assured. This wasachieved by exposing a column of earth and removing sections in core liners. Thisincreased Collembola extraction efficiency to 72.5 % but had little effect onProstigmata and Astigmata recovery.4.4.3 AnalysisThe main difficulty with the analysis is large variances associated with the meansof the groups. As explained earlier, this is characteristic of the soil fauna community.Reduction of the variance of the mean could occur through more samples and replicatesbeing taken, but this would increase the workload substantially.The level of taxonomic identification is another problem limiting the results.Identification to group combines a variety of species which exhibit differentcharacteristics, distributions and populations. If the level of taxonomy could beadvanced to genus and/or species level, species may exhibit a more confined rangewhich could show differences between the two forest types more readily. Hágvar(1983) stated that vertical separation between collembolan species may be morepronounced during maximum population density in an ecosystem. Usher (1970) lookedat the distribution of collembolan species in the top three cm and found smallervariances in mean depth. However, by pooling all these species together, the variationin vertical distribution (L e. mean depth) is larger because of the differences betweenthe species themselves.Calculating mean depth by Usher's method (1970) is complicated when soilsamples are compacted during sampling. Furthermore, the appearance of a few66individuals at great depth causes a noticeable downward shift of mean depth values(Price 1975). While soil depth may have some effect on faunal distribution, meandepth does not provide an adequate description of the habitat in which the fauna arefound. Furthermore, the variation encountered within the soil profile does not lenditself to using depth values to describe where individuals are found. Identification ofthe organic and mineral horizons where the organisms live makes more sense. Depthmeasurements would still be important to allow for mapping of the soil profile in thesampling area and, perhaps, be used to determine where, in a particular horizon,organisms are found. By relating certain organism to certain horizons, furtherinformation on habitat and biological activity could be discerned.Parameters other than soil moisture content must also be examined. While highertaxonomic identification might show some interaction with soil moisture, another factormight show a better interaction with faunal distribution. Although moisture contentaffects other ecological processes, it may not be the dominant factor in forestecosystems. This is not to say that it should not be measured but it should be done inaddition to another variable such as soil temperature. These measurements should becarried out in a manner that would reflect variations at the microhabitat level. Foodsources could be considered since they vary with species in habitats and time.Anderson (1977) stated that a more complex soil structure can allow more species toco-exist in the same habitat. However, the examination of bulk measurements of soil(e.g. soil moisture and temperature) could cloud relationships among organisms andtheir environments since most soil fauna are distributed in relation to microhabitats inthe soil profile.675.0 CONCLUSIONSOverall, numerical abundance and biomass values suggest that the HA supports alarger population of soil fauna than the CH. However, both forest types exhibit similardistribution patterns with respect to the abundance and biomass of various groups.Nematoda are the numerically dominant group in both forest types followed by theAcari, Collembola and Copepoda. The Diplopoda, however, maintain the greatestbiomass in both forest types. Enchytraeidae, Diptera larvae and Acari also account fora substantial portion to the biomass in the CH and HA.Similarity indices suggest that the soil fauna communities are similar with respectto numbers and relative abundance at the present level of taxonomy. However,biomass values reflect a slightly more varied distribution between the two forest types,and suggest the possibility of a difference between the two forest types during October1990 and, perhaps, in March 1991. Identification to genus or species could help toidentify the differences between the two forest types.The ANOVA performed on mean depth calculations suggests that the mean depthof soil fauna is not significantly different between the forest types. A significantdifference in mean depth between sampling times suggests some type of seasonal effecton vertical distribution in both forest types. Examining the distribution of severalgroups through sampled horizons supports this seasonal change in vertical distribution.Results indicated a decrease in the percent abundance of microarthropods in the upperorganic horizons during the summer with greater abundances in these horizons duringthe spring and fall for the CH and HA. Nematoda maintained a steady abundance inthe upper organic horizons through the sampling period for both forest types whichsuggests that soil moisture does not have a direct effect on mean depth. This wassupported by ANCOVA results that indicated soil moisture did not have a significanteffect on mean depth.68"As a general rule, the most successful faunal surveys are those devoted to aspecific organism..." (Murphy 1953). Following this "general rule", future researchwill require a number of people to work on separate groups to identify individuals tothe species or, at least, the genus level. A lack of knowledge of the biology of mostspecies of soil fauna also adds to the difficulties in interpreting information. Byspecializing on a specific group, sampling could be more intense, reducing variabilityand allowing for at least monthly sampling to observe trends in seasonal distribution ofthe populations, life cycles, feeding behaviour and other important biological factors.This could provide a better understanding of the soil fauna communities present withinthe Cedar-Hemlock and Hemlock-Amabalis fir forest types and the important role theyplay in these ecosystems.Soil fauna are not a dominant component of soil respiration (<10%) but they areimportant in modifying the soil, especially the organic matter for furtherdecomposition. Identification of the organic layers from which the organisms arecollected as well as examining the physical and chemical properties of these horizonscould provide more useful information than only measuring the depth of the sample.Micromorphological studies of the soil types could also provide valuable informationon the structure of microhabitats and possibly which organisms frequent these soiltypes. This would allow us to identify soil fauna associated with humus forms andfurther help to establish the role soil fauna have in these forest ecosystems.Microcosm studies could be beneficial in determining feeding preferences ofgroups or species, life cycles, identification of larval stages as well as speciesinteractions providing the microcosm community was complex enough. Furthermore,trials could be performed to study species effect on wood decomposition and nutrientcycling. This could indicate the role of soil fauna in nutrient retention within the CHand HA forest types and could provide further information on their importance withinthe soil ecosystem. The improvement of present silvicultural practices and increased69environmental protection would help secure prolonged benefits from soil fauna activityand maintain a healthy component of the environment.70LITERATURE CITEDAbrahamson, G. and Strand, L. 1970. Statistical analysis of population density data ofsoil animals with particular reference to Enchytraeidae (Oligochaeta). Oikos,21:276-284.Addison, J.A. 1977. Population dynamics and biology of Collembola on TrueloveLowland, Devon Island, N.W.T. In: Bliss, L.C. (ed.), Truelove Lowland,Devon Island, Canada: A High Arctic ecosystem. Univ. of Alberta Press,Edmonton, pp. 363-382.Anderson, J.M 1977. The organization of soil animal communities. In: Lohm, V. andPersson, T. (eds.), Soil organisms as components of ecosystems, Ecol. Bull.(Stockholm), 25:15-23.Anderson, J.M. 1988. Spatiotemporal effects of invertebrates on soil processes. Biol.Fert. Soils, 6:216-227.Andren, 0., Rajkai, K. and Katerer, T. 1991. A non-destructive technique for studiesof root distribution in relation to soil moisture. In: Crossley, D.A., Jr., Coleman,D.C., Hendrix, P.F., Cheng, W., Wright, D.H., Beare, M.H. and Edwards,C.A. (eds.), Modern techniques in soil ecology. Elsevier, New York, pp 269-278.Berthet, P. and Gerard, G. 1965. A statistical study of microdistribution of Oribatei(Acari). Part I. The distribution pattern. Oikos, 16:214-227.Brand, R.H. 1979. Effect of sampling time on estimates of soil Arthropod populations.Rev. Ecol. Biol. Sol., 16:499-515.Carter, A. and Cragg, J.B. 1976. Concentrations and standing crops of calcium,magnesium, potassium, and sodium in soil and litter arthropods and their food inan aspen woodland ecosystem in the Rocky Mountains (Canada). Pedobiologia,16:379-388.Coleman, D.C., Anderson, R.V., Cole, C.V., Elliott, E.T., Woods, L. and Campion,M.K. 1978. Trophic interactions in soils as they affect energy and nutrientdynamics. IV. Flows of metabolic and biomass carbon. Microbial Ecology,4:373-380.Danks, H.V. (ed.) 1979. Canada and its insect fauna. Memoirs of the EntomologicalSociety of Canada, No. 108.71Danks, H.V. 1988. Insects of Canada: A synopsis prepared for delegates to theXVIIIth International Congress of Entomology (Vancouver, 1988). BiologicalSurvey of Canada Document (Terrestrial Arthropods), series no. 1.Dash, M.C. and Cragg, J.B. 1972. Ecology of Enchytraeidae (Oligochaeta) inCanadian Rocky Mountain soils. Pedobiologia, 12:323-335.Edwards, C.A. 1991. The assessment of populations of soil-inhabiting invertebrates.In: Crossley, D.A., Jr., Coleman, D.C., Hendrix, P.F., Cheng, W., Wright,D.H., Beare, M.H. and Edwards, C.A. (eds.), Modern techniques in soilecology. Elsevier, New York, pp 145-176.Fjellberg, A. 1985. Recent advances and future needs in the study of Collembolabiology and systematics. Quaestiones Entomologicae, 21:559-570.Fox, C.A., Trowbridge, R. and Tarnocai, C. 1987. Classification, macromorphologyand chemical characteristics of Folisols from British Columbia. Can. J. Soil Sci.67:765-778.Germain, A.Y. 1985. Fertilization of stagnated Sitka Spruce plantations on NorthernVancouver Island. M.F. Thesis, Univ. of British Columbia.Gisin, H. 1943. Okologie und Lebensgemeinschaften der Collembolen imSchweizerischen Exkursionsgebiet Basels. Rev. Suiss.Zool., 50:131-224.Green, R.N., Courtin, P.J., Klinka, K., Slaco, R.J. and Ray, C.A. 1984. Sitediagnosis, tree species selection and slashburning guidelines for the VancouverForest Region. Land Manag. Handbk. #8. Ministry of Forest, Victoria, B.C.,Canada.Hägvar, S. 1983. Collembola in Norwegian coniferous forest soils II. Verticaldistribution. Pedobiologia, 25:383-401.Harmon, M.E., Franklin, J.F., Swanson, F.J., Sollins, P., Gregory, S.V., Lattin,J.D., Anderson, N.H., Cline, S.P., Aumen, N.G., Sedell, J.R., Lienkaemper,G.W., Cromack, K. Jr. and Cummins, K.W. 1986. Ecology of coarse woodydebris in temperate ecosystems. Advances in Ecological Research, 15:133-302.Hillel, D. 1980. Fundamentals of soil physics. Academic Press, Toronto.Hoffman, R.F. 1989. Diplopoda. In: Dindal, D.L. (ed.), Soil Biology Guide. JohnWiley & Sons, Toronto, pp. 835-860.72Huhta, V., Karppinen, E., Nurminen, M. and Valpas, A. 1967. Effect of silviculturalpractices upon arthropod, annelid, and nematode populations in coniferous forestsoil. Ann. Sool. Fenn., 4:87-145.Kitazawa, Y. 1967. Community metabolism of soil invertebrates in forest ecosystemsof Japan. In: Petrusewicz, J. (ed.), Secondary productivity of terrestrialecosystems. Polish Acad. Sci., Warsaw, pp. 649-661.Klinka, K., Nuszdorfer, F.C. and Skoda, L. 1979. Biogeoclimatic units of central andsouthern Vancouver Island. Province of British Columbia, Ministry of Forests.Klinka, K., Green, R.N., Trowbridge, R.L. and Lowe, L.E. 1981. Taxonomicclassification of humus forms in ecosystems of British Columbia. Ministry ofForests, Victoria, B.C., Canada.Klinka, K., Pojar, J. and Meidinger, D.V. 1991. Revision of biogeoclimatic units ofcoastal British Columbia. Northwest Science, 65:32-47.Krantz, G.W. 1978. A manual of acarology, 2nd ed. Oregon State University,Corvalis.Krebs, C.J. 1978. Ecology: The experimental analysis of distribution and abundance,2nd ed. Harper & Row, Publishers, New York.Krebs, C.J. 1989. Ecological Methodology. Harper & Row, Publishers, New York.Krivolutsky, D.A. and Pokarzhevsky, A.D. 1977. The role of soil animals in nutrientcycling in forest and steppe. Ecol. Bull. (Stockholm), 25:253-260.Lawrence, J.M. 1986. Soil fauna colonization of high elevation coal mine spoils in theCanadian Rockies. M.Sc. Thesis, University of Victoria.Lewis, T. 1982. Ecosystems of the Port McNeill block 4 of Tree Farm Licence 25.Int. report. Western Forest Products Ltd., Vancouver, B.C., Canada.Lussenhop, J. 1971. A simplified canister-type soil arthropod extractor. Pedobiologia,11:40-45.Marshall, V.G. 1974. Seasonal and vertical distribution of soil fauna in a thinned andurea -fertilized Douglas-fir forest. Can. J. Soil Sci., 54:491-500.McBrayer, J.F. 1977. Contributions of Cryptozoa to forest nutrient cycles. In:Mattson, J. (ed.), The role of arthropods in forest ecosystems. Springer-Verlag,New York.73McKay, W.P., Silva, S. and Whitford,W.G. 1987. Diurnal activity patterns andvertical migration in desert soil microarthropods. Pedobiologia, 30:65-71.McSorley, R. and Walter, D.E. 1991. Comparison of soil extraction methods fornematodes and microarthropods. In: Crossley, D.A., Jr., Coleman, D.C.,Hendrix, P.F., Cheng, W., Wright, D.H., Beare, M.H. and Edwards, C.A.(eds.), Modern techniques in soil ecology. Elsevier, New York, pp. 201-207.Messier, C. 1991. Factors limiting early conifer growth in salal-dominated cutovers onNorthern Vancouver Island, British Columbia. Ph.D. Thesis, University ofBritish Columbia.Metz, L.J. 1971. Vertical movement of Acarina under moisture gradients.Peclobiologia, 11:262-268.Mitchell, M.J. 1977. Population dynamics of oribatid mites (Acari, Cryptostigmata) inan aspen woodland soil. Pedobiologia, 17:305-319.Morisita, M. 1959. Measuring of introspective association and similarity betweencommunities. Mem. Fac. Sci. Kyushu Univ. Ser. E (Biol.) 3:65-80.Murphy, P.W. 1953. The biology of forest soils with special reference to themesofauna or meiofauna. J. Soil Sci., 4:155-193.Nuszdorfer, F.C., Kassay, K.L. and Scagel, A.M. 1985. Biogeoclimatic units of theVancouver Forest Region. B.C. Min. For., Queen's Printer, British Columbia.O'Connor, F.B. 1962. The extraction of Enchytraeidae from soil. In: Murphy, P.W.(ed.), Progress in soil zoology. Butterworths, London, pp. 279-285.Paoletti, M.G., Favretto, M.R., Stinner, B.R., Purrington, F.F and Bater, J.E. 1991.Invertebrates as bioindicators of soil use. In: Crossley, D.A., Jr., Coleman,D.C., Hendrix, P.F., Cheng, W., Wright, D.H., Beare, M.H. and Edwards,C.A. (eds.), Modern techniques in soil ecology. Elsevier, New York, pp. 341-362.Persson, T. 1980. Trophic structure, biomass dynamics and carbon metabolism of soilorganisms in a Scots pine forest. Ecol. Bull. (Stockholm), 32:419-459.Petersen, H. and Luxton, M. 1982. A comparative analysis of soil fauna populationsand their role in decomposition processes. Oikos, 39:287-388.Price, D.W. 1973. Abundance and vertical distribution of microarthropods in thesurface layers of a California pine forest soil. Hilgardia, 42:121-148.74Price, D.W. 1975. Vertical distribution of small arthropods in a California pine forestsoil. Ann. Entoml. Soc. Amer., 68:174-180.Proctor, D.L.C. 1977a. Nematode densities and production on Truelove Lowland. In:Bliss, L.C. (ed.), Truelove Lowland, Devon Island, Canada: A High Arcticecosystem. Univ. of Alberta Press, Edmonton, pp. 347-361.Proctor, D.L.C. 1977b. Invertebrate respiration on Truelove Lowland. In: Bliss, L.C.(ed.), Truelove Lowland, Devon Island, Canada: A High Arctic ecosystem.Univ. of Alberta Press, Edmonton, pp. 383-393.Proctor, J.R. and Marks, C.F. 1974. The determination of normalizing transformationsfor nematode count data from soil samples and of efficient sampling schemes.Nematologica, 20:395-406.Raw, F., 1959. Estimating earthworm populations by using formalin. Nature(London), 184 (Suppl. 21):1662.Robson, D.S. and Regier, H.A. 1964. Sample size in Petersen mark-recaptureexperiments. Trans. Amer. Fish. Soci. 93:215-226.Ryan, J.K. 1977. Synthesis of energy flows and population dynamics of TrueloveLowland invertebrates. In: Bliss, L.C. (ed.), Truelove Lowland, Devon Island,Canada: A High Arctic ecosystem. Univ. of Alberta Press, Edmonton, pp. 325-346.Schaefer, M. and Schauermann, J. 1990. The soil fauna of beech forests: Comparisonbetween a mull and a moder soil. Pedobiologia, 34:299-314.Schenker, R. 1984. Spatial and seasonal distribution patterns of oribatid mites(Acari:Oribatei) in a forest soil ecosystem. Pedobiologia, 27:133-149.Seastedt, T.R. 1984a. The role of microarthropods in decomposition and mineralizationprocesses. Ann. Rev. Entomol., 29:25-46.Seastedt, T.R. 1984b. Microarthropods of burned and unburned tallgrass prairie. J.Kansas Entomol. Soc., 57:468-476.Seastedt, T.R. and Crossley, D.A., Jr. 1981. Microarthropod response followinglogging and clear-cutting in the southern Appalachians. Ecology, 62:126-135.Smith, C.A.S., Tomlin, A.D., Miller, J.J., Moore, L.V., Tynen, M.J. and Coates,K.A. 1990. Large enchytraeid (Annelida:Oligochaeta) worms and associatedfauna from unglaciated soils of the northern Yukon, Canada. Geoderma, 47:17-32.75Springett, J.A., Britain, J.E. and Springett, B.P. 1970. Vertical movement ofEnchytraeidae (Oligochaeta) in moorland soils. Oikos, 21:16-21.Steinberger, Y. and Loboda, I. 1991. Nematode population dynamics and trophicstructure in a soil profile under the canopy of the desert shrub Zygophyllumdumosum. Pedobiologia, 35:191-197.Usher, M.B. 1970. Seasonal and vertical distribution of a population of soilarthropods: Collembola. Pedobiologia, 10:224-236.Usher, M.B. 1971. Seasonal and vertical distribution of a population of soilarthropods: Mesostigmata. Pedobiologia, 11:27-39.Wallwork, J.A. 1967. Acari. In: Burges, A. and Raw, F. (eds.), Soil Biology.Academic Press, New York, pp. 363-395.Wallwork, J.A. 1970. Ecology of soil animals. McGraw-Hill, London.Wallwork, J.A. 1976. The distribution and diversity of soil fauna. Academic Press,London.Walter, H. 1979. Vegetation of the Earth and ecological systems of the geo-bioshpere.2nd ed. Springer-Verlag, New York, pp. 25-28.Wilkinson, L. 1988. SYSTAT: The system for statistics. SYSTAT, Inc., Evanston, IL.Willard, J.R. 1972. Soil invertebrates. I. Methods of sampling and extraction. MatadorProject. Techn. Rep. 7, Saskatoon, Sask.Willard, J.R. 1973a. Soil invertebrates. III. Collembola and minor insects: Populationand biomass. Matador Project. Techn. Rep. 19, Saskatoon, Sask.Willard, J.R. 1973b. Soil invertebrates. IV. Acarina, minor Arachnida and Chilopoda:Populations and biomass. Matador Project. Techn. Rep. 20, Saskatoon, Sask.Willard, J.R. 1973c. Soil invertebrates. V. Nematoda: Populations and biomass.Matador Project. Techn. Rep. 21, Saskatoon, Sask.Willard, J.R. 1974a. Soil invertebrates. VII. Enchytraeidae (Annelida: Oligochaeta):Populations and biomass. Matador Project. Techn. Rep. 28, Saskatoon, Sask.Willard, J.R. 1974b. Soil invertebrates. VIII. A summary of populations and biomass.Matador Project. Techn. Rep. 56, Saskatoon, Sask.76Willard, J.R., Fisher, V. and Petrovick, M. 1973. Soil invertebrates: VI. Nematodes:trophic classification and structure of the population. Matador Project. Techn.Rep. 26, Saskatoon, Sask.Wolda, H. 1981. Similarity indices, sample size and diversity. Oecologia, 50:296-302.Yeates, G.W. and Coleman, D.C. 1982. Role of nematodes in decomposition. In:Freckman, D.W. (ed.), Nematodes in soil ecosystems. University of Texas Press,Austin. pp. 55-80.77Appendix AE uations for Similarity Indicesi) Mean depth (Usher 1970)M =  E dini N where k= number of layers of samplesdi= centre depths of sample layers (cm)ni= number of individuals in that layerN= total number of individuals in the coreii) Morisita's Index of Similarity (Morisita 1959)MIS = 2 E nii nki where X,. = E n.• (n.• - 1)J ^J1^Ji Ni (Ni - 1) ( Xj ± XI) NjNkwhere MIS = index of similarity between samples j and kn.•ii, nki = number of individuals of species i in samples j and kNJ.=^JiE n••= total number of individuals in sample jNk= E nki= total number of individuals in sample kiii) Simplified Morisita's Index of Similarity (Wolda 1981, Krebs 1989)SMIS calculated as MIS above except forA.. = I n2.•J^Ji N2.Jiv) Renkonen Index (Wolda 1981, Krebs 1989)RI = E minimum (Pii, Pki)where RI = percentage similarity between samples j and kPii= percentage of species i in community sample jPki= percentage of species i in community sample k78Appendix BGravimetric soil moisture contents (%) for the sampled horizonsMar 1991 May 1990 Jul 1990 Aug 1989 Oct 1990CH HA CH^HA CH HA CH^HA CH HAmat 424 313 256 175 138 176 189 142 330 227F NA NA NA NA 320 275 295 163 418 319H 351 285 NA NA 343 192 233 224 351 236Mineral 147 126 NA NA 78 75 132 92 90 71NA - not available79Appendix CMean depth (cm) ± 1 standard deviation of soil fauna for all sampling periods.August 1989 CHMean s.d.HAMean s.d.Astigmata 0.38 0.75Cryptostigmata 4.24 0.27 7.75 3.59Mesostigmata 3.77 1.73 6.46 3.28Prostigmata 6.19 1.71 6.29 1.10Undetermined 1.11 2.23total Acari 3.06 2.49 4.18 3.69Aranaea 16.13 28.31 6.00 11.02Pseudoscorpionida 4.25 2.50 10.27 6.50Copepoda 10.36 3.47 6.22 4.43Pauropoda 9.13 6.63Diplopoda 1.13 2.25Chilopoda 1.13 2.25 4.13 3.09Symphyla 2.78 2.43 6.79 3.97Protura 2.25 3.57 7.61 10.61Anuridae 1.50 2.12 21.46 37.06Entomobryidae 5.53 1.66 6.55 3.13Hypogasturidae 5.21 6.98 3.01 1.02Isotomidae 4.70 3.14 3.87 2.89Neanuridae 3.00 2.74 7.54 4.70Onychiuridae 6.15 1.67 10.10 3.91Tomoceridae 4.83 0.87 6.38 9.75Neelidae 9.25 13.03 16.21 11.86Sminthuridae 2.30 1.91 3.98 4.72Undetermined 7.46 6.23 8.67 1.43Total Collembola 4.99 2.34 8.77 5.87Diplura 0.38 0.75Coleoptera (L) 0.38 0.75 3.19 3.99Coleoptera (A)Diptera (L) 0.75 0.87 12.94 16.61Diptera (A) 1.13 2.25Hemiptera 1.88 3.75Thysanura 4.38 8.75Nematoda 10.92 5.00 13.12 3.74Enchytraeidae 10.53 5.63 12.19 4.60Rotifera 7.02 4.45 10.15 5.03Tardigrada 7.12 5.92 9.59 4.01TOTAL 4.54 3.73 6.70 4.9380May 1990CHMeanAppendix C (continued)HAs.d.^Mean s.d.Astigmata 4.95 5.90 2.87 0.30Cryptostigmata 3.77 1.53 4.14 1.73Mesostigmata 3.10 1.33 3.66 1.34Prostigmata 3.85 1.78 4.31 2.03Undetermined 1.88 3.75 3.19 5.42total Acari 3.51 1.13 3.63 0.61Aranaea 0.38 0.75 5.63 11.25Pseudoscorpionida 1.13 1.44 6.75 7.09Copepoda 4.65 0.72 6.29 3.53Pauropoda 3.23 4.27 9.50 5.48Diplopoda 0.38 0.75 0.75 0.87Chilopoda 2.63 2.56 0.94 1.13Symphyla 3.00 3.24 7.13 6.97Protura 0.38 0.75 0.38 0.75Anuridae 1.31 0.94 7.63 5.11Entomobryidae 1.88 3.75 5.30 7.55Hypogasturidae 2.99 1.79 3.73 3.27Isotomidae 2.69 0.81 4.52 2.62Neanuridae 0.38 0.75Onychiuridae 6.09 2.37 8.06 4.72Tomoceridae 2.92 1.40 1.69 1.28Neelidae 2.34 0.75 5.46 5.27Sminthuridae 5.25 4.50 0.38 0.75Undetermined 3.93 2.49 4.05 4.68Total Collembola 2.98 1.73 4.08 2.75Diplura 0.38 0.75 0.38 0.75Coleoptera (L) 11.63 23.25Coleoptera (A)Diptera (L) 4.50 5.34 1.88 3.75Diptera (A)HemipteraThysanuraNematoda 10.14 1.64 8.91 2.90Enchytraeidae 6.77 3.70 7.53 3.84Rotifera 4.35 4.25 5.79 3.37Tardigrada 3.71 2.77 3.88 4.52TOTAL 3.47 2.57 4.14 2.74July 1990CHMeanAppendix C (continued)HAs.d.^Mean s.d.Astigmata 0.88 1.44 3.00 6.00Cryptostigmata 3.50 0.95 5.37 3.46Mesostigmata 2.85 2.03 5.31 1.92Prostigmata 3.69 1.14 5.66 2.55Undetermined 2.40 2.76 4.29 3.51total Acari 2.66 1.12 4.73 1.10AranaeaPseudoscorpionida 2.00 2.31 4.75 4.86Copepoda 6.73 1.84 7.00 2.19PauropodaDiplopoda 0.38 0.75 0.25 0.50Chilopoda 0.75 1.50Symphyla 2.88 1.93 5.82 5.41ProturaAnuridae 1.63 2.14 1.65 1.41Entomobryidae 0.50 1.00 2.19 3.31Hypogasturidae 3.46 2.06 3.08 2.10Isotomidae 2.48 1.11 2.74 1.43Neanuridae 1.83 2.50 2.42 3.62Onychiuridae 4.09 2.09 7.43 3.79Tomoceridae 1.40 1.59 0.13 0.25Neelidae 3.88 5.20 2.63 3.09Sminthuridae 0.13 0.25 4.13 8.25Undetermined 2.27 1.49 3.75 1.31Total Collembola 2.17 1.35 3.01 1.91Diplura 0.13 0.25Coleoptera (L)Coleoptera (A) 6.44 11.74Diptera (L) 2.25 3.30Diptera (A)HemipteraThysanuraNematoda 6.62 2.34 9.11 2.02Enchytraeidae 6.47 2.83 9.32 4.92Rotifera 6.46 2.34 4.54 3.16Tardigrada 5.49 5.73 5.10 3.82TOTAL 2.56 2.14 4.03 2.46October 1990AstigmataAppendix C (continued)CH^HAMean^s.d.^Mean s.d.Cryptostigmata 3.54 2.04 2.57 1.05Mesostigmata 4.99 5.90 4.58 6.57Prostigmata 0.85 1.70 3.15 2.96Undetermined 1.00 1.15 1.47 2.93total Acari 2.07 2.10 2.35 1.73Aranaea 1.00 1.15 0.50 1.00Pseudoscorpionida 7.25 14.50 1.00 1.15Copepoda 4.59 1.37 4.51 4.85PauropodaDiplopoda 1.00 1.15Chilopoda 0.50 1.00 0.50 1.00Symphyla 5.13 10.25Protura 3.75 7.50Anuridae 0.50 1.00 0.38 0.75Entomobryidae 0.50 1.00 0.50 1.00Hypogasturidae 1.00 1.15 0.88 1.03Isotomidae 4.30 1.88 4.12 1.96Neanuridae 0.50 1.00 1.71 2.52Onychiuridae 4.17 3.07 3.78 4.30Tomoceridae 4.75 6.90 3.00 4.76NeelidaeSminthuridaeUndetermined 13.33 12.80 3.94 5.29Total Collembola 2.90 4.13 1.83 1.71Diplura 0.00 0.00 1.00 1.15Coleoptera (L)Coleoptera (A) 0.50 1.00 6.75 10.87Diptera (L) 0.50 1.00 6.00 5.89Diptera (A)HemipteraThysanuraNematoda 13.27 4.98 15.30 5.21Enchytraeidae 8.98 5.04 13.89 11.48Rotifera 2.50 3.79 1.92 1.35Tardigrada 11.51 7.24 10.52 4.88TOTAL 3.19 3.92 3.30 3.85March 1991AstigmataAppendix C (continued)^CH^HAMean^s.d.^Mean s.d.Cryptostigmata 3.62 1.26 5.47 2.56Mesostigmata 5.60 4.74 6.91 3.71Prostigmata 5.56 5.25 3.29 0.94Undetermined 0.50 1.00total Acari 3.06 2.69 3.13 3.14Aranaea 0.50 1.00 3.75 4.92Pseudoscorpionida 19.50 14.27 1.50 1.00Copepoda 9.31 4.59 8.54 7.25Pauropoda 0.50 1.00 0.50 1.00DiplopodaChilopoda 0.50 1.00 6.00 6.93SymphylaProturaAnuridae 1.00 1.15 3.38 4.19EntomobryidaeHypogasturidae 12.00 14.14 3.42 2.83Isotomidae 2.37 0.48 3.26 1.34Neanuridae 0.50 1.00Onychiuridae 9.96 6.87 12.91 2.70Tomoceridae 2.00 0.00 1.50 1.00Neelidae 0.50 1.00 8.50 15.70Sminthuridae 4.75 6.90Undetermined 5.33 5.42 4.42 2.20Total Collembola 3.37 4.34 4.21 3.94Diplura 3.75 4.92Coleoptera (L) 1.00 1.15Coleoptera (A)Diptera (L) 8.00 16.00 10.75 12.77Diptera (A)HemipteraThysanuraNematoda 16.13 6.85 13.11 2.38Enchytraeidae 6.58 1.85 7.56 4.83Rotifera 4.97 1.70 3.71 1.59Tardigrada 7.15 2.67 8.84 4.45TOTAL 4.43 5.07 4.63 3.73


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items