TEE SPATEAL FM’TEHHS 0F SEEEC’E‘EE SOEL ACRRE AM} CGiE‘EfiéBOLA EN AN ECOLGQICMEY EEAHIFHEATEB EKVERONMEM fine-ska if“ flue Deg?” of D5. D. EiECHIGAH ST "f‘fi UNWERSETY Jamea Henry Shacfidy - W76 I 11ng Wt IL TIL/I W111 ll Emil); j University w This is to certify that the thesis entitled THE SPATIAL PATTERNS OF SELECTED SOIL ACARI AND COLLEMBOLA IN AN ECOLOGICALLY MANIPULATED ENVIRONMENT presented by JAMES HENRY SHADDY has been accepted towards fulfillment Q ~ We requiremgs- “(a (W) ABSTRACT THE SPATIAL PATTERNS OF SELECTED SOIL ACARI AND COLLEMBOLA IN AN ECOLOGICALLY MANIPULATED ENVIRONMENT By James Henry Shaddy The spatial patterns of three groups of soil arthropods (two Acari; Gamasides, and Malaconothrus, and a species of Collembola; Mesaphorura granulata) were studied in an ecologically manipulated old field environment. Cultivation and the addition of water were used to manipulate the conditions. Plots were established with the following characteristics: uncultivated-no water added, uncultivateddwater added, cultivated-no water added, and cultivated-water added. A high gradient extractor was used to remove the soil animals from the soil samples. Extraction efficiency estimates were as follows: Gamasides - 90%, Malaconothrus - 100%, and Mesaphorura granulata - 93%. Cultivation had little effect on the time of peak density of the Gamasides but did result in a vertical redistribution with the largest population occurring at six inches (depth to which plots were plowed) ' and very few individuals present in the upper three inches of the soil. The spatial pattern of this group was mostly random except at the lower levels of the cultivated plots where aggregation occurred. A type II aggregation characterized this group. The addition of water had no conclusive effect. James Henry Shaddy The Malaconothrus was mostly aggregated in distribution exhibiting type II aggregation characteristics. Water had no apparent effect on seasonal abundance or aggregation characteristics. Mesaphorura granulata was mostly random in distribution. Water had no apparent effect. An initial reduction of numbers was observed in all groups in the cultivated plots. The most common size of aggregation in all groups was two inches or less and eight inches. THE SPATIAL PATTERNS OF SELECTED SOIL ACARI AND COLLEMBOLA IN AN ECOLOGICALLY MANIPULATED ENVIRONMENT By James Henry Shaddy A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Entomology 1970 Cgcwjflifl /"/ / 1', II a. f \ . yd ACKNOWLEDGEMENTS I wish to express my sincere appreciation to Dr. James Butcher for his encouragement and guidance throughout this study. A special note of thanks is extended to Dr. William Cooper, Dr. Paul Rieke, Dr. Gordon Guyer, and Mr. Richard Connin for serving on my guidance committee. Each member contributed considerable assis- tance throughout my study. Thanks is also extended to Mr. Ridhard Snider, Dr. James Truchan, Mr. Ernest Bernard, Miss Jacqueline Lorentzen, and others for their assistance during this study. A special note of gratitude is extended to my parents for their encouragement and to my wife for her patience and labor in the prepara- tion of this dissertation. ii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . v LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . vii INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 MATERIALS AND METHODS . . . . . . . . . . . . . . . . . . . . . . 4 Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Mathematical Manipulations . . . . . . . . . . . . . . . . . . . 17 RESULTS 0 I O O O O O O O O I O O O O O O O O O O O O O O O O O O 20 Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Soil and Soil Moisture . . . . . . . . . . . . . . . . . . . . . 22 Seasonal Abundance and Vertical Pattern . . . . . . . . . . . . 29 Gamasides . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Malaconothrus . . . . . . . . . . . . . . . . . . . . . . . . 29 Mesaphorura granulata . . . . . . . . . . . . . . . . . . . . 34 Horizontal Pattern . . . . . . . . . . . . . . . . . . . . . . . 34 Gamasides . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Malaconothrus . . . . . . . . . . . . . . . . . . . . . . . . 38 Mesaphorura granulata . . . . . . . . . . . . . . . . . . . . 42 DISCUSSION 0 O O O O O O O O O O O O O O O I O O 0 O O O O 0 O O O 48 Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Soil and Soil Moisture . . . . . . . . . . . . . . . . . . . . . 48 Seasonal Abundance and Vertical Pattern . . . . . . . . . . . . 49 Gamasides . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Malaconothrus . . . . . . . . . . . . . . . . . . . . . . . . 50 Mesaphorura granulata . . . . . . . . . . . . . . . . . . . . 51 Horizontal Pattern . . . . . . . . . . . . . . . . . . . . . . . 51 Gamasides . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Malaconothrus . . . . . . . . . . . . . . . . . . . . . . . . 52 Mesaphorura granulata . . . . . . . . . . . . . . . . . . . . 53 CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 iii POSSIBLE FUTURE STUDIES AND IMPROVEMENTS . LITERATURE CITED . . . . . . . . . . APPENDIX A . . . . . . . . . . . . . . . iv Page 57 58 62 Table 10. 11. LIST OF TABLES Sampling Scheme Showing the Block, Plots, Number of Replicates, and Depth of Each Sample on Each smp 1e Date I O O O O O O O O O I O O O O O O 0 Representative Daily Fahrenheit Temperatures at Points Monitored in Extractor During Extraction Period. Extraction Began Evening of Day 0 . . . Estimates of Extraction Efficiency for All Arthropods in Forty-two Hand Sorted Samples . . . List of Collembola Species Identified During Study 0 O O I O O O O O O O O O O O O O O O O O 0 Results of Soil Analysis From Each Plot. Each Figure Represents an Average of Two Samples . . . Total Rainfall and Water Added to Plots During his StUdy O O O O O O O I O O O O O O O O O O 0 Soil Moisture at a Depth of Four Inches in Each Plot on Each Sample Date. All Figures Represent an Average of at Least Four Samples . . . . . . . Results of a Two Way Analysis of Variance of Soil Moisture with Watered and Unwatered Treatments in Cultivated and Uncultivated Plots . . . . . . . . A Summary of the Spatial Patterns of Samples with Five or More Individuals per Sample of the Gamasides in All Plots . . . . . . . . . . . . . . . . . . The Relationship Between the Actual Number of Significantly Aggregated Samples, the Percent of Significantly Aggregated Samples and the Date of Sample for Gamasides in All Plots . . . . . . . . . . Summary of the Number and Size of Aggregates at Each Depth for Gamasides in All Plots . . . . . . . . Page 14 21 23 24 25 27 28 36 37 39 Table 12. l3. 14. 15. 16. 17. 18-87. A Summary of the Spatial Patterns of Samples With Five or More Individuals per Sample of the Malaconothrus in All Plots The Relationship Between the Actual Number of Significantly Aggregated Samples, the Percent of Significantly Aggregated Samples and the Date of Sample for Malaconothrus in All Plots Summary of the Number and Size of Aggregates at Each Depth for Malaconothrus in All Plots . . . A Summary of the Spatial Patterns of Samples With Five or More Individuals per Sample of Mesaphorura granulata in all plots The Relationship Between the Actual Number of Significantly Aggregated Samples, the Percent of Significantly Aggregated Samples and the Date of Sample for MeSaphorura'granulata in All Plots Summary of the Number and Size of Aggregates at Each Depth for Mesaphorura granulata in All Plots . Raw Data vi Page 40 41 43 44 45 47 62 Fig 10. ll. LIST OF FIGURES Figure Page 1. Location and Explanation of Experimental ManiPUIatj-ons O O O O O O O O O O ' O O O O O O O O O O O 5 2. The 4" x 4" x 2 1/2" Stainless Steel Soil Arthropod sampler C O O C O O O O O O I O C O O O O O O O O O O O O 8 3. The High Gradient Extractor Used to Extract Arthropods From Soil Samples . . . . . . . . . . . . . . ll 4. Demonstration of How Sample Was Inserted Into the Extractor O O C O O O C I O I O O O I O O O I O O O O O O 12 5. Shows How Sample Was Cut Into Subsamples and Placed Into Container for Extraction . . . . . . . . . . 15 6. Materials Used to Construct Device to Hold Subsample for Extraction: Left - Four Inch Polyethylene Tube; Right - Nylon Netting; and, Below - a 25 x 150 mm Test Tube . . . . . . . . . . . . . . . . . . . . . . . . l6 7. Time Relationship of Rainfall and Water Application to Arthropod and Soil Moisture Sampling . . . . . . . . . 26 8. The Vertical and Seasonal Distribution of Gamasides in CUltivated Plots 0 I O O O O O O O O O O O O O O O O 0 3O 9. The Vertical and Seasonal Distribution of Malaconothrus in Cultivated Plots . . . . . . . . . . . . 31 10. The Vertical and Seasonal Distribution of Mesaphorura granulata in Cultivated Plots . . . . . . . . 32 11. The seasonal Distribution of Gamasides, Malaconothrus and Mesaphorura granulata in Uncultivated Plots . . . . . . . . . . . . . . . . . . . 33 ' vii INTRODUCTION The Arthropoda of the soil, of which Acari and Collembola are a major component, is an extremely diverse complex of animals. Generally, the greatest complexity and abundance is attained in undisturbed habitats such as forests and grasslands, especially where the climate, vegetation and soil type combine to supply adequate moisture, tempera- ture and food (Burges and Raw, 1967). Their importance in this process of decomposition has been illustrated by Kubiena (1955), Edwards and Heath (1963), Ghilarov (1963), MacFadyen (1963), and Burges and Raw (1967), among others. It is generally known that the horizontal patterns of animal populations are not random, but are aggregated or clumped (Glasgow, 1939; MacFadyen, 1952; Hairston and Byers, 1954; Haarlov, 1960; Poole, 1961; Block, 1966; and Gerard, 1966). Klopfer (1969), among others, has suggested that heterogeneity of the environment has resulted in reduced interaction between Species consequently, permitting more species to coexist. If the environment was uniform, interspecific competition would be intensified resulting in a reduced number of dif- ferent species. Other factors, both biotic and abiotic that may affect, in different degrees, the distribution and abundance of soil animals are (1) competition for food, (2) reproductive behavior (i.e. issuance from egg masses or clumped eggs with limited dispersion of immature stages and clustering of progeny around parents), (3) mutual attraction 1 2 for other individuals of the same species, (4) response to daily and seasonal weather changes, and (5) increased survival of the species through clumping. Individuals in groups often experience a lower mor- tality rate during periods of stress (such as attacks by other organisms) than do isolated individuals, because the surface area exposed to the environment is less in prOportion to the mass and because the group may be able to favorably modify the microclimate or microhabitat (Allee, 1931; Odum, 1959). Several studies on the vertical distribution patterns of soil micro-arthropods have shown that the majority of the organisms occur in the uppermost layers of the soil, particularly in forest and grass- land regions (Glasgow, 1939; Nielsen, 1949; Kuhnelt, 1955; Murphy, 1955; and Poole, 1961). There is some disagreement concerning the effect of cultivation on soil populations. Edwards (1929) and Gisin (1955) have concluded that there is relatively little effect, while others have observed a deleterious effect (Karg, 1956; Sheals, 1956; and Strebel, 1957). Seasonal abundance of soil organisms have been correlated by numerous authors with such parameters as moisture and temperature (Ford, 1938; Weis-Fogh, 1948; MacFadyen, 1952; and Block, 1966). The effects of cultivation, moisture, temperature and many other factors have been reviewed thoroughly by Christiansen (1964) and Burges and Raw (1967). The non-economic soil arthopod fauna of the United States, have received little attention from zoologists. This study is intended to provide information on seasonal abundance, horizontal and vertical dis- tribution patterns of two groups of Acari (Gamasides and Malaconothrus) and one species of Collembola (Mesaphorura'granulata). 3 Other Acari and Collembola were not present in sufficient numbers to allow a sophisticated analysis. Therefore this information has been placed in Appendix A. The three above mentioned groups were studied in a cultivated and an uncultivated habitat which had water added during the driest part of the summer. These particular manipulations were studied since cultivation is a common practice in agricultural areas and its affect on the distribution of soil arthropods is poorly understood and at the same time, water may be an important limiting factor in their popula- tion growth and distribution (Christiansen, 1964; Knight and Chesson, 1966). Specifically, the research was designed to gain information on the following questions: (1) Does cultivation affect the horizontal aggregation pattern, the vertical pattern and seasonal abundance of the soil microarthropod fauna? (2) Does the addition of water during the summer affect the horizontal aggregation or vertical pattern, or the seasonal abundance? MATERIALS AND METHODS Plots In the Spring of 1968, eight plots were established in an old field that had not been disturbed for approximately 15 years, on the farm of Ray L. Cook, Chairman, Department of Soil Science, Michigan State University, located approximately ten miles North of Lansing, Michigan. Each plot was 22 feet wide and 30 feet long. On May 22, the disrupted plots were plowed to a depth of six inches and double disced to insure a mixture of plant residue with the soil. Beginning July 17th and ending September 26th, one inch of water was added to specific plots at seven to nine day intervals. Water was procured from a nearby drainage ditch and applied to the plots using oscillating sprinklers. Actual rainfall was recorded with a universal recording rain gauge. Figure 1 indicates the location of the experimental manipulations in the different plots. The four western plots are referred to as Block I and the four eastern plots are referred to as Block II. Each block contains an undisturbed plot, a disrupted plot, an undisturbed plot with water added, and a disrupted plot with water added. A lettering system was devised to identify each plot. This system is explained in Figure 1. The dominant plant fauna in the undisturbed plots consisted of Agropyron repens, Solidagg_sp., Poa compressa, and Phleum pratense. 4 ILOCK I BLOCK II No Water Wont Added ---CuI t ivotod -------- Uncul t ivctod--- "“01” i voted ---- "“0an t ivatod--- Wot. I Added Figure 1.--Location and explanation of experimental manipulations.1 1ACA = Block I, no water, cultivated. AWA = Block I, watered, cultivated. ACB = Block I, no water, uncultivated. AWB = Block I, watered, uncultivated. BWA 8 Block II, watered, cultivated. BCA = Block II, no water, cultivated. BWB = Block II, watered, uncultivated. BCB = Block II, no water, uncultivated. 6 ‘Agrgpyron repens and Ambrosia sp. were the dominant plants on the cul- tivated plots later in the summer. Soil samples were obtained from each plot for detailed chemical analysis to determine the homogeneity of the soil in the plots. Each sample consisted of 20 l-inch x 2—inch soil cores taken randomly over the plot. The cultivated plots were sampled to a depth of six inches with the two samples taken from each two inch interval. Two samples were taken from each uncultivated plot to a depth of two inches. The samples were analyzed by the Michigan State Soils Testing Laboratory at Michigan State University. The following methods were used in the analysis: extracted with neutral normal ammonium acetate and determined on a flame photometer. Potassium Phosphorus a Bray P. Nitrates = Brucine method. Carbon Leco Carbon Analyzer (combustion method). A gas-liquid chromatograph (Beckman GC-4) was used to check the soil and water source for possible insecticide pollutants. None were found. Sampling The peripheral three feet of each of the plots was excluded from the sampling scheme to eliminate any possible edge effect. The re- mainder of each plot was divided into one square foot grids. Stakes were placed along the edge of each plot at one foot intervals. Strings were placed between any pair of longitudinal stakes to determine the exact location of any point on the grid. Points on the grid in each 7 plot were selected at random from a table of random numbers with the restriction that a given point in any given plot could be sampled only once. Points chosen by this method determined the center of each sample location. Consequently, the center of each sample location was at least one foot from the center of the next closest sample. The device used to obtain samples of the soil for arthrOpod analysis was constructed from stainless steel. The penetrating edges were sharpened. The sampler had dimensions of 4-inch x 4-inch x 2 1/2- inch (Figure 2). In the cultivated plots, samples were taken to a depth of six inches, but only the top two inches of the soil was sampled in the undisturbed plots. It is known that the majority of animals are in the first two inches of an undisturbed soil (Glasgow, 1939; Gisin, 1943; Nielsen, 1949; Kuhnelt, 1955; and Poole, 1961). Since water was added to the uncultivated plots during this study, two samples were taken to a depth of six inches near the end of the sampling period to check the vertical distribution. It was found that over 75% of the Collembola and Acari were in the top two inches of the soil. The living vegetation was clipped at ground level and along with the litter, removed from the four inch square sampling point. The sharpened side of the sampler was pushed into the ground to a depth of two inches, removed with the sample intact, and placed in a styrofoam container to minimize desiccation during transport back to the laboratory. The hole that was left in the ground by removing the sample was immedi- ately filled with sand. On specific dates (Table 1), four samples were taken adjacent to one another at various depths to gather information on samples of a ‘o-‘L E : . p .. ‘3 .63 I ~ _. 1. v ' r216 . 3.“ \ sI . x 1 ‘9’“;5 \ “.3531“: .1 *5». Figure 2.--The 4" x 4" x 2 1/2" stainless steel soil arthropod sampler. TABLE 1.--Samp1ing scheme showing the block, plots, number of replicates, and depth of each sample on each sample date Date Block Plot Replicates Depth 6-7-68 I ACA, AWA I, II 1-6 I ACB, AWB I, II 1-2 6-20-68 ' II BCA, BWA I, II 1-6 II BCB, BWB I, II 1-2 7-2-68* I ACA I, II, III, IV 3-4 I AWA I, II, III, IV 3-4 7-17-68 I ACA, AWA I, II 1-6 I ACB, AWB I, II 1-2 7-31-68 II BCA, BWA I, II l-6 II BCB, BWB I, II 1-2 8-14-68* I ACA I, II, III, IV 5-6 I ACB I, II, III, IV 1-2 8-28-68 I ACA, AWA I, II 1-6 I ACB, AWB I, 11 1-2 9-12-68 II BCA, BWA I, II 1-6 II BCB, BWB I, II 1-2 10-16-68* I ACB I 1-6 II, III, IV 1-2 I AWB I 1-6 II, III, IV 1—2 11-11-68* I AWB I, II, III, IV 1-2 *Set of four adjacent samples were taken to form an eight inch square sample. 10 larger size. Thus an eight inch square sample instead of the standard four inch square sample was taken on those dates. Replicate samples were taken from each plot and are explained in Table 1. The raw data for the total of 16,797 micro-arthropods from the sample dates are presented in Appendix A. The moisture content of the soil was determined on each sampling date. Replicated samples were taken at a depth of four inches in each plot. The samples were brought back to the laboratory, weighed and dried in an oven at 100°C for 48 hours. The samples were then re- weighed to obtain per cent moisture. Extraction The arthropods were extracted from the soil samples by means of a modification of the Tullgren funnel described by MacFadyen (1961), Murphy (1962), and others. .The extractor used by the author is similar to the high gradient extractor which Kempson, Lloyd, and Ghelardi (1963, p. 5, Figure 5B) have described in detail (Figure 3). The extractor walls was constructed from 1/2-inch plywood. It differed from the extractor described by Kempson, Lloyd, and Ghelardi (1963) in the following respects: the baffle was built of 3/4-inch plywood as a separate unit which could be lifted out and replaced by a baffle of another design. The top side was covered with aluminum foil and the bottom covered with waterproof paint for insulation pur- poses. A total of 456 one inch holes were cut in the baffle in which the samples were placed (Figure 4). Two 43 1/2-inch L x 22 1/2-inch W x 6-inch H cooling baths were constructed with galvanized metal. A 1/2 horse power circulating 11 Figure 3.—-The high gradient extractor used to extract arthropods from soil samples. 12 .HOuomHuxm mfiu OUEH Umuhmmdfl wm3 . f . I”. ! ,£.L¢(¥ y! oHaEMm 30: mo coaumuumcosonll.q ouswflm refI ible homc wate mete time: peric once. therm coupl the m This ; samplc small becamt eXtra: over ; days ‘ the e: authol May a] 13 refrigeration unit was installed for each cooling bath. One submers- ible water pump was then placed in each cooling bath to insure a homogeneous temperature throughout the water. The temperature of the water at the beginning of each extraction period was maintained about 5°F below that of the soil temperature. A Honeywell 24 point potentio- meter was used to monitor temperatures within the extractor at all times. A total of seven points were monitored during each extraction period, each point or area within the extractor was replicated at least once. Four soil samples, each specially drilled to receive three thermocouples, were monitored during the extraction period. One thermo- couple was inserted near the top of the sample, one was inserted near the middle of the sample, and the third near the bottom of the sample. This allowed measurement of the temperature gradient through the soil samples throughout extraction. The gradient through each sample was small at the beginning of the extraction, usually about 4° - 6°F, but became more steep as extraction proceeded. At the end of the nine day extraction period, the temperature at the top of the soil sample was over 120°F while the temperature at the bottom was 85° - 90°F. Representative series of the temperature readings for different days during the extraction period along with the points monitored in the extractor are presented in Table 2. Although the temperatures in this study correspond closely to those obtained by Kempson, Lloyd and Ghelardi (1963), studies by these authors and Nef (1962) indicate that desiccation along with gravity may also be very important stimuli in the extraction process. Because 14 of the small size of the sample, it was impossible to monitor the moisture within the sample in this study. TABLE 2.--Representative daily fahrenheit temperatures at points monitored in extractor during extraction period. Extraction began evening of day 0 Point of Thermocouple l 2 3 4 5 6 7 8 9 Water bath 62 62 62 62 63 63 63 64 65 Air above bath 62 62 63 64 64 65 65 66 67 Air directly above sample 82 83 83 85 87 90 95 100 110 Air in extractor 82 83 83 86 87 89 91 92 99 Top of sample 75 76 77 83 86 94 100 111 120 Huddle of sample 73 74 75 76 79 87 91 97 107 Bottom of sample 68 69 70 73 73 75 77 8O 89 The samples were prepared for extraction by cutting each into 16 subsamples l-inch x l-inch x 2-inch in size or into 32 subsamples l-inch x l-inch x 1-inch (Figure 5) size. This allowed delineation of animal distribution patterns to a depth of six inches, in one inch increments in the cultivated plots. If the sample was dry and crumbly, distilled water was added to hold it together during the cutting process. Each subsample was placed in a polyethylene tube one inch in diameter and four inches long. This tube was then slipped over a 25 x 150 mm dis- posable culture tube with a three inch square piece of nylon netting placed between the culture tube and the polyethylene; thus, the sample was kept intact within the polyethylene tube (Figure 6). Each culture 15 Figure 5.—-Shows how sample was cut into subsamples and placed into container for extraction. l6 left - four Figure 6.-—Materials used to construct device to hold subsample for extraction inch polyethylene tube; right - nylon netting; and, below — a 25 x 150 mm test tube. l7 tube was filled to within 1 1/2 inches of the top with concentrated aqueous picric acid for collecting the soil arthropods. After all of the samples had been inserted into the baffle of the extractor, it was then lowered onto the extractor base. This allowed the culture tubes containing the picric acid to be immersed in the cooling bath. The extraction efficiency was determined by hand sorting 42 ran- domly selected subsamples after complete extraction in the extractor. This is one of the methods of determining extraction efficiency and although time consuming, has been used by many authors (Forsslund, 1949; MacFadyen, 1953; Hughes, 1954; Murphy, 1962). At termination of the extraction period, the samples were re- moved from the extractor and the culture tubes containing the animals were sealed with corks and stored. These were later counted and sorted under a binocular dissecting microscope. The Collembola in the samples were sorted to the species level using the classification of Snider (1967). The mites were identified to genus or family whenever possible. Other specimens were grouped under unidentified. The remaining arthrOpods were identified to genus, family, order, or class. Mathematical Manipulations The experimental design, sampling program, and extraction techniques were specifically designed for the use of Morisita's (1959) index of dispersion of individuals and analysis of distributional patterns. This index is: 18 N Z n (n -1) i=1 i i I = N x(x-l) Where N = total samples, n1 = numbers in the 1th sample and x = the sum of the numbers of individuals found in all the samples. This index has the advantage of being relatively independent of the type of dis- tribution, the number of samples, and the size of the mean (Southwood, 1966). When the distribution of the population is Poisson (random) the index will exhibit a value of unity. When the distribution is contagious (negative binomial) the index will be greater than one and if the distribution is regular (binomial) the index is less than one. Morisita (1962) has defined the exact mathematical relationship between the index and the parameters of these distributions. An F test developed by Morisita (1959) was used to test the significance of the departure from a random distribution indicated by the index: I(x-l) + N-x N-l F: The size of aggregations within samples can also be determined by using Mbrisita's (1959) I value. The formula is I(s)/I(Zs), where 1(a) is the I value of the quadrat size 8, and I(23) is that of the quadrat size 28. When these computed values are plotted, the peak or peaks of the curve will determine the aggregation size. Usher (1969) has developed a method of defining the type of aggregation exhibited by a soil animal. Type I aggregation is char- acterized by having a positive and significant correlation coefficient between the population density and the number of aggregated samples while the correlation between the papulation density and the mean 19 number of individuals per aggregated sample is either non-significant or negative and significant. Such an aggregation could result from some fixed attribute of the species such as the size of the egg cluster. A Type II aggregation is characterized by a non-significant correlation between papulation density and the number of aggregated samples and a positive and significant correlation between population density and mean number of individuals per aggregated sample. This type of aggre- gation would result from the aggregations being located either in par— ticularly suitable microenvironments or in relation to food supply, with the individuals moving there. A Type III aggregation would be characterized by both correlations between population density and the number of aggregated samples, and population density and the mean number of individuals per aggregated sample being positive and signifi- cant. This method has been employed in this study where sufficient data were available. An analysis of variance was used to test for significant dif- ferences among the ecological manipulations discussed earlier. RESULTS Extraction Generalizations concerning extraction efficiency of soil arthro— pods cannot be made since they are an extremely diverse group of animals. Such factors as behavior, size, and age, only to mention a few, will affect their egress from a soil sample. For example, it is known that Oribatid nymphs tend to be slower than adults in leaving the sample which results in their exposure to deleterious conditions for a longer period of time; this may increase the ratio of nymph to adult mortality (Haarlov, 1947; Murphy, 1958; Murphy, 1962). Other factors such as mineral and organic matter content of a soil will also influence ex- traction efficiency (Nef, 1960; Murphy, 1962). In determining the extraction efficiency in this study, dis- tinction was not made between adult and immature forms of the Acari and Collembola. Forty-two dried soil samples were placed in 70% alcohol after extraction was finished and later searched through for corpses of soil animals. The extraction efficiency estimates for all arthropods are presented in Table 3. Among the Acari, Malaconothrus was extracted with 100% efficiency. Oppia, Gamasides, and Scutacaridae were extracted with efficiency of 100, 90, and 89% respectively. Tetocepheus and unidentified Prostigmata groups were lowest with 75 and 73% efficiency. 20 21 TABLE 3.-—Estimates of extraction efficiency for all arthropods in forty two hand sorted samples No. No. Per Cent Animal Extracted Not Extracted Efficiency Acari Malaconothrus 28 0 100 Oppia 4 0 100 Gamasides 56 6 90 Scutacaridae 1 89 Oribatei 1 83 Tectocepheus l 75 Prostigmata 16 6 73 Average 89 Collembola Mesaphorura granulata 52 4 93 Isotoma notabilis 36 5 88 Protaphorura armata 14 2 88 Pseudosinella violenta 1 86 Megalothorax albus 5 l 83 Average 90 Other ArthrOpods Ants 8 0 100 Psocoptera l O 100 Hymenoptera 1 O 100 Paurogus 1 O 100 Campodea 1 O 100 Coleoptera 9 3 75 Heteroptera 21 9 7O Diptera 2 3 40 Average 75 Overall Average 87 22 The average efficiency for all Collembola was 90%. Mesaphorura granulata was extracted with an efficiency of 93% while Megalothorax albgg exhibited the lowest efficiency at 83%. A total of 26 species of Collembola were identified in this study (Table 4). Many of these did not occur in sufficient numbers to obtain an efficiency estimate. The average efficiency for arthropods other than Acari and Collembola was 75%. Soil and Soil Mbisture ‘ The results of the chemical analysis of the soil (Table 5) in- dicated that no important differences existed between plots with respect to percent carbon, potassium, phosphorous, nitrates, or pH. All of the above elements are present in minimal amounts1 which may explain the low populations of some soil arthropods. Based on these results, it is assumed the study plots are homogeneous, although it is not possible to obtain complete homogeneity in any natural soil. The mechanical analysis indicated the soil was 55 percent sand, 27 percent silt, and 18 percent clay. This represents a Washtenaw sandy loam soil type which is approaching a loam. A total of 13.9 inches of rainfall was recorded in the area of the experimental plots from June lst through September 30th. Nine inches of water was added artificially making a total of 22.49 inches of water on the watered plots (Table 6). Because of environmental and mechanical complications, it was impossible to begin adding the water to the plots earlier. If the additional water had been applied at a 1Personal communication from Dr. Paul Rieke, Professor of Soils, Michigan State University. 23 TABLE 4.--List of Collembola species identified during study Arrhopalites benitus Folsom Arrhopalites pygmaeus (Wankel) Bourletiella juanitae Maynard Entomobrya multifasciata (Tullberg) Entomobryoides purpurascens (Packard) Folsomia candida Willem Isotoma notabilis Shaffer Isotoma viridis Bourlet Lepidocyrtus lignorum (Fabricius) Lepidocyrtus paradoxus Uzel Lepidocyrtus violaceus (Geoffrey) Megalothorax albus Maynard Mesaphorura granulata (Mills) Metakatianna maggillivrayi (Banks) Neanura muscorum (Templeton) Neelus minutus Folsom Orchesella ainsliei Folsom Proisotoma minuta (Tullberg) Protaphorura armata (Tullberg) Pseudosinella rolfsi Mills Pseudosinella sexoculata Schott Pseudosinella violenta (Folsom) Sinella curviseta Brook Sminthurinus elegans (Fitch) Tomocerus flavescens (Tullberg) Willowsia platani nigromaculata (Lubbock) 24 .coHHHwa non muumm i wmvv¢ Hmumz «.0 N.NN 0N 00 0N.N 0.0 N.0 0N 00H 00.N N emum>fiuNauaa 0000< umumz oz 0.0 0.0 0.0N NNN 0N.N N.0 0.NN 0.0N 0.00N em.m N 0000>Nuaauaa 0.0 N.0N 0.N oN NN.N N.0 «.0N 0.0 00 NH.N 0 0.0 N.0 0.0 00 m0.N N.0 0.0 N 00 0N.N 0 0000< “mums 0.0 N.N a N0 N0.N N.0 0.m 0 00 oo.N N umum>fiuaso 0.0 o.NN 0.0 00 00.N 0.0 0.m 0.0 0N 00.N 0 0.0 N.N 0.0 00 NN.N 0.0 N.NN N 00 00.N 0 0000< “mums oz m.0 N.N 0 00 0N.N 0.0 0.N m 00 00.N N emua>NuN=o mm «002 «N Na No ms «002 am «a No enema uoNN NH xuoNN H xuon moamsmm oBu mo ommuo>o do muaomouaou ounmfim zoom .uoam some Bouw wfimhamam Hfiom mo muaamoMII.m mqm