PM: man FAUNA OE BEAVER ISLANDLZ-r.:‘jfifff [if m:- commsmom. momma“. :j . 2‘2 iifz’iii‘ * onsmwraow AND ORIGIN ,; _;_j * : 11¢“;ng 93m Dogma o9 Db. D. ‘5: ' MICHIGAN STATE UNIVERSITY : 3 : ' ‘ Lesllie C. Drew" . : 1965.: ; THESlS LIBRARY Michigan State University This is to certify that the thesis entitled The Spider Fauna of Beaver Island, Its Composition, Ecological Distribution and Origin presented by Leslie C. Drew has been accepted towards fulfillment of the requirements for Ph. D. degree mm Major professor Date August 6, 1965 0-169 am 88994. ABSTRACT THE SPIDER FAUNA OF BEAVER ISLAND, ITS COMPOSITION, ECOLOGICAL DISTRIBUTION AND ORIGIN by Leslie C. Drew A total of 211 species of spiders was collected from the 58-square mile Beaver Island located in northern Lake Michigan. Qualitative and quantitative collecting methods disclosed the ecological distribution of the species through- out the island. Quantitative collecting was performed in specific eco- logical communities to ascertain the structure of the spider population with reference to frequency, density and food- getting behavior. Diurnal and seasonal succession was studied in the Old Field community. The populations inhabiting the leaf litter stratum of the mixed hardwood-conifer community on the island and the adjacent mainland of the Lower Peninsula were compared. The statistical characteristics of these p0pu1ations are evaluated by means of Fisher's Index of Diversity. The evolution of this particular pOpulation on the island is discussed. Leslie C. Drew The origin of the spider fauna of Beaver Island is examined with reference to post-glacial history, current climate, agents of distribution and the influence of the surrounding mainlands of Wisconsin and Michigan. THE SPIDER FAUNA OF BEAVER ISLAND, ITS COMPOSITION, ECOLOGICAL DISTRIBUTION AND ORIGIN by f? >\ . “9 Leslie CE Drew A THESIS Submitted to ‘Michigan State University in partial—fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Zoology 1965 Acknowledgements First, I thank my wife, Lorraine, for her helpful assistance in many facets of this work and Miss Josephine C. Giambrone for assistance in the preparation of the manu- script. I am.especially grateful to Dr. T. wayne Porter for his continued efforts in stimulating my progress through the past several years. Sincere thanks go to Dr. Roland L. Fischer and Dr. Marvin M. Hensley for their advice and guidance. Greatly appreciated is the assistance given by Dr. Henrietta Band (Mrs. R. Neal Band) and the late Dr. Philip J. Clark. I am also grateful to: Dr. Ralph Mac'Mullin and Dr. Ralph Blouch of the Game Division of the Michigan Depart- ment of Conservation; Mr. Vernon Fitzpatrick, former Beaver Island Fire Officer; Mr. George L. Egbert, a summer resi- dent of Beaver Island; Mr. Wilton Ivie and Dr. W. J. Gertsch of the American Museum of Natural History; and Dr. Herbert W. .Levi and Dr. Arthur M. Chickering of the Museum.of Compara- tive Zoology, Harvard University. I would also like to thank Mrs. B. R. Henderson, secre- tary of the Department of Zoology for her assistance. ii CONTENTS ' ACKNOWLEDGEMENTS LIST OF FIGURES LIST OF TABLES INTRODUCTION A DESCRIPTION OF BEAVER ISLAND Physiography, Climate, Soils and Vegetation Post-Glacial History of Beaver Island MATERIALS AND METHODS RESULTS The Spiders of Beaver Island Spider Pepulations of Specific Communities Old Fields Upland Hardwood Forest (BeechrMaple) Juniper Bushes (Juniperus depressus) White Cedars (Thgig occidentalis) ' The Spider P0pulations of a SimilarHabitat on the Island and on the Mainland of Michigan CONSIDERATIONS ON THE ORIGIN OF THE ISLAND FAUNA Zoogeographic A5pects Wind as a Factor in Distribution Distribution Through Human Activities iii page ii vi 16 20 20 44 44 48 50 52 52 64 64 6o 72 CONTENTS (continued) page DISCUSSION 74 Composition of the Spider Fauna of Beaver Island 74 Spider Populations of Specific Communities 77 The Spider Populations of a Similar Habitat on the Island and on the Mainland of Michigan 78 SUMMARY AND CONCLUSIONS 81 LITERATURE CITED 84 iv FIGURE 10. ll. 12. LIST OF FIGURES Tepography of Beaver Island The soils of Beaver Island Major cover types and their distribution Lake Chippewa stage in Lake Michigan basin Diurnal succession of old field populations Populations in old fields, July POpulations P0pulations P0pulations POpulations in in of of Summer surface Variability of old fields, September upland hardwood forest junipers cedar trees currents summer surface currents page 14 45 47 49 51 53 54 69 7o TABLE LIST OF TABLES Major plant species and their areal dis- tribution Lake stages during post-glacial history Summary of the spiders of Beaver Island Webdweaving families of the Island,'Michi- gan and Wisconsin Hunting families of the Island, Michigan and Wisconsin POpulations of mixed hardwood-conifer forest Comparisons of Island and mainland leaf litter populations Spiders found on surface of Lake Michigan vi page 12 4O 42 43 59 62 71 INTRODUCTION Zoogeographershave used glacial andpost-glacial events to interpret existing distributional patterns of many animal groups. Smith(l957) in his studies.of verte- brates, primarily amphibiansand reptiles, distributed in northeastern United States, concludes that the ebb and _flow of changing climates during and following.the last glacial epochaccounts for the present disruptive patterns of several species. Van der Schalie (1963) makes use of post-glacial stream confluence to account for the distri- bution of freshewater'muSSels in the northern Great Lakes region. The araneologist.is concerned with ED 3 and from ‘whégg certain species came to the island. gggw.for'most species is aerial distribution; however, spidersas waifs can also gain access. These facts of spider distribution “are best exposed by quantitative census to determine what species are presentand then making use of known distribu- tion data. The present study is concerned primarily with the composition and origin of the spider fauna on Beaver Is- land. It also includes population profiles of established island habitats and a comparison of a Specific ecological situation on the island with that of a similar one on the adjacent‘mainlandof‘Michigan. 1 A DESCRIPTION OF BEAVER ISLAND Physiography, Climate, Soils and Vegetation M Beaver Island, with an area of 58.4 square miles (37,376 acres), occupies the position in northeastern Lake Michigan between 45°34l - 45°46} north latitude.and“85°29' - 85°38' west longitude. Politically, the island is part of Charlevoix County, Michigan. ,The island is 19 miles west of Emmet County in the Lower Peninsula of Michigan and 15 'miles south of MackinacCounty in the Upper Peninsula. Beaver Island is 13F2 miles long (northesouth axis) and 6.4 miles across at its greatest width. 'Elevations vary from 578 feet above sea.level, the level of Lake Michigan (1956 map-n4530-W8530/15, l-62,500, of the United States Geological Survey of Washington, C. C.), to 790feet, the site ofthe fire tower south of Fox Lake. The greater portionof the island.is 229172 feet above the level of the Lake. St. James, the only settlement on the island, is 15 feet above lake leveland is situated along the mar- gins of Beaver Island Harbor at the northeastern tip of the island. Two Silurian-Devonian limestone formations, Bois Blanc and Detroit River, underlie the island; the Bois Blane for- mation.is under the northern third and is exposed at two sites along the northern shoreline; the Detroit River 3 formation is under theremaining portion.of the island. Both are tilted and dip to the southeast. This bedrock is mantled with glacial till of varying thickness giving rise to the gentle rolling topography of the island (Fig. 1). Eight inland lakes differing in size,bottom~type'and depth are scattered over the island. Three major and a number of intermittent streams are located in the south-central and southeastern areas. These flow directly into either Lake Michigan or Lake Geneserath which in turn drains into Lake Michigan. The beaches, forming the shoreline, range frmm limestone outcrop and boulder to cobble, pebble and sand. Exposed dunes occur behind*the shoreline along the northern part of the western margin;and captured vegetated dunes occur just behind the south-central shoreline. The climate of Beaver Island is seasonally affected by Lake Michigan. Duringrthe summer,shore stations to the west, north and east of the island have their mean daily. maximum duringJuly. This daily maximum occurs in August on the island. However, during the winter when the Lake is partially or wholly covered by ice, the climate is contin- ental, similar to that of the mainland and the mean mnnthly temperatures are coincident. Annual precipitation shows further Lake influence. During the summer, this area of the Lake receives 6.2% less precipitation than the surround- ing shore stations, and 4.5% more winter precipitation (Blust Figure 1. The topography of Beaver Island. ‘ih.< Feat Lake Lake Geaeaerath too .978 Hundreds of Feet above aea level 3 6 f Y Y 7 Iii/u a l 3 ' I Baat-Weet traaaect acroaa Beaver lalaad at latitude 45.38' North, Lake Geneaerath 608' Font lake 619' V V I Iii/n I l J i I t I T v! a; - 7 fi North-South transect across Beaver Island at longitude 85°33' West, 5 and DeCook, 1960). This fluctuation.iswattributed to the presence of the vast mass ofopen water during the summer and its stabilizing or temperinginfluenceon atmospheric storm cells. During the winter, the island intercepts storms due to the unequal heating of the tree shrouded is- land area as opposed to the relatively uniformsurrounding ice. The average annual precipitation is 24.72 inches. September receives the maximumandJanuary the minimum. The mean annual temperature is 43.80F.; the mean daily maxi- mum 75.80F.; mean daily- minimum. 16.6°F. The highest maxi- mum reeorded is 97°F. and the lowest minimum.-280F. During the summer, the prevailing winds are from the west-south- west with.occasional periodsof variable winds.from the east-northeast. During the winter months, the prevailing winds are from the north-northwest. The growing season, extending from 14 May to 15 October, is 156 days long. This is comparable to that of the interior counties 200 miles to the south in the.Lower Peninsula. The soils of Beaver Island are agriculturally poor and are quickly depleted under cultivation. According to Veatch (1953), four types occur on the island. The distribution and classification of soils is given inFigure 2. The vegetation of the island has been studied by Fuller (1918), Darlington (1940) and the Game Division of the i .2 ~? I I l I Scale of Nile: Eastport, Wet land sands; primarily young lacustrine sands on both wet and dry sites. Kalkaska-Rubicon, Sands and light sandy loams, locally gravely and cobbly, Munuscong-Detour-Alpena-Rubicon, Wet flats of stony sandy loams and stony clay, Emmet-Roselawn, Mostly sandy loams. In part deep sands or reddish clay at shallow depths. Figure 2. The soils of Beaver Island according to Veatch, t953. 7 Michigan Conservation Department. Darlington (gp.lgi£)re- ported that the island has 30 of the 33 plant communities found on the mainland of Michigan. The Game Division mapped the cover types in 1945 and has updated the map periodically. Figure 3 depicts the current distribution of the major vege- tative types of the island. The areal distribution and ma- jor species of these types are presented in Table l. Post-Glacial History of Beaver Island. Studies evaluating the influences of pre- and post- glacial events as well as past climates on current distri- butional patterns have been reportedby many authors- Nota- ble among these are Dillon (1956), Martin and Harrell (1957), Smith(l957), Howden (1963) and van der Schalie (1963). Studies to evaluate these influences on the distribution of ' spiders are handicapped because: most of these organisms are passively dispersed by aerial means; relict populations have yet to be recognized; and, relatively few conclusive studies have been done ongeographic variations. Nonethe- less, some species must have been carried to and deposited on the varying amounts of exposed land of the northern basin of Lake Michigan during the post-glacial epoch. Therefore, the sequence of events that compose the post-glacial history of Beaver Island is pertinent to this study. - 33.325 - iii? Scale of Miles Figure 3. A generalized cover-type map of Beaver Island. 9 Table 1. Major plant Species and areal distribution of vegetative types on Beaver Island Area Cover types and major species sq. mi. % Cleared Lands 17.5 30.0 'apple (Pyrus malus)2, ground juniper (Juniperus communis var. depressa), blackberry (Rubus allegheniensis), bracken fern (Pteridium aquilinum), milkweed (Asclepias £22.), grasses (Poa'compressa, P. pratensis, Phleum pratense and Agropyron repens). Upland Hardwood 13.2 22.7 beech (Fagus grandifolia), sugar maple (Acer saccharum), eastern hem- lock (Tsuga canadensis). Mixed Hardwood-Conifer 17.0 29.1 balsam fir (Abies balsamea), white spruce (Picea glauca), northern white cedar (Thuja occidentalis), white and yellow birch (Betula papyrifera, P. lutea). Swamp Conifer 5.3 8.9 northern white cedar, eastern hemlock, balsam fir Upland Pines 2.9 4.8 red and white pine (Pinus resinosa, '2. strobus, red oak (Quercus rubra) Shoreline 2.5 4.4 willows (Salix cordata and S. glau- cophylloides var. glaucophylla), sand cherry (Prunus pumila), red osier (Cornus stolonifera), marram grass (Ammophila breviligulata), balsam- pOplar (POpulus balsamifera). '§§TZ 100% 1 Modified from Fuller (1918), Darlington (1940) and Ozoga (1963). 2 Species names follow those in Fernald (1950). 10 Geological evidence indicates portions of the present basin of northern Lake Michigan were alternately exposed and submerged. This evidence consists of beaches and wave- cut terraces, and shallow-water sediments underlying more recent deep-water sediments above and below the present lake level. Interpretations of this evidence-have been reported by Leverett and Taylor (1915), Stanley (1938) and Hough (1955, 1958, 1962, 1963). The sequence of events composing the post-glacial history is fairly well esta- blished. However, the duration of various stages is some- what controversial. Radiocarbon dating techniques, used since 1949, have given rise to diverSe chronologies. By way of illustration, the dates for wood samples from the Two Creeks Forest (the unofficial standard reference mate- rial) varies from 11,400 years before the present (Arnold and Libby, 1949, 1951) to 11,850 i 140 years Before Pres- ent (Broecker and Farrand, 1963). The variance is com- pounded by suggested dates for what is sometimes called the Two Creeks Interval or the Two Creeks Low-water Stage. Zumberge and Potzger (1956) date this at 11,000 Before Present; anthropologists use the date of 12,450 years Before Present (C. E. Cleland, pers. comm.). Consequent- ly, the time intervals of post-glacial events are, at best, tentative. The radiocarbon dates recorded by Hough (1958) 11 are used for the post-glacial events of interest to the present study. Concepts of post-glacial climates and concomitant forest types have been postulated from pollen profiles by Deevey (1949) for the northeastern region of the United States and by Zumberge and Potzger (1956) for two sites in southwestern.Michigan. There is considerable agree- ment between these two studies; and when differences do occur, it is probably regional. These data have been in- corporated in the following account with the realization that studies on post—glacial succession dynamics (Wright, 1964) may modify them somewhat. Lake Algonquin was the first major lake stage follow- ing theretreat of the Valders, the last glacial ice, from the Lake Michigan basin. It came into being 8,800 years Before Present and maintained a level of 6051 feet Above Tide for 800 years. Portions of the present Beaver Island were above this level (Table 2). Deevey postulates that a tundra flora followed the northward retreat of ice through the entire glaciated region. In the pollen profiles at South Haven, Michigan, Zumberge and Potzger have identified spruce-fir pollen at this age level and conclude, as does Deevey, that the climate was cool and moist. 1 The elevations are average values representing a verti- cal range of about five feet (Hough, 1958). 12 Table 2. Lake stages during the post-glacial history of Beaver Island ' Total No. of shoreline Lake stages Elevation Islands in miles Algonquin 605 ft. 5 11.2 miles Chippewa 230 ft. 11 185.0 miles Nipissing 605 ft. 12 59.7 miles Algoma 596 ft. 1 37.0 miles‘ Lake Michigan 578 ft. 1 41.6 miles 1 Really a peninsula (Fig. 4). The shoreline measured was that which extended from present Lake Charlevoix on the south, the Mackinac River on the north to the present Straits of Mackinac. 13 As new discharge routes were uncovered by the north- nOrtheastward retreating ice, the lake level dropped grad- ually to a low of 230 feet Above Tide. This recession, which occurred over a period of 1500 years, exposed in- creasing amounts of the former lake bed and, thus, united the former isolated islands into a peninsular body broadly connected to the present Lower Peninsula (Fig. 4). The pollen profiles reflect a climatic change during this stage. The Spruce-fir pollen is overlain by increasing amounts of pine pollen indicating a warmer and drier climate. Zumberge and Potzger identified a shift from jack-pine to white and red pine during this period. The low-water stage of 230 feet Above Tide, known as Lake Chippewa, remained for 700 years (6,500-5,800-Before Present). An increasing amount of oak pollen occurs with that of pine in the South Haven profile which indicates a continuation of the warm dry climate. lsostatic rebound in the northeastern portion of the Great Lakes region gradually decreased the northern dis- charge route at North Bay, Ontario. As a result, the water level in the northern Lake Michigan basin rose once again to 605 feet Above Tide. This latter level was deter- mined by the elevations of the discharge route from the 15 southern portion of the basin and that of Port Huron in Lake Huron. This gradual increase in water level, termed the Transition by Hough (1958), occurred over a period of 1800 years (5,800-4,000 years Before Present) and frag- mented the extensive continuous shorelines of the previous stage. The climate remained warm but became more humid as evidenced by pollen of oak, beech, hemlock and elm in the profiles. Smith (1957) and Hough (1958) report this to be the Climatic Optimum or Thermal Maximum.of the geologists. The re-establishment of the 605 feet Above Tide level, known as the Lake Nipissing stage, lasted for 1000 years (4,000-3,000 years Before Present). During this stage, Beaver Island was a series of 12 or more islands (Table 2) of various sizes and elevations. Deevey (1949) refers to this stage as the Thermal Maximum, as do Zumberge and Potzger (1956). The basis for this conclusion by Deevey is the occurrence of oak and hickory pollen. He also as- sumes a steppe-like vegetation developed or there was an extension of the western grasslands into the recently gla- ciated northeastern region. There is no palynological evi- dence to substantiate this assumption. However, the present distributional patterns of some organisms is best explained by the occurrence of such vegetation (Smith, 1957).‘ The profile at South Haven,Michigan records the occurrence of oak and pine during this stage. 16 Lake Nipissing was brought to a close by erosion of the sill in the discharge route at Port Huron. This re- sulted in a new level, 596 feet Above Tide, known as Lake Algoma, which lasted 800 years (3,000-2,200 years Before Present). During this stage, present day Beaver Island was exposed as one single island with a shoreline of 37 miles (Table 2). Alsoduring this stage, the climate became cooler and more moist much like that of the present. The evidence for this conclusion is that oak-hickory pollen deposits in the profiles of Wisconsin and Minnesota are overlain by pollen of species presently occurring in these areas (Deevey, 1949). Continued erosion at Port Huron in conjunction with a lateral shift of the Detroit River dropped the water level of the lake to 580 feet Above Tide-~present day Lake Michi- gan. There have been minor fluctuations during the last 2200 years but none of the magnitude of preceding stages. MATERIALS AND METHODS Qualitative and quantitative collecting was accom- plished with the following equipment: a 12-inch diameter insect net attached to a three-foot handle; a 30-inch square beating net of medium weight canvas suspended from the tips 17 of a wooden collapsible X-shaped frame; a wooden frame en- closing one square foot; two sifting boxes, one 13" x 13" with 5/16" mesh screen bottom, and the other 10" x 10" with 1/16" mesh screen; an aspirator; and a 5-cell head lamp. The methods varied according to the purpose for col- lecting. General collecting to determine the composition of the spider-fauna included visual searching during the day and night hours with and without the above equipment. Sweeping with the insect net, beating with the beating net and sifting leaf litter were the methods used to collect for the spider population profiles of specific ecological communities. In studying the pOpulation of old fields, use was made of the sweeping methods advocated by Barnes (1953). One sample includes those spiders collected in 200 strokes over a distance of 100 paces. Each stroke of the net covers an approximate 6-foot are through the standing vegetation; and a return stroke is taken before stepping forward one pace. The populations inhabiting clumps of junipers, white cedars and shrubs of forests were sampled by use of the beating net. The method used in these situations was to place the net under boughs or branches which were then struck twice with a three-foot, lZ-ounce stick. After two 18 such blows, the spiders were collected from the net by means of an aspirator. One sample consists of those spi- ders dislodged by 200 such blows which usually covered more than one bush or tree. Spider populations of the leaf litter stratum were studied by sifting the specimens out of one square-foot samples of litter. In an attempt to assure random selec- tions of samples, these procedures were followed: a two- digit random number (from the tables of Dixon and Massey, 1957) determined the number of paces along a directional line to the starting point; the end of a 100-foot steel tape was secured to the ground by means of a six-inch lag screw; the last digit of another two-digit random number determined the distance in feet along the tape to the sam- ple, where the wooden frame that encloses one square foot was placed. The leaf litter enclosed by this frame was gathered by hand and placed in a paper bag, the top of which was folded down several times, secured with paper clips, and marked as to distance from the starting point and sample number. Later, the contents of each bag was sifted and the spiders collected by use of the aspirator. All specimens from one bag, constituting one sample, were put into the same vial containing 70% alcohol for identifi- cation later. 19 The results of the quantitative sweeping, beating and sifting can be expressed in terms of frequency and density as applied by Barnes (1953) and Barnes and Barnes (1954, 1955) in studies of the spider populations in various real communities. Frequency as defined by Barnes and as used here is a statistical expression of the number of times a particular species occurs in a given number of samples taken from an existing community. Frequency is expressed as a percentage and is derived by using the following formula: No. of samples with particular species x 100 A Frequency = Total number of samples Density is defined as the average number of individuals of a particular species per sample and is calculated as follows: , 4 Total number of specimens of a particular species Den31ty ' Total number of samples These two expressions provide a graphic representation of the population structure of specific communities. The Index of Diversity was calculated for the species occurring in the leaf litter communities of mixed hardwood- conifer forests on Beaver Island and on Waugoshance Point, Emmet County--that portion of the Lower Peninsula closest to Beaver Island. Such an index is based on the relation- ship between the total number of individuals and the number of species contained in samples. This index was evolved by Fisher, Corbet, and Williams (1943) and is known as the 20 logarithmic series. It is expressed by this formula: S =CXln(1 + N/a). Whereincxis a constant independent of sample size and, thus, becomes an expression of the diver- sity of species in relation to the total number of individ- uals. The comparison of such indices from areas that were once geographically continuous could possibly reveal changes in the population structure of presently ecologically simi- lar communities. Quantitative sweeping and beating samples were taken during the course of one day. This was not feasible with the leaf litter samples taken on the island and on the mainland. However, these samples were taken within a one week period. Collecting on the island was done during the spring, summer and fall of 1961, 1962 and 1963. The specimens collected during this study have been deposited in the collections of the Entomology Museum, Michigan State Uni- versity, East Lansing, Michigan. RESULTS The Spiders of Beaver Island Of the 211 species known to occur on Beaver Island, 114 or 54.1% are web-weaving species and 97 or 45.9% are hunting species. These two descriptive categories, used .. 21 throughout the remainder of this report, require defini- tion: web-weaving species are those which capture their prey by entanglement in an aerial web; hunting species are those which capture prey by sheer physical acumen, speed or stealth, in the presence or absence of a web. These categories, derived in part from Comstock (1948) and Gertsch (1949), are applied to families known to occur in this temperate area. A listing of the families, genera and species which comprise the fauna of the island together with pertinent ecological data for Beaver Island is presented below. A summary table (Table 3) follows this listing. THE SPIDERS OF BEAVER ISLAND Class Arachnida Order Araneae Suborder Labidognatha (Araneomorphae) Section Ecribellatae Superfamily Epeiroidea Family Theridiidae—-Comb-footed Spiders Members of this family are usually associated with irregular or mesh webs placed in a variety of ecological situations. Achaearanae tepidariorum (C. L. Koch, 1841). Under logs and in abandoned buildings in swamp-conifer forest; in and on buildings in St. James and cottages along east coast a 22 Argyrodes trigona (Hentz, 1850). Shrubs and low tree branches of mixed hardwood-conifer forest. Crustulina stricta (O. P.-Cambridge, 1861). Juniper stands south of Font Lake. Dipoena nigra (Emerton, 1882). Low branches of coni- fers of mixed hardwood forest. Enoplognatha marmorata (Hentz, 1850). Under stones in gravel pits; on shrubs in upland hardwoods. Euryopsis funebris (Hentz, 1850). Leaf litter of up- land hardwoods; on sand under juniper bushes. Robertus laticeps (Keyserling, 1884). Litter of swamp-conifer forest. .3. riparius (Keyserling, 1886). Litter of upland hardwoods and mixed hardwood-conifer forests. Steatoda albomaculata (De Gear, 1778). Under debris at dump sites; under flat stones (2" to 6") behind fore- dune along northwest shoreline. ‘§. borealis (Hentz, 1850). Under bark and in litter of upland hardwoods; in rolls of birch bark in swamp-coni- fer forest. Theridion differens Emerton, 1882. In rock piles, juniper stands, and on shrubs of upland hardwoods. ‘1. frondeum Hentz, 1850. Low branches and shrubs of all forest types; vegetation of old fields. .1° glaucescens Becker, 1879. Vegetation of roadsides and old fields; cedar stands. T, murarium.Emerton, 1882. Juniper stands; shrubs of hardwood-conifer forests; cedars; on buildings where mature ‘males were occupying webs with immature females. ‘1. ornatum Hahn, 1831. Low branches and shrubs of upland hardwood-conifer forests; in crevices of stone wall. .1: petraeum.L. Koch, 1872. Rock piles; vegetation at meadow-cedar edge. 23 T. sexpunctatum.Emerton,1882. In rock piles shelter- ed by_ shrubs. Theridula emertoni Levi, 1954. On vegetation of clear- ings in upland hardwood forest. Family Linyphiidae--Sheet-web Weavers Most Species of this family construct rather conspic- uous dome-shaped or flat webs at various heights above ground level in standing vegetation. Some are found only in the litter on forest floors. Bathyphantes pallida (Banks, 1892). Leaf litter of _mixed hardwood-conifer forest. Centromerus cornupalpus (O. P.-Cambridge, 1875). Leaf litter of swamp-conifer forest. 9, latidens (Emerton, 1882). Leaf litter of mixed hardwood-conifer forest. 9. persoluta (O. P.-Cambridge, 1875). Leaf litter of swamp-conifer and mixed hardwood-conifer forests. Drapetisca alteranda Chamberlin, 1909. On webs close- 1y applied to bark of beech and sugar maple saplings. Frontinella communis (Hentz, 1850). (E. pyramitelal of Chamberlin and Ivie, 1944). Bowl-like web suspended in spaces of juniper bushes and bracken ferns. Helophora insignis (Blackwell, 1841). Flat webs in low Shrubs of upland hardwood forest. Lepthypantes appalachia Chamberlin and Ivie, 1944. Litter of swamp-conifer forest. L, nebulosa (Sundevall, 1830). ‘Webs in shrubs border- ing cleared fields and on juniper bushes. L. subalpina (Emerton, 1882). On webs in roadside vegetation and in rock piles. 1 Following the suggestion of Levi and Levi (1961) the name changes proposed by Chamberlin and Ivie (1944) are not used pending further study. 24 Linyphia marginata C. L. Koch, 1834. Under dome- shaped webs in roadside vegetation; on low shrubs in mixed hardwood-conifer forest; and cedars of swamp-conifer forest. L, waldea Chamberlin and Ivie, 1943. Under flat webs suspended in vegetation bordering lakes and marshes. Macrargus multesimus (O. P.-Cambridge, 1875). Leaf litter of hardwood-conifer forest. Microneta viaria (Blackwell, 1841). Leaf litter up- land hardwood forest. Pityohyphantes costatus (Hentz, 1850). Conspicuous flat webs on shrubs'and low tree branches of all forest types. Pusille mandibulata (Emerton, 1882). Flat webs sus- pended near bases of shrubs on sand beaches. Family Micryphantidae--Dwerf Spiders Most of the species belonging to this family are small, less than 2 mm., and live near the ground under vari- ous kinds of leaf litter. To avoid repetition in the following list, all species followed by a phrase describing a forest type were found in leaf litter. Exceptions to this will be described. Ceraticelus elticeps (Fox, 1891). Mixed hardwood-coni- fer forest. '9. atriceps (O. P.-Cambridge, 1874). Herbs of mixed hardwood-conifer forest. - .9. emertoni (O. P.-Cembridge, 1874). Vegetation of old fields. - ‘Q. fissiceps (0. P.-Cambridge, 1874). Shrubs of mixed hardwood-conifer forest. - g, laetebilis (O. P.-Cambridge, 1874). Herbs of apple orchards; shrubs and leaf litter of upland hardwood and hardwood-conifer forests. 25 g, minutus (Emerton, 1882). Upland hardwood and hard- wood-conifer forests. Ceratinella brunnea Emerton, 1882. Swamp-conifer and mixed hardwood-conifer forests. Ceratinopsis nigriceps Emerton, 1882. On webs at dis- tal ends of cedar and juniper branches. Cochlembolus pallidus (Emerton, 1882). Hardwood-coni- fer forest. Collinsia oxypaederotipus (Crosby, 1905). Upland hardwood forest. Corniculeria clavicornus Emerton, 1882. Hardwood- conifer forest. 9, indirecta (O. P.-Cambridge, 1874). Hardwood-coni- fer forest. 9, minute Emerton, 1882. Swamp-conifer forest. .9. pallida Emerton, 1882. Hardwood-conifer forest. Eperigone meculata (Banks, 1892). Upland hardwood forest./ E. trilobata Emerton, 1882. Shrubs and leaf litter of upland and hardwood-conifer forests. Erigone elsaide Crosby and Bishop, 1928. Leaves of Shrubs in upland hardwood forest at night. E. atra Blackwell, 1833. Upland hardwood forest. E. dentigera O. P.-Cambridge, 1874. Vegetation border- ing lake shores; herbs of upland hardwood forest. Floricomus plumalis (Crosby, 1905). Hardwood-conifer forest. Grammonota pictilis O. P.-Cambridge, 1875. On cedar trees and in leaf litter of swamp-conifer forest. ‘Q. spinimana Emerton, 1923. Hardwood-conifer forest. 26 Maso sundevalli (Westring, 1851). Swamp—conifer and hardwood-conifer forests. . Minyriolus castaneus (Emerton, 1882). Upland hardwood forest. Oreonetides vaginatus (Thorell, 1872). Mixed hardwood- conifer forest. Pelecopsis moestum.(Banks, 1892). Hardwood-conifer forest. Pocadicnmmis hertlandiana Emerton, 1913. Hardwood- conifer forest. P. pumile (Blackwell, 1841). Upland hardwood forest. Sciastes terrestris (Emerton, 1882). Hardwood-conifer forest. Sisicottus montanus (Emerton, 1882). Hardwood-conifer forest. Scironis tarsalis (Emerton, 1911). Mixed hardwood- conifer forest. S. montigenus Bishop and Crosby, 1938. Hardwood- conifer forest. Sisicus penifusiferus Bishop and Crosby, 1938. Hard- wood-conifer forest. Tapinocyba minute (Emerton, 1909). Hardwood-conifer forest. I. simplex (Emerton, 1882). Hardwood-conifer forest. Tunagyne debilis (Banks, 1892). Hardwood-conifer forest. Welckenaer vigilex (Blackwell, 1853). Vegetation of old fields. Family Epeiridae--Orb-web Weavers Some taxonomic schemes reduce this family to subfamily status under the family Argiopidae. I prefer the workable 27 arrangement proposed by Keston (1948) which gives full family status to Epeiridae and eliminates ArgiOpidee. The Species listed below all build orb webs of various size, complexity and inclination,though most are vertical or near vertical. Subfamily Argiopinae ArgiOpe trifesciate (Forskal, 1775). On vertical webs suspended in juniper bushes and standing vegetation of old fields. Subfamily Epeirinae Acanthepeira stellata (Welckenaer, 1805). Webs on vegetation of inland lake Shores and on shrubs and herbs of sandy beaches. Areneus diadematus Clerck, 1757. Webs in vegetation at old field-woods edge. ‘A. marmoreus Clerck, 1757. Shrubs of upland hardwoods, mixed hardwood-conifer forests and vegetation of old fields. A. nordmanni (Thorell, 1870). Shrubs of upland hard- wood forest; and beach. ‘ A. solitarius (Emerton,1884). Shrubs of upland hard- wood forest. ' 'A. thaddeus (Hentz, 1847). Shrubs of upland hardwood. forest. A. trifolium (Hentz, 1847). Herbs of open areas in up- land hardwood forest, old fields and beach Shrubs. Araniella displicata (Hentz, 1847). Vegetation of old fields; herbs and Shrubs of open woods. Cyclosa conica (Pallas, 1772). Shrubs of upland and mixed hardwood forests. Epeiracornute (Clerck, 1757). Shrubs of upland and mixed hardwood forests; beach shrubs. E, patagiate (Clerck, 1757). Herbs and shrubs of up- land and mixed hardwood forests; vegetation on beaches. 28 E. sericeta (Clerck, 1757). Trees and shrubs of up- land and mixed hardwood forests; frequently on buildings. Eustela anatera (Walckenaer, 1841). ‘Marsh vegetation; shrubs of upland hardwood forest. Larine borealis Banks, 1894. Vegetation of open areas in upland hardwood forest, junipers and grassy fields. Mangoragibberosa (Hentz, 1847). Vegetation of aban- doned orchards. Neoscone arabesce (Welckenaer, 1837). Shrubs and herbs of upland hardwood forest; junipers and old fields. Zygiella nearctica Gertsch, 1964. Low branches of trees and shrubs of mixed hardwood-conifer forests; herbs and shrubs of beach along eastern Shore line; on buildings. Family Tetragnathidae These species also spin orb webs which.may vary from near horizontal to vertical in position. Subfamily Tetragnathinae Leucauge venusta (Walckenaer, 1837). Shrubs and herbs of upland hardwood forest. Pachygnatha kuratai Levi, 1951. Vegetation bordering inland lakes and marshes. Tetragnatha elongate Walckenaer, 1805. Shrubs and herbs of hardwood forest; on cedar trees along stream course; vegetation of beaches. .1: extensa Linnaeus, 1757. Vegetation around build- ings; shrubs of upland hardwood forest. 2, guatemelensis O. P.-Cambridge, 1889. Shrubs and low tree branches of hardwood-conifer forests; beech shrubs. T. harrodi Levi, 1951. Cedar trees; vegetation of clearings in upland hardwood forest; vegetation on beach. T. laboriosa Hentz, 1850. Shrubs of upland hardwood forest; vegetation of old fields, beaches and roadsides. 29 I. pellescens F. O. P.-Cambridge, 1903. Juniper stands near open water and beach vegetation. .1: versicolor Welckenaer, 1841. On cedars bordering inland lakes and Lake Michigan; junipers and grass of open fields. Family Mimetidae--Pirate Spiders These Species are reported to prey exclusively on other spiders. Ero leonina (Hentz, 1850). Leaf litter of mixed hardwood-conifer forests. Mimetus epeirodes Emerton, 1882. Vegetation of old fields and under debris in dumps. Superfamily Lycosoidea Family Agelenidee--Funnel-web Weavers 'Most species of this group inhabit moist situations such as under leaf litter, under loose bark of logs and stumps in forests; others place their sheet webs at vari- ous heights above ground level in standing vegetation. Agelenopsis emertoni Chamberlin and Ivie, 1935. Vege- tation of old fields. ‘A. pensylvanica (C. L. Koch, 1843). On and near build- ings in St. James. .A. potteri (Blackwell, 1846). Shrubs of mixed hardwood- conifer and upland hardwood forests; in rock piles; and on buildings. ‘A. utahana (Chamberlin and Ivie, 1933). Shrubs and hollow stumps of upland hardwood forest; leaf litter of upland hardwood and hardwood-conifer forests. Cicurina arcuate Keyserling, 1887. Leaf litter of up- land hardwood and hardwood-conifer forests. 30 C. brevis (Emerton, 1889). Leaf litter of upland hardwood and mixed hardwood-conifer forests; vegetation of roadsides. C. lowriei Levi, 1951. Shrubs and herbs of mixed hardwood-conifer forest. Cores montanus (Emerton, 1889). In the leaf litter and hollow stumps of upland hardwood forest. Cryphoeca montana Emerton, 1909. Litter of swamp- conifer forest. Wedotes calcaratus (Keyserling, 1887). In brush piles and leaf litter of upland hardwood and mixed hardwood-coni- fer forests; under debris on beach along the west side of the island. \ Family Hahniidae These spiders Spin delicate small sheet webs in moist situations, especially in forest litter. Neoantistea agilis (Keyserling, 1887). Leaf litter of upland hardwood and mixed hardwood-conifer forests. ‘N. riparia (Keyserling, 1887). Leaf litter of upland hardwood forest. Family Pisauridae--Fisher Spiders or Nursery-web Weavers The species of this family do not build snares for capturing prey but rather are active hunters. ‘Most are associated with.moist habitats often near or on Open water. Dolomedes scopuleris C. L. Koch, 1848. Under vegeta- tive debris along shores of inland lakes and along sandy beaches on east coast. 2, scriptus Hentz, 1845. In shallow depressions under logs along shore line of Fox Lake. 2. tenebrosus Hentz, 1844. Under vegetative debris along shore line and on vegetation bordering shore line. 31 Family Lycosidae--Wolf Spiders Most members of this family are active hunters that pursue their prey. However, some are sedentary and build more or less vertical burrows that may be topped by turrets of vegetative debris. Subfamily Lycosinae Arctosa littorelis (Hentz, 1844). Under debris on sandy beaches during the day or running along water's edge at night. Geolycosa missouriensis (Banks, 1895). In vertical burrows with turrets in soils of mixed hardwood-conifer forests and juniper-grass fields. The deepest burrow was 11 inches; the average depth of six burrows was 8-3/4 in- ches. IQ. wrightii (Emerton, 1912). In vertical burrows of varying depths in sandy areas of beaches. The depths of the burrows seem to vary seasonally: shallow (9" to 15") in early summer; deeper (18" to 27") in the fall. Lycosa avide (Walckenaer, 1837). Under beach debris along northwest rocky shore line. L. beltimoriana (Keyserling, 1876). Under debris on sand beaches and wandering on soil surface under shrubs behind foredunes. L, frondicola Emerton, 1885. Leaf litter of upland hardwood forest. L. helluo walckenaer, 1837. Under debris on sandy beaches; in leaf litter and under stones in upland hardwood forest. L, pratensis Emerton, 1885. In upland hardwood forest litter; in crevices between cobblestones on beaches. Pirate arenicola Emerton, 1909. Under debris and on vegetation along shore lines of inland lakes. P. marxii Stone, 1890. Leaf litter of mixed hardwood- conifer forest. - 32 P. 'minutus Emerton,1885. In rock piles and in litter at old field-wood' s edge. P. montanus Emerton,1885. In sunflecks on leaf litter of upland hardwood forest. .2: pirate (Clerck, 1757). Between pebbles and rocks of beach along southeast shore line; under debris along west shore line. Tarentule eculeata (Clerck, 1757). On leaf litter of mixed hardwood-conifer forests; under beach debris; and in Open grassy fields. Subfamily’Pardosinae Pardosa distincta (Blackwell, 1846). In rock piles; leaf litter of mixed hardwood-conifer forests; vegetation along shores of inland lakes; under vegetative debris on sand beaches. P. fuscula (Thorell, 1872). Under debris on sand beaches. “P. lapidicina Emerton, 1885. On and between pebbles and rocks along Lake Michigan shore line. P. mackenziana (Keyserling, 1876). On shrubs of up- land hardwood forest, junipers, cedars and vegetation of roadsides. . P. milvine (Hentz, 1844). Old fields and shrubs of upland hardwood forest. 2. modica (Blackwell, 1846). Marsh vegetation. P. moesta (Banks, 1892)° Grass of old fields and shore lines- of inland lakes. P. saxatilis Hentz, 1844. Grassy clearings in upland hardwood forests and along shore lines of inland lakes; rock piles; soil surfaces of old fields. ‘P. xerampelina (Keyserling, 1876). In rock piles; on stony beaches; on roadside vegetation-and in litter of up- land herdwood forest. 33 Superfamily'Gnephosoidea Family Gnaphosidae--Running Spiders These species are primarily nocturnal, hunting spi- ders. During the day they may be found in shallow depres- sions or crevices or under stones, logs and debris. Subfamily Gnaphosinae Gnaphosa muscorum (L. Koch, 1866). Under stones and logs along shores of inland lakes; soil surface of old fields. Subfamily Drassodinae Drassodes neglectus (Keyserling, 1887). In rock piles; under logs of mixed hardwood-conifer forests; soil surface of old fields. Drassyllus niger (Banks, 1911). Leaf litter of up- land hardwood forest. Haplodressus signifer (C. L. Koch, 1839). In debris under cedar trees. Herpyllus vasifer (Welckenaer, 1805). In leaf litter of mixed hardwood-conifer forests; under stones in old fields. Zelotes subterraneus (C. L. Koch, 1839). Under de- bris in dumps; in rock piles; on vegetation of roadsides; and on shrubs of upland hardwood forest. Superfamily Clubionoidea Family Clubionidae--Sack Spiders Many of these hunting spiders may be found in silken retreats in rolled leaves of shrubs and weeds. Others are found under stones, boards and in leaf litter of several forest types. Subfamily Clubioninae Clubiona canadensis Emerton, 1890. On cedars; under loose bark; and on shrubs of swamp-conifer forest. 34 .Q. Chippewa Gertsch, 1941. On shrubs of upland hard- wood forest. .9. mixta Emerton, 1890. On beach vegetation along western shore line. ‘9. moesta Banks, 1876. On Shrubs of upland hardwood forest. C. obesa Hentz, 1847. On shrubs of upland hardwood forest; under debris and logs on beaches of inland lakes and Lake Michigan. ‘9. pallens Hentz, 1847. Shrubs of hardwood-conifer forests. lg. riparia L. Koch, 1866. On Shrubs of upland hard- wood forest; under stones behind foredune along western shore line. ‘Q. trivialis C. L. Koch, 1843. On cedars. Subfamily Liocraninee Agroeca ornate Banks, 1892. In leaf litter of upland hardwood and swamp-conifer forests. ‘A. pratensis Emerton, 1890. On shrubs and in leaf lit- ter of upland hardwood forest; in leaf litter of mixed hard- wood-conifer forest, Phrurotimpus alerius (Hentz, 1847). Leaf litter of upland and mixed hardwood forests. '2. borealis (Emerton, 1911). Leaf litter of mixed hardwood-conifer forest. Subfamily Micariinae Castianeira cinguleta (C. L. Koch, 1842). In leaf litter of mixed hardwood-conifer forests; under beach debris along northern portion of eastern shore line. C. descripta (Hentz, 1847). In litter and under stones of upland hardwood forest, under debris and amongst rocks of beaches surrounding island. Miceria emertoni Gertsch, 1935. In leaf litter of up- land herdwood forest. 35 M,.1ongipes Emerton,1890. On beach vegetation and under stones behind foredune along sandy western shore line. Superfamily Thomisoidea Family Thomisidae--Crab Spiders The spiders of this family capture their prey by either lying in ambush or wandering about. They do not makemwebs. Subfamily Misumeninae Misumena vatia (Clerck, 1757). Vegetation of old fields and beaches; on juniper bushes. ‘Misumenops oblongus (Keyserling, 1880). Vegetation of old fields. Oxyptila americana Banks, 1895. Leaf litter and de- bris of all forest types; relatively more common in decid- uous litter. - Xysticus elegans Keyserling, 1880. Vegetation bor- dering marshes and grass-juniper fields; litter and shrubs of upland hardwood forest. .K. ferox (Hentz, 1847). Under debris in dumps and on juniper bushes. g, gulosa Keyserling, 1880. Vegetation of old fields and under stones at field-wood's edge. 3;. luctans (c. L. Koch, 1845). Old fields and grassy juniper stands. ‘5. pellax O. P.-Cambridge, 1894. On juniper bushes. g, triguttatus Keyserling, 1880. Vegetation of old fields. Subfamily Philodrominae Ebo latithorax Keyserling, 1883. On beach vegetation. Philodromus cespiticolus (Walckenaer, 1802). Shrubs of upland hardwood forest; on cedars and junipers; vegeta- tion of old fields and beaches. 36 .P. imbecillus Keyserling, 1880. Vegetation of old fields; shrubs and herbs of mixed hardwood-conifer forest. P. infuscatus Keyserling, 1880. Vegetation of old fields; sandy surface and vegetation of beach along western shore line. .3. pernix Blackwall, 1846. Leaves of shrubs, bark of trees in upland hardwood forest; on cedar trees. 2. placidus Banks, 1892. On juniper bushes. P. rufus‘Walckenaer, 1825. Shrubs and herbs of up- land hardwood forest; on cedar trees; vegetation of old fields and roadsides. Thanatus formicus (Clerck, 1757). Vegetation of old fields; grassy Openings in mixed hardwood-conifer forest; on clumps of beach grass. Tibellus maritimus (Menge, 1874). Roadside vegeta- tion and juniper bushes. T. oblongus (Walckenaer,1802). Shrubs of upland hardwood and mixed hardwood-conifer forests; junipers; and vegetation of old fields. Superfamily Salticoidea Family Salticidae--Jumping Spiders These hunting spiders are reported to have the keen- est vision of all spiders. They hunt in broad daylight capturing their prey by stalking slowly and jumping quick- ly. Evarcha hoyi (Peckham, 1883). Vegetation of old fields; grassy openings in upland hardwood forest. Habrocestum pulex (Hentz 1846). On and in leaf litter of upland hardwood forest; in rock—piles of old fields. Habronattus agilis (Banks, 1893). Cedar trees and vegetation of old fields. .5. borealis (Banks, 1895). Herbaceous beach vegeta- tion along southeastern shore line; old fields near Open water. 37 ‘fl. calcaratus (Banks, 1904). Under stones in old fields. Icius elegans (Hentz, 1846). Vegetation of old fields and field-wood's edge. 1. similis Banks, 1895. Shrubs and herbs of upland hardwood forest. Metacyrba undata (De Geer, 1778). Under debris and on vegetation of old fields; on buildings. Metaphidippus flavipedes (Peckham, 1881). Most fre- quently on cedar trees; shrubs and herbs of upland and mixed hardwood forests; vegetation of most beaches along shore line. gs protervus (Welckenaer, 1837). Shrubs of upland and mixed hardwood-conifer forests; vegetation.of marshes and old fields. Neon nellii Peckham, 1888. Leaf litter of all forest types. Paraphidippus marginatus (Walckenaer, 1837). Shrubs and herbs of all forest types; vegetation of shore lines. Phidippus clarus Keyserling, 1884. Shrubs and herbs of upland hardwood forest; vegetation of clearings in old fields and flower gardens; under debris in dumps. '2. purpuratus Keyserling, 1884. Under stump frag- ments, bark, stones and boards in forests and fields. I Salticus.scenicus (Clerck, 1757). Shrubs of upland hardwood forest; vegetation of old fields, beaches and flower gardens; in and on buildings. ' Sitticus palustris (Peckham, 1883). Vegetation of old fields; shores of7in1and lakes and northern shore lines. 38 Section Cribellatae Superfamily Dictynoidea Family Dictynidae--Mesh-web Spiders Most of these species build small dense webs at the tips of dried weeds, at the distal ends of shrub and coni- fer branches, or under the rolled edges of leaves. Some are found only in.1eaf litter. Dictyga annulipes (Blackwall, 1846).' On grasses in clearings of upland hardwood forest; on beach vegetation. ‘2. bostoniensis Emerton, 1888. On cedar trees and herbaceous vegetation near open water; old fields. D. brevitarsus Emerton, 1915. Leaf litter of mixed hardwood-conifer forests. ‘2. coloradensis Chamberlin, 1919. On beach vegeta- tion; on junipers; vegetation of old fields and roadsides. Q, foliacea (Hentz, 1850). In grassy areas of old fields. D. maxima Banks, 1892. Under leaves of shrubs in upland- ~hardwood forest. 2. minuta Emerton, 1888. On cedar trees. D. sancta Gertsch, 1946. Herbaceous vegetation of beach_ along northeastern shoreline. D. sublata (Hentz, 1850). On cedars, shrubs and herbs of upland hardwood forest. D. volucripes Keyserling, 1881. Shrubs and herbs of upland hardwood forest; vegetation of grassy fields. Lathys foxii Marx, 1891. Leaf litter of all forest types. 6 - L, pallida (Marx, 1891). Leaf litter of all forest types. 39 Family Uloboridae--Hack1ed-band Orb Weavers Hyptiotes cavatus (Hentz, 1847). Low tree branches and shrubs of mixed hardwood-conifer forest. Family Amaurobiidae-~Hack1ed-band Weavers Subfamily Amaurobiinae Callobius bennetti (Blackwall, 1846). Under logs and in leaf litter of all forest types; in rock piles; on cedar trees. Subfamily Ixeuticinae Titanoeca americana Emerton, 1888. Leaf litter of mixed hardwood-conifer forests. walmus borealis (Emerton, 1909). Leaf litter of up- land and mixed hardwood forests. 40 TABLE 3. Summary of the spiders of Beaver Island Families Genera Species Theridiidae 10 18 Linyphiidae 11 16 Micryphantidae 22 37 Epeiridae 11 18 Tetragnathidae 3 9 Mimetidae 2 2 Agelenidae 5 10 Hahniidae 1 2 Pisauridae l 3 Lycosidae 6 23 Gnaphosidae 6 6 Clubionidae - 5 l6 Thomisidae 8 19 Salticidae 11 16 Dictynidae 2 12 Uloboridae 1 l Amaurobiidae 3 3 Total: 17 108 211 41 The number of species known to occur on the island is approximately one-half that which occurs on the main- land of Wisconsin or Michigan (Tables 4 and 5). This re- duced number not only reflects the absence of many species but of families as well. Of these, the absence of Oxyopi- dae and Anyphaenidae should be noted for representatives of these families are relatively abundant in specific ecological situations. Muma and Muma (1949), Whitcomb £5..§l., (1963) and Brady (1964) report species of OxyOpi- dae to be very abundant in old fields and grassy areas. Of the two oxyopid species occurring in Michigan (Drew, in preparation), Oxopes scalaris is fairly common in areas of the Southern Peninsula that are ecologically similar and geographically near the island. Therefore, its ab- sence is that much.more noticeable. The lack of repre- sentatives of the family Anyphaenidae is of equal interest for these are also relatively abundant (Lowrie, 1948). Here, however, we may be confronted with geographical distribution entirely, because the four species recorded for Michigan have been collected from southern counties only (Drew, gp. ‘giE.). This same distributional pattern applies also to Wisconsin (Levi and Field, 1954; and Levi, Levi and Kaspar, 1959). The absence of representatives of eight other families (Atypidae, Oon0pidae, Dysderidae, Scytodidae, Pholcidae, 42 Table 4. ‘Web-weaving families of Beaver Island, Michigan and‘Wisconsin 1 Families Beaver Island ‘Michigan Wisconsin Pholcidae - 1,12 2,2 Theridiidae 10,18 15,43 14,41 Nesticidae - - 1,1 Linyphiidae 11,16 14,23 16,33 Micryphantidae 22,37 30,56 37,67 Epeiridae 11,18 17,36 20,43 Theridiosomatidae - 1,1 1,1 Tetragnathidae 3,9 3,15 4,14 Oecobiidae - 1,1 - Dictynidae 1,1 3,25 4,15 Uloboridae 1,1 2,3 2,2 Amaurobiidae 3,3 3,4 3,3 63,114 89,208 104,222 54.1% 51% 51.2% 1 Modified from.Levi and Field (1954) and Levi, Levi, and Kaspar (1958). 2 The number preceeding the comma is the number of genera; that following the comma is the number of species. 43 Table 5. Hunting families of Beaver Island, Michigan and Wisconsin W Hunting families Beaver Island 'Michigan Wisconsin Atypidae - 1,1 1,1 Oonopidae - 1,1 - Dysderidae - - 1,1 Scytodidae - 1,1 1,1 Mimetidae 2,2 2,5 2,5 Agelenidae 5,10 5,15 6,15 Hahniidae 1,2 3,4 3,4 Pisauridae 1,3 2,7 2,5 Lycosidae 6,23 7,35 8,36 Oxyopidae - 1,2 1,2 Gnaphosidae 6,6 11,21 12,28 Clubionidae 5,16 8,29 7,28 Anyphaenidae - 3,4 2,3 Thomisidae 8,19 12,42 11,35 Salticidae 11,16 23,49 ' 28,48 45,97 80,216 85,212 45.9% 49.0% 48.8% Totall 108,211 169,426 189,434 1 Figures are total number of genera and species for each geographical area compared in Tables 4 and 5. 44 Nesticidae, Theridiosomatidae and Oecobiidae) (Tables 4 and 5) will be discussed below. Two species which are frequently found on the sur- rounding mainlands were not collected on the island. These are: Argiope aurantia, the banded orb-weaving gar- den spider; and Phidippus audax, a usually common and most conspicuous jumping spider. Despite repeated collecting attempts throughout the growing season, neither mature nor immature forms of these two species were found. Spider Populations of Specific Communities Old Fields. The spider populations representative of this community inhabit the upper portions of standing vege- tation. In order to learn more about diurnal and seasonal succession in the old field community, collecting was done during the morning and afternOon hours on the same day in July; and in July and September. On 13 July 1963, six sweeping samples were taken dur- ing two different periods of the day: three at 10:00 A.M. and three at 2:00 P.M. The afternoon samples were taken from the same areas as those in the morning. Of the 14 species collected, six were present in both samples, five unique to the morning samples and three unique to the afternoon samples (Fig. 5). Each series of samples had one species that was not collected in subsequent quantitative 10:00 A.M, 2:00 P.M. 2 Frequency 7. Frequency § 2 8 e 9. : s a :- 8 1 L 1 1 ‘3: 3‘ 1 L 1 1 9 h «‘6 :k" \ Tetragnatha laboriosa Diotyna volucripes J Phidippus olargg Neoecoy_a_ agabeggg L Dictyna foliacg 1:: Dictgne coloradensis ] r ALIS—1119222313 L Philodromus ruins r——-“ Xxeticus triguttatus 1 :::1: W098 gaperfiue Metaphidippus flavipedes C__,_m Araniella dieplicm _Icius el_ez.a_n.s . Theridion glaucegcens Jhl Figure 5. Frequency-density diagrams of the spider papulationa at different times on the same July day in the old field community. 46 collecting in this community. They were Metaphidippus flavipedes in the morning sample and Dictyna coloradensis in the afternoon sample. Tetragnatha laboriosa was the most abundant species in both series of samples. It had a frequency of 100% in both series; but a density of 4.0 in the A.M., and a density of 5.3 in the P.Mgl In the morning samples, Dictyna volucripes, Phidippus clarus and Neoscona arabesca, in this order, followed 1. laboriosa in abundance. In the afternoon series 3. arabesca fol- lowed 1. laboriosa at the same value of abundance as in the morning. .2. clarus, at a somewhat lesser figure, was next most abundant and replaced 2. volucripes in order of abundance. The abundance of the latter species was great- ly reduced and was equaled in abundance by a congenic species, 2. coloradensis, which was not present in the morning series. The first series of samples evaluating seasonal suc- cession was taken during morning hours in July 1963. Nine sweeping samples gave evidence that 26 Species were present. in this community (Fig. 6). The three most abundant species had densities greater than the remaining 23. Thirteen of thei26 species were web-weavers, including two of the most abundant. Neoscona arabesca the most abundant species had a frequency of 100% and a density of 5.1. Tetragnatha 47 Figure 6. Frequency-density diagram of the spider populations in the old field community, July 1963. 7. Frequency 00/ s s l l r 09 08 f) Q’s-u: 8' Neoscona arabesca TetraLgnatha laboriosa Phidippus clarus ._J Dictyna foliacea Misumena vatia —I Dictyna volucripes Philodromus rufus 4. Xysticus triguttatus Dictyna oolo radens is , l l Epeira sp. H Philodromus crespiticolus Misumenops asperatus Evarcha hoyi Metaphidippus flavipedes Metaphidippus protervus Tibellus oblonflg Theridion glaucescens '£h_:er1dion irondeum Frontinella communis Walckenaera vigilax Ceraticelgg emertoni. Araniella displioata Tetragnatha versicolor Thanatus" formiciggg Icius slogans Sitticus palustris TIIIIIIHTUUUhJ l 48 laboriosa had a greater density, 5.6, but a frequency of 77.8%. The two most abundant hunting species, Phidippus clarus and Misumena XEEEE followed but in greatly reduced abundance. A series of five sweeping samples taken in the morn- ings of the first week in September 1963, contained a total of 14 species (Fig. 7).) A considerable reduction in abundance as compared with the July samplings is appar~ ent. Also, there were present six species not collected in July. '3. laboriosa was most abundant with a frequency of 60% and a density of 4.6. It was followed by Philodro- 'mu§ crespiticolus, lgigg elegans, Dictyna bostoniensis and Acanthepeira stellata at much lower values. Of these last four species, only.l. elegans was present in July (Fig. 6), and then at a much lower abundance. Dictyna volucripes V was present but replaced by Q. bostoniensis in abundance. ‘E. arabesca was present but two other web-building species, Acanthepeira stellata and Araniella displicata were each more abundant. Of the 14 species collected in this series, seven were web-building Species and seven were hunting Spe- cies. Upland Hardwood Forest (BeechrMaple). Two strata of this community were sampled during August 1963. Eight beating samples of the herb-shrub stratum and 36 square- % Frequency 0\ Q a? 0/; 06 [flu/3,0 2' I Tetragnatha laboriosa Philodromus crespiticolus Icius elegans Dictyna‘bostoniensis Acanthepeira stellata Misumena vatia Araniella displicata Diotyna volucripes Neoscona arabesca Mimetus epeiroides Tibellus oblong__us Misumenops as‘peratus Xysticus gulosa Habronattus bo re 9.118 Figure 7. Frequency—density diagram of the spider populations in the old field community, September 1963. 00/ 50 foot samples of the leaf litter stratum were collected. Because two sampling techniques were used in these two distinct habitats, the results are not comparable. How- ever, together they reveal the population structure of this community. The results of the eight beating samples show a total of 17 species inhabiting the herb-shrub Stratum (Fig. 8A); eleven of these are web-building species and six hunting species. Pityohyphantes costatus was most abundant with a frequency of 50% and a density of 4.5. The hunting species, Paraphidippus marginatus was next most abundant followed by the web-building species Helophora insignis. The results of the 36 square-foot samples of leaf litter contained eleven Species, seven web-building and four hunting species (Fig. 8B). Of the 36 Samples taken; 17 did not contain Spiders; 12 contained one specimen each; six contained two specimens (not of the same species in each case); and one contained four specimens of four different species. Oxyptila americana was most abundant witha frequency of 16.7% and a density of 0.2. walmus borealis and Phrurotimpus alarius followed in equal but lower numbers. Juniper Bushes, (Juniperus depressus). During July 1963, a series of three beating samples were taken in Oxyptila americana 7. Frequency x. Q l l _ baa ) Walmus borealis Phrurotimpus alarius Eperigone trilobata Pooadionemig mil; Robertus riparius Neoantistea riparia Clubiona riparia Eperigone maoulata Cicurina brevis man—elm Leaf Litter Stratum B UUUUUUUUUU 2° [’0 [Int is P E00/ {7. J P ityohyphantes costatus Paraphidippus marginatu Helojhora insignia Neoscona arabggg; Philodromus orespitioolus Meta‘phidijpus protervus Clubiona mu Araniella displioata Arms: solita i Tetragnatha laboriosa Iberidmm Tux-azimuth mugging Arline—us mums. Acanthepeira 1111.11.11 _Agelenops is utahana Xystious slogans lily. aimilia Eustala anestera lllHU 1 Frequency 2883‘ All 2 T. 9/3/va _J__J ”“UHHUJUU Herb-shrub Stratum A Figure 8'. Frequency-density diagrams of spider populations inhabiting the herb-shrub and leaf litter strata of the upland-hardwood forest, August 1963. 52 juniper clumps (3' to 5' high and 5' to 12' in diameter) to ascertain the spider populations of this community. Eleven of the 15 species collected were web-building spe- cies, the remaining four were hunting Species (Fig. 9). However, the combined densities of the four hunting Spe- cies, 11.3, nearly equals that of the eleven web-building species, 13.5. Metaphidippus flavipedes was the moSt abundant with a frequency of 100% and a density of 8.0. Neoscona arabesca followed with the same frequency and a density of 4.7. The abundance of the remaining 13 Species was considerably lower than these two Species. White Cedars, (Thuja occidentalis). In late July 1963, a series of three beating samples were taken in this conifer community. Two of the three most abundant Species collected were hunting spiders, Metaphidippus flavipedes and Philodromus imbecillus (Fig. 10). Their combined densities, 6.3, was slightly less than that of Grammonota pictilis the most abundant Species with a denSity of 6.7 and a frequency of 100%. Among the 12 species collected, seven were hunting and five were web-building species. The Spider P0pu1ations of a Similar Habitat on the Island and on the Mainland of Michigan This investigation was done to ascertain the simi- larities and dissimilarities of the spider populations 1...“. k s % Frequency 00/ m ”é o " i l l h03' {yum}: 6' l s_:._a Metaphidippus flavipedes Neoscona arabesca Theridiqg murarium Clubiona p_allens Philodromus placidus Grammonata pictilis Tetraflatha guatemalensisl‘ Philodromus orespitioolus‘ Tibellus oblong) up Diotyna bostoniensis Tetrmatha laboriosg Tetragnatha versicolor Araniella displioata mutual“. communis Linyphia marginata UHHUUHHUU Figure 9'. Frequency-density diagram of the spider populations inhabiting jlmipers (Juniperus depressus). July 1963. 7. Frequency g» g 8‘ °o° 8 l l l p i 1 a N is s Grammonata pictilis Metaphidijpus flavipedes Ehilodromus imbecillus Habronattus agilis fl Tetragnatha guatemelensis Tetragnatha laboriosa Philodromus crespiticolus Paraphidippus marginatus : Pityohyphantes costatus 1:—_= Pardosa mackenziana =2 Tibellus oblongus 5:, Theridio; murarium ;:__-—_, Figure 10 . Frequency-density diagram of the spider populations inhabiting northern white cedars. August 1963. 55 inhabiting a specific ecological community occurring on the island and on the mainland. The comparison was made in terms of the number of individuals per species, the number of species in each community, and making use of these figures, the calculation of Fisher's Index of Divers- ity. This study was conducted during September 1963 by analyzing 110 square-foot samples of the leaf litter stra- tum of two stands of mixed hardwood-conifer forest: 55 samples from the northeast portion of the island; and 55 from a Similar stand, 19 miles over water to the east, on waugoshance Point in Emmet County. The sites studied have a number of sflmilarities, other than forest cover, and fewer dissimilarities. The readily apparent Similarities and their implications are: both are located at 45°45' north latitude and back of the shoreline, therefore, each is subjected to the same over- all meteorological factors; both are generally level and vary in elevation from two to five feet above lake level, thus each has been subjected equally to the recent fluctu- ations of Lake Michigan; both have the same soil type (Eastport), therefore the same general drainage features; both were submerged by Lake Algonquin, were part of the exposed land mass during the Lake Chippewa Stage and were 56 submerged once again by Lakes Nipissing and Algoma, there- fore, both sites have had equal time for vegetational and soil development; and both stands border and gradually merge, on their landward sides, into oak-pine forests. Over and above the fact that one site is on an island and the other is part of the mainland, the most conspicuous difference is that the island site is located on an east- facing coast while that on the mainland is located on a west-facing coast. No definite phytosociological analysis was made in either stand. However, the major dominant trees present are: balsam fir (Abigs balsamea), white spruce (£1293 glauca), northern white cedar (Thuia occidentaliS), white birch (Betula papyrifera), quaking aspen (PoEulus Eggmgy loides), bigtooth aspen (F. grandidentata). Scattered amongst these species are: white pine (Pigus strobus), red pine (P. resinosa), hemlock (Tsuga canadensis), and red oak (Quercusiggbga). Both stands are considered to be more closed than open. That is, the conifer Species are of such abundance that even though the deciduous trees may be bare of leaves, the forest still has the appearance of being in leaf-out condition during the entire year. As a consequence, the litter layer of this forest type is never exposed to wind penetration as is that of a homogeneous H‘ IIIIIImH 57 deciduous forests. Also, the litter layer is in relative- ly continuous shade. Both these factors, reduced wind penetration and abundant shade, would be conducive to restricting the spiders to the litter and preventing, or at least greatly reducing aerial dispersal of the Species present. For, as Duffey (1956) concluded, it takes a com- bination of rising temperatures which induce aeronautic behavior and winds of sufficient velocity to carry the ex- posed spiders to new or less densely populated habitats. It is assumed, therefore, that these populations present in this stratum.of mixed hardwood-conifer forest are con- fined populations. The leaf 1itter, or L layer, habitat of this forest is in part Type I Curled and in part Type II Needle ac- cording to classification by Heatwole (1961). That is the interstices, which are of importance to the resident fauna, vary from fairly large due to the curled or rolled decid- uous leaves to small and angular in the conifer litter. Further, because the samples were collected in September before leaf fall, the litter had been subjected to the chemical and physical processes of humification for at least a year. Sampling of the leaf litter habitat was done randomly eliminating the preference of either of the two litter 58 types. This sampling method satisfied the major criterion for using Fisher's Index of Diversity in that the speci- mens, and not the Samples, must be randomly selected. The assumption that the spiders in this specific community are randomly distributed is based on the conclusion put forth by Cole (1946), who found that of the cryptozoa distributed under various Sized boards placed on a forest floor, only spiders displayed a random distribution. The results revealed that a total of 78 Species of 52 genera representing 15 families inhabited this leaf lit- ter community. Table 6 contains a listing of these Species together with the number of individuals per species for each of the two sites. Of the 78 species present, 25 (32.1%) were common to both areas, 34 (43.6%) were present only in the island study site, and 19 (24.3%) occurred only On the mainland site. Further summation of the data is contained in Table 7. There were 14 families represented by 59 species and 424 specimens in the island's community, and 13 families re- presented by 44 Species and 162 individuals in the main- land site. Categorizedin terms of food-getting habits, 39 or 66.1% of the island's 59 Species and 25 or 56.8% of the mainland's 44 species were web-weavers; 20 or 33.9% of those on the island and 19 or 43.2% of those on the 59 Table 6. Species inhabiting leaf litter of mixed hardwood- conifer forest on BeaVer Island and on waugoshance Point Number of Individuals Species Beaver Island Mainland Theridiidae Euryopis argentea - Paidisca unimaculata Robertus laticeps , .§.;riparius 2 Theridion frondeum - T. ornatum l. sexpunctatum CNN 'UJl-‘Ni N Linyphiidae Centromerus latidens .9, persoluta Helophora insignis Lepthyphantes nebulosa L. zebra NHi-‘QCh HI NH! Micryphantidae Ceraticelus alticeps ‘Q. communis '9. fissiceps .9. laetabilis g. minutus .9. rugosa. Ceratinella brunnea Cochlembolus pallidus Corniculeria clavicornis Q, indirecta Q. minuta .9. pallida Eperigone'maculata ‘g. trilobata Floricomus plumalis Maso sundevalli Pelecopsis moestwm Pocadicnemis hartlandiana Sciates truncatus Scirites pectinatus Sisicottus‘montanus .§° monitgenus HH Ill-'45! lNl—‘l '—-I lNl—‘lbl H nan): $‘BJUIQ>UJFJBJUJI UJNJU1P‘I P‘UIhilld I IND—'3 60 Table 6. (continued) Species Number of Individuals Beaver Island Mainland 'Micryphantidae (cont'd.) Sisicus penifusiferus TapinOcyba'minuta _T. simplex Tunagyna debilis Epeiridae Epeira patagiata Mimetidae Ero leonina Tetragnathidae Tetragnatha harrodi ‘1. laboriosa Pachygnatha kuratai Agelenidae Agelenopsis utahana Cicurina arcuate Q. brevis .9. robusta Cryphoeca‘montana Hahniidae Hahnia cinerea Neoantistea riparia Lycosidae Lycosa gulosa L, pratensis Pardosa milvina Pirata'marxii Schizocosa crassipes Gnaphosidae Zelotes subterraneus updraraua IHIND HN DON HUNNI 61 Table 6. (continued) Species Number of Individuals Beaver Island Mainland Clubionidae Agroeca ornata Clubiona kastoni ‘9. pallens ‘Q. riparia Phrurotimpus alarius “P. borealis Scotinella divestus Thomisidae Oxyptila americana Philodromus aureolus Thanatus coloradensis Tibellus oblongus Xysticus elegans Salticidae Neon nellii ParaphidippuS'marginatus Dictynidae Dictypa bostoniensis '2. brevitarsus D. coloradensis Lathys foxii L. pallida AmaurObiidae Callobius bennettii Titanoica americana Walmus borealis 13 30 42 424 HJ-‘boowmn 3 \l NIHI H9 162 62 Table 7. Comparisons of the spider populations inhabiting leaf litter Beaver Island Mainland No. of No. of No. of No. of Families species individs. species individs. Salticidae 2 30 2 ' 5 Thomisidae 4 52 3 40 Clubionidae 3 15 6 22 Gnaphosidae 1 3 - - Lycosidae 4 6 1 3 5 Hahniidae 2 15 2 5 Agelenidae 3 18 3 7 Tetragnathidae 1 9 2 3 Mimetidae 1 l - - Epeiridae - - l 1 Micryphantidae 22 134 10 24 Linyphiidae 5 19 3 4 Theridiidae 4 26 4 9 Dictynidae 4 50 3 26 Amaurobiidae 3 46 2 11 Totals 59 424 44 162 Web-weaving Species 39 (66.1%) 25 (56.8%) Hunting species 20 (33.9%) 19 (43.2%) Index of ' Diversity 18.62 (17.07-20.17) 19.88 (17.20-22.36) Index of Diversity for combined Samples: 17.50 (16.53-18.47) 63 mainland were hunting species. These data indicate the mainland populations of web-weaving and hunting species are much more nearly equal in number. On the island, the number of web-weaving species dominates the hunting spe- cies nearly 2 to 1. In addition, the total number of individuals from the island sample is 2.62 times larger than that of the mainland. Of the 55 samples collected on the island, four were without spiders as were 15 of the 55 samples collected on the mainland. However, the Index of Diversity for each community showed no significant difference: 18.62 (17.07- 20.17) for the island; 19.88 (17.20-22.36) for the main- land. Further, when the Index is calculated on the com- bined samples (78 species and 586 individuals), again there was no significant difference; 17.50 (16.53-18.47). According to Williams (1944), these results would indicate that the samples were from the same p0pu1ation. These re- sults reflect a further characteristic of this Index: the values are small and there are a large number of individ- uals representing a much smaller number of species. CONSIDERATIONS ON THE ORIGIN OF THE ISLAND FAUNA Zoogeographic Aspects Studies by several authors on the impoverished verte- brate faunae of the island have indicated that relatively little morphological variation has taken place, and the species present are for the most part identical with those of the mainland of Michigan. Effects of isolation have been recorded by Hatt, 2;. 1. (1948) for one species of amphibian, the Redbacked Salamander (Plethodon cinereus cinereus (Green)), and two species of snakes, the Milk Snake (Lampropeltis triangulum.triangulum (Lacepéde)) and the Water Snake (Natrix sipedon sipedon (Linnaeus)). Hatt .25..§1. (pp. 315.) conclude that fifteen of the remaining 16 Species of the herpetological fauna have been derived from the Southern Peninsula of Michigan. The one excep- tion is the Painted Turtle (Chrysemys pigpa marginata Agassiz), which is Similar to those found in the Northern Peninsula of Michigan and in Wisconsin. According to Ozoga and Phillips (1964), the mammal pOpulations of the island most closely resemble those of the Northern Peninsula. With the exception of the Western House wren (Troglodytes aedon parkmanii Audubon), the avifauna resembles the South- ern Peninsula (Hatt, 22'.§l°: 1948). To date, 163 bird species have been recorded for the island (Drew and 64 65 Phillips, 1964). This figure represents less than half the number of birds that could reach the island (by fly- ing) from the adjacent mainland of Michigan. The data on the herpetological and mammal faunae suggest accidental introductions, with the exception of the amphibian and snakes mentioned above, rather than relict p0pu1ations derived from individuals which made use of the exposed dry land during the low-water or Lake Chippewa stage. The open water separating the island from the mainland masses is considered to be an effective bar- rier to most terrestrial vertebrates. However, during the winter when the island is ice locked to the mainland, some Species may use this bridge to gain access to or leave the island (Ozoga and Phillips, 1964). The relative ease of transport (winds/air currents) available to many species of spiders rules out obvious ’ Iterrestrial barriers. This is reflected by the fact that of the 211 Species presently known to occur on the island, 167 Species or 79.1% are common to Wisconsin to the west and Michigan to the north and east. Further, a western component, species occurring in Wisconsin and not in Michigan, is represented on the island by 20 species or 9.5% of the total. The majority of these 20 Species presently occur in northern Wisconsin (10 species) or are 66 scattered over the entire state (five species); and the remainder (five species) are either west-central or south- ern in distribution. There are 16 Species (7.6%) on the island that occur in Michigan and not in Wisconsin. Of this number, 15 occur on adjacent mainland to the north and east; and only one is from the southern portion of the state. In conjunction with these components, there are eight species or 3.8% of the total that are unique to the island; they have not been recorded for the surrounding mainlands of Wisconsin.or Michigan. These species current- ly have known distributions confined to the northeastern regions of the United States. The above figures strongly suggest the influence of the nearby mainlands on the island with reference to the source of spider species. That the vagaries of the wind may be important in distri- bution is reflected by the number of species presently unique to the island as well as those in common with Wis- consin and not_Michigan, and conversely. Wind as a Factor in Distribution Aerial dispersal by several species of spiders in boreal and temperate regions of North America and EurOpe have been recorded by many investigators. Air-borne spi- ders have been observed 60 miles from the nearest mainland 67 (Darwin, 1896) and at an altitude of 5,000 feet (Glick, 1939). This means of dispersal is characteristic of the juveniles of many species soon after hatching. Several species of the families Linyphiidae and Micryphantidae, all small (1 mm. to 3 mm.) spiders even as adults, make regular use of this means of dispersal to the point that members of these families are described as having aero- nautical behavior. This behavior appears to be correlated with population density (Duffey, 1956) and occurs during the fall, winter and spring. Consequently, year round fluctuating winds could be mixing spiders from several populations and depositing them in other areas. This means of dispersal undoubtedly accounts for the disruptive dis- tribution patterns of many species of spiders. Besides the direct influence of winds, there is the indirect influence as shown by surface current patterns on lakes. These patterns are induced by the transfer of kinetic energy from the wind to the surface of the water through wind stress. Because of this, currents respond quickly to changes in the intensity and direction of winds (Sverdrup, 35,131., 1942). The most recent study on the surface current pattern in Lake Michigan was done by the Great Lakes Research Institute (Ayers, 25..§1., 1958). This study revealed that 68 under the influence of the prevailing westerly winds of June, believed to approximate the normal summer circula- tion, the northern region of this Lake is dominated by the outflow current as it flows toward the Straits of Mackinac (Fig. 11). The current originates near the cen- ter of the western shore of the lake and dips into the lower end before flowing northward along the west coast of Michigan. This pattern coincides with that found by Harrington (1895). The second phase of the surface cur- rent study was conducted during August after the Straits region had been under the influence of an easterly wind for eight days. The pattern at this time was a strong southward current that flowed from the Straits along the eastern edge of the lake to the lower end before turning northward (Fig. 12). Evidence to support the role of surface currents in spider diSpersal was found during the course of this study. Living representatives of 17 species were found trapped in the surface tension 25 to 30 feet off the eastern shore- line of the island (Table 8). Of these, 14 species are known to occur on the island, Wisconsin and Michigan. The remaining three species have not been collected on the island. Of these, one species occurs in northern Wiscon- sin and two in the Southern Peninsula of Michigan. They —44° SURFACE CURRENTS sssssss n: to was was won res CLIENT MOTION mo FM MIT VELOCITY_ 88° 87° 86' 350 Figure 11. Summer surface currents in Lake Michigan according to Ayers, _e_i_:: 31;, 1958. 70 /' ,fi «\c/xz/Jp \K ’I‘ (u§ t / n‘”) '\ u, / 1‘ ‘ M W l\.'L//¢/ .‘ " ‘1' 1 SURFACE cusseurs {txlfikfil ' l svuonuc m 9.1," ‘,f)‘)~lx) & no AUGUST I958 _44. " a}; ”'11 l @3833ng 43° GRAPH M ”RENT VELOOITY g .2 9"" .4 -p§ 24 305F265!) KL 86' 05’ Figure 12. Variation of summer surface currents in Lake Michigan accordingto Ayers, _e__t. a1__., 1958. 71 Table 8. Species found on the surface of Lake Michigan No. of Families and species individuals Sexl Theridiidae Theridion frondeum 4 M .2: glaucescens 1 M Linyphiidae Bathyphantes pullatus 1 M B, pallidaz 2 M,F Micryphantidae Ceratinopsis laticeps3 l M Erigone atra 3 2M,1F ‘Minyriolus arenarius3 1 F Tetragnathidae Tetragnatha laboriosa 1 F Gnaphosidae Drassyllus Sp. 1 Imm. Clubionidae Micaria montana 1 Imm. Thomisidae Philodromus rufus 1 F Tibellus oblongus 1 Imm. Xysticus Spp. 3 Imm. Salticidae , Icius similis 2 M Phidippus rimator l Imm. Talavera minuta 1 M 1 ML= males, F = females, Imm. = immature. 2 Occurs in northern Wisconsin but not in Michigan nor the island. 3 Occurs in southern Michigan but not in WiSconsin nor the island. 72 evidently became trapped in the surface tension after being deposited by dying winds and were then carried by currents eddying off the shore of the island. There is no way of determining the period of time the spiders were trapped in the surface tension. However, the flexible metabolic requirements of spiders and the fact that their need for water is more critical than for food gives them a distinct advantage to survive such tranSport. It Should be noted that these specimens were found during periods of no appreciable wind. Possibly, with winds of sufficient velocity, Spiders could be extricated from the surface tension and continued on their passive journey. Distribution Through Human Activities Human activities may have influenced spider distribu- tion. Travel, trade and agricultural activities on the island and between the island and mainlands have existed for some time. Some species could conceivably have reached the is- land on the canoes and boats of Indians and early white settlers. Quimby (1960) maintains that the Chippewa, Potawatomi and possibly the Ottawa Indians used the is- lands in northern Lake Michigan as hunting and fishing sites during the summer months. Visitations to the island 73 by early white settlers for exploratory or social activi- ties are well known (Bald, 1961). Sporadic boat travel to and from Beaver Island continued and later, as it ex- ists today, became more regular. No doubt, species trav- eled aboard as accidental passengers. Early in the 20th century, large quantities of pota- toes and fish packed in boxes and crates were exported from the island to mainland ports (pers. comm., Capt. Forrest Pratt, 1962). It is conceivable that before these crates and boxes were returned, spiders could have se- creted themselves or laid egg sacs in or on them. During thelumbering days in Michigan and Wisconsin, some of the logs were floated down rivers that drained into Lake Michigan (i.e., the Muskegon, Pere Marquette, -Manistee and Menominee Rivers) (Bald, 1961). Some of these logs could have passed the mill sites and reached the open water of the Lake. Once there, currents could have carried the logs to the island. The logs, in turn, could have carried spiders or spider eggs in bark crev- ices; and when the logs were washed ashore, these waifs may have established themselves on the island. In summarizing the information on the origin of the spider fauna of Beaver Island: some species may have been present since portions of the present island were exposed 74 above post-glacial Lake Algonquin about 9000 years ago; other populations could have occupied the mass of exposed area during the low water stage which was later fraction- ated and isolated from the mainland mass by the formation of Lake Nipissing 3000 years ago; still others have been carried to the island by wind, air and lake currents dur- ing this latter time and up to the present; and some spe- cies were transported to the island through the activities of man. DISCUSSION Composition of the Spider Fauna of Beaver Island The composition of the current spider fauna of Beaver Island is due to the interaction of the present climate of the region, the influence of the surrounding mainlands, the ecological diversity of the island, and the impact of restricted area. The fauna is derived from at least four sources. Of the 211 Species known to occur on the island, 20 species or 9.5% are common to the mainland of Wisconsin; 16 Spe- cies or 7.6% are common to the mainland of Michigan; eight species or 3.8% are endemic to the island; and 167 or 79.1% presently occur on the mainlands of Wisconsin and Michi- gan. The large percentage (79.1%) in common with the two 75 surrounding mainlands is a direct result of the current climate in this region of the Great Lakes, the mixing of Species attributable to the winds sweeping over these areas and the similarity of post-glacial histories. Pol- len profiles indicate that the present cool, moist climate was initiated approximately 2200 years ago during the Lake Algoma stage in the Lake Michigan basin (Hough, 1958, Deevey, 1949). This in conjunction with the fact that species occurring in a given area are present as a conse- quence of similar adaptations to the physical factors of the environment (Allee 25..31., 1949) indicates that there has been sufficient time for the establishment of the cur— rent spider fauna in this area. A measure of the ecological diversity of the spider fauna of a region is to evaluate it on the basis of food- getting behavior, web-weaving or hunting. Basic to this criterion, is the assumption that during the initial colo~ nization of a pristine area there would be an imbalance of behavior types due to fortuitous immigration; but through time and biological maturity of the communities within an area, a balance of behavior types would be reached. This implies that available niches would be filled by Species that were structurally, physiologically and behaviorally adjusted; and competition for similar niches would occur and some forms would be absent because of unavailable niches. q 76 The application of this behavioral criterion to the surrounding mainlands discloses a Wisconsin fauna of 51.2% web-weaving and 48.8%phunting species, and aMichigan fauna of 51.0% web-weaving and 49.0% hunting Species (Tables 4 and 5). Thus the ecologically diverse mainlands support Spider faunae that are almost equal in web-weaving and hunting species. The spider fauna of Beaver Island is composed of 54.1% web-weaving species and 45.9% hunting species (Tables 4 and 5). These percentages indicate an ecologi- cal situation Slightly favorable to the web-weaving spe- cies which, in turn, possibly have easier access to the distributing winds than do hunting species. Implied in this evaluation is the assumption that some of the hunting species present on this island may have reached it by wandering over the exposed but presumably ecologically suitable peninsula during the 700-year long Lake Chippewa stage of the Lake Michigan basin. With the submergence of major portions of this land mass, the immigration of hunting species has been more limited than has that for the majority of web-weaving Species. Of the 27 families known to occur in this region (Tables 4 and 5), 10 are not represented on the island. The absence of these may be accounted for on the basis of 77 the restricted area of the island or, as Anderson (1960) defines this phrase, dearth of niches. This indicates that the available niches are occupied, leaving little if any chance for encroachment by ecologically similar but less favored forms. This may in part account for the ab- sence of species of Oxyopidae and Anyphaenidae. These Species are hunting Spiders that are usually abundant in grassy fields and shrubs of wooded areas. Quantitative sampling of these areas (Figs. 5-8) reveals that hunting Species of the families Salticidae, Thomisidae and Clubi- onidae are abundant and may be filling these niches quite adequately. Or, possibly, the absence of oxyopid and anyphaenid species represents a stage in the ecological evolution of the spider fauna. Perhaps these species occur in the relatively youthful stages of the development of a community; or, become established during the later or more mature stages. The remaining eight families are relatively rare on the surrounding mainlands where they are represented by one species each. The exception - Pholcidae - is repre- sented in Wisconsin by two Species of two genera, but neither are abundant (Levi and Field, 1954). Spider Populations of Specific Communities The structure of the spider p0pu1ations inhabiting- four real communities has been graphically presented in 78 Figures 5-10. These p0pu1ation profiles indicate that two or three species dominate each community in abundance. With reference to food-getting behavior, web-weaving species dominate the following communities: Old Field (September), 57.2%; BeechrMaple shrubs, 62.0%; and the Junipers, 66.7%. Hunting species dominate the leaf litter community of BeechrMaple forest, 54.6%; and the Northern White Cedars, 58.3%. The Old Field community in July was equally balanced with 50% of each type species. Diurnal succession in the Old Field community in July revealed that Dictypa coloradensis is more active later in the day than is Q. volucripes. This behavioral difference may be of value in future descriptions of these two mor- phologically similar Species. Two Species, Tetragnatha laboriosa and Philodromus crespiticolus occur, at varying rates of abundance, in all of the communities sampled. Evidently these communities —- '— are micrcclimatologically similar and/or these two species have a wide range of tolerance. The Spider P0pu1ations of a Similar Habitat on the Island and on the Mainland of Michigan The results of the logarithmic series of Fisher, 23. ‘31., indicate that the species-diversity values of the leaf litter p0pu1ations on the island and the mainland are 79 Similar. Moreover, the value obtained from the combined samples indicates the samples could be from the same popu- lation. This similarity of Spider faunae could only have developed during the time Since the two Sites were exposed after the demise of Lake Algoma 2200 years ago even though they have been separated for approximately 5000 years. Because both sites currently support ecologically similar communities, it is quite probable that the environ- mental factors of each are identical. Because theSe fac- tors may have adverse as well as beneficial effects on individuals occurring in specific habitats within the com- munities, the abundance of the Species present can be favorably or unfavorably affected. Clark‘gg..§1. (1964) maintain that effects of environmental factors tend to combine multiplicatively and are reflected by the relative abundance of the individuals of a Species which inhabits a community. Basically, it is this relative abundance that is evaluated by species-diversity values. Thus, the similarity of the spider populations inhabiting the leaf litter of these ecologically similar though geographically separated sites is the product of the interaction of envi- ronmental factors and the tolerance ranges of the individ- uals comprising a species. 80 This similarity of the populations inhabiting the separated communities does not mask a number of dissimi- larities. In 51 samples from the island, 424 individuals were collected and from 40 mainland samples, 162 individ- uals. These figures yield crude densities (Odum, 1959) of 8.3 spiders/sq. ft. of litter contrasted with 4.1 spiders/sq. ft. of litter. This dissimilarity indicates that the carry- ing capacity of the island is much greater than that of the mainland. Also, on the island, the web-weaving species dominate the hunting species almost 2 to 1 (1.95 - 1). The community on the mainland supports a more nearly balanced pr0portion of 1.32 - 1 a figure very similar to the ratio for the entire mainland of Wisconsin or Michigan. Thus the web— weaving Species are favored in some manner in the island community. These two dissimilarities appear to be reflections of a more biologically interesting feature--the establishment and subsequent divergence of the populations currently oc- cupying these similar but separated communities. The premise that the island populations originated independently is based on several geological features out- lined and discussed above. The island p0pu1ations are not isolated portions of established populations but rather 81 are the results of fortuitous immigrations from parental populations. This implies the formation and develOpment ofaa gene pool that initially would be quite different from that of a segment of the mainland populations. This possibility could be explained in terms of the founder principle of Mayr (1963) which postulates that the small gene pool would, in time, develop into a large population in which the genes have adaptive values different than those of the parental p0pu1ation. These changes are the end results of genetic drift, recombinations and to a lesser extent mutations. This genetic substrate in the confines of the simplified island environment and under the protection of isolation could contribute to high popu- lation densities. In addition, because of the aeronautic behavior of many families of web-weaving species, a pre- dominance of these over the hunting Species would be ex- pected in such isolated situations. SUMMARY AND CONCLUSIONS A total of 211 species representing 17 families and 108 genera have been collected on this 58-square mile island during the course of this study. Though this number may seem large, it is just less than one half of the number recorded for Wisconsin and for Michigan. This disparity in 82 species number takes on added interest because the island has 30 of the 33 plant communities found in Michigan and was once connected to that mainland. The primary means of dispersal available to most spiders is on wind and air cur- rents. Thus, the species found on the island have reached it either by this passive means or by individual effort-- wandering--during the 9000 years Since portions of the island were first exposed and, later, temporarily land- locked to the mainland of Michigan. The following State- ments summarize some of the findings concerning the Spider fauna of Beaver Island: 1. The fauna has four geographic components: species in common with both Wisconsin and Michigan; species in common with Wisconsin only; species in common with.Michigan only; and, species endemic to the is- land. The major portion (167 Species, 79.1%) is common with the surrounding mainlands; the minimal portion (8 species, 3.8%) represents the endemic component. 2. The Spider p0pu1ations inhabiting specific communi- ties are structured with reference to the parameters of frequency and density and the food-getting behav- ior of each species. 83 Positive evidence in the form of 17 species indicates access to the island is accomplished through indirect effect of the wind, namely, surface currents of Lake Michigan. The carrying capacity of the hardwood-conifer leaf litter community on the island is more than twice that of the mainland. There are no significant differences between Fisher's Index of Diversity values for the island and mainland populations inhabiting the hardwood-conifer litter. The calculated values indicate these populations could be segments of the same population even though they are currently separated by 19 miles of open water . The similarity of the Index values is attributed to the interaction of environmental factors and the tolerance ranges of the individuals composing the populations. These interactions accumulate multi- plicatively and determine the abundance of the p0pu1ations. The present populations of the leaf litter community are not an isolated segment of the mainland popula- tions but have been established independently. LITERATURE CITED Allee, W. C., Emerson, A. E., Park, 0., Park, T. and Schmidt, K. P. 1949. Animal ecology. W. B. Saunders Co., Phila- delphia xii+837 pp. Anderson, P. K. 1960. Ecology and evolution in island pppulations of salamanders in the San Francisco Bay region. Ecol. Mono. 30: 359-385. Arnold, J. R. and Libby, W. F. 1949. Age determinations by radiocarbon content: Checks with Samples of known age. Science 110: 678-680. 1951. Radiocarbon dates. Science 113: 115-118. Ayers, J. C., Chandler, D. C., Lauf, G. H., Powers, C. F. and Henson, E. B. . 1958. Currents and water masses of Lake Michigan. Univ. Michigan Gt. Lakes Res. Inst. Publ. 3, 169 pp. Bald, F. C. 1961. Michigan in four centuries. Harper and Bros. Publ., New York xii+528 pp. Barnes, B. M. and Barnes, R. D. 1954. The ecology of the spiders of maritime drift lines. Ecology 35: 25-35. Barnes, R. D. 1953. The ecological distribution of spiders in non- forest maritime communities at Beaufort, North Carolina. Ecol. Mono. 23: 315-337. Barnes, R. D. and Barnes, B. M. 1955. The spider population of the abstract broom- sedge community of the southeastern piedmont. Ecology 36: 658-666. Blust, F. and DeCook, B. G. 1960. Comparison of precipitation on islands of Lake Michigan with.precipitation on the perimeter 84 85 of the lake. Jour. GeOphys. Res. 65: 1565- 1572 0 Brady, A. R. 1964. The lynx spiders of North America, north of Mexico (Araneae:0xy0pidae). Bull. Mus. Comp. 2001. Harvard Univ., 131: 429-518. Broecker, W. S. and Farrand, W. P. 1963. Radiocarbon age of the Two Creeks foresu bed, Wisconsin. Bull. Geol. Soc. America 74: 795- 802. Chamberlin, R. F. and Ivie, W. 1944. Spiders of the Georgia region of North America. Bull. Univ. Utah, Biol. Ser. 8, 267 pp. Clark, P. J., Ekstrom, P. T. and Linden, L. C. . 1964. On the number of individuals per occupation in a human society. Ecology 45: 367-372. Cole, L. C. 1946. A study of the cryptozoa of an Illinois wood- land. Ecol. Mono. 16: 49-86. Comstock, J. H. 1948. The spider book. Comstodk Publ. Co., Inc., Ithaca xi+729 pp. Darlington, H. T. 1940.. Some vegetational aspects of Beaver Island, Lake Michigan. Pap. Michigan Acad. Sci., Arts and Letters 25: 31-37. ‘ Darwin, C. R. 1896. Journal of researches into the natural history and the geology of the countries visited dur- ing the voyage of the H. M. S. Beagle round the world, under the command of Capt. Fitz Roy, R. N. D. Appleton and Co., New York x+519 pp. Deevey, E. 8., Jr. 1949. Biogeography of the Pleistocene, Part 1: Europe and North America. Bull. Geol. Soc. America 60: 1315-1416. 86 Dillon, L. S. 1956. Dhmn,W.J. 1957. Drew, L. C. Drew, L. C. 1964. Duffey, E. 1956. Fernald, M4 1950. Wisconsin climate and life zones in North America. Science 123: 167-176. and Massey, F. J., Jr. Introduction to statistical analysis. McGraw- Hill Co., Inc., New York xiii+488 pp. The spiders of Michigan. (in preparation) and Phillips, C. J. Observations concerning the birds of Beaver Island, Charlevoix County, Michigan. Jack- Pine warbler 42: 234-238. Aerial dispersal in a known Spider population. Jour. Anim. Ecol. 25: 85-111. L0 Gray's manual of botany. American Book Co., New York lxiv+l632 pp. Fisher, R. A., Corbet, A. S. and Williams, C. B. 1943. The relation between the number of individuals and the number of species in a random sample of an animal population. Jour. Anim. Ecol. 12: 42-58. Fuller, G. D. 1918. Gertsch, W. 1949. Click, P. A. 1939. Harrington, 1895. Some perched dunes of northern Lake Michigan and their vegetation. Trans. Illinois Acad. SCio 11: 111-1220 J. American spiders. D. Van Nostrand Co., Inc., New York xiii+285 pp. The distribution of insects, spiders and mites in the air. U. S. Dept. Agric. Tech. Bull. 673, 150 pp. M. w. Surface currents of the Great Lakes, as deduced from the movement of bottle papers during the 87 seasons of 1892, 1893, 1894. U. S. Weather Bur. Bull. B (rev. ed.) 14 pp. Hatt, R. T., Van Tyne, J., Stuart, L. C., P0pe, C. H. and Grobman, A. B. 1948. Island life: a study of the land vertebrates of the islands of eastern Lake Michigan. Cran- brook Inst. Sci. Bull. 27, Cranbrook Press, Bloomfield Hills xi+l79 pp. Heatwole, H. 1961. Analysis of the forest floor habitat with a structural classification of the litter or L layer. Ecol. Mono. 31: 267-283. Hough, J. L. 1955. Lake Chippewa, a low stage of Lake Michigan indicated by bottom samples. Bull. Geol. Soc. America 66: 957-968. 1958. Geology of the Great Lakes. Univ. Illinois Press, Urbana xviii+313 pp. 1962. Geologic framework. In Pincus, H. J. (ed.), Great Lakes basis. American Assoc. Adv. Sci. Publ. 71: 3-27. ‘ 1963. The prehistoric Great Lakes of North America. American Sci. 51: 84-109. Howden, H. F. 1963. Speculations on some beetles, barriers, and climates during the Pleistocene and Pre- Pleistocene periods in some non-glaciated portions of North America. Syst. Zool._12: 178-201. Kaston, B. J. 1948. Spiders of Connecticut. Bull. Connecticut Geol. Hist. Surv. 70, 874 pp. Leverett, F. and Taylor, F. B. 1915. The Pleistocene of Indiana and Michigan and the history of the Great Lakes. U. S. Geol. Surv. Mem. 53, 529 pp. Levi, H. W. and Field, H. M5 1954. The spiders of Wisconsin. American Midl. Nat. 51: 440-467. 88 Levi, H. W. and Levi, L. R. 1961. Some comments on Walckenaer's names of American Spiders, based on Abbot's drawings. Psyche 68: 53-57. Levi, H. W., Levi, L. R. and Kaspar, J. L. 1959. Harvestmen and spiders of Wisconsin; addition- al Species and notes. Trans. Wisconsin Acad. Sci. 47: 43-52. Lowrie, D. C. 1948. The ecological succession of the spiders of the Chicago area dunes. Ecology 29: 334-351. Martin, P. S. and Harrell, B. E. 1957. The Pleistocene history of temperate biotas in Mexico and eastern United States. Ecology 38: 468—480. Mayr, E. 1 3. Animal species and evolution. Belknap Press of Harvard Univ. Press, Cambridge xiv+797 pp. Muma, H. and Muma, K. 1949. Studies on a population of prairie spiders. Ecology 30: 485-503° Odum, E. P. 1959. Fundamentals of ecology. W. B. Saunders Co., Philadelphia xi+546 pp. Ozoga, J. J. 1963. An ecological study of the coyote on Beaver Island, Lake Michigan. Unpubl. Diss., Michi- gan State Univ. viii+122 pp. Ozoga, J. J. and Phillips, C. J. 1964. Mammals of Beaver Island Michigan. Michigan State Univ., Publ. Mus., Biol. Ser. 2: 307- 347. Quimby, G. I. 1960. Indian life in the upper Great Lakes. Univ. Chicago Press, Chicago xv+182 pp. 89 Smith, P. W. 1957. Analysis of post-Wisconsin biogeography of the prairie peninsula region based on distribution- al phenomena among terrestrial vertebrate popu- lations. Ecology 38: 205-218. ’ Stanley, G. 1938. The submerged valley through.Mackinac Straits. Jour. Geol. 46: 966-974. Sverdrup, H. U., Johnson, M. W. and Fleming, R. H. 1942. The oceans, their physics, chemistry and gen- eral biology. Prentice-Hall, Inc., New York x+1087 pp. Thwaites, F. T. and Bertrand, K. 1957. Pleistocene geology of the Door Peninsula, Wisconsin. Bull. Geol. Soc. America 68: 831- 880. ' van der Schalie, H. 1963. Mussel distribution in relation to former stream confluence in northern Michigan, U. S. A. Malacologia 1: 227-235. Veatch, J. O. 1953. Soils and land of Michigan. Michigan State Coll. Press, East Lansing xi+24l pp. Whitcomb, W. H., Exline, H. and Hunter, R. C. 1963. Spiders of the Arkansas cotton field. Ann. Ent. Soc. America 56: 653-660. Williams, C. B. 1944. Some applications of the logarithmic series and the Index of Diversity to ecological prob- lems. Jour. Ecol. 32: 1-44. Wright, H. E., Jr. 1964. Aspects of early postglacial forest succession in the Great Lakes region. Ecology 45: 439-448. Zumberge, J. H. and Potzger, J. E. 1956. Late Pleistocene history of the Lake Michigan basin correlated with pollen studies. Bull. Geol. Soc. America 67: 271-288. "1111111111111111“