The GREAT LAKES ENTOMOLOGIST

Vol. 13, No. 4 Winter 1980 THE GREAT LAKES ENTOMOLOGIST

Published by the Michigan Entomological Society

Volume 13 So. 1 ISSN 0090-0222

TABLE OF CONTENTS

Seasonal Flight Patterns of Hemiptera in a North Carolina Black Walnut Plantation. 1. Pentatomoidea J. E. McPherson and B. C.Weber ...... IT-

Mixed Species Feeding Assemblages of Planthopper Nymphs (Hornoptera: Fulgoroidea) S. W. Wilson and J. E. McPherson...... 185

Observations on the Biology of the Slender Seedcorn Beetle, Cli~.inair?lpressjtions (Coleoptera: Carabidae) Robert D. Pausch and Lois M. Pausch...... 189

Localized Field Migration of the Adult Alfalfa Weevil, Clover Leaf Wee~il.and Clo~er Root Curculio (Coleoptera: Curculionidae), and its Implication for a Fall Pest Management Program R. D. Pausch, S. J. Roberts, R. J. Barney, and E. J. Armbrust ...... 195

Annotated List of Trichoptera Collected along Furnace Run of the Cuyahoga \-alley National Recreation Area in Northeastern Ohio C. Petersen and B. A. Foote ...... 201

Instar Head Widths, Individual Biomass, and Development Rate of Forest Tent Cater- pillar, Malacosorna disstria (Lepidoptera: Lasiocampidae), at Two Densities in the Laboratory Jocelyn M. Muggli and William E. Miller ...... 207

Validation of a PETE Timing Model for the Oriental Fruit Moth in Michigan and Central California (Lepidoptera: Olethreutidae) B. A. Croft, M. F. Michels, and R. E. Rice...... 211

The Biology of the White Pine Sawfly, Neodiprion pinetum, (Hymenoptera: Diprionidae) in Wisconsin Aunu Rauf and D. M. Benjamin...... 219

Distribution and Biology of Bumblebees (Hymenoptera: Apidae) in Michigan R. W. Husband, R. L. Fischer, and T. W. Porter...... 225

A Checklist of Illinois Centipedes (Chilopoda) Gerald Summers, J. A. Beatty, and Nanette Maguson...... 211

COVER ILLUSTRATION

A male northeastern pine sawyer, Monochamus notatus (Drury). Coleoptera: Cerambycidae) on a white pine log from which it emerged. Photograph by Nancy Gosling. School of Natural Resources. University of Michigan. THE MICHIGAN ENTOMOLOGICAL SOCIETY

I9XO-X I OFFICERS

President Gal-y A. Simmons President-Elect John A. Wirter Executive Secretary M. C. Nielsen Editor D. C. L. Gosling

The Michigan Entomological Society traces its origins ro the old Detroit Entomological Society and was organized on 4 November 1954 to ". . . promote the science of entomology in all its branches and by all feasible means, and to advance cooperation and good fellowship among persons interested in entomology." The Society attempts to facilitate the exchange of ideas and information in both amateur and professional circles. and encourages the study of insects by youth. Membership in the Society, which serves rhe North Central States and adjacent Canada, is open to all persons interested in entomology. There are four paying classes of membership:

Student (including those currently enrolled in college or graduare programsWues $4.00 Active-annual dues $8.00 Institutional-annual dues $10.00 Sustaining-annual contribution $25.00 or more

Dues are paid on a caIendar year basis (Jan. I-Dec. 3 1).

Memberships accepted before July I shall begin on the preceding January I: memberships accepted at a later date shall begin the following January I unless the earlier date is re- quested and the required dues are paid. All members in good standing receive the Nr~t~slrffrr of the Society, published quarterly. All Active and Sustaining Members may vote in Society affairs. All dues and contributions to the Society are deductible for Federal income tax purposes.

SUBSCRIPTION INFORMATlON

Institutions and organizations, as well as individuals not desiring the benefits of membership, may subscribe to Tllr Grrur Luhrs Enrornologisr at the rate of 910.00 per volume. The journal is published quarterly: subscriptions are accepted only on a volume (4 issue) basis. Single copies of Thr Grrur Luhrs Enrornologisr are available at $2.75 each. with a 20 per cent discount for 35 or more copies sent to a single address. MICROFILM EDITION: Positive microfilm copies of the current volume of Thr Grrur Luhrs Enrornologisr will be available at nominal cost, to members and bonafide subscribers of the paper edition only, at the end of each volume year. Please address all orders and inquiries to University Microfilms, Inc., 300 North Zeeb Road, Ann Arbor. Michigan 48106, USA. Inquiries about back numbers, subscriptions and Society business should be directed to the Executive Secretary. Michigan Entomological Society, Department of Entomology. Michigan State University. East Lansing, Michigan 48824, USA. Manuscripts and related correspondence should be directed to the Editor (see inside back cover).

]!?SO THE GREAT LAKES ENTOMOLOGIST

SEASONAL FLIGHT PATrERNS OF HEMIPTERA IN A NORTH CAROLINA BLACK WALNUT PLANTATION. 1. PENTATOMOIDEA

J. E. ~c~hersonland B. C. ~eber2

ABSTRACT

The seasonal flight patterns of 35 species and subspecies of Pentatomoidea collected in window traps in a North Carolina black walnut plantation are discussed. Flying height distributions and seasonal activities of the five species that were most frequently collected (i.e., Amnestus basidentatus, A. pallidus, A. spinifrons, Brochymena qltadripustulata and Euschistus servus) are considered in detail.

The increasing demand and decreasing supplies of black walnut (Juglans nigra L.) for timber and veneer require a better understanding of the potentially harmful and beneficial insects that inhabit walnut plantations. Most papers published, thus far, have dealt only with species actually found on the trees. For example. Nixon and McPherson (1977) compiled an annotated list of phytophagous insects found on black walnut in southern Illinois; no attempt was made to collect other species inhabiting the plantations but not occurring on the trees. Included in their list were 30 species of Hemiptera. In some instances (e.g., Corimelaena pulicaria), species found on the walnut trees represented overflow from the normal host plants within the plantations (Nixon et al. 1975). The present study was begun as an extension of a window trapping study by the junior author on Scolytidae (Coleoptera) in a North Carolina black walnut plantation. We found that several species and subspecies of Hemiptera were also present in the collections, and some in considerable numbers. The purpose of this paper is to provide a list of the Pentatomoidea occurring in this plantation during 1977 and 1978 and to present information on the flying height and flight activity period of each taxon. Later papers will deal with other groups of Hemiptera. All hemipteran specimens collected during this study are deposited in the Entomology Collec- tion, Zoology Research Museum, Southern Illinois University, Carbondale.

MATERIALS AND METHODS

&scription of the stody area. The black walnut plantation is located in McDowell County, North Carolina, 35"41'3(Y' N latitude and 8lo52'3(Y' W longitude, 1,100 ft elevation, ca. 40 miles east of Asheville. Total acreage of walnut is 12 acres but trapping in the present study was confined to a 5-acre area in the eastern part of the plantation. The trees were planted in 1973 at a 12 x 12 fmt spacing. Average height of the trees was 2.42 and 2.84 m in 1977 and 1978. respectively: average diameter at breast height was 2.43 and 3.12 cm in 1977 and 1978. respectively. The soil is sandy loam. Plant cover within the study area includes primarily Silene antirrhim L.. Solidago canadensis L., Oenothera laciniata Hill and Oenothera biennis L. However. the plantation is disked at least once each year for weed control.

'~epartmentof h1~.Southern Illinois University, Carbondale, IL 62901. 2~.~.~..4.Forest Sewice. North Central Forest Experiment Station, Forestry Sciences Laboratory, Carbondale. 1L 6--1. THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Fig. I. Window trap set at 1 rn

Widow flight traps. The traps (Figs. 1-2) were modified from Chapman and Kinghorn (1955) and Wilson (1%9). Each trap was constructed with a section of Plexiglas (76.2 x 76.2 x 0.3 cm) enclosed on the sides and top with a painted pine frame, and on the bottom with a galvanized metal pan (71 x 20 x 8 cm) bolted to the pine frame. An overflow hole 5 cm below the rim on the front and back sides, near opposite ends of the pan, was covered with nylon netting. Latex caulking was used to secure the Plexiglas in the frame and to seal all seams in the pan. The support frames consisted of 94 inch pipe mounted above bell reducers and attached to 1% inch base pipe. A brace piece on the taller traps (Fig. 2) provided additional support during high winds. Four guy wires on each trap were attached to metal stakes in the ground, and to the brace piece, to prevent twisting of the frame. A rope and pulley system permitted raising and lowering the traps to the desired height, and easy removal of insects. Twenty eight traps were randomly located within the 5-acre study area; four traps were suspended at each of the following heights: 1, 2, 3,4, 5,6, and 7 m. The height of each trap was determined by measuring from the mid-point of the Plexiglas to ground level. Each trap was partially filled with ca. 1% inch of ethylene glycol as the killing and preservative agent. Insects were removed weekly from 24 March to 14 October in 1977, and 24 March to 13 October in 1978. THE GREAT LAKES ENTOMOLOGIST

Fig. 2. Window trap set at 5 m. Note the rope and pulley system for raising and lowering the trap, and the brace piece below the trap for additional support.

RESULTS AND DISCCSSION

Thirty five species and subspecies were collected during the two years of this study with the pentatomids being best represented (Table 1 ). Several of the taxa were found at all seven heights: numbers of specimens collected for all taxa ranged from 1 to 609. The average flying heights of these insects show- no obvious general patterns. Body size, for example. was apparently wta factor in fl>-ingheight since both large (avg body length > 10.0 rnm) (e.g.. Ter?.ra bipunctata. Acrosrernum hilare, Brochymena arborea) and small (avg body length < 7.0 mm) 1e.g.. Galgupha ocalis. Amnestus basidentatus, A. pusillus, Melanaethus penn.lcanicus) insects were collected at average heights of 4 m or more; likewise. large 1e.g.. Broth>-mem p. punctata. Euschistus semus, E. variolarius) and small (e.g.. Corimelae~I. lateralis. Sehirus c. cinctus, Cosmopepla bimaculata, Mormidea lugens) insects were collected at average heights of less than 3 m. Variation within genera was also evident. For e-ample. the average heights at which all species of Euschistus were 180 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Table 1. Seasonal flight activity of Pentatomoidea during 1977-78 in a North Carolina black walnut plantation. Number Collection Height (m) Col- - Range of Taxon lected Kk SE Range Collection Dates SCUTELLERIDAE Pachy corinae Stetharrlax marmorata (Say) 1 29 Sept. Tetyra bipunctata (Henich- Schaeffer) 7 16 June4 Sept. CORIMELAENIDAE Corimelaena I. lateralis (Fabricius) 1 17 June C. pulicaria (Germar) 16 3 1 March-24 June Galgupha atra (Amyot & Serville) I 12 May G. ovalis (Hussey) 6 I April-15 April CYDNIDAE C ydninae Melanarthus pensylvunicus (Signoret) 1 26 May Amnestinae Amrlestus basidentatus Froeschner 609 I April-7 Oct. A. pallidus Zimmer 103 31 March-13 Oct. A. pusillus Uhler 3 12 May-11 Aug. A. spinifrons (Say) 78 3 1 March-26 May Sehirinae Sehirus c. cirzctus (Palisot de Beauvois) 6 I April-21 July PENTATOMIDAE Pentatorninae Acrosternurn hilare (Say) 48 20 May4 Oct. Brochymena arborea (Say) 17 6 May-4 Aug. B. p. putzcfata Van Duzee 18 14 April-7 July B. qrtadripustulatu (Fabricius) 5 13 14 April-7 Oct. Coerlus delius (Say) 2 14 July-21 July Cosmopepla bimaculata (Thomas) 5 9 June-12 Aug. Dendrocoris humeralis (Uhler) 5 20 May-22 July Euschistus serous (Say) 143 1 April-13 Oct. E. t. tristigmus (Say) 14 6 May-9 Sept. E. variolarius (Palisot de Beauvois) 25 7 April-15 Sept. Holcostethus limbolarius (St%l) 30 7 April4 Oct. Hymenarcys nervosa (Say) 4 22 April-14 July Mormidea lugens (Fabricius) I 17 June Oebalus p. pugnax (Fabricius) 2 15 July-28 July Proxys punctulatus (Palisot tie Beauvois) I 23 June Thyanta accerra (McAtee) 6 6 May-29 July T. calceata (Say) 9 7 April-15 July Trichopepla semivittata (Say) 6 14 April-29 Sept. Asopinae Perillus bioculatus (Fabricius) I 23 June Podisus maculiventris (Say) 49 7 April-9 Sept. P. placidus Uhler 2 19 May-2 June Stiretrus anchorago fimbriatus (Say) 14 22 April-2 Sept. ACANTHOSOMATIDAE Elasrnostethus cruciatus (Say) 5 28 July-;! Sept. THE GREAT LAKES ENTOMOLOGIST

Meters Meters

12-34567 Meters

Figs. 3-7. Flying height distributions of five pentatornoid species during 1977-78 in a North Carolina black walnut plantation. 3, Amni~sfusbasidenratus: 1. .a. pallidus: 5. .a. spinifrons: 6. Brochymrnu quadripustulata; 7, Eusrhistus serrus.

collected fell within a narrow range of 1.% to 2.10 m, while those of all species of Brochymena ranged between 2.33 and 1.65 m. It is still possible that general patterns are present but obscured by the marked difference in the number of specimens that were collected for each species and subspecies. The difference in the number of each taxon collected also makes conclusions about flight activity during the season difficult (e.g.. time of adult emergence from overwintering sites, number of generations per ?-earl.Howex-er. five species (i.e., Atnnestus basidentatus, A. pallidus. A. spinifrons. Broch\.rnena quadriplrstulata and Euschisrus servus) were collected in suff~cientnumbers (Table 1) to permit a more detailed discussion of flying height distributions and seasonal flight actix-ities. Most specimens of A. basidentarus uere collected at 4 to 5 m (Fig. 3),most of A. pallidus at I to 2 and 4 to 5 m (Fig. 41.and most of A. spinifrons at 1 m (Fig. 5). However, what is most interesting are flight activities during the season. A. pallidus and A. spinifrons showed a marked similarin- in actit-ities during the early part of the season (Figs. 9-10) with 91 and 1007. respectively. of the total ?-year capture collected between March and June; A. pallidus was again collected August to October which may only reflect the higher total THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

L?4%h?lAMJJASO

C l!L-f4JAMJJASO Et!%4s%J AMJJASO

Figs. &I?. Seasonal tlight activities of five pentatornoid species during 1977-78 in a North Carolina black walnut plantation. 8. Amnestus basidentatus; 9, A. pallidus; 10, A. spinifrons; I I, Brochymena quadri- prts~rtlu~u;I?. Euschisfus servus.

number of this species taken. A. basidentatus, by far the most abundant of the three species (N =609), showed a marked difference in its flight activity with most individuals (56%) collected from 8 to 23 September (Fig. 8). This is even more striking when both years are treated separately for in 1977,47.4% of the year's collection (N =468) emerged the week of 2 September and in 1978, 38.3% the week of 22 September. It would be interesting to learn if A. basidentatus has the same seasonal flight pattern when the three cydnid species do not occur together (i.e., this may be the result of competition for resources). Most specimens of Brochymena quadripustulata were collected at 3 m (Fig. 6), which may reflect this species' arboreal habits (recall the heights of the trees in this plantation). It was active throughout the season with the only gap in the weekly collections occumng between mid-August and early September (Fig. 11). Judging by its reported life cycle in southern Illinois (Cuda and McPherson 1976), the decrease in numbers collected in July corresponds to the death of overwintered adults and the period between September and October, the appearance of new adults. Over 50% of the Euschistus serous specimens were collected at 1 m (Fig. 9), with rela- 1980 THE GREAT LAKES ENTOMOLOGIST 183 tively few individuals collected above 3 m. which may reflect this species' usual occurrence on low gms) and herbaceous vegetation. It is bivoltine (e.g., McPherson and Mohlenbrock 1976) and the peaks in April. Aumst. and October probably correspond to the emergence of overwintered adults. and the adults of the 1st and 2nd generations, respectively (Fig. 12).

We wish to thank Dr. R. H. Mohlenbrock. Department of Botany, Southern Illinois University. Carbondale. for his help in identifying the plants given as plantation ground cover. We also wish to acknowledge Mr. D. Brenneman and the staff of the North Carolina Dik ision of Forestry, Morganton, for their help in collecting data and maintaining the window traps. This research was partially supported by the U.S.D.A. North Central Forest Experiment Station.

LITERATURE CITED

Chapman, J. A. and J. M. Kinghorn: 1955. Window flight traps for insects. Canadian Entomol. 87:46-47. Cuda, J. P. and J. E. McPherson. 1976. Life history and laboratory rearing of Brochymena quadripustulata with descriptions of immature stages and additional notes on Brochymena arborea (Hemiptera: Pentatomidae). Ann. Entomol. Soc. Amer. 69:977-983. McPherson, J. E. and R. H. Mohlenbrock. 1976. A list of the Scutelleroidea of the La Rue-Pine Hills Ecological Area with notes on biology. Great Lakes Entomol. 9:125-169. Nixon, P. L. and J. E. McPherson. 1977. An annotated list of phytophagous insects collected on immature black walnut trees in southern Illinois. Great Lakes Entomol. 10:211- 222. Nixon, P. L., J. E. McPherson, and J. P. Cuda. 1975. A list of the Scutelleroidea (Hemiptera) collected on immature black walnut trees in southern Illinois with some notes on biology. Trans. Illinois State Acad. Sci. 68:409-413. Wilson, L. F. 1969. The window-pane insect trap. Michigan Entomol. Soc. Newsletter 14(34):1-3. THE GREAT LAKES ENTOMOLOGIST

MIXED SPECIES FEEDING ASSEMBLAGES OF PLANTHOPPER NYMPHS (HOMOPTERA: FULGOROIDEA)l

S. W. Wilson2 and J. E. ~c~herson3

ABSTRACT

The occurrence of four species of southern Illinois fulgoroids in mixed species feeding assemblages is reported.

During a study of the biology of southern Illinois planthoppers conducted from I April to 1 November, 1977 through 1979, we observed and/or collected 969 nymphs, 322 (33%) of which were found in mixed species feeding assemblages on herbaceous (e.g., Rumex) and woody (e.g., Juglans, Morus) vegetation (Wilson 1980). To our knowledge, these assemblages have not previously been reported for North American planthoppers. The assemblages consisted of combinations of the following four species: Acanalonia conica (Say) (Acanaloniidae), Anormenis septenlrionalis (Spinola), Metcalfa pruinosa (Say), and Ormenoides uenusta (Melichar) (Flatidae) (Fig. I, Table I). The four species that comprised the feeding assemblages occupy similar habitats and overlap in seasonal distribution, food plants, and general oviposition sites (i.e., woody tissue) (Wilson 1980). Since so many aspects of their biologies are similar, it would seem disadvantageous for these planthoppers to form assemblages that put them in direct com-

Table 1. Number of planthopper nymphs occurring in various mixed species feeding assemblages.

Number of nymphs in each feeding assemblage combinations

Species composition of feeding assemblageb Ac As Mp Ov

Ac- AS-Mp-OV Ac-AS-Mp Ac-AS-OV As-Mp-Ov Ac-As Ac-Mp Ac-Ov AS-Mp AS-Ov MpOk Total 69 8 1 46 126 aAc = A. conica. As = .A. seprenrrionalis. Mp = .tf. pruinosa. OV = 0. uenusta. hhs hc-Mfl\ combination mas not found.

l~anof a dissertarion submitted to Southern Illinois University at Carbondale by the senior author in Fa1furnot of the requirements of the Ph.D. degree in Zoology. -Department of Biological Sciences, California State University, Chico, CA 95929. 3~e~artmentof Zoolog,'. Southern Illinois University, Carbondale. IL 62901. THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Fig. 1. Planthopper nymphs feedlng on black locust stem (Robinia pseridoucucia L.).A, A. conica; B, 0. cenusto; C. A. s~,ptenfrio~~ulis. petition for a food source. Feeding in assemblages must, therefore, outweigh any competi- tive disadvantage. For exampIe, the feeding assemblages were often surrounded by copious amounts of wax produced by the planthopper nymphs; this wax may provide a barrier to other arthropods (Hepburn 1967). or may reduce predation by concealing the nymphs from natural enemies. Also, nymphs in assemblages are clumped in space, which may reduce predation from predators that search randomIy. Further study of this phenomenon in planthoppers is needed. THE GREAT LAKES ENTOMOLOGIST

ACKNOWLEDGMENTS

We wish to thank the following faculty members of Southern Illinois University at Carbondale, for their critical reviewsof the manuscript: Drs. R. A. Brandon. B. M. Burr, W. G. Dyer. Department of Zoology, and R. H. Mohlenbrock. Department of Botany. We also thank Ms. B. C. Weber, Forestry Sciences Laboratory, North Central Forest Experiment Station, U.S.D.A., Carbondale, for the photograph of planthopper nymphs. This research was partially funded by the U.S.D.A. Forest Service (Cooperative Research Agreement No. 13-495).

LITERATURE CITED

Hepburn, H. R. 1967. Notes on the genus Epiptera (Homoptera: Achilidae). J. Georgia Entomol. Soc.2:78-80. Wilson, S. W. 1980. The planthoppers, or Fulgoroidea, of Illinois with information on the biology of selected species. Ph.D. thesis. S. Illinois University Carbondale. THE GREAT LAKES ENTOMOLOGIST

OBSERVATIONS ON THE BIOLOGY OF THE SLENDER SEEDCORN BEETLE, CLlVlNA IMPRESSIFRONS (COLEOPTERA: CARABIDAE)

Robert D. ~auschland Lois M. ~auschz

ABSTRACT

The slender seedcorn beetle, Clirina impressijirons, is known to attack germinating corn seed, but field and laboratory studies of its life history indicate it may sometimes play a beneficial role. Feeding studies show that adults and larvae are primarily carnivorous but will feed on corn seed when their primary food is absent or in short supply. Certain behav- ioral aspects indicate that populations of this beetle might be attracted to selected agricultural areas where adults and larvae could serve as predators of soil pests of corn. Much of the activity of these beetles is governed by temperature. with low temperatures below 16 and 10°C causing either a slowing or cessation respectively of several biological activities.

The slender seedcorn beetle, Clivina impressifrons Leconte (Coleoptera: Carabidae), is occasionally a severe pest on corn in Illinois (Forbes 1894. Webster 1906, Phillips 1909). The damage is done to seeds, usually during cool, damp weather. Phillips (1909) presented a general picture of the life history of impressifrons. Kirk (1971) seldom found impressifrons in cultivated fields in South Dakota. In Illinois it is commonly found in pitfall traps in row crops, and is abundant in light traps.

FIELD COLLECTIONS

Adults were collected with a black-light trap and several pitfall traps. Ten pitfall traps were placed randomly throughout a field with a history of continuous corn production and examined daily (except weekends) from mid-April to late October. The black-light trap was operated continuously during the same period and collections were gathered daily, including weekends. One hundred females from each collection were dissected to determine ovarian development and sex. The remainder were destroyed except for 200-300 adults. Those retained were placed in a 150 rnm diameter plastic container partly fdled with moist sand, maintained at approximately IYC, and supplied with fresh, heat-killed house fly (Musca domestics Linnaeus) eggs. Very little cannibalism occurred when there were enough house fly eggs.

LABORATORY REARING

When eggs were needed, adults were either collected from the field or transferred from storage to fresh containers at ZPC. Egg production by storage females was about half that of field-collected females. Oviposition began in 2-3 weeks, continued for 24weeks, then ceased and was not resumed. Eggs were processed as described by Pausch (1979). Larvae and pupae were reared as described by Pausch (1979). The Larvae of impressifrons are carnivorous, and if two or more are confined together, even with ample food, only one survives.

'Illinois Natural Histor) Sun-e)' and Illinois Agricultural Experiment Station, Urbana, IL 61801. ZOriginal Cataloging. Libran.. Cniversity of Illinois, Urban, IL 61801. 190 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Recording thermographs, located near the light trap, continuously recorded air temperatures at 1.5 m above the soil surface, and soil temperatures at a depth of 20 cm. Rainfall was also measured at the light trap every 24 h. Hourly weather data were also available from a National Oceanic and Atmospheric Administration weather station approximately 8 km from the study site.

LIFE HISTORY STUDIES

Temperature governs several aspects of the life history of impressifrons. The relationship between temperature and oviposition, egg hatch, larval development, and adult longevity was studied at 10, 16, 21, 27 and 32°C. Cultures of field-collected adult impressifrons were maintained at each test temperature and were examined twice-daily for newly laid eggs. All eggs found in the 0800 examination were discarded. Those eggs found in the 1600 examination were of known age and were retained. Each egg (no more than 8 h old) was placed on moist filter paper in a petri dish, maintained at one of the test temperatures. and observed daily for hatch. After four weeks it was assumed that any unhatched eggs were dead and they were discarded. Twenty-five larvae were maintained individually in sphagnum moss at each of the test temperatures, and transferred to a fresh vial twice-weekly to observe their condition. Every other day, approximately 25 heat-killed house fly eggs (more than a 3rd instar larva could consume) were dropped on the moss as food. Observations continued until death or pupation occurred. One hundred adults (50 males: 50 females) were placed in rearing containers, supplied with ample fresh, heat-killed house fly eggs, and maintained at the test temperatures for 20 weeks. Surviving beetles were counted every two weeks.

OVARIAN DEVELOPMENT AND SEX RATIO

Beetles were captured in the light trap from early spring until approximately mid-July. One hundred females from each week's collection were dissected and the state of ovarian development observed. The females were arbitrarily classified into four stages of ovarian development: (I) no apparent egg development; (2) slight swellings over part, but not all the ovarioles; (3) moderate swellings throughout all the ovarioles; and (4) at least one full-sized egg apparent. One hundred individuals, selected randomly from the light trap catch, were examined each week to determine the sex ratio.

SEASONAL DISTRIBUTION AND FLIGHT ACTIVITY

Seasonal occurrence of adult impressifrons was determined from pitfall and light-trap collections. The pitfall traps did not collect many impressifrons but they did record adult activity much earlier in the spring than did the light trap. Collections in the light trap began as soon as the nights were warm enough to allow flight. Nightly, and several instances of hourly, collections were correlated with weather data.

FOOD PREFERENCES

C. impressifrons adults attack seed corn during the spring but larvae are not known to do so. Adults and larvae, starved for 24 h were presented three types of food (commercially available seeds, artificial diets, and animal food) for 24 h. During and after this period, the food was examined for evidence of feeding. The artificial diets were a corn borer diet as described by Guthrie et al. (1%5) and a modification prepared with the addition of 25% by weight of homogenized house fly larvae. 1980 THE GREAT LAKES ENTOMOLOGIST 191

RESULTS

Most eggs were laid at 27" and 32°C. Several were oviposited at 2I0C but none at 10 and 16°C (Table 1). Although field-collected adults were used and it is unknown if any eggs had been laid previously in the field, it is nonetheless clear that temperature influences oviposition. Eggs developed faster at higher temperatures (Table 1). From 16°C to 32°C a high per- centage of hatch occurred with only the speed of development being governed significantly by temperature. No hatch occurred at 10°C. Temperature also influenced the rate of larval development to a great extent. At IWC, no larvae pupated and all were dead, still in the 1st instar, by the end of the third week. Approximately 20% pupated in an average of 43 days at 16°C. The speed of lava1 development increased and mortality declined at the higher temperatures. Eighty-eight percent of the larvae pupated in an average of 30 days at 27°C and 84% in 21 days at 32°C. The age of the beetles at the start of testing for longevity was unknown because they had been field-collected. The influence of temperature on the longevity of impressifrons was however quite evident (Table 2).

OVARIAN DEVELOPMENT AND SEX RATIO

Stage 4 females appeared in the light trap from May to late July (Table 3) and represented approximately 13% of the population. except for the first two weeks of collections. when they constituted 33% of the population. Adults were attracted to the light trap in an approximate 1-3 male to female ratio (Table 3).

SEASONAL DISTRIBUTION AND FLIGHT ACTIVITY

Adults first appeared in the light trap during May, when nighttime temperatures reached 20°C. Very few adults were captured after July even though nighttime temperatures remained high enough for flight activity in September and occasionally in October (Table 4). Over a 4 yr period an average of approximately 13,000 individuals was captured during the first week of July, 3600 the second week, 1800 the third week, and 500 the fourth week. Almost 70% of the July collection was taken in the first week. The reason for the cessationof flight after July is unknown. The occurrence of population peaks was influenced by temperature and reflected in the light trap collections. When spring temperatures averaged above normal, the peak popula-

Table 1. Relationship of temperature to oviposition and egg hatch of C. impressifrons

Egg ~atch~

No. Eggs Laida Days to Hatch Temperature Percent ("c) Total Range Mead: Mean Range Hatch

ratio 1 feutak 5 mak with I0 females used at each temperature. b10 -.remperature. Call indiduak dead b) 11th aeek dmr. 192 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Table 2. Relationship of temperature and adult longevity in C. impressifrons.a

Number Alive

Week 10°C 16°C 2 1°C 27°C 32PC

"100 field-collected adults were tested at each temperature

Table 3. Seasonal occurrence of ovarian stages and sex ratio in C. impressifrons as determined from pooled light trap collections, 1972-1975.a

- % of Females 9% of Collection Ovarian Stage Time Period #I #2 #3 #4 6 0

1-14 May 15-3 1 May 1-14 June 15-30 June 1-14 July 15-3 1 July Seasonal Mean 60 13 7 17 27 73

'1Collections after 1 August were insignificant and are not included in this table.

Table 4. Number of C. impressifrons adults captured in a black-light trap. Urbana, Illinois, 1972-1975.

Month 1972 1973 1974 1975 Mean Max.1-NightCollection

April 0 0 0 0 0 0 May 2 1468" 266 211 21199a 10786 8784 June 10236 21296a 5750 37978a 18815 9889 July 2629 46362a 16703a 9613 18826 14727 August 190 136 138 23 122 116 September 60 6 5 0 18 53 October 0 0 0 0 0 0 aMonths with above-average nighttime temperatures. 1980 THE GREAT LAKES ENTOMOLOGIST 193 tion occurred early, during May or June. When spring temperatures averaged below normal, the peak shifted to June or July (Table 4). Adults were active in the soil as early as mid-March as evidenced by pitfall collections, but flight did not begin until the nighttime air temperature reached approximately 20°C. The length of time that temperatures remained above this threshold determined the number of individuals attracted to the light trap. On nights when hourly collections were being made, impressifrons continued to be trapped until the air temperature dropped below 20°C and flight activity ceased abruptly. On nights when the temperature did not fall below this threshold, flight activity continued until dawn when the numbers being trapped gradually tapered off. Rainfall also modified flight activity. Light trap collections did not decline on nights with 0.4 inch or less precipitation. Flight activity declined drastically and remained very low for several days following a rainfall of 0.5 inch or more, particularly if the rainfall was intense and occurred over a short period of time. Presumably a heavy rainfall closed exit holes and cracks in the soil, and until the beetles were able to re-open these exits, no flight activity took place. Wind speeds of 13 mph or less, the presence or absence of moonlight or cloud cover, and fluctuations in barometric pressure had no apparent effect on flight.

FOOD PREFERENCES

House fly and northern corn rootworm (Diabrotica longicornis Say) eggs and larvae were by far the most preferred food of both larval and adult impressifrons. Feeding on corn seeds was minor and only softened kernels were attacked (Table 5), and only by adults. Larval irnpressifrons ignored commercial seeds and artificial diets but fed voraciously on animal foods. Larvae appear not to attack germinating corn.

Table 5. Food preferences of starved C. impressifrons adults and larvae.

Degree of Preferencea

Food Adults Larvae

Commercial seeds Corn kernels (hard) Corn kernels (softenedlb Soybeans (hard) Soybeans (softenedlb Alfalfa (hard) Alfalfa (softened$' Artificial diets Dog food pelletsc Com borer died Modified corn borer diet Animal food House fly eggs House fly larvaee Corn rootworm eggs Corn rootworn larvaee

a- = no apparent feeding. - = on]! slight el idence of feeding. -i = definite but moderate feeding, T - - - pronoud feeding actit~ities.- - - - = immediate and pronounced feeding. b~oiledfor 15 minutes. CFriskies Pupp! Food dCu~hrie.1%. eSeuIy harcbrd. THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 4

DISCUSSION

Although impress$runs may cause economic damage, evidence in this study indicates it might also be beneficial. Reports in the literature indicate that seed corn beetle damage is most severe during cool, damp springs. This may very well be true. but the actual reason for damage occurring under these climatic conditions is not completely understood. We have shown that impressijrons adults prefer animal food, but will attack plants (e.g., seed corn) when their preferred diet is not available or in short supply. We have also shown that adults and larvae will readily consume corn rootworm eggs and newly hatched larvae in the laboratory. Cold, damp spring weather supports seed corn beetle damage because the seed corn germinates slowly and there is a reduction of their primary animal food sources. It also prevents aerial dispersal of the beetles (flight activity occurs only when nighttime temperatures exceed 20°C) and slows down any movements out of the corn fields by crawling. Once the seed corn has germinated and is no longer attractive to the beetles, it can be assumed that adults and larvae will search out and eat a wide variety of animal food, with corn rootworm eggs and newly hatching lalrvae being at least a part of their diet if present in the soil. Wyman et al. (1976) attracted Agonoderus comma (Fabricius) and Agonoderus leconrei Chandoir (other seed corn beetles) into crucifer plots and found that cabbage maggot damage was significantly reduced. Wishart et al. (1956) also showed that several species of carabids would feed on large numbers of Hylemyu brassicae (Bouche) eggs. Since other carabids feed on edaphic pest insects, and impressijirons is strongly attracted to lights, particularly black light, the possibility of attracting these beetles into corn fields, after germination, to feed on corn rootworm eggs and larvae as well as other soil-inhabiting pests, should not be ignored.

LITERATURE CITED

Forbes, S. A. 1894. p. 15 in: Eighteenth report State Entomologist on the noxious and beneficial insects of the State of Illinois. Guthrie, W.D., E.S. Raun, F. F. Dicke, G. R. Pisho, and S. W. Carter. 1%5. Laboratory production of European corn borer egg masses. Iowa State J. Sci. 40:6543. Johnson, N. E. and R. S. Cameron. 1969. Phytophagous ground beetles. Ann. Entomol. Soc. Amer. 62:909-914. Kirk, V. M. 1971. Ground beetles in cropland in South Dakota. Ann. Entomol. Soc. Amer. 64:238-241. Pausch, R. D. 1979. Observations on the biology of the seed corn beetles. Srenolophus comma and Srenolophus leconrei. Ann. Entomol. Soc. Amer. 72:2&28. Phillips, W. J. 1909. The slender seed-corn ground-beetle. USDA Bull. 85, Part 11. Webster, F. M. 1906. The slender seed-corn ground beetle. USDA Circ. 78. Wishart, G., J. F. Doane and G. E. Maybee. 1956. Notes on beetles as predators of eggs of Hylemya brassicae (Bouche) (Diptera: Anthomyiidae). Canadian Entomol. 88:63&639. Wyman, J. A., J. L. Libby, and R. K. Chapman. 1976. The role of seed-corn beetles in predation of cabbage maggot immature stages. Environ. Entomol. 5:249-263. THE GREAT LAKES ENTOMOLOGIST

LOCALIZED FIELD MIGRATION OF THE ADULT ALFALFA WEEVIL, CLOVER LEAF WEEVIL, AND CLOVER ROOT CURCllLlO (COLEOPTERA: CURCULIONIDAE), AND ITS IMPLICATION FOR A FALL PEST MANAGEMENT PROGRAM'

Robert D. Pausch, Stephen J. Roberts, Robert J. Barney and Edward J. ~rmbrustz

ABSTRACT

The migration of the alfalfa weevil, clover leaf weevil and clover root curculio from aestivation sites into adjacent alfalfa fields was monitored with several different trapping systems. It was found that the initial dispersal back into alfalfa fields following aestivation was a slow gradual movement across the soil surface with flight occurring only after the insects had been in the alfalfa for several days.

The alfalfa weevil, Hypera posricn (Gyllenhal), clover leaf weevil, Hypera puncrara (Fabricius), and clover root curculio, Sirona hispidula (Fabricius) are pests of alfalfa in Illinois. The adults of each species leave the alfalfa field during the summer and return to the field later that same year following a period of aestivation. Of the three insects, only the movements of the alfalfa weevil are well documented. The migration of H. postica out of alfalfa fields into aestivation sites is well established (Prokopy and Gyrisco 1965, Prokopy et al. 1967). Also well documented is the reentry movement (Poinar and Gyrisco 1962, Prokopy and Gyrisco 1963, Prokopy et al. 1967, Barney et al. 1978). Prokopy et al. (1967) and Prokopy and Gyrisco (1%3) found in the state of New York that the reentry in the fall occurred as a flight. Field observations in Illinois concurrent with other H. posrica research, confirmed the movement back into the alfalfa fields in the fall but indicated that the reentry dispersal was not by flight but by a slow, gradual movement, initially and primarily on the ground. The purpose of this investigation was to examine in more detail, the reentry migration of all three species, to determine its applicability to afall adult control program. The movement of all three species was observed but the major emphasis of this investigation was directed toward H. posrica because it was the most economically important insect.

MATERIALS AND METHODS

This investigation was conducted in a 259 ha study area in Washington County in southern Illinois. The three alfalfa fields used were bounded on at least one side by an oak-hickory forest. Field #1 was 3.2 ha and bounded on the south by forest, on the east by corn, and the north and west sides by soybeans. Field #Z was 9.7 ha and bounded on the east by forest, on the north and south by soybeans, and on the west by another alfalfa field. Field #3 was 12.2 ha and bounded on the north and east by forest. on the south by pasture, and on the west by

hhis research has heen financed in part w-ith Federal funds from the Environmental Protection Agency under grant number L80014-5. and in part by the Illinois Agricultural Experiment Station and the Illinois Natural Histow Suney. The contents do not necessarily reflect the views and policies of the Environ- mental Protection A4gencynor does mention of tmde or commercial products constitute endorsement or re~ommendationfor use. -Illinois Natural Histon- Sun-e and Illinois Agriculrural Experiment Station, Urbana. IL 61801. 196 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4 farmstead and forest. The post-aestivational movements of adult weevils were monitored by the following sampling methods: sweeping, square foot samples, emergence traps, pitfall traps, and flight traps. Sweep Net and Square Foot Samples. Fields with populations of H. posticu larvae were monitored during the spring and summer. When sweep net samples of alfalfa fields indicated that the adult weevils had emerged from pupation and had left the fields for their aestivation sites, square foot samples were taken in the woodedges to determine the size of the population entering aestivation. Roberts et al. (1979) showed that the majority of each species aestivated in woodedges. The square foot sampling was done using the pyrethrin drench method described by Klostermeyer and Manglitz (1976) except that the drench was only 1.0 cm3 pyrethrin/3.78 litres water. Emergence Traps. One hundred emergence traps as described by Roberts et al. (1978) were placed randomly in the woodedge along the three alfalfa fields. Thirty traps were placed in field 2, and 35 each in fields 1 and 3. Pitfall Traps. Forty-two linear pitfall traps (Pausch et al. 1979) were placed in a staggered pattern (Fig. I) in the alfalfa in field 3. Each trap was aligned parallel to the woodedge with each successive row of traps being 5 ft farther out into the alfalfa field. The contents of each trap was collected weekly and taken to the laboratory for examination. Flight Traps. Eight flight traps as described by Roberts et al. (1978) were placed in each of fields 2 and 3. The contents of each trap was collected weekly and taken to the laboratory for examination. The relative location of all the trapping systems employed is shown in Figure I.

RESULTS

Square Foot Samples. Our square foot sampling showed the highest populations of H. postica and S. ltispidula to be located in fields 1 and 3 with field 2 having the largest number of H. punctatu (Table I). Emergence Traps. Data obtained from the emergence traps (Table 2) showed that S. hispidula and H. prcnctata emerge from aestivation at approximately the same time, while H. posricu leaves aestivation 34weeks later. Between 75% and 85% of the populations of

EMERGENCE TRAPS 0 0

0 f'L/ --.-/''-..-J~\-~~

0 0 ALFALFA 0

LINEAR PlTFALL TRAPS n C-, 0

Fig. I. Relative location of trapping systems in relation to aestivation sites (woods) and alfalfa field, Washington County, IL, 1978. 1980 THE GREAT LAKES ENTOMOLOGIST 197

Table 1. Mean number and standard error of H. postica, S. hispidula, and H. puncfata in square foot samples taken in woodedges in fall 1978, Washington County, IL.

H. postica S. hispidula H. punctata - - - Field No. x S.E. x S.E. x S.E.

Table 2. Total number of adult H. posrica, S. hispidula, and H. punctata captured in emergence traps set in aestivation areas, Washington County, IL, 1978.

August September October November

Field 1 9 16 22 29 6 12 19 26 3 10 18 24 1 8 15

H. posrica 2 0 5 1 0 54542 21 7 0 2 0 10 00117 00 0 1072 485148 281 0 S. hispidula 756 211 80 10 6 0 2 0 I 0 0 84 44 3713 2 3 13 00 0 180 172 163 57 6 1 1 2 0 0 0 H. puncrara 2 0 10 0010 000 15 6 73 3010 000 8 1181 0000 000 a = trap not yet in place.

H. posrica in all three fields emerged from aestivation during the period 10 October to 1 November. Although the emergence of S. hispidula appears to be much more extended in time (from early August to early October), an average of 714 emerged in the 3-week period of 29 August to 19 September. H. puncrara also emerged from aestivation earlier than H. postica. Data from the emergence traps showed that during the same three week period when the majority of S. hispidula were leaving aestivation, an average of 80% of the H. punctata population also left aestivation. Pitfall Traps. Data from the pitfall traps (Table 3) clearly show the gradual movement of all three species of beetles back into the alfalfa fields after aestivation. Although a few beetles were captured throughout the entire trapping zone during the monitoring period, the timing and location of the peak collections appear to be related to the relative time of emergence from aestivation. In the case of H. posrica, the largest pitfall trap collections were made during the period 19 October to 1 November. Emergence trap data show that 19 October was the date when the largest number of H. posrica left aestivation and entered the alfalfa field. After this point in time, the number of beetles entering the field declined as was shown in the pitfall trap collections. It was also reflected in the declining numbers of adults emergingfrom aestivation and being collected in our emergence traps. By 1 November the crest of the wave of beetles moving outward into the alfalfa fields reached the outer edge of the trapping zone and passed farther out into the field. Pitfall trap collections also declined at this time due 198 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Table 3. Average number of adult H. postica, S. hispidula, and H. punctata captured in pitfall traps located in alfalfa fields, Washington County, IL, 1978.

Distance into September October November alfalfa field (ft) 13 19 26 3 10 19 24 1 9 15

H. postica 5.0 11.7 4.8 17.8 1.7 7.8 .5 2.8 .7 3.0 .7 1.8 .5 .8 S. hispidula 10.5 3.7 10.7 3.3 21.2 11.5 15.7 7.5 17.2 15.3 13.7 14.7 10.7 22.0 H. punctata .5 .2 .3 .5 2.7 1.5

a = traps not yet in place possibly to a combination of several factors: (1) the movement of the beetles farther out into the field beyond our trapping zone, (2) all adults having left aestivation, (3) the start of flight activity (Table 4), and (4) the onset of cold weather. This post-aestivation migration was also apparent in the movements of both S. hispidula and H. punctata although the timing was slightly different. The movement of S. hispidula into the alfalfa fields started considerably earlier than H. postica as would be expected from their earlier emergence. Their movement through the field, however, was slower than H. postica since the crest of the movement reached the outer edge of the trapping zone only two weeks before H. postica after having started almost two months earlier. The movement of H. punctata was somewhat disjointed. Peak numbers of H. punctata did not reach the outer edge of the trapping zone until mid-November even though their entrance into the alfalfa fields started earlier than H. postica and their movement continued throughout the fall. Flight Traps. Flight activity of H. postica was not recorded until late October although they were moving out into the alfalfa fields from early September. S. hispidula entered the alfalfa fields earlier than did H. postica and their flight activity was first recorded at a proportionately earlier time. No flight activity was observed for H. punctata. 1980 THE GREAT LAKES ENTOMOLOGIST 199

Table 4. Total number of adult H. postica, S. hispidula, and H. punctata captured in flight traps located in alfalfa fields. Washington County. IL, 1978.

September October November

Field No. 13 19 26 3 10 18 24 1 8

H. postica 0 0 0 1 3 0 0 1 3 10 0 0 1 4 13 S. hispidula 7 0 0 0 5 12 0 0 1 4 19 0 0 1 9 H. punctata 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

CONCLUSIONS

The results from these studies clearly indicate that the movement of H. postica, H. punctata and S. hispidula into alfalfa fields after leaving aestivation is at first a gradual movement across the soil surface. Flight occurs only after the beetles have been in the field for a considerable length of time. Emergence of H. postica from aestivation occurs later in the fall than does the emergence of H. punctata and S. hispidula. It is our assumption, based largely on pitfall and flight trap data, that the initial dispersal into the alfalfa fields subsequent to aestivation has feeding as its prime mover. Since these adult beetles have not fed during the 4-5 months of aestivation, their stored energy resources in the form of fats are at a low level. This has been quite graphically demonstrated by Tombes (1964). Until adequate feeding has been accomplished and their energy resources restored, flight might not be possible for the adult beetles. In addition to the energy requirements for flight, prevailing temperatures may influence not only aerial but surface movement. During this post-aestivation dispersal back into alfalfa fields, the majority of the population, at some time, is concentrated in a relatively narrow band at the edge of the alfalfa fields bordering the woodedge. Since flight activity does not start until late in the migration, the beetles become concentrated, are accessible, and are relatively immobile within this narrow band. These are conditions which argue well for an adult fall control program with insecti- cides. Research is currently underway to more fully investigate various ways of controlling these three species of beetles in a fall adult control program.

LITERATURE CITED

Barney. R. J.! S. J. Roberts, R. D. Pausch, and E. J. Armbrust. 1978. Fall termination of aestivation and field dispersal of the alfalfa weevil (Coleoptera: Curculionidae) in Illinois. Great Lakes Entomol. 11:255-259. Klostermeyer, R. E. and G. R. Manglitz. 1976. Sampling for summer alfalfa weevils; pyrethrin drench vs. soil sampling. Proc. North Cent. Br., Entomol. Soc. Amer. 31:28. Pausch, R. D.. S. J. Roberts, R. J. Barney, and E. J. Armbrust. 1979. Linear pitfall traps, a modification of an established trapping method. Great Lakes Entomol. 12: 149-151. 200 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Poinar, G. 0..Jr. and G.G. Gyrisco. 1%2. Flight habits of the alfalfa weevil in New York. J. Entomol. Soc. Amer. 55:265-266. Prokopy, R. J. and G. G. Gyrisco. 1963. A fall flight period of the alfalfa weevil in New York. J. Entomol. Soc. Amer. 56:241. . 1965. Summer migration of the alfalfa weevil, Hypera posticn (Coleoptera: Curculionidae). Ann. Entomol. Soc. Amer. 58:63@-641. Prokopy, R. J., E. J. Armbrust. W. R. Cothran, and G. G. Gyrisco. 1967. Migration of the alfalfa weevil. Hypera posficn (Coleoptera: Curculionidae), to and from aestivation sites. Ann. Entomol. Soc. Amer. 60:2631. Roberts, S. J., R. D. Pausch, E. J. Armbrust, and R. J. Barney. 1978. Two trapping systems to determine incidence and duration of migration of adult alfalfa weevils, Hypern postica (Coleoptera: Curculionidae). Great Lakes Entomol. 11249-253. Roberts, S. J., E. J. Armbrust, R. J. Barney, and R. D. Pausch. 1979. Seasonal population census of the alfalfa weevil, clover root curculio, and clover leaf weevil (Coleoptera: Curculionidae) in southern Illinois. Great Lakes Entomol. 12: 141-148. Tombes, A. S. 1964. Respiratory and compositional study on the aestivating insect, Ifypern posticn (Gyll.) (Curculionidae). J. Ins. Physiol. 10:997- 1003. THE GREAT LAKES ENTOMOLOGIST

ANNOTATED LIST OF TRICHOPTERA COLLECTED ALONG FURNACE RUN OF THE CUYAHOGA VALLEY NATIONAL RECREATION AREA IN NORTHEASTERN OHIO'

C. petersen2 and B. A. ~oote3

ABSTRACT

Information on seasonal distribution of 46 species of Trichoptera occurring along a small, unpolluted, woodland stream in northeastern Ohio is presented. Two species. Hydropsyclle hageni and Polycentroprcs interruptus, are newly recorded for the state.

Approximately 165 species of Trichoptera have been reported from Ohio. Almost half of these have been recorded within the last five years from the West Branch of the Mahoning River in Portage County, a tributary of the Ohio River (McElravy 1976, McElravy et al. 1977, McElravy and Foote 1978). Ross' (1944) monograph of the Trichoptera of Illinois, Tkac's (1973) survey of Stebbins Gulch in Geauga County, Ohio, and the Lake Erie faunal studies of Marshall (1939) and Horwath (1964) record the remaining species. This paper summarizes the results of a faunal survey of the Trichoptera obtained from Furnace Run, a small tributary of the Cuyahoga River of the Saint Lawrence drainage.

DESCRIPTION OF STUDY AREA

Furnace Run originates in the southeastern corner of Cuyahoga County, Ohio, at an elevation of 375.6 m. The stream flows in a southeasterly direction through mesophytic forests and rural areas within the Cuyahoga Valley National Recreation Area of Summit County before joining the Cuyahoga River at an elevation of 207.3 m. Its total length is 5.5 km; its average gradient, 29.1 mlkm. It drruns a watershed of approximately 5.7 km2 (Frost and Smith 1959) and is a fourth order stream. The stream substrate consisted of cobble (Cummins 1962) at each of the four collecting stations included in the study (Petersen 1979). This rocky material is derived from Quater- nary glacial till of the Wbuck Lobe, except for a stretch between stations 3 and 4 where sandstone bedrock of the Pennsylvanian System has been exposed (Miller 1970). Of the four collecting stations, station 1 (near I80 crossing) was located nearest the source of the stream at an elevation of 315 m. The surrounding forest was dominated by oak. Brushwood Lake, of approximately 12,700 m2, was located nearby. From its origin and through s'tation 2 (near Brush Road crossing), the stream profile is relatively flat. After station 2, the topography becomes steep and hilly but then levels out again where the stream flows across the floodplain of the Cuyahoga River. Station 3 (near Rt. 303 crossing) was situated along the stretch of the stream having the steepest gradient, while station 4 (near Wheatly Road crossing) was located near its junction with the Cuyahoga River. Selected water quality measurements from Furnace Run (Table I), indicate that the stream is relatively unpolluted (Carlson 1978).

'~esearchsupported. in part. by a grant from the Oak Hill Center for Environmental Studies, PeFsula. Ohio. -Department of Entomology. University of Georgia, Athens, GA 30602. 3Dep~mentof Biologcal Sciences. Kent State University, Kent, OH 44242. 202 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Table 1. Selected water quality measurements (Station 4).

Fecal Coliform -Temp. BSD mg/l NoJ1000 ml Month X * SD X r SD X * SD

MATERIALS AND METHODS

Adult Trichoptera were collected between 7 June and 28 September 1978, and during May 1979. The collecting device, a Pennsylvania light trap (Frost 1957) fitted with a 15-watt incandescent light bulb, was suspended from a tree limb within 4 m of the stream. At least one sample was taken weekly from one of the four stations, with each sampling period lasting one hour. Larvae were collected between 6 June 1978 and 21 May 1979. All were obtained by random rock picking or by the use of a kick screen placed in riffle and pool areas of the stream. Ethyl alcohol (75%) was used to kill and preserve both adults and immatures. Adults were identified to species, whereas most of the larvae could only be identified to genus. Voucher specimens are on deposit in the Kent State University Collection.

ANNOTATED LIST OF SPECIES

Information on the species of Trichoptera is summarized for each of the four collecting stations. The flight period of each species is given, followed by the number of each sex collected.

Family RHYACOPHILIDAE

Rhyacophila curolir~aBanks. One male was collected from sta. 2 on 29 August 1978. At least in northeastern Ohio, this species is found only in streams of the highest water quality.

Family PHILOPOTOMIDAE

Chimarra atterima (Hagen). Larvae were taken from sta. 3. Adults are known to occur from the middle of April through May (Ross 1944). C. obscura (Walker). August; sta. 1 (2 males), sta. 3 (1 female).

Family HYDROFTILIDAE

Oxyethirajorciputa Mosely. September; sta. 3 (2 males, 11 females). This species was taken only along areas of steepest gradient. 0.pallidu (Banks). July-September; sta. 1 (1 male, 4 females), sta. 2 (1 male), sta. 3 (17 males, 87 females); sta. 4 (1 male). Emergence was heaviest during September. Orthotrichia aege$asciella (Chambers). June-September; sta. 3 (4 females). Hydroptila ajux Ross. August, September; sta. 3 (7 females). The species was far more abundant along the nearby Cuyahoga River, perhaps a reflection of its tolerance to organic pollution. H. armata Ross. July, August; sta. 3 (1 female). 1980 THE GREAT LAKES ENTOMOLOGIST 203

H. consirnilis Morton. JuneSeptember; sta. 1 (12 males, 228 females). sta. 2 (187 females), sta. 3 (12 males, 5091 females), sta. 4 (20 males. 183 females). This was, by far, the most abundant species of the order with a peak of emergence occurring during early August. Most adults were collected along areas of steepest gradient. H. hamata Morton. July; sta. 3 (1 female). H. perdita Morton. July, August; sta. 4 (1 male, 52 females). This may be a pollution- tolerant species, as most adults were taken along the lower end of Furnace Run near its junction with the Cuyahoga River. H. rr,aubesiana Betten. June, September; sta. 3 (3 females). Ochrotrichia spinosa (Ross) June, September; sta. 3 (2 males, 3 females), sta. 4 (1 male).

Family POLYCENTROPODIDAE

Polycentropus centralis Banks. JuneSeptember; sta. 1 (5 males, 2 females), sta. 2 (5 males), sta. 3 (I male, 55 females), sta. 4 (3 males, 7 females). Most of the specimens were taken in late August along the faster flowing waters of sta. 3. P. confusus Hagen. One male was captured at sta. 2 on 29 August 1978. P. interruptus (Banks). One male and one female were taken at sta. 4 on 7 June 1978. This is the first record of this species in Ohio. Nyctiophylax affinis (Banks). One female was collected along sta. 3 on 24 July 1978.

Family PSYCHOMYIIDAE

Lype diversa (Banks). One female was taken along sta. 3 on 14 August 1978.

Family HYDROPSYCHIDAE

Hydropsyche betteni Ross. June, August, September: sta. 1 (20 females), sta. 2 (1 female), sta. 3 (10 females), sta. 4 (2 females). H. bronta Ross. MaySeptember; sta. 1 (3 females), sta. 2 (2 males, 2 females), sta. 3 (4 males, 17 females), sta. 4 (22 females). Most of the adults were taken along the lower stretches of the stream. H. dicantha Ross. August: sta. 3 (1 female). H. hageni Banks. One female was collected at sta. 4 on 7 August 1978. This represents the first record of this species in Ohio. H. slossonae Banks. MaySeptember; sta. 1 (17 females), sta. 2 (6 males, 2 1 females), sta. 3 ( 16 males, 73 females). The peak of emergence occurred during August. Most of the adults were collected at sta. 3 along the steepest gradient of the stream. H. sparna Ross. JuneSeptember: sta. 1 (4 females), sta. 3 (1 male, 3 females), sta. 4 (6 males, I I females). Most of the specimens were taken during August along the slower waters of sta. 4. Cheumatopsyche aphanta Ross. JulySeptember; sta. 3 (5 females), sta. 4 (1 male, 6 females). C. campyla Ross. July, August; sta. 4 (2 females). This species was more common along the nearby Cuyahoga River, reflecting its tolerance to pollution. C. oxa Ross. MaySeptember; sta. 1 (12 Females), sta. 2 (2 males, 34 females), sta. 3 (17 females), sta. 4 (4 males, 7 females). C.pettiti (Banks). June, July; sta. 1 (2 males, 2 females), sta. 3 (1 male, 1 female), sta. 4 (2 females). C. speciosa (Banks). JuneSeptember; sta. 1 (12 females), sta. 2 (13 females), sta. 3 (30 females), sta. 4 (1 1 females). Most adults were collected during late August. Diplectrona modesta Banks. One female was taken from sta. 4 on 7 June 1978.

Family PHRYGANEIDAE

Phryganea sayi Milne. One female was collected at sta. 1 on 8 August 1978. THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Family LEPTOCERIDAE

Triaenodes abus Milne. One female was taken at sta. I on 8 August 1978. T. Jlauescens Banks. August; sta. 3 (1 male). T. ignitus (Walker). One male was collected at sta. 2 on 31 July 1978. T. tardus Milne. June, JulySeptember; sta. 1 (3 males), sta. 2 (3 males), sta. 3 (9 females), sta. 4 (6 females). Most adults were collected during August. Ceraclea tarsipunctata (Vorhies). One male was collected at sta. 4 on 19 July 1978. Lrptocerus americanus (Banks). August; sta. 3 (1 male, 1 female). Nectopsyche exquisita (Walker). All 5 males and 51 females were collected at sta. 4 on 7 June 1978. Oecetis cinerascens (Hagen). One female was collected at sta. 1 on 8 August 1978. 0.inconspicua (Walker). JuneSeptember; sta. 1 (3 males, 18 females), sta. 2 (1 female), sta. 3 (I male, 5 females). sta. 4 (8 males, 2 females). Adults were most common near Brush- wood Lake and along the slow flowing stretches of lower Furnace Run and the Cuyahoga River.

Family LIMNEPHILIDAE

Ironoquia punctatissima (Walker). One male and 1 female were taken on 7 September 1978, at sta. 3. Neophylax concinnus McLachlan. One female was collected at sta. 3 on 27 September 1978. Lirnnephilus consocius (Walker). A female was collected on 24 August 1878, at sta. 3. Pycnopsyche spp. Only larvae were collected during May 1978, at sta. 3 and 4. At least two species were represented.

Family HELICOPSYCHIDAE

Helicopsyche borealis (Hagen). Late August, early September; sta. 3 (2 males, 1 female). Adults and immatures were collected only along the steepest gradient of the stream.

LITERATURE CITED

Carlson, R. 1978. A comparison of water quality indices in the streams of the Cuyahoga Valley National Recreation Area. Report submitted to the Cuyahoga Valley National Recreation Area, Summit County, Ohio. Cummins, K. W. 1962. An evaluation of some techniques for the collection and analysis of benthic samples with special emphasis on lotic waters. Amer. Midl. Nat. 67:477-504. Frost, S. W. 1957. The Pennsylvania light trap. J. Econ. Entomol. 50:287-292. Frost, S. L. and R. C. Smith. 1959. Water inventory of the Cuyahoga and Chagrin River basins. Vol. I, Basin review. Div. Water, Ohio Dept. Nat. Res. Honvath, A. B. 1964. Insects of Gilbralter Island in relation to their habitat. M.A. thesis, Ohio State Univ. Marshall, A. C. 1939. A qualitative and quantitative study of the Trichoptera of western Lake Erie (as indicated by light trap material). Ann. Entomol. Soc. Amer. 32:665487. McElravy, E. P. 1976. The caddisflies (Trichoptera) occurring along the upper portion of the West Branch of the Mahoning River, Portage County, Ohio. M.S. thesis, Kent State Univ. McElravy, E. P., T. L. Arsufi, and B. A. Foote. 1977. New records of caddisflies (Trichoptera) for Ohio. Proc. Entomol. Soc. Washington. 79:5M. McElravy, E. P., and B. A. Foote. 1978. Annotated list of caddisflies (Trichoptera) occurring along the upper portion of the West Branch of the Mahoning River in northeastern Ohio. Great Lakes Entomol. 11:143-154. Miller, B. B. 1970. The Quaternary period. p. 149-164 in: P. 0. Banks and R. M. Feldman (eds.). Guide to the geology of northeastern Ohio. N. Ohio Geol. Soc. 1980 THE GREAT LAKES ENTOMOLOGIST 205

Petersen, C. E. 1979. Trichoptera and leaf processing in Furnace Run, Summit County, Ohio. M.S. thesis, Kent State Univ. Ross, H. H. 1944. Caddisflies, or Trichoptera, of Illinois. Illinois Nat. Hist. Surv. Bull. 23: 1-326. Tkac, M. A. 1973. The Plecoptera and associated aquatic insects of Stebbins Gulch. M.A. thesis, Kent State Univ. THE GREAT LAKES ENTOMOLOGIST

INSTAR HEAD WIDTHS, INDIVIDUAL BIOMASS, AND DEVELOPMENT RATE OF FOREST TENT CATERPILLAR, MALACOSOMA DlSSTRlA (LEPIDOPTERA: LASIOCAMPIDAE), AT TWO DENSI'TIES IN THE LABORATORY

Jocelyn M. Muggli and William E. ~illerl

ABSTRACT

Two colonies of forest tent caterpillar were reared at different densities on trembling aspen foliage in an environment chamber programmed for natural temperatures. Based on direct observation of molting, both colonies underwent five instars which overlapped neg- ligibly in head width. Instar head widths were relatively stable between colonies, but individual biomass was sensitive to the different densities. Instars 3-5 in the less dense colony were heavier than those of the more dense by factors of 1.3-1.9. Development to 10-20% of larvae spinning cocoons took 40 days in the less dense colony and 45 days in the more dense. Size of larvae and adults resembled that of field samples.

To produce forest tent caterpillar (Malacosoma disstria [Hiibner]) larvae of known age and development history for several research purposes, we reared colonies at two densities on natural food in the laboratory. Summarized here are instar head widths, individual biomass, and rate of development in these colonies.

METHODS

Eggs to start the rearings were obtained shortly before hatching from three neighboring trees of trembling aspen (Populus tremuloides Michx.) near Ear Falls, Ontario. In a sub- sample of 10 egg masses, total number of eggs per mass was 135-243, averaging 201 + standard deviation (SD) 38. To equalize genetic makeup, each colony of approximately 240 hatchlings was made up of aliquots from several concurrently hatching egg masses. The third day after hatching began was taken as the date of hatching (Lorimer 1979). Colonies were placed in transparent plastic garment boxes with openings in ends and tops covered by fine-mesh nylon screen for ventilation. Two box sizes were used with total volumes and volumes per hatchling as follows: 19040 cm3 and 79 cm3: 7722 cm3 and 32 cm3. Volume per larva increased as samples of larvae were removed, but a difference in density between the boxes continued. Lacking more exact density expressions, we refer to colonies in the larger and smaller boxes, respectively, as "less dense" and "more dense." Larvae were fed foliage of trembling aspen, the main host in the Lake States. Food was amply supplied in the form of branchlets 15-20 cm long held in stoppered vials of distilled water to maintain turgor. Food was replaced and boxes cleaned as necessary, never less often than every third day. The boxes were held in a 1.6 m3 controlled environment chamber programmed for a 12-h fluorescent light photophase at 22°C switching abruptly to a 12-h scotophase at IOaC. The temperatures simulated daily extremes near the Minnesota-Canada border during the natural larval period (Baker and Strub 1965).

l~orthCentral Forest Experiment Station. USDA Forest Service, 1992 Folwell Avenue, St. Paul, MN 55108. Present address of senior author: USDHEW Food and Drug Administration, 240 Hennepin Avenue. Minneapolis. MN 55401. 208 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

A sample of hatchlings was preserved when the colonies were started. Thereafter colonies were checked at intervals of 1-2 days for molting. Larvae aggregated at molting and rested for a time near their cast skins; this facilitated identification and sampling of recently molted individuals. At intervals of 1-7 days, equal samples of 10-20 larvae of known age and instar were removed simultaneously from each colony without replacement. This produced 1-5 samples per instar. All sample larvae were freeze-killed before they voided their alimentary tracts. Adults were freeze-killed on the day of emergence. All insects were stored in a freezer. Head widths were measured at 10X magnification to the nearest 0.05 mm; all sample larvae were measured except a few tenerals discarded because of distorted head capsules. Larvae and adults were weighed to the nearest 0.05 mg within I min of removal from the oven after drying for 24 h at 70°C. Weighings were of single individuals except for instar I larvae which were weighed in groups of 24because of small mass.

RESULTS AND DISCUSSION

Relative to the two rearing densities, number of molts was equal. head widths diverged little or none, development rate diverged some, and individual dry weight or biomass diverged greatly. We observed four larval molts and five instars before pupation. This confirms the number reported by Hodson (1941). The point when 10-20% of larvae were spinning cocoons was reached in 40 days in the less dense colony and in 45 days in the more dense. Instar head widths differed between colonies only in instar 5 (Table 1); mean head width in the less dense exceeded that in the more dense by 11%. Head widths of successive instars overlapped only once, in the more dense colony; 6% of instar 4 and 5 larvae were involved. As in the less dense colony, Baker (1969) reported no overlap among head widths of successive instars in the Great Basin tent caterpillar. Individual biomass differed between colonies in instars 3-5 (Table 2); mean dry weights in the less dense colony were greater than those in the more dense by factors of 1.3-1.9. Biomass attained near the end of instar 5 averaged 98 and 54 mg per larva, respectively, in the less dense and more dense colonies. Size of the larvae and adults was similar to that of two field samples. Fifty larvae collected

Table 1. Instar head widths in the two colonies.

Growth Head width (mm) Instar day nos. no. sampled . N Min. Max. Mean SD

Both Coloniesa I 0-2 49 0.39 0.49 0.43 0.02 2 7 18 0.63 0.78 0.70 0.06 More Dense Colony 3 11 8 1.22 1.32 1.26 0.04 4 16-2 1 62 1.90 2.49 2.25 0.11 5 27-44 43 2.44 4.05 3.08b 0.47 Less Dense Colony 3 I I 9 1.22 1.32 1.29 0.04 4 16-2 1 6 1 1.95 2.54 2.21 0.11 5 27-44 56 2.88 4.39 3.45b 0.38 aSarnples pooled because no difference expected between colonies in early instars b~eanssignificantly different (P, < 0.005). 1980 THE GREAT LAKES ENTOMOLOGIST 209

Table 2. Individual biomass of instars in the two colonies.

Growth Dry weight (mg) Instar day nos. no. sampled N Min. Max. Mean

Both Coloniesa 20 0. lob 10 0.30 More Dense Colony 8 0.80 36 4.50 30 8.70 Less Dense Colony 8 1.10 36 4.50 35 19.10 aSamples pooled because no difference expected between colonies in early instars. b~omputedfrom group weighings of 24individuals. CMeans significantly different between colonies (Pt < 0.001).

near International Falls, Minnesota, shortly before cocoon spinning had head width statis- tics as follows: range 2.00-3.90 mm; mean 3.35 -c SD 0.55. Except for one 2.00 mm larva which was in the instar 4 range, all these larvae were in the instar 5 range of both colonies (Table I). Seven male moths which emerged from cocoons collected near Cloquet, Minnesota, in an area heavily defoliated for three years, and 25 male moths from the less dense colony averaged, respectively, 18 + SD 7 and 28 + SD 4 mg dry weight per moth. The difference is similar to that between maximum larval weights in the less dense and more dense colonies, and suggests that individual biomass in the more dense colony was near that of undersized adults typical of older infestations. The forest tent caterpillar resembles those Lepidoptera which are fairly stable in number of molts and head widths when weight growth diverges (Fraisse 1953, Gruys 1970). Its growth in individual biomass seems particularly sensitive to rearing density, a trait common among Lepidoptera (review in Gruys 1970).

LITERATURE CITED

Baker, B. H. 1969. Larval instar determination of Malacosoma californicurn fragile on bitterbrush in southern Utah. J. Econ. Entomol. 62:1511. Baker, D. G. and J. H. Strub. 1%5. Climate of Minnesota. Part 111. Temperature and its application. Univ. Minnesota Agric. Expt. Sta. Tech. Bull. 248. Fraisse, R. 1953. La croissance de la t&techez la larve du born by.^ mori L. en fonction du rkgime alimentaire. Compt. Rend. Acad.Sci. (Paris) 236:1613-1614. Gruys, P. 1970. Growth in Bupaluspiniarius (Lepidoptera: Geometridae) in relation to larval population density. I. The influence of some abiotic factors on growth. 11. The effect of larval density. Verh. Rijksinst. Natuurbeh. 1. Hodson, A. C. 1941. An ecological study of the forest tent caterpillar, Malacosoma disstria Hbn., in northern Minnesota. Univ. Minnesota Agric. Expt. Sta. Tech. Bull. 148. Lonmer, N. 1979. Differential hatching times in the forest tent caterpillar (Lepidoptera: Lasiocampidae). Great Lakes Entomol. 12: 199-201. THE GREAT LAKES ENTOMOLOGIST

VALIDATION OF A PETE TIMING MODEL FOR THE ORIENTAL FRUIT MOTH IN MICHIGAN AND CENTRAL CALIFORNIA (LEPIDOPTERA: OLETHREUTIDAE)

B. A. ~roft.~M. F. ~ichels2and R. E. ~ice3

ABSTRACT

PETE (Predictive Extension Timing Estimator) model output for Grapholitha molesta (Busck) was within ca. 30 ? 46 degree-days or 1.4 2 2 days of predicting 50% adult moth activity as measured by pheromone trap catches in the spring and first summer generation in both Michigan apple and California Peach orchards. In subsequent summer generations. model output was less reliable.

Welch et al. (1978) described an extension timing model developed to more precisely schedule critical pest management operations (e.g.. sampling, pesticide treatment) for a multispecies pest complex associated with deciduous tree fruit crops (acronymed PETE, Predictive Extension Zming Estimator). They demonstrated a species model for the Ori- ental fruit moth (OFM), Grapholirhma molesra (Busck), a severe pest of peaches and apple worldwide, using two mathematical approaches including a k-th order distributed delay model developed by Manetsch and Park (1974). Since the initial report, further validation of the k-th order distributed delay model for the OFM has been completed. Validation experiments comparing this model when synchronized initially with first pheromone trap catch, are reported relative to seasonal pheromone and bait trap samples of adult moths taken from Michigan and California deciduous tree fruit orchards. Baker et al. (1980) showed that pheromone trap catches of the OFM corresponded closely to emergence of this species in early season and, therefore, probide reliable indices of pest activity and development.

METHODS

Data in Peterson and Haeussler (1930) and Dustan and Armstrong (1932) were the principal sources used in estimating developmental parameters for the egg, larval, and pupal stages of the OFM. Similar estimates for preoviposition and ovipositing females and adult males were calculated from unpublished data. Development thresholds used for all life stages were estimated to be KI, 7.2"C (45°F) lower and K3, 32.2"C (90°F) upper limit with a horizontal cut off using the methods of Baskerville and Emin (1969). The mean developmental time, parameter I/a (Welch et al. 1978), in F' degree days (dd) for each life stage in the model were eggs 143, larval (combined) 387, pupal 383, preoviposition adult 50, and adult 125. The k parameter for each life stage was 15; Welch et al. (1978) discussed the relationship between this parameter and the population variance. An oviposition function of the form proposed by Welch et al. (1978) was used. No mortality was applied to any life stage except the adults which died after 175 degree days.

l~ublishedas Journal Article 9019 of the Michigan Agric. Exp. Stn., East Lansing. This research was supported in part by a EPA Grant No. CR-806277-92-2 to Texas A&M University and Michigan State University. ZDepartmentof Entomology and Pesticide Research Center, Michigan State University, East Lansing, MI 48824. 3Department of Entomology. University of Califomia, Davis, CA 95616. 2 12 THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 4

For modet validations, OFM populations in Michigan (1971-78, apple) and California (1972-78, peach) were sampled from experimental research orchards which were pruned. fertilized, sprayed with fungicides, and otherwise recieved normal horticultural treatments during the study period. With one exception (see later discussion), no insecticidal sprays were applied in these plots during the study period. Orchard sizes in Michigan ranged from I to 15 ha. Data from California were taken in closely associated blocks of peaches of 1 and 1.5 ha. located at the San Joaquin Valley Research Center, near Parlier. Samples were made with pheromone traps (Michigan: ~ectar11~,4California: Pherocona I-~,5and rubber caps charged with a 1:3 ratio of the mixture, (a-8-dodecenyl acetate (95%), (E)-8-dodecenvl acetate (5%): dodecanol. This mixture is an effective lure for the OFM, although (a-8- dodecen-1-01 is also a pheromone component (Carde et al. 1979) and trap catches can be increased ca. 10-fold by addition of this compound (Baker and Carde 1979). Traps or trap- liners were changed every four weeks; pheromone caps were replaced every six weeks. Pheromone trap densities were from 0.5 to 2.0 per ha. Traps were placed in orchards 1-3 weeks before emergence of adult moths and were checked and moths removed at least weekly and as frequently as three timeslweek throughout the growing season. Weather data were collected via a standard weather station located within orchards or 100 m therefrom. Bait traps used in the California orchards were 946 ml pots containing ca. 473 ml of liquid bait (4660 ml H,O; 454 gm brown sugar; 2.5 ml terpinyl acetate; 0.3 ml emulsifier). Bait traps were changed and checked weekly. Degree day values were calculated using the method of Bakerville and Emin (1969). Running degree day totals were accumulated after 1 January in California and 1 April in Michigan and also after first month catch in pheromone traps (= Biofix) for both areas. Riedl et al. (1976) defined Biofix I for the codling moth, Laspeyresia pornonella (L.) as "the first male moth or moths in a pheromone trap with no significant interruption in catches thereafter." This definition was followed here; since catch data were taken at intervals longer than one day, the midpoint of the interval was used as the Biofix. Model output and validation data were segregated into generational classes. Both average weekly and triweekly counts and cumulative percent values for each area were compared. In the latter case degree day values for the 5, 50 and 95% levels of each generation were estimated by linear interpolation from the surrounding data points. The date of occurrence for each event was determined by the date on which the corresponding degree day value was reached. Errors in degree days and calendar days were estimated independently.

RESULTS AND DISCUSSION

In Figure 1, actual catch values for both pheromone (1972-78) and bait (1972-74) traps are given in calendar days relative to model outputs for the California orchards. Model outputs were standardized to the same area under the curve in each OFM generation as measured by pheromone catch. Model outputs agreed well with observed catches, especially in the initial spring and first summer generations. In fact, there was as much variation between the pheromone and bait trap catch estimates as there was between model outputs and either population sample. The model, which takes into account both male and female adult longevity, was either close to the observed or somewhat late when compared to male pheromone catch in early generations. However, the model almost always anticipated catch in bait traps which monitor both sexes. Bait traps were believed to be less sensitive trapping devices as compared to pheromone traps. Protandry also could be a factor influencing differences between model results and pheromone and bait trap catches. Model output relative to peaks and valleys between generations was reasonably congruent with that observed in the later generations (see exception in 1977, below). One might expect increasing error in predicting later generational events as simulation was extended further away from the synchronization point (Biofix). This could be due to compounding computational errors and population

4~ormerlyproduced by 3M Corporation, St. Paul. MN Sproduct of Zoecon Corporation. Palo Alto, CA. THE GREAT LAKES ENTOMOLOGIST 2 14 'THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

variance differences (owing to both intrinsic population differences and microhabitat environmental variations). Data for 1976 in California (not shown in Figure 1) were excluded from analysis owing to use of multiple chemical sprays which were applied for control of other pests and which caused mortality of OFM and thus confounded results. Also data in 1977 were influenced by a single spray of phosmet 70 WP @ 2.1 Ib A.I. per gal in 30 gal per acre applied on 19 May 1977 (Fig. 1). This treatment killed a large proportion of larvae hatching from eggs laid in late May-early June and influenced survivorship of adults during the first half of the next generation flight period 24 June4 July) and in later generations. One feature which critically influences the fit of the model relative to the catch of all generations is the accuracy of the first pheromone trap catch estimate (Biofix 1). When triweekly or weekly catches were taken in this study, the error in defining this event could have varied by 2 or 3.5 days. respectively. A good example of how this error may have influenced the model fit was in 1978 (Fig. I). It appeared that in this year, first catch probably was 1-3 days later than the value used to initiate the model based on using the midpoint. If daily catches in early season had been taken, then perhaps a better synchronization of the model in all generations would have been obtained. Later in the season, frequent monitoring of trap catch is less essential. Based on the comparisons made with the California data in Figure I and similarly with those from Michigan, the OFM model was initialized 50 degree days earlier for two reasons. One was to obtain a better least squares fit of the data, but also, it was advanced sufficiently so as to normally anticipate events as they occur in the field. This would provide a hedge against excessively late timing during those years which represent the extremes of early pest activity. In Figure 2, the model outputs and the observed data points plotted on a degree day (abcissa) and cumulative percent trap catch scale (ordinate) are given with the 50 dd ad- vancement included. The improved fit using a Biofix as opposed to no Biofix in Michigan is evident in the spring generation; however. there was less improvement in the variability expressed in the first summer generation when using the Biofix as opposed to not using it. It would appear that spring emergence of OFM populations in Calfornia is considerably more variable in relation to heat units than it is in Michigan even when Biofix is used for both areas (Fig. 2). This effect is also generally true for the remainder of the growing season, although the frequency of catch evaluations per week may have contributed partly to these differences (i.e., counts were taken more frequently in Michigan than in California). Comparisons of individual trap catch data points to the model curve in both locations gave similar evidence of the goodness of fit to estimates of moth activity (Table 1). In Michigan, without the Biofix, first emergence of moths at the 50Ch level was only 1.9 dd early. How- ever, fit at the 5 and 95% levels was less accurate (also see Figure 2). When using the Biofix, model error was reduced to 2 dd late (? 6.1) or to 0 ? 0.3 days from the observed. In California, use of a Biofix was essential in synchronizing the model since the number of degree days from 1 January to first moth catch ranged from 227 to 386 (T = 324.7) in this study. In the spring generation, model fits were also good at the 5m point and less accurate at the 5 and 95% levels. Egg hatch in first and second generation of OFM development are critical events for timing management operations in both study areas since upon hatching, larvae bore into either twigs or fruit. Model estimates (with Biofix) of adult activity in the first summer generation at the 50% levels were ca. +0.8 ? 33 dd (+ 0.3 +- 1.5 days) and -28.4 +- 46 dd (- 1.4 ? 1.8 day), respectively for Michigan and California data. Similarly, adult activity at the 5% level in the third adult flight was predicted well, especially in Michigan. Thereafter, at the 50 and 95% points and in subsequent generations in California, the model fit was poorer. For the California data, increasing error was positively correlated with the increasing time from Biofuc I. The California data showed an unexpected tendency to be consist- ently earlier than the model in generation 2, but later in generations 3 and 4 (see data in Table 1 and Figure 2). Several factors may contribute to the observed error in later generations in both areas: First, there usually is an inherent increasing variation associated with later generation owing to the intrinsic population and microhabitat differences mentioned earlier. At present the model does not include a diapause factor which likely has a differential influence on the developmental sequence of later generations (see discussion of this feature for the related THE GREAT LAKES ENTOMOLOGIST

00l45.0.901 AFTER APRIL 1 AFTER BIOFIX - HI

00l45.0.901 RFTER BIOFIX - CA

Figure 2 species L. pornonella in Riedl and Croft [1978]). The upper developmental threshold for G. r~zolestais not well defined and how this effect should be applied to California populations needs more study. Later in the season where generational overlap occurs, it is difficult to assign individuals into their respective generations. Error in defining these proportions undoubtedly influences the computational comparisons between model outputs and the observed data (Table 1). A better evaluation in these generations may be made between the peaks and valleys between generations which are apparent in Figure 1. The OFM model reported herein showed a good fit in early generations under Michigan and California conditions when validated against adult pheromone trap catch data (and bait trap catch in California). We believe it is sufficiently accurate to use it experimentally in the field for timing events in early season in each region. Areas needing further research are a comparison of model outputs with OFM oviposition and egg hatch data taken throughout the growing season and improved model behavior in the later generations (24). Also, there is a need to couple the timing model with a contact and residue spray model to predict moth phenology following pesticide applications. Table 1. Validation of a PETE model of the Oriental fruit moth (Grapholitha moiesta) against pheromone trap catch data taken in Michigan and central California.

Adult Michigan (no ~iofix)b Michigan (Biofix)c California (Biofix)e Pheromone Trap ~odeld ~odel~ Catch a dd Days dd Days dd dd Days dd

5 -48.4 2 16.28 -6.4 2 2.38 -44.5 t 5.2 -4.9 ? 1.0 -9 +4.6 ? 19.2 t0.6 ? 2.3 73 Spring 50 -1.9 ? 17.8 -0.9 ? 1.1 +2.0 ? 6.1 0.0 t 0.3 161 -2.6 ? 34.0 -0.4 ? 1.9 258 Emergence 95 -104 ? 27.0 +5.4 ? 1.3 -100.5 t 20.4 -5.4 t 1.0 369 -58.6 ? 33.3 -4.9 ? 2.5 480

5 +45.8?27.9 +1.8?1.1 +48.2?33.9 +1.9t1.2 873 t3.3 ? 32.9 -0.7 t 1.6 990 Gen 1 50 + 17.5 ? 31.1 +0.6 ? 1.5 +0.8 2 32.8 +0.3 ? 1.5 1165 -28.4 2 45.9 -1.4 ? 1.8 1256 95 -86.3 2 36.9 -3.5 ? 1.6 -79.9 ? 34.5 -3.7 t 1.5 1475 -72.4 2 33.4 -2.9 ? 1.0 1570

5 -38.9 2 21.4 -1.8 1.1 -3.4 ? 15.1 -0.3 ? 0.5 1787 +73.4 2 46.5 +2.4 ? 1.8 1904 Gen 2 50 -203 5 41.0 -8.6 ? 1.6 -229.7 ? 40.0 9.4 t 1.6 2149 t109.6 2 47.6 t3.4 ? 1.5 2252 95 -98.7 ? 59.5 -11.0 -C 3.3 -231.7 2 49.7 -13.3 ? 2.4 2531 +86.4 ? 92.3 +3.0 ? 2.5 2614

5 Gen 3 50 95

5 -148.0 ? 67.3 -4.8 ? 2.2 3872 Gen 4 50 -98.5 ? 80.3 -3.5 5 2.9 4249 95 -149.3 ? 86.2 -6.2 t 3.1 4669 aPercentage development in each generation. b~asedon 13 orchardlyear comparisons. CBased on 9 orchardlyear comparisons. d~asedon 7 orchardlyear comparisons. eFarenheit degree days after Biofix I (1st pheromone trap catch, see definition in text) degree day calculated using thresholds. 72°C (lower) and 32.?"C (upper). + (-) indicates model late (early) relative to observed event. gMean ? SE X. 1980 THE GREAT LAKES ENTOMOLOGIST 217

ACKNOWLEDGMENTS

Appreciation is expressed to A. J. Howitt for providing unpublished data and field plots for these experiments and R. T. Carde, H. W. Riedl and J. F. Brunner for their critical reviews of the manuscript.

LITERATURE CITED

Baker, T. C., and R. T. Carde. 1979. Analysis of pheromone-mediated behavior in male Grapliolitha molesta, the Oriental fruit moth (Lepidoptera: Tortricidae). Environ. Entomol. 8:95'6-968. Baker, T. C., B. A. Croft. and R. T. Carde. 1980. The relationship between pheromone trap capture and emergence of adult Oriental fruit moths, Grapholitlza molesta (Busck). Canadian Entomol. 112: 11-15. Baskerville, G. L., and P. Emin. 1%9. Rapid estimation of heat accumulation from maximum and minimum temperatures. Ecology. 50:5 16517. Carde, A. M., T. C. Baker and R. T. Carde. 1979. Identification of a four component sex pheromone of the female Oriental fruit moth, Grapholitha molesta (Lepidoptera: Tortricidae). J. Chem. Ecol. F423-427. Dustan, G. C., and T. Armstrong. 1932. Observations on the relation of temperature and moisture to the Oriental fruit moth. Entomol. Soc. Ontario Ann. Rep. 63:29-39. Manetsch, T. J.. and G. L. Park. 1974. Systems analysis and simulation with applications to economic and social systems. Part 11. Michigan State Univ. Dept. Elec. Eng. and System Sci. Peterson: A,, and G. J. Haeussler. 1930. Life history of the Oriental peach moth at Riverton N.J. in relation to temperature. USDA Tech. Bull. 183. Riedl, H. W., and B. A. Croft. 1978. The effects of photoperiod and effective temperatures on the seasonal phenology of the codling moth. Canadian Entomol. 110:455-70. Riedl. H. W.. B. A. Croft, and A. F. Howitt. 1976. Forecasting codling moth phenology based on pheromone trap catches and physiological time models. Canadian Entomol. 108:44940. Welch, S. M., B. A. Croft, J. F. Bmnner. and M. F. Michels. 1978. PETE: an extension phenology modeling system for management of a multispecies pest complex. Environ. Entomol. 7:482-94. THE GREAT LAKES ENTOMOLOGIST

THE BIOLOGY OF 'THE WHITE PINE SAWFLY, NEODlPRlON PINETUM (HYMENOPTERA: DIPRIONIDAE) IN WISCONSIN

Aunu Rauf and D. M. ~enjamin2

ABSTRACT

The white pine sawfly, Neodiprion pirletunl. is monophagous on eastern white pine, Pinus strobus, throughout its range in eastern North America. Localized outbreaks occur frequently, and small areas of white pine may be killed. The high population upsurge in Crawford County. Wisconsin, in 1978 was mitigated by the eulophid egg parasite, Closterocerus cinctipennis, when a mean 90.1 t 4.870 parasitization occurred. Larval parasitization in the Grant County infestation averaged 45.05%. and involved the ichneumonid, Olesicampe lophyri, the perilampid Perilamplts l~yalirllcs,and the tachinid Spathimeigenia erecta.

The white pine sawfly, Neodiprion pinetunl (Norton), is a major pest attacking eastern white pine, Pinus strobus L. It seldom causes widespread economic loss, but it can be locally destructive. Its larvae feed on both old and new needles. hence they are capable of completely defoliating and killing trees (Shenefelt and Benjamin 1955). Wisconsin is 4570 forested, and the importance of the white pine sawfly can be judged when it is realized that there are 174,000 acres of white pine type in the state (Stone and Thorne 1961). Local oubreaks of the white pine sawfly occur in Wisconsin almost every year, and it is becoming an increasingly important pest. Unfortunately, most literature describes general distribution and little is reported about its biology. In view of the scarcity of biological and ecological information, the investigation reported here was conducted in Wisconsin to determine its seasonal history and life cycle, its distribution in Wisconsin and in North America, and its enemies.

MATERIALS AND METHODS

The biological and seasonal history studies were conducted in 1978 on a 0.81 ha white pine plantation on the property of John Pelock in Crawford County, Wisconsin (T8N, R5W, Sec. 28). To study its seasonal history a weekly record was made, from May through October, of adults, eggs, and larvae. Needles bearing eggs were collected and reared for parasites. Studies of laval parasitism were conducted on a 0.81 ha white pine plantation in 1978 on the property of Harold Johnson in Grant County, Wisconsin (T3N. RIW, Sec. 5). Fifth and sixth instar larvae were collected and reared for parasites. In both locations, the 5- to 8-year old white pines had been defoliated heavily in 1977, one year before the research was conducted. Rearing of the insects was accomplished by cutting the twigs bearing larvae. These colonies were sustained on freshly cut foliage placed in a pint paper container filled with 7 cm of water; a lantern globe was placed in the inverted container top covering the water reservoir. The frass accumulating on the bottom of the cage served as the cocooning site for the larvae. Each cocoon was placed in a plastic capsule. The newly emerged adults were

l~esearchsupported by the School of hTatural Resources of the College of Agricultural and Life Sciences. University of Wisconsin-Madison, and the Midwest Universities Consortium for International Activities1U.S. Agency for International Activities/Indonesian Higher Agricultural Education Project. 2~epartmen~of Entomology, University of Wisconsin, Madison, WI 53706. 220 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4 transferred into cages provided with freshly cut foliage for mating and oviposition. Rearing \was conducted in a growth chamber at a constant temperature of 23°C and an 18-h photophase. Relative humidity was maintained as high as possible by placing dishes containing water in the growth chamber.

DISTRIBUTION AND HOST

The white pine sawfly occurs approximately throughout the natural range of eastern white pine in eastern North America. It is distributed throughout the eastern United States. westward into Minnesota and Iowa, southward into South Carolina, and northward into Ontario, Quebec, New Brunswick and Nova Scotia, Canada (Fig. I). In Wisconsin, it has been reported in 32 counties throughout the southern part of the state. No collections have been recorded in Vermont, New Hampshire, Massachusetts. and Maine. even though these states lie within the natural range of the eastern white pine. Since the sawfly has been reported from adjacent Canada southward into South Carolina, this omission is probably a question of detection rather than distribution. A similar situation existed in Pennsylvania and lower Michigan where the sawfly was reported in one and two counties respectively. N. pineturn is monophagous on eastern white pine; however, there are published records of its being found on red pine, P. resinosu Ait. (Britton 1926), mugho pine, P. nzrrgo Turra, pitch pine, P. rigida Mill., and short-leaf pine, P. echinata Mill. (Schaffner 1943). Snell (1919) reported N. pineturn on Ribes and suspected it to spread the pathogen causing white pine blister rust. Atwood (1961) criticized these reports of the white pine sawfly's feeding on hosts other than P. strobus, stating that such records probably are based on misidentifica- tions or upon the presence of partly grown larvae that migrated from eastern white pine upon which the eggs were laid.

Fig. I. Natural range of eastern white pine (cross hatching) and distribution of the white pine sawfly (dots) in North America. THE GREAT LAKES ENTOMOLOGIST

SEASONAL HISTORY AND BIOLOGY

The white pine sawfly ove~winteredas a prepupal larva within the cocoon spun by the non-feeding instar in the soil. Pupation occurred in the spring. Adult emergence began during the third week of May 1978, and continued for three weeks (Fig. 2). After mating, the females oviposited in the needles of the previous year's growth. The eggs were laid individually in a row of egg pockets cut in the edge of the needle. The egg pockets were separated by approximately the length of the female. The larval eclosion occurred during the third and fourth weeks of June and feeding continued until the first week of August. The non-feeding larval instars dropped to the ground where they spun cocoons and over wintered. In 1978 there was one generation. However, the abnormally warm summer temperatures in 1975, allowed a second generation to develop in Grant County, Wisconsin (Renlund 1975). A second generation also was reported from Michigan by McDaniel(1939) who studied the white pine sawfly in 1938. She noted that many first generation larvae spun cocoons in the foliage of the host tree. while those of the second generation all dropped to :he ground for spinning. The approximate sex factor of 0.65 was determined by measuring 996 cocoons obtained by rearing field-collected eggs and larvae (prior experience has shown that small cocoons yield males and large cocoons yield females). Adults that emerged from cocoons were transferred into the cages to where they soon mated. The males actively searched for a mate. Upon contact with a female, the male quickly crawled over her and copulated. Mating was completed with the sawflies facing in opposite directions. During mating, the females continued to walk and dragged the males behind them. The duration of copulation was about I2 to 15 minutes. Fecundity was determined by counting the eggs laid by laboratory reared females. Eggs remaining in ovaria after oviposition was completed also were counted; these were not included in fecundity data. The number of eggs produced per female ranged from 62 to 161, the average was 129.3 2 19.7. However, not all eggs were laid. The number of eggs laid per female ranged from 27 to 157, and the average was 116.0 2 29.9. The females oviposited in the needles of the previous year's growth. In selecting a needle for oviposition, the female began at the base of the needle, palpating it with her antennae, as

...-...... Jan Feb Ik Ap Mq h hb Avg Sep Oct Nor Doc

Fig. 2. Seasonal history of the white pine sawtly in Wisconsin-1978 222 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4 she crawled out to the tip. If the needle was unacceptble for oviposition. she turned around and returned to the base of the needle. and then moved out on another needle. This procedure was continued until an acceptable needle was found. Oviposition was accomplished in a manner as described by Benjamin (1955). An opening was made at the edge of the needle. The lancets were moved back and forth alternately as the ovipositor was moved forward until an egg pocket was formed. The ovipositor was returned back through the egg pocket as the egg was deposited. After oviposition of one egg was completed, the female moved forward and began to cut the next egg pocket. Unless disturbed, she would repeat this procedure until her full complement of eggs was laid. An average of 96.3 2 17.9 seconds was required to make a pocket and lay an egg. Based on observation of 24 twigs bearing 3046 eggs collected from the field. the mean number per needle was 2.2 2 0.5 (range 1-5). The average number of eggs per twig was 126.9 2 69.2. Since the mean number of eggs laid per female in the laboratory was 116.0 2 29.9, in the field a single female would probably oviposit her entire complement of eggs on one twig. In the laboratory, females laid eggs one to six days after they emerged from cocoons. and the average was 2.33 f 1.47 days (n=72). The mean longevity of ovipositing females was 3.73 2 1.47 days (n=56). The mean longevity of male adults was 2.80 i 0.19 days (n=35). After oviposition was completed, the female crawled down to the base of the needle, faced the twig, and remained in this position until she died. If mating did not occur after a few days, the female would begin oviposition without mating. All the sawflies that developed from parthenogenetically produced eggs were males. Non-ovipositing females died a little earlier; their mean longevity was 3.13 F .75 days. The first instar eclosion from the eggs and needles was accomplished by wriggling out between the edges of the egg pocket. Larvae crawled to the tip of the needle and began feeding. with their heads toward its tip, and progressed downward to the base. Often as many as 13 larvae were observed on the same needle. However, only about five or six of them fed; the remainder encircled the needle just behind the feeding larvae. The latter group might migrate to another needle or some of them would die from starvation. Since the maximum number of eggs per needle was five, larvae present in excess of five indicated that first instar larvae migrate and may not feed only on the needle in which their eggs were laid. The first instar, as well as the second instar larvae did not consume the entire needle, but fed only on the mesophyll tissue; they left a thin fibrous strand. The third. fourth, and fifth instars consumed the entire needle, and left a stub approximately 2-5 mm in length. They fed on old and new needles. In the laboratory, if needles were not available, they would feed on the bark. All feeding instars were gregarious, but as the larvae increased in size fewer larvae were found per needle (Fig. 3). For example, only three or fewer fifth instar larvae typically were observed feeding simultaneously on the same needle. The larvae exhibited the characteristic aIarm reaction when disturbed. The head and thorax were reared back and the end of the abdomen was raised slightly as they remained attached to the needle by the prolegs. A drop of resinous fluid was exuded form the mouth. Lyons (1962) suggested this behavior is a reaction to prevent attack by parasites and predators.

NATURAL CONTROL

Parasitism of white pine sawfly eggs by Closterocerits cinctipennis Ashmead (Hymenoptera: Eulophidae) was a significant factor in the sawfly population decline in Crawford County. Parasitism of 23 egg batches containing a total of 3052 eggs averaged 90.1 f 4.8% (range 78.4 to 100.0%) (Rauf et al. 1979). Observations conducted on 7 July 1978 revealed one tree with 15 to 30 second and third instar larvae. On several other trees only two to three larvae were present. In the spring of 1979 only a few male sawflies were attracted to traps containing N. pineturn pheromone, this indicates extremely low population levels.3 In Grant County, in late July 1978, about a month after the sawfly eggs hatched. several needles not devoured by larvae had parasite emergence holes. probably C.

3~ers.Comm., Mal-k Kraemer, Department of Entomology, Univei-sity of Wisconsin-Madison THE GREAT LAKES ENTOMOLOGIST 223

Fig. 3. Colony of fifth instar white pine sawfly larvae feeding on eastern white pine

cincfipennis. This indicates that the egg parasite was not as important in controlling sawfly populations in Grant County. Sawfly eggs parasited by C. cincfipennis turned black about the time normal for sawfly larval eclosion. Adult parasites soon emerged through tiny holes in the needle in which sawfly eggs were located. Emergence of these adult parasites was not synchronized with adult N. pinefum presence. This infers that C. cincfipennis must parasitize some other insect in addition to N. pinefum in order to survive. Possible insects among these other species as host are found in the literature. C. cincfipennis also was recorded as a parasite of N. swainei Middleton (Lyons 1962), N. lecontii (Fitch) (Benjamin 1955), N. virginianus Middleton, (Wilkinson et al. 1966), N. praffipratfi(Dyar), (Monis and Schroeder 1%6), and N. excifans Rohwer (Wilkinson and Drooz 1979). Also, in the absence of sawfly eggs, the parasites apparently survive on other hosts. Weis and Nicolay (1918) found C. cinctipennis parasitic on the eggs and larvae of Brachys ovatus (Weber), the eggs of B. aerosa Melsheimer (Coleoptera: Buprestidae), and the larvae of Phyllonorycter hamadryadella Clemens and P. ulmella Chambers (Lepidoptera: Gracillariidae), all of which are leaf miners. Parasitism of 624 fifth and sixth instar larvae collected in Grant County was 45.0%. The major species were Spathimeigenia erecta Aldrich (Diptera: Tachinidae) and Olesicampe lophyri (Riley) (Hymenoptera: Ichneumonidae), which caused 24.0 and 19.9% mortality, respectively. Perilampus hyalinus Say (Hymenoptera: Perilampidae) caused 1.1% mortality. No larval collections were made in Crawford County, because of the low sawfly population. Benjamin (1955) and Tripp (1960) found P. hyalinus as a hyperparasite of N. lecontei. However, Finlayson (1963), who studied parasites of some native diprionid sawflies in Canada, reported that in the case of N. pinetum, P. hyalinus was a primary parasite. Twelve parasites of N. pinetum larvae are recorded from Canada (Finlayson 1963, Raizenne 1957). Hemipteran predators, Podisus placidus Uhler (Pentatomidae) and Sinea diaderma (Fabricius) (Reduviidae) were observed preying on f&h instar larvae. Their influence was probably minimal. THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

LITERATURE CITED

Atwood, C. E. 1961. Present status of the sawfly family Diprionidae (Hymenoptera) in Ontario. Proc. Entomol. Soc. Ontario 91:205-215. Benjamin, D. M. 1955. The biology and ecology of the redheaded pine sawfly. U.S.D.A. For. Serv. Tech. Bull. 11 18. Britton, W. E. 1926. Shade and forest tree insects. Connecticut Agric. Exp. Sta. 25th Rep. Entomol. Bull. 275. Finlayson, T. 1963. Taxonomy of cocoons and puparia, and their contents, of Canadian parasites of some native Diprionidae. Canadian Entomol. 95:475-507. Lyons, L. A. 1962. The effect of aggregation on egg and larval survival in Neodiprion swainei Midd. Canadian Entomol. 94:4%58. McDaniel, E. I. 1939. Suggestions for the control of sawflies on conifers. Q. Bull. Michigan Agric. Exp. Sta. 21: 161-164. Morris, C. L. and W. J. Schroeder. 1966. Occur~enceof the egg parasite Closrerocerus cinc~ip~nnisin Virginia. J. Econ. Entornol. 59: 1533-1534. Raizenne. H. 1957. Forest sawflies of southern Ontario and their parasites. Canada Dept. Agric. Publ. 1009. Rauf, A., D. M. Benjamin, H. C. Coppel, and S. E. Banash. 1979. Parasitization of white pine sawfly and red-headed pine sawfly eggs by the egg parasite, Clos~erocerus cinctipennis Ashmead. Univ. Wisconsin For. Res. Notes. 227. Renlund, D. W. 1975. Forest pest condition in Wisconsin. Ann. Rep. Dept. Nat. Res. Schaffner Jr., J. W. 1943. Sawflies injurious to conifers in north-eastern states. J. For. 41:58&588. Shenefelt, R. D. and D. M. Benjamin. 1955. Insects of Wisconsin forest. Univ. Wisconsin Ext. Serv. Circ. 500. Snell. W. H. 1919. Observations on the relation of insects to the dissemination of Crorzartium ribicola. Phytopathology 9:451464. Stone, R. W. and H. W. Thorne. 1961. Wisconsin's forest resources. Lake States For. Exp. Sta. Pap. 90. Tripp, H. A. 1960. Spathimeigenia spinigera Townsend (Diptera: Tachinidae) a parasite of Neodiprion swainei Middleton (Hymenoptera: Tenthredinidae). Canadian Entomol. 92: 347-359. Weiss, H. B. and A. S. Nicolay. 1918. Notes on Closterocerus cincripennis Ashm. in New Jersey (Hymenoptera). Psyche 25: 128-130. Wilkinson, R. C. and A. T. Drooz. 1979. Oviposition, fecundity, parasites of Neodiprion excitans from Belize, C. A. Environ. Entomol. 8501-505. Wilkinson, R. C., G. C. Becker, and D. M. Benjamin. 1966. The biology of Neodiprion rugifrons (Hymenoptera: Diprionidae), a sawfly infesting jack pine in Wisconsin. Ann. Entomol. Soc. Amer. 59:78&792. THE GREAT LAKES ENTOMOLOGIST

DESCRIPTION AND BIOLOGY OF BUMBLEBEES (HYMENOPTERA: APIDAE) IN MICHIGAN

Robert W. Husband.1 Roland L. ~ischer?and T. Wayne porter3

The distribution of Michigan Bombinae was first studied by Milliron (1939). Records of Michigan bumblebees were included in Cockerell (1916) Franklin (1912). Lutz and Cockerell (1920). and Mitchell (1962). Other major studies in the Great Lakes Region were Chandler (1950) for Indiana, Medler and Carney (1%3) for Wisconsin, Husband (1966) for Michigan, Macfarlane (1974) for Ontario, LaBerge and Webb (1962) for Nebraska, Stevens (1948) for North Dakota, Frison (1926) for Illinois, and Neave (1933) for Manitoba. The purpose of this paper is to add to the information presented by Milliron (1939) and to complement the works by Chandler, Medler and Carney, and Macfarlane. Generic nomenclature follows Hurd in Krombein et al. (1979).

LIFE CYCLE

A generalized life cycle for Michigan bumblebees is summarized in Figure I. Variations in the basic pattern are related to differences in species, differences in latitude, and individual differences among bees. Holm (1960), for example, observed Bombus terrestris queens in Denmark emerging as early as 2 April and as late as 2 June. Under greenhouse conditions. 91% of queen B. terrestris had emerged by 3 May. The generalizations in Figure I are based on the pattern for Bombus vagans at 4?"N latitude (southern Michigan). When the temperature becomes warmer in the spring, queens emerge from hibernation. Hibernation sites are often in soil, 2.5-20.0 cm below the surface. They may also be found under logs or among the multiple, leaf-covered stems of shrubs. Survival rates of overwintering queens vary with environmental factors, physiological factors (fat body storage), and other factors, but may be as high as 84% (Holm 1960). As queens feed on cheny blossoms and other early blooming plants, their ovaries enlarge and rnature.,A period of nest-seeking begins. and queens may be observed flying close to the ground in a seemingly random fashion. A number of queens die without initiating nests. Deaths may be related to a small fat body or disease (Skou et al. 1963). to the fatal combats which queens have in competition for nest sites (Frison 1928, Plath 1934, Hobbs 1%7), or to other factors such as predation or acci- dental death. Porter and Husband found numerous dead queens washed up on the shore of Lake Superior. Nest sites in lower Michigan are frequently former rodent nests. Three such rodent species identified by skeletal remains, grass clippings, food remains, construction materials, or other nest characteristics were: Peromyscus leltcoplts (Rafinesque), Microtus pennsylvanic~ts(Ord) and Spertnophilus tridecemlineatus (Mitchill). Some species of bumblebees construct their own nests in insulation, dry grass. or other materials. After queens enter an abandoned rodent nest or other potential nest site, they rearrange the nest materials to form a brood chamber. Wax, which is exuded between abdominal segments, accumulates on the floor of the small cavity at the center of the nest. This is formed into a wax thimble-sized structure known as a honey pot. Egg cells may be formed before the honey pot. In the subgenus Pyrobombus, typically, two egg cells are formed from pollen, then closed after an egg is laid in each cell (Hobbs 1967). Three more cells are constucted in parallel on each side of the original two cells which are now covered with a

'Biology Department, Adrian College, Adrian, MI 49221. *Department of Entomology, Michigan State University, East Lansing, MI 48224. 3~epartmentof Zoology, Michigan State University, East Lansing, MI 48224. THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 4

Fig. 1. Generalized life cycle of a bumblebee based on Bomhus ruauns in Michigan at 12' N. latitude. wax-pollen canopy. The central cells are less elevated than the two lateral clusters of cells. This arrangement results in an incubation groove which deepens as the queen provides additional pollen to the lateral cells. The queen generates heat from the ventral part of the abdomen and this incubation results in an elevated temperature in the larvae (Heinrich 1977). The queen may open the wax-pollen canopy and regurgitate honey to the larvae. Larvae grow rapidly, spin cocoons, and pupate; workers emerge about three weeks after the egg stage. Cold temperatures result in longer developmental period. Nest temperatures and larval temperatures drop rapidly when the queen leaves the nest to forage. After the first brood emerges and begins to forage, the queen spends more time in the nest. Workers from subsequent broods take over feeding of larvae, incubation, foraging, and defending of the nest. The number of workers increases rapidly for five to six weeks after the first brood emerges. The nest reaches maturity %I2 weeks after the first eggs were laid. Males and new queens copulate outside the hive, remaining attached for hours. The queen may initiate digging of a hibernaculum on the initial day of mating or may mate again. In the case of multiple matings, entrance into hibernation may be delayed for several days after the initial mating. The nest declines as males and most new queens leave the nest and older workers die off. Parasitism by tracheal mites may speed the decline in Botnbus bimaculatr[s (Husband and Sinha 1970). Other factors which cause variations in patterns of nest development are weather factors, particularly long periods of wet weather; predators; nest invaders such as ants; parasites such as conopid and sarcophagid larvae, and eulophid larvae (Husband and Brown 1976); and a variety of diseases (Skou et al. 1%3, Skou 1967). THE GREAT LAKES ENTOMOLOGIST

SOCIAL PARASITISM BY PSITHYRUS

Social parasitism was defined by Wilson (1971) as "The coexistence in the same nest of two species of social insects, one of which is parasitically dependent on the other." Psithyrus bumblebees are social parasites which do not construct nests but displace queens from established nests. Some species of Bombus may do this regularly (Richards 1973) or occasionally (Hobbs 1967). Queen Psithyrus lack the corbiculas or pollen baskets which characterize female bumblebees of the genus Bornbus. Thus. Psithyrus is an obligate social parasite. Psithyrus queens emerge from hibernation later than other species (Table I). Queens enter established nests and, if not repulsed or killed, lay eggs which workers of the invaded colonies rear. If the Bombus queen remains alive in the nest. her eggs and larvae are usually destroyed by the Psithyrlts bumblebee. Psithyrus queens produce only new queens and males. Cumber (1949) recorded 33 ovarioles for one Psithyrus queen, many more than are found in Bombus. Alford (1975) reported a Psithyrus female observed by Sladen to lay 23 eggs in 6 min. Husband and Porter observed that in late May in the Upper Peninsula of Michigan, female Psithyrus were collected by the dozen on one apple tree and not collected at all in other localties. Thus, nests in some localities may have significantly higher concen- trations of social parasitism by Psithyrus than others. This may also explain why species like P. fernaldae are represented by such discontinuous distribution patterns.

DATES OF EARLIEST EMERGENCE OF MICHIGAN BUMBLEBEES

The specific dates for early emergence are given in Table 1. It is likely that some queen bumblebees may cmerge on warm days in March in lower Michigan. The mid-April records for six species are all from two Japanese weeping cherry trees on the south side of a limestone building in Adrian, Michigan. The 4 July record for male B. birnaculatus is from a nest collected in Adrian. New queens were observed in the same nest. In general, the dates in the column entitled males will represent the approximate early dates for new queens and the early date for nest maturity. As indicated earlier, there is a considerable variation in emergence from hibernation of the first and last queen of a given species.

Table I. The dates of the earliest records of appearance of bumblebees in Michigan.

Species Queens Workers Males

Bombus (Botnbias) nevadensis 5 May 30 June 10 Aug. Bombus (Bombus) affinis 14 Apr. 27 June 25 July Bombus (Bombus) terricola 14 Apr. 10 June 6 July Bornbus (Cullurnanobombus) rufocinctus 24 May 3 July 11 Aug. Bombus (Fervidobombus) fert'idus 6 May I June 16 Aug. Bombus (Fervidobombus) pennsylvanicus 30 May 28 June 26 Aug. Bombus (Pyrobombus) bitnaculatus 15 Apr. 16 June 4 July Bombus (Pyrobombus) impatiens 14 Apr. 17 June 5 Aug. Bombus (Pyrobombrts)perplexus 12 Apr. 2 June 5 July Bombits (Pyrobombrts) ternarius 3 May 10 June 22 Aug. Bombus (Pyrobomhus) vagarzs 17 Apr. 19 June 2 July Bombus (Separatobombus) griseocollis 5 May 10 June 10 July Bornbus (Subterrut~eobombus)borealis 13 May 30 June I I July Psithyrus ashtoni 21 May -a 23 July P. citrinus 2 July -a 28 July P. fertzcrldae 29 May -a 21 July P. insularis 3 1 May -a 6 Aug. aNo workers in this species. 228 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

In general, six species of Bombus first emerge in April: aflinis, terricola, bimacularus, imparietzs, perplexus and vagans. Five species emerge in early May: neuadetzsis, fervidus, ternarius, griseocollis and borealis. Bombus rufocinct~rsand pennsyluanicus first emerge in late May. All Psirhyrus first emerge in late May except citrinus in which the first record was 2 July. Medler and Carney (1963) reported early emergence for citrinus in Wisconsin as 6 June.

FLORAL HOSTS FOR MICHIGAN BUMBLEBEES

There are no studies of floral hosts of Michigan bumblebees comparable to those by Fye and Medler (1954) and Macior (1967) in Wisconsin. Two studies of specific plants pollinated by bumblebees in Michigan are those of Door and Martin (1966) for highbush blueberry and Stoutamire (1967) for the stemless lady's-slipper. A number of smaller unpublished studies exist.

IDENTIFICATION OF MICHIGAN BUMBLEBEES

Although bumblebees are sometimes difficult to identify on the basis of color. owing to much variation within species, this remains one of the most commonly used characteristics. Heinrich (1979) has a color plate of North American bumblebees (females only) which should be helpful to beginners. The key which follows utilizes a number of characteristics, including color for convenience. If doubtful specimens are collected, they should be sent to an authority for confirmation. Most characters used in the keys can be observed with a lox pocket lens. Bombus frarernrrs and P. variabilis are included in the keys as they may be found in the southern counties in specific habitats in climatically favorable years.

KEY TO MICHIGAN BUMBLEBEES

I. Seven visible abdominal tergites, abdomen bluntly rounded, 13 segmented antenna (males) ...... 2 1'. Six visible abdominal tergites, abdomen with a pointed cone and sting, I?-segmented antenna (females) ...... 3 2. Uniform distribution of branched hairs on hind tibia, genital capsule weakly sclerotized (Psirhyrus males) ...... 4 2'. Few, unbranched, hairs on central portion of outer surface of hind tibia, genital capsule strongly sclerotized (Bombus males)...... 12 3. Outer surface of hind tibia convex, covered with long hairs, no corbicula, sparse pile on abdomen (Psirhyrus females)...... 8 3'. Outer surface of hind tibia flat or concave, hairless, smooth (corbicula), thick pile on abdomen (Bombus females)...... 25 4. Anal segment black ...... 5 4'. Anal segment with an orange fringe ...... P. fernaldae 5. Fifth antenna1 segment about equal to length of 3rd and 4th combined...... 6 5'. Third and fifth segments of antenna equal ...... P. ashtotli 6. Abdominal tergite 4 with some yellow ...... 7 6'. Abdominal tergite 4 completely black ...... P. cirrinus 7. Mesopleura and most of metapleura usually yellow, thoracic pile shaggy and sparse ...... P. itzsularis 7'. Lower mesopleura, all of metapleura black, thoracic pile short and fine. P. variabilis 8. Occiput mostly yellow ...... 9 8'. Occiput mostly black, tergite 4 pale yellow ...... P. ashroni 9. Pleura yellow ...... 10 9'. Lower half of pleura dark, abdomen black, shiny...... P. variabilis 10. Tergite 4 nearly all black...... 11 THE GREAT LAKES ENTOMOLOGIST 229

OCCIPUT 1

COMPOUND EYE 0 ANTENNA1 SOCKET

CLYPEUS

MALAR SPACE

A MANDIRl E

Fig. 2. Head of a bumblebee (diagrammatic): A, anterior view; B, side view

10'. Tergite 4 yellow...... P. fernaldae I I. Postero-lateral margins of tergites 3 and 4 yellow ...... P. insularis 11'. Tergite 3 yellow or completely black...... P. citrinus 12. Tergite 2 with some yellow...... 13 12'. Tergite 2 entirely black ...... B. impatiens 13. Ocelli well below supraorbital line, eyes large-bulging from sides of head...... 14 13'. Ocelli near supraorbital line, eyes not bulging from sides of head ...... 16 14. Tergite 2 yellow ...... 15 14'. Tergite 2 rusty, posterolateral margins of tergite 2 black...... B. griseocollis 15. Tergites 1-3 yellow, malar space 3 times wider than long ...... B. neondensis 15'. Tergite 3 black, malar space at least 5 times wider than long ...... B. ji-aternus 16. Tergites I4without a pattern of yellow, orange, orange, yellow...... 17 16'. Tergites 14with a pattern of yellow, orange, orange, yellow; remainder of abdomen black ...... 18 17. Malar space as long as wide...... 18 17'. Malar space at least 2 times wider than long ...... B. rufocinctus 18. Tergite 2 completely yellow ...... 19 18'. Tergite 2 yellow in the center, black at the margins ...... B. bimacrrlarus 19. Tergite 1 yellow ...... 20 19'. Tergite 1 black, tergites 2, 3 yellow, tergites 4, 5, 6 black except for a yellow fringe of hairs at the apex of the abdomen ...... B. terricola 20. Occiput black ...... 21 20'. Occiput yellow ...... 22 21. Tergites 1-5 dull yellow, tergites 6-7 usually dull yellow or rust. pleura dark...... B. pennsylvar~icrrs 21'. Tergites 1-5 lemon yellow, tergite 6 (usually) and 7 (always) black, pleura yellow ...... B. fervidrrs 22. Pleura yellow to bases of legs ...... 23 22'. Lower '/2 of pleura dark...... B. borealis 23. Occiput distinctly yellow, face yellow, tergites 2 yellow, tergites 3-6 yellow or black ...... 24 23'. Black pile mixed with yellow on occiput, face black, tergite 2 tawny to rust . B. affinis 24. Tergites 1-3 typically yellow, tergite 6 black...... B. ~erplexus 24'. Tergite 3 typically black, if yellow then tergite 6 yellow...... B. vagans 25. Tergite 2 yellow, rust or orange...... 26 25'. Tergites 2-6 black...... B. impatiens 230 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

26. Ocelli well below supraorbital line ...... 27 26'. Ocelli placed near the supraorbital line ...... 29 27. Tergite 1 yellow, tergite 2 yellow or tawny ...... 28 27'. Tergite 1 black, tergites 2, 3 yellow, tergites 46 black ...... B. rlecadensis 28. Tergites 1, 2 entirely yellow, distinct interalar black band ...... B. fratern~ts 28'. Tergite I yellow, anteromedian part of tergite 2 tawny ...... B. griseocollis 29. Tergite 2 mostly yellow, orange or tawny...... 30 29'. Tergite 2 mostly black, with a anteromedian patch of yellow...... B. bimacul~ztus 30. Interalar band, if present, without V-shaped knotch, without a yellow. orange. orange, yellow pattern on tergites 1-4...... 31 30'. Interalar dark band with a distinct V-shaped posterior knotch. tergites 11with a yellow, orange, orange, yellow pattern, tergites 54black ...... B. ternarius 31. Pleura mostly black, yellow not extending to bases of legs ...... 32 31'. Pleura with yellow extending to bases of legs ...... 35 32. Occiput yellow ...... 33 32'. Occiput black ...... 34 33. Tergites 34black, face black ...... B. perp1e.m~ 33'. Tergites 1-4 yellow or tawny, face yellow ...... B. borealis 34. Apex of abdomen with a fringe of yellow hairs, malar space not longer than broad...... B. terricola 34'. Apex of abdomen black, malar space distinctly longer than broad. B. penns~lranicus 35. Malar space longer than broad ...... 36 35'. Malar space shorterthan broad ...... 37 36. Occiput black, tergites 1-4 yellow, tergites 54black ...... B. fercid~rs 36'. Occiput yellow, tergites 1-2 yellow, tergites 34black ...... B. cagans 37. Tergite 2 yellow or tawny with a distinct postero-median knotch. tergites .b5 black...... B. affinis 37'. Tergite 1 yellow, tergites 24may have varying patterns of yellou. red and black: if tergite 2 is yellow. then with shaggy pile and no distinct postero-median knotch ...... B. rltfocinct~ls

DISTRIBUTION

Of 18 species here reported from Michigan, 16 species have ranges that terminate within or close to the borders of the state. Only B. vagans and P. citrinus have ranges which extend far north and south of Michigan. Some northern species such as Bombus borealis. Bombus rufocinctus, and Psithyrus fernaldae are found south into North Carolina. following the Appalachian Mountains. These species are absent from most of Indiana. Ohio. and central Kentucky. One species. B. affinis, has a unique distribution in that it has both northern and southern limits within Michigan or near the borders. The remaining species may be classified as northern, southern and cosmopolitan species. Northern species are: B. borealis, B. perplexus, B. rufocinctus, B. ternarius, B. terricola, P. ashtoni, P. fernaldae and P. insularis. Southern species are: B. bimaculatus, B.fervidus, B.fraternus, B. griseocollis, B. impatiens, B. nevadensis, B. pennsylvanicus and P. variabilis. Cosmpolitan species which have ranges well north and south of Michigan are B. vagans and Psithyrus citrinrrs. These records are compiled from specimens in the collection of the Museum of Ento- mology, Michigan State University; the Museum of Zoology, University of Michigan: and the senior author. Collecting trips to the Upper Peninsula were made by Husband and Porter. However more than 90% of the bumblebees in the collection listed above are from the Lower Peninsula. Approximately 20,000 bumblebees have been examined. Medler and Carney (1963) described bumblebee distribution in relation to a tension zone for plants in Wisconsin. The relationships between plants and environment in Wisconsin were studied by Curtis and McIntosh (1951) and Curtis (1959). These authors described a tension zone without suggesting a cause-effect relationship. The tension zone in central Wisconsin is correlated with soil conditions, moisture, temperature, and plant and animal 1980 THE GREAT LAKES ENTOMOLOGIST 23 1

WISCONSIN

Fig. 3. The counties of the State of Michigan.

distributions. Medler and Carney (1963) pointed out that the 21°C July isotherm and the tension zone both occur in central Wisconsin. The 21°C isotherm for Michigan is shown in Figure 4. The 8°C annual isotherm is also included in Figure 4 because a shore effect is found which corresponds with bumblebee distribution. If a tension zone can be described in central Michigan in relation to bumblebees, eight species have ranges which terminate near such a zone: B. nevadensis, B. pennsyluanicus, B. rufocinctus, B. ternarius, B. terricola and P. ashroni, P. fernaldae and P. insularis. In the case of the two southern species (B. THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 4

Fig. 4. Average 8°C annual isotherm (line with dashes) and, aver-age 21°C July isotherm tsolid line) (Sornmers 1977). neuadensis and B. pennsylvanicus), there is a strong tendency for the species to be found farther north along the shores of lakes Michigan and Huron. In contrast. the six northern species tend to be found further south inland than along the shore. Distributions of bumblebees were compared to such factors as annual precipitation. last day of killing frosts, forest land, terrain elevation, and limits of rodent distribution. In many cases, a pattern of bumblebee distribution matches an environmental pattern or distribution pattern of a plant or animal. For example, the northern distributions of Perom.vsc~rs leucop~~sand Spermophilus rridecemlineatus closely approximate the northern distribution of Bombus qflinis, B. bimacularus, B. griseocollis and B. impatiens. However. there is no evidence that these four species are restricted to the use of C. tridecemlinearus or P. leucopus nests as nest sites. The rodents and bumblebees mentioned may have similar ranges because they respond to the same environmental factors. Temperatures which freeze a large number of hibernating queen B. bimacularus may also reduce a population of P. leucopus. Prolonged periods of rain in the spring, before food stores are built up, are particularly restricting to bumblebees. Also, rodents and bumblebees may compete for nest sites. Rodent and bumblebee species vary in resistance to a variety of stresses such as lack of food, cold temperatues, parasitism, disease and predation. For example, Plath (1934) described B.fervidus as one of the most vicious stinging bees and Franklin (1912) referred to B. prrplexus as docile. Both observations were based on experience with bees in established nests which were capable of repelling predators such as skunks and humans. The relative abundance of each species in the following list is based upon the number of specimens represented in the collections examined for this study: abundant = >3000, common = 100@3000, uncommon = 1W1000, rare =

Family APIDAE Subfamily BOMBINAE Genus BOMBUS Latrielle Subgenus BOMBIAS Robertson nevadensis Cresson 1874. (Fig. 5). Uncommon. A nest in lower Michigan (LP) in the insulation of a house contained two skeletons of Peromyscus leucopus. The subspecies in Michigan is B. n. auricomus (Robertson) 1903.

Subgenus BOMBUS Latrielle affinis Cresson 1863. (Fig. 6). Uncommon. Of two nests collected in the LP. one was underground and one in a garage. A few males with brown abdominal fourth tergites have been found. terricola Kirby 1837. (Fig. 7). Most common species in the Upper Peninsula (UP). Found commonly throughout the northern half of the Lower Peninsula (NLP), rare in the southern counties (SLP).

Subgenus CULLUMANOBOMBUS Vogt rufocinctus Cresson 1863. (Fig. 8). Uncommon throughout the state, seldom collected in the SLP.

Subgenus FERVIDOBOMBUS Skorikov fervidus (Fabricius) 1798. (Fig. 9). Common throughout Michigan, except seldom collected from the western half of the UP. Franklin (1912) referred to B. fervidus dorsalis as a subspecies. Steven (1948) collected a large nest which contained 12 specimens with the thoracic tergites completely yellow (dorsalis variety), intergrades, and typical B. fervidus with black interalar bands. Husband also collected a nest in Kalamazoo County in which 10% of the bees had yellow thoracic tergites, 10% were intergrades, and 80% were typical B. fervidus. Thus, bees with the yellow thoracic tergites are color varieties not subspecies. pennsylvanicus (DeGeer) 1773. (- B. americanorum Fabricius 1775). (Fig. 10). Abundant in the SLP. rare in the central region of the Lower Peninsula (CLP). 234 THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 4

[Subgenus FRATERNOBOMBUS Skorikov]

[fraternus Smith 1854. Milliron (1939) includedjruternus on authority of Franklin ( 1913) and Frison (pers. comm.) but saw no specimens. Since B. f~aternrcsis recorded from Illinois, and has been collected three times in Indiana, it may be found in the southern counties of Michigan.]

Subgenus PYROBOMBUS Dalla Torre

bimaculatus Cresson 1863. (Fig. 11). Common In the SLP, rare in the NLP. o* one record (St. Ignace) in the UP. It is one of the few bumblebees in Michigan that commonl) has the internal tracheal mite Locustucurus buchneri (Stammer). 1980 THE GREAT LAKES ENTOMOLOGIST 235 impatiens Cresson 1863. (Fig. 12). Common in the SLP, uncommon in the NLP, two records in the southern portion of the UP. One of the largest bumblebee nests collected in North America was a nest of impatiens (Husband 1977). perplexus Cresson 1863. (Fig. 13). Uncommon throughout the state, more often collected inland than along the shore in the SLP. ternarius Say 1837. (Fig. 14). Common in the UP and NLP, rare and more often collected inland in the CLP. No recent record in the SLP. vagans F. Smith 1854. (Fig. 15). Abundant throughout Michigan. The range continues north of Lake Superior and south to the southern United States. 236 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

Subgenus SEPARATOBOMBUS Frison griseocollis (DeGeer) 1773. (= B. separatus Cresson 1863) (Fig. 16). Common in the SLP, uncommon in the NLP. Only three records in the UP, all from the southern shore.

Subgenus SUBTERRANEOBOMBUS Vogt borealis Kirby 1837. (Fig. 17). Uncommon throughout Michigan; rare but widely distributed inland in the SLP.

Genus PSITHYRUS Lepeletier ashtoni (Cresson) 1864. (Fig. 18). Uncommon throughout the UP and NLP. rare in the SLP. The most often collected Psithyrus in the UP. Parasite of B. affinis and B. rerricola. 1980 THE GREAT LAKES ENTOMOLOGIST 237 citrinus (Smith). 1854. (Fig. 19). Uncommon throughout the LP, rare in the UP. The most often collected Psithyrus in the LP. Parasite of B. per~r~s.ylvanicus,B. bimaculatus. B. impatiens and B. vagans. fernaldae Franklin 191 1. (Fig. 20). Rare in the UP; represented by only three records in the NLP. Parasite of B. rufocinctus. insularis (F. Smith) 1861. (Fig. 21). Rare in the UP and NLP. Very rare in the SLP. Intensive collecting in Kalamazoo County yielded two specimens of insularis out of a total collection of about 5000 bumblebees. Parasite of B. rerricola, B. rufocinctus and B. ternarius. [variabilis Cresson 1872. Not reported from Michigan, but found in two counties in northem- most Indiana. Parasite of B. pennsylvanicus.] THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4

SUMMARY

Hundreds of county records have been added to the work on Michigan bumblebees by Milliron ( 1939). However it is not possible to draw conclusions about changes in the range of individual species with confidence. Based on our observations, it appears that the range limits are variable. For example, B. terricola is not expected in the southern tier of counties in Michigan. However, in a year with favorable weather, a tew specimens may move to the region. In 16 years of collecting from the same cheny trees, only in one year was B. terricola collected. One major finding of this study is the tendency for northern species to range in the inland counties and for southern species to range northward along the shores of the Lower Peninsula of Michigan. There is a considerable amount of research to be done if the biology of Michigan bumblebees is to be well understood. Floristic studies of Michigan bumblebees are relatively few. Heinrich (1979), Alford (1975), and Krombein (1979) are recommended as sources for additional literature on bumblebees.

ACKNOWLEDGMENTS

The authors thank Patricia Husband, who was particularly helpful in collecting bumblebees; Ruth Alford, Eastern Michigan University; and Thomas Moore and Gary Breitenbach of the University of Michigan for aid with the collections at their institutions.

LITERATURE CITED

Alford, D. V. 1975. Bumblebees. Davis-Poynter, London. Chandler, L. 1950. The Bombidae of Indiana. Proc. Indiana Acad. Sci. 60:167-177. Cockerell, T. D. A. 1916. Bees from the Northern Peninsula of Michigan. Occas. Papers Mus. Zool., Univ. Michigan 23. Cumber. R. A. 1949. Bumblebee parasites and commensals found within a thirty mile radius of London. Proc. Royal Entomol. Soc. London. Ser. A 24:119-127. Curtis, J. T. 1959. The vegetation of Wisconsin. Univ. Wisconsin Press, Madison. Curtis, J. T. and R. P. McIntosh. 195 1. An upland forest continuum in the prairie-forest border region of Wisconsin. Ecology 3 1:43&455. Door, J. and E. C. Martin. 1966. Pollination studies on the highbush blueberr).. \hccinium corymbosum L. Michigan Agric. Exp. Sta. Quart. Bull. 48:437448. Franklin, H. J. 1912. The Bombidae of the New World. Trans. Amer. Entomol. Soc. 38: 177- 486. Frison, T. H. 1919. Keys to the separation of the Bremidae, or bumblebees of Illinois. and other notes. Trans. Illinois State Acad. Sci. 12: 157-166. . 1926. Contribution to the knowledge of the interrelations of the bumblebees of Illinois and their animate environment. Ann. Entomol. Soc. Amer. 19205-335. . 1928. A contribution to the knowledge of -the life history of Bremus bimaculatus (Cresson). Entomol. Amer. (N.S.) 8:159-223. Fye, R. E. and J. T. Medler. 1954. Spring emergence and floral hosts of Wisconsin bumblebees. Trans. Wisconsin Acad. Sci., Arts and Letters. 43:75-82. Heinrich, B. 1977. The physiology of exercise in the bumblebee. Amer. Sci. 65:45545. . 1979. Bumblebee Economics. Harvard Univ. Press, Cambridge. Hobbs, G. ,A. 1%7. Ecology of species of Bombus (Hymenoptera: Apidae) in southern Alberta. VI. Subgenus Pyrobombus. Canadian Entomol. 99:1271-1292. Holm, S. N. 1960. Experiments on the domestication of bumblebees (Bombus Latr.) in particular B. lapidarius L. and B. terrestris L. Kgl. Vet. Landbohfijsk Arsskr. 1960: 1-19. Husband, R. W. 1966. Acarina associated with Michigan Bombinae. Ph.D. thesis. Michigan State University. . 1977. Observations on colony size in bumblebees (Bombus spp.). Great Lakes Entomol. 10:8345. 1980 THE GREAT LAKES ENTOMOLOGIST 239

Husband, R. W. and T. M. Brown. 1976. Insects associated with Michigan bumblebees (Bombus spp.). Great Lakes Entomol. 95742. Husband, R. W. and R. N. Sinha. 1970. A revision of the genus Lo,custacarus with a key to the genera of the family Podapolipidae (Acarina). Ann. Entomol. Soc. Amer. 64:1152- 1162. Krombein. K. V., P. D. Hurd, Jr., D. R. Smith, and B. D. Burks. 1979. Catalog of Hymenoptera in America north of Mexico. Smithsonian Press, Washington D.C. LaBerge, W. E. and M. C. Webb. 1962. The bumblebees of Nebraska. Nebraska Agric. Expt. Sta. Res. Bull. 205:l-38. Lutz, F. E. and T. D. A. Cockerell. 1920. Notes on the distribution and bibliography of North American bees of the families Apidae, Meliponidae, Bombidae, Euglossidae, and Anthophoridae. Bull. Amer. Mus. Nat. Hist. 42:461491. Macfarlane. R. P. 1974. Ecology of Bombinae (Hymenoptera: Apidae) of southern Ontario, with emphasis on their natural enemies and relationships with flowers. Ph.D. thesis. Univ. Guelph. Macior, L. W. 1967. Pollen-foraging behavior of Bombus in relation to pollination of noto- tribic flowers. Amer. J. Bot. 54:359-364. Medler, J. T. and D. W.'Carney. 1%3. Bumblebees of Wisconsin (Hymenoptera: Apidae). Univ. Wisconsin Res. Bull. 240. Milliron. H. E. 1939. The taxonomy and distribution of Michigan Bombidae, with keys. Papers Michigan Acad. Sci., Arts and Letters 24:167-182. Mitchell, T. B. 1962. Bees of the eastern United States, Vol. 2. North Carolina Agric. Exp. Sta. Tech. Bull. 152. Neave, F. 1933. The Bremidae of Manitoba. Canadian J. Res. 8:62-72. Plath, 0. E. 1934. Bumblebees and their ways. Macmillan: New York. Richards. K. W. 1973. Biology of Bomblrs polaris Curtis and B. hyperboreus Schonherr at Lake Hazen, Northwest Territories (Hymenoptera: Bombini). Quaest. Entomol. 9: 115- 157. Skou, J. P. 1967. Diseases in bumble-bee (Bombus Latr.), the occurrence, description and pathogenecity of five hyphomycetes. Kgl. Vet. of Landbohfijsk. Arsskr. 1967:13&153. Skou, J. P., S. V. Holm and H. Haas. 1963. Prel~minaryinvestigations on diseases in bumble-bees. (Bombus Latr.). Kgl. Vet. og LandbohBjsk. Arsskr. 1963:2741. Sommers, L. M. (ed.) 1977. Atlas of Michigan. Michgian State Univ. Press. Stevens, 0. A. 1948. Native bees. North Dakota Agric. Exp. Sta. Bimonthly Bull. 1151-54. Stoutamire, W. P. 1967. Flower biology of the lady's slippers (Orchidaceae: Cypripedium). Michigan Bot. 6: 159-175. Wilson, E. 0. 197 1. The insect societies. Belknap Press of Harvard Univ. Press: Cambridge, Mass. THE GREAT LAKES ENTOMOLOGIST

A CHECKLIST OF ILLINOIS CENTIPEDES (CHILOPODA)

Gerald ~ummers,lJ. A. ~eatt~,2and Nanette ~a~nuson2

ABSTRACT

Records for 45 centipede species occurring in Illinois are given with a brief discussion of general habitat features and the effects of certain environmental factors on specific distribution patterns. The chilopod fauna of Illinois includes species from southeastern, western, and northern physiographic regions and represents a richer mixture of chilopod species than is found in any adjacent state.

Although centipedes have been noted in the Illinois fauna since the earliest reports on the natural history of the state, published records have generally been based on incidental observations made in connection with studies of other animal groups. Notable exceptions to this pattern of anecdotal study are Wood's (1862) monograph on North American Chilopoda, which included several references to material from southern Illinois and the works of Lucien Underwood (1885. 1887) and C. H. Bollman (summarized in Underwood 1893). Chamberlin's early systematic papers on the Lithobiomorpha frequently cited Illinois records (e.g., 1913. 1922, and 1925a) and there have been some studies of local populations (Ostendorf 1939, Rapp 1946, Auerbach 1951), but no comprehensive reports for the state are available. The variety of natural areas in Illinois as well as its location between two major biomes suggest the potential for a unique assemblage of animal species. The following checklist confirms this diversity for centipedes and provides a more thorough report of centipede distribution than is currently available for any state. In addition to cataloguing the state's centipede fauna, the distributional information in this list gives a general indication of habitat requirements for individual species and provides an opportunity for further analysis of geographic patterns in a group of organisms of reputed low vagility (Crabill 1961). The natural areas of Illinois have been reviewed by Schwegman (1973) and are summarized in Figure 1. The principal habitats in Illinois are components of the Great Plains grassland biome and the eastern deciduous forest biome, but climatic and edaphic factors have contributed to a wide range of forest and prairie sites (Table 1). The presettlement vegetation of the central portion of the state was prairie, but scattered prairie groves and streamside forests provide woodland habitats throughout the grassland region. Hardwood forests include maple-basswood communities in the north and oak-hickory stands in western and central Illinois. Along the southeastern border, beech-maple and tuliptree forests are common and bald cypress swamps occur along the Ohio River. Additional habitats include hill prairies; extensive sand areas along the Mississippi, Illinois, and Kankakee rivers; and bogs. This habitat diversity is reflected in a richer centipede fauna than is known from adjacent states and includes representatives from southeastern, western, and northern faunas. The following checklist is based on collections at the Field Museum of Natural History, the Illinios State Natural History Survey, the Zoology Department of Eastern Illinois Uni- versity, the Department of Zoology at Southern Illinois University, and the personal collection of the senior author. We summarize the distribution of each species by natural divisions and provide a figure with locality records for those species which are not restricted to single

l~ivisionof Biological Sciences, University of Missouri, Columbia, MO 65201. 2~epartmentof Zoology, Southern Illinois University, Carbondale, IL 6 1901. -3 Table 1. Characteristics of the principal natural divisions of Illinois (adapted from Schwegman 1973). m;e Map Index Division Plant communities Topography Soils z0 A Northern Hardwoods Mesic sites: maple-basswood Rolling hills, ra- Loess 5 Dry sites: black oak, white oak vines, and river bluffs j: 7: B Northeastern Morainal Mesic sites: maple-basswood Moraines, dunes Drift, peat, g Dry sites: bur oak, white oak along Lake Michi- and lake bed !2 Prairie: little bluestem, side-oats gramma gan sediments 5 C Grand Prairie Prairie: big bluestem, Indian grass, prairie drop- Level to rolling; Drift, loess, seed, and switch grass moraines and lake bed r Hill prairies: little bluestem, side-oats gramma sediments 0 Bottomlands: oak-hickory with silver maple on floodplains 2 Bottomlands and River Margins Bottomlands: silver maple, ash, pin oak Fltx)dplains, gravel Alluvium and M:u-shes ;mtl hogs tcl-r;~ccs,;lnd river glacial out- hlul'fs wash; loess < along Wabash -Z?. W Sand areas Forests: hlack oak, hlack.iack oak I.cvcI to rolling Sand Z Prairie: little hluestcm, June yrttss, Indian glxss, sand plains; dunes ? and porcupine grass and blowouts P Western Forest Mesic sites: white oak, red oak, and basswood Strongly dissected Loess Dry sites: black oak, white oak, shagbark hickory plain Prairies: big bluestem, Indian grass, prairie drop- seed, and switch grass

Southern Till Plain Forest: black oak, shingle oak, mockernut hick- Hilly in south; flat Loess and till; ory, shagbark hickory in north claypan sub- Prairie: big bluestem, Indian grass, prairie drop- soil seed, and switch grass 4 Eastern Forest Bottomlands: pin oak, sweetgum, hackbeny, and Floodplains, ter- Loess and till silver maple races, and river ?i Uplands: white oak, red oak, and sugar maple bluffs % m Ozark Forest: red oak, sugar maple, basswood, Ohio Maturely-dissected Loess 3 buckeye, beech and tuliptree plateau: bluffs. ra- vines, and stream s' X canyons E Coastal Plain Uplands: black oak, white oak, red oak, tuliptree, Floodplain. river Loess and h shagbark hickory terraces gravel 2 Bottomlands: Shumard oak, swamp white oak, 0z ashes, hickories, beech, and tuliptree 0 Swamps: black cypress, tupelo gum, swamp r cottonwood F1 THE GREAT LAKES ENTOMOLOGIST

Fig. I. County map of Iuinois showing principal natural divisions. (Natural divisions are indexed and described in Table 1 .) (Adapted from Schwegman 1973) 1980 THE GREAT LAKES ENTOMOLOGIST 245 records. Selected literature records are noted where current material is lacking and we provide citations for this material in parentheses. The key to natural division abbreviations is given in Table I.

Order SCOLOPENDROMORPHA Family CRYPTOPIDAE Subfamily CRYPTOPINAE

Cryptops hyalinus Say 1821. (Fig. 2). B (Auerbach 1951).

Subfamly SCOLOPOCRYFTOPINAE

Scolopocryptops nigridius McNeill 1887. (Fig. 3). S. rubiginosus L. Koch 1878. (Fig. 3). S. sexspinosus (Say) 1821. (Fig. 4).

Subfamily THEATOPINAE

Theorops posticus (Say) 1821. (Fig. 5). T. spinicaudus (Wood) 1862. (Fig. 5). B (Auerbach 1951).

Family SCOLOPENDRIDAE

Hemiscolopendra punctiventris (Newport) 1844. (Fig. 6). Scolopendm viridis Say 1821. (Fig. 6). The only record for this species in Illinois is Chamberlin's (1944) report of a single specimen (Alto Pass, Union County, Natural Divi- sion F) in the Field Museum. We have not been able to locate this specimen and contirm the identification. THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 1 THE GREAT LAKES ENTOMOLOGIST

Order GEOPHILOMORPHA Family CHILENOPHILIDAE

Arctogeuphilus umbruficus (McNeill) 1887. (Fig. 7). Pachymerium ferrugineum (C. L. Koch) 1835. (Fig. 7). B (Auerbach 1951) Taiyunu opifu Chamberlin 1912. (Fig. 8).

Family DIGNATHODONTIDAE

Sfrigamiu bidens Wood 1862. (Fig. 8). S. bofhriopu Wood 1862. (Fig. 9). S. brunneri (Bollman) 1888. (Fig. 10). S. chionophilu Wood 1862. This species is easily confused with branneri and many previous reports of chio~~ophilu(e.g., Summers and Uetz 1979) have probably been based upon incorrect identifications. We have seen no specimens of this species from Illinois but we have confirmed the widespread occurrence of branneri and share the view of Crablll that "many distributional records presumably based upon rhionophi/a have really been founded upon specimens of brunneri. . . ." (1952: 112-1 13).

Family GEOPHILIDAE

Arenophilus bipuncticeps (Wood) 1862. (Fig. I I). Brachygeophilus purki Auerbach 1954. (Fig. 12). B. rupesrris Crabill 1949. (Fig. 12). Geophilus ampyx Crabill 1954. (Fig. 13). G. mordux Meinert 1886. (Fig. 13). G. vifaffus(Rafinesque) 1820. (Fig. 14). 248 THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 1 1980 THE GREAT LAKES ENTOMOLOGIST 249

FamiIy SCHENDYLIDAE

Escaryus nzissouriensis Chamberlin 1942. Although we have seen no specimens of this species, Crabill (1961) includes the state of Illinois in its range. The type locality is St. Louis, Missouri. Schendyla nemorensis (C. L. Koch) 1836. (Fig. 15).

Order SCUTIGEROMORPHA Family SCUTIGERIDAE

Scutigera coleoptera (Linnaeus) 1758. We have seen a number of specimens from several natural divisions throughout the state; however, this species (the "house centipede") is undoubtedly found in homes and buildings in every part of Illinois. It is rarely found outdoors in temperate regions, but it may occasionally be found in areas near human habitation (e.g., Lee 1980; Summers, personal observation).

Order LITHOBIOMORPHA Family HENICOPIDAE

Brlethobius huestoni Williams and Hefner 1928. (Fig. 16). Lamyctes firlvicornis Meinert 1868. (Fig. 16). D (Chamberlin 1912).

Family LITHOBIIDAE Subfamily ETHOPOLYINAE

Bothropolys multidentatus (Newport) 1845. (Fig. 17). A (Chamberlin 1925b). Garibius opirolens Chamberlin 1913. (Fig. 18). 250 THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 4 THE GREAT LAKES ENTOMOLOGIST

Subfamily LITHOBIINAE

Lithobius forjicatus (Linnaeus) 1758. (Fig. 19). A (Chamberlin 1925a). Nadabius anleles Chamberlin 1944. (Fig. 20). N. holzingeri (Bollman) 1887. (Fig. 21). N. iowensis (Meinert) 1886. (Fig. 22). A (Chamberlin 1922). N. pullus (Bollman) 1887. (Fig. 23). Nampabius virginiensis Chamberlin 1913. (Fig. 24). Neolithobius mordax (L. Koch) 1862. The Illinois State Natural History Survey has four specimens from Carbondale and Urbana that were identified as mordux by R. V. Chamberlin, but the specimens are missing the last two pairs of legs and we are unable to verify the identifications. This species occurs in a variety of locations in the Coastal Plain and Interior Lowland physiographic provinces (cf. Fennemann 1928) and it is not unlikely that it occurs in Illinois; however, these four specimens are the only records we found for the state. N. tyrannus (Bollman) 1887. (Fig. 25). N. voracior (Chamberlin) 1912. (Fig. 26). Paitobius juventus (Bollman) 1887. (Fig. 27). [Physobius rappi Chamberlin 1945. An examination of the type specimen suggests that this species is based upon an aberrant specimen of a locally-abundant centipede (R. E. Crabill, pers. comm.). Furthermore, extensive collections at the type locality have failed to yield topotypes and we therefore exclude it from our list.] Pokabius bilabiatr~s(Wood) 1867. (Fig. 28). A (Chamberlin 1922). Sigibius urbanus Chamberlin 1944. The only record for this species is the type specimen, a single female from Chicago (Natural Division B). Sonibus bius (Chamberlin) 1911. (Fig. 19). D (Chamberlin 1922). Sozibirrs proridens (Bollman) 1887. (Fig. 30). 252 THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 4 1980 THE GREAT LAKES ENTOMOLOGIST 253 254 THE GREAT LAKES ENTOMOLOGIST Vol. 13. No. 1

Tidabius plesius Chamberlin 1945. The only records for this species are three females collected at Urbana and Mahomet Illinois. Crabill and Lorenzo ( 19571ha~e suggested that critical study of the genus may reduce several specific names to s>non>m>.Although we have collected no specimens referable to this species. we retain the name until further study resolves the problem of specific names in this genus. T. suitus (Chamberlin) 191 I. (Fig. 31). T. tiuilrs (Chamberlin) 1909. (Fig. 31). D (Chamberlin 1913).

DISCUSSION

This checklist provides distributional records for 45 centipede species. including the pre- sumed occurrence of Escaryus missouriensis in the state and the unconfirmed resports of Neolithobius mordax and Scoloprndra heros. With the possible exception of Scurigera coleoptrata. no species occurs in all 10 natural divisions, but ZI species do occur in five or more of the areas noted in Figure I while nine are restricted to a single natural division. Our maps emphasize the relationship between species distribution and environmental factors rather than political subdivision and thus provide some indication of habitat facton which might affect centipede diversity. Although there are limitations in this presentation oa ing to unequal collecting effort over all areas, we believe consideration of species distributions in terms of natural divisions yields patterns suggesting features which can be specifically examined for their effect on centipede distribution. Water is the major environmental factor affecting centipede distribution and abundance. The limited development of a hydrofuge lipid layer covering the cuticle makes centipedes susceptible to both dessication and endosmosis (Blower 1955). Long-term studies have shown that centipede population size in deciduous forests is significantly related to annual variations in precipitation (Kendeigh 1979), and our data reflect this influence, but habitat 1980 THE GREAT LAKES ENTOMOLOGIST 255 stabilty and microclimatic or edaphic factors also appear to be important. The sand areas along the Mississippi, Illinois, and Kankakee rivers (Natural Division E) are surrounded by relatively rich bottomland forests (Division D, with 21 species), yet the sand areas are inhabited by only five centipede species. Soils in sand areas are dry, but they are also subject to a high degree of erosion and this instability may make centipede colonization difficult. The complete absence of geophilomorphs (generally a soil-dwelling group of centipedes), suggests that the instability of suitable habitats may be an important factor in this region. In contrast, soils in the Northern Hardwoods (Division A). Northeastern Morainal (Division B), Grand Prairie (Division C), and Western Forest (Division F) areas are relatively stable and these regions have greater species richness. Edaphic factors in maple- basswood forests throughout northern Illinois probably limit centipede diversity in general, but low scolopendromorph diversity in this region is undoubtedly due to climatic conditions. It is unlikely that additional collecting will add any species other than Scolopoctyptops sexspirlosrrs. which is common in southern Wisconsin (Matthews 1935), because scolo- pendromorphs are typically limited by extreme winter temperatures (Crabill 1952). The rich centipede fauna of the Grand Prairie division reflects extensive riparian woodlands which provide permanent habitat for litter organisms. Streamside forests represent the principal dispersal routes for trees entering the prairie following glaciation (Gleason 1922) and the centipede fauna of Illinois bottomland and streamside forests appears to demonstrate the same phenomenon. Scolopoctyptops rlrbiginosus is generally found in western localities (Crabill 1960) but it occurs in a number of widely separated sites along the Mississippi River and appears to have expanded its distribution along streamside forests (Fig. 3). Further evidence of riparian dispersal includes Pachymerium ferrugineum along the Illinois and Wabash river systems and the northern extensions of Arenophil~rsbiplrt~cticep.c (Fig. I I) and Tidubius tivius (Fig. 3 1). The Western Forest division (F) includes a rich centipede fauna with representative species from several physiographic provinces. The location of this division at the junction of the Central Lowlands. Coastal Plain, and Interior Low Plateau provinces (Fennemann 1928) provides a unique combination of habitats and results in a high diversity (30 species). Scolopendromorph centipedes are generaIly tropical centipedes and onIy 10 species are known from eastern North America; however, eight species occur in the Western Forest Division and this probably reflects the greatest scolopendromorph diversity in eastern North America. Additional species which are generally more abundant in areas outside of Illinois include Arctogeophilus umbruticrrs from the Interior Low Plateau and Appalachian provinces of southeastern United States and several lithobiids from the eastern United States (e.g., Burthobirrs huestoni, Guribius opicolens, and Numpubirrs virginiensis). The remaining four divisions and portions of the Bottomlands (D) and Western Forest (F) regions are located in the southern portion of the state, but each natural division varies in diversity and characteristic fauna. The Southern Till Plain (Division G) is adjacent to both the Western Forest and Grand Prairie divisions and combines elements of each region to result in the greatest diversity in the state (3 1 species). Fauna1 similarity may be compared through a variety of measures, but the most convenient index is the coefficient of community (Goodall 1978: also termed Faunal Similarity, Cf. Huey 1978). Coefficients of community (CC) have intermediate values (0.54.75) for all comparisons between the natural division of southern Illinois except for low similarity between the Eastern Forest (H) and Coastal Plain (J) divisions (CC=0.381) and a high association between Western Forest (F) and Southern Till Plain (G) (CC-0.82). Although Eastern Forest and Western Forest share many species (CC=0.694), a number of common Indiana centipedes were not found in our material. This may be due to inadequate collecting in Eastern Forest sites in the state, but it is also possible species richness is affected by the small area of the division in Illinois. Centipede colonization of appropriate microhabitats could be reduced by isolated sites that are remote from dispersal centers. A similar effect may account for low diversity in the Ozark (I) division. The Ozark division is similar to regions in Arkansas and southern Missouri which served as ecological refuges during glaciation and the area in Illinois is known for a number of unique animal species (Schwegman 1973). It is possible that additional collecting in this area might add several species to our list. The Coastal Plain division is the northernmost extension of the Gulf Coastal Plain physiographic province and has the 256 THE GREAT LAKES ENTOMOLOGIST Vol. 13, No. 4 mildest climate in the state; however, it is also an area of extensive bottomland swamps and bogs and these sites are unlikely to provide suitable habitats for soil or litter centipedes. Upland sites in this division do not include a very large area and are inhabited by the small number of species noted in our list (13 species). The collection records we report for Illinois centipedes indicate broad influences of environmental factors on species richness in distinctive habitats as well as variation in species distributions among these habitats. Such information is useful in the interpretation of broad geographic patterns (the y-diversity of Whittaker 1972) and suggests areas for further study in the community ecology of this important group of forest invertebrates. These records can serve as guidelines for selecting sites for field studies of specific relationships between centipedes and environmental factors. They also serve as patterns which can be analyzed for studies of centipede dispersal and colonization and the limiting factors for local populations. The availability of modem identification keys (e.g., Summers 1979) should facilitate these studies.

ACKNOWLEDGMENTS

We thank E. H. Smith, P. W. Smith, and R. C. Funk for access to material. A. .A. Weaver offered assistance during the early phases of this study and provided preliminar). identifications for many specimens; however, the senior author is responsible for all records noted here. Jeanine Kasprowicz produced the distribution maps and I. D. Unzicker offered useful suggestions during the preparation of the manuscript. B. Badortes offered his customary guidance and counsel.

LITERATURE CITED

Auerbach, S. I. 1951. The centipedes of the Chicago area, with special reference to their ecology. Ecol. Monogr. 21:97-124. Blower, J. G. 1955. Millipedes and centipedes as soil animals. p. 13E-151 in: D. K. M. Kevan (ed.), Soil zoology. Academic Press, New York. Chamberlin. R. V. 1912. The Henicopidae of America, north of Mexico. Bull. Mus. Comp. Zool. Harvard 57: 1-36. . 1913. The lithobiid genera Nampabius, Garibius, Tidabirrs, and Sigibius. Bull. Mus. Comp. Zool. Harvard 57:37-104. . 1922. Further studies on North American Lithobiidae. Bull. Mus. Comp. Zool. Harvard 57:259-382. . 1925a. The genera Lifhobius, Neolithobius. Gonibirrs, and Zitlapul~sin .herica north of Mexico. Bull. Mus. Comp. Zool. Harvard 57:441-504. . 1925b. The Ethopolidae of America north of Mexico. Bull. Mus. Comp. Zool. Harvard 57:385-437. . 1944. Chilopods in the collections of Field Museum of Natural Histon. Field Mus. Nat. Hist. Zool. Ser. 28:175-216. Crabill, R. E. Jr. 1952. The centipedes of northern North America. PhD thesis. Cornell Univ. . 1960. A new American genus of cryptopid centipede, with an annotated key to the Scolopendromorph genera from America north of Mexico. Proc. U.S. Natl. Mus. 111: I- 15. . 1961. A catalogue of the Schendylinae of North America, including Mexico, with a generic key and proposal of a new Simoporus (Chilopoda: Geophilomorpha: Schendylinae). Entomol. News 72:29-80. Crabill, R. E. Jr. and M. A. Lorenzo. 1957. On the identity of the Gunthorp types, 11, and some notes on plectrotaxic criteria (Chilopoda: Lithobiomorpha: Lithobiidae). Canadian Entomol. 89:428432. Fennemann, N. M. 1928. Physiographic divisions of the United States. Ann. Assoc. Amer. Geog. 18:261-353. 1980 THE GREAT LAKES ENTOMOLOGIST 257

Gleason, H. A. 1922. The vegetational history of the middle west. Ann. Assoc. Amer. Geog. 12:39-85. Goodall, D. W. 1978. Sample similarity and species correlation, p. 99-149 in: R. H. Whittaker (ed.), Ordination of plant communities. W. Junk, The Hague. Huey, R. B. 1978. Latitudinal pattern of between-altitude faunal similarity: mountains might be higher in the tropics. Amer. Nat. 112:225-229. Kendeigh, S. C. 1979. Invertebrate populations of the deciduous forest: fluctuations and relations to weather. Illinois Biol. Monogr. 50. Lee, R. E. Jr. 1980. Summer microhabitat distribution of some centipedes in a deciduous and coniferous community of central Ohio (Chilopoda). Entomol. News 91: 14. Matthews. D. C. 1935. The Chilopoda of Wisconsin. PhD thesis, Univ. Wisconsin. Ostendorf. M. L. 1939. Experimental and ecological studies of central Illinois forest millipedes and centipedes. MS thesis, Univ. Illinois. Rapp, J. L. C. 1946. List of myriapods taken in Champaign County, Illinois, during the fall and winter of 1944-1945. Amer. Midl. Nat. 36:66&667. Schwegman, J. 1973. Comprehensive plan for the Illinois Nature Preserves System, Part 2. The natural divisions of Illinois. Illinois Nature Preserves Comm., Rockford. Summers, G. 1979. An illustrated key to the chilopods of the north-central region of the United States. J. Kansas Entomol. Soc. 52:6%700. Summers. G. and G. W. Uetz. 1979. Microhabitats of woodland centipedes in a streamside forest. Amer. Midl. Nat. 102:34&352. Underwood, L. M. 1885. The North American Myriapoda. Entomol. Amer. 1:141-151. . 1887. The Scolopendridae of the United States. Entomol. Amer. 3:6145. . 1893. The Myriapoda of North America, by Charles Harvey Bollman. Bull. U.S. Natl. Mus. 46. Whittaker, R. H. 1972. Evolution and measurement of species diversity. Taxon 21.2 13-25 1. Wood. H. C. 1862. On the Chilopoda of North America, with a catalogue of all the specimens of the collection of the Smithsonian Institution. J. Acad. Nat. Sci. Philadelphia (N.S.) 5:s-52. INFORMATlON FOR AUTHORS

Papers dealing with any aspect of entomology will be considered for publication in 71ir Grrut LuXe.s Entomolo~isr.Appropriate subjects are those of interest to amateur and professional entomologists in the North Central States and Canada. as well as general paper\ and revisions directed to a larger audience while retaining an interest to readers in our geographical area. Manuscripts must be typed, double-\paced, with wide margins on white 8W x I I" or equivalent size paper. and submitted in d~rplic~ertr.Footnotes, legends, and captions for illustrations should be typed on separate sheets of paper. Titles should be conclse, ident~fy- ing the order and family discussed. The author of each species mentioned must be given fully at least once in the text. A common name for each species or group should be given at least once when such a name exists. The format of references should follow that used in recent issues. Photographs should be glossy. Drawings, charts, graphs, and maps must be scaled to permit proper reduction without loss of detail. Contributors should follow the recommenda- tions of the Council eJ' Biology Editors Style, Munuul. 4th ed. Papers published in T1ir Grrtrt LuXrs Enromologi.sr are subject to a page charge of 530.00 per published page. Members of the Soc~ety,who are authors without funds available from grants, institutions or industry and who are unable to pay costs from personal funds, may apply to the Society for financial assistance. Authors will receive page proof, together with an order blank for separates. Short papers on new ecological information, distribution records, and similar topics are welcomed for publication in the "Entomological Notes" section as space permits. Such papers are not subject to a page change. All manuscripts for The Great Lakes Entomologist should be sent to the Editor, David C. L. Gosling. 69063 Wallowa Road. White Pigeon. Michigan 49099. USA. Other correspondence should be directed to the Executive Secretary (see inside front cover).

CBE Style Manual 4th edition

This publication of the Council of B~ologyEditors is a guide for authors, editors and publishers In the b~o log~calsclences It was prepared by a cornmlttee of the council, w~ththe ass~stanceof sc~entlstsIibrari- ans, ed~tors and publishers Substanl~allyrevised th~snew ed~tioncontalns valu- able informatlor- on article planning and preparat~on ed~torialrevlew of the manuscript proofreading and Indexing It del~neatesgeneral style conventions as well as style In special f~elds abbrev~at~onssym- bols and word usage and features an annotated b~bl~ographyand a greatly expanded Index The CBE Style Manual should be read~lyaccess~ble to anyone involved In sc~entif~ced~t~ng wr~tlng and or publ~shing 265 pages (casebound) 512 00 Postage and handling rncluded A 10'0 discount is ava~lablelo. orders of 10 or more copies delivered lo one address Make checks money orders payable lo AlBS All orders must be prepaid. American Institute of Biological Sciences 1401 Wilson Blvd., Arlington, VA 22209