MSU RETURNING MATERIALS: Place in book drop to ”‘M ‘ ‘ ‘Itéa LIBRARIES remove this checkout from .—::—. your record. FINES will be charged if book is returned after the date " stamped be1ow. JUt17‘86 4 afitfi"”1¥flu§ THE INTERACTIONS OF ROOT MAGGOTS AND TWO PARASITOIDS ALEOCHARA BILINEATA (GYLL.) AND APHAERETA PALLIPES (SAY) By Eleanor Groden A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Entomology 1982 ABSTRACT THE INTERACTIONS OF ROOT MAGGOTS AND TWO PARASITOIDS ALEOCHARA BILINEATA (GYLL.) AND APHAERETA PALLIPES (SAY) by Eleanor Groden This thesis examines the temporal and spatial distributions of Aleochara bilineata (Gyll.) and Aphaereta pallipes (Say), two parasitoids Of the onion maggot (OM) Hylemya antiqua (Mg.). The effect Of temperature on A. bilineata develop- ment was quantified. Since the parasite develops slower than its host, a corrector was developed for the bias this imposes on field data. A. bilineata adults emerged in mid-June and were parasitizing larvae and were synchronized with first generation OM pupae. Cabbage maggot, Hylemya brassica (Bouche), and OM pupae provided a continuous supply of hosts for A. bilineata when radishes and onions were grown together. A. pallipes was not present in onion fields during the growing season, but it was Observed on cow dung. A. palligs activity increased dramatically after harvest on maggot-infested culls in the field. Both parasitoids showed increased activity near weedy field borders. DEDICATION To Nate and Brie, in hopes that they may live in a cleaner world. And to my mother, Helen P. Groden, who always presented me with a model of women as strong, intelligent and compassionate people. And to my sisters following in her footsteps. ii ACKNOWLEDGMENTS I would like to thank the entire faculty, staff, and students in the Entomology Department at Michigan State University, particularly Dr. Jim Bath, for creating an atmosphere conducive to professional growth and individual expression. I sincerely doubt that this educational opportunity could be exceeded elsewhere. It has been an honor to be a student of Dr. Dean Haynes. Dean's challenges and inspirations always kept the gears going, allowing one to realize more than I had thought possible. I would like to thank the other members of my guidance committee, Drs. Stuart Cage and Fred Stehr, for their advice and encouragement and for reviewing this manuscript. Also, I appreciate the advice and time spent reviewing this manuscript by Dr. Dick Casagrande. For their humor, encouragement, and discussions, I would like to thank my fellow graduate students: Dr. Mike Mispagel, Mr. Nick Bolgiano, Ms. Julia Pet, Drs. Ray Carruthers and Gary Whitfield. I would also like to thank Mr. Tom Ellis for his frequent assistance on this project, and Dr. Emmett Lampert and Mr. Ken Dimoff for their assistance and advice in the data analysis, and Ken for the assistance with the graphics and the hundreds of little computer-related prob- lems I inevitably encountered. I am indebted to Ms. Susan Battenfield, not only for the assistance in editing and pulling together this manuscript, but for her friendship which made it a more enjoyable process. Also, a considerable amount of work on this project was done by Ms. Robin Rosenbaum, my extra eyes and hands in the field. Robin's iii dedication and friendship proved essential. Most of all, I would like to thank my colleague and dearest friend, Frank Drummond. Often our fellow graduate students become our teachers, and this was frequently the case with Frank. His encouragement, companionship, and love were the true catalyst in the completion of this goal. iv TABLE OF CONTENTS LISTOFTABLES...................viii LIST OF FIGURES . . . . . . . . . . . . . . . . . . x I. INTRODUCTION . . . . . . . . . . . . . . . . . 1 [1. BACKGROUND . . . . . . . . . . . . . . . . . 13 \«A. Onion Maggot . . . . . . . . . . . . . . . . . 13 ’B. Aphaereta palliyes (Say) . . . . . . . . . . . . . 15 JC. Aleochara bilineata (Gyll.) . . . . . . . . . . . . 1 7 D. Alternate Hosts . . . . . . . . . . . . . . . . 20 III. LABORATORY INVESTIGATIONS . . . . . . . 21 A. A. bilineata Egg Development and Mortality . . . . 21 1. Materials and Methods. . . . . . . . . . . . . 21 2. Results and Discussion. . . . . . . . . . . 21 B. Temperature and Parasitoid Age Effects on Successful Parasitism by A. bilineata Larvae . . . . . . . . . 2 3 1. Methods. . . . . . . . . . . . . . . . . . 23 2. Results . . . . . . . . . . . . . . . . . . 27 C. A. bilineata Parasitoid Development and Mortality. . . . 3 1 1. Methods. . . . . . . . . . . . . . . . . . 31 2. Results . . . . . . . . . . . . . . . . . . 33 D. A. bilineata Fecundity and Predation as a Function of Temperature . . . . . . . . 3 5 1. Methods. . . . . . . . . . . . . . . 35 2. Results . . . . . . . . . . . . . . . . 40 E. A. bilineata Oviposition Preferences . . . . 4 5 1. Methods. . . . . . . . . . . . . . . . 45 2. Results . . . . . . . . . . . . . . . . . 46 F. Multiple Parasitism . . . . . . . . . . . . . . . 48 1. Methods. . . . . . . . . . . . . . . . . . 49 2. Results . . . . . . . . . . . . . . . . . . 49 G. Diet Influence on the Longevity of Adult A. pallipes . . . 4 9 1. Methods. . . . . . . . . . . . . . . . . . 50 2. Results . . . . . . . . . . . . . . . . . . 50 IV. FIELD INVESTIGATIONS . . . . . . . 52 I.Methods.......... A.SpringEmergence............... 52 2.Results............. B.SummerDynamics............... 56 TABLE OF CONTENTS (continued) 1. Methods . . . . . . . . . . . 56 a. Organic Farm, Eaton Rapids. . . . . . . . . . 58 1. Description of study area - - . - - - . - . 58 ii. Damage and population sampling . . . 58 b. MSU Muck Farm . . . . . . . . . . . 62 1. Description of study area . - - - - - - 52 ii. Damage and population sampling - - - - - - 65 c.Grant................ . 66 2. Results. . . . . . . . . . . . . . . . . 67 a. Evaluating Parasitism . . . . . . . . . . 67 b. Organic Farm, Eaton Rapids. . . . . . . . . 71 i. Seasonal OM damage . . . . . . . 7 1 ii. Parasitism of OM . . . . . . . . . 7 3 iii. CM damage on radishes . . . . . . . . 74 iv. Parasitism of CM. . . . . . . . . . . . 7 7 c. MSU Muck Farm . . . . . . 77 i. OM damage as affected by planting data and insecticide use. . . . . . . . . . . . 77 ii. Parasitism of OM pupae. . . . . . . 80 iii. CM infestations in radishes . . . . . . . 81 iv. Parasitism of CM. . . . . . . . . . 84 d. Grant. . . . . . . . . . . . . . . 90 i. Parasitism on OM . . . . . . . . 90 ii. CM infestation of radishes and parasitism . . . 92 e. Superparasitism . . . . . . . . . . . . . . 92 C. Fall Dynamics . . . . . . . . . . . . . . . . . 100 l.MethOds................ . 100 2. Results . . . . . . . . . . . . . . . 103 K/a. Predator and Parasitoid Surveys . . . . . . 103 Vb. Parasitism . . . . . . . . . . . . . . . . 107 \jD ..A pallipes Activity In and About Pastures and Onion Fields . 111 1. Methods . . . . . . . . . . . . . . . . . . 112 2. Results . . . . . . . . . . . . . . . . . . 113 V. DISCUSSION . . . . . . . . . . 119 A. The Influence of Planting Date on Root Maggot Infestations. 1 19 B. A. bilineata Synchrony With its Hosts . . . . . . . . . 120 C. Limitations on A. bilineata Parasitism . . . . . . . . 126 D. Fall Dynamics . . . . . . . . . . . . . . . . . 126 E. The Spatial Distribution of A. pallipes After Harvest . . . 128 F. Crop Diversity and Pest Management . . . . . . . . . 129 vi TABLE OF CONTENTS (continued) VI:SUMMARY....................132 Silk/CONCLUSIONS.................. 135 v111. LITERATURECITED................ 137 IX. APPENDICES A. RearingA.bilineata . . . . . . . . . . . . . . . 144 B. RearingA.Eallee . . . . . . . . . . . . . . . . 145 C. Precision Estimates for OM damage sampling at Eaton Rapids 147 D.Coenosiatigrina................. 150 vii LIST OF TABLES Table Page I. The common components in the onion, radish, and pasture (freshdung)habitats.. . . . . . . . . . . . . . . . . 9 2. A. bilineata egg development at constant temperatures. . . . . 22 3. The effect of temperature and parasitoid age on successful parasitismbyA.bilineata. . . . . . . . . . . . . . . . 28 4. The probability of mortality with age for A. bilineata first instarlarvae. 32 5. Development of A. bilineata parasitising CM at constant temperatures.................... 34 6. The influence of temperature on the fecundity Of A. biline- fl. O O O O O O O O O O O O O O O O O O O O O O O 43 7. The influence of temperature on the consumption Of third instar onion maggot larvae by A. bilineata adults. . . . . . . 43 8. , A. bilineata oviposition preference experiments. . . . . . . . 47 9. The influence of diet on the adult longevity of A. pallipe . . . . 51 10. Spring emergence of A. pallipes adults. . . . . . . . . . . 53 11. Degree day accumulation for OM, CM and A. bilineata on planting dates at MSU Muck Farm. . . . . . . . . . . . . 64 12. A. bilineata parasitism of OM at Eaton Rapids research site. . . 75 13. CM infestations in radishes at Eaton Rapids research sites. . . . 76 14. Parasitism of OM pupae, MSU Muck Farm, 1980. . . . . . . . 82 15. A. bilineata parasitism of CM ptpae in radish plots at the MSUMuckFarm,l980.. . . . . . . . . . . . . . . . 87 I6. Parasitism of second generation OM pupae in fields receiv- ing different insecticide treatments. . . . . . . . . . . . 91 17. A. bilineata parasitism Of CM pupae in Grant radishes. . . . . 93 viii LIST OF TABLES (continued) Table 18. I9. 20. 21. 22. 23. Predator densities in post harvest cull plots in Grants. , The relative abundance Of Staphylinidae and A. pallipes adults in post harvest cull plots, 1980. , , A. bilineata parasitism of fall collected OM pupae in 1979. A. pallipes activity in fresh dung in pastures abutting onion fields, I980. . . . . . . Parasitism by A. pallipes on trap crOp planted in different habitat types (post harvest). ix A. pallipes parasitism of OM in post harvest cull plots, I980. Page 104 105 108 110 114 116 LIST OF FIGURES Figure l. 10. ll. 12. Conceptualization of the onion agroecosystem showing levels of interaction within the Object of control and the monitored enVIronmento 0 0 0 I o o o o o o o o o Conceptualization of an onion-radish agroecosystem using Aleochara bilineata asa reference point. - - . . - - - Conceptualization of an onion-pasture agroecosystem using Aphaereta pallipes as a reference point. . . . . . . . A. bilineata egg development as a function of temperature. Standard error method for determination of developmental base temperature for A. bilineata eggs. . . . . . . . . A. bilineata egg mortality at constant temperatures. . . . Survival of first instar A. bilineata larvae under laboratory conditionSO O O O O O O O O O O O O O O O O O O Standard error determination of developmental base tem- perature for A. bilineata parasitizing life stages. . . . . A. bilineata parasitoid development as a function of temper- atureO O O O O O O O O O O O O O O O O O O O O A. bilineata parasitoid mortality at constant temperatures. A. M fecundity as a function of temperature. . . . 0) A. bilineata consumption of third instar OM larvae as function of temperature. - - - - - . . . . . . . . IBa-e. Oviposition preferences of A. bilineata. . . . . . . . 14. 15. l6O Standard error method for determination of developmental base temperature for overwintering A. pallig . . . . . . Location of research sites for onion agroecosystem project. Research fields at organic onion farm in Eaton Rapids, Michigan,1980. ................ Page 24 25 26 30 36 37 38 42 44 47 55 57 59 LIST OF FIGURES (continued) Figure 17. Research plots at the MSU Muck Farm in Laingsburg, MiChigan, 1980. o o o o o o o o o o o o o o o o 18. Correcting for redundancy in sampling due to differences in the rates Of development of A. bilineata and its host, CM. . 19. OM damage in Eaton Rapids research field, 1980. . . . . 20. A. bilineata parasitism of CM pupae in radish plots in Eaton Rapids, 1980. O O O O O O O O O O O O O O O O O 21. The influence of planting date and the use of Dyfonate’ on OM damage at the MSU Muck Farm, 1980. . . . . . . 22. CM damage and infestations in radish plantings at the MSU MUCk Farm, 1980. O O O O O O O O O O O O O O O 23. Estimation of early season CM pupal densities from third in star den SitieSO O O O O O O O O O O O O O O O O 24a-f. A. bilineata parasitism Of CM pupae in radish plantings at 25. 26. 27. 28. 29. 30. 31. 32. th_eMSUMuckFarm, 1980. . . . . . . . . . . . . A. bilineata parasitism at different host densities. . . . . A. bilineata superparasitism of CM pupae. . . . . . . . A. bilineata superparasitism of OM pupae. . . . . . . . The vertical distributions of OM and CM pupae in the soil. . A. pallipes post-harvest activity in relation to the onion field border. O O O O O O O O O O O O O O O O O A. pallipes post-harvest activity in different onion field bordersinGrant,l980... . . . . . . . . . . . . The rate of A. bilineata parasitism of CM and OM in reference to the seasonal abundance of OM pupae. . . . . Seasonal CM egg and larval activity on radishes relative to OM pupation. O O O O O O O O O O O O O O O O O xi Page 63 70 72 78 79 83 85 89 94 96 97 99 117 118 122 123 LIST OF FIGURES (continued) Figure Page 33. Timing of CM pupation relative to OM pupation , , , , , , , . 124 34a-b. Soil temperatures at a depth of 3 cm under the onion, radish,andoatcanopies. , , , , , , , , , , , , , , . , 125 xii INTRODUCTION In the U.S. and many developed countries, the structure of agriculture has changed in the last 50 years. Cheap and abundant fossil fuels (energy) increased chemical fertilizer and pesticide use and promoted farm mechanization (Edens and Haynes 1982). Cheap energy also lengthened food distribution networks, such that food grown in one part of the country could be shipped, cheaply, almost anywhere in the world. More importantly, however, old crop varieties were replaced with less stable varieties that required heavy fertilizer, water, and chemical pesticide use, and the heterogeneous farms were transformed into mono—cropping production systems. These structural changes have profoundly affected farm ecosystems. First, the less diverse habitat concentrates the food resources and attracts greater numbers of herbivorous insects. The habitat's design also makes it less likely that the insects will emigrate out of the system (Risch 1981). Second, pesticides eliminate the natural enemy complex. As a result, severe insect outbreaks occur in mono-cropping systems. Realizing the importance Of natural enemies in controlling pest insects, researchers have been improving application methods (e.g., eliminating some pesticides and using ULV application). However, merely changing or eliminating pesticides will not guarantee biological control, because agricultural systems are not designed to support natural enemies. Adult parasitoids do not solely depend on host distributions. Ecological factors influence adult distribution patterns. Abiotic characteristics of the habitat, particularly relative humidity, affect habitat preferences of some hymenopteran parasitoids (Weseloh 1972, 1975, 1979, Merritt and Anderson 1977, Townes 1957). The abundance and success of many parasitoids depends on the availability of a nectar and/or pollen source for adult parasitoid nutrition (Symes 1975). The effectiveness of some parasitoids in controlling pest populations depends on the proximity of habitats that support alternate hosts (Hsiao and Holdaway 1966, Doutt and Nakata 1973). Many of the essential links between the components of the ecosystem are lost in the simplified large scale monocul- tures. The onion production system typifies this scenario (Haynes et a1. 1980). The existing production system has resulted from decades of heavy reliance on organic insecticides to control the major insect pest, Hllemya antim (Meigen), the onion maggot (OM). Figure 1 (Haynes et a1. 1980) details the onion ecosystem with the onion plant as the reference point. However, to understand the dynamics of the parasitoids of the OM and the potential for parasitoid management, parasitoids must be the reference point (Ravlin 1981). Also, the system boundaries must include the alternate hosts Of the parasitoids and their habitat (Figures 2 and 3). From this perspective a considerable number Of common components emerge between the habitat types (Table 1). In a system including onions, radishes, and beef or dairy production (pasture), the only overlapping species within the "pest" complexes is H. m, the seed corn maggot (SCM). A minor pest in onions and radishes, the SCM is an alternate host for most of the parasitoids, predators, and diseases of the major pests in both crops. It is also a host for the OM and fl. brassicae (Bouche), the cabbage maggot (CM). The SCM may therefore be an important component in parasitoid dynamics. There is, however, considerable overlap between many of Figure l. Conceptualization of the onion agroecosystem showing levels of interaction within the Object of control and the monitored environment (Haynes et a1. 1980). . . . _ ..I ll 1. I 1 a _t o! ,, I. 1 n 1 T l _ — m ..l-F W 1—1l “ET — ‘00... u _ .1538. _ 1......“ . allalJ aflfilfi _afirm. . .31.. .1 n i _ “Ix-yum lllllll I“ .Iwwflrcllfi ..g .1111111 2.28... 8. 11.1 8 my»... - i .8 final _ gm g _ (await . so 1311. . 2...... " .Lofim- #1 It... an .1. 'all lay-1i?! 8:88.319! 1!... _ f _ 889.3 «63.3.1538 .— u do. 8.38! O C. I 1 l 41 I I g 2 €335 Hug Ouaégo Ill 2823.. .83 323 88.8.8. 59.9.8 2.... 829...: 8.8.: 123.3 33.28 :82. 19...... 2 1 115 “.31 'Irlliij . 11H 1- - --- .._-_-_._--.1.._ 8 E l to: .90!- !g 8.0 I Wig. -.zQ—......E.B.. 15¢. $1..me .3328 3.3153 :4 1.5a 8‘ gait-3:72:52: not; «Ev-$.51 8.92:: 93 Saginaw 33.7.8 9.8 8.333 9:85 - Sim coco gag 9:05 93.9mm... gang 9:! Ouagpg :31 7: 1,51] ' D g: 1 11:? __ iii - 7.1 I“. .._III ’35 it: IIIIIIILIIII IlvllPIIIIIIIIrI .I 88 $96 88 ~96 5.8 H a H a 580 H a H a 822qu I... llFIllll Jochzoo no #02300 hszZOKSZw 691.0220: Figure 2. Conceptualization of an onion-radish agroeco- system using Aleochara bilineata as a refer- ence point. (* refers to components common to both Aleochara bilineata and Aphaereta pallipes life systems.) ,oioooomIoIO-zoaoltc SbOOU: 23:2... 2...:- Uh..(l(& 0-2:... mars-03¢ gh-.('OOIUO>3 5.0.3.9.. 2:62.08 1 | l I I l I l i I l | | I | | i | I | l l | I OOh‘OUOO 9M... 2.33.20: 39:30.. ill- I Isa-o. tiwomooselII 8UO¥h¢t coo-.0390 3:20. #303. n (HOOD ————————1_-_p——-——-———_b——1b_‘p———————— I l I I I l '— I l l l 002:... 2.300.»:- 9:1 2...... 2.2.2.0 OOb‘OU: 2232.... 0.20003 lOr¢00fa§OE¢t¢e 0.30.... 0.03.0.3. Berg-at OOgOOO‘ OsO‘O0.00000: flog 90:02.: #8043 80.80 (I 23-3 .23.... r0993. :00 can. all i I I l I I I J _—————————d—_——_-— ————————— 0-2:... c2303: 3...: 2823: I I FOOO gdo 8.3“ .ocOIEIIoqumI-‘corod T msIIEmoIquIooorolomouI.“ POOR IO‘JO 8‘35 :33. newshodgma I .0. IoIsIcIurleoIiflm O‘DO xOhd.’ x. 52.0 , .. ..uoco h§<8 uO‘OO<0 ——_—_--T 3.0.3... 0.030003 .OrstOCOIOOh‘O-Ot >35 Ugguuut b_———————————-———_--———__—_——_——_—— Figure 3. Conceptualization of an onion-pasture agroeco- system using Aphaereta pallipes as a reference point. (* refers to components common to both Aleochara bilineata and Aphaereta pallipes life systems.) 1See Merrit (1974) for further description of interactions in the fresh cow dung habitat. .1 l I | l | | l _{ I I_a I @111. v 00:... 030000.00! I Shdnwt A 00300003000 (20.00000 80000080 0.200020 0.000020 80.00300130000030 I -——————_——‘ fiil 2.2.00.0. 2.0000 act-(C(t g!" 00.8008 002.5 I .000 005:2}; OOrQOUOO _ 03000000 00300200000008 0000.00.00 000., I 000008! _ >00 0000 _ 0.0.0.0003 0.000020 H _— ———-——_—_—————J 90:00:00180000030 00.00 20.00000 80.00300 030:0..00 00:00. UP-OCO‘O 0.3.00.0.0fl0000300 30000030 I D———————_——_——_ Elm-log. 5000.. .200 0000 0.0000023 00.000203 p 0200.000 0.10.2.0! 0.000 000000.203 800400: 8000080 000008! 0000 0.0.00 80.80 I r—__._._____________________________.__ 00;” “3.1.0101. 080.4. 0000 00000! 0.050601030 00400.0 ‘OOOO. éO-UOOQ 8800:(O 0008.3 00.00.08 I >801! 88.0 1__m___ 1. 0.02.3.0 0.000020 gh..¢¢(t n (Oh4000O ..mmameafllzI» g 00.0..» 80.30 000.003.. 2.000 OOh‘OUOO . 2000000000001 03020080000 80°00.»(t Ii 11' g’xhfld 0 00...: 0.000000 80000030 —__—_—p-—d—_ 1! ——-qF——_-——‘b— __—————_—— D 1— I 2.0.00.0: 000000.000 30000030 .0: 30.20.00. OOhCOU: > 800.0 2 8008 I I008 a .0030 . 8008 3 03.00! b—————————— Table l. The common components in the onion, radish and pasture (fresh dung) habitats. Onion Radish Pasture (fresh dung) Major Colonizers Oni_qn maggot, Hylemya antiqua — Seedcorn maggot, H. platura Cabbage maggot, H. brassicae Face fly, Musca autumnalis Ravinia species Orthella caesarion Flea beetle, Phyllotreta crucifera IIII-IIIIIIIIIIIIIIIIIIII IIII-I-IIl-I Onion thrips, Thrips tabaci — III-IIIIIIII Aphid, Brevicoryne brassicae — III-IIIIIIII Nematode, Meloidogyne hapla Natural Enemies Aphaereta pallipes- Al eochara bilinea ta — 11' bipustulata — A . bimaculata _ COGHOS-ia tigrina — Sca tophaga stercoraria — Bembidion species Entomophthora muscae —-------------I Trybliographa rapae 0------— 10 Table 1. (continued) Onion Radish Pasture (fresh dung) I Strongwellsca castrans _ III-IIII-IIII Heterotylenchus sp. Heterotylenchus autumnalis Eucoila impaticas Sphaeridium scarabacoides Philonthus species llllllllllllll Ill-IIIIIIIII III-llllI-I-Il I-III-I-IIIII Macrochelid mites _ Syrphidae spp. — Geocoris bullatus — Diaretiella rapae — Chrysopa sp. _ Entomophthora sperospera Adalia bieuneleda III-IIIIIIIII l-II-I-I-II-I MODOChUS papillatus _ Bacilus penetrans _ III-IIIIIIIIII Arthrobotrgs oliogospora 11 the major natural enemies in the three habitats. Given the complexity of the system, probably more interactions occur than are presently understood. The OM and onion thrips, Thrips tabaci, are the major herbivores in the onion habitat, with nine natural enemies. The radish habitat is slightly more complex with 3 major herbivores and 12 natural enemies. An extremely complex ecosystem lives in fresh dung. An estimated 68 species of dung feeders, 31 species of predators, and 18 species of parasitoids have been reported in fresh cow dung in California (Merritt 1977). Some species are predaceous as larvae yet dung feeders as adults (Sphaeridium spp.), and others are dung feeders as larvae yet predaceous as adults (Scatophaga spp.) The objectives of this study were (1) to investigate the biologies and dynamics of the two major parasitoids of the OM, Aphaereta pallipes (Say), and Aleochara bilineata (Gyll.) and (2) to elucidate ecological factors influencing their distributions within the onion agroecosystem. The dynamics of the parasitoids were examined with the OM and the major alternate hosts of each parasitoid in a multicropping system. The potential of parasitoid management as a major component of an integrated management program was also studied. This thesis represents one sub-project in an interdisciplinary study of the onion agroecosystem initiated at Michigan State University in 1976. A systems approach (Carruthers 1981, Haynes et al. 1980) has been used to examine the economics of onion production as well as the biotic and abiotic components of the onion ecosystem. Research cooperators include economists Mr. Alex Russell, Ms. Sharon Hart, and Dr. Thomas Edens; plant pathologists, Mr. Nick Bolgiano and Dr. Gene Safir; nematologists, Ms. An MacGuidwin and Dr. G. W. Bird; entomologists, Mr. Frank Drummond, Dr. Raymond Carruthers, Dr. Gary H. 12 Whitfield and this author; and system scientists, Dr. Lal Tummala and Ms. Julia Pet. The entire project was managed by Mr. Thomas Ellis and directed and inspired by Dr. Dean L. Haynes. l3 BACKGROUND ONION MAGGOT The OM has been the major insect pest in many onion growing regions in the United States and Canada since it was first introduced to North America from Europe in the mid 19th century (Fitch 1867). Scott (1969) compiled an extensive bibliography of the material published on this pest since 1826. Recent reviews of the literature pertaining to the biology and control of the OM have been done by Whitfield (1981), Carruthers (1981, 1979), Loosjes (1976), Perron (1972), and Ellington (1963). OM attack only the Allium species, of which the bulbing onion, Allium cepa L., is the prefered host (Ellis et al. 1979). Ellis and Eckenrode (1979) and Drummond (1982) suggest that the microbial activity on or near the plant, perhaps governed by the plant age and quality, determines its attractiveness to ovipositing flies. The distribution of OM damage in a uniform onion field is initially random, but due to the migratory behavior of the larvae, damage becomes contagious (Carruthers 1979). However, sprouting volunteer onions in a field act as attraction points for adult OM flies, which then oviposit in adjacent seedling onions (Whitfield 1981). Damage spreads from these focal points, because flies oviposit on previously damaged onions (Workman 1958, Armstrong 1921;, Loosjes 1976, Carruthers 1979). The CM overwinters as diapausing pupae in the soil, and emerges as an adult in mid to late spring. Spring emergence (50%) occurs around 390 degree- days (base temperature #.4°C) (Whitfield 1981, Carruthers 1979, Eckenrode 1975). Adult fly activity within the onion field follows a diurnal pattern, 14 particularly early in the growing season (Carruthers 1981, Whitfield 1981). Fly activity in the fields peaks between 5:30 - 9:30 AM and 5:30 - 8:30 PM; the remainder of the time is spent the cover along the field borders. After a preoviposition period of ca. 10 days (Theunissen 1976), gravid females lay their eggs in the soil around the base of the onion plant or on the leaves of the plant itself. OM eggs require 50 degree days with base tempera- ture 3.88°C (Carruthers 1979) to develop to eclosion. The OM larvae feed directly on the tissue of the onion bulb. In the early season when the food resource per plant is small, larvae migrate along the onion row consuming several seedling onions. A density-dependent migratory behavior is also exhibited later in the season as previously infested plants become loaded with new eggs (Drummond 1982). Carruthers (1979) reports that the three larval instars of OM complete development in 37, 89, and 161 degree days (base 4.#°C). Third instar larvae migrate from the onion plant and pupate in the soil. Pupal depths range from 5 to 15 cm in muck soils in Michigan (Carruthers 1979). Drummond (1982) discovered a shift in this profile with the third generation, such that more individuals pupated in the deeper zones. OM complete one to five generations per year in different regions around the world (Drummond 1982). In Michigan, OM regularly produce three genera- tions per year (Drummond 1982, Whitfield 1981, Carruthers 1981, 1979); how- ever, most investigations have focused on the first and the second generation as they coincide with the onion growing season. The reports on the percentage of diapausing individuals from each OM generation vary (Ellington 1963, Loosjes 1976). Whitfield (1981) reported only 396 of the second generation OM pupae entering diapause in Michigan. Drummond (1982), investigating the post-harvest 15 dynamics of the OM, reported that the larger proportion of the overwintering OM population develops in cull onions left in the field after harvest. Of the natural enemies attacking the OM, the fungal disease Entomoph- thora muscae (Cohn), and the parasitoids Aphaereta pallipes (Say) and Aleochara bilineata (Gyll.) appear to have the most impact on OM populations in Michigan. Several species of Carabidae have also been reported as important predators of OM eggs. Drummond (1982) reviewed the literature concerning carabids associated with OM and CM, and reported on trap catches of carabid species in and about onion field in Michigan. Carruthers (1981), investigating the biology and ecology of _E_. m in the onion agroecosystem in Michigan, reported that the border habitats about onion fields were of primary importance for the interaction between the host and pathogen. He suggests that field infection levels can be enhanced by maximizing the grassy border habitats. APHAERETA PALLIPES (SAY) A. m parasitize OM (Hammond 1924, Sleesman 1931, Workman 1958, Salkeld 1959, Perron 1972, Whitfield 1981). Parasitized pupae have been detected in the overwintering generation of OM in which A. 231113.235 parasitism has ranged from 12 percent (Perron 1972) to 29 percent (Salkeld 1959). Aphaereta pallipes (Say) is a member of the family Braconidae, subfamily Alysiinae. Wharton (1977) recently revised the genus Aphaereta to include ‘16 New World species. McComb (1958) details the species [EM- Shenefelt (1974) gives the most recent list of synonyms and host lists for the species. The more commonly appearing synonyms include _A_. auripes (Provancher), A. muscae l6 (Ashmead), A. californica (Ashmead), and A. sarcophagae (Gahan). Besides the OM, the most commonly reported hosts of A. pallipes include species of muscoid Diptera associated with fresh cow dung, particularly the face fly (Musca autumnalis, DeGeer), Orthellia caesarion (Meigen), several Ravinia species, and Sarcophaga kellyi (Aldr.) (Mohr 19143, Blickle 1961, Benson and Wingo 1963, Houser and Wingo 1967, Turner et al. 1968, Thomas and Wingo 1968, Poorbaugh et al. 1968, Garry and Wingo 1971, Merritt and Anderson 1977). A. pallipes has also been reported as a parasitoid of dipterous larvae on poultry manure (Abeles and Shepard 19M) and the agricultural pest, fl. brassicae, cabbage maggot (Wishart 1957). Although several studies of A. 2111235 biology have investigated its potential for biological control of the face fly, the parasitoid adults cannot emerge from face fly puparia (Blickle 1961, Benson and Wingo 1963). However, A. pilflpgs completing development on preferred hosts (W species and Q. caesarion) still exert considerable mortality on the overwintering population of the face fly Thomas (1967). Thomas (1967) also reported 21196 mortality of M. autumnalis larvae exposed to A. m from adult parasitoid oviposition. A. W is attracted to fresh manure very quickly after deposition, as is its preferred host, Ravinia querula. R. gu_e§u_l_a_ larvae oviposit on the fresh dung. The larvae must remain near the surface of the very moist manure, which exposes them to parasitoid attack (Houser and Wingo 1967). As the dung ages and begins to desiccate, dipterous larvae move deeper in the pat and are less susceptible to attack. Gary and Wingo (1971) reported that the presence of @aeridium scarabaeoides (I...) tunneling in dung increased contact between fly larvae and the parasitoid, resulting in increased parasitism. 17 Salkeld (1959) investigated the life history and behavior of A. m attacking OM in the laboratory. The adult A. m emerge from the fly puparium by chewing a hole in it (Salkeld 1959, Wharton 1977). An average of 10 adults emerge from one pupa, with a sex ratio of 2 males to 5 females (Salkeld 1959). A. m oviposits on all instars of its hosts, but prefers lst and 2nd instars of the face fly (Salkeld 1959, Garry and Wingo 1971). A. algae; may locate its host habitat with olfactory cues. It then detects the specific location of its host tactically as its host moves (Salkeld 1959). Once laid in the host larvae, the parasitized eggs increase in size, but do not hatch until the host pupates. The time required for egg development, therefore, varies with the age of the host. At 23°C, 5 to 8.5 days are required for egg development, 2 to 3 for larvae, and 7 to 9 for pupae in OM (Salkeld 1959). Houser and Wingo (1967) found that at 32°C and 55% RH, adult A. m lived an average of 4 days when supplied with fresh apple slices, fresh cow manure, and distilled water. Individuals supplied with distilled water but no food lived almost a week (Salkeld 1959). ALEOCHARA BILINEATA (GYLL) A. bilineata has been regarded as the most important parasitoid of the OM (Perron 1972). Most investigations on the biology of A. bilineata and its potential for biological control deal with predation and parasitism of the cabbage maggot (CM) (Wadsworth 1915, Colhoun 1953, Wishart et a1. 1956, Read 1962, Coaker 1965, Finlayson et al. 1976, Nair and McEwen 1977, Esbjerg and Bromand 1977, Bromand 1980). However, A. bilineata parasitism of OM has been reported (Hammond 192‘}, Smith 1922, Perron 1972, Loosjes 1976, Beglyarov and Smetnik 18 1977, Bromand 1980, Whitfield 1981). Aleochara bilineata (Gyllenhal) is a member of the family Staphylinidae, subfamily Aleocharinae. Of the 87 species of Aleochara reported in North America, all of the eight for which life histories have been investigated are ectoparasitoids on the pupae of Diptera within the puparium (Moore and Legner 1971). Though this parasitoid has been referred to in the literature as Baryodma ontarionis (Casey) and A. anthomyia (Sprague), it is now called A. bilineata. A. bilineata parasitism has been reported on several Hylemya species, most commonly H. m (OM), fl. brassicae (CM), and _lj. m (SCM). However, SCM is more commonly parasitized by the closely related species A. bipustulata (1..) (Finlayson and Campbell 1976). A. bilineata and A. bipustulata are frequently found in the same habitat, and although A. bipustulata parasitizes CM and OM, it is much less frequent than bilineata (Jorgenson 1979, pers. comm.; Wishart 1957). Adult A. bilineata are thought to be general predators, but are commonly associated with CM, OM, and SCM, the hosts of its parasitic larvae. A. bilineata is one of the most common predators trapped in CM-infested cropping systems (Coaker 1965, Wishart et al. 1956, Finlayson and Campbell 1976). Read (1962) determined that each A. bilineata adult can destroy an average of 23.8 CM eggs or first instar larvae per day, or 2.6 third instars per day. Rear-and-release programs for A. bilineata predation reduced CM populations. Beglyarov and Smetnik (1977) achieved full protection from CM by releasing 20,000 to l40,000 beetles per hectare at 2 to 3 times a season, beginning at egg laying. In addition, 33,000 beetles per hectare reduced OM egg densities by 7696 after three days. Bromand (1980) found that 20,000 beetles per hectare reduced CM damage on 19 cauliflower by 3096. Female A. bilineata lay their eggs in the soil about the base of the host plant (Wadsworth 1915). Each female lays an average of 10 to 15 eggs per day or 500 to 700 eggs over its lifetime (Colhoun 1953, Bromand 1980). Eggs hatch in about five days at 23°C (Colhoun 1953, Bromand 1980). Newly-hatched larvae burrow into the soil immediately, presumably a negative phototrophic response (Wadsworth 1915). On locating a fly puparium, the larva attaches itself to the puparium by the tip of the abdomen and gnaws an entrance hole. This process takes 12 to 36 hours; the average lifespan of the larvae is 8.7 days (Colhoun 1953). The larva punctures the cuticle of its host pupa as it enters the puparium; haemolymph wells out, coagulates, and effectively seals the hole. The larva crawls to the dorsal thoracic region of the host pupa where it feeds. After about eight days at 23°C, it molts into a more parasitic form (Colhoun 1953). A. bilineata overwinters as a first instar larva within the puparium. Second instars are intolerant of cold temperatures (Colhoun 1953). The second and third instars require ca. five and seven days, respectively, to complete development, and the pupa requires ca. 14 days (Colhoun 1953). The adult staphylinid emerges by gnawing an exit hole in the puparium. The hosts of A. bilineata are found in agroecosystems dominated by the heavy use of pesticides (Haynes et al. 1980, Harris et a1. 1967). Kirknel (1978) investigated the impact of several insecticides on CM parasitism by A. bilineata in the laboratory. Each insecticide lost toxicity proportional to the exposure time in the field. Eventually, parasitism in treated soil exceeded that of the control. The time span for this reversal was shortest for chlorfenvinphos, which 20 was also the least toxic to the parasitoids. Finlayson et al. (1980) reported that after seven days, chlorfenvinphos caused no mortality to A. bilineata adults, but the seasonal parasitism of CM on cauliflower was reduced 60% in chlorfenvin- phos-treated field plots. Whitfield (1981) found A. bilineata parasitism of OM to be significantly less in commercial fields receiving between three and five foliar spray applications of parathion or Sevin., than in a research field receiving no foliar insecticide treatments. ALTERNATE HOSTS The CM and SCM emerge in the spring two to three weeks earlier than the OM (Eckenrode 1972, Whitfield 1981). CM emerge around 2‘43 degree days, base temperature 6.1°C (Eckenrode and Chapman 1971). The SCM is a primary pest of seedling beans and corn, and is a secondary pest of onions and cruciferous crops (Miller and McClanahan 1960). The CM is the primary insect pest of most cruciferous crops. Of the root crops, CM prefers for rutabaga, Brassica napobrassica (Mill.), and turnip, B. rapa (I...) (Doane and Chapman 1962). CM is also highly attracted to radish, Rapharus sativus (1...), but the plant's age affects how attractive it is to the CM (Ellis et a1. 1979). 21 LABORATORY INVESTIGATIONS Most of the following laboratory investigations were to quantify relation- ships between temperature and A. bilineata development and survival. As the activity and development of insects is often correlated with temperature, models quantifying these relationships may allow for a more accurate interpretation of the population dynamics observed in the field. The other laboratory experi- ments, oviposition preferences of A. bilineata and the influence of diet on A. pallipes longevity, were designed to gain insight on factors influencing the spatial distribution of the parasitoids in a multicropping system. A. BILINEATA EGG DEVELOPMENT AND MORTALITY Materials and Methods Eggs were collected daily from isolated pairs of mating A. bilineata adults held in controlled temperature chambers at 13°, 17°, 23° and 29°C. Groups of up to 30 eggs per day were then held at the same temperatures and monitored daily to determine the number of days required to complete development and egg mortality at constant temperatures. Standard error determination (Casagrande 1971) was used to estimate the lower developmental threshold. A least squares regression analysis was performed to quantify the relationship between tempera- ture and percent development per day. Results and Discussion Table 2 shows the number of days required to complete A. bilineata egg development. These data agree with data by other investigators. Colhoun (1953) found 7696 of 2,880 eggs held at 23°C hatched on the fifth day, and Bromand 22 Table 2 . A, bilineata egg development at constant temperatures. Temperature (°C) ‘i'Days to Deve10p s n 29 3.25 .543 40 23 4.98 .699 193 17 7.70 .642 92 13 19.48 3.820 25 23 (1980) found that 8096 of 716 eggs held at 22°C also required five days to complete development. Percent development per day can be accurately pre- dicted with temperature over the range of temperatures tested; temperature explains 9996 of the variation in mean development (Figure ‘1). The estimated lower developmental threshold was calculated to be 10°C (Figure 5) with an average of 59.74 1 l4.59 degree days required to complete development. The time required to complete development varied more at the lower temperatures, ranging from 1‘4 to 27 days at 13°C. Egg mortality was lowest at 23°C with a mean of 3296 of the sample population never hatching (Figure 6). Slightly greater mortality was experienced at 17°C and 29°C with 36 and 3796, respectively, not completing development. Accurate data are not available for mortality at 13°C. TEMPERATURE AND PARASITOID AGE EFFECTS ON SUCCESSFUL PARASITISM BY A. BILINEATA LARVAE Materials and Methods Test A: Newly-hatched A. bilineata larvae were collected from eggs developing in controlled temperature chambers at 17° and 23°C. Larvae from each chamber were transferred with a fine point paint brush to petri dishes containing 12 OM pupae covered with moist sand. Eighty-one parasitoid larvae were distributed, 12 per dish (one dish contained only 9) and returned to the respective temperature chamber in which the larvae had originally hatched. Pupae were examined #8 hours later to determine the number of attacks on the pupae, the number of parasitoid larvae successfully entering the pupae, and the incidence of superparasitism. Test B: Newly-hatched A. bilineata larvae were collected from the 24 .wusumuwmamu mo cowuocsw m mm usmfiaoao>ov wwm wumocfiawn .< moo. mmphmmwmzwp .q muowfim mm MN “N “L m0 1m Tm mm. n m e junu xnné + anPl u x av lung AUG 83d lN3Nd013A30 1N33836 25 .mwwo wumosflawp .¢.uom oHSumumaamu mmmfi Houcmanam>ww mo coauNCwauwuwv uouuo wumvSMum .m ousmwm So. Quorwmmzh S S 3 m m b m _ p r a _ a 0 * S I.— x 19 W x 0 mm X 0 19 3 H x mu 0 H X 16 S x. X rm 26 .moHSumumaEmu ucmumcoo um huHHMuuoE www mummcfiawn .< .o musmwm meow mmapcmmm:MH mu mu 2 08 Al I 16.1.8011 lNEOHBd 27 controlled temperature chambers at 23° and 29°C. Larvae from each chamber were distributed 10 per dish (one dish contained only 5 larvae) in petri dishes with a small amount of moist sand. Larvae were kept in their respective controlled temperature chambers. On each of the following four days, ten pupae were added to one dish with parasitoid larvae at each temperature and covered with moist sand. (The experiment was conducted for only four days at 17°C due to a limited number of parasitoid larvae.) Four days following the addition of pupae, pupae were checked for successful attack by the parasitoids. The number of attack holes as well as parasitoid larvae within the fly puparium were seen by submerging puparia in water and viewing them under a dissecting microscope. Results and Discussion The results are presented in Table 3. The percentage of pupae attacked and parasitoids successfully parasitizing OM pupae did not differ significantly at 17° and 23°C. Thirty-three to 7596 of the available pupae were attacked at 23°C, while 2596 to 5896 were attacked at 17°C after #8 hours. Successful parasitism (parasitoids in the fly puparium and feeding) after #8 hours ranged from 8 to 80% at 23°C and 25 to 5096 at 17°C. These values do not include subsequent parasitoid mortality due to superparasitism. One fly puparium was entered by five parasitoid larvae. None of the 50 parasitoid larvae (of varying ages) successfully attacked fly pupae at 29°C. This dramatic decrease in parasitoid performance at the higher temperature suggests a thermal threshold between 230 and 29°C beyond which larval parasitoid activity is terminated. Although this could contribute to the varying impact A. bilineata has on its host populations from one year to the next (Nair and McEwen 1975), the abrupt drop in larval activity at 29°C could be due 28 Table 3. The effect of temperature and parasitoid age on successful parasitism by A. bilineata. A. Number of Number of Z Attached Z Successful Tempgrature Exposed Parasitoid _ Pupae Parasitoid Entries ( C) Pupae Larvae x1 s—- x2 s— x x 17 36 33 .417 .167 .389 .127 23 48 48 .583 .205 .375 .259 1 F ratio = 1.29 with 1 degree of freedom, = .05 2 F ratio = .01 with 1 degree of freedom, = .05 B. Number of Number of Age of Z Successful Tempegature Exposed Parasitoid Parasitoid Z Attached Parasitoid ( C) Pupae Larvae Larvae Pupae Entries (Days) 23 10 10 1 10 10 10 10 2 50 80 10 10 3 3O 30 5 5 4 60 60 29 10 10 l O 0 10 10 2 O 0 10 10 3 0 0 10 10 4 0 0 10 10 5 0 0 29 to a combination of moisture and heat stress. In this experiment all replicates received the same amount of moisture, yet the evaporation rate would have been greater in those at 290 than at the lower temperatures, which could have contributed to the decreased parasitoid activity. Within the time span of the experiment, successful parasitism did not appear to be related to the age of the parasitoid larvae. Sixty percent of the four-day-old larvae successfully parasitized OM pupae, while 3096 of the three- day-old and 50% of the two-day-old larvae were successful. Colhoun (1953) examined the mortality of 300 starved first-instar parasitoid larvae at 23°C with age. After four days, mortality was only 23%. The average life span was estimated at 8.7 days. As other factors besides parasitoid age are most probably acting on parasitism success, an experiment designed to reflect any differences in parasitism with age would have to be carried on over a longer period of time. The results of Colhoun's study are depicted in Figure 7 with percent survival as a function of age. A least squares regression analysis was performed to examine the relationship between age and mortality. The resulting regression equation follows: 2 = .98. Y=99.73 - 5.914x, r However, as the data represent cumulative survival over 20 days, each observa- tion depends on the other. Thus, the computed regression equation may not result in accurate predictions of survival at one point in time, as any error in the estimated survival in the first few days could dramatically shift the resulting curve. Another method of quantifying mortality, which does not require inde- pendent observations, is to assign a probability for mortality at each age. Using 30 .Ammma amonaoo Eoum mumav .wcofiufipcoo muoumuonma nova: wm>hma MummcHHHn .fl woumcfi umpflm mo Hm>H>h=m .n ouswwm mam ON mu Na 0 v o PIIJYIJQ _ p L _ nu 4 4 Q 4 4 a 13.. 4 4 4 S 10 4 4 Q 4 n. 4 1| Q 4 AV 1. 1% WUAIAEUS 1N3383d 31 data extrapolated from Colhoun (1953) (Figure 7), the probability density function for mortality and hazard rate was calculated for each day. The probability density function, f(tmi), is the estimated probability of an individual dying during a particular interval (day) and is computed as follows (Gross and Clark 1975): fitmi) = S(ti)qi/hi where: S(ti) = cumulative probability of surviving, q1 = the proportion dying in a given interval, and h1 = the width of the interval (day). The hazard rate (tmi) or the instantaneous death rate is an estimate of the probability that an individual who has survived to the beginning of a given interval (day) will die within that interval (day). The hazard rate is computed as follows: g(tmi) = 2 qi/hi (l + pi) where: pi = l - qi The results are presented in Table 4. The hazard rate for first instar A. bilineata larvae generally increases slightly with each day of age for the first seven days. There is then a marked increase in the hazard rate on the eighth and subsequent days. This corresponds with the estimated mean (8.7 days) and median (8.6 days) longevity for this life stage. The population of individuals surviving beyond the average experience a greater instantaneous death rate. A. BILINEATA PARASITOID DEVELOPMENT AND MORTALITY Materials and Methods The rate of development for _A_. bilineata life stages parasitizing OM was 32 Table 4. Probability of mortality with age for A. bilineata first instar larvae. Number alive Number dying Probability Interval at beginning during Hazard SE2 density SE2 (Day) of interval interval Rate function 1 306 12 .040 .012 .039 .0001 2 294 19 .067 .015 .062 .0002 3 275 17 .064 .015 .056 .0002 4 258 22 .089 .019 .072 .0002 5 236 16 .070 .017 .052 .0002 6 220 18 .085 .020 .059 .0002 7 202 18 .093 .022 .059 .0002 8 184 16 .091 .023 .052 .0002 9 168 27 .175 .033 .088 .0003 10 141 25 .195 .039 .082 .0002 11 116 30 .297 .054 .098 .0003 12 86 17 .219 .053 .056 .0002 13 69 26 .464 .089 .085 .0003 14 43 12 .324 .092 .039 .0001 15 31 16 .696 .163 .052 .0002 16 15 3 .222 .127 .010 .0000 17 12 6 .667 .257 .020 .0001 18 6 3 .667 .363 .010 .0000 19 3 3 0.000 0.000 0.000 0.0000 1 Data extrapolated from Colhoun (1953). 2 Standard Error (SE) computed as to formular in Cross and Clark (1975). 33 estimated from temperature dependent developmental data for A. bilineata parasitizing CM (Read 1962). It was assumed that as _A. bilineata lives as an external parasitoid within the puparium of several Hylemya species, its develop- mental zero and rate of development are not necessarily synchronized with its hosts. Read (1962) collected overwintering parasitized CM pupae from the field in early November. A. bilineata overwinters as a first instar larva within the puparium of its host. The pupae were stored at 2.2°C (36°F) for ca. two months before being transferred to soil cylinders in Wisconsin tanks where they were reared at constant temperatures ranging from 10° to 233°C (50° to 74°F). The pupae were monitored every 3 to 5 days for 90 days for emergence of adult 5. bilineata and cabbage maggot flies. These data were summarized as to the mean number of days required to complete development at each temperature. The standard error determination (Casagrande 1971) was used to estimate a lower developmental threshold and the mean degree day accumulation required to complete development of the parasitizing life stages. The percent development per day was regressed with temperature to quantify this relationship. Percent mortality as a function of temperature was extrapolated from Read's emergence data, given that he began with 160 parasitized pupae at each temperature. This relationship was ex- amined. Results and Discussion The mean days required to complete development at each temperature are given in Table 5. The lower developmental threshold was calculated to be 34 Table 5. Deve10pment of A. bilineata parasitizing CM at constant temperatures. eeeeeee Coiiiztiesgziiggzm (°F) (0c) :2 s 50 10 0.00 0.00 54 12.22 85.77 13.12 56 13.33 82.05 22.07 60 15.55 67.73 11.87 62 16.67 48.17 19.75 64 17.78 43.48 24.17 66 18.89 38.09 19.50 68 20.00 34.29 8.01 70 21.11 27.77 8.93 72 22.22 26.19 5.37 74 23.33 24.76 8.07 1 Data summarized from Read, 1962. 35 8.6°C (Figure 8) with a mean of 384 degree days required to complete development. This threshold and required degree day accumulation differs from that of two of its hosts, the OM and the CM. The OM requires an average of 306 degree days (base 4.4°C) (Carruthers 1979) and the CM requires 243 degree days (base 6.l°C) (Eckenrode 1972). As this parasitoid attacks the pupal stage of its host, it is not suprising that it would develop slower and hence emerge later in the spring than its host which must go through an adult, egg, and larval stage before it reaches a stage suitable for parasitism. The percent development per day for the parasitizing life stages of A. bilineata can be predicted with temperature (r2 = .97, see Figure 9). Over the range of 13° to 21°C there appears to be no appreciable difference in the _ mortality of A. bilineata parasitizing CM (Figure 10). The average mortality experienced over this range of temperatures was 38%. Mortality increased sharply to 100% at 10°C and decreased with higher temperatures of 22°C and 23°C. It is likely that at some higher temperature mortality increases again. Bromand (1980) reported 43 to 52% mortality to parasitoids in singly- attacked field-collected CM pupae. He noted that superparasitism was respon- sible for some of the mortality in the laboratory culture where parasitoid mortality in CM pupae was 64% and 52% in OM. A. BILINEATA FECUNDITY AND PREDATION AS A FUNCTION OF TEMPERATURE Materials and Methods Mating pairs of A. bilineata adults of various ages were collected from the laboratory colony and isolated by pairs into petri dishes lined with a three-layer disc of moist newspaper. The moist newspaper was thought to provide an easily 36 .mowmum omwa wcflufiuwmmumo mummcwawn .«.uom unnumuoaaou 1 ZS ommn HmquEooam>mc mo CowumGMEuouov Mom vosuofi uouuo vumvcmum .w opswfim so. wmahmmmmzwp «a 3 a o v _ . . L R X X X X X S I. * T“ “U x m X U .8 X 0 X 14. 3 w X 0 M N. S vmt 37 .ousumuma Iaou Mo cowuoanm m on ucoaaoao>ov vwoufimmuma mumoCHHHn .< .m ouswfim n Hoog mmzhmmumzm» vN NN ow g 3 3 Ma r L _ . p L O II km. "4. xmu. 4 and: u 15 AUG 83d lN3Hd013A30 1N3383d 38 .mousumuanou ucmumcoo um xuwamuuoe wfi0ufimmuma mumNCwHHn Agog mmahmmmmzmh .«..oa mnsmam mu «N ha mu m _ _ . b no XX 7.. filo: xxx um wx xm WX Q. 1.0 L 1c. AlITHlHON 1N3383d 39 examinable substrate for egg laying, as well as a moist environment where the food supply, OM larvae, could survive. Each pair of adults was supplied with three third instar OM larvae. The dishes were distributed in four controlled environmental chambers set at constant temperatures of 13°, 17°, 23°, and 29°C. Every 24 hours the dishes were removed, and the staphylinid pair and its food supply were transferred to another dish. The maggot consumption per pair was recorded and a maggot was replaced for every maggot consumed, such that at the beginning of each 24 hour period each A. bilineata pair always had three OM larvae for a food supply. The new dish was then returned to the respective temperature chamber. The dish from which the pair had come and both sides of each newspaper disc were examined under magnification for eggs. Egg laying activity and maggot consumption were monitored for five to ten days. For the data summary, the egg production per female per day was pooled over all pairs at each temperature. Egg production was evaluated from the time an individual began laying at least one egg. Therefore, egg production of zero for the first day for an individual was not included in the data set. Once an individual initiated egg laying, any subsequent days for which egg production equalled zero were included. A least squares regression analysis was performed on the mean number of eggs per female per day and temperature. The regression line was extended to its intersection with the x-axis to predict a lower thermal threshold for fecundity. An analysis of variance for repeated measures (Winer 1971) was performed on the egg production per female on each of the five to ten days. The intent of this procedure was to (1) evaluate the variability between individuals and within an individual through time, and (2) to determine if the data reflect an interaction between temperature and time. 40 The consumption of larvae per pair per day was also pooled over all pairs at each temperature for the data analysis. A least squares regression analysis was performed on the mean number of larvae consumed per pair per day and temperature. The regression line was extended to its intersection with the x- axis to determine a predictive lower thermal threshold for predation. Results and Discussion The mean number of eggs laid per female per 24 hours for each tempera- ture is listed in Table 6. The greatest number of eggs laid by a single female in one day was 46 at 29°C. At least one female laid no eggs on at least one day at every temperature except 29°C. At 29°C all females laid at least one egg every day. Bromand (1980) found an average of 9-10 eggs per female per day at 22°C and 500-600 eggs per female over a lifetime (ca. 55 days). Both Colhoun (1953) and Read (1962) found that females produced an average of 15 eggs per day and an average of more than 700 eggs during their lifetime. This agrees with the mean of 12.03 eggs per female per day at 23°C found by the author. Egg production per female per day varied at all temperatures tested. The variability within a single individual through time (mean square estimated, 52 = 358) is greater than that between individuals (mean square = 153). The analysis of variance for repeated measures indicates a significant interaction between temperature and time; however, at all temperatures except 13°C, egg production did not necessarily increase or decrease with time (age). Perhaps individual variability is more a function of food consumption or some unmeasured factor unknown to this author. Egg production per female per day did decrease with time at 13°C. This is thought to be due to the prolonged exposure to cold temperatures rather than age. 41 The mean egg production per female per day shows a positive trend with increasing temperature (Figure 11) and can be predicted (r2 = .98). Assuming a linear model at the lower temperatures, the extension of the regression line estimates a lower thermal threshold for egg production of 10°C. The mean number of third instar onion maggot larvae consumed per day is also linearly related to temperature (Table 7, Figure 12). However, a lower threshold for feeding could not be estimated by extending the regression line, as this method yields an estimate less than 0°C. Consumption does not appear to be linearly related to temperature below 13°C; therefore, the regression model cannot be used as an accurate predictor below this temperature. Read (1962), testing one individual per dish over time, presumably each at 23°C, found that A. bilineata adults destroyed an average of 2.6 third instar larvae or pupae per day. In this author's experiment, with three maggots available, a pair of staphylinid adults at 23°C rarely consumed their entire food supply ()1 consumption per pair per day = 1.13 maggots). From observations of predation behavior, when a staphylinid attacks a large larva, it does not necessarily consume the entire larva. Once satiated, the individual moves on, and may attack the next larva it encounters. Also, with more than one staphylinid adult per chamber, an individual was seen to frequently "join in the fight" when encountering another individual attacking a large larva. These observations suggest that absolute predation rates may be a complex function of A. bilineata adult density and prey density. This relation- ship needs to be investigated to quantify predation. From this data, predation is assumed to also vary with temperature, and this variation can be predicted within the range of temperatures tested. 42 .wuaumuanou mo cowuocam o no muwwczomm mumocaafifi Hoou mashcmmmEMH ma 8 w— P e L .< .HH muawwm ou xmm. + 061 I x 5 AUG 83d BWUHEJ 83d 9003 43 Table 6. The influence of temperature on the fecundity of A. bilineata. Temperature EggsiLaid PersDay Number of Females ngzir5::::530f 29 17.85 12.02 7 39 23 12.03 9.36 5 36 17 7.80 7.50 7 46 13 2.03 3.95 6 33 Table 7. The influence of temperature on the consumption of third instar onion maggot larvae by A. bilineata adults. Larvae Consumed Total Number of Temperature Per Pair Per Day Number of Pairs Observations i’ s 29 1.69 1.00 8 38 23 1.13 0.85 7 47 17 1.02 0.91 6 43 13 0.77 0.95 6 33 44 .ounumuoasou mo cofiuocsm m mm om>umH zo unumafi bHMSu mo coaumaamcoo mumwcHHwn .< .NH ouswfim noou mmahcmmmzme on ma on ma r _ c nm..u q. .60. + no.1 u x Cu SL'I AUG 83d NOIldHnSNOO WUAHHW X 45 A. BILINEATA OVlPOSITlON PREFERENCES Materials and Methods An oviposition chamber designed to evaluate preferred choices for oviposi- tion by A. bilineata females was constructed out of four, square, pint-sized, plastic food containers. Two adjacent sides of each container were cut down to 2 cm in height. The short sides of all containers were fastened together and sealed with duct tape to prevent any possible escape. The tall sides of adjacent containers were fastened together and sealed the same way. The bottom of the chamber was then filled just to the height of the inner walls (ca. 2 cm) with moist sand. The resulting four-sectioned chamber permitted free movement throughout the entire arena for A. bilineata adults traveling along the surface of the sand, but limited movement of A. bilineata larvae traveling through the sand to a single section. Oviposition preferences were evaluated as follows. One of each of the four treatments to be compared was placed in the center of one of the four sections. Four mating pairs of A. bilineata were released into the arena. The top of the chamber was secured with a cotton muslin lid. Adults were free to travel about the chamber for 48 hours, after which they were removed and 25 to 50 OM pupae were added to each of the four sections. (An equal number was added to each chamber in a single trial.) Pupae were gently buried within the sand as to not disturb any of the staphylinid eggs or larvae. The pupae were left in the chamber for six days, after which they were removed and examined for holes indicating A. bilineata larva attack and entry. As the parasitizing larvae could not travel between sections of the chamber in search for pupae, the number of parasitized pupae within a section was considered a relative measure of the 46 number of eggs laid within that section. Hence, the number of parasitoids per section was used as an indirect measure of oviposition activity and was compared between sections to reflect any preference in oviposition with the treatments tested. The treatments were randomly distributed among the sections for each experiment. The experiments are listed in Table 8. Results and Discussion The results of the A. bilineata oviposition preference experiments are presented in Figure 13. In the first experiment, a larger number of eggs (as reflected by successful parasitism) were laid around the control, a moist sponge, than around any of the other treatments. No preference between the other treatments tested (an onion, an onion with host pupae, or a maggot-infested onion) was seen. Preference for the sponge was assumed to reflect a preference for high moisture. The same test was repeated again in experiment 3, only particular attention was paid to maintaining the same high soil moisture in all four sections of the treatments, including the moist sponge. The resulting parasitism indicated no particular preference for any of the treatments, includ- ing the moist sponge. To further test the attraction of A. bilineata to moist areas for oviposition sites, a test was set up to evaluate any preference for oviposition between four levels of soil moisture (Experiment 4). As reflected in the resulting parasitism, preference increases with soil moisture. Twice as many larvae parasitized pupae in the low moisture section than in the dry, and twice again as many in the medium moisture section than the low. Parasitism between the medium and the high moisture levels showed little difference. 47 F‘- _‘ O 7' 2 / . g4 //é 8 / 7/ 8 2.4 8 / / / g / 7 7 ..7 g j a 1/71 . v7//,{ 17% o A /x. 8.. 1 2 J 4 4 mumrm :4 24 [Egg number of parasitoid attacks ”7 :::3 number of parasitoid larvae /§% - successfully entering pupae °7 1 A 2 s 4 ‘ TREAHMENT TREAJMENT Figure 13.1-13.5. Oviposition preferences of A. bilineata. Table 0. A. bilineata oviposition preference experiments. Experiment Treatment 1 2 3 4 1 control - moist onion with 0M onion maggot-infested sponge pupae onion 2 maggot-infested maggot-infested control - no control - no onion onion treatment treatment 3 control - molat onion with 0" onion maggot-infested sponge pupae onion 61 no moisture: low moiature: medium moisture: high moisture: dry sand 5 ml HZO/aection 10 ml Hzolaection 15 ml “Zolscction 5 radish maggot-infeated maggot-infested onion radiah with pupae onion with pupae 1 For experiment 4, each of the four aectiona of the chamber were intially filled with dry aand to a depth of ca. 2 cm. The above different volumea of water were added to the aand to create various levela of aoil moiature. "hen the adult ataphylinida were removed and the pupae were added, half the initial volume of water recieved by each aection was again added. 48 Alone, this method of evaluating oviposition preferences does not ade- quately test responses to differing levels of soil moisture. The differences in parasitism may also reflect differing egg and first instar larval survival under different soil moisture regimes. Experiment 1 indicates that A. bilineata females prefer areas of high soil moisture. Experiment 4 probably reflects these differences, but to accurately evaluate this test, egg and larval survival under such conditions must be examined. Experiment 5 compares A. bilineata preference for ovipositing about radishes and onions, and any added influence of maggots and pupae. Maggots infesting onions and maggots infesting radishes appear not to influence the attractiveness of their host plants as oviposition sites. However, radishes with or without maggots and pupae appeared to be more attractive than onions with or without maggots and pupae. This greater attractivity of radishes could be one factor responsible for the greater staphylinid parasitism in maggot-infected radishes than adjacent maggot-infected onions in the field study plots. The results of experiment 2 comparing the attractiveness of maggot- infested onions to a control or no treatment are puzzling as there appeared to be a greater response to the control than to the onions. lassumed that, as previous experiments indicated no ovipositional preference about onions with or without maggots or pupae, the greater parasitism in the control section is probably due to random chance, and, given adequate replication, there would be no significant difference between the onions and the control. MULTIPLE PARASITISM An experiment was conducted to determine whether A. bilineata larvae 49 searching for OM pupae would distinguish and select against A. pallipes parasit- ized OM pupae. Materials and Methods Fifty OM pupae, half parasitized by A. pallipes and half healthy, were buried in the sand in the A. bilineata colony for 48 hours. The pupae were then removed and examined for attack holes and A. bilineata parasitism. The experiment was repeated 3 second time with 24 OM pupae, half parasitized by A. pallipes and half not. Results and Discussion During the first trial, ten healthy OM pupae were parasitized by A. bilineata, and two healthy pupae showed signs of attack although they were not parasitized. Thirteen of the healthy pupae remained unparasitized. Only one of A. pallipes parasitized pupae showed any signs of A. bilineata attack, and none were parasitized. Similar results were obtained in the second trial. Seven healthy pupae were parasitized by A. bilineata, and five were not attacked. None of the 12 braconid-parasitized pupae were attacked. These results indicate that A. bilineata larvae recognize OM pupae attacked by the larval parasitoid A. pallipe . Bromand (1980) reported similar findings regarding A. bilineata attack on OM pupae previously parasitized by Tribliographa rapae. The mechanisms for detection are not known. DIET INFLUENCES ON THE ADULT LONGEVITY OF A. PALLIPES A carbohydrate food source for several adult hymenopteran parasitoids significantly increases the longevity and fecundity of the parasitoids (Leuis 1960, 1961, 1963, 1967). Several investigators suggest that the availability of 50 carbohydrates in the form of nectar and pollen sources influence the distribution and abundance of many hymenopteran parasitoids (Symes 1975). The influence of diet in the longevity of adult A. pallipes was investigated to determine if the availability of a food source might influence their distribution and abundance about the onion agroecosystem. Materials and Methods Forty-two newly-emerged adult A. pallipes were separated into groups of seven (3 males and 4 females) in rearing dishes covered with a fine cotton muslim cloth secured with an elastic band. The parasitoids were fed and watered every other day by touching a corner of the muslin cloth with the food mixture and moistening the sand in the rearing dish. Each dish of parasitoids was maintained on a different diet. The diets tested included: (1) a 10% honey solution, (2) a 10% sucrose solution, (3) a 10% sucrose solution with brewers yeast, (4) a 10% honey solution with brewers yeast, (5) a cut onion, and (6) a control of distilled water. Parasitoids were monitored daily for mortality. Results and Discussion A. pallipes longevity on diets of 10% sucrose or honey with or without brewers yeast was significantly longer (F5141 = 14.33) than with either an onion food source or just water (Table 9). There were, however, no significant differences in longevity between the different sugar diets. Adding brewers yeast to the diets did not increase longevity. Over all diets, the average male and female survival did not differ. The greater than three fold increase in adult A. pallipes longevity with a carbohydrate food source suggests that the availability of this food source in the field could influence the abundance and distribution of this parasitoid. 51 Table 9. The influence of diet on the adult longevity of A. pallipes . 3f longevity D' 1et (days) 10Z honey solution 13.29 a lOZ sucrose solution 15.43 a lOZ hgney solution with 14.86 a rewers yeast lOZ sucrose solution with brewers yeast 18.14 a Cut onion 5.00 b Control (distilled water) 4.57 b 1Treatments followed by the same letter are not significantly different at a = .01; Duncan's multiple range test. 52 FIELD INVESTIGATIONS SPRING EMERGENCE Materials and Methods In 1980, overwintering A. m and A. bilineata parasitized OM pupae were monitored for adult emergence. Parasitized pupae were collected from onion fields in the fall of 1979 and buried at 2.5 cm, 7.5 cm, 12.5 cm, and 17.5 cm depths in plastic buckets filled with muck soil. Each bucket contained ten parasitized pupae and each depth was replicated three times for A. m and twice for A. bilineata. The buckets were perforated with 1 cm holes to facilitate drainage, and buried in an onion field at the MSU Muck Farm, such that the top of each bucket was flush with soil surface. In late April 1980, emergence traps (Cage and Haynes 1975) were emplaced over the buckets, and the traps were monitored every three days for the emergence of the adult parasitoids. A. m adults were easily recognized by the yellow legs and basal antennal segment. No attempt was made to identify A. bilineata adults to species; all staphylinids of the appropriate size and color were recorded. Results and Discussion First emergence of A. m adults occurred on June 2. The mean emergence date was June 10 (sx=.58 dys.), a week earlier than that of A. bilineata. Female A. pallipes emerged an average four to five days earlier than male A. pallipes (Table 10). The date of emergence varied with the depth at which the pupae were buried. The parasitoids in pupae at the deeper depths emerged later and over a longer span of days than the shallower ones. A. bilineata emergence did not vary with depth. The first emergence of a 53 Table 10. Spring emergence of A. pallipes adults. Emergence Date Depth (cm) x s- n i s; n 2.5 157 .77 16 154 .31 36 7.5 161 - 1 156 .69 23 12.5 166 .99 18 164 .81 31 17.5 172 2.12 7 169 1.52 14 54 staphylinid adult was June 6 (287 degree days, base 8.6°C). Emergence continued through the first week in July, yet the mean emergence date was June 17 (Sx=.74 dys.; 387 degree days, base 8.6°C). Using the model for A. bilineata spring emergence developed from laboratory data in the literature and MSU Muck Farm soil temperatures monitored at a 10 cm depth, the predicted emergence date for this parasitoid is in exact agreement with the mean observed in the field. Fifty percent emergence for the OM (Whitfield 1981) averaged 16 days earlier than A. pallipes, and 23 days earlier than A. bilineata. Thus both parasitoids appear well synchronized with the OM life stage they attack. The spring emergence data was used to estimate a lower threshold for development and accumulated degree days required for emergence of A. pallipes adults. As soil temperatures were monitored at depths intermediate to those at which the pupae were buried, the mean emergence dates from two adjacent depths were calculated to generate estimated emergence dates for three A. pallipes populations to correspond with the three sets of temperature data. Modifying the method described by Casagrande (1971), hypothetical thresh- old levels were combined with soil temperature data to predict the accumulated degree days required for emergence. The mean and standard error for the predicted degree days at the different depths was calculated from a range of hypothetical thresholds. These standard errors were plotted against the thresh- olds (Figure 14), and the threshold corresponding to the lowest point in the parabolic curve was chosen in the best estimate. A lower developmental threshold of 9.0°C with 292 degree days required to complete development was chosen as the best predictor. This model was not validated and requires an .mwmwaamm .fl.wcwuoucw3hm>o How ououmquEou ammo HmucoanHo>mv mo coaumcfisnouow How bosons uouuo vumccmum .qa ouswwm moi washcmmmzuh o.- 0.: DA: o.m 0.0 0.5 Pi _ _ _ _ ma 8 ..IL x x x x x o m ”U N In. 0 X 3 U ud X 0 1L 3 .7 no no X 0 1h ud 1m. 1 S x. X 1.1. r. 8 X I 0'9 56 independent data set to do so. SUMMER DYNAMICS Materials and Methods Field investigations on the OM and CM parasitoid populations were conducted in three different Michigan onion production regions from the fall of 1979 through 1980 (Figure 15). The two primary study sites were the Michigan State University Organic Soils Research Farm in Laingsburg, Michigan (Bath Township, Clinton County), henceforth referred to as the MSU Muck Farm, and an organic onion production farm near Eaton Rapids, Michigan (Eaton Rapids Township, Eaton County). The research plots at the MSU Muck Farm were a combination of experimental plantings of onions and radishes, some with and some without insecticide inputs. All onion plantings received one or two herbicide treatments, but no fungicides. The radish plantings received neither. The organic onion production farm in Eaton Rapids was maintained free of any chemical inputs including pesticides and commercial fertilizers. The grower culturally controlled (crop rotation, delayed planting, strip planting, cultivation, etc.) insect, disease, and weed pests and soil fertility. Additional population monitoring and post-harvest experiments were conducted in the Rice Lake vegetable production region (Grant Township, Newaygo County) from the fall of 1979 through the fall of 1980. Detailed descriptions of each of the study areas follow. These study sites were utilized by several investigators researching the dynamics of the OM, natural enemies, plant growth, and several other biological parameters of interest in the overall onion agroecosystem project. 57 Location of researchx sites for onion agroecosystem project. Figure 15. 58 Organic Farm, Eaton Rapids Description of Study Area With the cooperation of the grower, Mr. Dale Kunkel, standard production practices at the organic farm were modified to incorporate two beds of radishes adjacent to onion plantings. Otherwise, this research was conducted in fields arranged by the grower (see Figure 16 for a map of the research site). The grower commonly planted crops in long, narrow fields, maximizing the inter- facing between different crops. The onion fields were planted the first week of May. With the exception of the fall surveys, most of the OM damage and population monitoring throughout the growing season was conducted in field three, as OM populations in fields one and two were too low to efficiently sample. Two plantings of radishes were sown along the east border of field one. Each planting consisted of three 15 m rows. The first crop was sown in early June (June 6) and the second in mid-July (July 14). The Charlotte weather station provided records of daily maximum and minimum temperatures through- out the year. Damage and Population Sampling Onions: Three methods for estimating OM damage were used at different times during the growing season to accommodate the needs of the study and varying OM densities. Initially, damage sampling was conducted as determined optimal for within-field damage estimates by Carruthers (1979). Ten randomly selected 30 m samples of onion row were examined for the number of OM- damaged plants. With experience, OM damage can be recognized by examining the onion leaves for loss of turgor and yellowing (Carruthers 1979). Percent damage was estimated from the number of damaged plants (on that sample date) 59 N 1') '2 .2 u. ‘2 e 8 o TF—T f— F—j E RADISH 8 3rd Planting N '- § ‘7’, T RADISH i: i: g g 3 2nd Planting "’ o 3 o 2 < E i— <3 C) <3 _ 1 RADISH E II) '- 1st Planting 1.5m Lp—-—24m——" n-Vm-ul L7m-——-1.7m Figure 16. Research fields at organic onion farm in Eaton Rapids, Michigan, 1980. 60 divided by the average plant density of 725 plants per 30 m. This sampling plan was carried out over three sampling dates from June 6 to June 27. After this time, damage sampling was stratified relative to the distance of the sample from the field's border to coincide with stratified sampling for pupal parasitism. (If the incidence of parasitism differed with distance from the field border, it would be important to know whether or not this spatial pattern was independent of the host's distribution.) Each strata consisted of six single rows (two beds) of onions parallelling the east-west field borders. Seven strata, encompassing the width of the field, were sampled. Five randomly-selected 8 m samples of row were examined per strata to estimate percent damage. Damage estimates were pooled over all strata to determine the field estimate of damage. OM damage was estimated as such from July 4 to July 18. OM damage and OM pupal density were estimated at the same time due to limited resources. With decreasing OM densities and the intensive efforts required in pupal sampling, a combination of distance and row sampling was the most efficient (see pupal sampling, methods). The distance down a row (north- south direction randomly selected) to the nearest damaged onion from a random point was measured. The number of OM-damaged plants and healthy plants within a .3 m section of row beyond the initially encountered damaged plant was recorded. The percent damage was calculated as: % damage = (number of infested p1ants)/(2 x distance (m) + .3) 7.25 plants/m) Five random samples per strata were collected on four sampling dates from July 18 through August 15. For a comparison of the two methods, row sampling by bed and distance-row sampling by bed were conducted on July 18. 61 No attempt was made to monitor densities of OM eggs and larvae. Maggots from 10 to 20 damaged plants were sampled routinely and aged to obtain an estimate of the age structure of the population to determine when to initiate first generation pupal sampling. Initially pupae were sampled by sifting through the soil under randomly selected damaged plants to a depth of ca. 15 cm. Samples were stratified relative to the distance of the sample from the field border. As in the damage sampling, each stratum consisted of six single rows of onions. Five to ten pupae were collected per stratum on four occasions between July 2 and July 11. Although this sampling method determined the percent parasitized pupae on any one day, estimates of the seasonal incidence of parasitism could not be made without a measure of parasitized and non-parasitized pupal density. Therefore, a pupal sampling strategy based on a measured area sample was developed. As OM pupae are not randomly distributed about a field, and the effort involved in sampling pupae is quite intense, sampling randomly selected unit area plots was too inefficient to be considered. An area with a high probability of containing pupae—usually around a damaged plant--was found. The distance up or down a row from a randomly selected point to the nearest damaged onion was measured, and the soil under a .3 m section of row beyond and including the encountered damaged onion was sampled to a depth of ca. 15 cm for OM pupae. Pupal density was then calculated as: density = (no. of pupae) divided by (2 x distance (m) + .3 m) Pupae were examined for signs of parasitism. Radishes: As at the MSU Muck Farm (see following section: MSU Muck 62 Farm), the radish beds in Eaton Rapids were sampled for damage and maggot population density by randomly selecting five samples of five adjacent plants per planting and examining these for feeding damage and maggots. The maggots were aged and identified to species. CM pupae were sampled by sifting through the soil to a depth of ca. 12 cm under each five-plant sample of radishes. Sampling of the first planting was initiated on July 18 and was continued once every two weeks through the growing season and the fall. Sampling of the second planting was initiated in mid-August and continued through the fall. MSU Muck Farm Description of Study Area Experimental plots were designed to investigate the effect of (1) the time of planting of onions and radishes, and (2) the use of the soil granular insecticide, Dyfonate, on maggot damage and A. bilineata production. Each plot was 14 m x 15 m large divided into four 6.75 m x 7.5 m sections (Figure 17). Four treatments (onion with Dyfonate, onion without Dyfonate, radish with Dyfonate, radish without Dyfonate) were randomly assigned. Each stratum contained four beds of plants with three single rows per bed. Each planting was bordered on the east and west by a 1.5 m x 15 m strip of oats. The first planting was sown on May 7, about the time most Michigan onion growers plant. The second planting was sown on May 14; the third planting was sown on May 28; and the fourth planting was sown on June 12. The corresponding degree day accumulations for OM, CM, and A. bilineata on each planting date are presented in Table 11. The Muck Farm weather station provided daily maximum and minimum temperatures. A three-point thermograph monitored soil temperatures at three .owaa .cnwneuwz .musnmewmg :H Show x052 sz ecu um muoHa Luumomwm .NH madman 2:52.. 8.2.2.. 8:521 8:5... .n. new new saw 63 also was Rum .\.\ Rx s \. RX AR \ Rx 9R Nan R._m.\\\.\.m.\\ \RwR \\ s _ KR S\&& 8190 9:90 ciao J_r——ums1--1——4mr2-——~. E 1‘? ’ 39.025 505.3 0.95:5 5.3 steo 64 Table 11. Degree day accumulation for OM, CM, and A bilineata on plant- ing dates at MSU Muck Farm, 1980. Degree-day Accumulation1 Planting Date onion maggot cabbage maggot A, bilineata base 4.4 C base 6.1 C base 8.6 C 1 May 7 117 86 49 2 May 14 168 125 72 3 May 28 341 275 186 4 June 12 557 465 339 lFrom soil temperatures monitored at a depth of 10 cm. 65 depths (5, 10, and 15 cm) from November 1979 through July 1980. In August and September, the three-point thermograph monitored the soil temperatures at a depth of 3 cm under the onion canopy, the radish canopy, and the oats. Damage and Population Samplim ME OM-damaged plant assessments were initiated in each planting when the onion plants reached 3-6 cm in height. This was on May 31 for the first and second plantings, June 23 for the third planting, and June 30 for the fourth planting. From May 31 through June 30, damage assessments were conducted weekly by examining every onion plant in each plot. As the rate of damage decreased, damage assessments were reduced to sampling one row per bed in each of the four beds per plot, once every two weeks. No attempts were made to obtain density estimates of OM life stages within the plots. Observations of the age structure of infesting maggots in 10 to 20 damaged plants were made routinely to estimate when to initiate pupal sampling. Five to ten pupae were sampled per bed in those plots with OM infestation substantial enough to warrant the intensive effort necessary for sampling pupae. Pupae were examined for the presence of parasitoids. Radishes: Unlike onions, in most instances, maggot damage to radish was detectable only by pulling up the radish plant and examining the bulb for maggot feeding. Therefore, it was necessary to choose a sample size that would give an accurate estimate of infestation levels, yet would not appreciably change the resource with repeated sampling through time. It was arbitrarily decided that five randomly selected samples of five adjacent plants per plot would meet these criteria. Sampling began on May 24 in the first and second plantings, June 16 in the third planting, and June 30 in the fourth planting. Plots were sampled 66 weekly from May 24 through June 30 and once every two weeks through the rest of the season. Sampled plants were returned to the laboratory where they were examined for feeding damage and the presence of maggots. The maggots were examined for species and age determinations. As the maggots in radishes began to pupate, samples of fifty pupae were collected in infested plots and examined for parasitism. Pupae were collected four times between June 21 and July 3. At this time, it was decided that infestation levels were high enough that pupal sampling could be combined with the radish plant sampling, giving a measure of pupal density, and still retrieving enough pupae to evaluate parasitism. Therefore, from July 14 through the rest of the growing season, the soil under the 5-p1ant samples of radishes to a depth of ca. 12 cm was sifted for pupae. These pupae were then examined for symptoms of parasitism. Grant In 1980, OM pupae were periodically sampled in a few commercial onion fields in Grant, Michigan, to determine a relative estimate of parasitism. First and second generation pupae were collected from a 10 hectare field which was treated with Dyfonate at planting but received no foliar sprays (field 1), and an adjacent field which was treated with Dyfonate and had seven foliar sprays of either Sevin or parathion throughout the season (field 2). Second generation pupae were also sampled from a small onion field receiving foliar sprays but no Dyfonate at planting (field 3). On June 3, four double rows of radishes were planted along the west border of field 1. These radish plantings were periodically sampled in the same manner as those at the Eaton Rapids research site. Maggot damage and infestations 67 were monitored and pupae were collected and examined for parasitism. After a portion of the radish plantings were destroyed by the onion grower's cultivation practices in July, new plantings were sown within the older plantings. Results and Discussion Evaluating Parasitism Parasitism of insect populations was frequently evaluated by percent of parasitized individuals in a single sample (usually taken at peak abundance of the attacked life stage). This measure of relative parasitism between comparisons (i.e., fields, habitats, hosts, etc.) can also measure a parasitoid's impact on its host population (if the population is sampled when all parasitized and non- parasitized individuals in a specific life stage are present, as would be during an overwintering stage). However, for multivoltine species and parasitoids attack- ing non-diapausing life stages, this estimate can be very misleading. Many parasitoids develop at different rates than their hosts, and without considering this, one cannot accurately estimate the proportion of the population actually parasitized. To obtain an unbiased estimate, the total seasonal production of parasitized hosts to the total seasonal production of all hosts (parasitized and non-parasitized) was calculated. The problem was similiar to that described by Southwood (1966, 1976), Helgesen and Haynes (1972), Kiritani and Nakasujii (1967) and Manly (1976), when estimating age-specific mortality in a population where all age classes occur simultaneously. If the population census is taken frequently, an occurrence curve can be established for each life stage. The total seasonal incidence can then be calculated by (Lampert 1980): 68 n-l Di. + Di'+l Tr=z [—Lf—i—)(DD.l-DD) 1:1 1+ 1 where: TIi = total seasonal incidence for life stage i Dij = density of the 1th life stage on the jth sample date DDj = accumulated degree days on the jth sample date The total seasonal incidence (Tli) estimates the stage-degree days. To determine the total seasonal production (TPi) of the stage, and correct for redundancy in sampling, divide the total seasonal incidence by the developmental time for the stage (DTi): T. TP.= __Il 1 DTi Lampert (1980) evaluated total life stage parasitism of the cereal leaf beetle (Oulema melanopus) larvae by multiplying this host density (Dij) times the sampled proportion parasitized (Ppij) to estimate the parasitized host density (pDij)' He then calculated the total seasonal incidence of parasitized hosts and, by dividing this by the total seasonal incidence of the life stage, determined the proportion of the life stage parasitized as follows: 2.. «Dij Ppij) + (D.. PP l))/2 (DDj+l - DDl) PPar.= 11 “+1 4+ 1 T5 If the developmental rates of the parasitoid and its host are synchronized, the ratio between total seasonal incidence of parasitized hosts (as calculated above) to the total seasonal incidence of the host life stage, accurately estimates the proportion of the population parasitized. However, if the 69 developmental rates of the parasitoid and its hosts differ, as with A. bilineata and its hosts, the OM and CM, the sampled proportion parasitized on the jth sample date is a biased estimate. To correct for this bias, the total seasonal incidence of parasitized host (TPIi) must be calculated from the sampled parasitized host density (spDij) and accumulated degree days based on the parasitoid developmental thresholds, as follows: n-l SPDij +5903.+1 TPlizs. ( 2 ) (DDp.+ -DD .) l 1 1 ':1 1 P1 The total seasonal production of parasitized hosts (TPPi) is then calculated by dividing this seasonal incidence of parasitized hosts by the developmental time of the parasitoid: TPI. TPP. = l 1 DT p To calculate the proportion of the life stage i parasitized, the total seasonal production of non-parasitized hosts (TNPi) must be calculated based on sampled densities of non-parasitized hosts and the host's developmental thresh- olds. The proportion parasitized is calculated by: TPP. PPar. : l 1 TPPi + TNPi This bias, however, can lead to incorrect conclusions. For example, figure 18 shows non-corrected and corrected parasitized pupal densities relative to the total sampled pupal density of CM in the first radish planting at the MSU Muck 70 .20 5.3: 3H can numoaadn .6. mo unoaaoamgov mo moumu on”. cw moocouommwc ou use mafiaaamm ca hocmvasvou wow wafiuoouuoo .wa ouawfim >¢o choap cow ovw owu cow cg :2 P c . . _ .. 10 T 1.6 r? £2.06 .83.. .8quan 6302.3 Queen Base _33 “.033 I 1 19 Queen .25.. «63:?an vouoPcooe: 0 Epcot Base .33 6302.35. XI 1 .X . r9 SlNUWd S / AlISNBU 71 Farm. To correct for the bias at each sample date, sampled parasitized pupal density was multiplied by the ratio of A. bilineata developmental time to CM pupal developmental time accumulated between successive sampling dates (j - l and j). Without correcting for this bias, the total seasonal proportion CM pupae parasitized was 62%. Yet, with the seasonal temperatures experienced, the parasitoid required 1.6 times as long to develop as its host, and therefore was more likely to be redundantly sampled. Correcting for this difference revealed a true proportion parasitized of 26%. Organic Farm - Eaton Rapids Seasonal OM Damage OM damage in the research field at the organic farm in Eaton Rapids reached 40% of the total crop (Figure 19) during the 1980 growing season. Eighty-one percent of the damage occurred during the first OM generation. This pattern was similar to the pesticide-free plots at the MSU Muck Farm, but differed from the pesticide-dominated commercial onion fields (Whitfield 1981). Most commercial onion growers plant their onions along with the systemic, soil granular insecticide, Dyfonate. Dyfonate reduced the impact of the first generation OM, but lost its effectiveness before second generation damage began. The effectiveness of chemical foliar sprays for second generation adults is questionable (Whitfield 1981), hence a significant portion of crop damage (39 to 79%; Whitfield 1981) happens during the second generation. Without chemicals, first generation damage was significant, but second generation damage was reduced. This reduction is partially due to the dynamics of the onion plant and OM adult oviposition behavior. As the bulb matures, each individual plant can support more maggots, and because female flies prefer to 72 nzwu . .owma .vawwm consume» mcwamm acumm cw ownEmu 20 .aa mnsmnm >¢o chuap cam om~ ova om“ . . . . 7» . nu um .6 IO 9 .Inu .7 10 9 BOUNHU 1N3383d 3A11610H03 73 oviposit on previously-infested onions, a loading effect occurs. Hence, the same population density causes less damage later in the season. It is also likely that a build-up of natural enemies in the pesticide-free environment affected OM mortality and hence reduced OM damage. An analysis of variance for factorial design was performed on the plant damage data to determine differences in OM damage with distance from the field border over time. No significant differences existed in OM damage levels between beds sampled (F6,112='996)° OM activity was evenly distributed about the width of the field. Only two out of 180 larvae sampled in Eaton Rapids during the first week of August were SCM; the overwhelming majority were OM. Some of the early season damage probably was caused by SCM, but larvae were not identified to species. Parasitism of OM Sampling absolute densities of OM pupae at Eaton Rapids was labor- intensive and did 92; yield an adequate data set for determining the seasonal incidence of parasitized and non-parasitized pupae. As a result, a relative measure of parasitism corrected for the differences in host and parasitoid developmental times was calculated from numbers of periodically sampled pupae. The ratio of accumulated degree days from one sample date to the next divided by the developmental time of the parasitoid (385 degree day, base 8.6OC) to accumulated OM degree days divided by the developmental time of OM pupae (306 degree day, base 4.4°C) was multiplied by the number of sampled para- sitized OM pupae to give the corrected value (n pl). The percentage of sampled OM pupae parasitized by A. bilineata in Eaton Rapids reached 17.0%. After correcting for the developmental times, it 74 decreased to 10.9% (Table 12). The largest proportion of hosts parasitized was sampled during the later part of the first generation of pupae. This might be expected, because the parasitoid continues to attack through this life stage. Also, because it takes longer to complete development, the parasitoid is likely to still be in the soil when healthy flies emerge. An analysis of variance reveals no significant differences in the proportion of OM pupae parasitized in the different beds around the width of the onion field (F6,24:'9M)' A. bilineata activity did not appear to be influenced by the proximity of the field border. The width of the onion field samples in Eaton Rapids is considerably less than most commercial onion fields. This may or may not represent the spatial distribution of parasitoid activity in larger fields. CM Damage in Radishes The first radish planting in Eaton Rapids had reached nearly 100% damage by CM by the first sample date (July 18, see Table 13). Seventy-six percent of the sampled radishes were infested with CM larvae, of which 96% showed some signs of feeding damage. Little damage was experienced in either the second or third radish plantings (8% and 4% respectively), and damage was not initiated in these plantings until September. Seven percent of larvae sampled in the radish plantings on July 18 and l of the 4 larvae sampled on August 29 were SCM. The majority of the maggots sampled were CM; no other species of radish-infesting maggots were detected. Unlike onion plants, maggot-infested radishes continued to grow. Plants attacked by CM larvae early in the season continue to attract CM flies for some time. As a result, although the first radish planting by July 18 had 100% damage, the food resource for maggots was not depleted, and these plants were 75 Table 12. ‘A.bilineata parasitism of OM at Eaton Rapids research site. Corrected OM Aleochara Proportion Proportion Date Degree Days Degree days n Parasitized Parasitized1 178 746 478 0 0 0 183 823 535 54 .09 .07 185 853 556 69 .07 .04 189 916 602 21 .24 .11 192 969 642 33 .33 .23 199 1098 742 5 .60 .36 203 1176 803 17 .18 .12 213 1337 923 67 .22 .11 226 1555 1086 45 .16 .07 222 1520 1092 12 0 0 Season Total 1570 1097 323 .17 .10 1Proportion parasitized corrected for differences in develop- mental times of the host and parasitoid (see text). 76 Table 13. CM infestation in radishes at Eaton Rapids research site. 1St Planting, 2nd Planting, 3rd Planting i z i z E z i z i z E 2 Date Damaged Infested Damaged Infested Damaged Infested plants plants plants plants plants plants 199 96 (2.0)3 76 (7.0) — - - _ 217 80 (6.0) 8 (4.0) 0 (0.0) 0 (0.0) - — 227 100 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) - - 242 76 (5.0) 4 (2.0) 15 (5.0) 4 (2.0) 0 (0.0) 0 (0.0) 255 92 (2.0) 0 (0.0) 8 (4.0) O (0.0) 8 (2.0) 0 (0.0) 303 80 (3.0) O (0.0) 20 (4.0) 0 (0.0) 8 (2.0) 4 (2.0) a standard error 77 reinfested. Parasitism of CM The ratio of the seasonal incidence of parasitized pupae to the seasonal incidence of all pupae determined the seasonal parasitism of CM pupae in radishes at Eaton Rapids. Corrected for differences in developmental times, an estimated 74% of the CM population was parasitized (Figure 20). Total seasonal production of parasitized pupae was estimated at 2.72 pipae/5 plants, compared with .94 non-parasitized pupae/5 plants. Though parasitoid densities were not directly measured, the A. bilineata population was fairly high. Eighty-one percent of the parasitized pupae were attacked by more than one parasitoid larvae. Parasitoid activity in the radish plot was well synchronized with its host population for the first generation; however, no parasitoid activity was detected in the fall generation. MSU Muck Farm OM Damage as Affected by Planting Date and Insecticide Use OM damage was first detected in plots with Dyfonate and plots without Dyfonate in both the first and second plantings on the same date (June 8, Figure 21). Damage levels were considerably higher in the non-Dyfonate plots and continued to be so through the end of June. After this time, the damage rate for all plots declined, corresponding with first generation pupation. Damage slightly increased during the second generation of OM in the first planting, non-Dyfonate plot. The third planting plots and the Dyfonate fourth planting also had damage. Little additional damage was incurred in either of the second planting plots. Five percent of the first generation maggots infesting onions at the MSU Muck Farm were SCM. The remainder from the first and subsequent generations 78 .6flOUflm Imuma 6:6 000: 0:0 mo ucoaaoao>06 mo moumw 0:0 cw 000cm Iuowmw6 you pouoowuoo mum mmfluwwamo .owma .m6fiamm :oufim aw muoaa nmwwmu cw woman 20 mo Bmfiuwmmuum mumocfiawn .< .om muswwm xco chuzp mum com mhw 0mm www com m: 7 Oh _ 0L h 7 r0 /. 2 33:06 _oaaa 602.3082. 6308.60 0 : . 90:06 69.... 6309.50 X r8 kl? SlNU'Id S / A119N30 .83 .86... 0.6:: am: 2: an amass... .8 no uuucomha mo 00: 0:0 van 0066 wcwuamaa mo mesmSHmafi 05H .HN ouswfih 9 328... .85.: 68 5... 95%... .4 6:6 368.5 5... 955... ..n x 368...... .85... 855... ..n a 322.6 5.: 95.6... ..N < 368}. 5.: 6.55... L a 365.8 565... 655... L < 6.65.5 365... 955... s. e 79 ovN w>¢o zc H .5... own cum o—N DON {PLID b h h Pb PL bL P*Pblrb h m m‘b h En MW in} cm" _ p P D can Ob— ..._ I11 fifi nsz SlNV'ld GEOVWVO .LNBOEBd BALLV'IFIWHO 80 were OM. OM damage in the different plots was strikingly different between planting dates and the soil granular insecticide, Dyfonate. In the non-Dyfonate plots, seasonal damage experienced in the second planting of onions was 40% of that experienced in the onions planted without Dyfonate seven days earlier. Onions planted three weeks later had 6% damage. This was less than the damage incurred in the first planting Dyfonate plot. The difference in damage levels between Dyfonate plots and non-Dyfonate plots of the same planting date decreased with delay of planting. Though OM damage was slightly higher in the non-Dyfonate plot than in the Dyfonate plot of the third planting, total cumulative damage levels for both over the entire season were still less than 4%. Neither of the fourth planting plots experienced any substantial OM injury. The percent cumulative damage for the Dyfonate plot and non-Dyfonate plot were .003% and 0%, respectively. Planting date and the use of the soil granular insecticide in early season plantings greatly affected OM damage. Yet, delayed planting alone probably will not reduce OM damage. The presence of a preferred resource in the form of earlier planted, perhaps previously-infested onions (Carruthers 1979) probably affects damage in later plantings. However, this indicated that delayed planting in conjunction with the use of an earlier planted trap crop could be a valuable component of an integrated management strategy for OM (see Discussion). Parasitism of OM Pupae OM densities in the experimental plots at the MSU Muck Farm were high enough to repeatedly sample pupae only in the earliest non-Dyfonate onion planting during the first generation. One sample was taken in the second non- Dyfonate onion planting about peak pupation of the first generation. The results 81 of OM pupae sampling are presented in Table 14. Parasitized pupae were very low in number, accounting for only 2.9% of all the pupae sampled. CM Infestations in Radishes CM infestations in radishes at the MSU Muck Farm are presented in Figure 22 a-h. CM egg laying was first detected in the earliest non-Dyfonated radish planting on May 30—three weeks after the plants were sown. Within ten days, more than 50% of the plants were infested. Damage was begun in the second and third non-Dyfonated plantings in early June, though the third planting incurred little first generation damage. Dyfonate delayed damage in the first and second plantings (Figure 22) such that these plantings had little first generation damage. The resulting damage in the Dyfonated plots (Figure 22) was considerably less than the same aged plantings without the insecticide treatment. Due to the different timed plantings, the CM food resource was continually changing, and as such, a CM incidence curve could not be constructed. However, it appears that the CM completed three generations in radishes at the MSU Muck Farm. The first and second generations attacked the first and second plantings, and a late summer-fall generation was evident in the third and fourth plantings. More SCM were observed infesting radishes at the MSU Muck Farm than in any other plantings of onions or radishes in the other regions. SCM comprised 25% of the radish-infesting maggots sampled in June, 4% in July, and 25% in August. No SCM were detected in the fall generation. Ellis et al. (1979) investigated the comparative attractiveness of different aged radish plants to adult CM for oviposition, and discovered two distinct peaks in attractivity. The plants become more attractive as the hypocotyls swell, and 82 Table 14. Parasitism of OM pupae, MSU Muckfarm, 1980. Julian number Day Plot n parasitized 184 lst planting; non-Dyfonated 21 0 191 lst planting; non-Dyfonated 30 0 192 2nd planting; non-Dyfonated 36 2 199 lst planting; non-Dyfonated 18 l 83 .‘ . P8 . mgr emuc (9% npsr mus (owomrr‘) ‘ 1 0‘ a ' R ,I" i \ r . ‘\ 1 ar’ *- a V T I l’ V ; I = r O 1 SECOND PLAMWG (nouo‘rromrt’) SECOND mus (mount?) ) l .7 '3 u l // J , L 0.1 L: t l c m ‘4 \ 2 3 o ~2- ‘l ‘~ . - 0’ ‘ c l _, e-o ' b- - - o- - _ - _J . v ‘ # r I I ’ j V T Q- 3‘1 :3 [Tunivuwmm(mmowmumfi l“ J C: . , E I c .1 ’3 l .x p . ,' -: I I J 7 'l .’ ’3 : , 5 I g : ’ '3 - ‘ I I O r : ° ‘ ”17W FUNDING (NONOYTONATE') FOURTH PLANTING (DYFONATE') .4 ha a" .8 o.“ l-g -‘ _: °m :6: 2;: in fins '10: - 258' 2‘20 ' :10 an" JULIRN DRY Figure 22. CM damage and infestation in radish plantings at the MSU Muck Farm, 1980. PERCENT INFESTEO PLHNTS 84 are most attractive as they reach marketable size. After this stage, the plant's attractivity decreases, but increases again as the plant produces flowering shoots. Oviposition preference was seen in the MSU Muck Farm plots. CM females were attracted to earlier plantings until the later plantings began to reach marketable size. Infestations then shifted. As the first radish planting was heavily infested before the hypocotyls had begun to swell, it appears that, without a choice CM adults will oviposit in whatever is available. These results indicate that the timing of planting and the use of Dyfonate can influence the timing and levels of CM infestations. Parasitism of CM CM pupal samples collected in the first radish planting at the Muck Farm between June 21 and July 3 were not taken per a measured unit area. Absolute densities over these sample dates were estimated from third instar density data. Third instar densities at 97.5 degree days, base 6.l°C (developmental time for third instar CM larvae) prior to each pupal sampling date were used to construct a hypothetical incidence curve for CM pupae (Figure 23). However, as A. bilineata parasitoid activity began during the later half of the first generation pupation period, and parasitized pupae require considerably more time to complete development than healthy CM pupae, it is unlikely that the pupal population goes to zero between generations, but rather overlaps generations. To reflect this situation and thereby make pupal density estimates more realistic, a line was fitted by connecting the estimated peak pupation with the first sampled pupal density estimate on July 14th. CM pupal densities for the interim sample dates were drawn off this resulting curve. 85 .mofiufimam6 “pumaw 6uwsu scum mowufiwco6 Honda :0 common mauuo mo cowumaaumm .o ..m mwcm. m>¢o mumomo com. com. com. own com a ' 9 b b P 9 ‘4‘ 4.. .v fcmzwo ..(nSm Oman; 4 rtmzwo ._._._mzmo ”3.52. can Quins—(m X .mN muswnm E nzum SlNU'lcl S /AlISN30 86 Seasonal production of A. bilineata parasitized and non-parasitized CM pupae and the resulting proportion parasitized in the infested radish plantings at the MSU Muck Farm are presented in Table 15 and Figures 24 a-f. The first non- Dyfonate planting showed the largest number of parasitized pupae and the largest proportion of parasitized host population. In the second non-Dyfonate planting, parasitoid activity appeared synchronized with the initial incidence of pupation; however, parasitism declined with the first pupal generation and remained low through the second, despite moderate CM pupal density. The earliest Dyfonate planting was also well synchronized with its host's pupation, and resulted in a relatively high proportion parasitized (.23). However, the CM population in this planting was low, and parasitoid activity was not evident until the second generation, resulting in a low seasonal production of parasitized pupae. The later-planted Dyfonate radish plots produced few CM pupae and no detectable parasitism. The third non-Dyfonate planting produced the largest overwintering CM pupal population, yet only 6.7% were parasitized. The fourth planting produced few CM pupae, and no detectable parasitism. The radish plots were small and close together; therefore, CM distribution between the plots may have been influenced by the presence of a preferred resource, the earlier planted radishes. However, given the resulting CM infestations and A. bilineata parasitism in the radish plots at Eaton Rapids and the MSU Muck Farm, it appeared that the timing of infestations and resulting parasitism was influenced by host plant availability, and thus planting date. The larval parasitoid of the CM, Trybliographa rapae, was found in pupal samples collected in August and September in the radish plots at the Muck Farm. Nine percent of the pupae collected in August were parasitized by _'I_'. ra ae, and 87 Table 15. .A. bilineata parasitism of CM pupae in radish plots at the MSU Muck Farm, 1980. Seasonal production/5 plants .A. bilineata parasitized Proportion Planting CM pupae CM pupae Parasitized l nonDyfonate 10.257 2.661 0.259 2 " 7.348 0.832 0.113 3 " 1.908 0.129 0.067 4 " .470 0.0 0.0 1 Dyfonate 2.991 .689 .230 2 " 1.098 0.0 0.0 88 Figure 24 a-f. A, bilineata parasitism of CM pupae in radish plantings at the MSU Muck Farm, 1980. Densities are corrected for differences in rates of development between the host and parasitoid. 89 >¢o 2¢H426 own ova ONN DON oo— . oo— Ova 1 €36,333 0:5:03 vcu .Ia. I/lw.\l IIIL . $365555 0:55.: 5... . cmw cow oo— Aeobeotoeoev 05:63 65 50:06 .69.: T guy—neg e a 33:06 .69.: X Aeoaocoxogev 0:5:63 vcd $365355 95:63 50— SlNUWd S / AlISNBU 90 1096 of those collected in September. Bromand (1980) found a lesser percentage of A. bilineata parasitism in 1. Egg; parasitized fl. 113.11g pupae than in non- parasitized pupae. He concluded that some selection for non-parasitized pupae may exist, but the process of recognition is not understood. Grant Parasitism of OM First generation pupae samples were pooled over fields 1 and 2. The first sample (n=l95), collected on July 3, revealed no A. bilineata and less than 196 fl. m parasitism. Of the pupae collected on the second sample date, July 16 (n=l70), 3.5396 (uncorrected) were parasitized by _l_\_. bilineata and none by A. m. If parasitism occurred, it should have been more detectable at this time, as July 16 (106# degree day, base (4.400 was just past peak pupation, and healthy flies had already begun emerging, whereas parasitized pupae would still be in the soil. Table 16 presents the results of the second generation pupal sampling. More parasitism was observed in the field without foliar sprays, though Whitfield (1981) reported no difference in OM infestations between the fields. These results may be slightly biased, as the Dyfonate field with foliar sprays, field 2, was sampled a week prior to the non-sprayed field 1, and during that time, parasitism could have increased. However, the field with no Dyfonate, but with repeated foliar spray applications, was sampled the same day as field I, and less than 196 parasitism was observed. Five percent of the first generation ptpae sampled were SCM, as were l#96 of the second generation. No parasitism of SCM was observed. 91 Table 16- Parasitism of second generation OM pupae in fields receiving different insecticide treatments. Percent parasitism2 Date Insecticide Use1 n ‘5, bilineata ‘5. pallipes 233 Dyfonate plus 417 4.31 .24 foliar sprays ‘ 240 Foliar sprays only 274 .36 O 240 Dyfonate only 116 19.83 .86 1Dyfonate is a soil granular applied at lanting, foliar sprays consisted of F-lO applications of Sevi or parathion throughout the season. 2Percentages not corrected for differences in developmental times. 92 CM Infestations of Radishes and Parasitism Eighty percent of the sampled radish plants from the plots in Grant showed CM damage by the first sample date, July 2‘4. Seventy-four percent of the sampled plants were infested. The maggot infestation decreased to 596 of the sampled plants on August 13, but increased to M96 by the end of August, and 2‘96 by the end of September. Thus, it appears that there were two generations of CM in the radishes in Grant. An earlier generation was probably missed due to the late planting date. A. bilineata parasitism of CM pupae is presented in Table 17. Parasitism levels were corrected for differences in the developmental times of the host and parasitoid by multiplying the number of sampled parasitized pupae times .53 (the mean ratio of physiological time calculated for different sample dates during July and August at the MSU Muck Farm and Eaton Rapids). (This ratio only varied between .51 and .54 over this time period.) No 1. 32a; parasitism was observed in Grant. SCM accounted for 1296 of the radish-infesting larvae in July 214, and 396 of the pupae on August 13. Otherwise, all other radish-infesting maggots were CM. Stperparasitism The corrected numbers of _A. bilineata parasitized pupae in the different non-Dyfonated radish plots at the MSU Muck Farm were examined in relation to the host densities. A least squares regression revealed a fairly exact linear relationship (r2=.980; Figure 25). This indicates that with increases in host densities, the proportion of pupae parasitized by a single A. bilineata population remains fairly constant. Examining the attack behavior of the first instar parasitoid larvae reveals a possible explanation for this relationship. 93 Table 17. A, bilineata parasitism of CM pupae in Grant radishes. corrected percent percent Date n parasitized parasitized 205 36 25.00 15.01 225 131 75.57 62.14 240 5 40.00 26.11 94 .mowuwmcmv umon ucmuwmwwv um Emfiuwmmumm mummcfiawn .< .mm muswfim whzcmm m mum mamam 5 «w m: m _ _ : x3. + we. r. a SlNUJd 9 83d aaana OEZliISUHHd 95 Bromand (1980) reported that under laboratory conditions, the maximum depth through which a parasitoid larva would burrow in sand to parasitize a pupa was 9.5 cm; in the field, parasitized pupae are not found below 20 cm. He also noted that if two host-pupae are placed in close proximity, one directly above the other, the lower one is seldom parasitized. Even with several parasitoid larvae present, the upper pupa is reattacked. Superparasitism or multiple attacks of the same pupae occurs on both CM (Bromand 1980) and OM (Whitfield 1981), and was prevalent on both hosts in this study. More than a single larva per pupa was much less common. With the inclusion of two years of data from Bromand (1980) on A. bilineata superparasitism of CM in swede fields in west Jutland, Denmark, and the data from radish plantings at the different research sites in this study (for each site data were pooled over all plots and sample dates), a least squares regression was performed to examine the relationship between the proportion of the host population parasitized and the proportion of the parasitized pupae superparasitized. The same analysis was performed in OM parasitism data for the Eaton Rapids research site and three different fields in the Grant data. (All values represented uncorrected numbers of parasitized individuals sampled.) If the proportion of host parasitized is constant at a given A. bilineata density irrespective of host density, we can assume the differences in the proportion of individuals parasitized represent relative differences in the _A, bilineata density. With both hosts, increases in the proportion of pupae parasitized correlated with 2=.7l&; OM r2=.88; Figures 26 and 27). The an increase in superparasitism (CM r regression equation predicted that for CM, the percentage of parasitized pupae superparasitized increased with the percentage of the host population parasitized 96 .mmasa :0 mo Emwuwmmumaumasm mumwcwawa .< .om muswwm 2m_._._m GEZIiISVHVdHBdnS SJVHGINGNI GBZIiISVHVd .LNEIOEBd 97 .66363 So no Emfiuwmmummuoasm oumosfiawn .< .NN madman 2m_._._m 8 x r0 GBZLLISVHVdEJEdflS S‘IVOGIAIGNI GHZliISVHVd 1N3083d 98 (A. bilineata density), and at 10096 parasitism, 9296 of the parasitized pupae would be superparasitized. The relationship between proportion of OM pupae parasitized and percent superparasitism is best described by a curvilinear function, in which at 3196 parasitism 10096 of the parasitized pupae would be superparasitized. Whitfield's findings (1981) agree with this model. He reported 29% A. bilineata parasitism of OM in a research field in Grant and noted that most of the attacked pupae were superparasitized. The vertical distributions of CM (Harris et al. 1967) and OM (Carruthers, 1979) pupae in the soil (Figure 28) differ considerably; more CM larvae pupate within the top 2 inches (5.08 cm.) of soil than do OM larvae. Within this shallow zone, pupae are more often attacked by A. bilineata ' larvae. As a larger proportion of the OM population pupates at a greater depth, a larger proportion is less likely to be attacked, and as A. bilineata populations increase, superpara- sitism of shallower pupae is more likely to occur. Read (1962) suggests that parasitoid larvae prefer unparasitized pupae, and that larvae will search the soil for unparasitized rather than enter a previously attacked puparium. However, his experiments were performed in the laboratory with CM pupae buried at a uniform depth. He also examined superparasitism concerning the number of larva entering the puparium-mot the multiple attacks on a host puparium. As stated previously, few multiply-attacked pupae had more than one parasitoid within the puparium. Hence, once the larvae gnaws through the puparium, it may detect the presence of the previous attacker. Multiple attacks result in reduced parasitoid survival. Bromand (1980) found that only 2296 of the superparasitized pupae produced A. bilineata adults, whereas in 5296 of the singly-attacked pupae parasitoid adults emerged. The om .H6 0 mwuum: 136m 93 u“ wanna 20 6cm 2 ZO:.<.Sn5n_ ._<._.O._. ..5 Emommm ON 0 SOIL DEPTH 6.636.: :oEo >\\\\\7 // \\\\\\\\\\\\\\\/////////////// \\\\\\\\\\///// \\\\\ mouw>fiuom uwm>hmclumom mmmHHHmm .< AEV $38 26:... mozfima 0+ on. 9. .mN munwam 39 (z) wsmsvava 1 9L l 001 118 30 1 20 1 \ x \\ . \‘ \\ PARASITISM (z) \ \ \\‘ \ ‘\ \ s: ‘ ~53 .78 ~\‘ \_\\\ ("\f- t), . \hs \ \\ \\\ C \‘C‘\‘i ‘ :\f\§'\$\v \ \ ~. \\ \ xgbs \ \\ . - \\\ \ \‘. ‘-\\ .\\.\ \3 \t \\ \ \ \€32~__g1 \ \ .\\\\ §:\ ‘\\\\ ~~ng iixfib .‘\ \ ‘ \\\‘ .\\‘ ‘»‘\‘ 0 0 l \ \ . . \AerR5\-A , ‘\\5§§8\§x§>‘§ \. . .\ \ \ .\\\ ‘ / (kCiixfs \ \" w. - f§§§\§§O;F \ \ \\\ ,‘j’ / ’?'/; 'r’l/ 4‘? ;A¢ /r/ ;Z/ /// /';5 -% / ,2 . /, - 4% f’/’ / ‘ /,’r‘// , [/1 '1’?” ,[>6/ 114%- , /’ ,;/2 '1, I“ r I’ " I; ’11/7’ g7 / l/ /. 1/1/3éfcz /¢ 71/ , g/ ,./. .1/ _ V . . \ ‘~ \~\ .‘7 . \ \\\\‘ V \‘ \\ . \ \\\\< If ,‘HZ // x PASTURE 1 2 3 4 5 ONION FIELD BORDERS 1 Borden: active cow pasture 3 Border: wooded area 2 Borden: active cow pasture 4 Borders wooded area 5 Borders adjacent onion field Figure 30. ‘A, pallipes post-harvest activity in different onion field borders in Grant, 1980. 119 DISCUSSION THE INFLUENCE OF PLANTING DATE ON ROOT MAGGOT INFESTATIONS Planting date greatly influenced OM infestations. OM damage decreased significantly with each weeks' delay of planting. Gary (1922) showed that early sown onions are most heavily infested, and later-sown onions are not damaged until the end of the first generation. However, the meaning of early and late planting is not clear from this report, nor if the early planting was in the vicinity of the later. First generation OM oviposition continued in the Muck Farm onion plots through late June. Due to the late planting date, the third and fourth onion plantings missed the major portion of the OM oviposition period. The second planting, however, was at a susceptible stage when OM damage was initiated, yet the flies appeared to prefer the earlier planting. Threherne and Ruhmann (1922) reported that OM are more attracted to the more vigorously growing plants. With the greater leaf tissue of the earlier planted onions, growth was likely more vigorous. Once damage was initiated, as flies were attracted to previously damaged onions, the earlier plants continued to accumulate most of the egg load. This pattern continued through the second generation. As the plants were susceptible, without the earlier planting the second onion planting probably would have experienced a heavier OM infestation. It cannot be concluded that delayed planting alone (unless delayed through most of the OM oviposition period) will reduce OM damage. Yet early-planted onions may be extremely valuable as a trap crop. 120 The research plots used for this study were very small relative to a commercial onion field. As any point attractant for the adult fly is limited by the range of fly dispersal, before early sown onions can be used as a trap crop for OM, more research must be focused on the scale where this method would be effective. CM infestations in radish plantings also varied with planting date, reflect- ing a known preference for older plants (Ellis et al. 1979). Yet, it can be seen from the heavy infestations of CM in late sown radishes in both Eaton Rapids and Grant that delayed planting alone does not reduce infestations. Planting date, however, does influence the timing of CM infestations, and hence its availability to parasitoids. A. BILINEATA SYNCHRONY WITH ITS HOSTS Aleochara adults emerged about mid to late June. With a preoviposition period of two to three days (Colhoun 1953) and five days to egg hatch, parasitizing larvae usually are not active until late June and early July. Peak pupation of first generation CM in the earliest radish planting occurred about June 23 (degree day, base 6.10 C). A. bilineata parasitism was not detected in this plot until June 27. Parasitoid activity appeared to be better synchronized with pupation of CM in the second radish planting. The seasonal proportion of individuals parasitized and total production of parasitoids was greater in the first than second planting. Yet without the influence of the earlier planting, infestation levels and para- sitism may have been greater in the later, as they were in Eaton Rapids and Grant. 12.1. When examining the synchrony of parasitoid activity with the CM popula- tion, it is important to address the temporal distribution of the target host, the OM (Figure 31). The OM with peak pupation in early July is well synchronized with Aleochara parasitoid activity. Emerging adult Aleochara seek a food source in the form of dipteran eggs and larvae and randomly deposit their eggs in the process. As the OM pupate, the food source declines. The timing of CM infestations at the Muck Farm indicate that second generation oviposition activity in radishes begins about this time. The temporal distributions of these organisms is dynamic as development is tied to temperature. Yet in 1980, second and third generation CM oviposition in radishes roughly synchronized with first and second generation OM pupation (Figure 32). Inte'rplanting radishes with onions may increase the food supply for adult Aleochara and hence increase the parasitoid abundance in onions. Aleochara females are long lived (average 45 to 50 days in the laboratory) (Calhoun 1953, Bromand 1980) and produce eggs during their entire life. Overlapping generations of Aleochara most likely occur, and dependent on host availability more than on the parasitoid availability. With CM pupation occurring between that of the OM (Figure 33), a continuous supply of hosts would be available to parasitizing larvae, resulting in a greater production of Aleochara in the system. General predator populations were greater within the radish plantings than the onions. This may reflect a more favorable microhabitat afforded by the radishes. Temperatures were cooler and fluctuated less within the radish canopy than within the onions (Figure 34). The broad leaves and growth habit of the radish provides greater protection from solar radiation and maintains higher 122 1-00 0.12 J c 1 \: \ 0.06 PERCENT PUPHE PER DRY OF TOTRL ( _. PROPORTION SERSONRL PRRRSITISN O °ieo 180 2700 250 210 280:; JULIRN DRY ii! Onion maggot 0 Cabbage maggot . Kunkel'e onions Muckfarm—1 radish planting A Cabbage mag at 0 Cabbage ma got . Kunkel'e radls es Muckfarm-Z radish planting Figure 31. The rate of A. bilineata parasitism of CM and 0M in reference to the seasonal abundance of 0M pupae. 123 .aoaumasa 0mm 20 Cu m>fiumaou mcnmfiwmu co huw>fiuom Hm>umH 6am wwc 20 Hmaommmm .Nm musmfim % fl 0 z < H A D h 00H ovH OON omH ovm CNN owm 20 ZOHBfimmZmD BmmHh ZO ZOHB<¢WZMO 0200mm ZOHfiémDm Mdmm ZOHB_i_lineata Gyll., a staphylinid parasite of Chortophila brassicae Bouche. J. Econ. Biol. 10:1- Weseloh, R.M. 1972. Spatial distribution of the gypsy moth (Lepitoptera: Lymantridae) and some of its parasitoids. Entomophaga. 17:339-51. Weseloh, R.M. 1979. Comparative behavioral responses of three Brachymeria species and other gypsy moth parasitoids to humidity and temperature. Environ. Entomol. 8:670-5. Wharton, R.A. 1977. New world Aphaereta Species (Hymenoptera: Braconidae: Alysiinae), with a discussion of terminology used in tribe Alysiini. Ann. Entomol. Soc. Amer. 70(5):782-803. Whitfield, G.H. 1981. Spatial and temporal population analysis of the onion maggot (mlemya anmua (Meigen)), in Michigan. Ph.D. thesis, Mich. St. Univ., E. Lansing. 379 pp. Wishart, G. 1957. Surveys of parasites of Hylemya spp. (Diptera: Anthomyiidae) that attack cruciferous crops in Canada. Can. Entomol. 89:450-54. Wishart, G., J.F. Doane and G.E. Maybee. 1956. Notes on beetles vs. predators of eggs of Hylemya brassicae (Bouche) (Diptera: Anthomyiidae). Can. Entomol. 88:634-39. Wishart, G. and E. Monteith. 1954. Trybliographa rapae (Westw.) (Hymenoptera: Cynipidae), a parasite of Hylemya spp. (Diptera: Anthomyiidae). Can. Entomol. 86:145-54. Workman, R.H., Jr. 1958. The biology of the onion maggot, Hylemya antiqua (Meigen), under field and greenhouse conditions. Ph.D. thesis, Oregon St. Call. 82 pp. APPENDIX A A- BILINEATA REARING 144 A colony of A. bilineata was maintained in a covered 35 x 22 x 7 cm plastic box. Small holes were punctured in the bottom and a portion of the cover was replaced with a fine mesh nylon screen to facilitate drainage and air flow. The box was filled to a depth of ca. 3 cm. with a mixture of moist sand and vermiculite. Adults were supplied with maggots on cut onions, which provided cover for the insects. Healthy OM pupae were buried in the sand, exposing them to A. bilineata larvae for in 48 hours. After this time, the pupae were removed and kept in a separate petri dish with moist vermiculite. When adults emerged, they were returned to the colony. Pupae left in the colony for much more than 48 hours usually suffered repeated attacks from parasitoid larvae resulting in reduced survival. With adequate food and moisture, A. bilineata were fairly long lived and easy to maintain. Bromand (1980) describes a more efficient method for mass rearing A. bilineata. APPENDIX B REARING A- PALLIPES 145 Field-collected A. pallipes emerged in rearing dishes consisting of a 20 cm glass lantern globe in a dish of moist sand. The top of the globe was covered with a fine cotton muslim cloth through which distilled water and sucrose solutions were fed. Sufficient moisture in the sand assured parasitoid survival. The sex ratio of emerging A. pallipes was 2 males to 5.5 females; an average of 8.3 parasitoids emerged from a single OM pupa. From a small percent of the field-collected pupae, only males emerged, indicating that the female had not mated. The size of the parasitoids depended on the number of parasitoids per pupa and probably the age of the onion when attacked. The average length was about 2 mm. Males were usually smaller than females and could be easily distinguished as the females ovipositor was visible without magnification. Mating was observed almost immediately on emergence, but the preoviposition period averaged 2.25 days. OM larvae on slices of onion were placed in the bottom of the rearing dish for exposure to A. pallipes adults. The parasitoids became excited with the introduction of the onion, and frequently mature females would approach within a few minutes. Females crawled between the leaves of the onion bulb in search of OM, often becoming matted with onion juice, resulting in their mortality. A. pallipes oviposited in both second and third instars of OM. OM pupae were sifted from the moist sand. Parasitized pupae were usually smaller than healthy pupae and frequently misshapen. The developing parasitoids were easily recognized through the fly puparium under a 40X magnification dissecting microscope. By the time the host pupated, the parasitoid larvae had hatched and occupied most of the space in the puparium. A detailed description 146 of the parasitoid life stages is given by Salkeld (1959). Though never quantified, it was observed that the oviposition process of A. pallipes frequently killed OM larvae. Dead larvae were often found in onions exposed to the parasitoids. Mortality due to A. pallipes oviposition on the face fly, Musca autumalis was reported (Thomas 1967). This phenomenon should receive further investigation. This mortality could result from multiple attacks of a single OM larvae in a caged situation which may not be as frequent in nature. APPENDIX C PRECISION OF OM DAMAGE SAMPLING AT EATON RAPIDS RESEARCH SITE 147 The precision of the different strategies for sampling within-field OM damage at the organic farm in Eaton Rapids was calculated to evaluate whether changes in observed plant damage actually reflected trends in the OM population or changes in the sampling strategies. Using the field level variance-to-mean relationships estimated by Carruthers (1979) for sample unit sizes of 100 ft (30 m) and 35 ft (10.5 m), and solving for a, precision, in Karandinos' (1976) equation for calculating optimal sample size, levels of precision can be estimated. a = 3. 4s2 - 2 x n where a : precision, $2 2 sample variance, and i 2 sample mean. With the relationship between the sample mean and variance determined as S2 = ax b and substituting, The precision levels for the first two sampling stategies for OM damage were calculated with x equal to the mean percent damage over the time period that sample unit size was utilized. The results are presented in Table C1 in comparision with Carruthers' (I979) recommended strategy of twenty 200 ft2 samples per field. The initial sampling technique of 10-100 ft2 samples, is, as one would expect, not as precise as Carruthers' recommended twenty samples of 100 ft of row, at the densities sampled. However, the stratified sampling technique, using 148 Table Clo Precision of onion maggot damage sampling. Sample Unit Size Precisionz (ft. of rows) n a1 b1 (expressed as proportion of x) 100 20 4.48 1.35 .5176 100 10 4.48 1.35 .7320 25 35 3.78 1.29 .3228 1From Carruthers, 1979 2Calculated with x equal to the mean percent damage observed over the time period that the sample unit size was utilized. 149 35 samples of 25 ft of row was more precise. A comparative precision level cannot be calculated for the third sampling technique which was a combination of distance and unit sampling. However, with confidence in the precision of the stratified unit sampling, we can compare the percent damage levels calculated from data called on the same date (July 18) using both stratified unit sampling, and the combination distance and unit sampling. With stratified unit sample, or damage a July 18 is estimated at 2.74 5 .80%, and with combination distance unit sampling, 1.87 5 1.78. Given knowledge of the distribution of OM damage (Carruthers 1979, Whitfield 1981), the combination distance-unit sampling could underestimate damagel, yet when comparing the results of the two different methodologies on one sample date this did not occur. lAs OM distribution about a field is highly clumped (Carruthers 1979), this sampling strategy of walking distance of row until encountering a damaged plant and then sampling .3 m of row beyond that point, may underestimate damage if the clump continues beyond the .3 m sampled. A more accurate methodology would entail measuring the size of the clump or increasing the sampled unit to a size that would most likely encompass the largest clump encountered. APPENDIX D Coenosia gg' rina, PREDATOR OF THE ADULT ONION MAGGOT 150 APPENDIX D Coenosia tigrina, predator of the adult onion maggot Adults of the tiger fly, 9. ti rina, have been reported as important predators of the OM fly (Perron and LaFrance 1952, Perron et al. 1956, LeRoux et al. 1960) and the SCM fly (Miles 1948, Miller and McClanahan 1960) in Canada and Europe. The tiger flies have also been noted as an important predator of muscoid flies associated with fresh dung (Thomas 1967). Hobby (1931, 1934) published a list of prey species reported for _C. M in Europe, which includes many anthomyiids. All life stages of the tiger fly have been found inoonion fields in Canada. 9. M lays its eggs loosely on or beneath the soil surface (LeRoux and Perron 1960). This species completes only one larval instar from egg hatch to pupation. Mature larvae overwinter in compost soils and pupate in the early spring (LeRoux and Perron 1960). Adults have been reported in fields in southwestern Ontario from early May to late October, their period of activity roughly coinciding with that of the OM (LeRoux and Perron 1960). Detailed descriptions of the immature life stages and adult 9. M are given by LeRoux and Perron (1960) and Perron et al. (1956). Tiger fly predation on adult onion maggots has been observed in caged situations (LeRoux et al. 1960), and C. M adults reduced the subsequent prey generation by 50%. To date, further field and laboratory investigations have been curtailed by a lack of information on the food habits of Q. grin—a larvae. Attempts to rear the larvae to maturity on various vegetable and animal diets were unsuccessful (Perron and LeRoux 1956). Recently, however, Q. tigrina 151 larvae were observed in the field feeding on the earthworm, Eisenia rosea (Yahnke and George 1972). Subsequent sampling supported this original observation and the larvae were successfully reared to pupation on earthworms in the laboratory. Yahnke and George (1972) observed the survival of Q. m larvae reared on E. ggsga in the laboratory to be significantly greater on bisected prey than live intact worms. However, these authors found that with the presence of the cluster fly, 3. Lu_d_ig, also a predator of _E. 51532;, C. t_ig£i__rla_1_ survival on live worms increased dramatically. The association of E. w and Q. _t_igri_n§ under field conditions is not known. During preliminary sampling of commercial onion fields in the Great Swamp area this past year, g. M were found in very low levels. However, these flies were relatively abundant (densities approaching that of the adult onion maggot) in monitored fields where no chemical pesticides were used (Carruthers 1979). It is known that earthworms are extremely susceptible to fungicide poisoning. In apple orchards, earthworms have been reduced by 90% following two years of standard fungicide spray programs (Stringer 1974). As earthworms are an important food source for immature 9. gm, fungicide applications in commercial onion fields may be reducing earthworm populations to levels unable to support high predator populations. Hobby, B. M. 1931. The prey of Coenosia tigrina F. Proc. Entomol. Soc. London 6:13-15. Hobby, B. M. 1934. Prey of Coenosia tigrina F. (Diptera: Anthomyiidae). J. Entomol. Soc. 5. Engl. 1:69-77. LeRoux, E. J., and J. P. Perron. I960. Descriptions of immature stages of Coenosia tigrina (F.) (Diptera: Anthomyiidae). With notes on hibernation of larvae and predation by adults. Can. Ent. 93:264-96. 152 Miles, M. 1948. Field observations on the bean seed fly (seed corn maggot), Chortophila cilicrura Rond., and Q. trichodactyla Rond. Bull. Ent. Res. 38:559-741. Miller, L. A., and R. J. McClanahan. 1960. Life history of the seed corn maggot, Hylemya cilicrura Rond. and fl. liturata (Mg.) (Diptera: Anthomyiidae) in southwestern Ontario. Can. Ent. 42:210-21. Perron, J. P., and J. LaFrance. 1952. A note on a Dipterous predator of the onion maggot, Hjlemya antiqua (Mg.). Can. Ent. 84:112. Perron, J. P., J. LeRoux, and J. LaFrance. 1956. Notes on Coenosia tigrina (F.) (Diptera: Anthomyiidae) mainly on habits and rearing. Can. Ent. 88:608-11. Stringer, A., and C. H. Lyons. 1974. The effect of benomyl and thiophanate- methyl on earthworm populations in apple orchards. Pestic. Sci. 5:189-96. Thomas, G. D. 1967. Natural enemies of the face fly, Musca autumnalis DeGeer, in Missouri. Ph.D. thesis, Univ. of Missouri, Columbia. 131 pp. Yahnke, W. E., and J. A. George. 1972. Earthworms as prey for larvae of Coenosia tigrina. J. Econ. Entom. 65:1478-79.