Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. Reproductive behaviour of Aphidius ervi Haliday (Hymenoptera: Aphidiidae) a thesis presented in partial fulfiment of the requirements for the degree of Doctor of Philosophy in Plant Science (Entomology) at Massey University Palmerston North New Zealand Xiong Zhao He 2008 i Abstract Aphidius ervi Haliday is a cosmopolitan parasitoid species of several major aphid pests on economically important crops. Prior to this research, little information was available on its reproductive behaviour. Emergence of A. ervi peaks during the first few hours of the photophase with males being protandrous. Females become sexually mature earlier than males and oviposit primarily in the photophase. Aphids parasitised in their early instars die before reproduction but those parasitised in later instars produce a limited number of progeny. Females pr efer aphids of 3- to 5-d-old over the younger and older aphids for oviposition. Female s ovipositing in 4- to 7-d-old aphids have more fitness gains in terms of progeny body size and egg load at emergence. Fertilised eggs are more likely deposited in large hosts and unfertilised eggs in small ones. Large individuals have greater longevity, large males father more progeny, and large females have higher fecundity, parasi tism and greater ability in host searching. However, with increasing body size females gain more than males in longevity and fecundity but males gain more than females in the number of female progeny. Males can inseminate up to nine females and they carry about 82% effective sperm at emergence and replenish about 18% sperm dur ing their adult life. Females adjust the oviposition and sex allocation strategies in response to increasing host density with higher number of aphids parasitised at higher host densities and lower proportion of female progeny produced at lower host densities. Males play an active role in mating behaviour. Males having mati ng experience, and being la rge or younger, respond to females more quickly and perform better cour tships resulting in higher mating success. Males prefer larger and younger females for mating probably because the latter have greater reproductive potential. Males optim ize the use of their sperm based on the availability of their sperm and the reproductive status (age) of females. The switching- off of female receptivity of male mating attempt after the mating is a gradual process. Some females accept the second males within 1 minute since the termination of the first mating. The shorter mating period in the second mating suggests that females remate probably due to the gradual process of switching-off of female receptivity rather than the insufficient sperm transformation during the first mating. Males prolong their mating duration in male-bia sed operational sex ra tio to reduce the probability of female remating. ii Acknowledgements I am extremely grateful to my supervisor, Professor Qiao Wang, for his invaluable time, effort and support provided throughout this research. Qiao always fed me with more scientific questions, challe nged my explanations , encouraged me to publish and present my results, and had an open-door policy with kind smile on his face. My gratitudes are also towards my co-supervisor, Dr. David A. J. Teulon (Crop and Food Research, Christchurch, New Zealand), for his time in discussing the results and making valuable comments on the thesis and manuscript papers. I am also thankful for his financial support in the earlier stage of this project. I appreciated very much Drs. G. P. Walker and N. Martin (Crop and Food Research, Auckland) for their valuable advi ce and comments on an early draft of the project. I am thankful for the excellent academic and tec hnical support from Ms.L. K. Davis and staff of INR, Masse y University. I also thank th e staff of the Plant Growth Unit, Massey University, for providing me with materials and their technical assistance. I am grateful to Drs. D. Hedderle y, A. Noble and A. Jiménez-Pérez for statistical assistance and advice throughout the research, and to the staff of Massey University Library and the IT service (INR) for their help with information access. I am very thankful to my family, esp ecially my wife Dong Xiao Feng and my two daughters Joanna and Ivy for their un conditional support during my studies. My wife always does her best to take care of our daughters and our family so that I can concentrate on my studies. I am also grateful to my brother and sister for taking care of our parents after we migrated to New Zealand. This thesis is dedicated to my loving parents who showed me the way to succeed in my life iii Table of Contents CHAPTER 1 GENERAL INTRODUCTION 1 1.1 Introduction……………………………………………….... … ……1 1.2 Important of the Study of Parasitoid Reproductive Behaviour………………………………………………… … … ……3 1.3 Relevance of This Research………………………… … … … ……..4 1.4 Aim and Objectives…………………….…………………... … ……6 CHAPTER 2 LITERATURE REVIEW 7 2.1 Introduction………………………… … … … … … … … … … … ……7 2.2 Taxonomy of A. ervi……… ………………………………… … ……7 2.3 Identification of A. ervi……… ……………………………… … ……8 2.4 Distribution of A. ervi……… ……………………………….. … ……9 2.5 Hosts and Host Preference of A. ervi……… ……………… … … ……9 2.6 General Biology of A. ervi……… ………………………… … ……10 2.6.1 Eggs………………………………………………………… ……10 2.6.2 Larvae………………………………………………………. ……10 2.6.3 Pupae……………………………………………………….. ……15 2.6.4 Adults………………………………………………………. ……15 2.6.5 Host Searching and Oviposition of A. ervi……… ………….. ……15 2.7 Reproductive Biology of Aphidiinae………………………. ……18 2.7.1 Reproductive Systems…………………................................ ……18 2.7.2 Mating Behaviour………………………………………….. .……20 2.7.3 Sex Allocation and Sex Ratio……………………………… ……20 2.8 Factors Affecting Reproductive Fitness of Parasitoids…….. ……21 2.8.1 Host Quality………………………………………………... ……21 2.8.2 Host Density………………………………………………... ……22 2.8.3 Parasitoid Body Size……………………… … … … … … ….. ……23 2.8.4 Mating History of Parasitoids……………………………… ……23 2.9 Sexual Selection………………………… … … … … … … …. ……24 iv CHAPTER 3 GENERAL BIOLOGY OF APHIDIUS ERVI 27 3.1 Introduction………………………………………………… ……25 3.2 General Methodology………………………… … … … … … . ……25 3.2.1 Materials……………………………………………………. ……25 3.2.2 Procedures………………………… … … … … … … … … ….. ……27 3.2.3 Environmental Conditions…………………… … … … … …. ……28 3.2.4 Definitions………………………… … … … … … … … … ….. ……28 3.2.5 Statistical Analysis and Reported Values………………….. ……28 3.3 Emergence, Sexual Maturation and Oviposition of Aphidius ervi 29 3.3.1 Introduction………………………………………………… ……29 3.3.2 Materials and Methods……………………………………... ……30 3.3.2.1 Experimental Parasitoids and Hosts………………………... ……30 3.3.2.2 Emergence……………………… … … … … … … … … … ….. ……30 3.3.2.3 Sexual Maturation………………………………………….. ……30 3.3.2.4 Oviposition…………………………………………………. ……31 3.3.2.5 Statistical Analysis……………………… … … … … … … …. ……31 3.3.3 Results……………………………………………………… ……32 3.3.3.1 Emergence……………………… … … … … … … … … … ….. ……32 3.3.3.2 Sexual Maturation………………………………………….. ……33 3.3.3.3 Oviposition…………………………………………………. ……34 3.3.4 Discussion………………………………………………….. ……35 3.4 Effect of Parasitism by A. ervi on Development and Reproduction of the Pea Aphid 37 3.4.1 Introduction………………………………………………… ……37 3.4.2 Materials and Methods……………………………………... ……37 3.4.2.1 Development and Reproduction……………………… … …. ……37 3.4.2.2 Population Growth………………………… … … … … … …. ……38 3.4.2.3 Statistical Analysis……………………… … … … … … … …. ……38 3.4.3 Results……………………………………………………… ……39 v 3.4.3.1 Development and Survival……………………… … … … …. ……39 3.4.3.2 Reproduction……………………………………………….. ……39 3.4.3.3 Population Growth………………………… … … … … … …. ……41 3.4.4 Discussion………………………………………………….. ……42 CHAPTER 4 FACTORS AFFECTING REPRODUCTIVE FITNESS OF APHIDIUS ERVI 44 4.1 Introduction………………………… … … … … … … … … … ……44 4.2 Host Age Preference and its Effect on Reproductive Fitness of A. ervi 44 4.2.1 Introduction………………………………………………… ……44 4.2.2 Materials and Methods……………………………………... ……45 4.2.2.1 Experimental Insects……………………… … … … … … ….. ……45 4.2.2.2 Preference Behaviour………………………… … … … … …. ……45 4.2.2.3 Effect of Age Preference on Reproductive Fitness………… ……46 4.2.2.4 Statistical Analysis……………………… … … … … … … …. ……47 4.2.3 Results……………………………………………………… ……48 4.2.3.1 Preference Behaviour………………………… … … … … …. ……48 4.2.3.2 Effect of Age Preference on Reproductive Fitness………… ……51 4.2.4 Discussion………………………………………………….. ……53 4.3 Effect of Host Density on Reproductive Fitness of A. ervi 56 4.3.1 Introduction………………………………………………… ……56 4.3.2 Materials and Methods……………………………………... ……56 4.3.2.1 Experimental Insects……………………… … … … … … ….. ……56 4.3.2.2 Experiment……………………… … … … … … … … … … …. ……56 4.3.2.3 Statistical Analysis……………………… … … … … … … …. ……57 4.3.3 Results……………………………………………………… ……58 4.3.4 Discussion………………………………………………….. ……60 4.4 Effect of Body Size on Reproductive Fitness of A. ervi 62 4.4. 1 Introduction………………………………………………… ……62 vi 4.4.2 Materials and Methods……………………………………... ……64 4.4.2.1 Measurement of Body Size…………………… … … … … … ……64 4.4.2.2 Experiments……………………… … … … … … … … … … … … … 6 4 4.4.2.3 Statistical Analysis……………………… … … … … … … … ……65 4.4.3 Results……………………………………………………… ……66 4.4.3.1 Male Reproductive Performan ce in Relation to Body Size... ……66 4.4.3.2 Female Reproductive Performance in Relation to Body Siz e ……66 4.4.3.3 Relative Fitness Gain of Male s and Females in Relation to Body Size…………………………………………………... ……68 4.4.4 Discussion………………………………………………….. ……70 4.5 Effect of Mate Age at Mating and Male Mating History on Reproduction of A. ervi 73 4.5.1 Introduction………………………………………………… ……73 4.5.2 Materials and Methods……………………………………... ……75 4.5.2.1 Experimental Insects……………………… … … … … … ….. ……75 4.5.2.3 Experiment……………………… … … … … … … … … … …. ……75 4.5.2.4 Statistical Analysis……………………… … … … … … … …. ……76 4.5.3 Results……………………………………………………… ……77 4.5.4 Discussion………………………………………………….. ……80 CHAPTER 5 MATING BEHAVIOUR AND SEXUAL SELECTION OF APHIDIUS ERVI 83 5.1 Introduction………………………………………………… ……83 5.2 General Methods………………………… … … … … … … … ……83 5.2.1 Data Recording………………………… … … … … … … ….. .……83 5.2.2 Behaviour Definition……...…………………… … … … …... ……84 5.3 Mating Behaviour and Effect of Male Mating History on Mating Success in A. ervi 85 5.3.1 Introduction……………………………….………………... ……85 5.3.2 Materials and Methods……………………………………... ……86 5.3.3 Statistical Analysis……………………… … …..…………... ……86 5.3.4 Results………………………………………….…………... ……86 vii 5.3.4.1 General Mating Behaviour………………………… …......... ……86 5.3.4.2 Mating Success in Mated and Virgin Males……………….. ……90 5.3.4.3 Male-Male Interaction………………………… … … … … … … … 9 0 5.3.5 Discussion…………………………………………….......... .…...91 5.4 Mate Choice of A. ervi in Relation to Body Size and Age 94 5.4.1 Introduction……………………………….………………... ……94 5.4.2 Materials and Methods……………………………………... ……94 5.4.2.1 Influence of Body Size on Mate Choice...……………...….. ……94 5.4.2.2 Influence of Mate Age on Mate Choice…..………………... ……95 5.4.3 Results………………………………………………….…... ……96 5.4.3.1 Influence of Body Size on Mate Choice...…………………. ……96 5.4.3.2 Influence of Mate Age on Mate Choice……………………. ……98 5.4.4 Discussion…………………………………...……………... …..101 5.5 Operational Sex Ratio and Density Influence Partial Local Mating Behaviour in A. ervi 105 5.5.1 Introduction………………….……………………………... …..105 5.5.2 Materials and Methods……………………………………... …..106 5.5.2.1 Experimental Insects……………………… …...…………... …..106 5.5.2.2 Experimental Design……………………...………………... …..106 5.5.2.3 Statistical Analysis……………..…………………… … …... …..107 5.5.3 Results……………………………….……………………... …..108 5.5.3.1 Homosexual Behaviour…………….……..………………... …..108 5.5.3.2 Competition for Mate…………………… …..……………... …..111 5.5.3.3 Mating Success…………………………………………….. .…..111 5.5.3.4 Mating Period………………………..……………………... …..114 5.5.3.5 Female Remating……………………...…………………... …..116 5.5.4 Discussion………………………………………………….. …..117 CHAPTER 6 GENERAL DISCUSSION AND CONCLUSION 120 6.1 Introduction………………………………………………… …..120 6.2 Reproductive Rhythms of A. ervi……… …………………… …..120 viii 6.3 Reproduction of A. ervi in Relation to Host Age an d density…………………………….………………………... …..121 6.4 Factors Affecting Mate Choi ce and Reproducrive Fitness o f A. ervi……… ……………………………………………….. …..122 6.5 Mating Behaviour of A. ervi……… ...………………………. …..124 6.6 Conclusion…………………………………………………. . …..126 REFERENCES 127 ix APPENDIX I: Published Papers From PhD Study 1 He X.Z. and Wang Q. 2008. Reproductive strategies of Aphidius ervi Haliday (Hymenoptera: Aphidiidae ). Biological Control 45: 281-28 7. 159 2 He X.Z. and Wang Q. 2006. Asymmetr ic size effect of sexes on reproductive fitness in an aphid parasitoid Aphidius ervi (Hymenoptera: Aphidiidae). Bi ological Control 36: 293-298. 161 3 He X.Z. and Wang Q. 2006. Host age preference in Aphidius ervi (Hymenoptera: Aphidiidae). New Zealand Plant Protection 59: 190- 194. 172 4 He X.Z., Teulon D.A.J. and Wang Q. 2006. Oviposition strategy of Aphidius ervi (Hymenoptera: Aphidiidae) in response to host density. New Zealand Plant Protection 59: 195-201. 179 5 He X.Z., Wang Q. and Teulon D.A.J. 2005. The effect of parasitism by Aphidius ervi on development and reproduction of the pea aphid, Acyrthosiphon pisum. New Zealand Plant Protection 58: 202-207. 184 6 He X.Z., Wang Q. and Teulon D.A.J. 2005. Host stage preference and reproductive fitness of Aphidius eadyi (Hymenoptera: Aphidiidae) on Acyrthosiphon pisum (Hemiptera: Aphididae) . New Zealand Journal of Agricultural Research 48: 157-163. 190 7 He X.Z., Wang Q. and Teulon D.A.J. 2004. Emergence, sexual maturation and oviposition of Aphidius ervi ( Hymenoptera: Aphidiidae). New Zealand Plant Protection 57: 214-220. 197 8 He X.Z., Wang Q. and Teulon D.A.J. 2003. Effect of parasitism by Aphidius eadyi (Hymenoptera: Aphidiidae) on reproduction of pea aphid, Acyrthosiphon pisum (Hemiptera: Aphididae). New Zealand Plant Protection 56: 185-189. 204 x APPENDIX II: Conference Presentations From PhD Study 1 He X.Z. and Wang Q. 2008. Operat ional sex ratio and population density influence partial local mating behaviour in Aphidius ervi (Hemenoptera Aphidiidae). Austra lia and New Zealand Biocontrol Conference, February, Sydney, Australia. 209 2 He X.Z. and Wang Q. 2006. Mate age at mating and male mating history affect mate choice and reproduction in Aphidius ervi Haliday (Hymenoptera: Aphidiidae). Australian and New Zealand Entomological Societies Conference, September, University of Adelaide, South Australia. 210 3 He X.Z. and Wang Q. 2006. Is large important in reproduction? MacDiarmid Young Scientists of the Year Award, July, Auckland, New Zealand. 211 4 He X.Z. and Wang Q. 2005. Reproductive response of Aphidius ervi (Hymenoptera: Aphidiidae) to pea aphid density. The Fifth Asia- Pacific Congress of Entomol ogy, October, Jeju, Korea. 213 5 He X.Z., Wang Q. and Teulon D.A.J. 2004. Effect of Body Size on Reproductive Fitness in Aphidius ervi (Hymenoptera: Aphidiidae). XXII International Congress of Entomology, August, Brisbane, Australia. 213 6 He X.Z., Wang Q. and Teulon D.A.J. 2003. Effect of aphid life stage and parasitoid adult age on parasitism and sex ratio of Aphidius eadyi (Hymenoptera: Aphidiiae). XV In ternational Plant Protection Congress, July, Beijing, China. 214 7 He X.Z., Wang Q. and Teulon D.A. J. 2003. Host stage preference by Aphidius eadyi (Hymenoptera: Aphidiid ae) and its effect on reproductive fitness. The 52 nd Annual Entomology Society of New Zealand, April, Wellington, New Zealand. 215 xi List of Tables Table 3.1 Effect of parasitism by A. ervi on survival (days), reproductive period (days) an d number of progeny of pea aphid 39 Table 3.2 Effect of parasitism by A. ervi on life table parameters of pea aphid 41 Table 4.1 Oviposition behaviour of A. ervi females in response to host ages 50 Table 4.2 Mean (± SE) head width (mm) of newly emerged A. ervi when parasitised hosts of different ages 50 Table 4.3 Effect of body size (head width, mm) on longevity (days) in A. ervi 68 Table 5.1 The mean (± SE) period (seconds) of A. ervi males performing mating behaviours 90 Table 5.2 The mean (± SE) period (seconds) of homosexual and competition behaviour performed by A. ervi males 91 Table 5.3 The number of small and large males (selectees) preformed mating behaviour in response to females (selectors) in A. ervi 97 Table 5.4 The mean (± SE) period (seconds) of mating behaviours performed by males (selectees) to the females (selectors) in A. ervi 97 Table 5.5 The number of males (sel ectors) preformed mating behaviour in response to small and large females (selectees) in A. ervi 98 Table 5.6 The mean (± SE) period (seconds) of mating behaviours performed by males (selectors) to the females (selectees) in A. ervi 98 Table 5.7 Mating behaviour of A. ervi males in the presence of females of different ages 99 Table 5.8 Mean ( ± SE) mating period (seconds) of males when they mated with females of different ages 100 Table 5.9 Number of males of different ag es that courted, mounted and mated with 1-d-old females 100 xii Table 5.10 The mean (± SE) period (seconds) of males of different ages performing mating behaviours to 1-d-old females in A. ervi 101 Table 5.11 Effect of OSR and population density on mating behaviour of A. ervi 109 Table 5.12 The proportion of mated females mounted by other males and females remated in A. ervi in various OSRs 116 Table 5.13 The mean ( ± SE) mating period (seconds) of A. ervi females in the first and second matings 117 xiii List of Figures Figure 2.1 Lateral view of tergum 1 of female: (A) A. eadyi (from Starý et al. 1980) and (B) A. ervi (from Kavallieratos et al. 2001) 8 Figure 2.2 Development of A. ervi eggs: mature eggs in ovaries (A), and eggs 1 h (B), 12 h (C), 24 h (D), 48 h (E) and 72 h (F) after oviposition. All pictures were taken in the present study 11 Figure 2.3 Development of A. ervi larvae: newly hatched larva (3 days after oviposition) (A), and larvae of 1 day old (B), 2 day old (C), 3 day old (D), 4 day old (E) and 5 day old (F). All pictures were taken in the present study 12 Figure 2.4 Pupation of A. ervi : mature larvae [9 days (A) and 10 days (B) after ovipositi on] and pupae [11 days (C) and 12 days (D) after oviposition] . All pictures were taken in the present study 13 Figure 2.5 A. ervi pre-adult (A) (13-14 days after oviposition), aphid mummy (B) and adults (C). All pictures were taken in the present study 14 Figure 2.6 Female reproductive system in A. ervi. This picture was taken in the present study 19 Figure 2.7 Male reproductive system in Nasonia vitripennis (Walker) (Hymenoptera: Pteromalidae) (from Sanger and King 1971) 19 Figure 3.1 Breeding cage used for rearing A. ervi and pea aphid 26 Figure 3.2 E xperimental cylinder 26 Figure 3.3 The percentage of female or male A. ervi offspring emerging throughout photophase or scotophase 32 Figure 3.4 The number of A. ervi males courting or mating at 0, 2, 4, 6, 8, 10 or 12 h after emergence. Within the same category (Courting or Mating) columns with the same letters are not significantly different (P > 0.05) 33 Figure 3.5 The number of A. ervi females responding to male courting or mating at 0, 2, 4, 6, 8, 10 or 12 h after emergence. Within the same category (Response to male courting or Mating) columns with the same letters are not significantly different (P > 0.05) 34 xiv Figure 3.6 The number of parasitisms and eggs laid by A. ervi females throughout photophase and scotophase. Within the same category (Photophase or Scotopha se), means (± SE) followed by the same letters within each line are not significantly different (P > 0.05) 35 Figure 3.7 Relationship between host age at parasitisation and reproductive period 40 Figure 3.8 Relationship between host age at parasitisation and reproduction of progeny 40 Figure 3.9 Relationship between host age at parasitisation and r m 42 Figure 4.1 Mean (± SE) number of aphids parasitised per A. ervi female in relation to host age. Columns with the same letters are not significantly different (P > 0.05) 51 Figure 4.2 Mean (± SE) number of eggs laid per A. ervi female (estimated from dissection) in relation to host age. Columns with the same letters are not significantly different (P > 0.05) 51 Figure 4.3 Mean (± SE) proportion of fe male progeny emerged from hosts of different age classes. Columns with the same letters are not significantly different (P > 0.05) 52 Figure 4.4 Mean (± SE) number of egg load per female progeny emerged from hosts of different age classes. Columns with the same letters are not significantly different (P > 0.05) 53 Figure 4.5 Relationship between body size a nd number of egg load per newly emerged female of A. ervi 53 Figure 4.6 Mean (± SE) number of aphids parasitised and eggs laid by A. ervi at different host densities (15, 25, 50, 75, 100 and 125 aphids/cylinder/day/female). Within the same category (Aphids parasitised or Eggs laid) columns with the same letters are not significantly different (P > 0.05) 58 Figure 4.7 Mean (± SE) number of eggs per parasitised aphid laid by A. ervi at different host densities. Columns with the same letters are not significantly different (P > 0.05) 59 Figure 4.8 Mean (± SE) daily parasitism rate of A. ervi at different host densities. Columns with the same letters are not significantly different (P > 0.05) 59 Figure 4.9 Mean (± SE) proportion of female progeny of A. ervi at different host densities. Columns with the same letters are not significantly different (P > 0.05) 60 xv Figure 4.10 Effect of body size of both sexes on the number of eggs laid (A), number of parasitism (B ), number of female progeny (C), proportion of female progeny (D) and days of proportion of female progeny > 50% (E). Columns with the same letters were not significantly different (P > 0.05). For A to D: F = 43.43, 36.12, 13.69, and 29.68 for A, B, C and D, respectively; df = 3,56; P < 0.0001. For E: F = 3.88; df = 3,56; P < 0.05. 67 Figure 4.11 Effect of female ( xf) and male ( xm) body size on fertility [A: y = exp(1.26 + 2.97 xf + 3.23 xm) (R 2 = 0.5222; F = 31.15; df = 2,57; P < 0.0001)] and fertility rate [B: y = exp(-0.67 - 2.25 xf + 2.77 xm) (R 2 = 0.6328; F = 48.96; df = 2,57; P < 0.0001)] in A. ervi 69 Figure 4.12 Relationship between body size and longevity of A. ervi. For female: y = 35.05 x – 9.16 (R 2 = 0.2895; F = 23.63; df = 1,58; P < 0.0001); for male: y = 16.54 x + 3.93 (R 2 = 0.1452; F = 9.85; df = 1,58; P < 0.01) 70 Figure 4.13 Effect of mate age (days) at mating on female offspring production of A. ervi: (A) proportion of female offspring produced in pairs of different age combinations of sexes (n = 90); (B) predicted proportion of female offspring produced by males and females of different ages 78 Figure 4.14 Proportion of female offspring of A. ervi males after successive matings. Means (± SE) followed by the same letters within each line are not significantly different (P > 0.05) 79 Figure 4.15 Relationship between proportion of female offspring and male mating history of A. ervi 80 Figure 5.1 E quipment used for recording mating behaviour 84 Figure 5.2 Mating behavioural sequences of A. ervi. Probabilities of a particular transition between stages are given 88 Figure 5.3 Comparison of the number of courtships and matings displayed by mated and virgin males when competing for a virgin female in 10 minutes. n = the number of males 89 Figure 5.4 Interactions of male (χ m) and female ( χf) densities affecting homosexual behaviour in A. ervi: (A) number of homosexual events performed by males in combinations of different male and female densities (y = exp(- 0.9372 + 0.6421 χm - 0.0262χ m2 - 0. 2866 χf + 0.0188 χmχf); (B) predicted number of homosexual events affected by male and female densities 110 xvi Figure 5.5 Interactions of male (χ m) and female ( χf) densities affecting mate competition in A. ervi: (A) number of mate competition events performed by males in combinations of different male and female densities (y = exp(- 2.1592 + 0.4421 χm - 0.0173χ m2 + 0.2742 χf - 0.0208 χf2 ); (B) predicted number of mate competition events affected by male and female densities 112 Figure 5.6 Interactions of male (χ m) and female ( χf) densities affecting mating success in A. ervi : (A) number of mating success in combinations of different male and female densities (CCD: y = exp(- 1.5704 + 0.2042 χm - 0.0092 χm2 + 0.3159 χf - 0.0126χ f2 ; F = 38.40, df = 4,201, P < 0.0001, R 2 = 0.4332); (B) predicted number of mating success affected by male and female densities 113 Figure 5.7 Interactions of male (χ m) and female ( χf) densities affecting mating period in A. ervi : (A) mating period in combinations of different male and female densities (y = exp(3.9891 + 0.0155χ m); (B) predicted mating pe riod affected by male and female densities 115 . General Introduction 1 CHAPTER 1 GENERAL INTRODUCTION 1.1 Introduction Aphidius ervi Haliday 1834 is an important biological control agent and usually introduced to control aphid species on lucerne, Medicago sativa (Leguminosae) (Cameron et al. 1979; Sandow 1981; Mackauer and Kambhampati 1986; Cameron and Walker 1989; Waterhous e and Sands 2001). Lucerne has been cultivated for thousands of years and is the most important forage/grazing legume (Grewal and Williams 2001). Lucerne, with it s high nitrogen fixation capacity, is also grown in rotation with wheat to improve the yield and protein content of wheat grain in Australia (Grewal and Williams 2001). It gives high yields of nutritious fodder that often out-yields pasture by 50~100% in New Zealand (McSweeney and Dunbier 1978; Rive 2003). It is grown today on all continents except Antarctica and more than 33 million hectares cultivated in the world (Seigler 2005). Both blue-green lucerne aphid, Acyrthosiphon kondoi Shinji and pea aphid, Ac. pisum (Harris) are the major pests on lucerne that reduce the yields and quality (Kain et al. 1977; Kain and Biggs 1980). Ka in et al. (1979a) indicate that, in New Zealand, infestation with blue-green lucerne aphids and pea aphids may result in large losses in herbage production (30~60% ) and reduction in plant density (18%). In the United States, pea aphid was responsible for lucerne production losses of $60 million in 1963 (Harper et al. 1978). Moreover, the field and glasshouse studies indicate that infestation of blue-green lucerne aphid a nd pea aphid caused increased coumestrol levels in severely aphid-damaged lucerne which were high enough to impair ewe fecundity (Kain and Biggs 1980). To control the cosmopolitan blue-green lucerne aphid and pea aphid, three common aphidiids: A. ervi , A. eadyi Stary, Gonzalez and Hall, and A. smithi Sharma and Subba Rao, were introduced as potential biological control agents in South and North America, Australia and New Zealand (Cameron et al. 1979; Sandow 1981; Mackauer and Kambhampati 1986; Came ron and Walker 1989; Waterhouse and . General Introduction 2 Sands 2001). A. eadyi and A. smithi are specialized species of pea aphid and A. ervi is polyphagous. Of these three species, A. ervi is the most abundant, and widely distributed and adapts to wide range of climates and habitats (González et al. 1978). In North America, A. ervi displaced A. smithi as the predominant species even if the latter was introduced first and rapidly became established (Mackauer and Kambhampati 1986). In Australia, A. ervi , A. eadyi , A. smithi and A. urticae Haliday were introduced and released specifically for control of pea aphid between 1980 and 1981 in New South Wales and Victoria (Sandow 1981; Milne 1986; Carver 1989). A. ervi spread up to 300 km in a year and is now widely distributed in Queensland, New South Wales, Victoria, Western Australia and Tasmania (Sandow 1981; Milne 1986; Carver 1989). A. smithi had become established in Tasmania only but not in the mainland Australia, and in Tasmania its population is lower than A. ervi (Carver 1989). In New Zealand, five species including three Aphidius species ( A. ervi , A. eadyi and A. smithi) and Ephedrus plagiator (Nees) and Praon barbatum Mackauer were imported from California for biological control of aphids on lucerne (Cameron et al. 1979; Cameron and Walker 1989). However, only A. ervi and A. eadyi successfully established (Cameron and Walker 1989). A. eadyi established easily and spread rapidly, but A. ervi eventually replaced A. eadyi as the predominant parasitoid in the North Island (Cameron and Walker 1989). This displacement was driven through direct and indirect larval interactions and possibly superior searching ability by A. ervi females in mixed populations (Chua et al. 1990; McBrien and Mackauer 1990; Bueno et al. 1993). A. ervi has significantly suppressed lucerne aphid populations in the field. In Australia, the parasitism rate of blue -green lucerne aphid was up to 70% in Queensland, New South Wales, Victoria and West Australia in 1981 (Sandow 1981). Surveys in 1982 and 1983 demonstrated the su ccessful establishment and dispersal of A. ervi in blue-green lucerne aphid and pea ap hid in the major lucerne-growing areas of New South Wales and also the ability of the parasitoid to built up populations rapidly again after a severe and widespread drought; sam pling in southern New South Wales and in costal areas of Australia from 1994 to 1996 revealed that high populations of blue-green lucerne aphid and pea aphid were undoubtedly parasitised by A. ervi (Waterhouse and Sands 2001). In Ne w Zealand, since 1979-80, populations . General Introduction 3 of blue-green lucerne aphid and pea aphid have steadily declined at a number of northern North Island sites. This decline lik ely resulted from a number of factors, including the establishment of the parasitoids (Cameron et al. 1979), and replacement of susceptible cultivars with aphid resistant lucerne (Dunbi er and Easton 1982), together with better management practices (Kain and Trought 1982); however the increase of parasitoid populations indicates that parasitoids have provided successful control of lucerne aphids (Cameron and Walker 1989). Compared to A. eadyi and A. smithi , the greater effectiveness of A. ervi in biological control of pea aphid and blue-g reen lucerne aphid may be due to high behavioural adaptation of A. ervi to the target aphids and biotypes, and phenological adaptations to the climates of aphid environment (Starý 1988). Moreover, the major aphid host plant, lucerne, is usually grown for many years so that the community system of plant-host and aphid- parasitoid is not disturbed by agricultural activities at the end of each year (Starý 1962). Finally, the high persistence of A. ervi within its hosts may be a key factor leading to succes sful biological control in the disturbed systems (Rauwald and Ives 2001). Although A. ervi has been effective in reducing aphid populations in the high rainfall areas, it is less effective in drier in land areas such as in Australia (Waterhouse and Sands 2001). Aphid populations are still high in some lucerne-growing areas, for example, in Palmerston North, New Zealand, the number of blue-green lucerne aphids can be as high as 120 individuals per stem on the susceptible cultivar, ‘Wairau’, in spring and summer 2003 (unpublished data), which is four times higher than the critical control threshold (i.e. 30 aphids per stem) (Kain and Trought 1982). Therefore, augmentation of field populations may be an appropriate option which is relevant to mass-rearing. 1.2 Importance of the Study of Parasitoid Reproductive Behaviour The study of reproductive behaviour is an important key for the understanding of how parasitoids should influence the population dynamics of their hosts and the structure of the insect communities in which they live (van Alphen and Jervis 1996), . General Introduction 4 and for the determination of what behaviours parasitoids adopt in order to maximize their reproductive success (Anon 2005). In parasitoids, the production of progeny directly reduces the host populations. Therefore, studying how parasitoids optimize their reproductive output should provide means to improve pest control efficacy in biological control programmes. Until the 1990s, behavioural ecologists largely focused on developing and testing theoretical predictions for the fitness consequences of behaviour but devoted less effort to the studying of behavioural mechanisms employed by the animals in order to optimize their reproductive output (Anon 2005). In the last decade, behavioural ecologists have tried to integrate the study of behavioural mechanisms with the functional approach to study the fitness consequences of behaviour (Anon 2005). In A. ervi , many studies have investigated the behavioural mechanisms employed by parasitoids to find hosts (Vinson 1984; Schmidt 1991; Vet and Dicke 1992; Guerrieri et al. 1993, 1997; Turlings et al. 1993; Godfray 1994; Du et al. 1996, 1997; Quicke 1997) and allocate sex of pr ogeny to the hosts (Micha et al. 1992; Battaglia et al. 1993, 1995, 2000; Pennacchio et al. 1994; Du et al. 1997; Weinbrenner and Volkl 2002), but how these mechanisms contribute to fitness or act as constraints on the optimization of reproductive behaviour of parasitoids is still poorly known. The study of parasitoid reproductive behaviour is essential for understanding and modeling life history evolution and behavioural decisions regarding host choice, fecundity, and sex ratio (Godfray 1994; Vi sser 1994; Cloutier 2000). It is also important in the development of theory in behavioural ecology that incorporates knowledge of the underlying mechanisms of parasitoid behaviour in models that predict maximal fitness (Anon 2005). Knowledge of these can be used to define more efficient biological control technologies of insect pests in agriculture. 1.3 Relevance of This Research Success in biological control is ofte n dependent on a thorough understanding of the organisms involved both injurious a nd beneficial species, and their intricate interactions. Few studies have investigated the reproductive behaviour of A. ervi , limiting the design, improvement and implementa tion of biological control strategies. . General Introduction 5 The study of A. ervi reproductive behaviour is relevant because: (1) Lucerne is an important cosmopolit an legume crop grown for both grazing and hay production in the world, and it can also improve soil quality, which consequently enhances agricultural profitabi lity (Hanson et al. 1988). However, some pests, such as blue-green lucerne aphid a nd pea aphid have continually caused serious damage resulting in the reduction of fresh or hay production and quality. Although the resistant lucerne cultivars have been introduced (Dunbier and Easton 1982) and better management practices have been developed (Kain and Trought 1982), around a third of lucerne grown in New Zealand is still the cultivar ‘Wairau’, which is susceptible to all major pests (Rive 2003). (2) Aphidius ervi is a cosmopolitan parasitoid (Marsh 1977) and an important biological control agent of several major aphid species on economically important crops such as legumes and cereals (Starý 1978; Powell 1982). However, its ability in suppressing aphid population decreases in drought prone regions (Waterhouse and Sands 2001). Lucerne is drought resistant (Seigler 2005), it is ex pected that more lucerne will be grown in drier regions if the global temperature continues to increases due to the ‘global greenhouse e ffect’. There is thus a need to find strategies to improve the success of pest control using A. ervi , especially in cropping systems where lucerne is grown for many years. (3) Development of IPM techniques, in cluding the application of biological control agents, can reduce the amount of chem ical used and increase the value of the harvested produce. The study of A. ervi reproductive behaviour provides information for the understanding of its biology and the development of environmentally friendly IPM practices. The host habitat loca tion and host searching behaviour of A. ervi have been well studied (Vinson 1984; Schmidt 1991; Vet and Dicke 1992; Guerrieri et al. 1993, 1997; Turlings et al. 1993; Godfray 1994; Du et al. 1996, 1997; Quicke 1997). However, little is known about its reproduc tive behaviour such as host preference, functional response and mating behaviour. Moreover, how these behaviours affect reproductive fitness of parasitoids and population dynamics of both aphids and parasitoids is not clear. Unraveling the factor s which govern the reproductive . General Introduction 6 behaviour and its impact upon the reproductive fitness of this species will provide valuable information for IPM developers in improving mass rearing and biological control strategies. 1.4 Aim and Objectives In order to provide usef ul information for the development of biological control strategies and theoretical research, the aim of this research is to study the reproductive biology and mating behaviour of A. ervi with the following objectives: (1) To investigate the basic biology of A. ervi; (2) To determine the factors affecting the reproductive fitness of A. ervi ; (3) To investigate the mating behaviour and sexual selection of A. ervi. . Literature Review 7 CHAPTER 2 LITERATURE REVIEW 2.1 Introduction This chapter reviews the current know ledge on Aphidiinae species that are relevant to my PhD studies. Special re ferences are given to known biology of A. ervi. 2.2 Taxonomy of A. ervi Among genera that attack aphids, Aphidius is the largest one in the Aphidiinae with over 70 species described worldwid e (Mescheloff and Rosen 1990) and the number of described species continues to increase (Kavallieratos et al. 2004, 2006; Tomanović et al. 2003). Aphidius ervi Haliday was first named in 1834, and has since been known by many synonyms (Marsh 1977) including: Aphidius ulmi Marshall, 1896 Aphidius medicaginis Marshall, 1898 Aphidius ( Aphidius ) pisivorus Smithi, 1944 Aphidius fumipennis Gyorfi, 1958 Aphidius ervi nigrescens Mackauer, 1962 Aphidius caraganae Starý, 1963 Aphidius mirotarsi Starý, 1963 The classification for this species is: Order: Hymenoptera Family: Braconidae Subfamily: Aphidiinae Genus: Aphidius Species: ervi . Literature Review 8 2.3 Identification of A. ervi A large number of Aphidius species have been recorded for pea aphids and cereal aphids including A. ervi , A. eadyi , A. smithi , A. picipes (Nees), A. pisivorus Smith, A. popovi Starý, A. nigripes Ashmead, A. uzbekistanicus Luzhetski, A. rhopalosiphi De Stefani Perez, A. matricariae (Haliday) and A. colemani Viereck (Pungerl 1983). As mentioned above, three Aphidius species and E. plagiator and P. barbatum had been imported into New Zealand to control pea aphids and blue-green lucerne aphids, but only A. ervi and A. eadyi had successfully established (Cameron et al. 1979; Cameron and Walker 1989). Because the breeding colony used in this study started from A. ervi emerged from blue-green lucerne aphids collected on lucerne (Section 3.2), I only identified the A. ervi and A. eadyi and keys for identification of other Aphidius species have been reported by Pungerl (1983). Figure 2.1 Lateral view of tergum of female: (A) A. eadyi (from Starý et al. 1980) and (B) A. ervi (from Kavallieratos et al. 2001). Routine examination of the two key-characters may identify A. ervi and A. eadyi , i.e. number of antennal segments and character on the anterolateral area of petiole (Marsh 1977; Starý et al. 1980; P ungerl 1983). The numbers of antennal segments are 19-22 and 17-19 for A. ervi male and female, respectively, and 20-24 and 20-21 for A. eadyi male and female, respectively. The anterolateral area of petiole . Literature Review 9 is costulate, bearing 7-9 ridges in A. eadyi , (Figure 2.1A), while the surface of the anterolateral area of petiole is coarsely rugose, with no ridge in A. ervi (Figure 2.1B). 2.4 Distribution of A. ervi Aphidius ervi is an aphidiid of Palaearctic origin. A. ervi can adapt to different climatic conditions: continental Europe, high plateaus, coastal and desert (González et al. 1978). Now it occurs in North and South America, Europe, North Africa, Middle East, Australia, New Zealand, India, Japa n and China (Marsh 1977; Cameron et al. 1979; González et al. 1979; Sandow 1981; Starý and Delfino 1986; Raychaudhuri 1990). 2.5 Hosts and Host Preference of A. ervi Aphidius ervi is a polyphagous parasitoid and has wide host range including: A. pisum, A. kondoi , A. caraganae (Cholodkovsky), Aphis sp., Aulacorthum solani (Kaltenbach), Macrosiphum euphorbiae (Thomas), Microlophium evansi (Theobald), Myzus certus (Walker), M. persicae (Sulzer) and Rhopalosiphum padi (L.) (Marsh 1977). Aphidius ervi can be reared either on blue-g reen lucerne aphid or pea aphid. However, host preference of A. ervi depends on the previous aphid host (Cameron and Walker 1989). For example, when reared on blue-green lucerne aphid for one year, A. ervi continues to prefer blue-green lucerne a phid rather than pea aphid; when reared on pea aphid, A. ervi prefers pea aphid in the first generation and performs as well on pea aphid as on blue-green lucerne a phid in the second generation, but its performance on pea aphid declines in the third generation (Cameron and Walker 1989). Thus, A. ervi is capable to switch its hos t preference behaviour by rapid responding to available host types, which enab les it to survive as a biological control agent in natural conditions. . Literature Review 10 2.6 General Biology of A. ervi The life cycle of A. ervi includes egg, larva, pupa a nd adult. It is a solitary endophagous species. At 20°C, eggs hatch in 3 days; larvae last for about 8 days, after which time, a cocoon is formed and the a phid is ‘mummified’ ( unpublished data). The parasitoid pupae last for about 5 days in aphid mummies before adults emerge. Adults may survive and oviposit for about 10 days. The low temperature thresholds for development is 3.9, 2.6 and 4.5ºC for A. ervi , blue-green lucerne aphid and pea aphid, respectively, thus it is well synchronized with their hosts in cool spring temperature (Cameron and Walker 1989). A. ervi overwinters as diapausing final instar larvae in its host aphids and its diapause is induced by short day length (vaz Nunes and Hardie 1996; Christiansen-Weniger and Hardie 1997). 2.6.1 Eggs Mature eggs in ovaries of A. ervi are elongate and oval in shape with one or both ends being more or less tapering (Spencer 1926), and 50 × 20 μ m in size (Figure 2.2A). During oviposition eggs are placed by parasitoids in the adipose tissue of aphids, usually with part of the egg proj ecting a short distance into the body cavity. Soon after oviposition the e gg absorbs liquid from the aphid by osmosis, and increases enormously in size, becomi ng lemon shape and as large as 130 × 60 μm (Figures 2.2B-F). 2.6.2 Larvae Aphidiine wasps are endoparasitoids that undergo larval development inside a living host. Three instars based on larval morphology have been proposed (Schlinger and Hall 1960; Starý 1966; Calvert a nd van den Bosch 1972; O'Donnell 1987; Pennacchio and Digilio 1990) and confirmed for A. ervi by counting the exuviae of in vitro reared larvae (Pennacchio and Digi lio 1990). The first instar larva is mandibulate-caudate with sickle-shaped mandibles, the second instar larva is hymenopteriform and amandibulate, and the third instar larva has well developed mandibles (Pennacchio and Digilio 1990). Af ter hatching, larvae feed primarily by . Literature Review 11 surface absorbing and ingesting host hemolymph, and begin to feed on host tissues and embryos using mouthparts 7 days after oviposition (Polaszek 1986; Pennacchio and Digilio 1990). The first, second and third instar la rvae are recorded 56, 90 and 110 h after oviposition, respectively at 26°C (Pennacchio and Digilio 1990). The development of A. ervi larvae at 20°C is shown in Figure 2.3. Figure 2.2 Development of A. ervi eggs: mature eggs in ovaries (A), and eggs 1 h (B), 12 h (C), 24 h (D), 48 h (E) and 72 h (F) af ter oviposition. All pictures were taken in the present study. 50μm 50μm 50μm 50μm F D C B 50μm 50μm E A . Literature Review 12 Figure 2.3 Development of A. ervi larvae: newly hatched larva (3 days after oviposition) (A), and larvae of 1 day old (B), 2 day old (C ), 3 day old (D), 4 day old (E) and 5 day old (F). All pictures were taken in the present study. 200μm 200μm 200μm 200μm 200μm 200μm A B C D E F . Literature Review 13 Figure 2.4 Pupation of A. ervi: mature larvae [9 days (A) and 10 days (B) after oviposition] and pupae [11 days (C) and 12 days (D) after oviposition]. All pictures were taken in the present study. 500μm 500μm 500μm 500μm A B C D . Literature Review 14 Figure 2.5 A. ervi pre-adult (A) (13-14 days after oviposition), aphid mummy (B) and adults (C). All pictures we re taken in the present study. 500μm 500μm A B 500μm C ♂ ♀ . Literature Review 15 2.6.3 Pupae The mature larva (Figures 2.4A and B) makes a hole at lower side of the aphid’s skin, fixes the aphid by the secretion of its salivary glands to the leaf surface, and then pupates inside the mummified aphid (Starý 1962). Th e pupa is initially whitish, gradually turning yellow (Figures 2.4C and D) and black (Figure 2.5A) before adult eclosing from aphid mummy. 2.6.4 Adults The eclosed adult cuts out a circular hole in the dorsal abdominal part of the host just above the siphons, and then emer ges (Figure 2.5B). The period of emergence usually lasts several minutes (Starý 1962). The newly emerged parasitoid is soft and in only about five minutes its wings expend and body becomes compact (Spencer 1926). During this period, the parasitoid sits on the plant, moves its legs and antenna e and cleans itself intensively. In about 10 minutes after emergence, it is able to fly. A certain period for the maturation of eggs seems necessary (Spencer 1926). A. ervi adult is a long (4~5 mm), slender, and black wasp (Figure 2.5C). 2.6.5 Host Searching and Oviposition of A. ervi In general, the foraging behaviour of hymenopteran species is divided into three mechanistic steps: (1 ) host habitat loca tion, (2) host loca tion and (3) host acceptance (Doutt 1959; Vinson 1976). Chemical and physical cues appear to play a major role at almost every level of th e foraging behaviour (Vinson 1984; Schmidt 1991; Vet and Dicke 1992; Turlings et al. 1993; Godfray 1994; Quicke 1997). Success of host searching depends upon the execution of appropriate behavioural responses to an array of cues which are available during foraging. Host habitat location. Plant volatiles emanating from the host’s food and host odors have been shown to be important cues in host habitat location at long distances . Literature Review 16 for a number of hymenopteran parasitoids. In a wind tunnel bioassay, naïve A. ervi females responded poorly to undamaged bean plants and to pea aphid isolated from the plant, but they showed strong oriented flight responses to aphid-infested plants and to aphid-damaged plants from which the aphids had been removed (Guerrieri et al. 1993; Du et al. 1996). When offered a choi ce, more parasitoids flew toward and landed on the host-infested or host-damaged plants than on undamaged plants (Du et al. 1996). These results suggest that A. ervi females use host-induced plant volatiles as host habitat location cues. The compounds of volatile extracts from pea aphid infested plants that could be involved in parasitoid attraction, ar e identified as 6-methy1-5-hepten-2-one, linalool, ( E)- β -ocimene, ( Z) - 3 -hexeny acetate, ( Z) - 3 -hexen-1-ol, and ( E) - β -farnesene, with A. ervi appearing to use the more specific, 6-methy1-5-hepten-2-one, which is not induced in plants infested by the non-hosts, such as black bean aphid, Aphis fabae Scopoli (Du et al. 1998). Thus, 6-methy1-5-he pten-2-one is potentially one of the volatile components that allow A. ervi to distinguish between plants infested with hosts and non-hosts (Du et al. 1998), and its ab ility has been demonstrated in the wind tunnel study (Du et al. 1996). The parasitoid experience enhances sear ching ability in at least 20 species (Turlings et al. 1993). In the wi nd tunnel study, responses of A. ervi females to volatiles emanating from undamaged, host-infested and damaged or host-damaged plants are enhanced by previously exposi ng the females to the plants infested by aphids (Du et al. 1997; Guerrieri et al. 1997). Thus the plasticity of parasitoid foraging behaviour may provide ways of enhancing the efficiency of A. ervi in pest management (Guerrieri et al. 1997). Moreover, long range attraction of A. ervi toward monochromatic light in the green region of the spectrum (514 nm) has been reported (Goff and Nault 1984), which suggests that this parasitoid employs a visual response to green vegetation during the host habitat location phase of host selection. The pea aphid sex pheromones as a potential host habitat location cues have also been demonstrated for A. ervi in a wind tunnel study (Glinwood et al. 1999). . Literature Review 17 Host location and acceptance. In a short range, host location and acceptance by A. ervi are influenced by the physical cues. A. ervi females show strong attack responses toward green pea aphid even without physical contact, indicating the use of visual cues during host location (Battaglia et al. 1995). A. ervi females react to yellow pigment with repeated oviposition attack re sponses, but they do not react to green pigment (Battaglia et al. 2000). The spectrum of reflected light from the yellow pigments is very similar to that from the ‘green’ natural pea aphid, with a high proportion of the total radiation energy being emitted in the yellow-orange wavebands (580~660 nm) (Battaglia et al. 2000). The host location and acceptance by A. ervi are also regulated by semiochemical cues. It has be en shown that the cornicle secretion, cuticula and honeydew of pea aphid act as contact kair omones and elicit an intense oviposition attack response by A. ervi females, which appears to be innate (Battaglia et al. 1993; Pennacchio et al. 1994; Du et al. 1997). A. ervi females do not attack wet pea aphid that are washed previously with water, but one hour later this phenomenon disappears and A. ervi females attack washed aphids to the same degree as dry ones (Weinbrenner and Volkl 2002), confirming the chemical cues on the cuticula of aphids as contact kairomones. In the field, A. ervi females may encounter parasitised and unparasitised hosts and the parasitised hosts may have been attacked by conspecific or heterospecific parasitoids. Host discrimination (i.e. the ability to distinguish parasitised and unparasitised hosts) is a co mmon phenomenon among Hymenopt eran parasitoids (van Lenteren 1981). In A. ervi -blue-green lucerne aphid system, newly parasitised hosts appear to be susceptible for superparasitism but parasitoids tend to avoid superparasitism six hours after the hosts are attacked (Micha et al. 1992). This suggests that A. ervi females do not mark their hosts with an oviposition-deterring pheromone to limit short term superparasitism, and the trend to avoid superparastism at longer intervals is probably attributed to changes occurring in the host as a response by the aphid to parasitism or to changes induced by the presence of a parasitoid egg within the host (M icha et al. 1992). A. ervi females generally avoid ovipositing in a pea aphid which has been parasitised by another species, such as Aphelinus asychis Walker, but they can survive in the hosts superparasitised by A. asychis, killing the . Literature Review 18 competing larvae by physical and possibly physiological suppression (Bai and Mackauer 1991). 2.7 Reproductive Biology of Aphidiinae 2.7.1 Reproductive System The female reproductive system of Aphidiinae usually comprises the ovipositor, ovaries, oviducts, vagina, spermatheca and associated glands, and the venom gland (with its associated reservoir) (Quicke 1997). The ovarial imaginal bud gives rise to the lateral oviducts and the posterior ends of the lateral oviducts unite to form the common oviduct, sometimes referred to as the vagina. There is only one pair of meroistic ovarioles in Aphidius species (Quicke 1997). Matu re eggs move from the ovarioles and pass through the lateral oviduct toward the common oviduct (see Figure 2.6 for A. ervi ) and then the ovipositor. The spermathecal complex consists of a small, nearly spherical reservoir which opens into a narrow spermathecal duct which opens into the anterior end of the common oviduct; sperm release for fertilization is stimulated by movement of an egg into the oviduct (Quicke 1997). The venom gland may insert on to the venom reservoir which opens either into the distal part of the vagina or into the egg canal of the ovipositor via the dorsal ovipositor valve (Quicke 1997). A. ervi ’s venom is injected in to the body of aphids when ovipositing, and is an imported regulatory factor of the target host ovaries and responsible for pathological syndrome which results in developmental arrest (Digilio et al. 1998). The female reproductive system of A. ervi is shown in Figure 2.6. The male reproductive system in Hyme noptera comprises paired testes, vas deferens, accessory glands and ejaculator y duct which leads to the aedeagus of the external genitalia (Sanger and King 1971; Quicke 1997) and contents of the accessory glands mix with sperm from the vas defere ns or seminal vesicle before ejaculation (Quicke 1997). An example of the reproduc tive system in male Hymenoptera is shown in Figure 2.7. . Literature Review 19 Figure 2.6 Female reproductive system in A. ervi . This picture was taken in the present study. Figure 2.7 Male reproductive system in Nasonia vitripennis (Walker) (Hymenoptera: Pteromalidae) (from Sanger and King 1971). Ovaries Venom gland Venom reservoir Spermatheca Lateral oviducts Common oviduct Ovipositor . Literature Review 20 2.7.2 Mating Behaviour The primary mating behaviour has been described by Spencer (1926), Mackauer (1969), Battaglia et al. (2002) and McClure et al. (2007). Mating in A. ervi is mediated by sex pheromones produced by females. As found in another Aphidius species, A. nigripes (McNeil and Brodeur 1995), virgin females of A. ervi produce a long-distance sex pheromone that stimulates upwind flight and then produce a close- range pheromone that elicits courtship behaviour by males (McClure et al. 2007). After they detect the presence of females within a short distance of seven to ten millimeters (Spencer 1926), males perform courtship behaviour such as wing fanning, approaching and mounting (Battaglia et al . 2002; McClure et al. 2007). After a male mounts a female, he crawls forward and moves his antennae rapidly from side to side and touches those of the female. When the female accepts the male, the latter moves backward, curves his abdomen, and inserts his aedeagus into her genitalia. The female terminates the mating by walking away from the copulation site (M cClure et al. 2007). 2.7.3 Sex Allocation and Sex Ratio In Hymenoptera, unmated females lay unf ertilised haploid eggs that develop into males, and mated females can lay either unfertilised eggs that develop into males or fertilised diploid ones that develop into females. Because of the haplo-diploid system of sex determination in Hymenoptera, sex ratio in parasitoid wasps can vary considerably. In the laboratory mass-rearing of parasitoids, male-biased sex ratios are undesirable because males produced in excess of those required for ensuring that all females are inseminated are useless for biological control. Thus the study of sex allocation in parasitoids is an important area of reproduction not only for those interested in evolutionary ecology, but also for practitioners of biological control (Luck 1990). Natural selection acts upon oviposition behaviour to sex allocation that is adaptive in relation to different conditions. When only one female colonises a patch, she should produce only enough sons to fertilise all her daughters (Hamilton 1967). However, sex allocation of a searching fe male may be affected by the density of . Literature Review 21 female population. When several females co lonise a patch, each female needs to produce sufficient sons to compete with the sons of other females for mating, the evolutionarily stable sex ra tio for a parasitoid population should switch to be male- biased (Hamilton 1967). This is one reason why the mass-rearing of parasitoids often does not result in desired female-biased sex ratios. Moreover, sex allocation is affected by host quality. Hosts provide the only resource for the development of immature stages of parasitoids, and they vary in the amount of resource they contain (e.g. vary in size). Charnov et al. (1981) and Charnov (1982) suggest that parasitoids lay fertilized eggs in larger hosts resulting in larger female progeny. This may be that females have more to gain from developing in large host than do males, namely that major com ponents of female fitness, in particular fecundity, are more strongly correlated w ith their body size than major components of male fitness, in particular mating ability (van den Assem et al. 1989). 2.8 Factors Affecting Reproductive Fitness of Parasitoids 2.8.1 Host Quality The hosts encountered by parasitoids will often vary in their quality as food for their larva development, in the time n eeded for their attack and parasitism, and possibly in the risks of their mortality dur ing oviposition. Variation in host quality often depends on the age of the host. Most hosts are immature insects and grow in size as they age; larger hosts are often better food sources for parasitoid larvae (Charnov et al. 1981; Liu 1985), though they may be better defended, both physically and physiologically. According to the optimal diet model (Stephens and Krebs 1986), parasitoids should attack the best hosts and this is assumed to maximize the gain of a quantity or quality such as the number or size of offspring. Thus, a parasitoid may show a preference for a particular host type available in the environment (Godfray 1994). In koinobiont parasitoids, where the hos ts continue developing for a certain period of time after parasitisation, host sele ction may not reflect current host quality but may be based on an assessment of future growth rates and resources available for the developing larvae (Rivero 2 0 0 0 ). For some solitary species of parasitoids, hosts . Literature Review 22 selected for oviposition are determined by host size (Kouamé and Mackauer 1991). Many studies demonstrate that host size prefer ence by parasitoids affect their progeny fitness, such as the body size (Liu 1985; Lampson et al. 1996), progeny sex ratio (Godfray 1994; Napoleon and King 1999; Ue no 1999), and egg load at emergence (Liu 1985; Visser 1994; Mills and Kuhlmann 2000). 2.8.2 Host Density The impact of a parasitoid on its host population depends upon its ability to find and parasitise hosts and to increase the number of its offspring when needed (Mackauer 1983; Waage and Hassell 1982). This is largely determined by parasitoid’s functional response to host density (density-dependent pa rasitisation) and density- dependent sex ratio (Hassell and Waage 1984). The functional response is defined as the number of hosts parasitised per parasitoid per unit of time as a function of host density and is one of the essential ingredients in modeling any host interaction (Hassell 1986). It is important to identify the response form in population models, fr om which prescriptions for augmentative biological control can be developed (Mills and Lacan 2004). Three types of functional response of parasitoids are classified by Ho lling (1959): Type I, where there is a linear increase to a maximum in the number of hos ts parasitised as host density increases; Type II, where the response increases non-linearly towards a maximum value; and Type III, where the response is sigmoi d, again approaching an upper asymptote. In analytical host-parasitoid models, changes in the density-dependent sex ratio influence the level of equilibrium popula tions as well as the stability of the host- parasitoid relationships (Ha ssell and Waage 1984), and in turn determine the success or failure of biological control (Waage and Hassell 1982). Both functional response and density-dependent sex ratio together affect the dynamic interaction between parasitoids and pests, and determine parasitoid population size of future generations. . Literature Review 23 2.8.3 Parasitoid Body Size Parasitoid quality control is important to ensure the success of a biological control programme. One straightforward indi cator of parasitoid quality is size, as there is a positive correction between body size or weight and one or more fitness variables for many parasitoids (Jervis and Copland 1996) . Body size of insects has usually been considered to be a key dete rminant factor for many physiological and fitness characters (Thornhill and Alcock 1983; Honék 1993; Cloutier et al. 2000; Jiménez-Pérez and Wang 2004). For example, the reproductive fitness of females in terms of searching rate, longe vity, and fecundity, and ability to parasitise hosts, is often positively correlated with their body size (van den Assem et al. 1989; Honék 1993; Godfray 1994; Visser 1994; Kazmer and Luck 1995; Cloutier et al. 2000; Sagarra et al. 2001; Arakawa et al 2004). La rge females are able to regenerate eggs faster when required (Cloutie r et al. 2000). Large males usua lly have greater ability to inseminate and compete for mates (van den Assem et al. 1989; Kazmer and Luck 1995) or have better genes and more sp erm supply (van den Assem et al. 1989). Therefore, knowledge of these biologica l parameters is important for the implementation of an efficient mass-rearing system as it leads to better understanding of the progeny production and sex ratio. It shou ld also enhance the efficiency of field releases by ensuring the quality of the released parasitoids. Moreover, the fitness consequences of size and its correlates, es pecially the supply of eggs or sperm and adult longevity, are important in popul ation dynamics and essential to the understanding and modelling of life history evolution and behavioural decisions (Visser 1994; Cloutier et al. 2000). 2.8.4 Mating History of Parasitoids In hymenopteran species, no eviden ce is found that females obtain a nutritional contribution from males during copulation (Mad el et al. 1990; Godfray 1994; Fauvergue et al. 1998). Thus, mating a ffects parasitoid reproductive fitness mainly in terms of egg fertilization. In a review of mating behaviour among parasitoids, Gordh and Debach (1978) f ound that multiple mating by females occurred in only five out of thirty-four species, while multiple mating by males occurred in all . Literature Review 24 cases. Ridley (1993) has collected inform ation from the literature on the female mating behaviour of ninety-nine species of parasitoid wasps, of which nearly 80% were reported to mate only once; furthe rmore, there was a strong tendency for gregarious females to mate several times and for solitary females to be monandrous. Therefore, in aphidiide species, mating hi story affecting reproductive fitness may only be relevant to males, and male mating hist ory or age may affect reproductive fitness by the supply of sperm during copulation. 2.9 Sexual Selection Sexual selection is one of the most active areas in studying behavioural ecology because it directly influences the reproductive fitness of insects. Individuals select their partners because potential mates vary in quality, quantity and availability. Sexual selection may occur through intrasexual selection where males or females compete for mates or through intersexual selection where females or males choose their mates with certain characteristics (Jennions and Petrie 1997; Panhuis et al. 2001). For aphidiide species, females are monandrous and males are polygynous. Therefore, there is no issue about sperm competition. For females, premating mate choice is extremely important especially in species with very female-biased sex ratio; they should be very choosy theoretically and mate with ‘best’ possible mates. Age and body size may be the major characteristics th at affect the sexual selection. It is expected that females shoul d choose large and young males to obtain more sperm, and males should select large and young females to fertilize more and better eggs. . General Biology of Aphidius ervi 25 CHAPTER 3 GENERAL BIOLOGY OF APHIDIUS ERVI 3.1 Introduction This chapter described the general methodology applied throughout this research, investigated the general biology of A. ervi in emergence, sex maturation and oviposition patterns of A. evi and then determined the effect of parasitism on host reproductive potential. 3.2 General Methodology The materials, procedures, environmental conditions and definitions detailed in this section were used throughout the thesis. 3.2.1 Materials Breeding colony : A breeding colony started from A. ervi emerged from blue- green lucerne aphid, collected on luce rne in Palmerston North, New Zealand in December 2002. Both blue-green lucerne aphid and pea aphid are hosts of A. ervi , and lucerne is the host plant of both species; however, because of the difficulty of growing lucerne in greenhouse or laboratory, pea aphid was used as host of A. ervi and broad bean, Vicia faba L. cv. Pride, was used as host plant of pea aphid. The A. ervi colony was reared on pea aphid for five generations before used in experiments. Breeding cage : A. ervi and pea aphid were maintained separately in aluminium framed cages (64 cm in length × 45 cm in width × 40 cm in height) with fine metal screen (0.2 mm in length × 0.2 in mm in width) on th e back and both sides and perspex on the top and front and alumin ium alloy on the bottom (Figure 3.1). . General Biology of Aphidius ervi 26 Figure 3.1 Breeding cage used for rearing A. ervi and pea aphid. Figure 3.2 E xperimental cylinder. Experimental cylinders: All experiments were carried out in transparent plastic cylinders (8.5 cm in diameter × 10.5 cm in height) with gauze-covered holes, one in the top (5 cm in diameter) and two (2 cm in diameter) in opposite sides for ventilation (Figure 3.2). A plastic contai ner (6.5 cm in diameter × 8.5 cm in height) was firstly inserted into a cardboard (10.5 cm in length × 10.5 cm in width) via hole (6.5 cm in diameter) cut in the centre before the contai ner was filled with tap water and covered. A broad bean cutting was placed in the plastic container via a hole (0.6 cm in diameter) cut in the centre of a cover. The plastic cylinder was then placed on and sustained by Plastic cylinder Cover of plastic container Plastic container Cardboard . General Biology of Aphidius ervi 27 the cardboard (Figure 3.2). The broad bean cutting was replaced when necessary. Honey solution (10%) was supplied for para sitoids as food daily in a cotton wool wick (1 cm in length), inse rted through a hole (0.6 cm in diameter) in the top of the cylinder (Figure 3.2). Microscope : A stereomicroscope (Leica MZ12, German) equipped with a micrometer eyepiece was used for dissecting parasitised aphids and measuring body size of newly emerged adults. A compound microscopy (Olympus, GH, Japan) equipped with transmitted light and a micrometer eyepiece was employed for counting the number of eggs in newly emerged A. ervi females. 3.2.2 Procedures Eggs laid: To determine the number of eggs laid by A. ervi in a parasitised aphid, aphids were dissected in a droplet 70% alcohol 4 d afte r parasitisation under the stereomicroscope and the number of larvae in each aphid was counted. The number of parasitoid larvae recorded from dissecting was assumed equal to the number of eggs laid (Bueno et al. 1993). Egg load : To determine the egg load of A. ervi females in ovaries at emergence, females were killed by freezing at -20 °C and dissected in a droplet of 70% alcohol on a slide under the stereomicroscope. One droplet of acid fuchsin was added to the alcohol for staining of the ovaries. After 3 ~ 5 mi n, the ovaries were covered with a slide cover and spread by gently pressing the slide cover. The number of eggs in the ovaries was counted and recorded under a compound microscope. Emergence : A. ervi adults used for experiment s emerged from mummies that were individually maintained in glass vials (1.5 cm in diameter, 5.0 cm in height). Mummies from each experiment were removed from bean plants and maintained in Petri dishes (5.5 cm in diameter × 1.3 cm in height) for emerging, and emerged adults were sexed. . General Biology of Aphidius ervi 28 Sex identification: The females could be distinguished from males by the sharp abdomen tip. 3.2.3 Environmental Conditions The colonies were maintained and expe riments were carried out in bioassay rooms at 20 ± 1°C and RH 60-70% with a photoperiod of 16:8 h L:D. Lighting was provided by high frequency broad-spectru m biolux tubes (Osram, Germany). 3.2.4 Definitions Number of parasitism: number of aphids parasitised by female A. ervi . Parasitism rate: percentage of aphids parasitised by female A. ervi. Superparasitism: when two or more eggs were laid in an aphid. Fecundity: the total number of eggs laid detected by dissecting. Fertility rate : the proportion of female progeny. Reproductive fitness: female parasitoids - fec undity, fertility, longevity, number and body size of progeny, and egg lo ad of newly emerged female progeny; male parasitoids – longevity, number of ma tes inseminated and progeny fathered, and body size of progeny. 3.2.5 Statistical Analysis and Reported Values All statistical analysis were set at P < 0.05 and carried out using SAS STAT (SAS Institute 2006). Unless stated otherwis e, all reported values are means ± SE. . General Biology of Aphidius ervi 29 3.3 Emergence, Sexual Maturation and Oviposition of A. ervi 3.3.1 Introduction Insect emergence events are usually rhythmic (Saunders 1982). In the parasitic hymenopterans, such rhythmicity is of ten synchronised with mating (Gordh and DeBach 1976; Nadel and Luck 1985) and ovi position (Armstrong et al. 1996; Couch et al. 1997) activities for an op timal reproductive fitness. Daily activity patterns have been studied in detail in some parasitic hymenopterans (Vogt and Nec hols 1991; Armstrong et al. 1996; Couch et al. 1997). Quicke (1997) suggested that for many para sitoid species, most oviposition occurs in the morning, for example, the squash bug egg parasitoid, Gryon pennsylvanicum (Ashmead) (Vogt and Nechols 1991). However, in some other species, such as the Sitona weevil parasitoid, Microctonus aethiopoides Loan, oviposition may occur during light and dark with the circadian oviposition activity corresponding with its host activity (i.e. feeding and oviposition) (Armstrong et al . 1996; Couch et al. 1997). Knowledge of parasitoids’ emergence, mati ng and oviposition patterns is vital to an understanding of the ecology and evolution of their reproductive strategies, which in turn contributes to the development and implementation of biological control programmes. Michaud and Mackauer (1994) reported that A. ervi could successfully oviposit in pea aphid in both photophase and scotophase, but the circadian emergence and oviposition patterns were still unknown prior to this study. To provide information for the development of mass-r earing and field releasing techniques and better understanding of biological control ecology of A. ervi , the circadian patterns of emergence and oviposition, and sexual maturation in A. ervi were investigated. . General Biology of Aphidius ervi 30 3.3.2 Materials and Methods 3.3.2.1 Experimental Parasitoids and Hosts Parasitoids used for experiments emerge d from mummies that were parasitised in the third instar (3 d old), and the third in star pea aphids were used as hosts in all experiments. 3.3.2.2 Emergence To observe the 24-h emergence patterns of A. ervi , two bioassay rooms were set up. The photophase in one room was set from 0800-24 0 0 hours (normal-light regime) and in the other room the photophase was between 1800-100 0 hours (reverse- light regime). To obtain parasitised aphids, a mated female parasitoid (<12 h old) was introduced into a Petri dish (5.5 cm in diameter × 1.3 cm in height) containing 25 third instar aphids for a period of 5 h. Fi fty-eight females were used for this study. Fifty parasitised aphids were reared on the bean plant in a transparent plastic cylinder (Figure 3.2). Fifteen and nine cylinde rs were maintained in the normal- and reverse-light regimes, resp ectively. Parasitoid emergence was observed from 675 mummies in the photophase in the normal-li ght regime and from 378 mummies in the scotophase in the reverse-light regime. Th e emergence incidence was recorded hourly and emerged adults were sexed. Developmenta l time from eggs to adults of both sexes was also recorded. 3.3.2.3 Sexual Maturation Because most matings occurred during the photophase (unpublished data), all experiments on sexual maturation were carri ed out during the photophase. To detect the sexual maturation period of adult parasi toids, two experiments were set up, each with 7 treatments. In the first experiment 12-h-old virgin females were paired with 0 (newly emerged), 2, 4, 6, 8, 10 and 12-h- old virgin males, and in the second . General Biology of Aphidius ervi 31 experiment 12-h-old virgin males were pair ed with 0 (newly emerged), 2, 4, 6, 8, 10 and 12-h-old virgin females. A virgin male was paired with a virgin female in a clear glass vial (1.5 cm in diameter, 5.0 cm in height) with a 0.5 cm mesh covered hole in lids. Twenty pairs were established for each treatment. The sexual behaviour of both sexes in a 30- minute period was observed and the number of courtship displays (male wing fanning) and matings (insemination) was recorded. 3.3.2.4 Oviposition To determine the oviposition patterns of A. ervi on a 24 h basis, an experiment was carried out with 20 A. ervi females in the normal-light regime and another 20 A. ervi females in the reverse-light regime. Each mated female (<12 h old) was introduced into a Petri dish containing 20 healthy aphids on a broad bean leaf and allowed to stay for 2 h (first oviposition bout ). She was then transferred to another Petri dish with 20 healthy aphids and allowed to stay for 2 h (second oviposition bout). This was repeated until 8 and 4 oviposition bouts were completed in the photophase and scotophase, respectively. All aphids from each oviposition bout (in a single Petri dish) were transferred to and reared in a plastic cylinder (Figur e 3.2). Because females might lay more than one egg in a host aphids, to determine the number of eggs laid, 10 aphids from each oviposition bout were dissected 4 days after parasitisation and the number of larvae in each aphid was counted under the stereomicroscope. The remaining parasitised aphids were reared until mummification. The number of parasitisms in each oviposition bout was recorded as the number of aphids parasitised. 3.3.2.5 Statistical Analysis A chi-square test was used to determine the difference in emergence incidence between the photophase and scotophase. The rejection level was when 2χ > 2 05.0,1 χ = . General Biology of Aphidius ervi 32 3.84. The Marascuilo procedure of the nonparame tric analysis (Daniel 1990) was used to assess the sexual maturation. The rejection level was when '0U > 2 05.0,6 χ = 12.59. All other data were normally distributed and analysed using ANOVA followed by a Tukey’s studentised range test. Data of propor tion of female offspring were subject to arcsine transformation before ANOVA, but untransformed means are presented. 3.3.3 Results 3.3.3.1 Emergence The developmental time from oviposition to emergence was significantly shorter for males (mean ± SE, 13.94 ± 0.03 d) than that for females (14.31 ± 0.03 d) (ANOVA: F = 99.60; df = 5,541; P < 0.0001). The hourly emergence rate (mean ± SE) was significantly higher in the photophase (5.99 ± 0.94%) than in the sc otophase (0.74 ± 0.32%) (ANOVA: F = 14.94; df = 1,22; P < 0.001). About 95% of pa rasitoids emerged in the photophase in both light regimes. On a 24-h basis, the ma le emergence peaked 2 h after light-on and the female emergence peaked betwee n 3-6 h after light-on (Figure 3.3). 0 2 4 6 8 10 12 14 16 18 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 5 6 7 8 Hour P er ce nt ag e of e m er ge d of fs pr in g Male: Female: Photophase Scotophase Figure 3.3 The percentage of female or male A. ervi offspring emerging throughout photophase or scotophase. . General Biology of Aphidius ervi 33 3.3.3.2 Sexual Maturation Newly emerged males were able to perf orm their courtship display to 12-h-old females but failed to mate until they were 4-h-old (Figure 3.4). Both courtship and mating increased with age (h); however, 10- and 12-h-old males were significantly more likely to court females than ≤ 4-h-old males ( '0U = 105.22 > 2 05.0,6 χ ; df = 6; P < 0.0001); mating success of 12-h-old males was al so significantly higher than that of ≤ 4-h-old ( '0U = 80.01 > 2 05.0,6 χ ; df = 6; P < 0.0001) (Figure 3.4). Newly emerged females were able to entice the courtship display by 12-h-old males and successfully mate (Figure 3.5). Twelve-h-old female s were significantly more likely to respond to ma les’ courtship display than ≤ 4-h-old females ( '0U = 32.35 > 2 05.0,6 χ , df = 6, P < 0.0001); mating success of 12-h-old females was significantly higher than that of ≤ 2-h-old females ( '0U = 14.97 > 2 05.0,6 χ ; df = 6; P < 0.0001). Male age (h) 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 0 2 4 6 8 10 12N o. o f m al es c ou rt in g an d m at in g Mating a a ab ab bc bc c a ab abc abc bc c c Courting Figure 3.4 The number of A. ervi males courting or mating at 0, 2, 4, 6, 8, 10 or 12 h after emergence. Within the same categor y (Courting or Mating) columns with the same letters are not significantly different (P > 0.05). . General Biology of Aphidius ervi 34 Female age (h) 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 0 2 4 6 8 10 12N o. o f f em al es r es po nd in g to m al e co ur tin g an d m at in g Response to male courting a a ab ab ab ab ab ab ab b b b b b Mating Figure 3.5 The number of A. ervi females responding to male courting or mating at 0, 2, 4, 6, 8, 10 or 12 h after emergence. Within the same category (Response to male courting or Mating) columns with the same le tters are not significantly different (P > 0.05). 3.3.3.3 Oviposition Females oviposited in both photophase and scotophase. In the photophase, the mean (± SE) number of eggs laid ( 11.73 ± 2.73) and parasitism (8.6 ± 1.03) per oviposition bout were significantly greater than that in the scotophase (4.18 ± 0.88 eggs and 3.43 ± 0.69 parasitism, respectively) (ANOVA: F = 7.87 and 24.19 for number of eggs laid and parasitism, respectively; df = 1,22; P < 0.01). In the photophase, the numbers of eggs laid and pa rasitism were significantly higher in the first oviposition bout (ANOVA: F = 40.95 and 12.64 for number of eggs laid and parasitism, respectively; df = 7,72; P < 0.0001) (Figure 3.6). However, no significant difference in number of eggs laid or parasitism between oviposition bouts was detected in the scotophase (F = 2.39 and 2.77 for number of e ggs laid and parasitism, respectively; df = 3,36; P > 0.05) (Figure 3.6). . General Biology of Aphidius ervi 35 -5 0 5 10 15 20 25 30 35 2 4 6 8 10 12 14 16 2 4 6 8 Hours N o. o f p ar as iti sm s an d eg gs la id No. of parasitisms No. of eggs laid Photophase Scotophase a b bc bc bc accc a a ab a ab b b b b b aa a Figure 3.6 The number of parasitisms and eggs laid by A. ervi females throughout photophase and scotophase. Within the same category (Photophase or Scotophase), means (± SE) followed by the same letters within each line are not significantly different (P > 0.05). 3.3.4 Discussion Among parasitoid insects, adult emergen ce restricted to certain times of the day is known in several species, including Trichogramma minutum Riley (Corrigan et al. 1995), Telenomus busseolae Gahan (Fantinou et al. 1998) and Encarsia formosa Gahan (van Lenteren et al. 1992). The resu lts of the present study demonstrate that A. ervi males and females have an emergence peak during the firs t few hours of the photophase (Figure 3.3). This suggests that th e onset of light may act as a signal that synchronises the rhythmic function of A. ervi adult emergence. This study has shown that A. ervi is a protandrous species with males emerging ca 9 h earlier than females. Singer (1982) argued that selection for protandry can only occur when generations are discrete. A. ervi does not appear to match that hypothesis because generations usually overlap in the field and females are readily available. However, A. ervi females mate only once during their lifespan, limiting the chance for males to encounter virgin females. Quicke (1997) suggested that the protandry is common in many parasitoids because a late-emerging male is . General Biology of Aphidius ervi 36 likely to encounter already mated females a nd he is genetically doomed, as he cannot get any matings. It has been reported that for some pa rasitic hymenopterans, a period of sexual maturation is necessary, for exampl e, the bean weevil parasitoid, Chryseida bennetti Burks (Perez-Lachaud and Campan 1994) and the tortrix moth parasitoid, Ascogaster reticulatus Watanabe (Kainoh 1986). In the present study, newly emerged A. ervi females were able to mate with males but males need at least 4 h to become sexually mature. Therefore, the early emergence of A. ervi males may be a selective strategy for higher reproductive fitness. For exampl e, early emerged males have a better chance to encounter virgin females (Nadel and Luck 1985, 1992), inseminate a greater number of the females they encounter (Waage and Ng 1984), and reduce the risk of females’ death before oviposition (Fagerström and Wiklund 1982). Moreover, since virgin females start to lay eggs within 30 min after emergence (unpublished data), early emerged males are able to mate with females before their oviposition. Diel periodicities of insect activity ar e often determined by a combination of endogenous and exogenous rhythms (Beck 1 980). The decreasing oviposition and parasitism of A. ervi during the photophase may be the result of a decreasing load of mature eggs, as reported in the sycamore aphid parasitoid, Monoctonus pseudoplatani ( Marshall) (Collins and Dixon 1986). Furthermore, visual cues play an important role in host finding and attacking by A. ervi (Michaud and Mackauer 1994). In this study, A. ervi females laid fewer eggs and attacked fewer aphids during the scotophase even though they had sufficient eggs. This suggests that the oviposition pattern of A. ervi is determined by an exogenous factor, the light regime. In conclusion, light appears to entrai n emergence and oviposition rhythms of A. ervi. Emergence of the adults in the morning probably coincides with more favourable conditions, such as the light, in which pa rasitoids may increase the chance for host habitat location, host searching and ovi position (Battaglia et al. 1995, 2000). The findings of this study have implications for laboratory mass rearing and field release of A. ervi. For example, newly emerged parasito ids should be stored for 12 h for copulation to occur before release, and the parasitoids should be released the following morning to achieve the higher reproductive output. . General Biology of Aphidius ervi 37 3.4 Effect of Parasitism by A. ervi on Development and Reproduction of Pea Aphid 3.4.1 Introduction Many studies have reported that hymenopteran endoparasitoids can reduce the reproductive potential of aphids (Campbell and Mackauer 1975; Liu and Hughes 1984; Mackauer and Kambhampati 1984; Sequeira and Mackauer 1988; Tang and Yokomi 1996; He et al. 2003; Lin and Ives 2003). The majority of these st udies indicate that aphids parasitised in their early instars die before reproduction but those parasitised at later stages of development may reach the adult stage and produce a limited number of progeny before mummification. Most aphi d populations consist of multiple stages of development. Therefore, knowledge of th e effect of parasitism on development, reproduction and population growth of aphids of different ages is critical to the success of biological control (Tsai and Wang 2002; Zhang and Hassan 2003). Several authors (Pennacchio et al. 1995; Digilio et al. 1998; Rahbé et al. 2002) have studied biochemical changes in pea aphid caused by A. ervi. However, little is known about how parasitism by A. ervi affects the population of the pea aphid, making it difficult to evaluate the biological co ntrol efficiency of this parasitoid. This section investigated the effect of parasitism by A. ervi on the development, reproduction and population growth of pea aphid at different stages of development. 3.4.2 Materials and Methods 3.4.2.1 Development and Reproduction The effect of parasitism by A. ervi on the growth and reproduction of pea aphids at different stages of development (1~10 d old = 1 st instar to adult, Table 3.1) was investigated. To obtain parasitised aphids of the same age class, a mated female was released into a Petri dish (8.5 cm in diameter × 1.3 cm in height) containing 10 aphids of the same age. Aphids that received a single oviposition sting from the female parasitoid were removed and replaced with unparasitised individuals of the same age until 15 parasitised aphids were collected. Parasitised aphids were placed . General Biology of Aphidius ervi 38 individually onto a bean plant cutting which was held in a transparent plastic cylinder (Figure 3.2). There were 10 treatments (age classes) and 15 replicates (number of parasitised aphids) for each treatment in this study. Survival, development and reproduction of parasitised aphids were monitored at 24 h intervals. Any offspring from these aphids were counted and removed. In addition, the survival and reproductive periods of aphids were also recorded. Fifteen unparasitised aphids (start at 1 day old) reared individua lly in cylinders served as controls and their survival and reproduction were monitored as mentioned above. 3.4.2.2 Population Growth The daily survival rate and reproduction of aphids were compiled into a life table to assess the population growth according to the method of Jervis et al. (2005). The intrinsic rate of increase (female/female/day; r m) was estimated by solving the Lotka-Euler equation ( ∑e-r mxlxmx = 1). Other calculations included the net reproductive rate (females/females/generation; R 0 = ∑ lxmx) , mean generation time (days; T = loge(R 0 )/r m) and doubling time (days; DT = log e(2)/r m). In these calculations x is the pivotal age, l x is the proportion of the females surviving to age x , and m x is the number of offspring produced per female at age x. For each treatment and the control, a jackknife method (Caswell 2001) was used to estimate the means and standard errors of the above life table parameters by partitioning the 15 parasitised aphids into 5 groups of 3 individuals. 3.4.2.3 Statistical Analysis Linear regression analysis was used to identify the relationships between aphid age at the point of parasitism, repr oductive period and the number of progeny produced. A logistic regression was applied to analyse the relationships between aphid age ( x) at the point of parasitism and r m [i.e. r m = c/(1+exp(a+b*ln( x ) ) ) ] , where the constant c represents the maximum possible value of r m; and a and b are constants to fit the curve. All other data were normally distributed and analysed using ANOVA followed by a Tukey’s studentised range test. . General Biology of Aphidius ervi 39 3.4.3 Results 3.4.3.1 Development and Survival Aphids parasitised at 1 and 2 d old (1 st and 2 nd instar) became mummified in the 4th instar. However, aphids parasitised at 3 to 6 d old (3 rd and 4 th instar) continued development to the adult stage. For all ages, paras itised aphids died about 7 days after being parasitised and their longevity was significantly shorter than unparasitised aphids (ANOVA: F = 419.15; df = 10,154; P < 0.0001) (Table 3.1). Table 3.1 Effect of parasitism by A. ervi on survival (days), re productive period (days) and number of progeny of pea aphid. Aphid age at parasitisation (days) Longevity (days) Reproductive period (days) No. of progeny 1 (1 st instar) 8.13 ± 0.09 j --- --- 2 (2 nd instar) 9.13 ± 0.09 ij --- --- 3 (3 rd instar) 10.20 ± 0.11 hi --- --- 4 (3 rd instar) 11.13 ± 0.09 gh 2.07 ± 0.15 g 4.93 ± 0.43 g 5 (4 th instar) 12.07 ± 0.07 fg 2.33 ± 0.16 fg 10.07 ± 0.90 g 6 (4 th instar) 13.13 ± 0.09 ef 3.13 ± 0.13 f 19.40 ± 1.13 f 7 (adult) 14.07 ± 0.07 de 4.20 ± 0.18 e 30.47 ± 1.48 e 8 (adult) 15.13 ± 0.09 cd 5.20 ± 0.11 d 39.07 ± 1.49 d 9 (adult) 16.20 ± 0.11 c 6.33 ± 0.19 c 52.33 ± 0.88 c 10 (adult) 17.47 ± 0.17 b 7.60 ± 0.19 b 62.60 ± 1.92 b Control 27.87 ± 0.82 a 15.60 ± 0.38 a 111.20 ± 3.17 a Means ( ± SE) followed by the same letters in columns are not significantly different (P > 0.05). 3.4.3.2 Reproduction Only those aphids parasitised when ≥ 4 d old (late 3 rd instar) produced progeny. Most of these aphids started reproducti on from 8 d old. Parasitised aphids had a significantly shorter reproductive period and produced significantly fewer progeny . General Biology of Aphidius ervi 40 than healthy aphids (ANOVA: F = 483.15 and 452.23 for reproductive period and number of progeny, respectively; df = 7,112; P < 0.0001) (Tab le 3.1). For the parasitised aphids, the reproductive peri od and the number of progeny produced increased significantly with the age at parasitisation (Linear regression: F = 845.95 and 155.17 for reproductive period and number of progeny, respectively; df = 1,103; P < 0.0001) (Figures 3.7 and 3.8, respectively). y = 0.9524x - 2.2571 R 2 = 0.8913 0 2 4 6 8 1 0 3 4 5 6 7 8 9 10 Aphid age (days) at parasitisation R ep ro du ct iv e pe ri od (d ay s). Figure 3.7 Relationship between host age at parasitisation and reproductive period. y = 9.90x - 38.03 R 2 = 0.9379 0 20 4 0 6 0 8 0 1 0 0 3 4 5 6 7 8 9 10 Aphid age (days) at parasitisation N o. o f p ro ge ny Figure 3.8 Relationship between host age at parasitisation and reproduction of progeny. . General Biology of Aphidius ervi 41 3.4.3.3 Population Growth Parasitism by A. ervi significantly affected the population growth of pea aphids, reflected in changes in the intrinsic rate of increase (r m) , net reproductive rate (R 0 ), generation time (T), and doubling time (DT) (Table 3.2). Aphids parasitised as immature stages (≤ 6 d old) had a significantly lower r m and R 0 , shorter T, and longer DT than those parasitised as a dults and healthy aphids (ANOVA: F = 206.33, 520.21, 202.18 and 46.63 for r m , R 0 , T and DT, respectively; df = 7,32; P < 0.0001). The r m increased rapidly with the increase of aphids’ age when they were parasitised and reached a plateau at 9 and 10 d old (Logistic regression: F = 688.84; df = 2,33; P < 0.0001) (Figure 3.9). Table 3.2 Effect of parasitism by A. ervi on life table parameters of pea aphid. Aphid age at para- sitization (days) r m R 0 T DT 4 (3 rd instar) 0.1799 ± 0.0083 e 4.93 ± 0.39 g 8.71 ± 0.09 d 3.95 ± 0.21 a 5 (4 th instar) 0.2573 ± 0.0050 d 9.13 ± 1.17 g 8.86 ± 0.07 d 2.94 ± 0.35 b 6 (4 th instar) 0.3398 ± 0.0075 c 19.40 ± 1.18 f 8.69 ± 0.02 d 2.05 ± 0.05 c 7 (adult) 0.3705 ± 0.0056 b 30.47 ± 0.85 e 9.21 ± 0.07 c 1.87 ± 0.03 d 8 (adult) 0.3924 ± 0.0072 ab 39.07 ± 1.53 d 9.33 ± 0.07 c 1.77 ± 0.03 d 9 (adult) 0.3972 ± 0.0031 a 52.07 ± 0.67 c 9.95 ± 0.05 b 1.74 ± 0.01 d 10 (adult) 0.3979 ± 0.0033 a 62.60 ± 2.52 b 10.29 ± 0.06 b 1.73 ± 0.01 d Control 0.4018 ± 0.0023 a 113.00 ± 2.51 a 11.89 ± 0.13 a 1.74 ± 0.02 d Means ( ± SE) followed by the same letters in columns are not significantly different (P > 0.05). . General Biology of Aphidius ervi 42 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 3 4 5 6 7 8 9 10 Aphid age (days) at parasitisation r m rm = 0.4139/(1 + Exp(6.3283 - 4.3277*ln(x))) R 2 = 0.9694 Figure 3.9 Relationship between host age at parasitisation and rm. 3.4.4 Discussion Results of this study show that, in comparison with th e unparasitised pea aphid controls, parasitism by A. ervi significantly reduced pea aphids’ survival period, reproductive potential and population growth. However, when compared to Praon pequodorum Viereck (Sequeira and Mackauer 1988), another parasitoid of pea aphid, A. ervi has greater potential to suppress aphid population because those parasitised by P. pequodorum produce 20 to 30 more progeny than those parasitised by A. ervi. The effect of parasitism by A. ervi on pea aphid reproduction is similar to that by A. eadyi (He et al. 2003). The efficiency of parasitoids in the control of aphids depends on their ability to suppress aphids’ population growth. Aphids parasitised by A. ervi continued to feed and grow until they were killed by developing parasitoid larvae. Because the time from being parasitised to mummification is about 7 days for all parasitised aphids, their survival and reproduction depend on the time when they are parasitised. Aphids ≤ 3 d old at parasitisation died at the fourth instar or early adult stage. In contrast, those parasitised when ≥ 4 d old were able to reach the adult stage and reproduce. This may be because the aphid embryos in the later instar and adult stages have developed a resistant cuticle by the time of oviposition by the parasitoid (Polaszek 1986) and have thus escaped consumption by the developing parasitoid larvae. . General Biology of Aphidius ervi 43 Results also indicate that development, survival, reproduction and population growth of parasitised aphids vary with their age at the time of parasitism. It appears that attacking young hosts will be more e ffective in suppressing aphid populations than attacking the adult aphids. According to Schowalter (2000), ear ly in the season, growing populations have a high proportion of aphids in young age classes. Thus, the application of A. ervi early in the season could significantly suppress populations of pea aphid in the field, inhibiting its dispersal and population build-up later in the season. . Reproductive Fitness of Aphidius ervi 44 CHAPTER 4 FACTORS AFFECTING REPRODUCTIVE FITNESS OF APHIDIUS ERVI 4.1 Introduction The effectiveness of potential biological control agents greatly depends on their fecundity, longevity and sex allocation strategy which are affected by host quality (i.e. age or size) and density, a nd parasitoid body size and age of both sexes (Jervis et al. 2005). Understanding how para sitoids optimise their reproductive output in response to these factors can improve pest control efficacy in biological control programmes. Thus, this chapter investigated how these factors affect the reproductive fitness of A. ervi. 4.2 Host Age Preference and its Effect on Reproductive Fitness of A. ervi 4.2.1 Introduction Many studies have demonstrated that parasitism can influence development, fecundity, and population growth of the host aphids (Tsai and Wa ng 2002; He et al. 2003; Lin and Ives 2003). Generally, aphi ds parasitised in early instars are mummified before reaching reproductive maturity, whereas those parasitised in late instars are able to reach adult stage and produce progeny. Therefore, parasitoids’ impact on the aphid population growth largely depends on the pattern of the host age selected for parasitisation (Hågvar and Hofsvang 1991; Tsai and Wang 2002) and host age preference is critical to the success of biological control of aphids (Tsai and Wang 2002). Foraging parasitoids usually encounter ho sts of different ages or sizes and have opportunities to select the most suitable hosts to maximize their reproductive fitness. For some solitary species of parasitoids, hosts selected for oviposition are determined by host size (Kouamé and Mackauer 1991) because large hosts contain . Reproductive Fitness of Aphidius ervi 45 more resources for parasitoid progeny development than small hosts (Charnov et al. 1981; Liu 1985). Many studies suggest that host size prefer ence by parasitoids affects progeny fitness of parasitoids, such as the body size (Liu 1985; Lampson et al. 1996) and egg load at emergence (Liu 1985; Mills and Kuhlmann 2000). Host age may also affect sex allocation of para sitoids (Godfray 1994). Accordin g to Charnov et al. (1981) and Charnov (1982), parasitoids can make effi cient use of the size variation in the hosts encountered by allocating fertilized diploid eggs to large hosts and unfertilized eggs to small ones. Previous studies have demonstrated that the impact of A. ervi on population growth of the pea aphid largely depends on the host age at parasitism (Section 3.4), and that body size of pea aphid is positively correlated with its age (Sequeira and Mackauer 1992). However, so far little is known about the host age preference by A. ervi. Knowledge of host preference would lead to a better understanding of the population dynamics of the host and parasitoid (Nechols a nd Kikuchi 1985). Therefore, to provide useful informati on for the development of biology control strategies, this section investig ated the host age preference by A. ervi and its effect on the reproductive fitness and sex allocation. 4.2.2 Materials and Methods 4.2.2.1 Experimental Insects Parasitoid adults used for the experiments emerged from pea aphid mummies parasitised at the third instar (3 d old). 4.2.2.2 Preference Behaviour To observe the host age preference behaviour, 30 mated females (< 12 h old) were individually released into a closed Petri dish (5.5 cm in diameter × 1.3 cm in height) containing seven healthy aphids of different age from 1 to 7 d old. The oviposition behaviour was observed for 10 minutes and recorded using Panasonic SVHS Camcorder (MS-4) (Panasonic, Japan) (see Section 5.2.1). Behavioural . Reproductive Fitness of Aphidius ervi 46 interactions between hosts and parasitoids were defined as following (Gerling et al 1990; Chau and Mackauer 2001): (a) Encounter – a female met an aphid; (b) Non-response – a female walked away from an aphid after encounter; (c) Non-response rate – percentage of non-response per encounter; (d) Attack attempt – a female exposed he r ovipositor and tried to attack an aphid; (e) Sting – a female inserted her ovipos itor into an aphid regardless of successful oviposition or not; (f) Sting rate – percentage of s ting per attack attempt; (g) Oviposition – egg(s) laid into an aphid detected by dissecting 4 days later; (h) Oviposition rate – percentage of laying egg(s) per sting; (i) Escape – an aphid walked away to avoid being attacked by a female; (j) Kick – an aphid rised its legs and swayed its abdomen abruptly to prevent from being attacked by a female; (k) Searching time – the period of a fema le walking and searching for hosts; (l) Resting time – the period of a female staying for more than 5 seconds to clean her antennae and legs after performing attack(s); (m) Handling time – the period between encounter and attack by a female regardless of successful oviposition or not. Aphids from each dish were separated soon after observation and individually placed on a bean plant cutting. These aphids were dissected four days after oviposition to detect the number of eggs laid. 4.2.2.3 Effect of Age Preference on Reproductive Fitness A positive relationship was detected between aphids’ age (1 to 7 d old) and their body size (i.e. body weight, mg) (R A: body weight = - 0.0976 + 0.2917age, R 2 = 0.7365, F 1,5 8 = 162.09, p < 0.0001). To determine host age preference by A. ervi in relation to host size and its effect on A. ervi reproduction, mated females were supplied with host aphids of seven different ages from 1 to 7 d old. Ten females (replicates) were tested in this experime nt. For each replicate, one mated female . Reproductive Fitness of Aphidius ervi 47 parasitoid (< 12 h old) was introduced into an above-mentioned plastic cylinder (Figure 3.2) with 105 healthy aphids (15 aphids of each age class) feeding on a bean plant cutting. The female was allowed to stay in the cylinder for 24 h, and then moved to another cylinder with 105 healthy aphids of the same age structure, etc. until she died. After the removal of the female parasitoid, aphids of different age classes were separated and transferred to an uninfested bean plant in a cylinder. To estimate the number of eggs laid, five aphids of each age class were randomly selected from each cylinder 4 days after the removal of the female parasitoid, and dissected in 70% alcohol under a stereomicroscope. The number of eggs laid in an aphid was recorded (Bueno et al. 1993). The remaining aphids were reared until mummification. The emerged progeny were counted and sexed. To determine the effect of host age (p reference) on reproductive fitness of A. ervi , the head width (a measure of body size) of 30 newly emerged progeny of each sex randomly selected from each host ag e class was measured using the above stereomicroscope. The egg load of parasitoid females at emergence was determined by dissecting 30 newly emerged females random ly selected from each host age class. They were dissected in 70% alcohol on a s lide under the above stereomicroscope. One drop of acid fuchsin was added to the alcohol and allowed to stand for 3 to 5 min for staining. The number of eggs in the ovaries was counted under a compound microscope. 4.2.2.4 Statistical Analysis A goodness-of-fit test was used to test the distribution of data before analysis. Data of egg load were not normally distri buted even after transformation and thus analysed using the nonparametric Kruskal-Wallis test followed by Dunn’s procedure for multiple comparisons (Zar 1999). Other data were normally distributed and analysed using ANOVA followed by a Tukey' s studentized range test. Data of proportion of female offspring were arcsine transformed prior to analysis. The Paired t-test was applied to analyse the relationship between th e number of encounters and . Reproductive Fitness of Aphidius ervi 48 attack attempts. The relati onship between body size and eg g load of newly emerged females was analysed using regression analysis (RA). 4.2.3 Results 4.2.3.1 Preference Behaviour Analysed data of host age preference and oviposition behaviour are summarised in Table 4.1. In most cases, searching females switched from random walking to attack position when the host was within a range of 1 cm. Females usually attacked hosts without antennal examinati on. They were significantly more likely to encounter and attack 5- to 7-d-old aphids than 1- and 2-d-old aphids (Kruskal-Wallis test: H = 49.88 and 58.06 > 2 05.0,6 χ = 12.59, for encounter and attack attempt, respectively; df = 6,203; P < 0.0001) (Table 4.1). Encounter did not always el icit attack, the mean number of attack attempts (3.71 ± 0.72) was significantly lower than th at of encounters (5.47 ± 0.95) (t-test for paired two samples: t = 7.12; df = 6; P < 0.001). A parasitoid might not respond to a host and the host might escape from attack ing by walking away. The number of non- response was not significantly different between hosts of different age (Kruskal- Wallis test: H = 1.82 < 2 05.0,6 χ ; df = 6,203; P > 0.05), but the probability of parasitoids that did not respond a 1-d-old aphid upon enc ounter was significantly greater than 6- and 7-d-old aphids (Kruskal-Wallis test: H = 18.59 > 2 05.0,6 χ ; df = 6,189; P < 0.01). Six- and 7-d-old aphids were significantly more likely than 1- to 4-d-old aphids to escape from parasitoids’ attack (Kruskal-Wallis test: H = 54.67 > 2 05.0,6 χ ; df = 6,203; P < 0.0001) (Table 4.1). Older aphids had a greater ability to defend themselves by kicking during attack by a parasitoid (Kru skal-Wallis test: H = 89.55 > 2 05.0,6 χ ; df = 6,181; P < 0.0001). Parasitoids stung signif icantly more older aphids (5 to 7 d old) than younger ones (1 and 2 d old) (Kruskal-Wallis test: H = 24.35 > 2 05.0,6 χ ; df = 6,203; P < 0.0001); however, the sting rate was significantly hi gher in younger aphids (1 to 4 d old) than . Reproductive Fitness of Aphidius ervi 49 in older ones (6 and 7 d old) (Kruskal-Wallis test: H = 56.18 > 2 05.0,6 χ ; df = 6,178; P < 0.0001). The fact that the number of stings wa s greater that the number of eggs laid indicated that a female might deposit no egg in a host when stung, and the oviposition rate was significantly higher in 1- and 2-d-ol d aphids than in 7-d-old aphids (Kruskal- Wallis test: H = 18.95 > 2 05.0,6 χ ; df = 6,186; P < 0.0001). However, among hosts of different ages, the number of eggs laid was significantly higher in 4- and 5-d-old aphids (Kruskal-Wallis test: H = 25.15 > 2 05.0,6 χ ; df = 6,203; P < 0.0001) (Table 4.1). Parasitoids allocated significantly longer time in host searching (Mean ± SE: 6.19 ± 0.35 mins) than in resting (3.01 ± 0.37 ) and the latter was significantly longer than handling time (0.80 ± 0.07) (Kruskal-Wallis test: H = 58.00 > 2 05.0,2 χ ; df = 2,87; P < 0.0001). Parasitoids had significant longer ha ndling time when attacking the 5- to 7- d-old aphids (Kruskal-Wallis test: H = 85.67 > 2 05.0,6 χ ; df = 6,203; P < 0.0001) (Table 4.1). . Reproductive Fitness of Aphidius ervi 50 Table 4.1 Oviposition behaviour of A. ervi females in response to host ages. 1 d old 2 d old 3 d old 4 d old 5 d old 6 d old 7 d old No. of encounters 2.83 ± 0.37 c 2.60 ± 0.39 c 3.76 ± 0.38 bc 3.83 ± 0.41 bc 5.50 ± 0.61 ab 6.03 ± 0.59 ab 9.03 ± 1.23 a No. of attack attempts 1.30 ± 0.22 c 1.47 ± 0.25 c 2.97 ± 0.36 bc 2.90 ± 0.37 bc 3.90 ± 0.41 ab 4.50 ± 0.67 ab 7.97 ± 1.41 a No. of escapes 0.10 ± 0.07 c 0.00 ± 0.00 c 0.07 ± 0.05 c 0.17 ± 0.07 bc 0.77 ± 0.21 ab 1.23 ± 0.24 a 1.97 ± 0.34 a No. of kicks 0.35 ± 0.12 d 0.64 ± 0.17 cd 0.86 ± 0.17 cd 1.81 ± 0.28 bc 2.52 ± 0.30 ab 3.17 ± 0.35 ab 5.21 ± 0.78 a No. of non-responses 1.33 ± 0.26 a 1.33 ± 0.34 a 1.07 ± 0.17 a 1.30 ± 0.24 a 0.97 ± 0.19 a 0.93 ± 0.21 a 1.07 ± 0.22 a Non-response rate 0.44 ± 0.07 a 0.38 ± 0.07 a 0.29 ± 0.04 ab 0.31 ± 0.06 ab 0.17 ± 0.03 b 0.16 ± 0.04 b 0.14 ± 0.03 b No. of stings 1.30 ± 0.21 b 1.33 ± 0.21 b 2.50 ± 0.31 ab 2.13 ± 0.31 ab 2.83 ± 0.30 a 2.80 ± 0.28 a 3.20 ± 0.61 a Sting rate 0.97 ± 0.02 a 0.95 ± 0.03 a 0.89 ± 0.04 a 0.79 ± 0.05 ab 0.75 ± 0.05 ab 0.63 ± 0.05 bc 0.45 ± 0.05 c No. of eggs laid 0.93 ± 0.16 c 1.13 ± 0.13 bc 1.77 ± 0.23 ab 2.00 ± 0.21 a 1.90 ± 0.24 a 1.43 ± 0.21 ab 1.13 ± 0.20 bc Oviposition rate 0.81 ± 0.07 a 0.82 ± 0.06 a 0.69 ± 0.07 ab 0.70 ± 0.06 ab 0.63 ± 0.06 ab 0.58 ± 0.07 ab 0.46 ± 0.07 b Handling time (sec.) 2.27 ± 0.54 c 1.50 ± 0.26 c 4.53 ± 0.61 bc 4.67 ± 0.65 bc 6.63 ± 0.67 ab 11.07 ± 1.74 ab 17.33 ± 2.81 a Means (± SE) followed by the same letters within rows were not significantly different (P > 0.05). Table 4.2 Mean (± SE) head width (mm) of newly emerged A. ervi when parasitised hosts of different ages. 1 d old 2 d old 3 d old 4 d old 5 d old 6 d old 7 d old Male 0.535 ± 0.0034 d 0.556 ± 0.0032 c 0.580 ± 0.0031 b 0.601 ± 0.0037 a 0.599 ± 0.0045 a 0.596 ± 0.0049 ab 0.598 ± 0.0051 a Female 0.585 ± 0.0027 d 0.613 ± 0.0032 c 0.642 ± 0.0031 b 0.660 ± 0.0037 a 0.656 ± 0.0029 ab 0.646 ± 0.0054 ab 0.649 ± 0.0059 a b Means (± SE) followed by the same letters within rows were not significantly different (P > 0.05). . Reproductive Fitness of Aphidius ervi 51 4.2.3.2 Effect of Age Preference on Reproductive Fitness Aphidius ervi females significantly preferred aphids that were 3 to 5 d old for oviposition with significantly higher number of aphids parasitised (Figure 4.1) and eggs laid (Figure 4.2) (ANOVA: F = 35.5 and 9.95 for the number of aphids parasitised and eggs laid, resp ectively; df = 6,63; P < 0.0001). 0 20 40 60 80 1 0 0 1 2 3 4 5 6 7 Host age (days) A ph id s pa ra si tis ed a a ab ab bc cd d Figure 4.1 Mean (± SE) number of aphids parasitised per A. ervi female in relation to host age. Columns with the same letters are not significantly different (P > 0.05). 0 5 10 15 20 2 5 3 0 1 2 3 4 5 6 7 Host age (days) E gg s la id a a abbc bc bcc Figure 4.2 Mean (± SE) number of eggs laid per A. ervi female (estimated from dissection) in relation to host age. Columns with the same letters are not significantly different (P > 0.05). . Reproductive Fitness of Aphidius ervi 52 Host age had significant effect on proportion of female progeny produced (ANOVA: F = 35.4; df = 6,63; P < 0.0001), whic h was highest in 6-d-old aphids and lowest in 1-d-old aphids (Figure 4.3). 0 20 40 60 80 1 0 0 1 2 3 4 5 6 7 Host age (days) F em al e pr og en y % aab bbb c c Figure 4.3 Mean (± SE) proportion of female progeny emerged from hosts of different age classes. Columns with the same letters are not significantly different (P > 0.05). The head width of newly emerged adults of both sexes significantly increased with the increasing host age (from 1 to 4 d old) at the time of parasitisation (ANOVA: F = 40.18 and 45.80 for male and female, resp ectively; df = 6,203; P < 0.0001) (Table 4.2). Regardless of the host age when parasi tised, females were si gnificantly larger than males (ANOVA: F = 130.91, 161.95, 200.13, 125.96, 114.15, 46.56 and 42.18 for 1- to 7-d-old aphids, respectively; df = 1,58; P < 0.01) (Table 4.2). The egg load of newly emerged females was also significantly higher when parasitoids attacked aphids that were 3 to 7 d old (ANOVA: F = 49.29; df = 6,203; P < 0.0001) (Figure 4.4). There was a positive linear relationshi p between the body size and egg load of newly emerged females (RA: F = 220.13; df = 1,208; P < 0.0001) (Figure 4.5). . Reproductive Fitness of Aphidius ervi 53 0 20 40 60 8 0 1 2 3 4 5 6 7 Host age (days) E gg lo ad a a a a a b b Figure 4.4 Mean (± SE) number of egg load pe r female progeny emerged from hosts of different age classes. Columns with the same letters are not significantly different (P > 0.05). y = 427.45x - 220.10 R 2 = 0.5142 0 20 4 0 6 0 8 0 1 0 0 1 2 0 0.55 0.6 0.65 0.7 0.75 Female head width (mm) E eg g lo ad Figure 4.5 Relationship between body size and number of egg load per newly emerged female of A. ervi. 4.2.4 Discussion Aphidius ervi females attacked pea aphids of different ages. Once aphids are detected, hymenopteran parasitoids are usua lly able to evaluate host species and quality by antennal examination (Desneux et al. 2004; Mehrnejad and Copland 2006) , allowing them to verify the presence of the physiological conditions necessary for the development and growth of their progeny. However, A. ervi females usually performed attack position wit hout any physical contact. It has been reported that the visual cues (colour and shape of pea a phid) play important roles in the host recognition and acceptance in A. ervi (Battaglia et al. 1995, 2000). The behaviour . Reproductive Fitness of Aphidius ervi 54 switching from random searching to attack position suggests that A. ervi females recognise hosts by visual and/or odour cues before physical contact when the host is within a range of 1 cm. Although it was found that the parasitoid was interested in attacking larger hosts with more attack atte mpts, the larger aphids were more capable of physically defending themselves from parasitisation (i.e. ‘escape’ when encountered and ‘kicking’ when attacked). Therefore, larger hosts would cost more to the A. ervi females in oviposition (i.e. lower sting and oviposition rates and longer handling time). Similar cases are also reported in Monoctonus paulensis (Ashmead) attacking pea aphid (Chau and Mackauer 2001) and in Psyllaephagus pistaciae Ferrière parasitising the common pistachio psylla Agonoscena pistaciae Burckhardt and Lauterer (Mehrnejad and Copland 2006). The immature pea aphids grow in size as they develop to adult stage and larger aphids have more food resource for A. ervi larvae (Li et al. 2002; Section 3.4). According to the theory of optimal host acceptance in parasitoids, which is based on the model of optimal diet in predators (Stephens and Krebs 1986), A. ervi females should prefer larger aphids for oviposition. However, the present results indicate that A. ervi preferred aphids that were 3 to 5 d old to younger (1 to 3 d old) and older ones (6 and 7 d old) for oviposition. Therefore, the food resource at the point of oviposition is not the only factor that determines the oviposition decision by parasitoids, and they may select hosts according to the optimal tradeoff between food supply and oviposition costs. The results in this study also show that host age preference by A. ervi is related to its progeny’s reproductive fitness. The body size of progeny and egg load of newly emerged females were similar when aphids were attacked at 4 to 7 d old, suggesting that the quantity and quality of food resource for the parasitoid progeny are similar in aphids of this age range. The increase in proportion of female progeny of A. ervi with host size supports the host size-dependent sex al location theory in parasitic Hymenopter a (Charnov et al. 1981; Charnov 1982). There is also field evidence that the host age plays a role in sex . Reproductive Fitness of Aphidius ervi 55 ratios, as reported for A. ervi parasitising pea aphid, wh ere a male-biased population emerged from small aphids in the early season (Sequeira and Mackauer 1993). Although the body size of parasitoids emer ged from aphids parasitised at 4 d old was similar to that emerged from aphids parasitised at 6 d old, the proportion of female offspring was significantly lower when parasitoids selected 4-d-old aphids for oviposition than that when parasitoids selected 6-d-old a phids. Sex ratio data suggest that the body size of hosts at the point of oviposition is the major factor affecting the sex allocation by A. ervi. The findings of this study have implications for laboratory mass rearing and field release of A. ervi. For example, aphids that are 3 to 5 d old appear to be the most appropriate hosts in the mass-rearing progr am because they gave the best fitness return for the parasitoids. . Reproductive Fitness of Aphidius ervi 56 4.3 Effect of Host Density on Reproductive Fitness of A. ervi 4.3.1 Introduction The impact of a parasitoid on its host population greatly depends upon its ability to find and parasitise hosts and to increase offspring numbers in response to increasing host density (Waage and Hassell 198 2; Mackauer 1983). It is important to identify the form of the parasitoid response to prey density in population modeling, from which biological control programmes can be developed (Mi lls and Lacan 2004). In analytical host-parasitoid models, cha nges in the density-de pendent sex ratio of parasitoids influence the level of host population equilibrium and the stability of the host-parasitoid relationships, thus affecti ng the success of biological control (Waage and Hassell 1982; Hassell and Waage 1984). Ives et al. (1999) studied some aspects of the functional response of A. ervi on pea aphid, with emphasis on the parasitoid’s behavioural decisions and their influence on the parasitism rate. In this section, the relationship between host density and reproductive fitness of A. ervi was investigated by determining how host density affected parasitoid reproductive potential and sex allocation. The aim was to provide further information for the assessment and improvement of A. ervi ’s effectiveness in biological control. 4.3.2 Materials and Methods 4.3.2.1 Experimental Insects All parasitoids used for experiment s emerged from mummies that were parasitised in the third instar (3 d old), and the third inst ar pea aphids were used as hosts in all experiments. 4.3.2.2 Experiment To determine whether and to what exte nt host density affected reproduction of A. ervi , six densities of aphids (15, 25, 50, 75, 100 and 125 a phids/cylinder/parasitoid female/day) were tested, with 10 re plicates for each density. One mated A. ervi female . Reproductive Fitness of Aphidius ervi 57 (< 12 h old) was introduced into an expe rimental cylinder with aphids feeding on a bean plant cutting. The female was allowed to stay in the cylinder for 24 h, and then moved to another cylinder with the same number of healthy third instar aphids, etc. until she died. Because superparasitism was comm on under laboratory conditions, the oviposition potential of A. ervi was estimated by counting both the number of eggs laid (fecundity) and aphids parasitised (par asitism). To estimate the daily number of eggs laid in a parasitised aphid, five aphi ds from each cylinder at densities of 15 and 25 aphids, 10 aphids at dens ity of 50 aphids, and 20 aphi ds at densities of 75, 100 and 125 aphids were randomly selected from each cylinder 4 days after the removal of the female parasitoid. These selected aphids were dissected in 70% alcohol under the stereomicroscope. The numbers of parasitoid larvae recorded from dissecting were assumed equal to the number of eggs laid (Bueno et al. 1993). The remaining aphids were reared until mummification. The number of eggs laid was assumed equal to the total number of aphids parasitised (i.e. th e sum of mummies and aphids parasitised detected by dissecting) × the average numbe r of eggs laid in a parasitised aphid detected by dissecting. The emerged offspring were counted and sexed. 4.3.2.3 Statistical Analysis Goodness-of-fit test was us ed to test whether the data were normally distributed. Effect of aphid density on the number of parasitoid eggs laid was assessed by ANOVA and means separated by Tukey’s st udentised range test. Data for percent female progeny, parasitism rate and eggs per parasitised aphid were not normally distributed after transformation and were thus analysed using the nonparametric Kruskal-Wallis test. Means were subseque ntly separated by Dunn’s procedure for multiple comparisons (Zar 1999). A signifi cance level of P < 0.05 was used for all tests. In this study, more than 80% of eggs were laid within the first 8 days, thus all data presented were from the first 8 days. . Reproductive Fitness of Aphidius ervi 58 4.3.3 Results Mean total number of a phids parasitised and eggs laid (actual fecundity) during A. ervi females’ life time increased with the increase of host density (Kruskal- Wallis test for parasitism: F = 68.39, df = 5,64, P < 0.0001; ANOVA for eggs laid: F = 14.65, df = 5,64, P < 0.0001) (Figure 4.6). Fec undity reached a plateau at host densities of 75/cylinder and above (Figure 4.6). The number of e ggs laid in each parasitised aphid decreased with the increase in host density from 15 to 75/cylinder, after which no further decrease occurred [Kruskal-Wallis test: H = 51.96 > 2 05.0,5 χ ( = 11.07); P < 0.0001] (Figure 4.7). The daily para sitism rate decreased when the host density increased to 50/cylinder (Kruskal-Wallis test: H = 44.59 > 2 05.0,5 χ ; P < 0.0001) (Figure 4.8). The proportion of female prog eny was greatest at host densities of 50 and 75/cylinder, and least at extremes of low and high host densities (Kruskal-Wallis test: H = 35.30 > 2 05.0,5 χ ; P < 0.0001) (Figure 4.9). 0 100 200 30 0 4 0 0 5 0 0 6 0 0 7 0 0 Aphids parasitised Eggs laid N o. o f a ph id s pa ra si tis ed a nd e gg s la id aa b b c c a a ab bc cd d 15 25 50 75 100 125 15 25 50 75 100 125 No. aphids/cylinder Aphids parasitised Eggs laid Figure 4.6 Mean (± SE) number of aphids parasitised and eggs laid by A. ervi at different host densities (15, 25, 50, 75, 100 and 125 aphids/cylinder/day/female). Within the same category (Aphi ds parasitised or Eggs la id) columns with the same letters are not significantly different (P > 0.05). . Reproductive Fitness of Aphidius ervi 59 0 1 2 3 4 15 2 5 5 0 7 5 1 0 0 1 2 5 No. aphids/cylinder N o. o f e gg s/ pa ra si tis ed a ph id a b ab bc bc c Figure 4.7 Mean (± SE) number of eggs per parasitised aphid laid by A. ervi at different host densities. Columns with the same letters are not significantly different (P > 0.05). 0.4 0.6 0.8 1 15 25 50 75 100 125 No. aphids/cylinder M ea n da ily p ar as iti sm r at e a a b bc bc c Figure 4.8 Mean (± SE) daily parasitism rate of A. ervi at different host densities. Columns with the same letters are not significantly different (P > 0.05). . Reproductive Fitness of Aphidius ervi 60 20 30 40 50 60 70 80 15 25 50 75 100 125 No. aphids/cylinder M ea n fe m al e pr og en y% a a ab ab b b Figure 4.9 Mean (± SE) proportion of female progeny of A. ervi at different host densities. Columns with the same letters are not significantly different (P > 0.05). 4.3.4 Discussion High fecundity and parasitism rates are considered necessary for parasitoids to be able to respond rapidly to the increases in pest density (Waage and Hassell 1982; Waage 1990). The results of the present study indicate that A. ervi can adjust oviposition strategy in response to increasing host density through increasing parasitism and decreasing superparasitism (Figures 4.6 and 4.7). These data suggest that this parasitoid has higher potential to suppress the aphid population when the latter increases. However, when the host population was 50/cylinder, mean daily parasitism rate significantly decreased (Figure 4.8). The behaviour of A. ervi in this experiment is consistent with the classic Holling type II functional response (Holling 1959), i.e. the number of aphids parasitise d increased with an increase in aphid density but at a progressively decreasing rate. A type II functional response was also found in some other Aphidius species, such as the A. smithi (Mackauer 1983) and A. sonchi Marshall (Liu 1985). Results of this study suggest that a host density equivalent to 50 aphids/cylinder is the highe st critical density where the parasitoid could maximise its control efficiency. Th is information could be important when considering host-parasitoid density ratio in biological control programmes (Hassell and Waage 1984). For example, to maximise biological control efficiency of pea aphids, parasitoids could be released at a rate of about one female parasitoid/50 aphids. . Reproductive Fitness of Aphidius ervi 61 The proportion of female progeny developed from fertilised eggs quickly and increased significantly with the increase of host density from 15 to 50 aphids/cylinder, after which it gradually declined. This suggests that the highest potential proportion of female progeny that A. ervi can produce is about 70% and sperm limit occurs when host density reaches 50 to 75 aphids/cyl inder. Offspring sex ratio of other Aphidius species, such as A. smithi (Mackauer 1983) and A. sonchi (Liu 1985), has also been found to vary with host density. In hymenopter an parasitoids, males can mate several times while females mate only once. It is thus to the female’s benefit that she produces the minimum number of males possible. However, when hosts occur at low density, a highly female-biased sex ratio sh ould reduce the ability of parasitoid persistence (Tripathi and Singh 1991), leading to extinction of th e host population and eventually the parasitoid population. This suggests that the female parasitoids can adjust the sex allocation strategy in response to the host density. A similar case was found in Aphelinus mali Haldeman, which attacks the woolly apple aphid (Mueller et al. 1992). . Reproductive Fitness of Aphidius ervi 62 4.4 Effect of Body Size on Reproductive Fitness of A. ervi 4.4. 1 Introduction Body size of insects has usually been cons idered to be a ke y trait potentially affecting fitness in many ways, both dir ectly and indirectly through correlated physiological parameters (Cloutier et al . 2000; Jiménez-Pérez and Wang 2004). Many studies have provided strong support for the ‘s ize-dependent fitness’ model, i.e., large individuals often have physiological and behavioural advantages (Visser 1994; Cloutier et al. 2000). For example, in hym enopteran parasitoids the reproductive fitness of females in terms of fecundity, parasitism and longevity is often positively correlated with their body size (Sagarra et al. 2001; Arakawa et al. 2004); large males usually have greater ability to inseminate and compete for mates (Kazmer and Luck 1995) or have better genes and more sp erm supply (van den Assem et al. 1989). Therefore, the fitness cons equences of body size and its correlates, especially the supply of eggs or sperm and adult longevity, are important in population dynamics and essential for understanding and modeling the life history evolution and behavioural decisions. On the basis of generally accepted ‘s ize-dependent fitness’, Charnov’s ‘variation in fitness’ model predicts that para sitoids lay fertilized eggs in larger hosts resulting in female progeny, and unfertilized eg gs in small hosts giving rise to male progeny, and female fitness increases more rapidly with body size than does male fitness (Charnov et al. 1981; Charnov 1982). However, most studies on size-fitness relationships focus on females, neglecting males (King 1987; Visse r 1994). Studies on the relationship between body size of male parasitoids and reproductive fitness usually concentrate on male mate-searc hing behaviour, competitive ability and copulation capacity (van den Assem et al . 1989; Kazmer and Luck 1995), and rarely address the effect of male size on the other fitness parameters such as the number of resulting female progeny (Ode et al. 1996) . To my knowledge, Heinz (1991) was the first author to consider the effect of male body size on the reproductive fitness in a wasp species, Diglyphus begini (Ashmead). He compared the male and female reproductive success in relation to their body size, and suggested that females gained more in terms of the number of progeny with increasing body size than did males. . Reproductive Fitness of Aphidius ervi 63 However, like many parasitic wasps, D. begini is a species of haplodiploid sex determination. Heinz’s results had thus overestimated the effect of female body size on reproductive fitness because the number of fathers’ progeny (only female progeny) was always fewer than that of mothers’ progeny (both female and male progeny). A positive relationship was found between A. ervi body weigh and pea aphid size at parasitisation (Sequeira and Mackau er 1992). However, no study has addressed whether and how body size of both A. ervi males and females affects their reproductive fitness. The fitness of parasitoids of different sizes needs to be known when calculating the value of particular hosts to foraging female parasitoids, which is important in modeling host selection strategies (Charnov and Stephens 1988; Visser 1994). Moreover, body size-fitness relationships are also relevant to mass-rearing programmes, as body size is commonly monito red as an indicator of parasitoids’ quality (Jervis and Copland 1996; Sagarra et al. 2001). In the laboratory A. ervi females mated only once while males could inseminate up to 8 females if the mated male was offered a virgin female daily; the first 6 inseminated females produced similar number and proportion of female progeny (Section 4.5). Similar results were also reported for another parasitoid species, Trichogramma evanescens Westwood (Jacob and Boivin 2004). This information has two implications. On the one hand, the gr eater longevity of adult wasps may be important not only for females to lay more eggs (Cloutier et al. 2000) but also for males to inseminate more females. On the other hand, a male in the field may not be able to inseminate as many females as in the laboratory because there are simply not enough virgin females available (i.e. the proportion of female in the field population is about 66%) (Sequeira and Mackauer 1993); ther efore, the first or first few matings may be the most important for the male’s re productive success. In this study, the data from the first insemination were recorded and analyzed only. This section reports the relationship between the body size of both sexes and reproductive fitness in A. ervi , with two objectives: (1) to determine how body size affected reproductive fitness in terms of fecundity, para sitism, number and proportion of female progeny produced, searching efficiency, and longevity, and (2) to evaluate whether body size of both sexes affected reproductive fitness in similar way. . Reproductive Fitness of Aphidius ervi 64 4.4.2 Materials and Methods 4.4.2.1 Measurement of Body Size On the basis of previous work reported by various authors (e.g. Cloutier et al. 2000; Arakawa et al. 2004; Paine et al. 2004), head width of adults was used as the index of body size for A. ervi. The body size of emerging A. ervi wasps is host size-dependent. In this experiment, the mean head width (mean ± SE) of parasitoids emerged from aphids parasitised at the first instar (male, 0.547 ± 0.003 mm; female, 0.588 ± 0.004 mm) was significantly smaller than that from aphids parasitised at the fourth instar (male, 0.612 ± 0.004 mm; female, 0.663 ± 0.003 mm) (ANOVA: F = 184.14 and 218.55 for male and female, respectively; df = 1,58; P < 0.0001). The former was defined as “small” (S) and the latter as “large” (L) in this section. 4.4.2.2 Experiments To determine whether and to what exte nt parasitoids’ body size affected their reproductive fitness, four treatments were se t up with fifteen replicates per treatment: small female (SF) × small male (SM), large female (LF) × large male (LM), SF × LM and LF × SM. For each replicate, one mated A. ervi female (< 12 h old) was introduced into a plastic cylinder with 50 healt hy third instar aphids (3 d old) feeding on a bean plant cutting. The female was allowed to stay in the cylinder for 24 h, and then moved to another cylinder with 50 healthy third instar aphids, etc. until she died. Data on the daily basis and all data collected during females’ lifetime were recorded and included in the analysis. The total number of aphids used in the experiment was 8350, 8800, 10550 and 10600 for treatments SF × SM, SF × LM, LF × SM, and LF × LM, respectively. The once-mated males were numbered in correspondence with the females they mated with and reared individually in glass vials (1.5 cm in diameter, 5.0 . Reproductive Fitness of Aphidius ervi 65 cm in height) with a 0.5 cm mesh-covered hole in lids, and supplied with 10% honey solution in a cotton wool ball (0.5 cm in diameter) daily. To estimate the number of eggs laid, 10 aphids from each cylinder were randomly selected 4 days after the removal of the female parasitoid and then dissected in 70% alcohol under the stereomicr oscope. In total 1680, 1640, 2000 and 2030 aphids were dissected for above four treatments, respectively. The number of parasitoid larvae recorded from dissecting was assumed equal to the number of eggs laid (Bueno et al. 1993) or fecundity. The remaining aphids were reared until mummification. The number of parasitism is the sum of the number of parasitised aphids detected by both dissecting and mummies. The number and proportion of female progeny were counted and calculated from emerged progeny. Longevity of both sexes was also recorded for all treatments. 4.4.2.3 Statistical Analysis A goodness-of-fit test indicated that data were normally distributed and thus analyzed using ANOVA followed by a Tukey' s studentized range (HSD) test. The data on the proportion of female progeny were arcsine transformed prior to analysis. A central composite design (CCD), i.e., response surface (Box and Draper 1987) was used to analyze the differential si ze-dependent fitness gain of sexes in terms of the number and proportion female progeny produced. The relationship between parasitoid body size and reproductive fitness is given by the polynomial equation: fitness = Exp( β 0 + β 1 xf + β 2 xm + β 11 xf 2 + β22 xm 2 + β 12 xf xm), where β 0, β1 , β2 , β 12 , β 11 and β 22 are model parameters, and xf and xm are female and male body size, respectively. Only significant terms, afte r running the full regression models, were kept in the final models. A log likelihood ratio test (McCullagh and Nelder 1989) was then applied to determine whether body size of sexes had different effect on the number and proportion female progeny produced. The slopes of the regression lines of size-longevity relationships of sexes were also analyz ed using an analysis of covariance (ANCOVA). . Reproductive Fitness of Aphidius ervi 66 4.4.3 Results 4.4.3.1 Male Reproductive Performance in Relation to Body Size Male body size had no significant eff ect on fecundity (Figure 4.10A) and parasitism (Figure 4.10B). However, females mated to large males produced significantly more female progeny (Fi gure 4.10C) and had significantly higher proportion of female progeny (Figure 4.10D) than those mated to small males. Moreover, females mated to large ma les produced female-biased progeny for significantly longer period than those mated to small males (Figure 4.10E). 4.4.3.2 Female Reproductive Performance in Relation to Body Size Large females had significantly higher fecundity (Figure 4.10A), parasitised significantly more aphids (F igure 4.10B), and produced significantly greater number of female progeny (Figure 4.10C) than sm all females. The proportion of female progeny was significantly higher for small females than that for large females (Figure 4.11D). However, female body size did not affect how long females sustained the production of female-biased progeny (Figure 4.10E). . Reproductive Fitness of Aphidius ervi 67 LM S M S F LF 0 50 100 150 No. o f e gg s la id a a b b A LM SM S F LF 0 10 0 20 0 30 0 40 0 500 No. o f p ar as iti s m a a b b B LM S M S F LF 0 50 100 15 0 20 0 No. o f f em al e pr og en y b a c b C No. o f f em al e pr og en y LM S M S F LF 0 20 40 60 80 No. o f f em al e pr og en y c b b a D Fem al e pr og en y% LM S M S F LF 0 2 4 6 8 1 0 No. o f f em al e pr og en y b a b a E Day s pr od uc in g fe m al e pr og en y Figure 4.10 Effect of body size of both sexes on the number of eggs laid (A), number of parasitism (B), number of female progeny (C), proportion of female progeny (D) and days of proportion of female progeny > 50% (E). Columns with the same letters were not significantly different (P > 0.05). For A to D: F = 43.43, 36.12, 13.69, and 29.68 for A, B, C and D, respectively; df = 3,56; P < 0.0001. For E: F = 3.88; df = 3,56; P < 0.05. . Reproductive Fitness of Aphidius ervi 68 4.4.3.3 Relative Fitness Gain of Males and Females in Relation to Body Size The number of female progeny significantly increased with the body size of both sexes (Figure 4.11A) but log likelih ood ratio test shows that males had significantly more gain than females in the number of female progeny with the increasing body size ( χ 2 = 234.15; df = 2; P < 0.0001). Unlike males, the female size had negative effect on the proportion of female progeny (Figure 4.11B). Although large males and females lived significantly longer than small males and females (ANOVA: F = 4.35, df = 3,56, P < 0.01 for female; F = 3.15, df = 3,56, P < 0.05 for male) (Table 4.3), females gained dispropo rtionately greater longevity with the increase of body size than did males (ANCOVA: F = 19.17; df = 2,117; P < 0.0001) (Figure 4.12). Table 4.3 Effect of body size (head widt h, mm) on longevity (days) in A. ervi. SF × SM SF × LM LF × SM LF × LM Female 11.13 ± 0.74 b 11.73 ± 0.63 b 14.07 ± 0.74 a 14.13 ± 0.86 a Male 12.13 ± 0.63 b 13.87 ± 0.52 ab 13.67 ± 0.45 ab 14.40 ± 0.45 a Means ( ± SE) followed by the same letters in rows were not significantly different (P > 0.05). . Reproductive Fitness of Aphidius ervi 69 100 15 0 2 0 0 2 5 0 30 0 0.50 0.54 0.58 0.62 0.66 0.70 0.50 0.54 0.58 0.62 0.66 0.70 N o. o f f em al e pr og en y Fem ale he ad wid th ( mm )Male head width (mm) A 0.4 0.6 0.8 1.0 1.2 0.50 0.54 0.58 0.62 0.66 0.70 0.50 0.54 0.58 0.62 0.66 0.70 P ro po rt io n of fe m al e pr og en y Fem ale he ad wid th ( mm )Male head width (mm) B Figure 4.11 Effect of female ( xf) and male ( xm) body size on fertility [A: y = exp(1.26 + 2.97 xf + 3.23 xm) (R 2 = 0.5222; F = 31.15; df = 2,57; P < 0.0001)] and fertility rate [B: y = exp(-0.67 - 2.25 xf + 2.77 xm) (R 2 = 0.6328; F = 48.96; df = 2,57; P < 0.0001)] in A. ervi. . Reproductive Fitness of Aphidius ervi 70 5 10 15 20 0.45 0.5 0.55 0.6 0.65 0.7 0.75 Head width (mm) L on ge vi ty (d ) ● —— Female ○ - - - - - Male Figure 4.12 Relationship between body size and longevity of A. ervi. For female: y = 35.05 x – 9.16 (R 2 = 0.2895; F = 23.63; df = 1,58; P < 0.0001); for male: y = 16.54 x + 3.93 (R 2 = 0.1452; F = 9.85; df = 1,58; P < 0.01). 4.4.4 Discussion The body size of A. ervi progeny is influenced by the host stage or age at parasitisation (Sequeira and Mackauer 1992). Like many hymenopteran species (Honék 1993; Godfray 1994; Vi sser 1994; Sagarra et al. 2 001; Arakawa et al. 2004), the reproductive fitness of A. ervi females and males depends on their body size. The increase in the number of eggs laid with the increasing female body size in A. ervi may result from increasing longevity and greater ability to generate eggs when needed, as detected in A. nigripes (Cloutier et al. 2000). A higher number of eggs laid by large females may also translate into a higher parasitism due to their greater longevity. However, unlike many othe r insects where nutrients contributed by males through copulation are used by females for egg production (Huignard 1983; Simmons and Gwynns 1993; Jiménez-Pérez and Wang 2004) , large A. ervi males do not increase the number of eggs laid by females. This suggests that A. ervi females do . Reproductive Fitness of Aphidius ervi 71 not obtain a nutritional contribution from males during copulation as suggested by Godfray (1994), Fauvergue et al. (1998), Cheng et al. (2003) and Kh anh et al. (2005). With the increase of body size, males gained more in the number of daughters and less in the longevity than did females. It is indicated that males concentrate their resource gain (body size) more on sperm production and insemination than on longevity, suggesting that the benefit from longevity is less certain than inseminating virgin females they can find. The higher number of female progeny produced by larger females may be due to the greater storage capacity of the spermatheca and more production of eggs in those females (Lauzière er al. 2000). As found in other hymenopteran species, for example, the encyrtid parasitoid, Anagyrus kamali Moursi (Sagarra et al. 2001), the proportion of female progeny of A. ervi is higher for small females than that for large females. This may be because small females live shorter and deposit fewer eggs than large females and never become sperm limited. Furthermore, females mated with large males laid female-biased eggs for a longer period of time, suggesting that large males supply more sperm and thus sustain a female-biased population longer than small males. The fitness data of A. ervi are generally in agreement with the main premise of Charnov’s ‘variation in fitness’ model, wh ich concerns the differential size-fitness relationship in males vs. females (Charnov et al. 1981; Charnov 1982). However, this study demonstrates that body size of A. ervi females and males affects reproduction in different ways, i.e. female size has major effect on fecundity, and male size has more impact on paternity. The asymmetrical size-fitness between sexes may be because reproductive success of parasitoids directly relies on females’ potential fecundity and their ability to search and parasitise hosts whereas the success of male reproduction absolutely relies on, and is mediated, by females. In addition, virgin females still reproduce with the absence of males. Therefore, the differe nt reproductive strategies of different sexes may result in the asymmetrical size-dependent fitness. In conclusion, the body size of both sexes contribu tes to population growth and sex ratio di stribution in A. ervi. Results of the asymmetric size-dependent fitness . Reproductive Fitness of Aphidius ervi 72 in different sexes provide us useful informa tion when we manage to predict parasitoid population dynamics and practice effective mass-rearing and release programmes. . Reproductive Fitness of Aphidius ervi 73 4.5 Effect of Mate Age at Mating and Male Mating History on Reproduction of A. ervi 4.5.1 Introduction Hymenopteran parasitoids are usually a rrhenotokously parthenogenetic, where fertilized eggs produce diploid females a nd unfertilized eggs produce haploid males; virgin females give birth to only sons while inseminated females produce both sons and daughters (Quicke 1997). As a result, the reproductive success of females is mainly affected by egg production patterns and fertilization (Godfray 1994; Boivin et al. 2005) and that of males mainly by the number of daughters they father (Quicke 1997; Roitberg et al. 2001; Jacob and Boivin 2004). For the parasitoid species with monandrous females and polygamous males, the net reproductive gain of females from mating may depend on the age of sexe s at mating and male mating history, and that of males on the age of sexes at mating, the number of females they inseminate, and oviposition history of their mates. The ways and extent of these effects may vary considerably between sexes as well as betw een species, depending on their life history strategies (Pandey et al. 1983; Srivas tava and Singh 1995; King 2000; Jacob and Boivin 2004; Damiens and Boivin 2005). The understanding of reproductive strategies of hymenopteran parasitoids is ultimately important for the success of biological control (Gordh and DeBach 1976; Nadel and Luck 1985; Godfray 1994; King 2 000; Roitberg et al. 2001; Jacob and Boivin 2004; Damiens and Boivin 2005). In pro-ovigenic species, females have all their egg complement mature on emergence while in synovigenic females they mature eggs during adult stage (Fla nders 1950; Quicke 1997; Boiv in et al. 2005). However, most species probably fall in between these two extremes, and are termed pro- synovigenic, i.e. females carry some matu re eggs on emergence but continue to produce and mature more eggs throughout adult lifespan (Quicke 1997; Jervis et al. 2001, 2003). Both pro-ovigenic and synovigeni c traits are found in an encyrtid parasitoid, Coccidoxenoides peregrinus (Timberlake) (Ceba llo and Walter 2004), suggesting that female parasitoid life history strategies may be far more complicated than what we understand today. Prospermatogenic males have fully developed their . Reproductive Fitness of Aphidius ervi 74 testes (which have fixed number of sperm) by the time they reach maturity, whereas synspermatogenic males produce sperm during adult stage (Gerling and Legner 1968; Gordh and DeBach 1976; Nadel and Luck 1985 ; Jacob and Boivin 2004; Boivin et al. 2005). Similar to females, many species appe ar to be moderately synspermatogenic, i.e. males emerge with some mature sperm, and can continue to produce and mature sperm throughout their lifespan (Boivin et al. 2005). Prospermatogeny is expected to be advantageous for gregarious wasps where mating takes place mostly at emergence, and for the solitary species synspermatogeny should be advantageous as it enables males to disperse first and then locate mates (Boivin et al. 2005). Boivin et al. (2005) argue that species that are prospermatogenic should also be pro-ovigenic and vice versa because the constraints linked to life history parameters are likely to act on gamete production in both sexes. Hymenopteran parasitoids may develop gregariously, solitarily or quasi- gregariously. Gregarious parasitoid species favor sibmating on the natal patch and local mate competition (Mackauer and Völkl 2002), and usually produce strongly female-biased broods, e.g. only one or two males in each brood (Hardy et al. 1998). Solitary parasitoid species, in contrast, favor dispersal from natal patch and outbreeding (Nadel and Luck 1992), and their offspring sex ratio is usually not strongly biased (Fisher 1930). If the hosts of a solitary parasitoid species are gregarious such as aphids or coccids, quasi-gregarious broods of the parasitoid offspring can be produced (van den Assem et al. 1980), where partial sibmating and local mate competition may occur (Mackauer and Völkl 2002). Furthermore, in quasi- gregarious parasitoid species, sex ratio is often moderately female-biased and unfertilized eggs are usually laid earlier than fertilized eggs (Hardy 1992). Aphidius ervi is an endophagous, solitary and koinobiont parasitoid. The field sex ratio is ca. 1 male:1.9 females (S equeira and Mackauer 1993). Males are polygynous (Starý 1962), and females monandrous (Mackauer 1969). Under the conditions similar to the present study, A. ervi adults can live for 11-14 d (Section 4.4); the female carries about 60 mature eggs on emergence (Section 4.2) but can parasitize more than 300 aphids in her lifespan (Section 4.3), indicating that A. ervi females are pro-synovigenic with an ovigeny index of ca. 0.2 (calculated according to Boivin et al. . Reproductive Fitness of Aphidius ervi 75 2005). On the basis of the above property of A. ervi , it should be a quasi-gregarious species, i.e. a solitary parasitoid species attacking gregarious hosts. In this study, I postulate that, in A. ervi , (1) the age of males and females at mating has a differential effect on fecundity and daughter production because the number of matings varies between sexes, and (2) males are synspermatogenic or pro- synspermatogenic due to the quasi-gregari ous and polygynous property of this species. These hypotheses were tested through investigating how the age at mating and male mating history affected reproductive outputs. Knowledge of these reproductive strategy parameters is essential to the understanding of life history evolution and manipulation of daughter production for mass-reared and field populations of A. ervi. 4.5.2 Materials and Methods 4.5.2.1 Experimental Insects Parasitoids used for experiments emerge d from mummies that were parasitised in the third instar (3 d old), and the third in star pea aphids were used as hosts in all experiments. 4.5.2.3 Experiments To examine whether mate age at mating affected A. ervi female reproductive output and sex allocation, nine treatments were set up, i.e. virgin males and virgin females mated at nine combinations of three age classes: 1, 3 and 5 d old. There were ten replicates for each treatment. This experiment was carried out in transparent plastic cylinders (Figure 3.2). A male and a female were maintained in a glass vial (1.5 cm in diameter, 5.0 cm in height with a 0.5 cm mesh-covered hole in lids) for mating. After mating, the female was introduced into a plastic cylinder with 50 healthy third instar aphids feeding on a bean plant cutting. The female was allowed to stay in the cylinder for 24 h, and then moved to another cylinder with 50 healthy third instar aphids, etc. until . Reproductive Fitness of Aphidius ervi 76 she died. Females mated at 3 and 5 d old were also individually provided with 50 health aphids daily for oviposition before and after mating. Honey solution (10%) was supplied daily as food for parasitoids. For all females, the number of aphids parasitised was recorded and newly emerged offspring were sexed. The number of progeny produced by females after they mated was applied to calculate the proportion of daughters produced. Ten virgin females were used as control. To determine how male mating history affected female offspring production of A. ervi , two experiments were set up: (1 ) hourly mating and (2) daily mating. In the hourly mating experiment, a male was offered with a 1-d-old virgin female hourly in an above mentioned glass vial. Immediately after each mating, the mated males were removed, and individually kept in another glass vial. The number of females that a male mated with during one day (16 h phot ophase) was recorded. In the daily mating treatment, a male was offered a 1-d-old vi rgin female once a day during his life time. The number of females that a male could mate with was recorded as mentioned above. The males were supplied with 10% hone y solution as food by dropping on a cotton wool ball (0.5 cm in diameter). This experiment was conduced during photophase and ten males were tested for each experiment. Soon after mating, each female in both hourly and daily mating experiment was provided with 50 healthy th ird instar aphids daily as mentioned above, until she died. Ten percent honey solution was also supplied as food for females as mentioned above. The number of aphids parasitised and proportion of female offspring produced were recorded. 4.5.2.4 Statistical Analysis Goodness-of-fit test was used to test the distribution of data. The data on the number of parasitism were normally distributed and analysed using ANOVA, followed by Tukey’s studentised range te st. The data on proportion of female offspring were not normally distributed even after arcsin transformation and thus analyzed using the nonparametric Kruskal-Wallis test followed by Dunn’s procedure for multiple comparisons (Zar 1999). . Reproductive Fitness of Aphidius ervi 77 A central composite design (CCD), i.e., response surface (Box and Draper 1987) was used to analyze the effect of mate age at mating on proportion of female offspring produced. The relationship between mate age and proportion of female offspring produced is given by the polynomial equation: female offspring% = exp( β 0 + β 1 xf + β 2 xm + β 11 xf 2 + β 22 xm 2 + β 12 xf xm), where β 0, β 1 , β 2 , β12 , β11 and β22 are model parameters, and xf and xm are female and male age, respectively. Only significant terms, after running the full regression models , were kept in the final models. A log likelihood ratio test (McCullagh and Nelder 1989) was then applied to determine whether age of sexes had different eff ect on the proportion female offspring. The relationship between the proportion of female offspring produced and number of mates a male mated with was analysed using regression analysis (RA), in both hourly and daily mating experiments. ANCOVA was used to analyse the slopes. 4.5.3 Results Age of mates at mating and male mating history had no significant effect on the number of aphids parasitised (fro m 326.8 ± 11.6 to 352.6 ± 9.4) which was similar to that for virgin females (341.4 ± 5.1) (ANOVA: F = 0.18; df = 26,183; P > 0.05). The proportion of female offspring produced significantly decreased with the increasing age of both sexes at mating (CCD: F = 7.14; df = 2,87; P < 0.0001; R 2 = 0.1410) (Figure 4.13A). However, female age had significantly more effect on the proportion of female offspring produced than did male age ( likelihood rate test: 22χ = 96.18; P < 0.0001) (Figure 4.13B). When mated with 5-d-old females, 40, 20 and 30% 1-, 3- and 5-d-old males did not transfe rred sperm to their mates, respectively. . Reproductive Fitness of Aphidius ervi 78 30 45 60 75 1 2 3 4 5 1 2 3 4 5 F em al e of fs pr in g% Fe ma le ag e Male age A 30 45 60 7 5 1 2 3 4 5 Mate age F em al e of fs pr in g% Female = 1 day old Male = 1 day old B y = exp(4.34 - 0.02 male age - 0.12 female age) Figure 4.13 Effect of mate age (d ays) at mating on female offspring production of A. ervi: (A) proportion of female offspring produced in pairs of different age combinations of sexes (n = 90); (B) predic ted proportion of female offspring produced by males and females of different ages. . Reproductive Fitness of Aphidius ervi 79 The proportion of female offspring produced started to become significantly lower at fourth and seventh mating of a male in hourly and daily experiments, respectively (Kruskal-Wallis test: H = 59.75 > 2 05.0,8 χ = 15.51 for hourly experiment and H = 21.78 > 2 05.0,7 χ = 14.07 for daily experiment; P < 0.0001) (Figure 4.14). Furthermore, regression analysis shows that in both hourly and daily experiments, the proportion of female offspring produced significantly decreased with the increasing number of matings in males (RA: F = 132.62 and df = 1.59 for hourly experiment, and F = 34.77 and df = 1,47 for daily experime nt; P < 0.0001) (Figure 4.15), while it decreased significantly faster in hourly than in daily experiments (ANCOVA: F = 77.57; df = 2,107; P < 0.0001). 0 20 40 60 80 1 0 0 1 2 3 4 5 6 7 8 9 Mating F em al e of fs pr in g% Hourly mating Daily mating a a a a ab ab ab ab b b ab bc cd cd d d d 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Figure 4.14 Proportion of female offspring of A. ervi males after successive matings. Means (± SE) followed by the same letters within each line are not significantly different (P > 0.05). . Reproductive Fitness of Aphidius ervi 80 y = - 2.34x + 65.73, R 2 = 0.4252 y = - 9.72x + 79.79, R 2 = 0.6921 0 20 4 0 6 0 8 0 1 0 0 0 1 2 3 4 5 6 7 8 9 1 0 Mating F em al e of fs pr in g% Hourly mating: Daily mating: 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Figure 4.15 Relationship between proportion of female offspring and male mating history of A. ervi. 4.5.4 Discussion In insects, males may transfer both sperm and nutrition or sperm only during copulation. In the aphidiid Lysiphlebus delhiensis (Subba Rao and Sharma), female age rather than male age at mating significantly reduces the fecundity (Srivastava and Singh 1995), suggesting that males of this species may transfer egg production stimulants and/or nutrients to females during mating, and if mating occurs when females are young, these materials could significantly increase egg production. However, like the aphelinid A. asychis Walker (Fauvergue et al. 1998), neither parental age at mating nor male mating history had any significant influence on the female fecundity in A. ervi. This implies that mating does not stimulate egg production and males do not transfer male-derived nutrients during mating in A. ervi. In our study of body size effect on fecundity (Section 4.4), results also suggest that there may be no nutrient transfer from males during mating. The effect of mate age and male mating history on sperm production by male and sperm utilization by females is genera lly reflected in the proportion of female offspring produced. The proportion of female offspring declined with parental age at mating in this study. Similar to the aphidiid L. delhiensis (Srivastava and Singh 1995), . Reproductive Fitness of Aphidius ervi 81 the lower production of daughters by older males may not be attributed to the sperm depletion in A. ervi. Rather, this could be due to diminishing sperm viability and mobility related to aging (S rivastava and Singh 1995). For the aging females, the reduction of daughter production could be expl ained by the constraints in fertilization process in A. ervi. Flanders (1946) found the de pletion of spermathecal gland secretion resulting in a reduced availability of activated sperm for fertilization of ovulating eggs. According to Pandey et al. (1983), eggs may cont ain certain kind of sperm attracting chemicals, which deplete w ith the female age, leading to a reduced fertilization. This study also indicated that female age at mating had more severe effect than male age on the production of daughters in A. ervi; between 20 and 40% of females when mating at the age of 5 d did not pr oduce any daughters but none of the 5-d-old males failed to produce daughters. This suggests that fertilization process in females may be more sensitive to aging than sperm vigor in males. It is also likely that males adjust the amount of sperm transfer accordi ng to female age, delivering less sperm to older females (Srivastava and Singh 1995). In the hourly mating experiment, A. ervi males started to show sperm depletion after three successive matings, wh ile in the daily mating experiment, males still had sufficient sperm supply until having mated six times. In addition, the production of daughters decreased significantly faster in hourly mating than in daily mating treatments. These findings have two implications. First, males of this species appear to be at least moderately synspermatogenic because they can replenish sperm if allowed 24-h recovery period between matings up to six matings. As a quasi- gregarious species moderate synspermatogeny should be advantageous to some extent because it enables males to disperse and locate mates (Boivin et al. 2005) apart from partial sibmating and local mate competition (Mackauer and Völkl 2002). Moderate synspermatogeny also has been reported in a number of other parasitic hymenopteran families (Nadel and Luck 1985; review by Boivin et al. 2005). Second, because females are monandrous it could be difficult for A. ervi males to find a number of virgin females and mate several times within a day in the field. Therefore, A. ervi males may seldom suffer sperm depletion in the field. The reduction of daughter production in the daily mating experiment ma y be caused by the reduction in sperm . Reproductive Fitness of Aphidius ervi 82 transfer and quality due to a decrease in males’ vigour with the increased age (Srivastava and Singh 1995; King 2000; Jac ob and Boivin 2004; Damiens and Boivin 2005) rather than sperm depletion. In hymenopteran parasitoids, the numbe r of daughters produced may be an estimate of the maximum effective quantity of sperm a male can carry or transfer (Henter 2004). The number of sperm carried by A. ervi males at emergence may be detected by the number of daughters produced in the hourly experiment. The mean number of daughters produced was 757.50 ± 72.05 and 920.50 ± 146.50 for males in hourly and daily mating experiments, respectively. The results imply that males can carry 82% mature sperm at emergence and produce only about 18% sperm during adult life. Higher amounts of sperm carried at emergence may enable males to maximize the reproductive fitness of both se xes by taking advantage of more females being inseminated early in their life, becau se males’ vigour usually decreases over their age as discussed above and females carry mature eggs in ovaries and start oviposition soon after emergence. In conclusion, females are pro-synovigenic and males are at least moderately synspermatogenic in A. ervi , supporting Boivin et al.’s (2005) prediction. The fact that A. ervi adults do not feed on hosts (unpublished observation) and both sexes produce gametes during the adult stage suggests that adult food sources containing amino acids may boost egg production and sperm replenishment in this parasitoid. As a result, supply of adult food such as pollen may increase reproductive potential of both sexes in this species. Aging a ffects sexes in different ways : for old males, diminishing sperm vigor may be the main cause of the reduced daughter production while for old females constraints of egg fertilization proce ss in relation to aging may be the reason for the lower daughter proportions. Informati on provided in this study should be taken into consideration in manipulating optimal daughter production in both mass-reared and field populations of A. ervi. . Mating Behaviour o f Aphidius ervi 83 CHAPTER 5 MATING BEHAVIOUR AND SEXUAL SELECTION OF APHIDIUS ERVI 5.1 Introduction Mating is a central component of se xual reproduction in insects. Sexual reproduction involves separate reproductive strategies for males and females. Since Darwin (1859) highlighted the evolutionary importance of competition for fertilising the female ova, the study of mating behaviour and sex selection has received massive attention. In Hymenoptera, unmated females la y unfertilised haploid eggs that develop into males, and mated females can lay either unfertilised eggs that develop into males or fertilised diploid ones that develop into females (Quicke 1997). Thus, the mating behaviour contains fundamental information needed to understand parasitoids’ reproductive biology, and the discov ery of the traits selected for mating preference by males and females provides valuable information necessary for the development of biological control strategies. 5.2 General Methods 5.2.1 Data Recording E xperiments were conducted in Petri dishes (8.5 cm diameter × 1.3 cm height) that maintained experimental parasitoids. The mating behaviour was observed for 10 minutes and recorded using a Panasonic S VHS Camcorder (MS-4) (Panasonic, Japan) connected with a Samsung digital video cassette recorder ( DVD-V5 30, Korea) and images were viewed on a Panasonic colour monitor (TC-21T1Z, Japan) (Figure 5.1). Experiments were carried out in the first four hours after light was on at 8:00am as A. ervi generally mate in the morning (Mackauer 1969). . Mating Behaviour o f Aphidius ervi 84 Figure 5.1 Equipment used for recording mating behaviour. 5.2.2 Behaviour Definition Behavioural interactions between male a nd female parasitoids were defined as followings: (a) Courtship - a male rapidly fanned his wings within 1 cm from a female; (b) Approaching - a male approached a female with his wings fanning; (c) Mounting - a male mounted a female; (d) Homosexual behaviour - a male c ourted and mounted another male; (e) Mating success - a male inserted his aed eagus into a female’s genitalia and copulated; (f) Mate competition - a male mounted another male when the latter had mounted and/or was mating with a female; (g) Precourtship period - the period of a male needed to start wing fanning after he was released into a Petri dish where a female was preset; (h) Homosexual period - the period of a male performing courtship display to and mounting another male; (i) Courtship period - the period betw een initiation of wing fanning and mating; (j) Competition period - the period of a ma le mounting another male that had mounted and/or was mating with a female; (k) Mating period - the period from ma ting initiation to termination. . Mating Behaviour o f Aphidius ervi 85 5.3 Mating Behaviour and Effect of Male Mating History on Mating Success in A. ervi 5.3.1 Introduction Studies on mating behaviour are essential for understanding the reproductive behaviour of a parasitoid species. Mating behaviour can be studied for practical purposes, for example, the development of efficient techniques for the mass-rearing of large numbers for field or glasshouse release in a biological control program (Hardy et al. 2005). A major problem with mass-rearing of parasitoids is an overproduction of males (Waage et al. 1985; Waage 1986). A study of mating behaviour can help solve this problem and provide the key to the optimal production of males. Thus it can provide valuable information for the implementation and development of biological control programmes. The primary mating behaviour of A. ervi has been described (Spencer 1926; Mackauer 1969; Battaglia et al. 2002; Mc Clure et al. 2007) and three steps are identified: (1) females release sex pheromones, (2) males detect the sex pheromone and court the female, and (3) mating. The effect of male mating history on the reproductive fitness of both sexes has been studied (Section 4.5). However, no quantitative information on the behavioural se quences is available, making it difficult to understand the interactions between males and females and between males. Aphidius ervi is a solitary endophagous parasitoid species with females being monandrous and males polygynous. Because its hosts are gregarious, it is a quasi- gregarious species (i.e. solitary species th at develop in hosts that are gregarious) (Mackauer and Völkl 1993) where partial lo cal mating may be common (Hardy et al. 2005). Under natural field conditions, virg in females may encounter mated and/or virgin males, and male-male competition may occur. Therefore, mated and virgin males and virgin females were used for this experiment. The aim of the section was to investigate the mating behaviour of A. ervi based on observations of male-female and male-male interactions in relation to male mating history. . Mating Behaviour o f Aphidius ervi 86 5.3.2 Materials and Methods Parasitoids emerged from aphid mummies that were parasitised at third instar (3 d old) and were all 1 day old when used. To observe A. ervi mating behaviours defined in Section 5.2.2, a virgin male and a mated male were introduced into an above mentioned Petri dish containing a virgin female. Mating sequences were observed and recorded for 10 minutes as mentioned above. For each replicate, before introduced into the Petri dish, one of the two test males was randomly selected and marked on the front wings with Radiant Color (Magruder Color Company, NJ). The mated males were used 1 h later after individually copulating with a virgin female. Sixty-five replicates were performed. 5.3.3 Statistical Analysis The number of mated and virgin males that undertook wing fanning, approaching, homosexual behaviour, moun ting, competition and mating was analysed using a Chi-square test. A goodness-of-fit test was used to test the distribution of following data before analysis. Data on precourtship period were not normally distributed even after transformation and thus analysed using the nonparametric Kruskal-Wallis test. Data on homosexual behaviour, competition and mating period were normally distributed and analysed using ANOVA. Data on courtship period were square-root transformed prior to ANOVA. 5.3.4 Results 5.3.4.1 General Mating Behaviour The observed behaviours of all individuals were pooled to illustrate the general mating behavioural sequences of A. ervi (Figure 5.2). The mating behaviours can be divided into two phases: courtshi p and mating. Courtship included male wing fanning, approaching and mounting. . Mating Behaviour o f Aphidius ervi 87 After released into a Petri dish, males walked randomly in the dish and about 50 seconds later, 86% of them became excited and started fanning their wings upon encounter or detection of virgin females. Wh en a male first approached a female with his fanning wings and attempted to mount the latter, the female often ran or flew away. After an average of 4.2 ± 0.4 attempts (mean ± SE), the male usually mounted the female successfully from rear or side, cr awled forward, and grasped the female on the thorax and abdomen with his fore- and mid-legs. The male moved his antennae rapidly from side to side and touched those of the female. If the female rejected the male, she walked and tried to prevent the co ntact of their external genitalia by bending her abdomen downward. When the female accepted the male, she lowered her antennae and held them still, and raised her abdomen. When the female was in her receptive position, the male crawled backward, curved his abdomen, and inserted his aedeagus into her genitalia. The female usually terminated the mating by moving her antennae and walking away. After mating, bot h sexes usually cleaned their heads, mouthparts and antennae using their forelegs. There were 13.8 ± 4.6% males that did not perform courtship behaviour in 10 minutes. Among the courting males, 48.3 ± 14.4 and 42.9 ± 12.0% of them successfully mounted and mated with the females, respectively. Mating lasted an average of 59.8 ± 1.1 seconds. When the courting males encountered other males, 30% of them showed homosexual behaviour by displaying courts hip behaviours to each other. A male might compete for mating by mounting another male that had already mounted a female, displaying the same courtship beha viour and attempting to mate with the female. This behaviour might cause the couple to deleave from each other if mating had not occurred. However, in most cases (15/16) , the male that first mounted the female would mate, and once the mating occu rred, the subsequent male could not split the couple. . Mating Behaviour o f Aphidius ervi 88 14% Figure 5.2 Mating behavioural sequences of A. ervi. Probabilities of a particular transition between stages are given. ♂ wing fanning ♂ approaching ♀ ♂ mounting ♀ mating ♂ walking 100% ♂ homosexual behaviour ♂ failing to mate ♂ competition 86%15 % 7% 63% 3 0 % 8 5 % 56% 44% 60% 29% 11% 7% 93% 13% . Mating Behaviour o f Aphidius ervi 89 Behaviour Figure 5.3 Comparison of the number of courtships and matings displayed by mated and virgin males when competing for a virgin female in 10 minutes. n = the number of males. ♂ wing fanning Virgin male (n = 65) n = 53 Chi-square test: χ2 = 1.61, P > 0.05 n = 49 Chi-square test: χ2 = 3.17, P > 0.05 n = 18 Chi-square test: χ2 = 12.61, P < 0.001 n = 16 Chi-square test: χ2 = 9.55, P < 0.01 Mated male (n = 65) n = 59 n = 56 n = 37 n = 32 ♂ approaching ♀ ♂ mounting ♀ mating ♂ walking . Mating Behaviour o f Aphidius ervi 90 5.3.4.2 Mating Success in Mated and Virgin Males The number of males that displayed courtship behaviours with wing fanning and approaching females was not significantly different between mated and virgin males (Figure 5.3). However, compared to the virgin males, mated males were significantly more likely to mount and mate with the females (Figure 5.3). Mated males had significantly shorter precourtship (Kruskal-Wallis test: H = 4.35 > 2 05.0,1 χ = 3.84; df = 1,110; P < 0.05) and cour tship period (ANOVA: F = 5.43; df = 1,46; P < 0.05) than did virgin males (Table 5.1). However, there was no significant difference in mating period between mated and virgin males (ANOVA: F = 3.39; df = 1,46; P > 0.05) (Table 5.1). Table 5.1 The mean (± SE) period (seconds) of A. ervi males performing mating behaviours. Precourtship Courtship Mating Mated male 36.10 ± 4.35 b 134.00 ± 17.20 b 58.50 ± 3.35 a Virgin male 64.77 ± 11.65 a 200.33 ± 33.71 a 63.13 ± 4.84 a Means ( ± SE) followed by the same letters in columns are not significantly different (P > 0.05). 5.3.4.3 Male-Male Interaction There was no significant difference in the number of males preforming homosexual (n = 13 and 16, for mated and vi rgin males, respectively) and mating competition behaviours (7 mated and 9 virgin males, respectively) between mated and virgin males ( χ 2 = 1.433 and 0.001 for homosexual and mating competition behaviours, respectively; df = 1,110; P > 0.05). After mating occurred, no mated or virgin males could separate the mating couples. Virgin males spent a slightly longer period in homosexual behaviour and mating competition but no significant difference was found when compared with . Mating Behaviour o f Aphidius ervi 91 mated males (ANOVA: F = 0.097 and df = 1,17 for homosexual period; F = 1.519 and df = 1,14 for competition period; P > 0.05) (Table 5.2). Table 5.2 The mean (± SE) period (seconds) of homosexual and competition behaviour performed by A. ervi males. Homosexual Competition Mated male 62.56 ± 15.56 a 52.25 ± 10.51 a Virgin male 69.67 ± 16.65 a 69.13 ± 8.78 a Means ( ± SE) followed by the same letters in columns are not significantly different (P > 0.05). 5.3.5 Discussion Behavioural observations indi cate that males appear to be more active than females in the mating behavioural sequence of A. ervi. This may be because a virgin female can reproduce without mating but the success of male reproduction absolutely relies on, and is mediated by females (Quicke 1997). Moreover, as detected in A. nigripes (McNeil and Brodeur 1995), A. ervi females release a sex pheromone that triggers the male behavioural sequence of wing fanning, appr oaching and mounting (Battaglia et al. 2002; McClure et al. 2007) . My results show that the courtship behaviour by males (i.e. wi ng fanning and approaching) is non-random, indicating that in A. ervi , females produce signals and males do the searching. Tactile stimulation may play an important role in mating success of A. ervi. The male moved his antennae rapidly from side to side and contacted those of the female. This behaviour seemed to have qui etened the female. Previous reports show that males of many hymenopteran wasps have glands in their antennae (Isidoro and Bin 1995; Isidoro et al . 1996; Battaglia et al. 2002). Males of A. ervi emit secretion from their antennae when they are exposed to virgin females; the secretion functions specifically as a contact pheromone and the antennae play a pivotal role in sex recognition and partner acceptance (Battaglia et al. 2002). . Mating Behaviour o f Aphidius ervi 92 In theory, female choice is clearly an important driving force in sexual selection in animals, and a few examples of adaptive choice are known in natural enemies, particularly fly predators such as the black-tipped hanging fly Hylobittacus apicalis (Hagen) (Thornhill 1988) and scorpion fly Panorpa vulgaris Imhoff and Labram (Thornhill and Sauer 1991). However, in parasitoids there is no evidence that females exhibit any active choice (Hardy et al. 2005). A virgin parasitoid female will usually mate with the first conspecific male encountered, for example, in a gregarious parasitoid N. vitripennis (van den Assem et al. 1980). In the present study, A. ervi females usually mated with the first mounting males regardless of their virginity. This may be because of the difficulty of finding another suitable partner (van den Assem et al. 1986). As found in other Aphidius species (Mackauer 1969), A. ervi females usually became receptive after several approachings by males. It is still not clear whether the females have means to assess their partners’ quality during this period. Obvious male aggression behaviour was not observed in A. ervi , but male- male interactive behaviour such as male courtship display and mating attempt to the rival was common when males encountered each other. Homosexual mounting between males is found in a number of insect species including aphidiids (Starý 1970; Thornhill and Alcock 1983; van den A ssem 1986; Kaneshiro and Giddings 1987; Gillespie 1991; Wang et al. 1996; Serrano et al . 2000; Cheng et al. 2003; Switzer et al. 2004). It is expected that mated males have learnt from the previous mating in discerning the potential mates and rivals and avoid homosexual mounting; however, results indicate the freque ncy of homosexual mounting between mated and virgin males was similar. Homosexual interactions between males in A. ervi may be elicited by the female sex pheromone, as reported in another aphidiid, Praon aguti Smith (Sekhar 1957; Starý 1970). Visual cues may also result in male mistake in courtship in some insects (Thornhill and Alcock 1983; Wang et al. 1996), including some hymenopteran parasitoids (van den Assem 1986). According to Thornhill and Alcock (1983), althought males are at the risk of making some mistakes, mounting in A. ervi may be an important process in discriminating a mate. Male mating experience is a factor that affects mating success in A. ervi. Results of hourly mating experiments (Sectio n 4.5) indicate that females can produce similar number of female offspring when mated with virgin males or once-mated . Mating Behaviour o f Aphidius ervi 93 males, thus females are not expected to have preference between virgin and once- mated males for mating. In the present study, virgin males invested more energy in courtship (i.e. wings fanning and approaching) for courting females than mated males. Mated males responded to females more quickly and performed fewer approaches to females before mounting, and achieved more matings. These suggest that mated males have learnt from the previous mating, so th at they respond to females more quickly and perform better courtships than naïve males. It has be en reported that in some hymenopteran parasitoids (Perez-Lachaud and Campan 1995; Fischer and King 2005) and other insects (Schwartz 1991; Cook 1995) male mating experience enhances mating success. However, McClure et al. (2007) indicated that under no-choice situation, A. ervi males that have previously mated have longer pre-courtship period but have similar mating success when compared to virgin males of similar age. Thus, it is suggested that under competitive situation, mated A. ervi males may sense the presence of rivals, court females quickly and then achieve better mating success. . Mating Behaviour o f Aphidius ervi 94 5.4 Mate Choice of A. ervi in Relation to Body Size and Age 5.4.1 Introduction Mate choice is an important factor in the mating system that influences the reproductive fitness of both sexes. Accord ing to Darwin (1859), females should be choosier than males, because eggs cost mo re to produce than sperm, females mate fewer times than males and females endure most of the parental care in order to achieve the maximum fitness return. This leads to the conclusion that females have more to lose from suboptimal matings than males do. However, in Hymenoptera, virgin females can still produce progeny without mating; success of male reproduction absolutely depends on whether they can inseminate the females. In general, studies on mate choice have focused on female choice of males and paid less attention to male choice of females (Rosenqvist and Berglund 1992; Ahnesjö et al. 1993; Jennions and Petrie 1997, 2000; Bonduriansky 2001). Male choice of females has been reported for a range of taxa, although few have been studied in hymenopteran species (Bonduriansky 2001). Stud ies on mate choice in both sexes in a single insect species are uncommon (Bonduriansky 2001), especially in parasitoid species (Godfray 1994; King et al. 2005). In many parasitoids such as A. ervi , mate choice is poorly known. Body size and age of A. ervi males and females are two important factors that affect the reproductive fitness (see Sections 4.4 and 4.5). This section investigated the role of body size and age in the ma te choice for both sexes of A. ervi. 5.4.2 Materials and Methods 5.4.2.1 Influence of Body Size on Mate Choice The body size of “small” (S) and “large” (L) adults of both sexes was defined in Section 4.4.2.1. To investigate whether and to what extent mate body size influenced mate choice, four treatments were set up: (1) small female × (small male + large male), i.e. SF × (SM + LM); (2) large female × (small male + large male), i.e. LF × (SM + LM); (3) SM × (SF + LF); and (4) LM × (SF + LF). Parasitoids were all . Mating Behaviour o f Aphidius ervi 95 virgin and one day old when used for testing. There were 80, 75, 70 and 75 replicates for treatments (1), (2), (3) and (4), respect ively. To observe the mate choice behaviour, in each replicate, female(s) and male(s) were released into a closed Petri dish and mating sequences were observed and r ecorded for 10 minutes (Section 5.2.1). The number of small and large males that undertook courtship (wing fanning), mounting and mating was analysed using a Chi-square test. A goodness-of-fit test was used to test the distribution of other data (i.e. precourtship, courtship and mating periods) before ANOVA. Data were base- 10 logarithm transformed before ANOVA if they were not normally distributed. 5.4.2.2 Influence of Mate Age on Mate Choice E xperimental parasitoids emerged from aphid mummies that were parasitised at third instar (3 d old). To determine wh ether mate age affected mate choice, two treatments were set up: (1) one 1-d-old male + three females of 1, 3 and 5 days of age; (2) one 1-d-old female + three males of 1, 3 and 5 days of age. There were 78 and 61 replicates for treatments (1) an d (2), respectively. To obser ve mate choice behaviour, female(s) and male(s) were released into a closed Petri dish and mating sequences were observed and recorded for 10 minutes (S ection 5.2.1). Before in troduced into the Petri dish, two of the three test mates of different ages were randomly selected and marked on their front wings with Radiant Color (Magruder Color Company, NJ). In this experiment, parasitoids were fed with 10% honey solution. They were all virgin and females did not lay eggs (no aphids provided) before testing. The Marascuilo procedure of the nonparametric analysis (Daniel 1990) was used to assess the number of males performing mating behaviours (courtship, mounting and mating). The re jection level was when '0U > 2 05.0,2 χ = 5.99. A goodness- of-fit test was used to test the distribution of other data. Data of the mean number of a male approaching a female were not normally distributed even after transformation and thus analysed using the nonparametric Kruskal-Wallis test. All other data were analysed using ANOVA followed by a Tukey's studentized range (HSD) test. Data of . Mating Behaviour o f Aphidius ervi 96 precourtship and courtship periods were base-10 logarithm transformed before ANOVA. 5.4.3 Results 5.4.3.1 Influence of Body Size on Mate Choice When a small male and a large male were provided to a female, similar numbers of small and large males courted, mounted and mated regardless of female size (P > 0.05) (Table 5.3). No significant difference was detected in precourtship period between the larger and small males (Table 5.4). Compared to small males, large males had significantly shorter courtship and mating periods regardless of female size (Table 5.4). When a small female and a large female were provided to a male, large males started courtship display (i.e . wings fanning) significantly earlier than did small ones (precourtship period = 45.95 ± 2.40 and 52.86 ± 7.90 seconds, n = 66 and 59 for large and small males, respectively) (ANOVA: F = 5.58; df = 1,123; P < 0.05). Small males did not have significant preference for mates of different size for mating (Chi-square test: χ 2 = 0.63 and 1.67 for mounting and mating, respectively; df = 1; P > 0.05) but large males significantly preferred large females for mounting and mating (Chi-square test: χ 2 = 4.32 and 7.35 for mounting and mati ng, respectively; df = 1; P < 0.05) (Tables 5.5). Regardless of male size, both courtship and ma ting periods were significantly longer when mating with large females than when mating with small females (Table 5.6). . Mating Behaviour of Aphidius ervi 97 Table 5.3 The number of small and large males (selectees) preformed mating behaviour in response to females (selectors) in A. ervi. Selectees Selectors Small female (n = 80) Large female (n = 75) courtship mounting mating courtship mounting mating Small males 61 a 31 a 24 a 59 a 29 a 21 a Large males 68 a 38 a 32 a 63 a 33 a 29 a Chi-square test: χ 2 = 1.44, 0.92 and 1.35 for courtship, mounting and mating, respectively for small females; χ2 = 1.10, 0.25 and 1.47 for courtship, mounting and mating, respectivel y for large females; df = 1; P > 0.05. Table 5.4 The mean (± SE) period (seconds) of mating behaviours performed by males (selectees) to th e females (selectors) in A. ervi. Selectees Selectors Small female Large female Precourtship* Courtship* Mating Precourtship* Courtship Mating Small male 76.31 ± 8.54 a 202.63 ± 26.49 a 60.13 ± 1.25 a 73.56 ± 7.60 a 262.86 ± 29.15 a 65.52 ± 2.03 a Large male 69.65 ± 9.34 a 117.16 ± 12.50 b 54.00 ± 1.17 b 61.27 ± 9.51 a 120.45 ± 11.67 b 58.83 ± 1.48 b F 2.57 9.82 12.53 2.96 25.36 11.58 df 1, 127 1, 54 1, 54 1, 120 1, 48 1, 48 P > 0.05 < 0.01 < 0.001 > 0.05 < 0.0001 < 0.001 Means ( ± SE) followed by the same letters in columns ar e not significantly different (ANOVA: P > 0.05). * Data were base-10 logarith m transformed before ANOVA. . Mating Behaviour of Aphidius ervi 98 Table 5.5 The number of males (selectors) pref ormed mating behaviour in response to small and large females (selectees) in A. ervi. Selectees Selectors Small male (n = 70) Large male (n = 75) mounting mating mounting mating Small female 24 a 19 a 23 b 17 b Large female 27 a 24 a 33 a 29 a Numbers followed by the same letters in co lumns are not significantly different (P > 0.05). Table 5.6 The mean (± SE) period (seconds) of mating behaviours performed by males (selectors) to the females (selectees) in A. ervi. Selectees Selectors Small male Large male Courtship* Mating Courtship* Mating Small female 92.95 ± 13.17 b 54.37 ± 1.92 b 86.24 ± 8.42 b 51.18 ± 1.81 b Large female 167.71 ± 21.45 a 67.50 ± 2.25 a 123.76 ± 12.52 a 59.45 ± 1.46 a F 10.14 18.48 4.41 12.30 df 1,41 1,41 1,44 1,44 P < 0.01 < 0.001 < 0.05 < 0.001 Means ( ± SE) followed by the same letters in columns are not significantly different (ANOVA: P > 0.05). * Data were base-10 logarith m transformed before ANOVA. 5.4.3.2 Influence of Mate Age on Mate Choice When a 1-d-old male was provided with three females of different ages, 74.4% (58/78 ) of males courted. The number of males approaching 1- and 3-d-old females was significantly higher than that approaching 5-d-old females (Kruskal-Wallis Test: H = 9.32; df = 2,150; P < 0.01) (Table 5.7) . Males tended to mount younger females more often but there was no significant difference in mounting between females of different ages ( '0U = 3.24 < 2 2,05.0 χ = 5.99; df = 2; P > 0.05); however, upon mounting, . Mating Behaviour of Aphidius ervi 99 the probability of males dismounting their mates was significantly higher when they mounted 3- and 5-d-old females than when they mounted 1-d-old ones ( '0U = 11.47; df = 2; P < 0.001) (Table 5.7). Males signifi cantly preferred young than old females for mating ( '0U = 10.49; df = 2; P < 0.001). Within 10 minues observation period, significantly more males recourted and remated (with other females) after they had mated with 3- or 5-d-old females than after they mated with 1-d-old females ( '0U = 37.45 and 29.51 for recourting and remating, resp ectively; df = 2; P < 0.0001) (Table 5.7). Table 5.7 Mating behaviour of A. ervi males in the presence of females of different ages. Female age No. of approaching No. of mounting Dismounting % (n*) No. of mating Recourting % (n) Remating % (n) 1 d old 2.25 ± 0.18 a 24 a 4.1 (1) b 23 a 16.7 (4) b 28.3 (2) b 3 d old 2.45 ± 0.21 a 20 a 30.0 (6) ab 14 ab 78.6 (11) a 71.4 (10) a 5 d old 1.66 ± 0.20 b 15 a 40.0 (6) a 9 b 88.9 (8) a 66.7 (6) a Means ( ± SE) followed by the same letters in columns are not significantly different (P > 0.05). *n was the number of male s performed dismounting, recourting or remating behaviour. Males started to court an average of 49.47 ± 4.25 seconds (n = 58) after being released into the Petrel dishes and recourted 212.48 ± 14.69 seconds (mean, n = 23) after the termination of first mating. The mean courtship period was 128.78 ± 9.80 and 102.22 ± 22.17 seconds for the first and se cond mating, respectively. In the first mating, mating period was significantly longer when males mated with 1-d-old females than with 5-d-old ones (ANOVA: F = 6.31; df = 2,43; P < 0.01) (Table 5.8). However, no significant difference was f ound in mating duration between different female ages in the second mating (ANOVA: F = 0.72; df = 2,15; P > 0.05) (Table 5.8). . Mating Behaviour of Aphidius ervi 100 Table 5.8 Mean ( ± SE) mating period (seconds) of males when they mated with females of different ages. Female age 1 d old 3 d old 5 d old 1 st mating 53.08 ± 1.73 a 49.64 ± 1.59 ab 42.67 ± 2.46 b 2 nd mating 41.71 ± 2.68 a 43.25 ± 2.29 a 39.14 ± 1.71 a Means ( ± SE) followed by the same letters in lows are not significantly different (P > 0.05). When a 1-d-old virgin female was offere d to three virgin males of different ages, significanly more 1-d-old males cour ted, mounted and mated with the females than did 5-d-old males ( '0U = 14.71, 10.91 and 7.77 > 2 05.0,2 χ = 5.99 for mounting and mating, respectively; df = 2; P < 0.05) (Table 5.9). Before mounting, the 1-d-old male approached the female significantly more frequently than did 5-d-old male (mean ± SE = 3.34 ± 0.22, 2.97 ± 0.28 and 2.46 ± 0.24 for 1-, 3- and 5-d-old males, respectively; Kruskal-Wallis Test: 2χ = 6.33; df = 2,106; P < 0.05). The 1-d-old males had significantly shorter precourtship period but mated for a significantly longer period than did 5-d-old males (Table 5.10). There was no signifi cant difference in courtship period between the courting males of different ages (Table 5.10). Table 5.9 Number of males of different ages that courted, mounted and mated with 1- d-old females. Male age Courtship Mounting Mating 1 d old 47 a 20 a 16 a 3 d old 35 ab 12 ab 10 ab 5 d old 28 b 7 b 6 b Numbers followed by the same letters in co lumns are not significantly different (P > 0.05). . Mating Behaviour of Aphidius ervi 101 Table 5.10 The mean (± SE) period (seconds) of males of different ages performing mating behaviours to 1-d-old females in A. ervi. Male age Precourtship Courtship Mating 1 d old 37.17 ± 7.48 b 135.63 ± 19.64 a 58.50 ± 1.73 a 3 d old 45.57 ± 9.45 ab 177.20 ± 21.51 a 54.20 ± 1.25 ab 5 d old 60.29 ± 10.14 a 156.33 ± 24.35 a 51.00 ± 2.22 b F 3.66 1.37 4.03 df 2,107 2,28 2,29 P P < 0.05 P > 0.05 P < 0.05 Means ( ± SE) followed by the same letters in columns are not significantly different (ANOVA: P > 0.05). 5.4.4 Discussion Previous studies have demonstrated that large A. ervi males have more sperm supply during mating (Section 4.4). Results of this section (Section 5.5) indicate that large males had shorter courtship and mating periods (Table 5.4) which may help them to gain access to mates and reduce the predation risk comparing to small males. However, when provided two males of diff erent size, females did not significantly prefer large males for mating (Table 5.3). Generally, female mating preferences may be due to direct benefits which increase females’ fecundity or survival as detected in some other insect species (Huignard 1983; Simmons and Gwynns 1993; Jiménez- Pérez and Wang 2004). In A. ervi , however, mating does not increase females’ fecundity and longevity, and females do no t obtain a nutritional contribution from males during copulation (Section 4.4). Moreover, A. ervi is arrhenotokously parthenogenetic, where inseminated fema les produce both sons and daughters and virgin females can still give birth to sons. Finally, A. ervi females do not suffer sperm insufficiency (i.e. proportion of female o ffspring larger than 50%) when they mate with small or large males (Section 4.4). Therefore, it is not surprising that A. ervi females do not show any observable active mate choice. There have been many studies showing that larger males have a mating advantage over smaller males in mating competition and success, but most of them . Mating Behaviour of Aphidius ervi 102 have described aggressive male-male compe tition, defence of terr itory, or resource provision (Alcock 1979, 1984; McCauley 1979; Hughs and Hughs 1982; Johnson 1982; Karban 1983; Thornhill and Alcock 1983; Juliano 1985; Alcock andGwynne 1987; Vencl and Carlson 1998; Bateman 2000), and such behavioural characters are not observed in A. ervi. Thus it is suggested that similar to Lariophagus distinguendus ( F örster) (van den Assem et al. 1989), body size is not important for A. ervi males to compete successfully. When provided two females of different size, both small and large males preferred large females to small ones for mating, although they invested longer time to court the larger females (Table 5.6). Th is is probably because large females have higher reproductive potential and thus males mating with large females obtain maximum reproductive fitness return in te rms of their female progeny (Section 4.4). Moreover, when compared to small males, la rge males were more likely to mate with large females, which may be attributed to males’ superior behaviours (i.e. courted earlier with shorter courtship period). This may also explain the greater success of large males in mate searching when compared to small males, which has been demonstrated in other parasitoid species (van den Assem et al. 1989; Kazmer and Luck 1995). Regardless of female size, larger male s had shorter copulations than did small males. Regardless of male size, males copulated longer when they mated with larger females than with small ones. This may be a general rule in insects including parasitoids, for example, L. distinguendus (van den Assem et al. 1989). Usually insects will terminate mating when the maxi mum sperm has been transferred, or when the spermatheca is full (van den Assem et al. 1989). The relationship between body size of mates and copulation duration may be explained by three assumptions: (1) larger females have a larger spermatheca, (2) larger males have more sperm supply, and (3) larger males transfer more sperm per unit of time (van den Assem et al. 1989). The shorter copulation duration detected for larger males but longer copulation duration for larger females (this study) and greater num ber of daughters produced by larger males and females (Section 4.4) corroborates these assumptions. Thus, larger males may transfer more sperm at a faster rate than do small males, and males may need longer time to fill up a larger female’s spermatheca. It is expected that the . Mating Behaviour of Aphidius ervi 103 shorter copulation plus shorter courtship period may enable larger males to have sufficient time in replenishing sperm and searching for a new mate. Thus, large body size seems to directly benefit a male’s fert ilization success because of more frequent matings, which has been found in many hymen opterans (van den Assem et al. 1989; Kazmer and Luck 1995; Sagarra et al. 2001) and also other insect species (Pitnick 1991; Warner et al. 1995; Danielsson 2001). Results indicate that mate choice is influenced by mate age. A. ervi males were more likely to approach and mount young females. A. ervi females release a sex pheromone that triggers males’ behavioural sequence such as wing fanning, approaching and mounting (Battaglia et al . 2002; McClure et al. 2007). However, as reported in hymenopterans (Simser and Coppel 980; McNeil and Brodeur 1995; Schworer et al. 1999) and other insect spec ies (Teal et al. 1990; del Mazo-Cancino et al. 2004; Showler et al. 2005), sex phe romones produced and released by A. ervi females decrease with their age (McClure et al. 2007). Therefore, older females are less attractive to the males. On the other hand, because sperm is a limited resource in A. ervi males (Section 4.5), males preferring young females for mating have greater reward as younger females have greater potential fecundity due to their longer longevity and thus will produce more daughters. As a result, although males did not suffer sperm limitation, they dismounted th e old females more frequently, shortened the mating period, and were more likely to perform the second mating when they mated with old females. It is suggested th at males could discriminate the quality of their mates and optimize the use of their sperm based on the reproductive status of females (Pitnick and Markow 1994; Srivastava and Singh 1995). Male age of A. ervi had a negative effect on courtship and mating success (Tables 5.8 and 5.9). As reported by Mc Clure et al. (2007), under no-choice conditions, significantly fewer 8-d-old males performed courtship behaviour (wing fanning and approaching) and mate su ccessfully than did 1-d-old males. These may be the results of decreased ability to recognize and respond to females’ sex pheromone (Vetter and Visscher 1997; McClure et al. 2007) or decreased activity with aging (Cheng et al. 2003). The fact that young male s achieve more matings than do old ones has also been reported in another parasitoid species, Cephalonomia tarsalis (Ashmead) (Cheng et al. 2003). It is expe cted that young males transfer more sperm than old ones . Mating Behaviour of Aphidius ervi 104 during copulation, which may explain why younger males were more likely to be accepted by females for mating. . Mating Behaviour of Aphidius ervi 105 5.5 Operational Sex Ratio and Density Influence Partial Local Mating Behaviour in A. ervi 5.5.1 Introduction Hardy (1994) summarised three mati ng strategies in the hymenopteran parsitoids: (1) panmixis, which assumes that individuals mate randomly throughout the population (Fisher 1 930); (2) fully local mating or lo cal mate competition (LMC), which assumes that all matings are confined to the natal patch and only mated females disperse (Hamilton 1967); and (3) partial local mating, which is defined as an intermediate mating structure between panmixis and fully local mating (Hardy 1994). Theoretical and empirical studies have shown that LMC is common in parasitoid groups with female-biased sex ratio (Ki ng 1987; Godfray 1994; Ha rdy 1994). Martel and Boivin (2004) indicate that whether LMC occurs in a species depends on whether the species is truly solitary, quasi-gregarious or gregarious. The difference between solitary, quasi-gregarious or gregarious has been discussed in Section 4.5. Partial local mating is influenced by the operational sex ratio ( O S R ) which is defined as the ratio of males searching for mates to females available for mating (Hardy et al. 2005). The OSR composes of two component variables, the male and female densities (Alonso-Pimentel and Papaj 1996). Local population density is defined as the total number of individuals that are ready to mate in a given space (Wang et al. 2008). OSR and local population density are temporally and spatially dynamic in the natural environment (Law rence 1986; Krupa and Sih 1993; Wang and Chen 2005). Because parasitoid females are usually monandrous and males polygynous, the number of females available fo r insemination is usually far less than the number of males ready to mate (Hardy et al. 2005). Therefore, the OSR can be very male-biased even in those cases where the primary sex ratio is female-biased (van den Assem et al. 1980; Werren 1980). The influence of OSR and local population density on mating behaviour has been well studied in many insects (Arnqvist 1992; Parker and Simmons 1994; Joblo ński and Vepsäläinen 1995; Alonso- Pimentel and Papaj 1996; Simmons 2001; Wang et al. 2008) but little is known in parasitoid species. Knowledge of how OS R and population density affect LMC and . Mating Behaviour of Aphidius ervi 106 mating success of parasitoids may provide valuable information in optimizing sex ratio in mass rearing programmes. Aphidius ervi is an arrhenotokously reproducing species, and females are monandrous and males are polygynous. The field populations of A. ervi are female- biased (65.7%) (Sequeira and Mackauer 1993) and parasitis ed aphids often show a clumped distribution (Mackauer and Völkl 1993). Furthermore, males and females may emerge in temporal and spatial synchrony (Section 3.3). These conditions may favor local mating before dispersal. According to Hardy (1994) and van den Assem (1996), although the assumption of fully local mating may be incorrect for A. ervi, partial local mating may be common. This section investigated the effect of OSR and population density on the mating behaviour of A. ervi. 5.5.2 Materials and Methods 5.5.2.1 Experimental Insects Parasitoids used in this study emerged from aphid mummies parasitised at third instar (3 d old). Parasitoids were all virgin and females did not lay eggs before being used for experiment. 5.5.2.2 Experimental Design To determine how OSR and population de nsity affected mating behaviour of A. ervi , four experiments were set up as follows: Exp. A, fixed female density (n = 3) with increasing sex ratios ( ♂: ♀): 1:3, 3:3, 6:3, 9:3, 12:3 and 15:3 with 20, 16, 15, 15, 16 and 16 replicates, respectively; Exp. B, fixed male density (n = 3) with decreasing sex ratios (♂: ♀): 3:1, 3:3, 3:6, 3:9, 3:12 and 3:15 with 16, 16, 15, 15, 16 and 14 replicates, respectively; . Mating Behaviour of Aphidius ervi 107 Exp. C, even sex ratio (1 ♂: 1 ♀) with increasing density: 3:3, 6:6 and 9:9 with 16, 15 and 17 replicates, respectively; Exp. D, fixed overall density with various sex ratios ( ♂: ♀) : D 1 (density = 12), 3:9, 6:6 and 9:3 with 15 replicates for each treatment; D 2 (density = 18), 3:15, 9:9 and 15:3 with 14, 17 and 16 re plicates, respectively. In each replicate, experimental parasi toids were released into an above mentioned Petri dish, and mating sequences were observed and recorded for 10 minutes as described above. Data presented in this section were the mean number of behavioural events (i.e. homosexual behavi our, mate competition and mating success) which occurred in a 10-minute period. The pr obability of female remating, and mating durations of first and second matings by females were also recorded. 5.5.2.3 Statistical Analysis A goodness-of-fit test was used to test the distribution of data before analysis. Data of mating duration were normally distributed and analysed using ANOVA followed by Tukey’s studentized rang (HSD) test. Other data that were not normally distributed even after transformation were analysed using the nonparametric Kruskal- Wallis test. The interactions between male and female densities affecting homosexual behaviour, mate competition, mating success and mating duration were tested using the CCD model as described in Section 4.4. A log likeliho od ratio test (McCullagh and Nelder 1989) was then applied to de termine whether density of sexes had different effect on the mating behaviour of parasitoids. The Marascuilo procedure of the nonparametric analysis (Daniel 1990) was us ed to assess the frequencies of female remating with a rejection level of '0U > 2 05.0,2 χ = 5.99. . Mating Behaviour of Aphidius ervi 108 5.5.3 Results 5.5.3.1 Homosexual Behaviour The number of homosexual events depended on OSR, male density and population density (Table 5.11). It increas ed significantly with increasing male density when female density fixed to 3 (Kruskal-Wallis Test: H = 53.98 > 2 05.0,4 χ = 9.49; df = 4,73; P < 0.0001) and when population density was 12 and 18 (Kruskal-Wallis Test: H = 36.36 and 38.49 98 > 2 05.0,2 χ = 5.99; df = 2,42 and 2,44 for population density of 12 and 18, respectively; P < 0.0001). When the sex ratio was even (i.e. 1♂: 1 ♀) , the number of homosexual events signi ficantly increased with the increasing population density (Kruskal-Wallis Test: H = 18.85 > 2 05.0,2 χ = 5.99; df = 2,45; P < 0.0001). However, when male density was 3, increasing female density significantly reduced the number of homosexual events (Kruskal-Wallis Test: H = 16.23 > 2 05.0,5 χ = 11.07; df = 5,86; P < 0.01). The CCD model also indicates that both male and female densities significantly affected the homosexual behaviour of A. ervi (CCD: F = 163.98; df = 4,181; P < 0.0001; R 2 = 0.7837) (Figure 5.4A) with male density positively and female density negatively effecting the number of homosexual events (likelihood rate test: 22χ = 1080.09; P < 0.0001) (Figure 5.4B). . Mating Behaviour of Aphidius ervi 109 Table 5.11 Effect of OSR and population density on mating behaviour of A. ervi. OSR Number of Number of ma te Number of Mating period ( ♂: ♀) homosexual events competition events mating successes (seconds) Exp. A: fixed female density (n = 3) with increasing sex ratios 1 : 3 0.65 ± 0.11 c 52.77 ± 1.28 b 3:3 0.63 ± 0.24 c 0.55 ± 0.21 b 0.88 ± 0.18 bc 53.71 ± 1.47 b 6:3 5.13 ± 1.13 bc 2.30 ± 0.40 ab 0.93 ± 0.23 bc 56.36 ± 1.32 b 9:3 11.00 ± 1.00 ab 2.43 ± 0.84 ab 1.60 ± 0.24 a 61.94 ± 1.36 ab 12:3 15.44 ± 1.72 a 3.31 ± 0.41 a 1.31 ± 0.20 ab 63.43 ± 1.69 ab 15:3 16.63 ± 1.14 a 3.64 ± 0.75 a 1.31 ± 0.24 ab 67.10 ± 1.86 a Exp. B: fixed male density (n = 3) with decreasing sex ratios 3 : 1 1.81 ± 0.55 a 0.56 ± 0.18 a 0.56 ± 0.13 d 61.44 ± 3.02 a 3:3 0.63 ± 0.24 ab 0.55 ± 0.21 a 0.88 ± 0.18 cd 53.71 ± 1.47 a 3:6 0.33 ± 0.19 ab 0.54 ± 0.18 a 1.47 ± 0.22 bc 56.27 ± 2.63 a 3:9 0.27 ± 0.12 ab 0.85 ± 0.33 a 2.07 ± 0.25 ab 57.13 ± 1.82 a 3:12 0.25 ± 0.11 ab 0.38 ± 0.13 a 2.56 ± 0.13 a 56.32 ± 1.73 a 3:15 0.07 ± 0.07 b 0.29 ± 0.16 a 2.36 ± 0.17 a 56.91 ± 2.47 a Exp. C: even sex ratio (1♂:1♀) with increasing population density 3 : 3 0.63 ± 0.24 b 0.55 ± 0.21 b 0.88 ± 0.18 c 53.71 ± 1.47 b 6:6 2.33 ± 0.35 b 2.73 ± 0.41 a 2.07 ± 0.34 b 59.58 ± 2.17 ab 9:9 5.29 ± 1.07 a 3.24 ± 0.55 a 4.12 ± 0.43 a 63.30 ± 2.82 a Exp. D: fixed population density with varying sex ratios D1: population density = 12 3 : 9 0.27 ± 0.12 c 0.85 ± 0.33 b 2.07 ± 0.25 a 57.13 ± 1.82 a 6:6 2.33 ± 0.35 b 2.73 ± 0.41 a 2.07 ± 0.34 a 59.58 ± 2.17 a 9:3 11.00 ± 1.00 a 2.43 ± 0.84 a 1.60 ± 0.24 a 61.94 ± 1.36 a D2: population density = 18 3:15 0.07 ± 0.07 c 0.29 ± 0.16 b 2.36 ± 0.17 b 56.91 ± 2.47 b 9:9 5.29 ± 1.07 b 3.24 ± 0.55 a 4.12 ± 0.43 a 63.30 ± 2.82 ab 15:3 16.63 ± 1.14 a 3.64 ± 0.75 a 1.31 ± 0.24 c 67.10 ± 1.86 a In each experiment, means (± SE) followed by the same letters in columns are not significantly different (P > 0.05). . Mating Behaviour of Aphidius ervi 110 0 5 1 0 1 5 2 0 1 2 3 4 5 6 7 8 9 1 0 11 1 2 1 3 1 4 1 5 Density No. o f h om os exu al e ve nt s . Male (female density = 3) Female (male density = 3) B Figure 5.4 Interactions of male ( χm) and female ( χf) densities affecting homosexual behaviour in A. ervi: (A) number of homosexual ev ents performed by males in combinations of different male and female densities (y = exp(- 0.9372 + 0.6421 χm - 0.0262χ m2 - 0. 2866χ f + 0.0188 χmχf); (B) predicted number of homosexual events affected by male and female densities. . Mating Behaviour of Aphidius ervi 111 5.5.3.2 Competition for Mate As shown in Table 5.11, the number of mate competition events increased significantly with increasing male density when female density was fixed to 3 (Kruskal-Wallis Test: H = 17.17; df = 4,60; P < 0.01), when population density was 12 (Kruskal-Wallis Test: H = 8.31; df = 2,39; P < 0.05) and 18 (Kruskal-Wallis Test: H = 13.06; df = 2,39; P < 0.01) or when the OSR was even (Kruskal-Wallis Test: H = 14.47; df = 2,40; P < 0.001). However, when male density was fi xed, no significant difference was found in the number of competition events between various OSRs (Kruskal-Wallis Test: H = 4.34; df = 5,70; P > 0.05). Both male and female densities significantly affected mate competition behaviour of A. ervi (CCD: F = 26.28; df = 4,157; P < 0.0001; R 2 = 0.4011) (Figure 5.5A) but male density had significantly mo re effect on competition behaviour than did female density (likelihood rate test: 22χ = 34.28; P < 0.0001) (Figure 5.5B). The number of mate competition events was higher when female density was from 6 to 9 and male density from 12 to 15 (Figure 5.5A). 5.5.3.3 Mating Success Mating success depended on the density of both sexes (Table 5.11, Figure 5.6). Mating success increased significantly with increasing density of males from 1 to 9 and females from 1 to 12, after which no future increase occurred (Kruskal-Wallis Test: H = 14.82 and 51.46 and df = 5,92 and 5,8 6 for male and females, respectively; P < 0.01). The likelihood rate test indicated th at female density had significantly more effect on mating success than did male density ( 22χ = 9.84; P < 0.0001) (Figure 5.6B). At even OSRs, mating success increased significantly with increasing population density (Kruskal-Wallis Test: H = 25.65; df = 2,45; P < 0.0001) (Table 5.11). At the population density of 12, no significant difference was found between OSRs (Kruskal-Wallis Test: H = 2.28; df = 2,42; P > 0.05); however, at the population density of 18, significantly higher mating success was detected in the OSR of 9 ♂: 9 ♀ (Kruskal-Wallis Test: H = 24.58; df = 2,47; P < 0.0001) (Table 5.11). . Mating Behaviour of Aphidius ervi 112 0 1 2 3 4 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Density No. o f c om pe tit io n ev en ts . Male (female density = 3) Female (male density = 3) B Figure 5.5 Interactions of male (χ m) and female ( χf) densities affecting mate competition in A. ervi : (A) number of mate competiti on events performed by males in combinations of different male and female densities (y = exp(- 2.1592 + 0.4421 χm - 0.0173χ m2 + 0.2742χ f - 0.0208 χf2 ); (B) predicted number of mate competition events affected by male and female densities. . Mating Behaviour of Aphidius ervi 113 0 1 2 3 4 5 1 2 3 4 5 6 7 8 9 1 0 11 1 2 13 1 4 15 Density No. o f m at in g su cc es s . Male (female density = 3) Female (male density = 3) B Figure 5.6 Interactions of male ( χm) and female ( χf) densities affecting mating success in A. ervi : (A) number of mating successes in co mbinations of different male and female densities (CCD: y = exp(- 1.5704 + 0.2042 χm - 0.0092 χm2 + 0.3159 χf - 0.0126χ f2 ; F = 38.40, df = 4,201, P < 0.0001, R 2 = 0.4332); (B) predicted number of mating successes affected by male and female densities. . Mating Behaviour of Aphidius ervi 114 5.5.3.4 Mating Period At the fixed female density of 3, mati ng period significantly increased with male density (ANOVA: F = 5.25; df = 5,103; P < 0.01); but at the fixed male density of 3, increasing female density had no effect on mating period (ANOVA: F = 0.52; df = 5,145; P > 0.05) (Table 5.11). The CCD model a nd likelihood rate te st also indicate that only male density had significant effect on mating period (CCD: F = 25.73, df = 1,291, P < 0.0001, R 2 = 0.0805; likelihood rate test: 22χ = 605.74, P < 0.0001) (Figure 5.7). At even OSR, mating period significan tly increased with increasing population density (ANOVA: F = 3.52; df = 2,57; P < 0.05), and at the population density of 18, mating became significantly longer with the increasing male density (ANOVA: F = 3.45; df = 2,78; P < 0.05) (Table 5.11). . Mating Behaviour of Aphidius ervi 115 50 55 60 65 70 75 80 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Density Mat in g pe rio d (se co nd s) . Male (female density = 3) Female (male density = 3) B Figure 5.7 Interactions of male ( χm) and female ( χf) densities affec ting mating period in A. ervi: (A) mating period in combinations of di fferent male and female densities (y = exp(3.9891 + 0.0155 χm); (B) predicted mating period affected by male and female densities. . Mating Behaviour of Aphidius ervi 116 5.5.3.5 Female Remating Female remating was observed in most OSRs with the exceptions of 1:3, 3:1 and 3:15 ( ♂: ♀) (Table 5.12). The proportion of mated females mounted by other males and females remated was significantly higher in male-biased OSRs than in female-biased ones (Table 5.12). Fe males remated 50.67 ± 9.39, 47.08 ± 3.66 and 47.50 ± 3.60 seconds after the first mating at female-biased, even and male-biased OSRs, respectively, with no significant di fference (ANOVA: F = 0.09; df = 2,27; P > 0.05). For the remated females, mating peri od in the second mating was significantly shorter than in first mating (Table 5.13). Table 5.12 The proportion of mated females mounted by other males and females remated in A. ervi in various OSRs. OSR ( ♂: ♀) Total no. mated females Mated females mounted by other males% Females remated% Female-biased (3:6, 3:9, 3:12) 94 3.19 b (n* = 3) 3.19 b (n = 3) Even (3:3, 6:6, 9:9) 115 11.30 b (n = 13) 11.30 ab (n = 13) Male-biased (6:3, 9:3, 12:3) 81 30.86 a (n = 25) 17.28 a (n = 14) ' 0U 28.03 12.52 df 2 2 P < 0.0001 < 0.01 Percentage values followed by the same letters in columns are not significantly different (P > 0.05). *n is the number of mated females mounted by other males or remated. . Mating Behaviour of Aphidius ervi 117 Table 5.13 The mean (± SE) mating period (seconds) of A. ervi females in the first and second matings. Female-biased OSR Even OSR Male-biased OSR First mating 55.67 ± 7.31 a 60.54 ± 4.67 a 71.36 ± 4.40 a Second mating 17.00 ± 7.00 b 31.15 ± 5.15 b 30.07 ± 5.74 b F 14.59 17.86 32.62 df 1,4 1,24 1,26 P P < 0.05 P < 0.001 P < 0.0001 Means ( ± SE) followed by the same letters in columns are not significantly different (ANOVA: P > 0.05). 5.5.4 Discussion In A. ervi , the OSR of 1♂: 3 ♀ may reflect the sex ratio in field condition (i.e. 65.7% females, Sequeira and Mackauer 1993). However, OSR varies from female- biased to male-biased because of the declin ing number of virgin females. Results of this study show that A. ervi males tended to have more homosexual and competition events with the increasing of OSR and male density, especially when the OSR skewed to be male-biased. The increasing male homosexual and male mate competition events may be due to the increasing encounter rate between searching males. Thornhill and Alcock (1983) proposed an evolutionary interp retation of insect homosexual mounting behaviour. They assume that if the male’s main cost of reproduction is the time consumed in discerning the sex of potential mates, increasing the number of mountings per unit of time may be more advantageous rather than carefully choosing a partner of the right sex. Competition for mate increasing with OSR may be a general rule which has been demonstrated in many animals including insects (Hamilton 1979; Lawrence 1986; Enders 1993; Krupa and Sih 1993; Dick and Elwood 1996; Michner and Mclean 1996; Parker and Simmons 1996; Clutton-Brock et al. 1997; Jennions and Petr ie 1997; Flanagn et al. 1998 ; Jirotkul 1999, Abe et al. 2003; Shuker et al. 2006). Mating success increased with the increasing density of males or females and population density in even OSR. This may be because of the increasing encounter rate between males and females leading to the easier mate location. Furthermore, CCD model and the lik elihood rate test indicated that male . Mating Behaviour of Aphidius ervi 118 density had more effect on homosexual behaviour than did female density, but female density had more effect on mating success than did male density. It is further confirmed that production of highly female-biased offspring is advantageous to A. ervi in successful reproduction and improves mass-rearing programmes. Generally, the switching-off of parasitoid females’ receptivity following the end of a mating results from the presence of sperm in the spermatheca or the additional substances introduced by the male during copulation (Manning 1962; Merle 1968). It has been found in many pa rasitoid species, immediately following mating no further receptive signaling or response could be provoked, for example, in the Spalangia cameroni Perkins, Asaphes vulgaris Walker and Anogmus strobilorum Thompson (van den Assem 1986). Previous observations (personal observation in Section 4.5) showed that A. ervi females were unattractive to males one hour after mating. In this study, most mated females of A. ervi ( > 82%) were unattractive to males soon after mating and no longer mated, indicating that they have switched off their receptivity. Furthermore, males are le ss likely to mount and mate with mated females, suggesting that A. ervi males may be able to discriminate females’ mating status. Male discrimination against females’ mating status has been demonstrated in some parasitoid species such as L. distinguendus (Ruther et al. 2000) and Spalangia endius Walker (King et al. 2005). Female unreceptivity and male discrimina tion may benefit both sexes (van den Assem 1986; King et al. 2005) . To mated females of A. ervi , because they have received sufficient sperm during the first mating (Sections 4.4 and 4.5), preventing males from mating may enable them to concentrate on host searching and oviposition. To males, female remating may provide lo w fitness to them because possible first- male sperm precedence may occur (El Ago ze et al. 1995). Sperm precedence has not been examined in A. ervi , but first-male sperm precedence has been reported in many hymenopteran species (Wilkes 1966; Parker 197 0; Holmes 1974; van den Assem et al. 1989; Allen et al. 1994; El Agoze et al. 1995) . El Agoze et al. (1995) indicated that the difference in sperm precedence level between species may depend on variables such as the number of matings in nature and the time between successive matings. Results in this study have demonstrated that a mated A. ervi female is usually unattractive to males after 1 minute follo wing a mating, and remating in females in . Mating Behaviour of Aphidius ervi 119 the field is thus rare. Therefore, female s switching off receptivity soon after mating and male discrimination of females’ mating status may reduce the probability of males mating with the already mated females and hence, allow them to avoid the sperm wasting and save time and energy for searching for other mates. However, in some hymenopteran species, female unreceptivity after a mating is a gradual process which proceeds with time. For example, in N. vitripennis, responsiveness is switched off within 1 minute following the termination of mating (King 1960), and when more courtship is preformed by males, a second or third receptive signal could be provoked (van den Assem and Visser 1976). Results in this study showed that some females (< 18%) were still receptive to other males within 1 minute after the first mating. Moreover, in this study mating period of females in the second mating was significantly shorter than that in the first mating, and a short mating period of about 50 seconds (Section 5.4) is sufficient to fertilise a female’s life (Section 4.4). Thus female remating may be due to the gradual process of switching off female receptivity rather than the insufficient sperm transferred during the first mating. A small proportion of females that remated (3.2~6.7%) is also reported in Aphytis melinus DeBach (Allen et al. 1994). However, McClure et al. (2007) reported that in A. ervi , 25% (4/16 ) females remated 24 hours after the first mating. They supposed that remating in their study is possibly the result of an unsuccessful first mating or due to a low quality mate. Results from Sections 4.4 and 4.5 indicate that one mating of females could fertilise more than 50% of their fecundity, and thus do not support McClure et al .’s (2007) supposition. Obvious post-copulatory guarding is not common in the majority of parasitoid species (van den Assem 1986). In this a nd previous (Sections 5.3 and 5.4) studies, postcopulatory courtship and contact or non-contact mate guarding were not observed after the termination of mating. Therefore, mate guarding, if any, should occur during copulation in this species. Like some in sects (Arnqvist 1992; Parker and Simmons 1994; Joblo ński and Vepsäläinen 1995; Alonso-Pi mentel and Papaj 1996; Simmons 2001; Wang et al. 2008), A. ervi prolonged copulation duration under male-biased OSR and high male density. The reason may be that when the OSR and male density increase, the probability that a male will find another mate decreases but possibility that his mate will remate increases. . General Discussion and Conclusion 120 CHAPTER 6 GENERAL DISCUSSION AND CONCLUSION 6.1 Introduction In this thesis, I report how A. ervi reduces aphids’ re productive potential, investigate the factors that affect the reproductive fitness of parasitoids and examine the mating behaviour of A. ervi. Such knowledge is important for a better understanding of biological control ecology of A. ervi. In this chapter, I summarise and discu ss my main findings and their relevance to the reproductive biology of A. ervi in order to provide valuable information necessary for the development of biological control strategies. 6.2 Reproductive Rhythms of A. ervi In insects, adults’ circadian activities are usually correlated with reproductive rhythms, such as mating and oviposition. A. ervi is an arrhenotokous and quasi- gregarious species (Section 4.5) with monandrous females and polygynous males (Mackauer 1969). Males’ reproductive su ccess mainly depends on the number of mates they inseminate. Males emerge earlier but become sexual mature later than females. On the basis of the above reproductive property of A. ervi , the emergence pattern is advantageous to males where they can coincide with the sexual receptivity of females and mate before the latter disperse and start oviposition. Moreover, A. ervi adults usually emerge during the first few hours of the photophase and females mainly oviposit throughout the photophase. Virgin females may start oviposition within 30 min after emergence (unpublished data). Females attack more aphids and lay more eggs in aphids in the photophase than in the scotophase, suggesting that besides the chemical and physical cues (Vinson 1984; Schmidt 1991; Vet and Dicke 1992; Turlings et al. 1993; Godfray 1994; Quicke 1997), A. ervi females may also employ visual cues in host searching and oviposition (Battaglia et al. 1995). The oviposition pattern of A. ervi females ensures they reserve . General Discussion and Conclusion 121 energy and produce mature gametes during the scotophase. Therefore, the emergence of A. ervi synchronises the rhythmic function of mating and oviposition. 6.3 Reproduction of A. ervi in Relation to Host Age and Density Aphidius ervi significantly reduces the reproductive potential of pea aphids in terms of longevity and progeny production (S ection 3.4). Aphids parasitised in their early instars (1 to 3 d old) die before repr oduction but those parasitised at later instar stages (4 to 6 d old) may reach the adult stage and produce a limited number of progeny before mummification; parasitised a dults (7 to 10 d old) have significantly shorter reproductive period and produce significantly fewer progeny. These results suggest that attacking young hosts will be more effective in suppressing aphid populations. However, A. ervi females prefer aphids that are 3 to 5 d old over the younger (1 and 2 d old) and older (6 and 7 d old) aphids for oviposition (Section 4.3). It is found that females that oviposit in 4- to 7- d-old aphids have more fitness gains in terms of progeny body size and egg load in female progeny (Section 4.3) and sperm load in male progeny at emergence (detected by proportion of female progeny in Section 4.4), but older aphids are more cap able of physically defending themselves and escaping from parasitisation and thus cost more energy and time to the females in oviposition. Therefore, females’ preference for 3- to 5-d-old a phids for oviposition may be based on the optimal balance between costs in terms of host defense and fitness gains in terms of progeny quality. Moreover, A. ervi more likely deposits fertilised eggs in large hosts and unfertilised eggs in small ones, supporing the host size-dependent sex allocation theory in parasitic Hymenoptera (Charnov et al. 1981; Charnov 1982). The success of a parasitoid species in suppressing host populations greatly depends on its ability to find and parasitise hosts and to increase offspring numbers in response to increasing host density (W aage and Hassell 1982; Mackauer 1983). The number of aphids parasitised and eggs laid by A. ervi significantly increases with an . General Discussion and Conclusion 122 increase of host density. However, the number of eggs laid in each parasitised aphid significantly decreases (from 2.9 to 1.3 eggs) with the increase of host density from 15 to 75 aphids. These results suggest that the parasitoid adjusts oviposition strategy in response to increasing host density, and a host density equivalent to 50 aphids is the highest critical density where a parasitoid could maximise its control efficiency in 24 h period. Moreover, the highest proportio n of female progeny is produced at host densities of 50 and 75 aphids, suggestin g that the sperm limit occurs when host density reaches these density levels. Howeve r, at low host density of 15 and 25 aphids, the proportion of female progeny is also low, indicating that A. ervi females can adjust the sex allocation strategy in response to the host density rather than the limitation of sperm supply; this may be advantageous to parasitoids to persist under low host density conditions (Tri pathi and Singh 1991). On the basis of the above reproductive strategy, it is expected that parasitoids released in the early season when growing populations have a high proportion of young aphids (Schowalter 2000) should be more effective in inhibiting aphids’ population build-up than later in the season. Moreover, in the labor atory mass rearing, aphids that are 3 to 5 d old appear to be the appropriate hosts because they give relatively higher quality of progeny, and 50 to 75 aphids /female/day may be provided to parasitoids as hosts to produce higher female progeny populations. 6.4 Factors Affecting Mate Choice and Reproductive Fitness of A. ervi Factors affecting mate choice and reproductive fitness of A. ervi include mate body size and age, and male mating history. Reproductive fitness of both sexes is pos itively correlated with their body size. For example, large individuals have grea ter longevity, large males father more progeny, and large females have higher fecund ity and parasitism and greater ability in host searching and handling (Section 4.4). Ther efore, it is not su rprising that when given a choice of small and large females, both small and large males prefer large females for mating (Section 5.4), because these females had higher reproductive . General Discussion and Conclusion 123 potential and in turn males have maximum re productive fitness return in term of their female progeny. With increasing body size, females gain more than males in longevity and fecundity but males gain more than females in the number of daughters produced (Section 4.4). The asymmetrical size-fitn ess between sexes may be because the different reproductive strategies of different sexes, i.e. the reproductive success of A. ervi females directly relies on their potential fecundity and their ability to search and parasitise hosts whereas the success of male reproduction is absolutely mediated by females, and virgin females still reproduce with the absence of males. Thus, when given a choice of small and large males, both small and large females do not show any preference for small or large males for mating (Section 5.4). The production of daughter declines with mate age at mating. For the aging males, the lower production of daughters may be due to the diminishing sperm viability and mobility related to aging rather than sperm depletion as reported in L. delhiensis (Srivastava and Singh 1995). Evidence for this is found in the daily experiment (Section 4.5), where a male c ould inseminate up to six females with similar number of daughters produced (> 50%) if allowed 24-h recovery period between matings. When three males of different ages compete for one mate, old males achieve significantly fewer matings (Sec tion 5.4). These may be the results of decreased ability to recognize and res pond to females’ sex pheromone (Vetter and Visscher 1997; McClure et al. 2007) or decreased activity with aging (Cheng et al. 2003). For the aging females, the reducti on of daughter production could be explained by the constraints in fertilization process in A. ervi because in parasitoid females, the depletion of spermathecal gland secretion may result in a reduced availability of activated sperm for fertilization of ovulating e ggs (Flanders 1946), and eggs may contain certain kind of sperm-attr acting chemicals, which decrease with the female age, leading to a reduced fertilization (Pande y et al. 1983). A. ervi males may deliver fewer sperm to old females as found in L. delhiensis (Srivastava and Singh 1995). Mating behaviour observation (Section 5.4) shows that although 1-d-old males do not suffer from sperm limitation, they di smount the old females more frequently, . General Discussion and Conclusion 124 shorten the mating period, and are more like ly to perform the second mating after they have mated with the old females within a 10 minute period. Results also indicate that between 20 and 40% of females when mating at the age of 5 d old do not produce any daughters (Section 4.5). A. ervi males are more likely to approach, mount and mate young females, because younger females have greater pote ntial fecundity. Moreover, in A. ervi , females release a sex pheromone that tri ggers males’ behavioural sequence (Battaglia et al. 2002; Mc Clure et al. 2007), but se x pheromones produced and released by females decrease with their age (McClure et al. 2007). Therefore, older females are less attractive to the males. Mating in A. ervi does not increase females’ fecundity and longevity (Section 4.4 and 4.5), suggesting that A. ervi females do not obtain a nutritional contribution from males during copulation. These results support Godfray’s (1994) conclusion that this may be a general rule in hymenopteran parasitoids, which is found in other parasitoid wasps (Madel et al. 1990; Srivastava a nd Singh 1995; Fauvergue 1998; Cheng et al. 2003; Jacob and Boiv in 2004; Khanh et al. 2005). 6.5 Mating Behaviour of A. ervi Aphidius ervi males appear to play a more active role in mating behaviour. This may be because a virgin female can reproduce without mating but the success of male reproduction absolutely relies on, a nd is mediated by females (Quicke 1997). The courtship behaviours of males (i.e. wing fanning, approaching and mounting) are associated with the release of female sex pheromone (McClure et al. 2007), indicating that in A. ervi , females produce signals and males do the searching. It is found that both virgin and once-mated males father similar number of their progeny (hourly mating experiments in Section 4. 5), thus females are not expected to have preference between virgin and once-mated males for mating. However, once-mated males achieve more matings. These results suggest that mated males have learnt from the previous mating so that they may respond to females more quickly and perform better courtships than naïve males (Perez-Lachaud and Campan 1995; Fischer and King 2005). . General Discussion and Conclusion 125 Homosexual behaviour and mate competition between A. ervi males are common and tend to be more intensive when the OSR, male density and population density increase, especially in the extr emely male-biased conditions. These may be due to the increasing encounter rate between searching males. Thornhill and Alcock (1983) propose that in insects, male homosexual mounting may be an advantageous strategy of mate location, if the males’ main cost of reproduction is the time consumed in discerning the sex of potential mates. Evidence also shows that external stimuli such as visual cues may result in male mistake in courtship in some insects (Thornhill and Alcock 1983; Wang et al. 1996) including hymenopteran parasitoids (van den Assem 1986). Parasitoid females usually switch off their receptivity following mating due to the presence of sperm in the spermatheca or the additional substances transformed by the male during copulation (Manning 1962; Merle 1968). More than 82% mated females of A. ervi are unattractive to males soon after mating, suggesting they have switched off their responsiveness. Switching-off of receptivity by females may benefit both sexes (van den Assem 1986; Ki ng et al. 2005). To females of A. ervi , because they have received sufficient sperm during the first mating (Sections 4.4 and 4.5), preventing further mating may enable them to concentrate on host searching and oviposition. To males, the female unreceptivity may reduce the probability of males to mate with the already mated females and hence avoid sperm competition. Under the male-biased OSR, a proportion of females (3~17%) accept the second males for mating within 1 min after the termination of the first mating, suggesting the female’s switching-off of recep tivity is a gradual process. The shorter mating period in the second mating suggests that female remating could result from the above gradual process rather than the insufficient sperm transformation during the first mating, because one mating is suffici ent to fertilise a female for life. Under the male-biased OSR, high male de nsity or when other rivals compete for mating, A. ervi males prolong copulation duration. The reason may be that under these circumstances, the probability that a male will find another mate decreases but possibility that his mate will remate increases. In Aphidiidae, males will successfully . General Discussion and Conclusion 126 deliver sperm when copulation lasts for longer than 25 seconds (Wiackowski 1962; Straý 1970). Thus, sperm may be added by the second male if A. ervi females remate. Therefore, A. ervi males’ prolonged copulation may se rve as mate-guarding in order to reduce the probability of female remating and thus sperm competition. 6.6 Conclusion In this thesis, I have reported and discussed the main findings of the reproductive behaviour, and factors a ffecting the reproductive fitness of A. ervi. This work has provided a much firmer basis of knowledge of this sp ecies, and a rounded perspective of its reproductive biology. Such knowledge is vital, as noted in this thesis, to appraising prospects for further investigation in improvement of biological control strategies. . References 127 REFERENCES Abe J., Kamimura Y., Kondo N. and Shimada M. 2003. Extremely female-biased sex ratio and lethal male-male combat in a parasitoid wasp, Melittobia australica (Eulophidae). Behavioral Ecology 14: 34-39. Ahnesjö I., Vincent A., Alatalo R.V., Halliday T. and Sutherland W.J. 1993. The role of females in influencing mating patterns. Behavioural Ecology 4: 187- 189. Alcock J. 1979. The behavioural consequences of size variation among males of the territorial wasp Hemipepsis ustulata ( Hym., Pompilidae). Behaviour 71: 3 2 2 - 335. Alcock J. 1984. Long term maintenance of size va riation in two populations of Centris pallida ( Hymenoptera: Anthophoridae). Evolution 38: 2 2 0 - 2 2 3. Alcock J. and Gwynne D.T. 1987. Courtship feeding and mate choice in thynnine wasps (Hymenoptera: Tiphiidae). Australian Journal Zoology 35: 3 9 - 4 7. Allen G.R., Kazmer D.J., and Luck R.F. 1994. Postcopulatory male behavior, sperm precedence and multiple mating in a solitary parasitoid wasp. Animal Behaviour 48: 635-644. Alonso-Pimentel H. and Papaj D.R. 1996. Operational sex ra tio versus gender density as determinants of copulation duration in the walnut fly, Rhagoletis juglandis (Diptera: Tephritidae ). Behavioural Ecol ogy and Sociobiology 39: 171-180. Anon 2005. Behavioural ecology of insect para sitoids. European Science Foundation. http://bepar.sophia.inra.f r/telechargement/bepar_brochure.pdf. Accessed on 28 June 2007. Arakawa R., Miura M. and Fujita M. 2004. Effects of host species on the body size, fecundity, and longevity of Trissolcus mitsukurii (Hymenoptera): parasitoid of stink bugs. Applied Entomo logy and Zoology 39: 177-181. . References 128 Arakawa R., Miura M. and Fujita M. 2004. Effects of host species on the body size, fecundity, and longevity of Trissolcus mitsukurii (Hymenoptera): parasitoid of stink bugs. Applied Entomo logy and Zoology 39: 177-181. Archibald R.D., Cox J.M. and Deitz L.L. 1979. New records of plant pests in New Zealand. III. Six sp ecies of Homoptera. New Zeal and Journal of Agricultural Research 22: 201-207. Armstrong S.M., Barratt B.I.P. and Evans A.A. 1996. Circadian pattern of oviposition in the parasitoids Microctonus aethiopoides Loan and M. hyperodae Loan (Hymenoptera: Braconidae) , in relation to host activity. Proceedings of the Fourty-ninth New Zealand Plant Protection Conference: 280-284. Arnqvist G. 1992. Pre-copulatory fighting in a water strider: inter-sexual conflict or mate assessment? Animal Behavior 43: 559-567. Bai B. and Mackauer M. 1991. Recognition of heterospecific parasitism: competition between aphidiid ( Aphidius ervi ) and aphelinid ( Aphelinus asychis) parasitoids of aphids (Hymenoptera: Aphidiidae; Aphelinidae). Journal of Insect Behavior 4: 333-345. Bateman P.W. 2000. The influence of weapon asymmetry on male-male competition success in a sexually dimorphic insect the African king cricket Libanasidus vittatus ( Orthoptera: Anostostomatidae) . Journal Insect Behavior 13: 157-163. Battaglia D., Isidoro N., Romani R., Bin F. and Pennacchio F. 2002. Mating behaviour of Aphidius ervi (Hymenoptera: Braconidae) : the role of antennae. European Journal of Entomology 99: 451-456. Battaglia D., Pennacchio F., Marincola G. and Tranfaglia A. 1993. Cornicle secretion of Acyrthosiphon pisum (Homoptera: Aphididae) as a contact kairomone for the parasitoid Aphidius ervi (Hymenoptera: Braconidae). European Journal of Entomology 90: 423-428. . References 129 Battaglia D., Pennacchio F., Romano A. and Tranfaglia A. 1995. The role of physical cues in the regulation of host recognition and acceptance behaviour of Aphidius ervi Haliday (Hymenoptera: Brac onidae). Journal of Insect Behavior 8: 739-750. Battaglia D., Poppy G., Powell W., Romano A., Tranfaglia A., Pennacchio F. 2000. Physical and chemical cues influencing the oviposition behaviour of Aphidius ervi. Entomologia Experimentalis et Applicata 94: 219-227. Beck S. D. 1980. Insect photoperiodism. Academic Press, New York. Blackman R.L. and Eastop V.F. 2000. Aphids on the world’s crops: an identification and information guide. John Wiley and Sons, New York. Boivin G., Jacob S. and Damiens D. 2005. Spermatogeny as a life-history index in parasitoid wasps. Oecologia 143: 198-202. Bonduriansky R. 2001. The evolution of male mate choice in insects: a synthesis of ideas and evidence. Biolgical Reviews 76: 305-339. Box G. E.P. and Draper N.R. 1987. Empirical model-buildi ng and response surfaces. John Wiley and Sons, New York. Bueno B. H. P., Gutierrez A. P. and Ruggle P. 1993. Parasitism by Aphidius ervi (Hym.: Aphidiidae): preference for pea aphid and blue alfalfa aphid (Hom.: Aphididae) and competition with A. smithi. Entomophaga 38: 273-284. Calvert D.J. and van den Bosch R. 1972. Behavior and biology of Monoctonus paulensis (Hymenoptera: Braconidae), a parasite of dactynotine Aphids. Annals of the Entomological Society of America 65: 773-77 9. Cameron P.J. and Walker G.P. 1989. Release and establishment of Aphidius spp. (Hymenoptera: Aphidiidae), parasitoids of pea aphid and blue green aphid in New Zealand. New Zealand Journal of Agricultural Research 32: 281-290. . References 130 Cameron P.J., Allan D.J., Walker G.P. and Wightman J.A. 1983. Management experiments on aphids ( Acyrthosiphon spp.) and beneficial insects in lucerne. New Zealand Journal of Experimental Agriculture 11: 343-349. Cameron P.J., Thomas W.P. and Hill R.L. 1979. Introduction of lucerne aphid parasites and a preliminary evaluation of the natural enemies of Acyrthosiphon spp. (Hemiptera: Aphididae) in New Zealand. Proceedings of the Second Australasian Conference on Grassl and Invertebrate Ecology 219-223. Campbell A. and Mackauer M. 1975. The effect of parasitism by Aphidius smithi (Hymenoptera: Aphididae) on reproducti on and population growth of the pea aphid (Homoptera: Aphididae). Canadian Entomologist 107: 919-926. Carver M. 1989. Biological control of aphids. pp. 141-165, In A.K. Minks and P. Harrewijn [eds.], Aphids: th eir natural enemies and control. Volume C. World crop pests. Amsterdam, New York, Elsevier. Caswell H. 2001. Matrix population models: constructi on, analysis, and interpretation. Sinauer Associates, Sunderland, Massachosetts. Ceballo F.A. and Walter G.H. 2004. Life history parameters and biocontrol potential of the mealybug parasitoid Coccidoxenoides peregrinus (Timberlake) (Hymenoptera: Encyrtidae): asexuality, fecundity and oviposition patterns. Biological Control 29: 235-244. Charnov E.L. 1982. The theory of sex allocation. Princeton University Press, Princeton, New Jersey. Charnov E.L. and Stephens D.W. 1988. On the evolution of host selection in solitary parasitoids. American Naturalist 132: 707-722. Charnov E.L., Hartogh R.L., Los-den Jones W.T. and van den Assem J. 1981. Sex ratio evolution in a variable environment. Nature 289: 27-33. . References 131 Chau A. and Mackauer M. 2001. Host-instar selection in the aphid parasitoid Monoctonus paulensis (Hymenoptera: Braconidae, Aphidiinae): assessing costs and benefits. Canadian Entomologist 133: 549-564. Cheng L., Howard R.W., Campbell J.F., Charlton R.E., Nechols J.R. and Ramaswamy S. 2003. Behavioral interac tion between males of Cephalonomia tarsalis (Ashmead) (Hymenoptera: Bethylidae) competing for females. Journal of Insect Behavior 16: 625-645. Christiansen-Weniger P. and Hardie J. 1997. Development of the aphid parasitoid, Aphidius ervi , in asexual and sexual females of the pea aphid, Acyrthosiphon pisum, and the blackberry-cereal aphid, Sitobion fragariae. Entomophaga: 165-172. Chua T.H., Gonzalez D. and Bellows T. 1990. Searching efficiency and multiparasitism in Aphidius smithi and Aphidius ervi ( Hymenoptera, Aphidiidae), parasi tes of pea aphid, Acyrthosiphon pisum ( Homoptera, Aphididae). Journal of Applied Entomology 110:101- 1 0 6. Cloutier C., Duperron J., Tertuliano M. and McNeil J.N. 2000. Host instar, body size and fitness in the koinobiotic parasitoid Aphidius nigripes. Entomologia Experimentalis et Applicata 97: 29-40. Clutton-Brock T.H., Rose K.E. and Guinness F.E. 1997. Density related changes in sexual selection in red deer. Proceedings of the Royal Society of London, Series B 264: 1509-1516. Collins M.D. and Dixon A.F.G. 1986. The effect of egg depletion on the foraging behaviour of an aphid parasitoid. Journal of Applied Entomology 102: 342- 352. Cook D.F. 1995. Influence of previous mating experience on future mating success in male Lucilia cuprina (Diptera: Calliphoridae). Jour nal of Insect Behavior 8: 207-217. . References 132 Corrigan J.E., Laing J.E. and Zubricky J.S. 1995. Effects of paras itoid to host ratio and time of day of parasitism on development and emergence of Trichogramma minutum (Hymenoptera: Trichogramma tidae) parasitizing eggs of Ephestia kuehniella (Lepidoptera: Pyralidae). A nnals of the Entomological Society of America 8 8 : 773-78 0. Couch K.M., Cresswell A.S., Barratt B.I.P. and Evans A.A. 1997. Implications of host weevil circadian activity for parasitism by Microctonus aethiopoides (Hymenoptera: Braconidae). Proceedings of the Fiftieth New Zealand Plant Protection Conference: 227-231. Damiens D. and Boivin G. 2005. Male reproductive strategy in Trichogramma evanescens: sperm production and allocation to females. Physiological Entomology 30: 241-24 7. Damiens D. and Boivin G. 2006. Why do sperm-depleted parasitoid males continue to mate? Behavioral Ecology 17:138-143. Daniel W.W. 1990. Applied nonparametric statistics. Boston, MA: PWS-KENT Danielsson I. 2001. Antagonistic pre- and post-copul atory sexual selection on male body size in a water strider ( Gerris lacustris). Proceedings of the Royal Society of London, Series B 268:77-81. Darwin C. 1859. The origin of species. John Murray, London. David S.K. and Ghorpade K.D. 1974. Two species of aphids (Homoptera: Aphididae) new to India and four othe rs new to Southern India. Oriental Insects 8: 195-198. DeBach P. and Rosen D. 1991. Biological control by natural enemies. Cambridge University Press, Cambridge. del Mazo-Cancino A., Malo E.A., Cruz-López L. and Rojas J.C. 2004. Diel periodicity and influence of age and mating on female sex pheromone titre in . References 133 Estigmene acrea (Lep., Arctiidae). Journal of Applied Entomology 128: 459- 463. Dent D.R. 1991. Insect pest management. CAB International, Wallingford. Dent D.R. 1995. Integrated pest management. Chapman and Hall, London. Desneux N., Wajnberg E., Fauvergue X., Privet S. and Kaiser L. 2004. Oviposition behaviour and patch-time allocation in two aphid parasitoids exposed to deltamethrin residues. Ento mologia Experimentalis et Applicata 112: 227-235. Dick J.T.A. and Elwood R.W. 1996. Effects of natural vari ation in sex ratio and habitat structure on mate-guarding decisions in amphipods (Crustacea). Behaviour 133: 985-996. Dickson R.C. 1975. Identity, origin and host range of the blue alfalfa aphid. The Fifth California Alfalfa Symposium: 2 2 - 2 3. Digilio M.C., Pennacchio F. and Tremblay E. 1998. Host regulation effects of ovary fluid and venom of Aphidius ervi (Hymenoptera: Braconidae). Journal of Insect Physiology 44: 779-784. Dixon A.F.G. 1985. Aphid ecology. Glasgow, Blackie. Dolphin K. and Quicke D.L.J. 2001. Estimating the global species richness of an incompletely described taxon: an example using parasitoid wasps (Hymenoptera: Braconidae). Biological Journal of the Linnean Society. 7 3 : 279-286. Doutt R.L. 1959. The biology of parasitic Hy menoptera. Annual Review of Entomology 4: 161-182. Du Y.J., Poppy G.M. and Powell W. 1996. Relative importance of semiochemicals from first and second trophic levels in host foraging behavior of Aphidius ervi. Journal of Chemical Ecology 22: 1591-16 0 5. . References 134 Du Y.J., Poppy G.M., Powell W. and Wadhams L.J. 1997. Chemically mediated associative learning in the host foraging behavior of the aphid parasitoid Aphidius ervi (Hymenoptera: Braconidae). Jour nal of Insect Behavior 10: 509- 522. Du Y.J., Poppy G.M., Powell W., Pickett J.A., Wadhams L.J. and Woodcock C.M. 1998. Identification of semiochemical s released during aphid feeding that attract parasitoid Aphidius ervi. Journal of Chemical Ecology 24: 1355- 1368. Dunbier M.W. and Easton H.S. 1982. Longer stand life with new cultivars. pp. 121- 126, In R.B. Wynn-Williams [ed.], Lucerne for the 80’s. Agronomy Society of New Zealand, Special Publication 1. El Agoze M., Poirie M. and Periquet G. 1995. Precedence of the first male sperm in successive matings in the Hymenoptera Diadromus pulchellus. Entomologia Experimetales et Applicata 75: 251-255. Enders M.M. 1993. The effect of male size and opera tional sex ratio on male mating success in the common spider mite, Tetranychus urticae Koch (Acari: Tetranychidae). American Naturalist 46: 835-846. Fagerström T. and Wiklund C. 1982. Why do males emerge before females? Protandry as a mating system strategy in male and female butterflies. Oecologia 52: 164-166. Fantinou A.A., Alexandri M.P. and Tsitsipis J.A. 1998. Adult emergence rhythm of the egg-parasitoid Telenomus busseolae. BioControl 43: 141-151. Fauvergue X., Hopper K.R., Antolin M.F. and Kazmer D.J. 1998. Does time until mating affect progeny sex ratio? A manipulative experiment with the parasitoid wasp Aphelinus asychis. Journal of Evolutionary Biology 11: 611- 622. Fischer C.R. and King B.H. 2005. The effect of experience with a female on male mating behavior in the parasitoid wasp Spalangia endius. Tewnty-fifth . References 135 Midwest Ecology and Evolution Conf erence (Abstract), March 11-13, Southern Illinois University, Carbondale, USA. Fisher R.A. 1930. The genetical theory of natural selection. Clarendon Press. Flanagn K.E., West S.A. and Godfray H.C.J. 1998. Local mate competition, variable fecundity and information use in a parasitoid. Animal Behaviour 56: 191-198. Flanders S.E. 1946. Control of sex and sex-limited polymorphism in Hymenoptera. Quarterly Review of Biology 21: 135-143. Flanders S.E. 1950. Regulation of ovulation and e gg disposal in the parasitic Hymenoptera. Canadian Entomologist 82:134- 1 4 0. Gerling D. and Legner E.F. 1968. Developmental history and reproduction of Spalangia cameroni , parasite of synanthropic flies. Annals of the Entomological Society of America 61: 1436-1 4 4 3. Gerling D., Roitberg B.D. and Mackauer M. 1990. Instar specific defense of the pea aphid, Acyrthosiphon pisum: inference on oviposition success of the hymenopterous parasite Aphelinus asychis. Journal of Insect Behaviour 3: 501-514. Gillespie R.G. 1991. Homosxual mating behavior in male Doryonychus Raptor (Araneat, Tetragnathidae). The J ournal of Arachnology 19: 229-230. Glinwood R.T., Du Y.J. and PowellW. 1999. Responses to aphid sex pheromones by the pea aphid parasitoids Aphidius ervi and Aphidius eadyi. Entomologia Experimentalis et Applicata 92: 227-232. Godfray H.C.J. 1994. Parasitoids: behavioral and e volutionary ecology. Princeton University Press, Princeton, New Jersey. Goff A.M. and Nault L.R. 1984. Response of the pea aphid parasite Aphidius ervi Haliday (Hymenoptera: Aphidiidae) to transmitted light. Environmental Entomology 13: 595-59 8. . References 136 González D., Miyazaki M., White W., Takada H., Dickson R.C. and Hall J.C. 1979. Geographical distribution of Acyrthosiphon kondoi Shinji (Homoptera: Aphididae) and some of its parasites and hyperparasites in Japan. Kontyu 47: 1-7. González D., White W., Hall J. and Dickson R.C. 1978. Geographical distribution of Aphidiidae (Hym.) imported to Ca lifornia for biological control of Acyrthosiphon kondoi and Acyrthosiphon pisum (Hom.: Aphididae). Entomophaga 23: 239-248. Gordh G. and DeBach P. 1976. Male inseminative potential in Aphytis lingnanensis (Hymenoptera: Aphelinidae). Cana dian Entomologist 108: 583-589. Gordh G. and Debach P. 1978. Courtship behavior in the Aphytis lingnanensis group, its potential usefulness in taxonomy, and a review of sexual behavior in the parasitic Hymenoptera (Cha lc.; Aphelinidae). Hilgardia 4 6 : 37-75. Greathead D.J. and Pschorn-Walcher P. 1976. Apple woolly aphids - Erisoma lanigerum (Hausm.) (Aphididae, Hem.). In D.J.Greathead [ed.], A review of biological control in Weastern and Southern Europe. Commonwealth Institute of Biological Control, Tec hnical Communication 7: 4-13. Grewal H.S. and Williams R. 2001. Lucerne varieties differ in their response to liming on an acid soil. In B.A Rowe [ed], Proceedings of the 10th Australian agronomy conference (CD-ROM). The Australia Society of Agronomy, Hobart. Guerrieri E., Pennacchio F. and Tremblay E. 1993. Flight behaviour of the aphid parasitoid Aphidius ervi (Hymenoptera: Braconidae) in response to plant and host volatiles. European Journal of Entomology 90: 415-421. Guerrieri E., Pennacchio F. and Tremblay E. 1997. Effect of adu lt experience on in-flight orientation to plant and plant-host complex volatiles in Aphidius ervi Haliday (Hymenoptera, Braconida e). Biological Control 10: 159-165. . References 137 Gutierrez A.P. 1987. Systems analysis in solving pest management problems. pp. 71- 82, In V. Delucchi [ed.], Protection intgre: quo vadis? - “Parasitis 86”, Parasitis, Genva. Hågvar E.B., Hofsvang T. 1991. Aphid parasitoids (Hyme noptera, Aphidiidae): biology, host selection and use in biol ogical control. Biocontrol News and Information 12: 13-41. Hall R.W. and Ehler L.E. 1979. Rate of establishment of natural enemies in classical biological control. Bulletin of the En tomological Society of America 25: 280- 282. Hall R.W., Ehler L.E. and Bisabri-Ershadi B. 1980. Rate of success in classical biological control of arthro. Bulletin of the Entomological Society of America 26: 111-11 4. Hamilton W.D. 1967. Extraordinary sex ratios. Science 156: 477-488. Hamilton W.D. 1979. Wingless and fighting males in fi g wasps and other insects. pp 167-220 , In M.S. Blum and N.A. Blum [eds], Sexual selection and reproductive competition in insects. Academic Press, New York. Hanson A.A., Barnes D.K. and Hill R.R. Jr. 1988. Alfalfa and alfalfa improvement. American Society of Agronomy, Madison, Wisconsin, USA. Hardy I.C.W. 1992. Non-binomial sex allocation and brood sex ratio variances in the parasitoid Hymenoptera. Oikos 65: 143-158. Hardy I.C.W. 1994. Sex ratio and mating structure in the parasitoid Hymenoptera. Oikos 69: 3–20. Hardy I.C.W., Dijkstra L.J., Gillis J.E.M. and Luft P.A. 1998. Patterns of sex ratio, virginity and developmental mortality in gregarious parasitoids. Biological Journal of the Linnean Society 64: 239-270. . References 138 Hardy I.C.W., Ode P.J and Siva-Jothy M.T. 2005. Mating behaviour. pp. 219-260, In M.A. Jervis [ed.], Insects as natu ral enemies: a practical perspective. Springer, Dordrecht, New York. Harper A.M., Miska J.P., Manglitz G.R., Irwin B.J. and Armbrust E.J. 1978. The literature of arthropods associated with alfalfa. III. A bibliography of the pea aphid, Acyrthosiphon pisum (Harris) (Homoptera: Aphididae). Special Publication, Agricultural Experiment Station, College of Agriculture, University of Illinois at Urbana-Champaign. Hassell M.P. 1986. Parasitoids and population regulation. pp. 201-224, In J. Waage and D. Greathead [eds.], Insect parasitoids. Academic Press, London. Hassell M.P. and Waage J.K. 1984. Host-parasitoid populati on interactions. Annual Review of Entomology 29: 89-114. He X.Z., Wang Q. and Teulon D.A.J. 2003. Effect of parasitism by Aphidius eadyi (Hymenoptera: Aphidiidae) on reproduction of pea aphid, Acyrthosiphon pisum (Hemiptera: Aphididae). New Zealand Plant Protection 56: 185-189. He X.Z., Wang Q. and Teulon D.A.J. 2004. Emergence, sexual maturation and oviposition of Aphidius ervi (Hymenoptera: Aphidiidae). New Zealand Plant Protection 57: 214-220. Heinz K.M. 1991. Sex-specific reproductive consequen ces of body size in the solitary ectoparasitoid Diglyphus begini. Evolution 45: 1511-1515. Henter H.J. 2004. Constrained sex allocation in a para sitoid due to variation in male quality. Journal of Evol utionary Biology 17: 886-896. Hirose Y. 2006. Biological control of aphids and coccids: a comparative analysis. Population Ecology 48: 307-315. Holling C.S. 1959. Some characteristics of simple types of predation and parasitism. Canadian Entomologist 91: 385-39 8. . References 139 Holmes H. B. 1974. Patterns of sperm competition in Nasonia vitripennis. Canadian Journal of Genetics and Cytology 16: 789-795. Honék A. 1993. Intraspecific variation in body size a nd fecundity in insects: a general relationship. Oikos 66: 483-492. Hughs A. L. and Hughs M. K. 1982. Male size, mating succe ss, and breeding habitat partitioning in the white spotted sawyer Monochamus scutellatus ( Say) (Coleoptera: Cerambycidae). Oecologia 55: 258-263. Huignard J. 1983. Transfer and fate of male secretions deposited in the spermatophore of females of Acanthoscelides obtectus Say (Coleoptera Bruchidae). Journal of In sect Physiology 29: 55-63. Isidoro N. and Bin F. 1995. Male antennal gland of Amitus spiniferus (Brethes) (Hymenoptera: Platygastridae), likel y involved in courtship behavior. International Journal of Insect Morphology and Embryology 24: 365-373. Isidoro N., Bin F., Colazza S. and Vinson B.S. 1996. Morphology of antennal gustatory sensilla and glands in some parasitoid Hymenoptera with hypothesis on their role in sex and host recogniti on. Journal of Hymeno ptera Research 5: 206-239. Ives A.R., Schooler S.S., Jagar V.J., Knuteson S.E., Grbic M. and Settle W.H. 1999. Variability and parasitoid foraging efficiency: a case study of pea aphids and Aphidius ervi. American Naturalist 154: 652-673. Jacob S. and Boivin G. 2004. Insemination potential of male Trichogramma evanescens. Entomologia Experimentalis et Applicata 113: 181-186. Jennions M.D. and Petrie M. 1997. Variation in mate choice and mating preferences: a review of causes and consequences. Biological Reviews of the Cambridge Philosophical Society 72: 283-327. Jennions M.D. and Petrie M. 2000. Why do females mate multiply? A review of the genetic benefits. Biolgical Reviews 75:21-65. . References 140 Jervis M.A., Copland M.J.W. and Harvey J.A. 2005. The life cycle. pp. 73-165, In M.A. Jervis [ed.], Insects as natural enem ies: a practical pers pective. Springer, Dordrecht, New York. Jervis M.A., Ferns P. and Heimpel G.E. 2003. Body size and the timing of reproduction in parasitoid wasps: a co mparative analysis. Functional Ecology 17: 375-383. Jervis M.A., Heimpel G.E., Ferns P., Harvey J. and Kidd N.A.C. 2001. Life- history strategies of parasitoid wasps: a comparative analysis of ‘ovigeny’. Journal of Animal Ecology 70: 442-458. Jiménez-Pérez A. and Wang Q. 2004. Effect of body weight on reproductive performance in Cnephasia jactatana (Lepidoptera: Tortrici dae). Journal Insect Behavior 17: 511-523. Jirotkul M. 1999. Operational sex ratio influences female preference and male-male competition in guppies. Animal Behaviour 58: 287-29 4. Jobloński P. and Vepsäläinen K. 1995. Conflict between sexes in the water strider, Gerris lacustris: a test of two hypotheses for male guarding behaviour. Behavioral Ecology 6: 388-392. Johnson L.K. 1982. Sexual selection in a tropica l brentid weevil. Evolution 36: 251- 262. Juliano S.A. 1985. The effects of body size on mating and reproduction in Brachinus lateralis (Coleoptera: Carabidae). Ecological Entomology 10: 2 7 1 - 2 8 0. Kain W.M. and Biggs D.R. 1980. Effect of pea aphid and bluegreen lucerne aphid ( Acyrthosiphon spp.) on coumestrol levels in herbage of lucerne ( Medicago sativa). New Zealand Journal of Agri cultural Research 23: 563-568. Kain W.M. and Trought T.E.T. 1982. Insect pests of lucerne in New Zealand. pp. 49-59, In R.B. Wynn-Williams [ed.], Lucerne for the 80's. Agronomy Society of New Zealand, Special Publication 1. . References 141 Kain W.M., Atkinson D.S. and Oliver M.J. 1979a. Seasonality of blue green lucerne and pea aphid in the southern North Island of New Zealand. Proceedings of the Thirty-second Ne w Zealand weed and pest control conference: 180-185 . Kain W.M., Atkinson D.S., Marsden R.S., Oliver M.J. and Holland T.V. 1977. Blue-green lucerne aphid damage in lucerne crops within southern North Island. Proceedings of the Thirtieth New Zealand Weed and Pest Control Conference: 1 7 7 - 1 8 1. Kainoh Y. 1986. Mating behavior of Ascogaster reticulatus Watanabe (Hymenoptera: Braconidae), an egg-larval parasitoid of the smaller tea tortrix moth, Adoxophyes sp. (Lepidoptera: Tortricidae). I. Diel patterns of emergence and mating, and some conditions for mating. Applied Entomology and Zoology 21: 1-7. Kaneshiro K.Y. and Giddings L.V. 1987. The significance of asymmetrical sexual isolation and the formation of new species. Evolutionary Biology 21: 29-43. Karban R. 1983. Sexual selection, body size and sex-related mortality in the cicada Magicicada cassini. American Midland Naturalist 109: 324-330. Kavallieratos N.G., Lykouressis P., Sarlis G.P., Stathas G.J., Sanchıs A. and Athanassiou C.G. 2001. The Aphidiinae (Hymenoptera: Ichneumonoidea: Braconidae) of Greece. Phytoparasitica 29: 306-340. Kavallieratos N.G., Tomanović Ž., Sarlis G.P., Fasseas C. and Emmanouel N.E. 2006. A review of the genus Aphidius Nees in Greece (Hymenoptera: Braconidae: Aphidiinae) w ith the description of a new species. Journal of Natural History 40: 1179-119 7. Kavallieratos N.G., Tomanović Ź., Starý P., Athanassiou C.G., Sarlis G.P., Petrović O., Niketić M. and Anagnou-Veroniki M. 2004. A survey of aphid parasitoids (Hymenoptera: Braconidae: Aphidiinae) of southeastern Europe and their aphid–plant asso ciations. Applied Entomo logy and Zoology 39: 527- 563. . References 142 Kazmer D.J. and Luck R.F. 1995. Field tests of the size -fitness hypothesis in the egg parasitoid Trichogramma pretiosum. Ecology 76: 412-425. Khanh H.D.T., Bressac C. and Chevrier C. 2005. Male sperm donation consequences in single and double matings in Anisopteromalus calandrae. Physiological Entomology 30: 29-35. King B.H. 1987. Offspring sex ratios in parasitoid wasps. Quarterly Review of Biology 62: 367-396. King B.H. 2000. Sperm depletion and mating behavior in the parasitoid wasp Spalangia cameroni ( Hymenoptera: Pteromalidae). Great Lakes Entomologist 33: 117-127. King B.H., Saporito K.B., Ellison J.H. and Bratzke R.M. 2005. Unattractiveness of mated females to males in the parasitoid wasp Spalangia endius. Behavioural Ecology and Sociobiology 57: 350-356. King P.E. 1960. The passage of sperms to the spermatheca in Nasonia vitripennis. Entomologists’ Monthly Magazine 96: 163. Kouamé K.L. and Mackauer M. 1991. Influence of aphid size, age and behaviour on host choice by the parasitoid wasp Ephedrus californicus : a test of host-size models. Oecologia 88: 197-203. Krupa J.J. and Sih A. 1993. Experimental studies on water strider mating dynamics: spatial variation in density and sex ratio. Behavioral Ecology and Sociobiology 33: 107-120. Lampson L.J., Morse J.G. and Luck R.F. 1996. Host selection, sex allocation, and host feeding by Metaphycus helvolus (Hymenoptera: Encyrtidae) on Saissetia oleae (Homoptera: Coccidae) a nd its effect on parasitoid size, sex, and quality. Environmental Entomology 25: 283-294. Lauzière I., Perez-Lachaud G. abd Brodeur J. 2000. Effect of female body size and adult feeding on the fecundity and longevity of the parasitoid . References 143 Cephalonomia stephanoderis Betrem (Hymenoptera: Bethylidae). Annals of the Entomological Society of America 93: 103-10 9. Lawrence W.S. 1986. Male choice and competition in Tetraopes tetraophthalmus: effects of local sex ratio variation. Behavioral Ecology and Sociobiology 18: 289-296. Leatemia J.A., Laing J.E. and Corrigan J.E. 1995. Production of exclusively male progeny by mated, honey-fed Trichogramma minutum Riley (Hym.: Trichogrammatidae). Journal of A pplied Entomology 119: 561-566. Li S., Falabella P., Giannantonio S., Fanti P., Battaglia D., Digilio M.C., Volkl W., Sloggett J.J., Weisser W. and Pennacchio F. 2002. Pea aphid clonal resistance to the endophagous parasitoid Aphidius ervi. Journal of Insect Physiology 48: 971-980. Lin L.A. and Ives A.R. 2003. The effect of parasitoid host-size preference on host population growth rates: an example of Aphidius colemani and Aphis glycines. Ecological Entomology 28: 542-55 0. Liu S.S. 1985. Aspects of the numerical and functional response of the aphid parasite, Aphidius sonchi , in the laboratory. Entomologia Experimentalis et Applicata 37: 247-256. Liu S.S. and Hughes R.D. 1984. Effect of host age at parasitization on the development, survival, and reproduction of the sowthistle aphid, Hyperomyzus lactucae. Entomologia Experimentalis et Applicata 36: 239-24 6. Luck R.F. 1990. Evaluation of natural enemies for biological control: a behavioural approach. Trends in Ecology and Evolution 5: 196-199. Mackauer M. 1969. Sexual behaviour of and hybridi zation between three species of Aphidius Nees parasitic on the pea a phid (Hymenoptera: Aphidiidae) Proceedings of the Entomological Society of Washington 71: 339-35 1. . References 144 Mackauer M. 1983. Quantitative assessment of Aphidius smithi (Hymenoptera: Aphidiidae): fecundity, intrinsic rate of increase, and fu nctional response. Canadian Entomologist 115: 399-41 5. Mackauer M. and Kambhampati S. 1984. Reproduction and longevity of cabbage aphid, Brevicoryne brassicae (Homoptera: Aphidida e), parasitized by Diaeretiella rapae Hymenoptera: Aphidiidae). Canadian Entomologist 116: 1605-16 1 0. Mackauer M. and Kambhampati S. 1986. Structural changes in the parasite guild attacking the pea aphid in North Am erica. pp. 471-474 , In I. Hodek [ed.], Ecology of Aphidophaga. Academia, Prague and W. Junk, Dordrecht. Mackauer M. and Völkl W. 1993. Regulation of aphid population by aphidiid wasps: does parasitoid foraging behaviour or hyperparasitism limit impact? Oecologia 94: 339-350. Mackauer M. and Völkl W. 2002. Brood-size and sex-rati o variation in field populations of three species of solitary aphid parasitoids (Hymenoptera: Braconidae: Aphidiinae). Oecologia 131: 296-305. Madel G., Mühlen D. and Happe M. 1990. Diadegma semiclausum Hellen (Hymenoptera, Ichneumoni dae): copulation, spermatophore transfer, and offspring. Zeitschrift für Angewandte Zoologie 77: 347-56. Manning A. 1962. A sperm factor affecting the receptivity of Drosophila melanogaster females. Nature 194: 252-253. Marsh P.M. 1977. Notes on the taxonomy and nomenclature of Aphidius species (Hym.: Aphididae) parasitic on the pea aphid in North America. Entomophaga 22: 365-372. Martel V. and Boivin G. 2004. Premating dispersion in the egg parasitoid Trichogramma ( Hymenoptera: Trichogrammatidae). Environmental Entomology 33: 855-85 9. . References 145 McBrien H. and Mackauer M. 1990. Heterospecific larval competition and host discrimination in two species of aphid parasitoids: Aphidius ervi and Aphidius smithi. Entomologia Experimentalis et Applicata 56: 145-154. McCauley D.E. 1979. Geographic variation in body size and its re lation to the mating structure of Tetraopes populations. Heredity 42: 1 4 3 - 1 4 8. McClure M., Whistlecraft J. and McNeil J.N. 2007. Courtship behavior in relation to the female sex pheromone in the parasitoid, Aphidius ervi (Hymenoptera: Braconidae). Journal of Ch emical Ecology 33: 1946-19 5 9. McCullagh P. and Nelder J.A. 1989. Generalized linear models. Chapman and Hall, New York. McNeil J.N. and Brodeur J. 1995. Pheromone-mediated mating in the aphid parasitoid Aphidius nigripes (Hymenoptera, Aphididae). Journal of Chemcal Ecology 21: 959-972. McSweeney E.B. and Dunbier M.W. 1978. Blue-green lucerne aphid damage to lucerne seedlings - cultivar differen ces. New Zealand Journal of Ecology 1: 74-76. Mehrnejad M.R. and Copland M.J.W. 2006. Host-stage selec tion and oviposition behaviour of Psyllaephagus pistaciae, parasitoid of the common pistachio psylla Agonoscena pistaciae. Biological Control 36: 139-146. Merle J. 1968. Fonctionnement ovarien et réceptivité sexuelle de Drosophila melanogaster après implantation de fragment s de l’appareil genital male. Journal of insect Physiology 14: 1159-116 8. Mescheloff E. and Rosen D. 1988. Biosystematic studies on the Aphidiidae of Israel (Hymenoptera: Ichneumonoidea) 1. In troduction and key to genera. Israel Journal of Entomology 22: 61-73. . References 146 Mescheloff E. and Rosen D. 1990. Biosystematic studies on the Aphidiidae. 4. The genera Pauesia, Diaretus, Aphidius and Diaeretiella. Israel Journal of Entomology 24: 51-91. Micha S.G., Wellings P.W. and Morton R. 1992. Time-related rejection of parasitised hosts in the aphid parasitoid, Aphidius ervi. Entomologia Experimentalis et Applicata 62: 155-161. Michaud J.P. and Mackauer M. 1994. The use of visual cues in host evaluation by aphidiid wasps. I. Co mparison between three Aphidius parasitoids of the pea aphid. Entomologia Experimentalis et Applicata 70: 273-283. Michner G.R. and Mclean I.G. 1996. Reproductive behaviour and operational sex ratio in Richardson’s ground squirrels. Animal Behaviour 52: 743-758. Mills N.J. and Kuhlmann U. 2000. The relationship between egg load and fecundity among Trichogramma parasitoids. Ecological Entomology 25: 315-32 4. Mills N.J. and Lacan I. 2004. Ratio dependence in the functional response of insect parasitoids: evidence from Trichogramma minutum foraging for eggs in small host patches. Ecological Entomology 29: 208-216. Milne W.M. 1982. Imported parasites help contro l lucerne aphids. Agricultural Gazette of New South Wales 93: 15-17. Milne W.M. 1986. The release and establishment of Aphidius ervi Haliday (Hymenoptera: Ichneumonoidae) in lucerne aphid in eastern Australia. Journal of the Australian Entomo logical Society 25: 123-130. Mueller T.F., Blommers L.H.M. and Mols P.J.M. 1992. Woolly apple aphid ( Eriosoma lanigerum Hausm, Hom, Aphidae) parasitism by Aphelinus mali Hal (Hym, Aphelinidae) in relation to host stage and host colony size, shape and location. Journal of A pplied Entomology 114: 143-15 4. Nadel H. and Luck R.F. 1985. Span of female emergence and male sperm depletion in the female-biased, quasi-gregarious parasitoid, Pachycrepoideus vindemiae . References 147 (Hymenoptera: Pteromalidae). Annals of the Entomological Society of America 78: 410-4 1 4. Nadel H. and Luck R.F. 1992. Dispersal and mating structur e of a parasitoid with a female-biased sex ratio: implications for theory. Evolutionary Ecology 6: 270- 278. Nechols J.R. and Kikuchi R.S. 1985. Host selection of th e spherical mealybug (Homoptera: Pseudococcidae) by Anagyrus indicus (Hymenoptera: Encyrtidae): influence of host stage on parasitoid oviposition, development, sex ratio, and survival. Environmental Entomology 14: 32-37. Ode P.J., Antolin M.F. and Strand M.R. 1996. Sex allocation and sexual asymmetries in intra-brood competition in the parasitic wasp Bracon hebeter. Journal of Animal Ecology 65: 690-700. O'Donnell D.J. 1987. Larval development and the determination of the number of instars in aphid parasitoids (Hymenoptera: Aphidiidae). International Journal of Insect Morphology and Embryology 16: 3-15. Paine T.D., Millar J.G. and Hanks L.M. 2004. Effect of variati on in host size on sex ratio, size, a nd survival of Syngaster lepidus, a parasitoid of Eucalyptus longhorned beetles ( Phoracantha spp.): II. Biological Control 30: 374-381. Pandey R.K., Singh R., Kumar A., Tripathi C.P.M. and Sinha T.B. 1983. Bionomics of Trioxys ( Binodoxys) indicus , an aphidiid parasitoid of Aphis craccivora. 15. Influence of paras itoid’s age on its rate of oviposition and sex ratio of the offspring. Biological Agri calture and Hortic ulture 1: 211-218. Panhuis T.M., Butlin R., Zuk M. and Tregenza T. 2001. Sexual selection and speciation. Trends in Ecology and Evolution 16: 364-371. Parker G.A. 1970. Sperm competition and its evol utionary conseq uence in the insects. Biological Reviews 45: 525-567. . References 148 Parker G.A. and Simmons L.W. 1994. Evolution of phenotypic optima and copula duration in dung flies. Nature 370: 53-56. Parker G.A. and Simmons L.W. 1996. Parental investment and the control of sexual selection: Predicting the direction of sexual competition. Proceedings of the Royal Society of London, Series B 263: 315-321. Passlow T. 1977. The blue-green aphid - a furthe r new pest of lucerne. Queensland Agricultural Journal 1 0 3 : 403-404. Pennacchio F. and Digilio M.C. 1990. Morphology and development of larval instars of Aphidius ervi Haliday (Hymenoptera, Br aconidae, Aphidiinae). Bolletino del Laboratorio di Entomol ogia Agraria, Filippo Silvestri 46: 163- 174. Pennacchio F., Digilio M.C. and Tremblay E. 1995. Biochemical and metabolic alterations in Acyrthosiphon pisum parasitized by Aphidius ervi. Archives of Insect Biochemistry and Physiology 30: 351-367. Pennacchio F., Digilio M.C., Tremblay E. and Tranfaglia A. 1994. Host recognition and acceptance behaviour in two aphid parasitoid species: Aphidius ervi Haliday and Aphidius microlophii Pennacchio and Tremblay (Hymenoptera, Braconidae). Bulletin of Entomological Research 84: 57-64. Perez-Lachaud G. and Campan M. 1994. Sexual behaviour and reproductive strategy in Chryseida bennetti Burks (Hymenoptera: Eurytomidae), a parasitoid of the bean weevil. I. Effect of partner age. Canadian Journal of Zoology 72: 126-134. Perez-Lachaud G. and Campan M. 1995. Influence of previous sexual experience and post-emergence rearing conditio ns on the mating behavior of Chryseida bennetti. Entomologia Experimentalis et Applicata 76: 163-170. Pitnick S. 1991. Male size influences mate fecundity and remating interval in Drosophila melanogaster. Animal Behavor 41: 735-745. . References 149 Pitnick S. and Markow T.A. 1994. Male gametic strategies: sperm size, testes size, and the allocation of ejaculate among successive mates by the sperm limited fly Drosophila pachea and its relatives. American Nauralist 143: 785-819. Polaszek A. 1986. The effects of two species of hymenopterous parasitoid on the reproductive system of the pea aphid, Acyrthosiphon pisum. Entomologia Experimentalis et Applicata 40: 285-292. Powell W. 1982. The identification of hymenopterous parasitoids attacking cereal aphids in Britain. Systematic Entomology 7: 465-473. Pungerl N.B. 1983. Variability in characters commonly used to distinguish Aphidius species (Hymenoptera: Aphidiidae). Systematic Entomology 8: 425-43 0. Quicke D.L.J. 1997. Parasitic wasps. Chapman and Hall, London. Rahbé Y., Digilio M. C., Febvay G., Guillaud J., Fanti P. and Pennacchio F. 2002. Metabolic and symbiotic interactions in amino acid pools of the pea aphid, Acyrthosiphon pisum, parasitized by the braconid Aphidius ervi. Journal of Insect Physiology 48: 507-516. Rauwald K.S. and Ives A.R. 2001. Biological control in disturbed agricultural systems and the rapid recovery of parasitoid populations. Ecological Applications 11: 1224-1234. Raychaudhuri D. 1990. Aphidiids (Hymenoptera) of Northeast India. Indira Publishing House, Michigan. Ridley M. 1993. Clutch size and mating frequency in parasitic Hymenoptera. American Naturalist 142: 893-91 0. Rive K. 2003. Lucerne good option for drier regions. Country-Wide Northern. http://www.country-wide.co.nz/article/197.html. Accessed on 28 June 2007. Rivero A. 2000. The relationship between host selection behaviour and offspring fitness in a koinobiont parasitoid. Ecological Entomology 25: 467-472. . References 150 Roitberg B.D., Boivin G. and Vet L.E.M. 2001. Fitness, parasitoids and biological control: an opinion. Canadian Entomologist 133: 429-438. Rosenqvist G. and Berglund A. 1992. Is female sexual behaviour a neglected topic? Trends in Ecology and Evolution 7: 174-176. Ruther J., Homann M. and Steidle J. L. M. 2000. Female-derived sex pheromone mediates courtship behaviour in the parasitoid Lariophagus distinguendus. Entomologia Experimentalis et Applicata 96: 265-27 4. Sagarra L. A., Vincent C. and Stewart R.K. 2001. Body size as an indicator of parasitoid quality in male and female Anagyrus kamali (Hymenoptera: Encyrtidae). Bulletin of Entomo logical Research 91: 363-367. Sandow J.D. 1981. Can parasites and resistant plants control exotic lucerne aphids? Journal of Agriculture, We stern Australia 22: 65-67. Sanger C. and King P.E. 1971. Structure and function of the male genitalia in Nasonia vitripennis (Walker) (Hym.: Pteroma lidae). Entomologist 104: 137- 149. SAS Institute 2006. Users´ manual. SAS Institute Inc, Cary, N.C. Saunders D.S. 1982. Insect clocks. Pergamon Press, Oxford. Schlinger E.I. and Hall J.C. 1960. The biology, behavior, and morphology of Praon palitans Muesebeck, an internal para site of the spotted alfalfa aphid, Therioaphis maculata (Buckton) (Hym enoptera: Braconidae, Aphidiinae). Annals of the Entomological Society of America 53: 144-16 0. Schmidt J.M. 1991. The role of physical factors in tritrophic interactions. Redia 74: 43-93. Schowalter T.D. 2000. Insect Ecology: an ecosystem approach. Academic Press, San Diego, CA. . References 151 Schwartz J.M. 1991. Effect of sexual experien ce on male mating success in Drosophila silverstris. Animal Behaviour 42: 1017-101 9. Schworer U., Volkl W. and Hoffmann K.H. 1999. Foraging for mates in the hyperparasitic wasp, Dendrocerus carpenteri : impact of unfavourable weather conditions and parasitoid age. Oecologia 119: 73-80. Seigler D.S. 2005. Integrative biology 363, anthropology 378, plants and their uses: forage crops. University of Illinois, Urbana, Illinois. http://www.life.uiuc.edu /ib/363/forage.html. Accessed on 3 May 2007. Sekhar P.S. 1957. Mating, oviposition, and discrimination of hosts by Aphidius testaceipes (Cresson) and Praon aguit Smith, primary parasites of aphids. Annals of the Entomological Society of America 50: 370-375. Sequeira R. and Mackauer M. 1988. Effects of parasitism by Praon pequodorum on age-specific fecundity and popula tion growth of the pea aphid, Acrythosiphon pisum. Entomologia Experimentalis et Applicata 48: 179-185. Sequeira R. and Mackauer M. 1992. Covariance of adult size and development time in the parasitoid wasp Aphidius ervi in relation to the size of its host, Acyrthosiphon pisum. Evolutionary Ecology 6: 34-44. Sequeira R. and Mackauer M. 1993. Seasonal variation in body size and offspring sex ratio in field populations of the parasitoid wasp, Aphidius ervi (Hymenoptera: Aphidiid ae). Oikos 68: 340-346. Serrano J.M., Castro L., Toro M.A. and López-Fanjul C. 2000. Inter- and intraspecific sexual discrimi nation in the flour beetles Tribolium castaneum and Tribolium confusum. Heredity 85: 142-146. Showler A.T., Salgado E., Fraser I. and Robacker D.C. 2005. Effect of aging on pheromone emission from a commercial beet armyworm (Lepidoptera: Noctuidae) lure and trap efficiency. Journal of Economic Entomology 98: 373- 377. . References 152 Shuker D.M., Pen I. and West S.A. 2006. Sex ratios under asym metrical local mate competition in the parasitoid wasp Nasonia vitripennis. Behavioral Ecology 17: 345-352. Simmons L.W. 2001. Sperm competition and its evol utionary consequences in the insects. Princeton University Press, Princeton. Simmons L.W. and Gwynns D.T. 1993. Reproductive investment in bushcricket: the allocation of female and male nutrients to offspring. Proceedings of the Royal Society of London, Series B 252: 1-5. Simser D.H. and Coppel H.C. 1980. Female-produced sex pheromone in Brachymeria lasus and B. intermedia [ Hym.: Chalcididae]. BioControl 25: 373-380. Singer M. C. 1982. Sexual selection for small size in male butterflies. American Naturalist 119: 440-443. Smith R.F. and Reynolds H.T. 1966. Principles, definitions and scope of integrated pest control. Proceedings of FAO (United Nations Food and Agricultural Organisation) Symposium on Integr ated Pest Control 1: 11-17. Spencer H. 1926. Biology of the parasites and hyperp arasites of aphids. Annals of the Entomological Society of America 19: 119-15 7. Srivastava M. and Singh R. 1995. Influence of age of parents Lysiphlebus delhiensis (Subba Rao and Sharma) (Hym., Aphid iidae) during copulation on progeny production and offspring sex ratio. Journal of Applied Entomology 119: 73-77. Starý P. 1962. Hymenopterous parasites of the pea aphid Acyrthosiphon onobrychis ( Boyes) in Czechoslovakia. I. Bionomics and ecology of Aphidius ervi. Folia Zoologica, Brno. 11: 265-278. Starý P. 1966. Aphid parasites of Czechoslovaki a. A review of the Czechoslovak Aphidiidae (Hymenoptera). Publishing House of the Czechoslovak Academy of Sciences, Prague. . References 153 Starý P. 1970. Biology of aphid parasites (Hymenoptera: Aphidiidae) with respect to integrated control. The Hague, Junk. Starý P. 1978. Seasonal relations between lucerne, red clover, wheat and barley agro- ecosystems through the aphids and parasitoids (Homoptera, Aphididae: Hymenoptera, Aphidiidae). Acta En tomologica Bohemoslovaca 75: 296-311. Starý P. 1988. Natural enemies – parasites. pp. 171-184 , In A.K. Minks and P. Harrewijn [eds.], Aphids, their biology, natural enemies and control. Volume B. World crop pests. Amsterdam, Elsevier. Starý P. and Delfino M.A. 1986. Parasitoids (Hym., Aphid iidae) of aphids (Hom., Aphididae) in Tucuma ń, Argentina. Bollettino del Laboratorio di Entomologia Agraria, Portici 43: 41-50. Starý P., Gonzalez D. and Hall J.C. 1980. Aphidius eadyi n. sp. (Hymenoptera: Aphidiidae), a widely distributed parasitoid of the pea aphid, Acyrthosiphon pisum (Harris) in the Palearctic. En tomologica Scandinavica 11: 473-480. Starý P., Lyon J.P. and Leclant F. 1988. Biocontrol of aphids by the introduced Lysiphlebus testaceipes (Cress.) (Hym., Aphidiidae) in Mediterranean France. Journal of Applied Entomology 105: 74-87. Stephens D.W. and Krebs J.R. 1986. Foraging theory. Princeton University Press, Princeton, New Jersey. Switzer P.V., Forsythe P.S., Escajeda K. and Kruse K.C. 2004. Effects of environmental and social conditions on homosexual pairing in the Japanese beetle ( Popillia japonica Newman). Journal of Insect Behavior 17: 1-16. Tang Y.Q. and Yokomi R.K. 1996. Effect of parasitism by Aphelinus spiraecolae (Hymenoptera: Aphelinidae) on devel opment and reproduction of spirea aphid (Homoptera: Aphididae). Environmental Entomology 25: 703-70 7. Teal P.E.A., Tumlinson J.H. and Oberlander H. 1990. Endogenous suppression of pheromone production in virgin female moths. Experientia 46: 1047-10 5 0. . References 154 Thornhill R. 1988. Mate choice in Hylobittacus apicalis. Evolution 34: 519-538. Thornhill R. and Alcock J. 1983. The evolution of insect mating systems. Harvard University Press, Cambridge, MA. Thornhill R. and Sauer P. 1991. The natal organ of the scorpion fly Panorpa vulgaris: an adaptation to coerce mating duration. Behavioural Ecology 2: 156-164. Tomanović Ź., Kavallieratos N.G., Athanassiou C.G. and Stanisavljević L.Ź. 2003. A review of the West Palaearctic aphidiines (Hymenoptera: Braconidae: Aphidiinae) parasitic on Uroleucon spp. with a description of a new species. Annales de la Société Entomologique de France 39: 343-353. Tripathi R.N. and Singh R. 1991. Aspects of life-table studies and functional response of Lysiphlebia mirzai. Entomologia Experimentalis et Applicata 59: 279-287. Tsai J.H. and Wang J.J. 2002. Host age choice for parasitism by Lysiphlebia mirzai and its effect on the development and reproduction of brown citrus aphid. BioControl 47: 645-655. Turlings T.C., Wacker F.L., Vet L.E.M., Lewis W.J., and Tumlinson J.H. 1993. Learning of host-finding cues by hy menopterous parasitoids. pp. 51-78, In D.R. Papaj and A.C. Lewis [eds.], Insect learning: ecological and evolutionary perspectives. Chapman and Hall, New York. Ueno T. 1999. Host-size-dependent sex ratio in a parasitoid wasp. Special issue. Parasitoids: a model system to answer questions in behavioral, evolutionary and population ecology. Researches on Population Ecology 41: 47-57. van Alphen J.J.M. and Jervis M.A. 1996. Foraging behaviour. pp. 1-62, In M. Jervis and N. Kidd [eds.], Insect natural enemies: practical ap proaches to their study and evaluation. Chapman and Hall, London. . References 155 van den Assem J. 1986. Mating behaviour in parasitic wasps. pp. 137-167 , In J.K. Waage and D. Greathead [eds.], Insect parasitoids. Academic Press, London. van den Assem J. and Visser J. 1976. Aspects of sexual receptivity in female Nasonia vitripennis. Biologie du Comportement 1: 37-56. van den Assem J., Gijswijt M.J. and Nübel B.K. 1980. Observations on courtship and mating strategies in a few species of parasitic wasps (Chalcidoidea). Netherlands Journal of Zoology 30: 208-227. van den Assem J., van Iersel J.J.A. and los-den Hartogh R.L. 1989. Is being large more important for female than for male parasitic wasps? Behaviour 108: 160- 195. van Driesche R.G. and Bellows T.S. Jr. 1996. Biological control. Chapman and Hall, New York. van Lenteren J.C. 1981. Host discrimination by parasitoids. pp. 153-179 , In D.A. Nordlund, Jones R.L. and Lewis W.J. [eds.] , Semiochemicals, their role in pest control. Wiley, New York. van Lenteren J.C., Szabo P. and Huisman P.W.T. 1992. The parasite-host relationship between Encarsia formosa Gahan (Hymenoptera, Aphelinidae) and Trialeurodes vaporariorum (Westwood) (Homoptera, Aleyrodidae). XXXVII. Adult emergence and in itial dispersal pattern of E. formosa. Journal of Applied Entomology 114: 392-399. vaz Nunes M.V. and Hardie J. 1996. Differential photoperiodic responses in genetically identical winged and wingless pea aphids, Acyrthosiphon pisum, and the effect of day length on wing development. Physiological Entomology 21: 339-343. Vencl F.V. and Carlson A.D. 1998. Proximate mechanisms of sexual selection in the firefly Photinus pyralis (Coleoptera: Lampyridae). J ournal Insect Behavior 11: 1 9 1 - 2 0 7. . References 156 Vet L.E.M. and Dicke M. 1992. The ecology of infochemical use by natural enemies of herbivores in a tritrophic contex t. Annual Review of Entomology 37: 141- 172. Vetter R.S. and Visscher P.K. 1997. Influence of age on antennal response of male honey bees, Apis mellifera, to queen mandibular pheromone and alarm pheromone component. Journal of Chemical Ecology 23: 1867-18 8 0. Vinson S.B. 1976. Host selection by insect parasitoids. Annual Review of Entomology 21: 109-34. Vinson S.B. 1984. Parasitoid–host relationships. pp. 205-33, In W.J. Bell and R.T. Cardé [eds.], Chemical ecology of insects. Chapman and Hall, London. Visser M.E. 1994. The importance of being large: th e relationship between size and fitness in females of the parasitoid Aphaereta minuta (Hymenoptera: Braconidae). Journal Animal Ecology 63: 963-978. Vogt E.A. and Nechols J.R. 1991. Diel activity patterns of the squash bug egg parasitoid Gryon pennsylvanicum (Hymenoptera: Scelionidae). Annals of the Entomological Society of America 84: 303-30 8. Waage J.K. 1986. Family planning in parasitoids: adaptive patterns of progeny and sex allocation. pp. 63-96, In J.K. Waage and D. Greathead [eds.], Insect parasitoids. Academic Press, London. Waage J.K. 1990. Ecological theory and the selectio n of biological control agents. pp. 135-158, In M. Mackauer, L.E. Ehler and J. Ronald [eds.], Cr itical issues in biological control. Intr ecept Press, Andover. Waage J.K. and Godfray H.C. 1985. Reproductive strategies and population ecology of insect parasitoids. pp. 449- 470, In R.M. Sibly and R.H. Smith [eds.], Behavioural ecology: ecological consequences of adaptive behaviour. Blackwell Science, Oxford. . References 157 Waage J.K. and Hassell M.P. 1982. Parasitoids as biological control agents: a fundamental approach. Parasitology 84: 241-268. Waage J.K. and Ng S.M. 1984. The reproductive strategy of a parasitic wasp. I. Optimal progeny allocation in Trichogramma evanescens. Journal of Animal Ecolology 53: 401-415. Wang Q. and Chen L.Y. 2005. Copulation behavior of a flower-visiting longhorn beetle Zorion guttigerum (Westwood) (Coleoptera: Cerambycidae: Cerambycinae). Naturwissenschaften 92: 237-241. Wang Q., Chen L.Y., Li J.S. and Yin X.M. 1996. Mating behavior of Phytoecia rufiventris Gautier (Coleoptera: Cerambycidae) . Journal of Insect Behavior 9: 4 7 - 6 0. Wang Q., Yang L.H. and Hedderley D. 2008. Function of prolonged copulation in Nysius huttoni White (Heteroptera: Lygaeid ae) under male-biased sex ratio and high population density. Journal of Insect Behavior 21: 89-99. Warner R.R., Shapiro D.Y., Marcanato A. and Petersen C.W. 1995. Sexual conflict: males with highest mating success convey the lowest fertilization benefits to females. Proceedings of the Royal Society of London, Series B 262: 135-139. Waterhouse D.F. and Sands D.P.A. 2001. Classical Biological Control of Arthropods in Australia. ACIA R Monograph Series Number 77. Weinbrenner M. and Volkl W. 2002. Oviposition behaviour of the aphid parasitoid, Aphidius ervi : are wet aphids recognized as host. Entomologia Experimentalis et Applicata 103: 51-59. Werren J.H. 1980. Sex ratio adaptation to local mate competition in a parasitic wasp. Science 208: 1157-115 9. Wiackowski S.K. 1962. Studies on the biology and ecology of Aphidius smithi Sharma & Subba Rao (Hymenoptera, Braconi dae), a parasite of the pea aphid, . References 158 Acyrthosiphon pisum (Harr.) (Homoptera, A phididae). Polskie Pismo Entomologiczne 32: 253-31 0. Wilkes A. 1966. Sperm utilization following multiple insemination in the wasp Dahlbominus fuscipennis. Canadian Journal of Genetics and Cytology 8:451- 461. Zadoks J.C. 1993. Crop protection: why a nd how. pp. 48-60, In D. J. Chadwick and J. Marsh [eds.], Crop protection and sust ainable agriculture. John Wiley and Sons, Chichester. Zar J.H. 1999. Biostatistical analysis . Prentice Hall, Upper Saddl e River, New Jersey. Zhang W.Q. and Hassan S.A. 2003. Use of the parasitoid Diaeretiella rapae (McIntoch) to contro l the cabbage aphid Brevicoryne brassicae (L.). Journal of Applied Entomology 127: 522-526. . Appendix I: Published Papers 159 . Appendix I: Published Papers 160 . Appendix I: Published Papers 161 . Appendix I: Published Papers 162 . Appendix I: Published Papers 163 . Appendix I: Published Papers 164 . Appendix I: Published Papers 165 . Appendix I: Published Papers 166 . Appendix I: Published Papers 167 . Appendix I: Published Papers 168 . Appendix I: Published Papers 169 . Appendix I: Published Papers 170 . Appendix I: Published Papers 171 . Appendix I: Published Papers 172 . Appendix I: Published Papers 173 . Appendix I: Published Papers 174 . Appendix I: Published Papers 175 . Appendix I: Published Papers 176 . Appendix I: Published Papers 177 . Appendix I: Published Papers 178 . Appendix I: Published Papers 179 . Appendix I: Published Papers 180 . Appendix I: Published Papers 181 . Appendix I: Published Papers 182 . Appendix I: Published Papers 183 . Appendix I: Published Papers 184 . Appendix I: Published Papers 185 . Appendix I: Published Papers 186 . Appendix I: Published Papers 187 . Appendix I: Published Papers 188 . Appendix I: Published Papers 189 . Appendix I: Published Papers 190 . Appendix I: Published Papers 191 . Appendix I: Published Papers 192 . Appendix I: Published Papers 193 . Appendix I: Published Papers 194 . Appendix I: Published Papers 195 . Appendix I: Published Papers 196 . Appendix I: Published Papers 197 . Appendix I: Published Papers 198 . Appendix I: Published Papers 199 . Appendix I: Published Papers 200 . Appendix I: Published Papers 201 . Appendix I: Published Papers 202 . Appendix I: Published Papers 203 . Appendix I: Published Papers 204 . Appendix I: Published Papers 205 . Appendix I: Published Papers 206 . Appendix I: Published Papers 207 . Appendix I: Published Papers 208 . Appendix II: Conference P resentations 209 . Appendix II: Conference P resentations 210 Mate age at mating and male mating history affect mate choice and reproduction in Aphidius ervi Haliday (Hymenoptera: Aphidiidae) Xiong Zhao He and Qiao Wang Institute of Natural Resources, Mass ey University, Palmerston North, Private Bag 11222, New Zealand Corresponding author: q.wang@massey.ac.nz Hymenopteran parasitoids are usually a rrhenotokously parthenogenetic, where fertilized eggs produce diploid females and unfertilized eggs give rise to haploid males; the reproductive fitness of males depends on the number of daughters they father. The effect of mate age at mating and male mating history on mate choice and reproductive fitness of Aphidius ervi Haliday (Hymenoptera: A phidiidae), was studied in the laboratory. Both sexes prefer 1-d-old to 3-d-old pa rtners for mating if given a choice, and 1 or 2 d mating delay of a ny sex reduces the proportion of daughters produced. Furthermore, mating delay affects females more severely than males in production of daughters. Once-mated males ha ve significantly greater mating success than virgin males, suggesting that males l earn from previous mating(s). If allowed 24- h recovery time, the proportion of daughters produced only slowly decreases with the increase of the number of matings males have had but males can still supply enough sperm to produce female-biased progeny after seven matings. This suggests that males suffer little sperm depletion if allowed 24-h recovery time between matings and that the decline of daughter production with increasing matings may be caused by the increasing male age. If only allowed 1 h between matings, males would experience significant sperm depletion after three matings. The implications of these results in the biological control are discussed. Keywords: Aphidius ervi , age, mating history, sperm, proportion of female progeny . Appendix II: Conference P resentations 211 . Appendix II: Conference P resentations 212 Reproductive response of Aphidius ervi (Hymenoptera: Aphidiidae) to pea aphid density X. Z. He 1 , Q. Wang 1 , and D. A. J. Teulon 2 1 Institute of Natural Resources, Massey University, Palmerston North, Private Bag 11222, New Zealand, Q.Wang@massey.ac.nz 2 New Zealand Institute for Crop & Food Research Ltd, Private Bag 4704, Christchurch, New Zealand The reproductive response of Aphidius ervi Haliday to the density of pea aphid, Acyrthosiphon pisum (Harris) was investigated. Si x aphid densities (15, 25, 50, 75, 100 and 125 aphids), each kept on a broad bean cutting in a plastic cylinder (6.5 × 8.5 cm), were examined. Mean number of a phids parasitised and eggs laid by A. ervi significantly increased with the increase of host density from 15 to 100 aphids, after which no further increase occurred. However, the number of eggs laid in each aphid significantly decreased with the increase of host density from 15 to 75 aphids, after which no further decrease occurred. These results suggest that the parasitoid quickly adjusts oviposition strategy in response to increasing host density through increasing parasitism and decreasing superparasitism, and has high ability to suppress the increasing aphid population. The proportion of female progeny quickly increased (up to 70%) with the increase of host density from 15 to 50 or 75 aphids, after which it gradually declined, suggesting that the sperm limit occurs when host density reaches 50 to 75 aphids. When the host density was low (such as 15 or 25 aphids), female parasitoids lowered the allocation of fertilized eggs. This suggests that the female parasitoids can predict the immediate future need of sex ratio, i. e. the need of the number of female progeny. Key words: Aphidius ervi , host density, reproduction, sex ratio . Appendix II: Conference P resentations 213 Effect of Body Size on Reproductive Fitness in Aphidius ervi (Hymenoptera: Aphidiidae) X. Z. He, Q. Wang and D. A. J. Teulon Institute of Natural Resources, Massey University, Private Bag 11222, Palmerston North, New Zealand, xiong.he.1@uni.massey.ac.nz Aphidius ervi Haliday is an important parasitoid of several aphid species on economically important crops such as legumes and cereals. To improve its mass rearing and field release efficiency, we studied the re lationships between body size and reproductive fitness of A. ervi on pea aphid, Acyrthosiphon pisum (Harris), in the laboratory. Large females lived significan tly longer (14.10 ± 0.72 d), had significantly higher searching efficiency (1.50 ± 0.37 10 -4 min), fecundity (574.25 ± 23.76 eggs) and fertility (338.74 ± 19.41 eggs), and para sitized significantly more aphids (410.70 ± 17.74) than did small females (11.60 ± 0.57 d, 0.97 ± 0.09 10 -4 min, 302.74 ± 17.61 eggs, 213.39 ± 15.17 eggs and 257.07 ± 14.18 aphids for longevity, searching efficiency, fecundity, fertility and parasitism, respectively). However, fertility rate was significantly higher for small females (0.6923 ± 0.0238) than large females (0.5819 ± 0.0252). Male size was positively corr elated with fertility ranging from 246.60 ± 20.43 eggs for small males to 305.53 ± 20.02 eggs for large males, and with fertility rate ranging from 0.5771 ± 0.0276 for small males to 0.6971 ± 0.0199 for large males. Male size had no effect on fec undity and parasitism. Implications from this study were discussed. . Appendix II: Conference P resentations 214 EFFECT OF APHID LIFE STAGE AND PARASITOID ADULT AGE ON PARASITISM AND SEX RATIO OF APHIDIUS EADYI Xiongzhao HE 1 , Qiao WANG 1 and David TEULON 2 1 Institute of Natural Resources, Massey University, Private Bag 11222, Palmerston North, New Zealand 2 Crop & Food Research, Private Bag 4704, Christchurch, New Zealand Email: Q.Wang@massey.ac.nz Aphidius eadyi (Hymenoptera: Aphidiidae) is a specific parasitoid of pea aphid, Acyrthosiphon pisum ( Homoptera: Aphididae). The effect of A. eadyi adult age and A. pisum life stage on A. eadyi parasitism and sex ratio was investigated in transparent plastic cylinders (85 mm diameter by 120 mm high), each having a potted broad bean plant. Both choice (100 cylinders, each hol ding a mixture of 5 life stages with 10 aphids in each stage) and non- choice (500 cylinders, each containing 50 aphids of one of the five stages) experiments were carri ed out. Ten parasitoids of 10 ages (1-10 d old) were used for each experiment. Each pa rasitoid of a certain age was allowed to be in a cylinder for 24 h. W ith the increase of parasitoid age, parasitism rate significantly decreased and the proportion of male progeny significantly increased in both experiments. With the increase of aphi d life stage, parasitism rate significantly increased in the choice experiment but significantly decreased in the non-choice experiment. The preference of older aphids by parasitoids in the choice experiment may be interpreted as that the older aphids have more nutrition for their offspring. The decline in parasitism rate on older aphids in the non-choice experiment may be explained as that (1) older a phids have greater defense ability so that the parasitoids need more time and energy to parasitize th em, and (2) the number of aphids of a particular life stage was 5 times greater than that in the choice experiment so that the parasitoids attacked fewer ol der aphids within 24 h. In both experiments sex ratio of the parasitoid progeny was female-biased except for those from 1 st instar aphids, whose sex ratio was close to 1:1. Our results predicted that the parasitoid female laid about 66.3% of her eggs in the fi rst 5 d of her lifetime when 4 th instar aphids were offered, and that the sex ratio of progeny pr oduced by 6-d-old or older parasitoids was male-biased. . Appendix II: Conference P resentations 215