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. Aspects of the Ecology of Trachymela catenata Chapuis ( Coleoptera : Chrysomelidae) in New Zealand. A thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Ecology at Massey University Douglas Paul Barrett 1998 Table of Contents Abstract lV Acknowledgements v Chapter 1. Introduction The Eucalypts Eucalypt forest insects Eucalypt plantation forest pests and biocontrol The Paropsina History of Trachymela catenata in New Zealand Study sites References Chapter 2. Ecology of the recently established paropsine Trachymela catenata Chapuis (Coleoptera: Chrysomelidae) in New Zealand: Life history, Phenology, Mortality and Development. Abstract Introduction Methods Results Discussion References 1 1 3 7 14 17 18 22 32 32 33 36 45 56 68 Chapter 3. Larval performance and oviposition preferences 73 of Trachyrnela catenata Chapuis (Coleoptera: Chrysomelidae) on eight species of Eucalyptus. Abstract 73 Introduction 74 Methods 77 Results 81 Discussion 87 References 93 Chapter 4. Comparative studies of the effects of the parasitoid 98 Enoggera nassaui Girault (Hymenoptera: Pteromalidae) on Trachyrnela catenata Chapuis and Paropsis charybdis Stahl (Coleoptera: Chrysomelidae). Abstract 98 Introduction 99 Methods 104 Results 106 Discussion 109 References 113 Chapter 5. Conclusions 117 Location of Plates Plates 1, 2, 3 and 4 " 5 and 6 between pages " " " " " 21-22 38-39 iv Abstract A member of the Eucalyptus defoliating Australian tortoise beetles Trachymela catenata was recorded in New Zealand in December 1992. To date Eucalyptus viminalis, E. macarthurii and E. macarthurii x botryoides are known hosts. Investigations of a range of ecological parameters for T. catenata are presented in order to provide information with which to assess the potential status of this recent introduction. Life history as for all other paropsina comprises eggs, four larval instars, prepupa, pupa and adult Adults overwinter and emerge during October to lay first generation eggs in November/December. An estimated generation time of 50 days means a second generation lays eggs during February, indicating a bivoltine life history. Females are as fecund as some other paropsines which erupt to pest levels in other countries. Larval mortality is highest during the first instar and 45.8% mortality occurred during pupation. Developmental thresholds and development times indicate that thermal requirements for completion of two generations will be met throughout most of New Zealand. Laboratory trials to determine female oviposition preference and larval performance on eight potential host eucalypts indicate E. nitens (an important commercial species) and E. coccifera to be equally as suitable hosts as those currently utilised. Trachymela catenata is therefore polyphagus and field monitoring of these two potential hosts is needed. The hymenopteran pteromalid egg parasitoid Enoggera nassaui was trialed in a study comparing parasitization of T. catenata eggs with those of Paropsis charybdis, a known host The parasitoid had no apparent effect on T. catenata eggs and offers no potential control of T. catenata populations. v Acknowledgements I am pleased to acknowledge Professor Brian Springett for his supervision, guidance and encouragement through all my investigations and thesis. Prof. kindled my interest in "insects on trees" with his enthusiasm for the insect fauna and the important role they play in ecosystems. I am also pleased to acknowledge Dr. Murray Potter for his supervision and guidance through the entire project, especially with data analysis. Both Prof. and Murray provided valuable constructive comment and discussion on the manuscript. I am especially grateful to Malcolm Kay of the Forest Research Institute (FRINZ), Rotorua for his assistance and advise through the programme. Malcolm initiated the programme, willingly provided field equipment, advised with field work and data collection, provided literature and finally, provided valuable comment on the manuscript. I am also grateful to Chris Ecroyd (FRINZ) for identification of eucalypt species I wish to thank FRINZ and the Massey University Postgraduate Research Fund for providing valuable financial assistance for the project. I also acknowledge Mr Rodney Alexander and Mr Robert Crawshaw of Morere who kindly provided access to their properties as study sites. I especially thank Rodney and his late wife Marj for their hospitality and occasionally a bed after the long nights drive to Morere. I particularly wish to thank Liz Grant for her excellent illustrations in this thesis, Dr. Peter McGregor of Land Care Research for valuable advice and discussion, my wife Claire, who, along with her own busy life, continues to support and encourage me to achieve my goals and my children Lorelle and Brad for assistance in the field and company on the long drives to Morere. This thesis is for Claire, Lorelle and Brad and my parents. vi vii " the field of forest entomology must in its very nature rest upon an ecological foundation" Samuel A. Graham, 1956, Annu. Rev. Entomol., Vol. 1. Chapter 1 Introduction Introduction The Eucalypts The genus Eucalyptus contains approximately 700 species (Brooker and Kleinig 1990) of which almost all occur in forests and woodlands in Australia. Ten species occur only in northern Australia and New Guinea and four tropical species are endemic to northern New Guinea and southern Indonesia. In Australia, eucalypts dominate the coastal forests (often categorised as tall open­ forests and open-forests) and woodland regions (trees 10-25 m in height often associated with grassland). They also the comprise vast fil'eas of mallee shrublands of dry inland southern regions of the continent (Williams and Brooker 1997). It is however, the tall open-forests of coastal Australia (known also as wet sclerophyll forests due to the usually mesic sites on which they grow) where large single stemmed eucalypts dominate the forest canopy and form tall majestic stands, with tree heights in excess of 30 m and some that exceed 90-100 m. These forests occur in a large coastal strip on the eastern side of the continent from the higher rainfall regions of northern Queensland (lat. 170 S) to Tasmania (lat. 42° S), are absent across the southern xeric regions of South and Western Australia, but also occur in the south west of Western Australia (lat. 35° S) (Ashton and Attiwill 1994). A floristic discontinuity Chapter 1 Introduction 2 exists between these major areas whlch can be divided into three broad groups. In southwest Western Australia Eucalyptus diversicolor, E. calophylla, E. guilfoylei and E. jacksonii are the predominant species. Predominant species associated with Tasmania, Victoria and parts of the northern New South Wales highlands, comprise one major east coast group consisting of E. regnans, E. viminalis, E. obliqua, E. globulus, E. fastigata, E delagatensis, cypellocarpa, dalrympleana, and E. nitens. The remaining group, occurring from central New South Wales to southern Queensland, consists of E. cloeziana, microcorys, pilularis, E. saligna and E. grandis (Ashton and Attiwill 1994). Many of the tall open-forest eucalypt species have long been recognised as a valuable timber resource. Uses for eucalypts include saw logs for construction, furniture and joinery timber; pulpwood for paper manufacture; shade, shelter and fuel wood; and volatile oils for pharmaceutical and industrial use (Eldridge et al. 1993). Eucalypts (particularly those in the sub-genus Symphyomyrtus) are now the most widely propagated tree genus in the world and can be grown in most tropical and temperate climatic regions between lat. 450 S and lat. 40° N. Plantation eucalypts in Brazil cover a huge 3x106 ha. India, Spain, Portugal, South Africa, Angola and Chlna, all have between 3- 4x105 ha whlle Ethiopia and Argentina each have 2.4x105 ha. These plantations, along with more than 5x105 ha, in 60 other countries attest to the tremendous adaptability of thls genus (Eldridge et al. 1993). Chapter 1 Introduction 3 The success of eucalypts to adapt to new environments in both Australia and exotic sites is suggested to have evolved from the extremely variable environmental conditions of the past. Through millions of years of cyclic wet fertile to dry infertile and more recently (mid miocene), marked seasonal dry periods with increased frequency of fire, eucalypts have developed strategies which allow them to take advantage of periods of favourable growth between periods unfavourable for growth (Eldridge et al. 1993; Wardell-Johnson et al. 1997). Rapid growth of plantation eucalypts in exotic locations is often attributed to the lack of environmental constraints that these species experience in Australia. Conditions of higher fertility, more reliable and higher rainfall, well prepared weed free sites and absence of fire may all contribute to such accelerated growth (Eldridge et al. 1993). Pryor 1976 (Eldridge et al. 1993) postulated that eucalypts grow better in exotic sites due to lack of naturally occurring insect pests and diseases. Ohmart (1984) however maintains that insufficient data exist to validate these claims and cites a number of examples where eucalypt growth trials in Australia, unprotected from insect pests, produce similar results to comparable overseas trials. Eucalypt forest insects A large and diverse suite of insects belonging to a number of functional or feeding guilds (e.g. sap-suckers, seed-eaters, Gall-formers, leaf-miners, leaf­ chewers and wood-borers) are associated with eucalypt forests in Australia Chapter 1 Introduction 4 (Majer et al. 1997; Ohm.art and Edwards 1991). Sap-feeders include the hemipteran psyllids (Psyllidae), leafhoppers (Eurymelidae), scale insects (Eriococcidae) and coreid bugs (Coreidae). These species have a range of feeding strategies utilising the phloem of leaves and shoots and many produce large quantities of honeydew. Psyllid species, most notably the White Lace Lerp (Cardiaspina albitextura) and the scale insect Eriococcus coriaceus, during eruptive outbreaks can cause considerable damage to trees (Landsberg and Cork 1997). Seed-eaters associated with eucalypts are mainly small beetles and wasps and seed exclusion experiments have shown they can cause up to 66% seed loss. Ants are also responsible for consuming a large amount of the seed that falls to ground (Landsberg and Cork 1997). Gall-formers belong mostly to the hymenopteran Chalcid wasps, hemipteran coccoid scale insects, dipteran cecidomyiid gall-forming flies and a number of psyllids. Galls resulting from these insects however are not considered to be particularly harmful to eucalypts (Elliott and de Little 1984; Landsberg and Cork 1997). Eucalypt leaf-miners belong primarily to the Lepidoptera (including families Cosmopterygidae, Incurvariidae, Nepticulidae, Gracillariidae and Nolidae) as well as sawfly larvae (Hymenoptera : Perigidae) (Landsberg and Cork 1997). One species which causes considerable damage to E. camaldulensis is the gum Chapter 1 Introduction 5 leaf skeletonizer Uraba lugens (Lepidoptera : Nolidae) (Campbell 1962) and it is known to erupt to outbreak levels on eight other eucalypt species from many areas of Australia (Ohmart and Edwards 1991). A guild of considerable importance is the leaf-chewers (defoliators) and of these the Coleoptera comprising mainly Scarabaeidae, Chrysomelidae and Curculionidae are the most important groups (Ohmart and Edwards 1991). These beetles, particularly the chrysomelids are implicated in many outbreaks causing serious defoliation of eucalypts throughout Australia (Bashford 1993; de Little 1983; Elliott and de Little 1984; Greaves 1966; Ohmart 1990; Ohmart and Edwards 1991). Stick insects (Phasmatodea : Phasmatidae), sawflies (Hymenoptera: Pergidae) and many lepidopteran caterpillars from the families Geometridae, Nolidae, Anthelidae, Limacodidae, Saturniidae, and Lasiocampidae are also notable defoliators (Landsberg and Cork 1997; Ohmart and Edwards 1991). Termites (Isoptera) are the most important group of wood-boring insects associated with eucalypts. According to Perry et al. 1985 (Landsberg and Cork 1997) these insects are not usually associated with healthy living trees but invade trees that have suffered fire or physical damage to the trunk. They can, after invasion, cause the death of burnt or damaged trees. Many weevils (Coleoptera: Curculionidae) feed on the bark, roots and shoots of eucalypts but damage is often only restricted to young plantation trees (Landsberg and Cork 1997). Chapter 1 Introduction 6 Studies of the impact of insect defoliation on tree growth have shown eucalypt species to be highly variable. According to Carne et al. (1974), defoliation levels of less than 50% are unlikely to impair subsequent growth of E. grandis. Studies by Readshaw and Mazanec 1969 however, (reported in Landsberg and Cork 1997) showed that defoliation by the phasmatid Didymuria violescens over a 16 year outbreak caused a reduction in radial growth increment of paucifiora and E. stellulata of 20%. Further, using the same eucalypt species during a non outbreak year, foliage insecticide treatment of branches produced a growth ring increment of more than double that of the untreated branches (Landsberg and Cork 1997). These naturally occurring insect communities and guilds have been implicated in important eucalypt forest ecological and population processes. Processes such as nutrient cycling (Springett 1978; Ohmart et al. 1983), eucalypt species geographic distribution limits and influences on the coexistence of eucalypt species in mixed forest stands (Burdon and Chilvers 1974a, 1974b; Landsberg & Cork 1997; Morrow 1977; Williams 1990) have all been postulated. Evidence for many of these processes in the Australian eucalyptus open-forests is equivocal, but in disturbed regrowth or plantation forests, insect outbreaks are known to cause severe economic damage to tree growth (Bashford 1993; Elliott and de Little 1984; Ohmart 1990). Chapter 1 Introduction 7 Eucalypt plantation forest pests and biocontrol In Australian tall-open forests, the suite of insects associated with this habitat are almost without exception subject to many constraining factors of naturally occurring predators, parasitoids and pathogens (Landsberg and Cork 1997). In countries (including Australia) where eucalypts are established as monospecific plantation forests, reduction of biodiversity and alteration of the ecosystem mean that such constraints are often no longer present. Such habitats can allow an introduced insect species population to establish unimpeded and often reach eruptive and sometimes pest status. One of the most serious and economically important insect guilds associated with loss of production from, or the death, of plantation eucalypts are the defoliators, especially leaf chewers (Bashford 1993; de Little 1989; Elliott and de Little 1984; Elliott et al. 1993; Ohmart 1990). It is this guild we will focus on for the rest of this section. In Tasmania, the most notable of the leaf chewing defoliators are the paropsine leaf beetles (Coleoptera : Chrysomelidae). The paropsine Chrysophtharta bimaculata is capable of serious defoliation to plantation eucalypts especially E. delegatensis, E. regnans and E. obliqua (Bashford 1993; de Little 1989; Elliott and de Little 1984; Elliott et al. 1993; Ohmart 1990) with C. agricola, and C. variicollis also capable of notable damage (Elliott and de Little 1984). Paropsis porosa is known to cause damage to young eucalypt seedlings (Elliott and deLittle 1984). Chapter 1 Introduction 8 Other defoliating coleoptera are the eucalyptus weevil Gonipterus scutellatus (Curculionidae), scarabaeid beetles (many of the genus heteronyx), the green scarab Diphucephala colaspidoides (Elliott and de Little 1984) and in mainland Australia Christmas beetles Anoplognathus spp. (Carne et al. 1974; Elliott and de Little 1984; Ohmart 1990). Lepidopteran caterpillars particularly the gum leaf skeletonizer Urba lugens, the autumn gum moth Mnesampela privata and the helena gum moth Antheraea helena are all common defoliators of a variety of eucalypts. Leaf miner species include Lepidoptera of the Acrocercops spp. and hymenopteran sawfly larvae Phylacteophaga eucalypti (Elliott and de Little 1984; Ohmart 1990). While many of these eucalypt pests in Australian plantations are controlled to a large extent by natural enemies, many of the controlling parasitoid and predator species are themselves suppressed by their natural enemies (e.g. parasitoids such as Aridelus sp. (Hymenoptera : Braconidae) being suppressed by a hyperparasitoid such as Perilampus sp. (Hymenoptera : Pteromalidae). Known predators of chrysomelid eggs and young larvae are the cantharid Chauliognathus lugubris (Coleoptera : Cantharidae) (Shohet and Clarke 1997) and the ladybirds Cleobora mellyi and Harmonia conformis (Coleoptera : Coccinellidae). Both adult and larval ladybirds are predatory and laboratory studies have shown that adults of both species can consume up to 25 C. bimaculata eggs per day. Field studies have shown that these coccinellids can achieve up to 80% predation of C. bimaculata eggs (Elliott and de Little 1984). Chapter 1 Introduction 9 In some seasons such controls can break down (Landsberg and Cork 1997) and eruptive pest outbreaks occur. Integrated Pest Management (IPM) strategies are being developed (particularly for C. bimaculata) not only to take advantage of naturally occurring biocontrol agents but to also enhance pest control using various insecticidal compounds such as Bacillus thuringiensis (Greener and Candy 1994) at strategic times (Clarke et al. 1997; Elliott et al. 1992; Harcourt et al. 1996). Eucalypt plantations in other countries also experience insect pest problems. Most, but not all problems are associated with specialist eucalypt insects that have accidentally been introduced from Australia. paropsine leaf chewing defoliator Trachymela tincticollis Blackburn has established and reached pest status on commercially grown eucalypts in South Africa. Trachymela tincticollis was first recorded in the Cape Town region in 1982 and had by 1985 spread up the east coast to Piketberg and West as far as East London (Cillie 1981; Tribe and Cillie 1985a, 1989). Trachymela tincticollis has been found on a number of Symphyomyrtus eucalypts in that country but the most severe defoliation occurs on E. gomphocephala. Four hymenopteran egg parasitoids were released in South Africa in 1986 as possible biocontrol agents for T. tincticollis. Those released were the three Chapter 1 Introduction 10 pteromalids, Enoggera. reticulata, E. nassaui and Neopolycystus insectifurax and an encrytid species Procheiloneurus (G. Tribe pers. comm.). Enoggera reticulata (initially reported as E. polita) was the only species to establish. Establishment and spread was rapid, with consequent high rates of parasitism (97% of egg batches, 93% of eggs per batch) of that target species. The success of E. reticulata is attributed to the absence of hyperparasitoids and the initial high density of the target species (Tribe and Cillie 1989; Urban, et al. 1987). The eucalyptus weevil Gonipterus scutellatus Gyllenhall was first recorded in South Africa in 1916 and caused considerable damage wherever eucalypts were grown. It has subsequently spread throughout Africa, Madagascar and Mauritius (Majer et al. 1997) and into Italy and France making eucalypt plantations in all these areas almost non-viable (Ohmart and Edwards 1991). This species also established in New Zealand with similar consequences to its host eucalypts. Successful biological control of G. scutellatus in South Africa (Ohmart and Edwards 1991) and New Zealand (Flux et al. 1993; Miller 1984) was achieved with the egg parasitoid Patasson nitens (Hymenoptera : Myrmaridae). The most pervasive and wide spread eucalypt pest globally is Phoracantha semipunctata (Coleoptera : Cerambycidae). This Australian beetle has established in all major eucalypt growing areas of the world except India. First recorded in South Africa, it has spread throughout Africa, the middle east, the Chapter 1 Introduction 11 Mediterranean, South America, California (Ohmart and Edwards 1991) and Madagascar (Majer et al. 1997). In Australia, P. semipunctata is associated with debilitated trees particularly during drought conditions and indeed this species has its greatest ecological and economic impact in countries such as the xeric Mediterranean regions of Spain and Portugal where droughts are relatively common (Ohmart and Edwards 1991). New Zealand eucalypt plantations have and continue still to suffer from introduced insect pests. It appears relatively easy for Australian eucalypt insect fauna to invade and establish in this country. Up until 1993, 15 species from 4 orders (Lepidoptera, Coleoptera, Hemiptera and Hymenoptera) have established and become pests of New Zealands eucalypt forest industry (Flux et al. 1993). The eucalypt leaf chewing defoliators established in New Zealand other than Opodiphthera eucalypti (Lepidoptera : Saturniidae), Strepsicrates macropetana (Lepidoptera : Tortricidae) G. scutellatus and the leaf miner Phylacteophaga froggatti (Hymenoptera : Pergidae) belong to the paropsina (Flux et al. 1993). Paropsis charybdis Stahl was the first, recorded at Banks Peninsula in 1916 (Clark 1930) and by 1955 it had colonised the entire South Island (Bain 1977; Bain and Kay 1989). In 1956 P. charybdis was recorded in the North Island and is now present over the entire island wherever eucalypts are grown. Paropsis charybdis reached serious pest status causing severe defoliation and tree loss to eucalypt Chapter 1 Introduction 12 farm shelter and plantation forestry. The species E. globulus, E. viminalis, E. macarthurii, E. nitens, and E. ovata, all of which belong to the sub-genus Symphyomyrh1s (section Maidenaria), all suffered damage. Attempts at biological control of P. charybdis were undertaken as early as 1932 (Bain and Kay 1989) when seven cocoons of the larval parasitoid Meteorus sp. (Hymenoptera : Braconidae) were sent from Canberra. What happened to these is unknown. During 1934-35 importations of Aridelus (Hymenoptera : Braconidae ), Froggattimyia tillyardi, Froggattimyia spp. and a Paropsivora sp. (Diptera : Tachinidae) were of no use as they contained hyperparasitoids and sufficient numbers could not be reared for release (Bain and Kay 1989). In 1975 the first release of a biocontrol agent was the larval parasitoid F. tillyardi but the species has not been recorded since. In 1979 Cleobora mellyi the egg and larval predator of paropsines alluded to earlier was released, but it has only established in the Marlborough region (Bain and Kay 1989). Cleobora mellyi appears to need additional prey in the form of psyllids to enable it to achieve the fecundity and survival necessary to maintain an adequate population (Bain et al. 1984; Bain and Kay 1989). It was not until 1988 with the release of the egg parasitoid Enoggera nassaui Girault (Hymenoptera : Pteromalidae) that P. charybdis was brought under control (Kay 1990). Enoggera nassaui was released together with another pteromalid Neopolycystus insectifurax during the same season, but N. insectifurax has not been recorded since. In contrast E. nassaui Chapter 1 Introduction 13 has established over almost the entire country, causing a dramatic decline in P. charybdis populations (Kay 1990). In 1976 the paropsine, Trachymela sloanei was recorded in the Auckland region (Bain 1977) but this species has been slow to spread and has not reached pest status. Its southernmost population occurs in the Gisborne region (M. Kay pers. comm.). This species has established in California where its impact is causing much concern to foresters and biological control strategies are being considered (J. Millar pers. comm.). Trachymela catenata was first recorded in the Gisborne/northern Hawkes Bay region in December 1992. This paropsine was recorded from a number of the Symphyomyrtus eucalypts that P. charybdis also favours i.e. E. viminalis, E. macarthurii; and E. macarthurii x botryoides (M. Kay pers. comm.). The current range of this species is still within the original area recorded. The future status of T. catenata in relation to its impact on plantation eucalypts was unknown and was the prime reason for the present investigation. The most recent paropsine introduction, Dicranosterna semipunctata, was recorded in from Auckland in 1997 and has spread rapidly into other regions. This species, and a Pyrgoides sp. recorded from Auckland in 1976 (and which still occurs only in that region), are associated with Acacia (M. Kay pers. comm.) and will not be considered further. They do however serve to Chapter 1 Introduction 14 illustrate the variable pervasiveness of the paropsina genera and that the biology and ecology of any of this taxon that establish in New Zealand needs to be understood. The Paropsina In Australia1 the term paropsine refers to members of the two subtribes Paropsina and Dicranosternina within the tribe Chrysomelini (Reid 1992). This tribe is in the largest Australian sub family Chrysomelinae (with over 50 described genera and more than 600 species) of the family Chrysomelidae (Lawrence and Britton 1991). The Chrysomelidae (leaf beetles) comprise one of the biggest families (with about 3000 species) of the Australian Coleoptera (Lawrence and Britton 1991). The paropsina form the most important group of Australian Chrysomelinae as it contains several large genera including Paropsis, Trachymela, Chrysophtharta, Paropsisterna, and Stethomela (together comprising 138 species) with the majority of them feeding on eucalypts (Lawrence and Britton 1991; Majer et al. 1997). In their Australian habitat, the majority of paropsina are relatively rare and are often treated as a functional group when considering their role in native forest ecosystems. This possibly explains the general paucity of literature available (given the size of the taxon) for this group. A few species however, because of their pest status in plantation forests or availability for laboratory studies, have received much attention. Chapter 1 Introduction 15 Considerable literature is available on the biology and ecology of Paropsis atomaria 01. The ecology (Carne 1966a; Tanton and Khan 1978) and food utilisation efficiency (Carne 1966b; Fox and Macauley 1977; Morrow and Fox 1980) of P. atomaria has been extensively studied. The effects of leaf age/toughness and nitrogen concentration on larval performance (Larsson and Oh.mart 1988; Ohmart et al. 1985a), population dynamics (Ohmart et al. 1987; Oh.mart 1991), and fecundity (Ohmart et al. 1985b) as well as the effects of parasitization (Tanton and Epila 1984) on P. atomaria have all received attention. The establishment of Paropsis charybdis in New Zealand in 1916 is documented in several publications since that time (Bain 1977; Clark 1930, 1938; Gurr 1957; Thompson 1922). The impact of P. charybdis on eucalypts since then has prompted much research effort into the biology and ecology of this species. Life history and ecology (McGregor 1989), and host plant relationships (Steven 1973) were the subject of two Ph. D. theses and experimental work on energy and nitrogen budgets using P. charybdis were conducted by Edwards and Wightman (1984). Much work in relation to pest management of P. charybdis in New Zealand has accrued since the 1970s. Larval mortality and pest control (Styles 1970), susceptibility to Bacillus thuringiensis Gackson and Poinar 1989), a historical review of biological control (Bain and Kay 1989), a report of successful biocontrol (Kay 1990) and predation by hemipteran pentatomids (Edwards and Suckling 1980) of P. charybdis have all been produced. Chapter 1 Introduction 16 Biology of Chrysophtharta bimaculata Olivier has received attention from de Little (1979, 1983), de Little et al. (1990). Oviposition/host preference work has been done by de Little and Madden (1975) and Steinbauer et al. (1998) while Patterson et al. (1996) have considered larval performance on a range of eucalypt species. Dispersal behaviour of C. bimaculata and its implication for IPM strategies has been provided by Clarke et al. (1997) while Clarke et al. (1998) has provided information on adult overwintering sites. Defoliation and the pest status of C. bimaculata has been alluded to in a previous section. Life cycle, development and fecundity estimates of C. agricola have now been provided by Ramsden and Elek (1998) and one of the few studies of West Australian paropsina focuses on host preference studies of C. debilis and P. elytrura (Hall 1992). The only literature available on the Trachymela genus consists of a brief descriptive bulletin of T. tincticollis (Cillie 1981), some short scientific notes (Tribe and Cillie 1985a, 1985b, 1989; Urban et al. 1987) and recently an ecological paper on T. tincticollis in South Africa (Tribe and Cillie, 1997). No information is available on the biology or ecology of T. catenata. Chapter 1 Introduction 17 History of Trachymela catenata in New Zealand During routine port inspections of Gisborne city in December 1992, Ministry of Forestry staff collected an unidentified paropsine beetle in eucalypt plantations from Hospital Hill and Kaiti Hill reserves. New Zealand Forest Research Institute (NZFRn sent adult and larval specimens to Dr. Chris Reid at the Australian National University, Canberra, for identification. Dr. Reid identified the species as Trachymela catenata Chapuis. Past history of the closely related P. charybdis in New Zealand provided knowledge of possible pest status and prompted a wider search for this new introduction. In addition to the original collection of specimens around Gisborne, T. catenata was also recorded from sites 40 km west of Gisborne at Otoko and 40 km to the south at Morere. Both adults and larvae were recorded from E. viminalis, E. macarthurii and E. macarthurii x botryoides in these regions. Feeding damage on young eucalypt foliage by T. catenata is typical of the paropsine group. The margin of the leaves become deeply scalloped suffering severe damage and some are completely consumed, leaving many of the growing tips of the sprig totally denuded (Plate 3). The actual year of establishment of T. catenata is unclear. Discussion with Messrs Crawshaw and Alexander at Morere indicate that the species may have been established at the Morere sites for two seasons previous to discovery by Ministry of Forestry /NZFRI. As the time of establishment is unclear, the idea was also postulated that the T. catenata population had been established for some years but remained latent due to interspecific competition Chapter 1 Introduction 18 with P. charybdis. The decline of P. charybdis populations since the release of the parasitoid E. nassaui, could have freed T. catenata from that competitive restraint and allowed the species to increase in numbers. Before assessment of the future status of T. catenata in New Zealand could be evaluated specific areas of ecological information on the species were necessary. To this end I focussed on the areas of life history, phenology, mortality of larval and pupal stages and development as a function of temperature which comprise chapter two of this thesis. Chapter three focuses on the potential of eight eucalypt species as hosts which included assessing larval performance and female ovipositional preferences on each. In chapter four, by comparison with P. charybdis, the potential impact of the solitary egg parasitoid E. nassaui on T. catenata was evaluated. sy nthesises the information from the preceding chapters. Study sites Finally chapter five Field studies were conducted on the southern most population at Morere (177° 47' E, 380 59' S) in northern Hawkes Bay New Zealand (Fig. 1.1). Two sites were maintained over the period 1August1993 to 4 March 1995. Site AL was at Alexanders farm, and site CR at Crawshaws farm, Morere. Site AL comprised a fenced area ~ 350 m 2 under five mature E. viminalis on the north facing edge of a small woodlot. Inspection of this site during April 1993 revealed numbers of T. catenata adults greater than we had observed at any Chapter 1 Introduction 19 other site in the Gisborne/northern Hawkes Bay region. Overwintering adults were readily observed in the leaf litter on 24 July. Fencing of the site, installation of a continuous recording thermograph and erection of five cone traps (to monitor prepupa fall into the leaf litter) was completed on 1 August 1993 (Plate 1). Site CR comprised a fenced area""' 550 m2 under a mixed stand of E. viminalis and E. macarthurii also on the north facing edge of a woodlot approximately three times larger than that at site AL. Smaller numbers of adult T. catenata were observed at site CR in April and overwintering adults were observed on 25 Julv. Fencing, installation of a continuous recording thermograph and erection of five cone traps (under E. viminalis only) was completed 29August1993 (Plate 2). Cone traps were constructed from moulded polythene, and were erected on four 1 m long timber stakes driven into the ground. The top diameter of each cone was 600 mm (2827 cm2) with a tapered bottom opening of 75 mm reducing to 70 mm (outside diameter). Over the bottom opening was placed a PVC pipe "catch" cup 73 mm (inside diameter) x 180 mm long. The cup had a fine gauze bottom inserted to allow water to pass through but retain all leaf material and larvae that dropped from the foliage. Thermograph records were obtained continuously from 1 August to 11 June for both 1993/94 and 1994/95 seasons. Graphical presentation of these data includes the mean, rounded mean, maximum and minimum based on the mean reading of each, for each 7 day interval (Fig. 1.2a&b). Dates included for these Chapter 1 Introduction 20 data are the end date for each interval and give a guide to the monthly temperature changes. .. -- ·--- ' · · · - -·-~ . .. - ·· / . !Opotoki ·· ...... : .. <' .. Otoko ·-- ---·----.. , Major Roads -···-··- Fig. 1.1 Gisborne/Hawkes Bay region of North Island New Zealand. Current range of T. catenata is within a 40 km radius (west to Otoko and south to Morere) of Gisborne city. 30 J3 25 -~ 20 :J ... n:I Q; 1 5 a. E ! 10 c: n:I Q) ~ 5 OU - ~ :J ..... n:I ... Q) 0. E Q) ..... c: n:I Q) :!!! 0 30 25 20 1 5 1 0 5 0 ·1·····i-··· : : : : : .(a) !S~ason. ~ 993/,94 ····i····i···· .... :··· ·:·· lJill~;~ CJ) ::i <( .,- : : : : ~ (b) !season: 1994/95 : ·- -~-----~·····-~-----~----·i----·i---·-i- --~------~-----~· - ········ .. -~ .... ! .... ! ..... ~ .. 0.. Q) (/) LO ..... u 0 C") . .. : .... -~ .. > 0 z I'- u Q) 0 LO c ro ., .,- : : : ..0 Q) l.J... LO ..... ro :? LO , :~·rr~~FH8~b:" ITTm(smoothed) ' ' ; . ~ · ~ .. :Min ..... 0.. <( N >. ro :? I'- ...... ; ····!·· c ::i ., .q- ., .. Min Fig. 1.2a&b Seasonal temperatures in 7 day intervals at site AL for seasons 1993/94 and 1994/95. Notes: Dates are the last day of each interval - Dotted line = actual mean temp. 9 Dl '"Cl rt >-t ..... ....... ::s ::r 0 p.. s::: () ::t: § N ..... a: u Jew Zealand. Bain, J.; Singh, P.; Ashby, M.D. and Van Boven, R.J. 1984 Laboratory rearing of the predatory coccinellid Cleobara mellyi (Col. : Coccinellidae) for biological control of Paropsis clzarybdis (Col. : Chrysomelidae) in New Zealand. Entomophaga 29(2): 237-244. Bain, J. and Kav, M.K 1989 Paropsis charybdis Sral, eucalyptus tortoise beetle (Coleoptera: Chrvsornelidae) In Cameron, P.J.; Hill, R.L.; Bain, J.; Thomas, VV.P. (eds.) (1989) "A revieu> of biological control of pests and weeds in New Zealand 1874 to 1987" Technical communication No. 10. C.A.B. international & DSIR, Wallingford, U.K. Bashford, R. 1993 Insect problems of eucalypt plantations in Australia: 4. Tasmania. Australian forestry 56(4): 375-377. Brooker, M.I.H. and Kleinig D.A. 1990 Field guide to eucalypts (Vol. II) south western and southern Australia. Inkata Press, Sydney. Burdon, J.J. and Chilvers, G.A. 1974a Fungal and insect parasites contributing to niche differentiation in mixed stands of eucalypt saplings. Australian journal of botany 22: 103-114. Chapter 1 Introduction 23 Burdon, J.J. and Chilvers, G.A. 1974b Leaf parasites on altitudinal populations of Eucalyptus paucifiora Sieb. Ex Spreng. Australian journal of botany 22: 265-269. Campbell, K.G. 1962 The biology of Roeselia lugens (Walk.), the gum-leaf skeletoniser moth, with particular reference to the Eucalyptus camaldulensis Dehn river red gum forests of the Murray valley region. Proceedings of the Linnean society of New South Wales 87: 316-318. Carne, P.B. 1966a Ecological characteristics of the eucalypt-defoliating chrysomelid Paropsis atomaria 01. Australian journal of zoology 14: 647-672. Carne, P.B. 1966b Growth and food consumption during the larval stages of Paropsis atomaria (Coleoptera : Chrysomelidae). Entomologia experimentalis et applicata 9: 105-112. Carne, P.B.; Greaves, R.T.G. and Mcinnes, R.S. 1974 Insect damage to plantation - grown eucalypts in north coastal New South Wales, with particular reference to christmas beetles (Coleoptera: Scarabaeidae). Journal of the Australian entomological society 13: 189-206. Cillie, J.J. 1981 The eucalyptus tortoise beetle Trachymela tincticollis Blackburn (Chrysomelidae: Coleoptera ). Pamphlet 273, Pests and diseases of South African forests and timber. Department of environment affairs, Pretoria. Clark, A.F. 1930 Paropsis dilitata Er. in New Zealand. Preliminary account New Zealand journal of science and technology 12(2): 114-123. Clark, A.F. 1938 A survey of the insect pests of eucalypts in New Zealand. New Zealand journal of science and technology 19(12): 750-761. Chapter 1 Introduction Clarke, A.R.; Zalucki, M.P.; Madden, J.L.; Patel, V.S. and Paterson, S.C. 1997 Local dispersion of the Eucalyptus leaf-beetle Chrysophtharta bimaculata (Coleoptera: Chrysomelidae), and implications for forest protection. Journal of applied ecology 34: 807-816. Oarke1 A.R.; Shohet, D; Patet V.S. and Madden1 J.L. 1998 Overwintering 24 sites of Chrysophtharta bimaculata (Olivier) (Coleoptera: Chrysomelidae) in commercially managed Eucalyptus obliqua forests. Australian journal of entomology 37: 149-154. de Little, D.W. and Madden, J.L. 1975 Host preference in the Tasmanian eucalypt defoliating paropsini (Coleoptera: Chrysomelidae) with particular reference to Chrysophtharta bimaculata (Olivier) and C. agricola (Chapuis). Journal of the Australian entomological society 14: 387-394. de Little, D.W. 1979 Taxonomic and ecological studies of the Tasmanian Eucalyptus - defoliating paropsids (Coleoptera: Chrysomelidae), with particular reference to Chrysophtharta bimaculata (Olivier). Ph.D. thesis, University of Tasmania, Australia. de Little, D.W. 1983 Life cycle and aspects of the biology of Tasmanian eucalyptus leaf beetle, Chrysophtharta bimaculata (Olivier) (Coleoptera : Chrysomelidae). Journal of the Australian entomological society 22: 15-18. de Little, D.W. 1989 Paropsine chrysomelid attack on plantations of Eucalyptus nitens in Tasma_nia. New Zealand journal of ferestry science 19(2/3): 223-227. Chapter 1 Introduction de Little, D.W.; Elliott, H.J.; Madden, J.L. and Bashford, R. 1990 Stage­ specific mortality in two field populations of immature Chrysophtharta bimaculata (Olivier) (Coleoptera: Chrysomelidae). Journal of the Australian entomological society 29: 51-55. 25 Edwards, P.B. and Suckling, D.M. 1980 Cermatulus nasalis and Oechalia schellembergii (Hemiptera: Pentatomidae) as predators of eucalyptus tortoise beetle larvae, Paropsis charybdis (Coleoptera: Chrysomelidae), in New Zealand. New Zealand entomologist 7: 158-164. Edwards, P.B. and Wightman, J.A. 1984 Energy and nitrogen budgets for larval and adult Paropsis charybdis Stal (Coleoptera: Chrysomelidae) feeding on Eucalyptus viminalis. Oecologia (Berlin) 61: 302-310. Eldridge, K.G.; Davidson, J.; Harwood, C.H. and van Wyk, G. 1993 Eucalypt domestication and breeding. Oxford University Press, New York. Elliott, H.J.; Bashford, R.; Greener, A. and Candy, S.G. 1992 Integrated pest management of the Tasmanian leaf beetle, Chrysophtharta bimaculata (Olivier) (Coleoptera: Chrysomelidae). Forest ecology and management 53: 29-38. Elliott, H.J.; Bashford, R. and Greener, A. 1993 Effects of defoliation by the leaf beetle Chrysophtharta bimaculata, on growth of Eucalyptus regnans plantations in Tasmania. Australian forestry 56(1): 22-26. Elliott, H.J. and de Little, D.W. 1984 Insect pests of trees and timber in Tasmania. The Forestry Commission, Tasmania. Flux, A.; Gadgil, P.; Bain, J. and Nuttall, M. 1993 Forest Health: Forest, tree and wood protection in New Zealand. Ministry of Forestry, Wellington. Chapter 1 Introduction 26 Fox, L.R. and Macauley, B.J. 1977 Insect grazing on Eucalyptus in response to variation in leaf tannins and nitrogen. Oecologia (Berlin) 29: 145-162. Greaves, R. 1966 Insect defoliation of eucalypt regrowth in the Florentine valley, Tasmania. Appita 19: 119-126. Greener, A. and Candy, S.G. 1994 Effect of the biotic insecticide Bacillus thuringiensis and a pyrethroid on survival of predators of Chrysophtharta bimaculata (Olivier) (Coleoptera: Chrysomelidae). Journal of the Australian entomological society 33: 321-324. Gurr, L. 1957 A note on the occurrence of the Eucalyptus tortoise beetle (Paropsis dilatata) in the North Island. New Zealand journal of science and technology 38A(8): 807. Hall, G.P. 1992 Host preferences of paropsini (Coleoptera: Chrysomelidae) in south-western Australia. Journal of the royal society of western Australia 75: 19-20. Harcourt, R.L.; Llewellyn, D.; Morton, R.; Dennis, E.S. and Peacock, W.J. 1996 Effectiveness of purified Bacillus thuringiensis berliner insecticidal proteins in controlling three insect pests of Australian eucalypt plantations. /ournal of economic entomology 89(6): 1392-1398. Jackson, T.A. and Poinar, G.O. 1989 Susceptibility of eucalyptus tortoise beetle (Paropsis charybdis) to Bacillus thuringiensis var. San Diego. Proceedings of the 42nd weed and pest control conference. Pp. 140-142. Kay, M. 1990 Success with biological control of the eucalyptus tortoise beetle, Paropsis charybdis. lNhats new in forest research No. 184. Rotorua, Forest Research Institute of New Zealand. Chapter 1 Introduction 27 Landsberg, J.J. and Cork, S.J. 1997 Herbivory: interactions between eucalypts and the vertebrates and invertebrates that feed on them. In Williams, J. and \'Voinarski, J. (eds.) Eucalypt ecology: individuals to ecosystems. Cambridge University Press. Larsson, S. and Ohmart, C.P. 1988 Leaf age and larval performance of the leaf beetle Paropsis atomaria. Ecological entomology 13: 19-24. Lawrence, J.F. and Britton, E.B. 1991 Coleoptera (Beetles) In Naumann, I.D. (ed.) The insects of Australia (2nd ed.) Melbourne University Press: Carlton. Majer, J.D.; Recher, H.F.; Wellington, A.B.; Woinarski, J.C.Z. and Yen, A.L. 1997 Invertebrates of eucalypt formations. In Williams, J. and Woinarski, J. (eds.) Eucalypt ecology: individuals to ecosystems. Cambridge University Press. McGregor, P.G. 1989 Ecology of Paropsis charybdis Stal, (Coleoptera: Chrysomelidae) a Eucalyptus defoliator in New Zealand. Ph.D. thesis I\fassey University, New Zealand. Miller, D. 1984 Common insects in Neu> Zealand. A.H. & A.W. Reed Ltd. Wellington. Morrow, P.A. 1977 Host specificity of insects in a community of three co­ dorninant Eucalyptus species. Australian journal of ecology 2: 89-106. Morrow, P.A. and Fox, L.R. 1980 Effects of variation in Eucalyptus oil yield on insect growth and grazing damage. Oecologia (Berlin) 45: 209-219. Chapter 1 Introduction 28 Ohmart, C.P. 1984 Is insect defoliation in eucalypt forests greater than that in other temperate forests. Australian journal of ecology 9: 413-418. Ohmart, C.P. 1990 Insect pests in intensively-managed eucalypt plantations in Australia: Some thoughts on this challenge to a new era in forest management. Australian forestry 53(1 ): 7-12. Ohmart, C.P. 1991 Role of food quality in the population dynamics of chrysomelid beetles feeding on Eucalyptus. Forest ecology and management 39: 35-46. Ohmart, C.P. and Edwards, P.B. 1991 Insect herbivory on eucalypts. Annual review of entomology 36: 637-657. Ohmart, C.P.; Stewart, LG. and Thomas, J.R. 1983 Leaf consumption by insects in three Eucalyptus forest types in southeastern Australia and their role in short-term nutrient cycling. Oecologia (Berlin) 59: 322-330. Ohmart, C.P.; Stewart, LG. and Thomas, J.R. 1985a Effects of food quality, particularly nitrogen concentrations, of Eucalyptus blakelyi foliage on the growth of Paropsis atomaria larvae (Coleoptera : Chrysomelidae). Oecologia (Berlin) 65: 543-549. Ohmart, C.P.; Stewart, LG. and Thomas, J.R. 1985b Effects of nitrogen concentrations of Eucalyptus blakelyi foliage on the fecundity of Paropsis atomaria (Coleoptera: Chrysomelidae). Oecologia (Berlin) 68: 41-44. Ohmart, C.P.; Thomas, J.R. and Stewart LG. 1987 Nitrogen, leaf toughness and the population dynamics of Paropsis atomaria Olivier (Coleoptera : Chrysomelidae) - A hypothesis. Journal of the Australian entomological society 26: 203-207. Chapter 1 Introduction Patterson, K.C.; Oarke, A.R.; Raymond, C.A. and Zalucki, M.P. 1996 Performance of first instar Chrysophtharta bimaculata larvae (Coleoptera: Chrysomelidae) on nine families of Eucalyptus regnans (Myrtaceae). Chemoecology 7: 94-106. 29 Ramsden, N. and Elek, J. 1998 Life cycle and development rates of the leaf beetle Chrysophtharta agricola (Chapuis) (Coleoptera: Chrysomelidae) on Eucalyptus nitens at two temperature regimens. Australian journal of entomology 37: 238-242. Reid, C.A.M. 1992 Descriptions of the pupae of nine genera of Australian Paropsine Chrysomelinae (Coleoptera: Chrysomelidae). Proceedings of the Linnaean society of NeuJ South Wales 113(4): 311-337. Shohet, D. and Clarke, A.R. 1997 Life history of Chauliognathus lugubris (F.) (Coleoptera: Cantharidae) in Tasmanian forests. Australian journal of entomology 36: 37-44. Springett, B.P. 1978 On the ecological role of insects in Australian eucalypt forests. Australian journal of ecology 3: 129-139. Steinbauer, M.J.; Clarke; A.R. and Madden, J.L. 1998 Oviposition preference of a Eucalyptus herbivore and the importance of leaf age on interspecific host choice. Ecological entomology 23: 201-206. Steven, D. 1973 The host-plant relationships of Paropsis charybdis Stal (Coleoptera: Chrysomelidae). Ph.D. thesis, Lincoln University, New Zealand. Styles, J.H. 1970 Notes on the biology of Paropsis charybdis Stal (Coleoptera: Chrysomelidae). New Zealand entomologist 4(3): 103-111. Chapter 1 Introduction 30 Tanton, M.T. and Epila, J.S.O. 1984 Parisitisation of larvae of Paropsis atomaria OJ. (Coleoptera: Chrysomelidae) in the Australian Capital Territory. Australian journal of zoology 32: 251-259. Tanton, M.T. and Khan, S.M. 1978 Aspects of the biology of the eucalypt­ defoliating chrysomelid beetle Paropsis atomaria 01. in the Australian Capital Territory. Australian journal of zoology 26: 113-120. Thompson, G. M. 1922 Naturalisation of animals and plants in New Zealand Cambridge University Press, Cambridge. Tribe, G.D. and Cillie, J.J. 1985a The eucalyptus tortoise beetle: a recent arrival in South Africa. Proceedings, 5th entomological congress of tlze entomological society of South Africa. Pp. 39-40. Tribe, G.D. and Cillie, J.J. 1985b A device to monitor larvae of the eucalvptus tortoise beetle, Trachymela tincticollis (Chrysomelidae: Paropsini). Journal of the entomological society of Southern Africa 48: 213. Tribe, G.D. and Cillie, J.J. 1989 Biological control of the eucalypt defoliating Australian tortoise beetle (Coleoptera : Chrysornelidae). Proceedings, 5th entomological congress of the entomological society of South Africa. p. 118. Tribe, G.D. and Cillie, J.J. 1997 Biology of the Australian tortoise beetle Trachymela tinctico/lis (Blackburn) ( Chrysomelidae: Chrysomelini : Paropsina), a defoliator of Eucalyptus (Myrtaceae), in South Africa. African entomology 5(1): 109-123. Chapter 1 Introduction 31 Urban, A.J.; Tribe, G.D. and Cillie, J.J. 1987 Dispersal of the introduced parasitoid, Enoggera polita (Hymenoptera: Pteromalidae), and its initial colonization of eucalyptus tortoise beetle, Trachymela tincticollis, (Coleoptera: Chrysomelidae), in the Southern Cape province. Proceedings, 6th entomological congress of the entomological society of South Africa. Pp. 77-78. Wardell-Johnson, G.W.; Williams, J.E.; Hill, K.D. and Cumming, R. 1997 Evolutionary biogeography and contemporary distribution of eucalypts In Williams, J. and Woinarski, J. (eds.) Eucalypt ecology: individuals to ecosystems. Cambridge University Press. Williams, J.E. 1990 The importance of herbivory in the population dynamics of three sub-alpine eucalypts in the Brindabella range, south-east Australia. Australian journal of ecology 15: 51-55. Williams, J.E. and Brooker, M.I.H. 1997 Eucalypts: an introduction In Williams, J. and Woinarski, J. (eds.) Eucalypt ecology: individuals to ecosystems. Cambridge University Press. Chapter 2 A . . ,. Ecology of Trachymela catenata Ecology of the recently established paropsine Trachymela catenata Chapuis (Coleoptera : Chrysomelidae) in New Zealand: Life history, Phenology, Mortality and Development. Abstract: Manv members of the Australian paropsina (Coleoptera : Chrysomelidae) are serious defoliating pests of Eucalyptus trees. Knowledge of the ecology of Trachymela catenata, was necessary to assess the potential pest status of this recent introduction to New Zealand. Field records show T. catenata to be feeding on members of the Symphyomyrtus sub-genus of eucalypts, especially E. viminalis. T rachymela catenata life stages are similar to other paropsina and females are as fecund as many of those which have reached pest status in other countries. Phenological data shows overwintering T. catenata adults emerge and lay eggs in spring with resulting larvae and adults grazing Eucalyptus foliage over spring/summer causing damage to host trees. Similar to other paropsina, evidence points to T. catenata being bivoltine. Over a season, mortality is highest in first instar larvae (60%) compared 14% and 13% for second and third instars respectively. Using a simultaneous combination of cone and emergence traps, mortality of the prepupal/pupal stage was Chapter 2 Ecology of Trachymela catenata 33 estimated at 45.8%. Using a range of six temperatures (range 7 - 32 °C) developmental thresholds were determined for each life stage. Thresholds were 4.95 °C for eggs, 6-7 °C for larval stages and 8-9.5 °C for pupal stages. Development rates were shown to be similar (including a decline in rate at the highest temperature trialed) to results obtained for other paropsina. Thermal conditions in the current geographical range and that of all (except colder alpine areas) of New Zealand should not be limiting to T. catenata's potential spread. Introduction: In December 1992, a tortoise beetle (Coleoptera : Chrysomelidae) previously unrecorded in New Zealand, was found in the Gisborne and Northern Hawkes Bay areas. Specimens of this Eucalyptus defoliator were identified by Dr. Chris Reid (Australian National University) as Trachymela catenata Chapuis. Trachymela catenata is one of a large group of species belonging to the Australian paropsine group comprising members of the subtribes Paropsina and Dicranosternina (Reid, 1992). Chapter 2 Ecology of Trachymela catenata 34 Both paropsine larvae and adults feed on leaves of the host tree and can be serious pests of plantation eucalypts in Australia (Bashford 1993; de Little 1989; Elliott et al. 1992, 1993; Greaves 1966; Kile 1974; Ohmart and Edwards 1991), South Africa (Cillie 1981; Tribe and Cillie 1985, 1997) and New Zealand (Styles 1970; Steven 1973; Bain 1977; Bain and Kay 1989; McGregor 1989) as well as acacias throughout Australia (Elliott and de Little 1984). Five members of the Paropsina have been recorded in New Zealand since early this century (see Chapter 1) and all appear to have established stable populations (M. Kay, pers. comm.). The most notable of these was Paropsis charybdis Stahl first recorded at Banks Peninsula in 1916 (Clark 1930, 1938; Bain 1977; Bain and Kay 1989). Paropsis charybdis reached serious pest status and caused severe defoliation and tree loss, wherever establishment of farm and commercial forestry plantings of eucalypts (particularly those belonging the Symphyomyrtus sub-genus - Section Maidenaria) were attempted (Bain 1977; Bain and Kay 1989; McGregor 1989; Steven 1973; Kay 1990). These host species include E. globulus, E. viminalis, E. macarthurii, E. nitens, and E. ovata. Paropsis charybdis is relatively rare in Australia and little information about its biology exists from that country. A number of studies conducted since the 1970s however, provide considerable information on the biology of P. charybdis in New Zealand (Edwards 1982; Edwards and Suckling 1980; Edwards and Wightman 1984; McGregor 1989; Steven 1973; Styles 1970). The pest status of P. charybdis, was such that since 1933 a series of attempts at finding a suitable Chapter 2 Ecology of Trachymela catenata 35 biocontrol agent for this species were undertaken (Bain and Kay 1989). In 1988, release of the egg parasitoid Enoggera nassaui (Hymenoptera : Pteromalidae) resulted in a severe decline in P. charybdis numbers (Kay 1990) throughout most areas of New Zealand. Trachymela sloanei, first recorded in New Zealand in 1976 (Bain 1977), utilises similar eucalypt host species to P. charybdis (i.e. E. macarthurii, E. viminalis, globulus, E. pulchella, E. ovata and E. botryoides) (M. Kay pers. comm.) but has not reached pest proportions, has been slow to disperse, and is presently limited to the northern half of the North Island. Like P. charybdis, catenata is relatively rare in Australia and little information is available on the biology or ecology of this species. This work is intended to provide information on the biology and ecology of catenata which can be compared to what is known of P. charybdis and other paropsina in an effort to help evaluate the future status of this species in New Zealand. Key areas of investigation were to : 1, determine the life stages and seasonal population patterns and the habitats used for these; 2, investigate temperature requirements (a well documented correlate associated with development and important when predicting population growth and potential geographic distribution); and 3, investigate reproductive capacity and mortality (important when considering potential population growth). Chapter 2 Ecology of Trachymela catenata 36 This paper gives a brief morphological description, reports on life history, habitat use, phenology, development, fecundity and mortality of T. catenata using both field and laboratory data and observations. Methods: Adult Identification Adult Trachymela catenata and T. sloanei are morphologically very similar and the two species coexist in localised areas of the Gisborne region. Dorsally the pronotum of both species extends well over and around the head, obscuring the posterior margin of the eyes and leaving only the anterior portion of the head, the filiform antennae and labrum visible. The elytra are oval and convex with the elytral suture extending to the tips. The elytra also have a large epiplural area, with a heavily reinforced margin that extends well below the level of the abdominal sternites thus covering the entire abdomen and completely obscuring the legs (when at rest) giving the beetle its typical "tortoise-like" appearance. Unlike T. sloanei, the adult head, pronotum and elytra of T. catenata are shiny dark to rusty brown with black marking varying in size between individuals. The black markings of the pronotum usually form a central, anteriorly oriented "W" shape (Fig. 2.la). with a black area also Chapter 2 Ecology of Trachymela catenata 37 at the lateral margin. The elytral markings generally consist of two semicircular black areas at the apex of the suture which form a central 'spot' and anterior to this are similarly sized black areas sometimes extending to the anterior margin (humeri) (Fig. 2. la). Black areas of varying size also extend from the humeri to approximately halfway along the epiplurae (Fig. 2.lb) and around the posterior epiplural margin. Trachymela sloanei are of a similar base colour but have small, uniformly distributed black specks over the entire surface of the elytra. A number of Trachymela species (including T. catenata and T. sloanei) have secretor.· glands on the pronotum and elytra which (in mature adults) produces a waxy secretion (Selman 1988) that covers the entire pronotum and elytral surface giving the animal a dusty flour like appearance (Plates 5 & 6). For both these species this masks many of the features described above and identification is only possible when the secretion is removed. Chapter 2 Ecology of Trachymela catenata 38 Ill 3: Q) ·s; -.c ('C I.. Cl) -ro "O c:: - ('C .c -('C - ('C Ill I.. 0 "O Cl;l -Cl;l t:: Cl) -Cl;l u \1 ....: -::l "O <( .c ~ ('C .,... N Cl E u.. E l.() -('C Plate 5. Teneral adult T.catenata showing no wax secret ion on pronotum or elytra . Plate 6. Mature adult T. catenata with wax secretion covering pronotum and elytra. Chapter 2 Ecology of Trachymela catenata 39 Life history and Phenology: Field studies were conducted at Morere (177° 47' E, 38° 59' S) in northern Hawkes Bay, New Zealand, from 1August1993 to 4 March 1995. Two sites were maintained. One, site AL (Alexanders) comprised of a fenced area of ""' 350 m 2 under five mature E. viminalis trees on the north facing edge of a small woodlot The second, CR (Crawshaws), comprised a fenced area of""' 550 m 2 under mature E. viminalis on the north facing edge of a much larger woodlot This woodlot also included a small number of macarthurii. Continuous thermograph records were maintained at both sites. Life history of T. catenata was determined from field observations and cultures reared in the laboratory. The number of larval stages was determined from these cultures by observation and by recording head capsule measurements. Similarly, pupal characteristics were determined from laboratory and field observations. Phenological data were obtained from site AL at Morere from April 1994 to March 1995. Data on overwintering adults were obtained from monthly leaf litter searches over the period June to November 1994. These searches consisted of recording all live adult T. catenata from five 50 cm2 quadrats each consisting of the leaf litter layer and the top 2 cm of soil. Estimates of the mean number of live adults per quadrat were used to assess the overwintering population over time. Chapter 2 Ecology of Trachymela catenata 40 Foliage surveys provided data on oviposition patterns, with the number of eggs per batch and the age class of leaf upon which each batch was laid being recorded. Age classes of leaves were determined as: class 1 the terminal leaf, class 2 the first pair of leaves below the terminal leaf, class 3 the second pair below the terminal leaf (adjacent to leaf class 2) and so on down to the leaf class 7. All classes after that pair were termed >7. Mean egg batch size was then calculated along with the proportion of egg batches laid on each leaf-class over the season. Foliage surveys also provided data on adult phenology during spring/ summer 1994/95. These surveys comprised of recording adult numbers from each of 10 permanently tagged branches, thus standardising the leaf area surveyed. The mean number of adults from the 10 branches for each survey was calculated and then expressed as the proportion of total mean number over the season to give relative seasonal abundance (RSA). Prepupae fall from the foliage to pupate in the leaf litter. This allowed prepupal numbers to be recorded from five cone (litter-fall) traps (600 mm top opening tapering to a 73 mm diameter gauze bottomed "catch cup" (see Chapter 1)) erected under E. viminalis foliage. Dead fourth instar larvae are morphologically similar to the prepupal form and difficult to tell apart. All specimens of this size and morphology were therefore considered prepupae. Chapter 2 Ecology of Trachymela catenata 41 Dead larvae representing all instars were also obtained from these traps and together with the prepupal catch provided phenological data. Trap samples containing the larvae and prepupal fall were collected monthly from September 1993 to April 1994 and every 7 -10 days from September 1994 to April 1995 and preserved in 70% EtOH until counting. The 1994 to 1995 samples for 1st instar and prepupae were used to provide data on larval phenology. As for the RSA for adults (above), the mean number of larvae from the five traps for each sample period was obtained and expressed as the proportion of total mean number over the season to give relative seasonal abundance (RSA). RSA= Spx = L:T1-s Tn Spx L: Spx,1-12 WhereSpx Tl ... Mean number trapped for sample period x Trap numbers 1-5 Total number of traps Tn Spx,1 12 = Mean number trapped for sample periods 1 - 12 Development trials: Trials testing larval and pupal development time as a function of temperature were initiated using newly hatched T. catenata larvae which had had the opportunity to consume the egg chorion i.e. :S 24 h old. One larva was placed in a petri dish containing a layer of damp paper towelling (to maintain humidity) and a small adult leaf of flush E. viminalis. One replicate comprised ten such dishes. Three replicates were run at each of six temperatures i.e. 7, 12, 17, 22, 27 and 32 °C (± 1.5 OC) and 16 h photoperiod. During the larval growth stages (1st - 4th instars) leaves were replaced every 48 h and enough moisture replaced to keep the towelling damp but not "wet". Pupal stages Chapter 2 Ecology of Trachymela catenata 42 continued to be maintained in the damp dish conditions. Each individual was monitored twice daily, assessed as live or dead, and its instar recorded. Monitoring was maintained until adult emergence. Development time (t) for each of the six instars, at each temperature, was established (to the nearest half day) using the value at which 2: 50% of larvae/ pupae had moulted to the next instar. Using the three replicates, mean and standard errors (S.E.) were obtained. To test the validity of the effect of temperature and instar on development time, two way ANOV A including an interaction term was run on the development times for each of the six instars. Results from trials at 7 °C were not included as complete development did not occur for any instar at this temperature. Trials to determine development time of eggs (laying to edosion) were conducted over the same range of temperatures as that used for larval/pupal development Three replicates with 30 - 68 eggs in each, were set up for each temperature using eggs of a known age (range 8 - 24 h). Eggs were monitored twice daily until eclosion. The number of days on which the greatest number of eggs eclosed was adjusted for age at beginning of trial and the mean and standard error of the three replicates obtained. Total development time (egg lay to adult emergence) was obtained from the combined times of egg, larval and pupal development trials. Chapter 2 Ecology of Trachymela catenata 43 To determine developmental thresholds for each instar, development rate needs to be determined from development time at each temperature. Development rate was determined as the inverse of those values obtained for development time Development rate 1/ t and the mean and standard errors (S.E.) established over the three replicates. These data were subjected to regression analysis of the four linear points i.e. 12, 17, 22 and 27 co and developmental thresholds taken as the x intercept of the regression line (Harman et al. 1989; Petitt et al. 1991) The mean developmental threshold over all stages was then obtained. Mortality and infertility: Egg infertility data were gained from the various trials where 1st instar larvae were required. The number of larvae hatched in relation to the number of eggs collected from cultures provided these estimates. Unhatched eggs were checked under the microscope to confirm that no development had occurred. Leaves with eggs intact were recovered from cone trap samples. These provided an estimate of egg losses (mortality) in the field due to infertility and leaf abscission. Stage-specific mortality rates were not recorded for larvae in the field, but cone trap samples provided a measure of larval mortality (again expressed as RSA) and estimates of the mean percent mortality for 1st, 2nd and 3rd instar larvae Chapter 2 Ecology of Trachymela catenata 44 were gained over the entire sampling period. To investigate differential mortality between 1st and 2nd (spring and summer) generations these data were divided into two appropriate groups. 1st generation data were taken from 3 December to 31 December and 2nd generation from 18 January to 4 March. This removed 26 November data which contained no 3rd instars, removed 7 January to reduce overlap of the two generations and provided an equal number of sample periods in each group. The sum of the mean frequency of mortality for each generation for each instar was obtained and subjected to Chi-square analysis to test the null hvpothesis that there was no difference in the frequency of mortality for each instar between generations. Mortality data for the prepupal/pupal stage were obtained from the addition of emergence traps, (prior to spring 1995) to the five cone traps (each 600 mm diameter (2827 cm2) previously alluded to. This set up consisted of a series of five gauze-covered emergence traps being placed around each of the above cone traps (Plate 4). Emergence traps were 50 x 50 cm (2500 cm2) at the base tapering to a 12 mm funnel hole, opening into a glass jar (Plate 4). At the beginning of prepupal fall, one emergence trap was shifted back, thus exposing the quadrat it previously covered, and allowing prepupa to fall into that area. At seven-day intervals, the adjacent traps were sequentially shifted back to cover the previously exposed quadrat and expose the next one for prepupa to fall into. In this manner cone traps provided a weekly estimate of the number of prepupae falling into the leaf litter and emergence traps an estimate of live adults emerging. Numbers in the 2500 cm2 emergence trap were reduced by Chapter 2 Ecology of Trachymela catenata 45 the ratio of the difference in cm2 of the larger cone trap, then percent mortality gained for the five replicates. Mean mortality over the five data sets was then estimated. Results: Considerable variation in catenata population sizes existed between the two study sites despite them comprising the same dominant mature E. viminalis species, being just 7 km apart and at the same elevation. Site AL, while small, maintained a much larger population of T. catenata than site CR. Site CR, (set up as a replicate of AL), provided very little information from litter surveys or cone trap samples. Life History: Eggs, once hatched, progress through four larval instars. A prepupal stage precedes the pupal form, which then emerges into an adult Egg batches (mean number of eggs= 7.67 ± 0.27 S.E., n = 76, range 3 -16) are laid in a single row and with an apparent preference for the flush 3rd and 4th leaf pair (age class) of the sprig (Fig 2.2) . Chapter 2 (!) 0 c: (ti 50 ~ 40 :::J .c (ti ~ 30 0 IJ) co (!) :.. 20 0 c: 0 ~ 10 0.. 0 .... a.. 0 c1 c2 Ecology of Trachymela catenata 46 c3 c4 c5 c6 c7 c>7 Leaf age class Fig. 2.2 Proportion of egg batches laid on each leaf class. (Classes 1-7 =current seasons growth and class >7 = previous seasons growth). Leaf age classes are described in methods. Observations showed that newly hatched larvae consume the egg chorion and then feed exclusively on flush foliage showing a strong preference for the soft tissue of the terminal, and 2nd and 3rd leaf pairs during the 1st and 2nd instars. Third and 4th instars also feed on softer leaves but as the larvae mature, they shift to older, thicker, less succulent leaves. An indication of female fecundity was determined from cultures of T. catenata maintained at 27 ± 1.5 °c. Three females produced mean egg batch sizes of 7.63 ± 0.19 S.E. eggs batch-1 (n = 205). The mean number of eggs laid per female per day was 6.9 with oviposition periods ranging from 65 - 88 days. This provided an estimate of 522 ± 89 S.E. eggs per female over the laying period and indicates they lay approximately one batch per day. Chapter 2 Ecology of Trachymela catenata 47 The prepupal stage was characterised by the larvae ceasing to feed, taking on a curved semi-circular appearance and dropping to the leaf litter. The pupal form then develops a few days later within the leaf litter. No evidence was found of pupae forming pupal cells in the litter or soil as has been described for P. atomaria (Carne 1966) and T. tincticollis (Tribe and Cillie 1997). Adults emerged from the leaf litter to then feed on the foliage. Observations from the laboratory cultures (at 22 °C) showed that mating and oviposition can occur in 7 - 12 days. Duration from adult emergence to mating and oviposition under variable field temperatures however is not known. Phenology: At the AL site Morere, observations from monthly leaf litter survevs for overwintering adults suggests a preference for deeper moist conditions (presumably to avoid desiccation) and quantified a progressive decline in mean number per quadrat from 24 July 1994 (mean = 4.12) until 30 October 1994 when no more were found (Fig. 2.3 c). October 30 foliage surveys however, revealed a substantial increase in the Relative Seasonal Abundance (RSA) of adults (RSA = 10.1, Fig. 2.3 c). This shift of habitat coincided with a progressive increase in mean daily temperatures (as determined from the thermograph records) from 9 oC (winter) to 13 °C in October - November (spring) (see Chapter 1). Chapter 2 Ecology of T rachymela catenata 48 30 j (a) " u 25 1 • c J " 'O c " 20 .0 " • :;::; • •• ..!!! 5.0 l . "+ - - • -" '+ c:: • 0.0 1 --,--·--·--, - 30 1 (b) " u 25 -c "' ~ 'O t. 5 .0 20 "' "' c 15 0 _-J • "' "' • " "' 1 0 • " .~ • lii 5.0 ~ • " c:: •• .. .. 0.0 30 ' (c) Key: t 8 Q) 1 Black diamond = RSA adults " " " Open diamond = Overwinter adults ::: .~ 7 -"' 25 ~ 'O c 6 'O " " "' .0 20 ' ::J "' i I F- C' 5 ~ .. "' :;::; -1 y ......... ' f " .!!! 5.0 . \ -· - ... ., 1 ::;: Q) • . a:: . t • • ·- .. -',--·~·--+- ... ::'- .... " 0.0 . ' I 0 Q) ~ 0. u u > > " u " " " c c c .0 .0 ~ c 0 0 Q) Q) Q) Q) Q) "' ro "' " " "' " " Q) 0 0 z z 0 0 0 0 0 ..., ..., ..., u.. u.. ::;; ..., ..., (jJ " N 0 m CD "" 0 "' " ,.. "" m m "' " CD N N "" N N "" N N Fig. 2.3 a,b,c. Relative seasonal abundance of 1st instar (a), 4th instar/pre pupa (b) and adult (c) T. catenata over 1994/95 season. Overwintering adults equal mean number (including S.E.) per quadrat Chapter 2 Ecology of Trachymela catenata 49 Egg batches for the 1994/95 season were first recorded on 30 October. Batch numbers then increased steadily over time with a substantial increase recorded on 10 December. First and 2nd instar larvae were observed during the above period with 3rd and 4th instars becoming more readily seen as time progressed. These observations are supported by RSA from the cone trap samples for the period 26 November - 10 December (Figs. 2.3 a&b) where a number of 1st instar larvae were caught but few 4th instar/prepupa stages are evident. Numbers for 1st instars peaked on 24 December with a second, though smaller peak evident 47 days later on 9 February (Fig. 2.3 a). No egg batches were observed from 18 December through to 18 January. Egg batches were readily observed again on 29 January some 11 days before the second 1st instar peak. Adult numbers (from standard branch counts) revealed a substantial increase on 7 January (Fig. 2.3 c) some 14 days after the peak in 4th instar/prepupa fall. An increase in adult numbers is evident on the 4 March when sampling ceased. The foliage of the tagged standard count branches at this time held very little flush foliage, however increased adult numbers were observed on many small areas of flush not included in these tagged areas. These individuals had not yet developed the waxy elytral secretion indicating they had only recently emerged so are taken to be teneral second generation adults. Chapter 2 Ecology of Trachymela catenata 50 Development: Mean total development time decreased as temperature increased until a temperature of 32 °C was attained, after which development time increased (egg to adult time Table 2.1). No development occurred in any life stage at 7 oc and development did not proceed past the prepupal stage at 12 °C (Table 2.1). First instar development time continued to decrease over all temperatures. The 2nd and 3rd instars were consistently the shortest life stages except at 32 °C. A marked slowing of development occurred in the 4th instar also at 32 oc which accounts for the greatest part of the increase in development time as compared to 27 °C. Two way ANOV A indicated highly significant effects of temperature (F 3, n, = 32.71, P = 0.0001), instar (F s, 11, = 42.05, P = 0.0001) and temperature x instar interaction (F 1s, 71, = 7.22, P = 0.0001) on development time. A significant effect for trial was also obtained, seemingly due to the high mortality and slow growth rates at 12 °C, which for later instars in some trials produced highly variable data with zero data points and/ or very high single values. Exclusion of the 12 °c data set then, produced a non significant result (F 2, 71, = 0.40, P = 0.6742) for the effect of trial. ANOVA grouped 2nd and 3rd instars together as not being significantly different Table 2.1 Mean number of days (±5.E.) required for each instar of Trachymela catenata to develop under a range of constant temperatures. Temperature (OC) Stage 7 12 17 22 27 32 Eggs - 15.2±0.59 6.86±0.35 5.28±0.39 3.63±0.27 3.86±0.14 Ins tar 1st - 12.83±2.16 5.42±0.55 5.16±0.44 3.50±0.5 2.16±0.16 2nd - 10.33±1.45 4.00±0.57 3.00±0.28 2.16±0.17 2.33±0.16 3rd - 10.75±0.2 4.00±0.0 3.16±0.58 2.33±0.17 2.66±0.33 4th - 15.00±2.0 5.50±0.5 4.83±0.17 3.50±0.29 7.83±0.83 Pre pupa - 42.00±0.0 6.66±0.17 4.50±0.29 3.50±0.29 2.83±0.16 Pupa - - 12.66±1.92 7.50±0.5 5.66±0.33 5.16±0.66 Egg to adult - - 45.10±2.18 33.43±0.33 24.28±0.43 26.83±0.77 Total rate 0.022 0.029 0.041 0.037 I I g D.> '"\j m- N tTJ (") 0 ,_. 0 ~ 0 ........ '.:;l i:::i (') ~ ~ ('1;) ~ (') i:::i ~ s ~ (J1 ~ Chapter 2 >- 0.61 ('(! 0.51 "'O Qi -('(! .. 0.4 -i:::: (!) 0.3 i E c. 0 (!) > (1) 0.2 -j Cl 0.1 j i:::: ('(! (!) ~ ~ 0 I 0 Fig. 2.4 Ecology of Trachymela catenata y = -0.0953 = 0.0136 x x intersept = 7 5 10 15 20 25 30 35 Temperature 0c Egg development rate as a function of temperature and developmental threshold for T. catenata. (vertical bars= S.E.) 52 Development rate regression analysis reveals a developmental threshold of 7 °C for eggs (Fig. 2.4) and 2nd 3rd and 4th instar larvae (Fig. 2.5) of 6 - 7 °C but only 4.95 °C for the 1st instar stage (Fig. 2.5). Thresholds for prepupa and pupa are slightly higher at 8 - 9.5 °C. The mean development threshold for the species is 7.10 °C ± 0.57 (S.E.). Development rate is linear for eggs and all instars from 12 to 27 °C but a general decline in rate occurs (with the exception of 1 instar) at 32 °c (Fig. 2.5). Mortality and infertility: Estimates of losses during the egg stage were based on two factors: One; - egg infertility; and two - losses due to abscission of leaves upon which eggs were laid. Percentage of infertile eggs from laboratory trials was 1.5 - 2% (98% fertility). Chapter 2 0 0.6 J 0.5 i 0.4 i ~ 0.3 J ~ 0.2 J ~ 0.1 i 1st instar y = -0.0654 + 0.0132 x x intersept = 4.95 5 10 15 3rd instar y = -0.152 + 0 0219 x x intersept = 6.94 • • Ecology of Trachymela catenata 20 25 30 35 0 2nd instar y = -0.167 + 0.0235 x x intersept = 7.10 5 10 15 Temperature (°C ) i J J 4th instar y = -0 0807 + 0.0138 x x intersept = 5.85 20 25 o~,~~~.......,.~~~~~~~~~~~~~ 0 0.6 1 ~ 0.5 ~ 0.4 ~ 03~ 0.21 0.1 -=1 ~ 0 5 10 15 Pre pupa y = -0.168 + 0.0174 x x intersept = 9.65 • • 5 10 15 20 25 20 25 30 35 0 5 10 15 Temperature (°C ) j pupa y = -0.0785 + 0.0095 x x intersept = 8.22 20 25 ~ , . , . 30 35 0 5 10 15 20 25 Temperature (0 c ) 53 30 35 30 35 30 35 Fig 2.5 Larval and pupal development rate as a function of temperature and developmental thresholds forT. catenata. (vertical bars = S.E.) Chapter 2 Ecology of Trachymela catenata 54 Leaves with eggs intact occurred in four of the larval cone trap samples, thus some loss due to leaf abscission was evident. Total egg loss due to these two factors is estimated at 5%. Mean number (per trap) of dead larvae for each instar for each sampling period over the season is shown in Fig. 2.6. These samples provided estimates of percent mortality for each instar. Mean seasonal percentage was 60%, 14% and 13% for 1st, 2nd and 3rd instars respectively. An estimate of 4th instar mortality from the development trials however suggest that this stage suffers a similar mortality rate as the 3rd instar. 0.. ('(I .... -... Q) 0.. .... Q) .c E :::i c c ('(I Q) ~ :: l Key ~ 3rd lnstar • 2nd lnstar D 1st lnstar . 20 ~ j 15 ~ 10 i 5 0 > (.) (.) (.) (.) (.) c c c ..0 ..0 '- 0 (].) (].) (].) 7. Chrysophtharta bimaculata oviposits rows of eggs (de Little 1979a) while C. agricola deposits eggs in an untidy dump (Ramsden and Elek 1998) on flush foliage, similar to T. catenata. Aggregation behaviour of T. catenata larvae appears not to be as strongly developed as it is for some of the other paropsina. First instar larvae aggregate on and consume the chorion of the egg cluster before they begin feeding on foliage. During this instar, they were often observed in small aggregations but from the 2nd instar onwards, appeared to disperse singly Chapter 2 Ecology of Trachymela catenata 58 through out the foliage, and are only occasionally observed in two's or three's. T. tincticollis larvae are reported to conceal themselves in bark crevices during the day, only emerging at dusk or before dawn to feed (Tribe and Cillie, 1997). Conversely T. catenata larvae of all stages can be observed on the foliage at any time of the day. Carne (1966) reported that several P. atomaria larvae of the same or different instars aggregate whenever they are not feeding. Paropsis charybdis is similar in habit to T. catenata in forming loose aggregations while consuming the egg chorion and thereafter dispersing (McGregor 1989). I have however observed, in very windy conditions, small (n ~ 3) tightly aggregated groups of 3rd and 4th instar P. charybdis larvae in the axis of stems and small branches. Phenology: Trachymela catenata adults overwinter in the leaf litter apparently maintaining metabolic processes from the extensive fat bodies laid down from the previous autumns feeding (Carne 1966). Like P. atomaria (Carne 1966) adult T. catenata may be observed in late winter - early spring on unseasonally warm days, attached to logs or foliage but feeding was never observed at this time. Spring emergence appears to be temperature related (Carne 1966, McGregor 1989). Thermograph records from site AL for 1993 and 1994 indicated an increase in mean ambient temperature during October of about 3-4 °C. During October 1994, a marked decline in overwintering adult numbers in the leaf litter was matched by a definite increase in adult numbers on the foliage (Fig. 2.3 c). Chapter 2 Ecology of Trachymela catenata 59 Eggs (2 small batches) were first observed on 30 October and the number of egg batches increased slowly through to December. On 10 December 1994 a substantial increase in egg batch numbers was observed and numbers decreased rapidly thereafter. At 17 °C (the mean weekly temperature at that time) we expect egg development to take 6.9 days thus producing an increase in 1st instar numbers the following week. Field data support this expectation with the peak in 1st instar mortality being recorded for the 24 December sample period (Fig. 2.3 a). The second increase in egg numbers was recorded on 29 January. Using similar assumptions as for the first generation (but with a slightly shorter development time allowing for mean weekly temperature now at 19 °C) we would expect the increase in 1st instar mortality around 9 February (Fig. 2.3 a). The new egg batches observed on 29 January is 50 days after 10 December and along with the duration between 1st instar peaks may be indicative of the field generation time. From the temperature vs development trials, the estimated development time for a complete generation (at 17 to 19 °C ) is"" 41 days for eggs to adult plus 10 days adult emergence to oviposition time totals 51 days. Given sampling intervals of 7 to 11 days in my field data, variation of ± ?:.7 days may exist in the 50 day duration between the two egg/1st instar seasonal peaks. The estimated generation time of 51 days however, lies within this possible variation and is as close as these data allow. First generation adult numbers peak approximately 13 days after the maximum fall of 4th instar / prepupa (Fig. 2.3 b&c). The subsequent sampling period revealed a substantial decrease in adult numbers (Fig. 2.3 c). In Tasmania, Chapter 2 Ecology of Trachymela catenata 60 major dispersal events are characteristic of C. bimaculata (Clarke et al. 1997; de Little 1979a). Clarke et al. (1997) showed that C. bimaculata adults will disperse after egg laying and as well, found a negative correlation between adult numbers and tree damage or larvae present. During this time at site AL, very little flush foliage remained available as female oviposition sites but whether a similar dispersal event could explain this decline in numbers is not known. The second peak of 4th instar / prepupa appears only 22 days after the peak of the first generation adults. At 19 oc, adult maturation to oviposition, egg and larval development are estimated to take ;::; 33.5 days. Assuming those adults were responsible for the second peak in egg laying and hence the second generation, the estimated duration is unacceptably reduced. If the time is taken from the 31 December however (when adult numbers were increasing rapidly), that date provides a more acceptable 29 day period. The 4th instar / prepupa peak appearing earlier than predicted may well be error due to the problem of sampling interval previously alluded to. Adult numbers resulting from the second generation fall of 4th instar/prepupa were difficult to quantify because the standard branch count areas at that time (4 March) were no longer in a flushed state and few adults remained at these sample points. However, on 4 March, increased numbers of teneral adults were readily observed on adjacent small areas of flush foliage but obviously could not be included in these data. Chapter 2 Ecology of Trachymela catenata 61 This phenological pattern indicates that T. catenata appears to follow the bivoltine life history established for many of the other paropsine taxa. A single trial to mimic adult overwintering and the importance of flush feed for the initiation of spring oviposition was conducted in the laboratory. Overwintering adult T. catenata were collected and maintained in leaf litter for three months at a constant 12 °C. Over a period of 10 days (commencing on 15 October) the temperature was progressively increased to 22 °c. Once emerged from the overwintering state these adults, survived for> 8 weeks while being fed only last seasons sclerophyllous foliage. While on this diet both male and female adults formed waxy elytral secretions, appeared active and healthy but did not mate or oviposit After five weeks this culture was divided. One group of males and females was fed flush foliage while the other (males and females) remained on sclerophyllous feed. Adults on flush feed mated and commenced egg laying in 10 - 11 days whilst those remaining on old foliage layed no eggs at all. Carne (1966) suggests, that emergence from overwintering is initiated by the depletion of fat bodies, and my observation of fat bodies from dissected overwintering adults supports this hypothesis. The result of this trial may indicate that overwintering T. catenata adults can emerge in the spring and await the onset of flush foliage. The precise timing of emergence to coincide with eucalypt flush growth therefore, may not be an obligate requirement Such a strategy may confer considerable adaptive advantage to a phytophage Chapter 2 Ecology of Trachymela catenata 62 that is constrained to a single or a small range of host eucalypts that display considerable seasonal variability in the onset of spring growth. Development Results of the development versus temperature trials (where development time progressively decreased with increasing temperature), are comparable with development rates attained for C. bimaculata (de Little and Madden 1975) and C. agricola (Ramsden and Elek 1998) and produced a useful working model which will assist in estimating field based generation times. The major assumption with the working model is that growth rates and hence development time under constant temperatures are similar to those obtained in fluctuating field temperatures. Developmental studies on various insect species using constant temperatures have, when exposed to fluctuating temperatures, resulted in a decrease (Hagstrum and Leach 1973) or m some cases no difference (Kitching 1977) in development time. The effect of fluctuating temperatures on development time for T. catenata are unknown. Fourth instar development time at 32 °C increased markedly when compared to the development time of other stages at that temperature. Like many insects the greatest increase in volume and mass occurs during the final instar and any impairment to metabolic rate will presumably have its greatest effect at that stage. Interestingly, at 32 °C the development time of prepupa and pupa Chapter 2 Ecology of Trachymela catenata 63 continues to decrease which is reflected in an almost continual linear trend in development rate (Fig 2.5). Over all, egg and larval development rate increased as a function of temperature up to 27 °C with a decline (particularly during the larval stages) thereafter. The decline in development rate at 32 °c is in line with similar experimental work on P. atomaria (Carne, 1966). Maximum intrinsic growth rate for T. catenata therefore is at temperatures approaching the high 20s. In its present northern Hawkes Bay /Gisborne range, daily maximum temperatures can reach in excess of 27 °C but sustained mean temperatures at that level are unlikely. In terms of thermal requirement, population growth for T. catenata will most likely remain below its maximum potential. The relatively low developmental threshold (4.95 °C) for 1st instar larvae seems to arise from the relatively low development rates at 22 °C and 27 °c which affects the slope of the regression line. Explanation of these values is difficult Controlled temperature conditions as for all other trials, remained within± 1.5 0 c, humidity was maintained with damp paper towelling and foliage was carefully selected to maintain similar quality and sourced from the same tree for all trials in each replicate. However, with increased development rates at 22 oc and 27 °C the threshold rate is unlikely to increase to greater than 7 °C and given that all other larval thresholds are around 7 °C, 100% mortality of 1st instar larvae under controlled conditions is to be expected. Chapter 2 Ecology of Trachymela catenata 64 The overall developmental threshold for larval stages is similar to that reported by McGregor (1989) for P. charybdis. Results from the development trials indicate T. catenata larvae at 12 °C fail to complete development and need temperatures greater than 12 °C to persist. By the time 1st generation larvae are developing at site AL, mean daily temperatures are 15 - 17 °C and larvae are able to develop well. With an almost exponential rate of increase in development time from 17 down to 12 0 c, T. catenata is not likely to persist at mean weekly temperatures in the low teens because the generation time would be such that completion of one generation per season is unlikely. The mean spring/summer temperatures (17 - 19 °C) of the current range of T. catenata are not dissimilar to those in a large portion of New Zealand. The potential therefore for T. catenata to establish over a wide area of the country should not be limited by thermal requirement. The developmental threshold for prepupa/ pupal development was the highest of any stage at about 9.6 °c and development was not completed at 12 °C in the trials. Conceivably, these instars may be vulnerable to a period of very cold temperatures (perhaps not likely in its current range) which could have an important impact on the number of pupa entering adulthood. Further work is required to investigate how long T. catenata prepupa/ pupa can sustain temperatures around this threshold level and survive to continue development when conditions are again favourable. Chapter 2 Ecology of Trachymela catenata 65 Mortality: The laboratory results for egg fertility for T. catenata of 98% are similar to those reported for other paropsina (e.g. Ramsden and Elek 1998) and my estimate of an overall 5.0% loss due to infertility and leaf abscission may well be conservative. McGregor (1989), using marked egg masses, reported egg survival for P. charybdis of~ 94% in the field and 98.5% from laboratory trials. His field estimate does not appear to account for losses due to leaf abscission, so the survival rate of total eggs laid may be less than his estimate suggests. The verv low mortalitv for 2nd and 3rd instar larvae earlv in the season (26 - - - November - 10 December) is expected due to the time lag for the population to attain those stages. This is in line with the phenology data. Thereafter, 1st instar mortality appears relatively high (as compared to 2nd and 3rd instars) in the spring. Flush foliage appears abundant at this time, however minimum daily temperatures do drop to around 5 °C. It is conceivable that such temperatures could reduce the larvae' s metabolic rate and impair their ability to maintain a hold on the leaf and be instrumental in the high mortality rate for this stage. The Chi-square analysis comparing larval mortality between instars of the 1st and 2nd generations was not significant The test was justified on the basis that for the second generation, mortality rate of 2nd and 3rd instar larvae appeared high compared to 1st instars when compared with the 1st (spring) generation. During the second generation period 18 January to 4 March Chapter 2 Ecology of Trachymela catenata 66 considerable defoliation from 1st generation larvae and newly emerged teneral adults was apparent This appeared to result in much poorer feed quality and quantity (this however was not quantified) which may conceivably manifest itself in increased mortality, especially for the larger instars which require ever increasing quantities of quality food. Mortality during the pupation stage of 45.8% is considerably less than that reported for P. charybdis at~ 89% (McGregor, 1989). McGregor (1989) shifted his emergence traps every fortnight My method of emergence trapping may have caught a greater number of emerging adults (thereby decreasing the estimated mortality rate) due to the traps being left in place through out the entire sampling period. Sixty five percent of emerging T. catenata adults caught were in the period 31 (± 7) days after the emergence traps were placed and sealed down. This time span is