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. Morphology, Phylogeography and Drumming Behaviour of a New Zealand Ground Weta, Hemiandrus pallitarsis. A thesis presented in partial fulfillment of the requirements for the degree of Master of Science in Conservation Biology at Massey University, Palmerston North, New Zealand Esta Monique Chappell 2008 ii Abstract Species are one of the fundamental components of biology and the accurate delimitation of species is important in evolutionary, systematic and ecological studies, yet there is still confusion over how species can be recognised. Examining different characters allows multiple lines of evidence for successful and accurate species delimitation and identification. In this thesis, morphological, genetic and behavioural variation is investigated within an endemic species of ground weta, Hemiandrus pallitarsis , in the North Island, New Zealand. Twelve morphological characters were measured, and mitochondrial cytochrome oxidase I DNA sequences were analysed from populations across the distributional range of H. pallitarsis. Both methods provide no evidence of a species complex within H. pallitarsis . Instead, the morphometric results suggest females are significantly larger than males, and ground weta in Palmerston North are significantly smaller than weta further north. Additionally, genetic analyses found substantial population structuring, large genetic distances, and an historical south to north pattern of movement in the North Island. The pattern of vibratory drumming behaviour followed that predicted by morphology and geographic proximity – drumming signals were more similar between geographically close populations and did not match the patterns of genetic isolation. Overall, this thesis was able to show that H. pallitarsis is morphologically, genetically and behaviourally variable across the North Island. iii Acknowledgements I would like thank my supervisors Dr Steve Trewick and Dr Mary Morgan- Richards for their constant enthusiasm and effortless and endless support throughout this project. Thank you also for funding the genetic sequencing, without which my life would have been that much more difficult. Thank you to my co - supervisor, Dr Isabel Castro, for comments and advice on the weta drumming behaviour. I would like to acknowledge Dr Ian Stringer for giving me the original idea to study ground weta and also for taking me on many Middle Island tusked weta trips. Thank you to Dr Darryl Gwynne for his encouragement and exuberant enthusiasm for weta and their sexual behaviour. To al l those people who help me in the field: the “Kelso Lane weta hunters”, Rob Chappell , Lorraine Cook, Kaye Rabarts, Susan Cunningham, Jonathan Tonkin, and Louisa Robertson, thank you all so much. For help in the lab, thank you to everyone in the Allan Wils on Centre, especially to Renae Pratt, Simon Hills, Gillian Gibb and Carlos Lehnebach. I would like to acknowledge John Early at the Auckland Museum for providing weta specimens, and the Entomological Society of New Zealand and the Orthopterist’s Society f or their financial contributions to this project. A big thank you goes to everyone in the Ecology building and the B.E.E.R.S group. Finally and most importantly, thank you to all my family and friends for your love and support. In particular, thanks to Dad for taking me out to some wonderful and exiting islands and helping me whenever I needed it, I hope this thesis makes you proud, and thanks to Mum and Pete for providing a fantastic weta population in their back garden and for the many hours of encouragement, love and support, couldn’t have done it without you. iv CONTENTS Abstract .............................................................................................................................. ii Acknowledgements ........................................................................................................... iii List of Tables ................................................................................................................... vii List of Figures ................................................................................................................... ix Chapter 1: Thesis Introduction....................................................................................... 1 1.1 Species concepts ................................................................................... 1 1.2 New Zealand ground weta ................................................................... 2 1.3 Taxonomic history and systematics of Hemiandrus pallitarsis ........... 3 1.4 Mating behaviour of Hemiandrus pallitarsis ....................................... 4 1.5 Thesis outline and chapter aims ........................................................... 5 1.6 References ............................................................................................ 8 Chapter 2: Morphometric variation in Hemiandrus pallitarsis 2.1 Introduction .................................................................................................... 13 2.2 Methods .......................................................................................................... 16 2.2.1 Sampling sites and collections ........................................................ 16 2.2.2 Body measurements ........................................................................ 17 2.2.3 Measurement error .......................................................................... 20 2.2.4 Statistical analyses .......................................................................... 20 2.3 Results ............................................................................................................ 22 2.3.1 Measurement error .......................................................................... 22 2.3.2 Sexual size dimorphism .................................................................. 22 2.3.3 Morphometric variation among populations ................................... 23 2.3.4 Population classification ................................................................. 27 v 2.3.5 Variation in body size with latitude ................................................ 28 2.3.6 Variation in accessory organ shape ................................................. 31 2.4 Di scussion ...................................................................................................... 32 2.4.1 Sexual size dimorphism .................................................................. 32 2.4.2 Geographical morphometric variation ............................................ 33 2.5 References ...................................................................................................... 36 Chapter 3: Phylogeography of a ground weta, Hemiandrus pallitarsis 3.1 Introduction .................................................................................................... 42 3.1.1 North Island Plio- Pleistocene geophysical history ......................... 44 3.1.2 Objectives ....................................................................................... 50 3.2 Methods .......................................................................................................... 52 3.2.1 Taxonomy of ground weta .............................................................. 52 3.2.2 Sampling sites and collections ........................................................ 52 3.2.3 DNA extraction ............................................................................... 55 3.2.4 PCR amplification ........................................................................... 56 3.2.5 DNA sequencing ............................................................................. 57 3.2.6 Nuclear copies ................................................................................. 58 3.2.7 Phylogenetic analyses ..................................................................... 59 3.2.8 Population genetic analyses ............................................................ 60 3.2.9 Molecular dating ............................................................................. 62 3.3 Results ............................................................................................................ 63 3.3.1 Phylogenetic analysis ...................................................................... 63 3.3.2 Population genetic analysis ............................................................. 68 3.3.3 Isolation - by- distance ....................................................................... 72 3.3.4 Genetic distances and molecular clock ........................................... 73 3.4 Discussion ...................................................................................................... 75 vi 3.4.1 Phylogenetics .................................................................................. 75 3.4.2 Genetic distances ............................................................................. 75 3.4.3 Population structure and gene flow ................................................. 76 3.4.4 Phylogeography .............................................................................. 77 3.4.5 Molecular clock and its caveats ...................................................... 79 3.4.6 DNA barcoding and morphology .................................................... 80 3.5 References ...................................................................................................... 81 Chapter 4: Vibratory drumming variation in three populations of Hemiandrus pallitarsis 4.1 Introduction .................................................................................................... 93 4.1.1 Vibrational communication ............................................................. 93 4.1.2 Species recognition sig nals ............................................................. 94 4.1.3 Factors shaping the evolution of sexual signals .............................. 95 4.1.4 The New Zealand ground weta: Hemiandrus pallitarsis ................ 97 4.1.5 Hypotheses ...................................................................................... 98 4.2 Methods .......................................................................................................... 99 4.2.1 Study populations ............................................................................ 99 4.2.2 Recording techniques and drumming measurements .................... 100 4.2.3 Statistical analy ses ........................................................................ 102 4.3 Results .......................................................................................................... 104 4.4 Discussion .................................................................................................... 109 4.5 References .................................................................................................... 113 Thesis Conclusions ........................................................................................................ 121 References….. ................................................................................................................ 123 Appendixes…. ................................................................................................................ 124 vii List of Tables Table 2.1. Means and standard deviations (SD ) for six morphometric traits of adult male and female H. pallitarsis …………………………………………………...22 Table 2.2. Measurements of adult male H. pallitarsis from ten populations before log transformation. All measurements are in millimetres (mm). The statistical significance of differences among populations has been assessed using one- way ANOVA……………… …………………………………………………………..25 Table 2.3. Measurements of adult female H. pallitarsis from nine populations before log transformation. All measurements are in millimetres (mm). The statistical significance of differences among populations has been assessed using one- way ANOVA……………… …………………………………………………………..26 Table 2.4. Percentage correct a posteriori classification of populations based on the eight morphometric variables for male H. pallitarsis.............................................27 Table 2.5. Percentage correct a posteriori classification of populations based on the eight morphometric variables for female H. pallitarsis................................................................................................................28 Table 2.6. Correlation coefficients (linear regression) between morphometric characters and latitude (Lat) in males (including and excluding the Wellington population) and females of H. pallitarsis ………………………………………...29 Table 2.7. Pearson correlation r of length of the female accessory organ (LAO) with length of pronotum, metafemur an d body length………………………………...31 Table. 3.1. Sampling locations for Hemiandrus specimens. * Individuals used in Bayesian analysis of the entire COI gene fragment. † Locations above sea level viii during the early Pliocene (Fleming, 1979; Lewis & Carter, 1994) (see Fig. 3.7)..........................................................................................................................53 Table 3.2. Collection locations, number of individuals collected, molecular indices and haplotype codes. * remove d from population- level analyses due to sample size of one…………………………...……………………………………………….......69 Table 3.3. Pairwise ΦST values among populations of Hemiandrus pallitarsis. Comparisons that were significant at the P < 0.05 level are indicated by an asterisk....................................................................................................................70 Table 3.4. Analysis of molecular variance (AMOVA) using ARLEQU IN (Excoffier et al. 2005) among ground weta from 13 populations in NI, New Zealand...................................................................................................................71 Table 4.1. Mean values (±SE) for the drumming bout components of male ground weta H. pallitarsis from three different populations. Differences between populations and changes in drumming measurement with temperature shown on the right. * p ≤ 0.05; ns p > 0. 05…….………………………………………………………...108 ix List of Figures Figure 2.1. Collection locations of Hemiandrus pallitarsis in North Island, New Zealand. Inset shows collection locations on the Coromandel Peninsula………………………………………………………………….………17 Figure 2.2a-h. Morphometric characters measured for Hemiandrus pallitarsis specimens. Abbreviations as follows: metafemur (Mf), length subgenital plate (Sg), maximum width subgenital plate (Mx Sg), minimum width subgenital plate (MnSg), length of r ight (LAO1) and left (LAO2) tine of the accessory organ fork, width of accessory organ (WAO), head length (Hl), head width (Hw), fastigium width (Fa), distance between third tergite and anterior fastigium (Bl), length pronotum (Pr)…………………………………………………………………….19 Figure 2.3. Principal component analysis on eight morphological traits of adult male H. pallitarsis. Percent of total variation marked on each axis....................................24 Figure 2.4. Principal component analysis on eight morphological traits of adult female H. pallitarsis. Percent of total variation marked on each axis. ..............................24 Figure 2.5. Relationships between the means of morphometric characters and latitude in populations of H. pallitarsis from North Is land, New Zealand. Females, ●; males, ○. * indicates characters significantly correlated with latitude in female populations (Pearson r; p < 0.05). All regressions are linear. Latitude is expressed in minutes from Great Barrier Island locality (0 ′) as this is the northern most location………………………………………………………………...…………30 Figure 2.6. Images of dissected female accessory organs showing variability in shape from Cuvier Island, Middle Island, Taiharuru, Coromandel Town and Palmerston North…………………………………………………………………………...…31 x Figure 3.1. Palaeogeographic maps of the lower North Island (outline), showing land above sea level (grey) in the early Pliocene (A), end of the Pliocene (B), and in the early- mid Pleistocene (C) (modified from Lewis & Carter, 1994)……...… …….45 Figure 3.2. Map showing glaciers and vegetation zones in New Zealand at the Last Glacial Maximum (modified from Barrell et al., 2005)………….………………47 Figure 3.3. (A) Map showing the area of North Island covered by pyroclastic rock (dark grey) and as h (light grey) following the Taupo eruption ~1850 years ago. (B) Approximate numbers of endemic vascular plants according to Wardle 1963, and percentage of endemic insect species from Craw, 1988…………..……………...49 Figure 3.4. Collection locations of Hemiandrus pallitarsis ( ●), and outgroup species ( ■) in New Zealand. Inset shows collection localities in the Coromandel…………………………………………………………………..…..55 Figure 3.5. Diagram of the cytochrome oxidase I gene showing positions and sequencing directions of the primers used in this study……….…………… ……56 Figure 3.6. Neighbour Joining (NJ ) tree with H. bilobatus as outgroup. Symbols indicate alternative mtDNA COI sequences from the same individuals: * Cuvier Island,  Middle Island. Arrow shows position of sequence d Numts ……………………………………………………………………...……...59 Figure 3.7. Map showing hypothesized distribution of land in the North Island at the early Pliocene (grey areas) (modified from Morgan - Richards et al., 2001). Sampling locations within each region are indicate d by white circles………… ……… ………………………………………………...………...61 Figure 3.8. Map of NI, New Zealand showing populations sampled at three geographical scales: A) Coromandel Peninsula and offshore islands (dark grey); B) Taupo, Lake Waikaremoana and populations in A (medium grey); C) Taranaki, Palmerston North, Wellington and populations in A and B (light grey)…………………… ……………………………………………………..…..62 xi Figure 3.9. Bayesian consensus network showing the evolutionary relationships among short- ovipositor ground weta of New Z ealand: Hemiandrus pallitarsis (large grey box), H. bilobatus, H. “onokis”, H. “horomaka”, H. “promontorius” and H. “vicinus”. Bayesian analysis based on 1432 bp of the COI mitochondrial gene, implemented with MrBayes v3.1.2. (Ronquist & Huelsenbeck, 20 05) using HKY (nst=2) model, consisting of 6 million generations with two independent, simultaneous runs with three heated chains and a sample frequency of 1000 generations, and a 10% burn in. Node stability is indicated by posterior probabilities……………………………...……………………………………….64 Figure 3.10. Maximum likelihood tree for Hemiandrus pallitarsis based on 638 bp of the COI mitochondrial gene. Tree is midpoint rooted due to long branch lengths of outgroup species. An arrow indicates position of root in analyses that include outgroup Hemiandrus spp. Modeltest v3.06 selected TVM + I + G model ( - ln L = 3057.3407), with a gamma distribution shape parameter of 0.5777, 0.5548 proportion of invariable sites, and base frequencies of A = 0.26410, C = 0.25660, G = 0.16600, and T = 0.31330. Three measures of branch support are indicated with 1000 NJ and MP bootstrap replicates shown above edges, respectively, and 100 ML bootstrap replicates shown below edges. Terminal labels indicate sampling location and specimen code…………… ……………...……………….66 Figure 3.11. Unrooted consensus network for Hemiandrus pallitarsis COI mtDNA sequences. Bayesian analysis implemented with MrBayes v3.1.2. (Ronquist & Huelsenbeck, 2005 ) using HKY (nst=2) model, consisting of 6 million generations with two independent, simultaneous runs with three heated chains and a sample frequency of 1000 generations, and a 20% burn in. Values on branches are Bayesian credibility values. Terminal labels indicate individual codes and sampling locations are indicated by ellipses……………………………………...67 Figure 3.12. Isolation - by- distance plots for (A) Little Barrier Island, the Coromandel Peninsula and surrounding islands - Z score = 13648, R 2 = 0.137551, r = - 0.370878, p = > 0.05, (B) the above plus Taupo and Lake Wai karemoana - Z score = 5040, R 2 = 0.062945, r = 0.250889, p = > 0.05, and (C) the above plus Taranaki, xii Palmerston North and Wellington - Z score = 126 9, R 2 = 0.036622, r = 0.191369, p = > 0.05 (see Fig . 3.8)………………………………………………..…………72 Figure 3.13. Frequenc y distribution of pairwise uncorrected p genetic distances from COI mtDNA sequences from four Hemiandrus species, showing pairwise distances between (dark grey bars) and within species…………………… …..…74 Figure 4.1. Trees illustrate predicted patterns of drumming behaviour in Hemiandrus pallitarsis: a) Morphometric/geographic hypothesis – smaller weta from geographically proximate populations will show behavioural similarities; b) Genetic hypothesis – genetically similar populations (Coromandel and Palmerston North, north of the river) will have drumming patterns more similar to each other than to Palmerston North, south of the river…………...………………………...98 Figure 4.2. Sampling and recording locations of Hemiandrus pallitarsis, in North Island, New Zealand…………………………………………………...…………99 Figure 4.3. The recording setup showing a male weta inside a plastic container sitting on top of the Piezo film (beneath bla ck electrical tape)…………………… …...103 Figure 4.4. Oscillograph of a single drumming bout produced by a H. pallitarsis male showing measurements taken: top trace shows bout length, inter - pulse interval and number of pulses; bottom trace shows five pulses from within the drumming bout, indicating the number of oscillations within each pulse……………...................101 Figure 4.5. Scatter plots using the raw data not adjusted for temperature, for each drumming component measured at each of the three populations. Symbols indicate individual weta; identical symbols indicate multiple drumming bouts recorded for a single individual………………….……………………………...106 Figure 4.6. Morphometric/geographic hypothesis. - smaller weta from geographically proximate populations show more simila r drumming patterns…………..……..109 Chapter 1: Thesis Introduction Anterior view of an adult Hemiandrus pallitarsis found on a gravel track at Stony Bay, Coromandel Photo: E.Chappell 1 Chapter 1: Thesis Introduction 1.1 Species concepts Since before Darwin, there has been debate and disagreement concerning the meaning of the word ‘species’. Darwin himse lf recognised the fact that there were multiple definitions used by naturalists, but that everyone had the same general concept of what a species was (Darwin, 18 59). Over the last 60 years, there has been a plethora of species concepts formulated, such as the Biological species concept (Mayr, 1942), the Recognition species concept (Paterson, 1985), the Ecological species concept (Van Valen, 1976), and the Evolutionary species concept (Simpson, 1951), to name just a few. In fact, there are at least 24 diff erent named species concepts (Mayden, 1997). Much of the argument surrounding the debate has been over theoretical concepts versus methodological ones, that is, concepts with specific species identification criteria. Nevertheless, De Queiroz (2007) sugge sts there is a common element among all contemporary concepts and proposes a unified species concept in which species are “…separately evolving metapopulation lineages, or more specifically, segments of such lineages” (p. 880- 881 ). This concept, De Queiroz points out, separates the issue of species concept from species delimitation, where boundaries and numbers of species are inferred. Therefore, the general consensus seems to be that members of a species have an evolutionary history more similar to members in the same species than they do with other species (Hey, 2006 ). Despite the controversy surrounding species concepts, some authors have separated species identification and delimitation from the more conceptual aspect and have focused solely on identification (operational) criteria (e.g. Sites & Marshall, 2003; Winston, 1999). However, the operational criteria used in species delimitation have also come under scrutiny (Wiens & Servedio, 2000). DNA sequence data is now commonly used to infer species boundaries (Besansky et al. , 2003; Puorto et al. , 2001 ), although most species continue to be identified based on morphological descriptions. Indeed, several authors have stressed the importance of examining multiple different characters in order to gain evidence for the existence of separate species (e.g. De Queiroz, 2007; Roe & Sperling, 2007). The concordance of characters (e.g. Chapter 1: Introduction 2 morphological, genetic, behavioural) can form crucial evidence on which species status decisions are made. Accordingly, this t hesis sets out to document the degree of morphological, genetic and behavioural variation within an endemic speices of ground weta, Hemiandrus pallitarsis , in NI, New Zealand. 1.2 New Zealand ground weta The Southern Hemispheric group of ensiferan Orthoptera, which form a major group in New Zealand, are the weta, or king crickets. Previous taxonomic classification placed all New Zealand weta into the family Stenopelmatidae (Hutton, 1896), which was divided into two sub- families, Anostominae (with pads on lower surface of tarsi) and Dolichopodinae (without any pads). Johns (19 97), however, raised the sub- family Anostominae to full family status – Anostostomatidae, and suggested weta in this family could be distinguished from Stenopelmatidae based on fea tures of the fastigium, coxae, foretibia, metasternum and hindfemur. There are four groups of Anostostomatid weta in New Zealand: ground weta ( Hemiandrus ), tree weta ( Hemideina ), giant weta ( Deinacrida), and tusked weta ( Motuweta, Anisoura); of these four groups, ground weta are the most speciose. Ground weta are nocturnal flightless insects, ranging in size from 12 to 45 mm long. During the day they seal themselves in burrows in the ground, and at night they feed primarily on other invertebrates, but als o eat fruit and seeds. Like cave weta (Orthoptera: Rhaphidophoridae), they do not possess tibial tympanal organs and so presumably are deaf to airborne sound, they do however have abdominal stridulatory structures, usually reduced to conical pegs or small rounded knobs on the first two or three tergites (Field, 20 01). Ground weta inhabit a wide variety of environments including urban gardens, grasslands, shrublands and forests, and are found throughout New Zealand, from the subantarctic islands to Northla nd. There are currently nine named and more than 28 undescribed Hemiandrus species in New Zealand and several species in Australia (Jewell, 2007; Johns, 1997, 2001). The species of interest in this study was Hemiandrus pallitarsis , a ground weta with a characteristically short ovipositor and modified sixth abdominal sternite. There are seven species of short ovipositor weta in New Zealand (two described and five undescribed) (Johns, 2001). The majority of these species are found only in the South Chapter 1: Introduction 3 Island: Hemiandrus “turgidulus” is found in the Cromwell river sand dunes, an area protected for the Cromwell chafer beetle ( Prondontria lewisii ); H. “horomaka” is endemic to Banks Peninsula; H. “onokis” occurs throughout the North Canterbury area; H. “promontori us” is found only at Cape Campbell and Marfell’s Beach; and H. “vicinus” is found in the Marlborough Sounds area. There are two short ovipositor species in the NI which are sympatric: H. bilobatus occurs in the Wellington region as far north as Paraparamu and is found on Mana Island, Matiu/Somes Island and the Brothers Island on the south side of the Cook Strait; and H. pallitarsis is found from Wellington north to Auckland, and has been recorded on several offshore islands – Poor Knights Islands, Great Ba rrier Island and Middle Island in the Mercury group (Johns, 2001). Three long ovipositor species also occur in the NI, H. maculifrons, H. “elegans” and H. “otekauri”, the latter two species are only found in northern NI. 1.3 Taxonomic history and systematics of Hemiandrus pallitarsis The taxonomic history of Hemiandrus pallitarsis has been turbulent, due in part to poor species descriptions and confusion over early species names. Walker (18 69), described two species: Ceuthophilus (? ) lanceolatus from a male weta collected by H. Drew Esq. half a mile into a cave; and Libanasa pallitarsis from a male, the collection locality unknown. In 1896, Hutton synonymised C. (? ) lanceolatus with a cave weta species, Macropathus edwardsii, as it was supposedly collected from the same location as three of Walker’s other Macropathus species. However, the location and collection information of the C. (? ) lanceolatus holotype was likely confused. Kirby (1906), re - examined Walker’s type specimens and placed C. (? ) lanceolatus into the genus Onosandrus Ståll, 1876, and also synonymised Libanasa pallitarsis under the new combination Onosandrus (? ) lanceolatus Walker, 1869. Later Hutton (1899), synonymised O. (? ) lanceolatus with O. pallitarsis. So at this stage in the taxonomy, both C. (? ) lanceolatus and L. pallitarsis appear to be synonymous with O. pallitarsis. In 1938, Ander described a new genus and three new species in New Zealand; Hemiandrus furcifer, the type species for the genus which was collected from Palmerston North, H. similis collected in Wellington, and H. bilobatus also collected in Wellington . Following on from Ander, Salmon (1950) described nine new species and introduced a new genus, Zealandosandrus based on ovipositor length (fully developed Chapter 1: Introduction 4 in Zealandosandrus and extremely short in Hemiandrus ). In his revision of the New Zealand wetas, Salmon misidentified specimens and synonymised O. pallitarsis with Z. maculifrons on the basis of the shape of the subgenital plate. It was not until 1961, when Ramsay re - examined Walker’s type specimen of Ceuthophilus (? ) lanceolatus that he found it was actually an immature female (not a male), and possessed a very short ovipositor. He transferred C. (? ) lanceolatus into the genus Hemiandrus Ander, 1938, and so it became H. lanceolatus (Walker, 1869). Ramsay observed that H. Drew Esq., the original collector of the holotype, did most of his collecting in the Wanganui area so C. (? ) lanceolatus is possibly a conspecific of H. furcifer Ander, 1938. More recently , Johns (1997), placed all species into Hemiandrus based on initial phylogenetic research (Gerber, 1996 ) and a broader perspective on the systematics of the Anostostomatidae. Johns (1997), also synonymised H. furcifer with the new combination H. pallitarsis , and later suggested that H. lanceolatus may in fact be a doubtful name (Johns, 2001). 1.4 Mating behaviour of Hemiandrus pallitarsis The reproductive behaviour of Hemiandrus pallitarsis involves complex signals, interesting reproductive structures, and paternal and maternal investment (Gwynne, 2002, 2004, 2005). Vibratory drumming: Although it has been suggested that pheromones may be involved in long distance communication, vibratory signals appear to be used for local mate attraction (Gwynne, 2005). Males drum their abdomens on the substrate, usually a plant leaf or stem, and females drum in apparent response to the male. Several other short ovipositor species also exhibit vibratory signalling – H. bilobatus (D.T., Gwynne, pers. comm. E.M., Chappell, pers. obs.) , H. “promontorius”, H. “onokis”, and H. “vicinus” (Gwynne, 2005), and it has also been found in two long ovipositor species – H. subantarcticus (Butts, 1983), and H. maculifrons (Cary, 1981). Vibratory drumming in H. pallitarsis appears to be a pre- copulatory mechanism and does not occur after mating. Chapter 1: Introduction 5 Copulation: Copulation involves the production and transfer of a spermatophore from male to female. Like many Orthopterans, the spermatophore of H. pallitarsis consists of a sperm ampulla (containing the sperm) and a sperm - free spermatophylax (food gift) that the female feeds on while the sperm ampulla is expelled into her reproductive tract. However, unlike many Orthopterans, the two parts of the spermatophore are separate entities in H. pallitarsis (Gwynne, 2002 ). Males attach first to an accessory organ located on the ventral side of the female abdomen. This accessory organ is an elaborate elbowed structure, a modification of the sixth sternite. Whilst attached to this structure , the male then arches back and using the dorsal parts of his genitalia, attaches to the female’s primary genitalia where he delivers the sperm ampulla. Once transferred, he then detaches from this area, remains attached to her accessory organ, and delive rs the spermatophylax to her mid abdominal area (Gwynne, 2002, 2005). It has been suggested that this organ may be under sexual selection to acquire thesis nuptial gifts from males (Gwynne, 20 05). After the male has separated completely from the female, she then promptly reaches under and grasps the food gift and proceeds to spend the next hour or so feeding upon it. During this time the male exhibits a mate guarding behaviour and remains close to the female, seemingly to prevent other males copulating with the her (Gwynne, 2005). H. pallitarsis also shows maternal care. Females lay their eggs into the bottom of their burrow and apparently stay with the eggs and nymphs for an extended period of time (Gwynne, 2005). 1.5 Thesis outline and chapter aims The overall aim of this thesis is to examine intraspecific variation in H. pallitarsis from a broad perspective, encompassing morphological, genetic and behavioural characters. Chapter two sets out to document the morphological diversity across the entire range of H. pallitarsis. At present, there is some suggestion of regional morphological variation within this species (Johns, 2001), and a previous study found the size of the female accessory organ varied between geographically proximate populations (Gwynne, 2005 ). Morphometric variation across a landscape can be correlated with geographical differences in temperature (Atkinson, 1994), latitudinal or altitudinal differences (De Oliveira et al., 2004), or as a result of sexual selection (Fairbairn, 1997 ). The suggested Chapter 1: Introduction 6 morphological variation, and the potential for sexual selection on the female accessory organ in H. pallitarsis , suggests this is an interesting species in which to study morphological variation. This chapter specifically focuses on the morphometric variability, sexual size dimorphism and variation in the female accessory organ across a range of locations. Chapter three focuses on the genetic variation and phylogeographic structure within H. pallitarsis populations, and how past geophy sical events may have impacted current phylogeographic patterns. On the whole, there are few phylogeographical studies that have explicitly examined how patterns of genetic divergence within and among species correspond to Plio- Pleistocene geological and/ or climatic changes in NI, New Zealand (Apte et al., 2007; Morgan - Richards et al. , 2001; Stevens & Hogg, 2004). The majority of research has focused on the effects of mountain building and glaciation in the South Island, however important changes in the N I landscape have also occurred. For example, marine transgression in the lower NI during the Pliocene, habitat shifts during the glacial- interglacial cycles in the Pleistocene, and more recent volcanic activity in central NI. This chapter tested several hypotheses related to the geophysical history of the NI: 1) H. pallitarsis would have been restricted to the northern NI, then dispersed south as land emerged above sea level during the Plio- Pleistocene in what is now the lower NI ; or 2) H. pallitarsis originated in the northern SI and range expansion has occurred in a south to north pattern following land uplift in the Pliocene; or 3) Pliocene impacts on the phylogeographic structure in H. pallitarsis has been masked by more recent climatic events and/or v olcanic activity. These predictions were tested by analysing mitochondrial DNA cytochrome c oxidase subunit I (COI) sequence fragments from populations across the distributional range of H. pallitarsis. Finally, the last chapter in this thesis investigate d the variation in male vibratory signals between three populations of H. pallitarsis – Coromandel and two locations in Palmerston North, one north and one south of the Manawatu River . Geographical variation in sexual communication systems has been explai ned by environmental factors, such as temperature (Gerhardt & Huber, 2002; Toms et al. , 1993; von Helverson & von Helverson, 1994; Weissman, 2001), and social factors, such as sexual selection (Orci, 2007; Wilczynski & Ryan, 1999). The specific aim of thi s chapter was to test two hypotheses: 1) the two Palmerston North populations would exhibit more similar drumming behaviour patterns compared to the Coromandel population due to similarity in morphometrics and geographic proximity, or 2) drumming patterns on the Chapter 1: Introduction 7 northern side of the Manawatu River would be more similar to those in Coromandel based on genetic similarities. Collectively, these three chapters aim to give an insight into the degree of intraspecific variation that can be found in a single species. At the same time some interesting biological, behavioural and historical phylogeographic patterns are found. Chapter 1: Introduction 8 1.6 References Apte, S., Smith, P. J. and Wallis, G. P. (2007) Mitochondrial phylogeography of New Zealand freshwater crayfishes, Paranephrops spp. Molecular Ecology, 16: 1897- 1908. Atkinson, D. (1994) Temperature and organism size - a biological role for ectotherms? . Advances in Ecological Research, 25: 1- 58. Barrell, D. J. A., Alloway, B. V., Shulmeister, J. and Newnham, R. M. (eds) (2005) Towards a climate event stratigraphy for New Zealand over the past 30,000 years. Institute of Geological and Nuclear Sciences, Science Report, SR2005/ 07 (includes poster). Besansky, N. J., Krzywinski, J., Lehmann, T., Simard, F., Kern, M., Mukabayire, O., Fontenille, D., Toure, Y. and Sagnon, N. F. (2003) Semipermeable species boundaries between Anopheles gambiae and Anopheles arabiensis : evidence from multilocus DNA sequence variation. Proceedings of the National Academy of Sciences, 100: 10818- 10 823. Bu tts, C. A. (1983) The biologies of two species of weta endemic to the Snares island Zealandosandrus subantarcticus Salmon (Orthoptera: Stenopelmatidae) and Insulanoplectron spinosum Richards (Orthoptera: Rhaphidophoridae). B.Sc. Zoology Thesis, University of Canterbury, Christchurch, New Zealand. Cary, P. R. L. (1981) The biology of the weta Zealandosandrus gracilis (Orthoptera: Stenopelmatidae) from the Cass region. M.Sc. Thesis, University of Canterbury, Christchurch, New Zealand. Craw, R. (1988) Continuing the synthesis between panbiogeography, phylogenetic systematics and geology as illustrated by empirical studies on the biogeography of New Zealand and the Chatham Islands. Systematic Zoology, 37: 291- 310. Chapter 1: Introduction 9 Darwin, C. (1859) On the origin of species by m eans of natural selection. Murray, London. De Oliveira, C. M., Lopes, J. R. S., Dias, C. T. D. S. and Nault, L. R. (2004) Influence of latitude and elevation on polymorphism among populations of the corn leafhopper, Dalbulus maidis (DeLong and Wolcott) (He miptera: Cicadellidae), in Brazil. Environmental Entomology, 33: 1192- 1199. De Queiroz, K. (2007) Species concepts and species delimitation. Systematic Biology, 56: 879- 886. Fairbairn, D. J. (1997) Allometry for sexual size dimorphism: patterns and process es in the coevolution of body size in males and females. Annual Review of Ecological Systems, 28: 659- 687. Field, L. H. (2001 ) Stridulatory mechanisms and associated behaviour in New Zealand wetas. In: The biology of wetas, king crickets and their allies ( ed. L. H. Field). CABI Publishing, Wallingford. Gerber, A. S. (1996) Phylogenetics of New Zealand ground wetas (conference abstract). Phylogeny of Life Symposium: Research Training Group in the Analysis of Biological Diversification, University of Arizona. Gerhardt, H. C. and Huber, F. (2002) Acoustic communication in insects and anurans. Common problems and diverse solutions. University of Chicago Press, Chicago, IL. Gwynne, D. T. (2002) A secondary copulatory structure in a female insect: a clasp for a nuptial meal? Naturwissenschaften, 89: 125- 127. Gwynne, D. T. (200 4) Reproductive behaviour of ground weta (Orthoptera: Anostostomatidae): drumming behaviour, nuptial feeding, post - copulatory guarding and maternal care. Journal of the Kansas Entomological Society, 77: 414- 428. Chapter 1: Introduction 10 Gwynne, D. T. (2005) The secondary copulatory organ in female ground weta Hemiandrus pallitarsis , Orthoptera: Anostostomatidae): a sexually selected device in females? Biological Journal of the Linnaean Society, 85: 463- 46 9. Hey, J. (2006) On the failure of modern species concepts. Trends in Ecology & Evolution, 21: 447- 450. Hutton, F. W. (1896) The New Zealand Stenopelmatidae. Transactions and Proceedings of the New Zealand Institute, 29: 208- 244. Hutton, F. W. (1899 ) Supplement to the Stenopelmatidae of New Zealand. Transactions of the New Zealand Institute, 31: 40- 43. Jewell, T. (2007) Two new species of Hemiandrus (Orthoptera: Anostostomatidae) from Fiordland National Park, New Zealand. Zootaxa, 1542: 49- 57. Johns, P. M. (1997) The G ondwanaland weta: family Anostostomatidae (formerly in Stenopelmatidae, Henicidae or Mimnermidae): nomenclatural problems, world checklist, new genera and species. Journal of Orthoptera Research, 6: 125 - 138. Johns, P. M. (2001) Distribution and conservation status of ground weta, Hemiandrus species (Orthoptera: Anostostomatidae). Science for Conservation Report 180. Department of Conservation, Wellington, New Zealand. Kirby, W. F. (1906 ) A synonymic catalogue of Orthoptera. II. Orthoptera Saltatoria: I. ( Achetidae et Phasgonuridae). British Museum (Natural History), London. Mayden, R. L. (1997) A hierarchy of species concepts: the denouement in the saga of the species problem. In: Species: The units of biodiversity (eds. M. F. Claridge, H. A. Dawah and M. R . Wilson), pp. 381- 424. Chapman and Hall, London. Mayr, E. (1942) Systematics and the Origin of Species. Belknap Press. Morgan- Richards, M., Trewick, S. A. and Wallis, G. P. (2001) Chromosome races with Pliocene origins: evidence from mtDNA. Heredity, 86: 303- 312. Chapter 1: Introduction 11 Orci, K. M. (2007 ) Female preferences for male song characters in the bush- cricket Isophya camptoxypha ( Orthoptera, Tettigonioidea). Journal of Insect Behaviour, 20: 503- 513. Paterson, H. E. H. (1985 ) The recognition concept of species. In: Specie s and speciation (ed. E. S. Vrba), pp. 21- 29. Transvaal Museum Monograph No. 4, Pretoria: Transvall Museum. Puorto, G., Salomao, M. D., Theakston, R. D. G., Thorpe, R. S. and Warrell, D. A. (2001) Combining mitochondrial DNA sequences and morphological dat a to infer species boundaries: phylogeography of lanceheaded pitvipers in the Brazilian Atlantic forest, and the status of Bothrops pradoi (S quamata: Serpentes: Viperidae). Journal of Evolutionary Biology, 14: 527- 538. Roe, A. D. and Sperling, F. A. H. (2007) Patterns of evolution of mitochondrial cytochrome c oxidase I and II DNA and implications for DNA barcoding. Molecular Phylogenetics and Evolution, 44: 325- 345. Salmon, J. T. (1950) Revision of the New Zealand wetas - Anostostomainae (Orthoptera: Steno pelmatidae). Dominion Museum Records in Entomology, 1: 121- 177. Simpson, G. G. (1951) The species concept. Evolution, 5: 285- 298. Sites, J. and Marshall, J. C. (2003 ) Delimiting species: a Renaissance issue in systematic biology. Trends in Ecology & Evolution, 18: 462- 470. Stevens, M. I. and Hogg, I. D. (2004) Population genetic structure of New Zealand's endemic corophiid amphipods: evidence for allopatric speciation. Biological Journal of the Linnean Society, 81: 119- 133. Toms, R. B., Ferguson, J. W. H. a nd Becker, S. (1993) Relationship between body temperature in stridulating male crickets (Orthoptera, Gryllidae). South African Journal of Zoology, 28: 71- 73. Van Valen, L. (1976) Ecological species, multispecies, and oaks. Taxon, 25: 233- 239. Chapter 1: Introduction 12 von Helverson, O. and von Helverson, D. (1994) Forces driving coevolution of song and song recognition in grasshoppers. In: Neural Basis of Behavioural Adaptations, Progress in Zoology (eds. K. Schildberger and N. Elsner), pp. 253 - 284. Gustav Fischer Verlag, Stuttgart. Walker, F. (1869) Catalogue of the specimens of Dermaptera Saltatoria and supplement to the Blattariae in the collection of the British Museum. I. British Museum (Natural History), London. Wardle, P. (1963) Evolution and distribution of the New Zealand f lora, as affected by Quartenary climates. New Zealand Journal of Botany, 1: 3- 17. Weissman, D. B. (2001) Communication and reproductive behaviour in North American Jerusalem Crickets ( Stenopelmatus ) (Orthoptera: Stenopelmatidae). In: The biology of wetas, king crickets and their allies (ed. L. H. Field), pp. 351–373. CABI, Wallingford, UK. Wiens, J. J. and Servedio, M. R. (2000) Species delimitation in systematics: Inferring diagnostic differences between species. Proceedings: Biological Sciences, 267: 631- 636. Wilczynski, W. and Ryan, M. J. (1999) Geographic variation in animal communication systems. In: Geographic variation in behaviour: perspectives on evolutionary mechanisms (eds. S. A. Foster and J. A. Endler), pp. 234- 261. Oxford University Press, New York. Winston, J. (1999) Describing Species. Columbia University Press, New York. Chapter 2: Morphometric Variation Photos: E.Chappell Dissected female Hemiandrus pallitarsis accessory organs showing morphological variation 13 Chapter 2: Morphometric variation in Hemiandrus pallitarsis 2.1 Introduction When we look at an organism, the first thing we tend to notice is morphology, that is, colour, size and shape. Traditionally, species were described as ‘kinds’ or ‘types’ based on morphological differences, a concept commonly referred to as the morphological species concept in which “Population A … was determined to be a different species from population B … if it was deemed to be sufficiently different from it by morphological characters.” (Mayr, 1996 , p. 269). Additionally, Darwin’s morphological species concept (1859) described species as “varieties” between which there are no or few morphological intermediates. Although in some cases, changes in morphology, such as body size, can lead to changes in performance, and can result in evolutionary novelties (Koehl, 1996). The morphological species concept can be misleading as individuals, populations and species are often morphologically variable, and yet often species classified under the biological species concept are morphologically identical (i.e. cryptic species). However, detailed knowledge of the degree of morphological diversity among populations is always an important step towards a clear understanding of the species status of distinct populations. Thus the collection and description of morphological variation will always be an important element in evolutionary, taxonomic and ecological studies. Within insect species, variation in morphological characters can be correlated with a range of environmental factors such as latitude and altitude (Cushman et al. , 1993; Hawkins & Lawto n, 1995; Krasnov et al. , 1 996; Smith et al. , 2000). In a study investigating the corn leafhopper, Dalbulus maidis, across a range of latitudes (5 - 28° S) and altitudes (16 - 1628 m a.s.l) in Brazil, De Oliveira et al. (2004) found individual leafhoppers were larger, heavier and had darker pigmentation at higher latitudes, and Chapter 2: Morphology 14 were also heavier at higher altitudes. This pattern follows Bergmann’s rule which predicts increased body size at higher latitudes (Bergmann, 1847), and has been found in other insect groups including European ants (Cushman et al., 1993) and some species of butterflies (Hawkins & Lawton, 1995). Ectotherms, however, generally follow the converse to Bergmann’s rule - smaller individuals towards the poles (Mousseau, 1997; Park, 1949; Pa rtridge & Coyne, 1997; Van Voorhies, 1997). This pattern has been attributed to season length as opposed to temperature as shorter seasons at higher altitudes result in reduced growth and development time (Blanckenhorn & Demont, 2004). Morphological char acters are also affected by sexual selection, where selective processes can result in sexual dimorphism (Fairbairn, 1997) . The most obvious example of sexual dimorphism in animals is where one sex is considerably larger than the other, termed sexual size dimorphism (SSD). Males tend to be larger than females in birds and mammals (Alexander et al., 1979; Payne, 1984; Ralls, 1977), whereas among invertebrates, females tend to be larger (Andersen, 1994; Blackith, 1957; Hurlbutt, 1987; Vollrath & Parker, 1992). Interestingly , however, among invertebrates both male - biased and female- biased SSD can occur, sometimes within the same family or subfamily (Fairbairn, 1990; Tseng & Rowe, 1999). This thesis focuses on a species of ground weta in North Island, New Zea land, Hemiandrus pallitarsis (Orthoptera: Anostostomatidae). Like many Orthopterans, male ground weta produce a spermatophore which is transferred to the female genitalia during mating (Gwynne, 2002). As described earlier, H. pallitarsis males deliver the sperm ampulla to the female’s primary genitalia and deliver the food gift (spermatophylax) to her mid abdomen area, approximately where the accessory organ is located. It has been suggested that this accessory organ may be under sexual selection to acqu ire these nuptial gifts from males (Gwynne, 2005). Morphometric variation in this accessory organ has been studied in two H. pallitarsis populations approximately 40 km apart, Palmerston North and Kiriwhakapapa (Gwynne, 2005 ). The organ was found to var y by more than 50% of its total length (range 0.9 – 1.4 mm), and although differences between the two populations were not explored, the data suggested both populations have overlapping Chapter 2: Morphology 15 ranges in accessory organ length (Palmerston North, 0.9 - 1.3 mm; Kiriw hakapapa, 1.1 - 1.4 mm). As well as variation in the female accessory organ, several authors have alluded to other morphological variation within H. pallitarsis. For example, Johns (2001) noted that populations on the islands close to Northland and the Coromandel Peninsula have slight differences. In addition, Hemiandrus has been recorded on many islands east of the Coromandel Peninsula, all of which are either nature reserves, and/or wildlife sanctuaries, but doubt exists in the identification of ground weta on some of these islands. For example, Hemiandrus similis was recorded on Red Mercury Island in the Mercury group (Johannesson, 1972; Moeed & Meads, 1987). H. similis is synonomous with H. bilobatus (Salmon, 1950) and is otherwise only known from the Wellington region, Matiu/Somes Island, Brothers Island in the Cook Strait and Mana Island off the Kapiti coast. Therefore , it would be unexpected to find H. bilobatus on an island over 500 kilometres to the north. However, if the species of ground we ta on Red Mercury Island is correctly identified as H. bilobatus (syn. H. similis ), then this would present an interesting species disjunction similar to that found in Whitaker’s skinks (Towns, 1985). Therefore , it is important to accurately establish the taxonomic status of ground weta on these islands, and the mainland, before undergoing any genetic investigations. Given the great diversity of accessory organ size, and suggestions of regional morphological variation within H. pallitarsis (Johns 2001), the current study documents morphological diversity within the entire range of H. pallitarsis from latitude 36° to 41°. Specifically, patterns of morphometric variability, sexual size dimorphism, and variation in the female accessory organ are studied. Chapter 2: Morphology 16 2.2 Methods 2.2.1 Sampling sites and collections A total of 170 Hemiandrus pallitarsis individuals were collected between November 2006 and April 2007 from 15 locations in North Island (NI), New Zealand (Fig. 2.1). Specimens from two mainland locations ( Wellington and Palmerston North) were available in the collection of Steve Trewick and Mary Morgan - Richards at the Allan Wilson Centre for Molecular Ecology and Evolution, Massey University, Palmerston North. Specimens from Great Barrier Island were provi ded by John Early at the Auckland Museum. In total, eleven mainland sites and four islands were sampled across NI. Weta were hand collected by searching the ground, low shrubs and tree trunks after dark with a head lamp or torch light, and were found in a variety of habitats ranging from urban gardens to native forest. At most locations, sufficient numbers of weta could be collected within 2 - 3 hours; some areas where ground weta were abundant over 20 adults could be found within 20 minutes. Specimen s were initially frozen at - 20°C and were later preserved in 95% ethanol. Individuals were identified to species level using the following morphological characters: one foretibial spine, three prolateral spines and four retrolateral spines on the mesotibia, 70% cover of fine pilosity on the fourth maxillary palps and bare third maxillary palps (Johns, 2001). Weta were deemed to be adult based on descripti ons by Johns (2001) - in the penultimate instar in males, the falci (hook - like structures) on the last tergite are simple rounded knobs, in adults the falci become blackened hooks, whereas adult females have reduced or absent apical styles on the ovipositor but were also judged on development of the sixth abdominal sternite (accessory organ). Chapter 2: Morphology 17 Figure 2.1. Collection locations of Hemiandrus pallitarsis in North Island, New Zealand. Inset shows collection locations on the Coromandel Peninsula. 2.2.2 Body measurements 91 adult weta (51 males and 40 females) were initially measured for fourteen morphological traits (Fig. 2.2a - h). Morphological structures were measured with digital callipers (AIA) accurate to 0.01mm. Each female accessory organ was dissected and fixed onto a slide with mounting fluid and then measured using a stereomicroscope (Nikon Alphaphot YS) fitted with a calibrated eyepiece. A correlation matrix for nine of the fourteen morphological traits (male and female reproductive characters were not used in the correlation analysis), reduced the number of measurable non - reproductive traits to six, with the length of the metafemur used as an indicator of the length of all other appendages (Appendix A). In fact, the metafemur length was highly correlated (i.e. 90%) to all other leg measures (protibia, mesotibia and metatibia length). Desp ite the high correlation between the width and length of the head (93.8%), both measurements were included in further analyses to determine whether there was any shape relationships between the head and body measurements. Subsequently, eleven morpholog ical traits were measured in 170 adult weta (97 males and 73 females) from ten populations. Measurements for males included the pronotum length (Pr), metafemur length (Mf), fastigium width (Fa) - distance between Chapter 2: Morphology 18 lateral ocelli - , head width (Hw) and length ( Hl), distance from fastigium to hind coxa (used as a descriptor of body length (BL), due to distortions of the abdomen associated with alcohol fixation), length of the subgenital plate (Sg), maximum (Mx Sg) and minimum (MnSg) width of the subgenital plate. The first eight of these measurements were also taken for females, as was the mean length of the left and right tines in the fork of the accessory organ (LAO), and width of the accessory organ at the base of the fork (WAO). Chapter 2: Morphology 19 Figure 2.2a-h. Morphometric characters measured for Hemiandrus pallitarsis specimens. Abbreviations as follows: metafemur (Mf), length subgenital plate (Sg), maximum width subgenital plate (MxSg), minimum width subgenital plate (MnSg), length of right (LAO1) and left (LAO2) tine of the accessory organ fork, width of accessory organ (WAO), head length (Hl), head width (Hw), fastigium width (Fa), distance between third tergite and anterior fastigium (Bl), length pronotum (Pr). Chapter 2: Morphology 20 2.2.3 Measurement error Of the 169 weta that were measured, fifteen were selected at random (via a random number generator) and were re - measured for all eleven traits without reference to the original measurements. For each of the eleven traits, the first measurement taken was subtracted from the second measurement and the mean of the difference was calculated for each trait. A t- test comparison was employed using the statistical software SPSS version 4.0 (SPSS, Chicago, Ill., USA), to test the hypothesis that the mean was equal to zero. If the mean of the dif ference between the first and second measure was not equal to zero, then that trait was dropped from subsequent analyses. 2.2.4 Statistical analyses MIN ITAB v. 14.0 software was used to generate standard descriptive statistics and perform multivariate analysis of the morphometric data. A paired t-test was used to confirm the observation that females are larger than males. The data were log- transformed to improve normality as several of the traits did not follow a normal distribution (Anderson- Darling Normality Test; p > 0.10). Variability of morphological characters among populations was analysed using both univariate and multivariate methods. Univariate comparisons were performed by one - way analysis of variance (ANOVA). If significant differences a mong the populations were found ( p < 0.05), then Tukey’s posthoc test was used to determine which populations did not differ from each other and which populations were significantly different from all other populations. Multivariate analysis included pri nciple component analysis (PCA); an ordination technique which is a form of data reduction that still maintains the maximum information of the dataset. PCA summarises multivariate information (in this case: eleven body measurements, nine locations, and tw o sexes) and allows the detection of similarities and differences in the data. These patterns are then represented graphically by plotting the first principle component against the second principle component. Points on the graph which are close together are more similar morphometrically than those which are further apart. Principle components can be used to give descriptions of the general morphology of the organism (e.g. body size, body shape). Post hoc tests and PCA were not performed for the length or width of the accessory organ (LAO and Chapter 2: Morphology 21 WAO, respectively) as the accessory organ could not be dissected out of the museum specimens (i.e. Great Barrier Island individuals). Discriminant function analysis (DFA) was performed to determine which variables best predict the population origin of ground weta. DFA calculates the percentage of each population which is correctly assigned to their population of origin based on the morphometric measurements. Cross- validation was employed whereby one case is omitted at a time, the remaining data are used to recalculate the classification function, and then the omitted case is classified. Variation in body size with latitude was determined by plotting each population mean for each morphometric character against latitude, which ranged from 36° 18 ′ (Great Barrier Island) to 41° 16 ′ (Wellington). Pearson r correlation coefficients were used to evaluate the relationship between body size and latitude. Chapter 2: Morphology 22 2.3 Results 2.3.1 Measurement error For ten of the eleven morphom etric measurements, the mean was not significantly different from zero (Appendix B). Therefore all traits except the minimum width of the male subgenital plate were used in further morphometric analyses. 2.3.2 Sexual size dimorphism Analysis of all six traits measured for both males and females confirmed the observation that females are significantly larger in all body measurements than their male counterparts ( t-test, p < 0.001, Table 2.1). Because of the significant sexual dimorphism, males and fe males were treated separately in further analyses. Table 2.1. Means and standard deviations (SD) for six morphometric traits of adult male and female H. pallitarsis. Mean (mm) ±SD Paired t- test p Males ( n = 97 ) Females ( n = 73 ) Pronotum (Pr) 4.82± 0.37 5.29±0.40 < 0.001 Metafemur (Mf) 12.66± 0.84 13.48± 0.91 < 0.001 Fastigium (Fa) 1.40±0.11 1.60±0.12 < 0.001 Body length (BL) 10.96± 1.01 11.90± 0.77 < 0.001 Head width (Hw) 4.91±0.32 5.53±0.34 < 0.001 Head length (Hl) 7.36±0.44 8.23±0.49 < 0.001 Chapter 2: Morphology 23 2.3.3 Morphometric variation among populations One- way ANOVA revealed that all morphometric variables differed significantly among the ten populations of male weta (Table 2.2), and between the nine populations of female weta (only one adult female was measured from Wellington and so was not used in morphometric analyses) (Table 2.3). Tukey’s posthoc test showed that for six of the eight morphometric variables, male weta from Palmerston North were significantly smaller than male weta from all other populations and that the eight Coromandel Peninsula populations and Wellington are relatively poorly differentiated from each other. In contrast to males, post hoc tests showed that all female weta populations are relatively poorly differentiated from each other. For male weta, principal component analysis showed that the first and second principle components, PC1 and PC2, explained 68.3% and 11.2% of the total variance, respectively (Fig. 2.3). All morphometric characters had positive loadings on PC1, in which the highest loadings were associated with the head size and the length of the pronotum (Appendix C). The highest loadings on PC2 were the length of the subgenital plate and the width of the fastigium. Head shape, body length and the width of the fastigium had negative loadings on PC2. Scores from PC1 were used as general descriptors of body size, while scores from PC2 did not show a distinct relationship between measurements. PCA also supports the separation of the Palmerston North population. Adult male weta from Palmerston North are smaller in overall body size than weta from all other populations (Fig. 2.3). For the female weta, all variables loaded positively on PC1; head length and head width yielded the highest loadings, followed by the length of the pronotum (Appendix D). Once again, scores from PC1 were used as general descriptors of body size. The length and maximum width of the subgenital plate had the highest loadings on PC2; all other scores on PC2 had negative loadings. Therefore, PC2 describes the size of the subgenital plate in relation to all other body parts. PC1 and PC2 accounted for 78.2% of the total morphometric variability. Figure 2.4 shows several outliers from Great Barrier Island, and a single outlier from Kauaeranga Vall ey. These individuals have smaller subgenital plates than all other weta. Overall, the majority of adult female weta from Palmerston North were smaller in body size than weta from the Coromandel Chapter 2: Morphology 24 populations, although, unlike males, there is overlap in body size and shape with other populations (Fig. 2.4). 5.02.50.0-2.5-5.0-7.5 2 1 0 -1 -2 -3 PC 1 (68.3%) PC 2 ( 11 .2 % ) Wellington Great Barrier Is. Moehau Cuvier Island Taiharuru Middle Is. Coromandel Town Ruamahuanui Is. Kauaeranga Valley Palmerston North Location Figure 2.3. Principal component analysis on eight morphological traits of adult male H. pallitarsis. Percent of total variation marked on each axis. Chapter 2: Morphology 25 420-2-4-6 4 3 2 1 0 -1 -2 -3 PC 1 (64.7%) PC 2 ( 13 .4 % ) Great Barrier Is Moehau Cuvier Is. Taiharuru Middle Is. Coromandel Town Ruamahuanui Is. Kauaeranga Valley Palmerston North Location Figure 2.4. Principal component analysis on eight morphological traits of adult female H. pallitarsis. Percent of total variation marked on each axis. Table 2.2. Measurements of adult male H. pallitarsis from ten populations before log transformation. All measurements are in millimetres (mm). The statistical significance of differences among populations has been assessed using one- way ANOVA. Pronotum (Pr) Metafemur (Mf) Fastigium (Fa) Body length (BL) Head width (Hw) Head length (Hl) Subgenital plate (Sg) Width Subgenital plate (MxSg) Mean ± SD Great Barrier Island 5.03 ±0.27 12.90± 0.37 1.37±0.07 11.09± 0.69 4.72±0.15 7.33±0.23 2.86±0.19 3.11±0.10 Cuvier Island 5.23 ±0.29 12.64± 0.49 1.46±0.09 11.14± 0.46 5.02±0.17 7.47±0.30 2.7 8±0.15 2.85±0.17 Moehau 5.60 ±0.20 12.48± 0.61 1.43±0.03 11.66± 0.45 4.87±0.38 7.49±0.32 2.80±0.32 2.72±0.16 Middle Island 5.40 ±0.21 13.20± 0.60 1.37±0.09 11.27± 0.69 5.05±0.22 7.44±0.32 2.83±0.08 2.90±0.18 Coromandel Town 5.74 ±0.16 12.57± 0.23 1.48±0.04 12.3 1± 0.80 5.27±0.18 7.67±0.24 2.69±0.09 3.15±0.12 Taiharuru 5.61 ±0.24 13.08± 0.43 1.35±0.10 10.73± 0.76 4.92±0.18 7.35±0.28 2.72±0.12 2.77±0.23 Ruamahuanui Island 5.64 ±0.22 12.45± 0.21 1.49±0.05 11.77± 0.65 5.29±0.15 7.78±0.25 2.80±0.09 2.91±0.16 Kauaeranga Valley 5.07 ±0.37 13.82± 0.58 1.42±0.09 11.71± 0.81 5.18±0.28 7.87±0.37 2.85±0.22 3.01±0.11 Palmerston North 4.68 ±0.22 11.06± 0.89 1.31±0.11 9.43±1.05 4.41±0.39 6.70±0.66 2.44±0.14 2.39±0.21 Wellington 5.06 ±0.23 12.79± 0.41 1.56±0.09 11.26± 0.79 5.03±0.25 7.55±0 .35 2.74±0.18 2.98±0.16 ANOVA p < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 Table 2.3. Measurements of adult female H. pallitarsis from nine populations before log transformation. All measurements are in millimetres (mm). The statistical significance of differences among populations has been assessed using one- way ANOVA. Pronotum (Pr) Metafemur (Mf) Fastigium (Fa) Body length (BL) Head width (Hw) Head length (Hl) Subgenital plate (Sg) Width Subgenital plate (MxSg) Length Accessory Organ (LAO) Width Accessory Organ (WAO) Mean ± SD Great Barrier Island 5.03 ±0.27 13.12± 0.99 1.46±0.11 12.23± 0.91 5.15±0.26 7.81±0.44 1.77±0.56 2.25±0.47 - - Cuvier Island 5.23 ±0.29 13.42± 0.60 1.61±0.07 11.51± 0.30 5.49±0.29 8.21±0.46 2.12±0.29 2.5 5±0.13 2.49±0.21 1.28±0.13 Moehau 5.60 ±0.20 13.90± 0.30 1.78±0.06 12.11± 0.83 5.76±0.08 8.57±0.22 2.03±0.20 2.52±0.14 2.22±0.15 1.27±0.16 Middle Island 5.40 ±0.21 13.91± 0.37 1.59±0.06 12.22± 0.46 5.57±0.22 8.35±0.24 2.28±0.12 2.55±0.14 2.70±0.20 1.37±0.20 Coromandel Town 5.74 ±0.16 13.88± 0.32 1.66±0.03 12.90± 0.68 6.00±0.13 8.86±0.27 2.08±0.04 2.62±0.19 2.61±0.12 1.42±0.16 Taiharuru 5.61 ±0.24 14.50± 0.47 1.62±0.09 12.08± 0.59 5.76±0.20 8.51±0.27 2.02±0.29 2.53±0.40 2.67±0.46 1.28±0.20 Ruamahuanui Island 5.64 ±0 .22 13.50± 0.59 1.67±0.10 12.31± 0.53 5.79±0.14 8.66±0.22 1.95±0.11 2.45±3.7 2.33±0.10 1.39±0.18 Kauaeranga Valley 5.07 ±0.37 13.04± 0.98 1.51±0.09 11.70± 0.69 5.44±0.34 8.08±0.40 1.92±0.41 2.29±0.45 2.58±0.22 1.24±0.14 Palmerston North 4.68 ±0.22 12.11± 0.54 1 .52±0.09 10.93± 0.56 5.16±0.27 7.61±0.44 1.87±0.21 2.32±0.19 2.19±0.18 1.13±0.25 ANOVA p < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.01 5 0.01 1 - - Chapter 2: Morphology 29 2.3.4 Population classification The Discriminant Function Analysis (DFA) on 97 adult males using the ten collection localities resulted in 63.3% correct classification using cross - validation (Table 2.4). All individuals from Palmerston North were correctly assigned to their population of origin and Great Barrier Island, Ruamahuanui Island, and Kauaeran ga Valley populations only had one individual misclassified in each location. The Discriminant Function Analysis on 72 adult females correctly classified 42.9% of adult female weta to nine populations (Table 2.5). As for the males, Palmerston North had t he highest percentage of correctly classified individuals (55.6%). In general, however, females were not as well classified into their populations of origin as the males. Table 2.4. Percentage correct a posteriori classification of populations based on the eight morphometric variables for male H. pallitarsis. Classified as True group 1 2 3 4 5 6 7 8 9 10 1 Great Barrier Island 11 1 1 0 0 0 0 0 0 0 2 Moehau 0 6 1 1 0 1 0 0 0 2 3 Cuvier Island 1 2 0 0 0 2 0 0 0 1 4 Taiharuru 0 1 0 2 0 2 0 0 0 0 5 Middle Island 0 0 0 0 3 0 1 0 0 0 6 Coromandel Town 0 1 0 2 0 19 0 1 0 0 7 Ruamahuanui Island 0 1 0 0 2 0 4 0 0 1 8 Kauaeranga Valley 0 0 0 1 0 0 0 3 0 0 9 Palmerston North 0 0 0 0 0 0 0 0 11 0 10 Wellington 0 4 0 0 0 1 0 0 0 2 % Correct 91.7 37.5 0 33.3 60 76 80 75 100 33.3 Chapter 2: Morphology 30 Table 2.5. Percentage correct a posteriori classification of populations based on the eight morphometric variables for female H. pallitarsis. Classified as True group 1 2 3 4 5 6 7 8 9 1 Great Barrier Islan d 3 0 0 0 0 0 0 0 0 2 Moehau 0 6 1 0 0 1 0 1 0 3 Cuvier Island 0 1 4 0 0 1 1 0 0 4 Taiharuru 5 2 1 5 0 3 0 1 0 5 Middle Island 0 0 0 1 1 0 1 0 0 6 Coromandel Town 0 0 0 3 1 4 1 1 0 7 Ruamahuanui Island 0 0 2 1 2 0 1 0 1 8 Kauaeranga Valley 1 1 0 0 0 0 0 1 3 9 Palmerston North 0 2 0 0 0 0 0 1 5 % correct 33.3 50.0 50.0 50.0 25.2 44.4 25.0 20.0 55.6 2.3.5 Variation in body size with latitude There was no significant correlation between any of the eight morphometric characters measured and latitude for male weta when the Wellington population was included (Table 2.6). Re - running the analysis without the Wellington population results in significant negative correlations between six of the eight characters with latitude; pronotum length, m etafemur length, body length, head length, subgenital plate length and width of the subgenital plate decreased with increasing latitude (Table 2.6; Fig 2.5). Four out of the ten measurements for females are significantly negatively correlated Chapter 2: Morphology 31 with latitude; pronotum length, metafemur length, body length and width of the accessory organ were all smaller at higher latitudes (i.e. further south) (Fig. 2.5). Table 2.6. Correlation coefficients (linear regression) between morphometric characters and latitude (Lat) in males (including and excluding the Wellington population) and females of H. pallitarsis. Character Males (incl. Wellington) Males (excl. Wellington) Females N r2 p N r2 p N r 2 p Pr/Lat 10 0.2510 0.14 0 ns 9 0.6823 0.00 6 * 9 0.4449 0. 05 0 * Mf/Lat 10 0.2144 0.17 8 ns 9 0.5491 0.02 2 * 9 0.5640 0.02 0 * Fa/Lat 10 0.0313 0.62 5 ns 9 0.3025 0.12 5 ns 9 0.1056 0.39 4 ns Bl/Lat 10 0.2714 0.12 2 ns 9 0.6448 0.00 9 * 9 0.5213 0.02 8 * Hw/Lat 10 0.1452 0.27 7 ns 9 0.4225 0.05 8 ns 9 0.2218 0. 20 1 ns Hl/Lat 10 0.1875 0.21 2 ns 9 0.5730 0.01 8 * 9 0.3411 0.09 9 ns Sg/Lat 10 0.3697 0.06 2 ns 9 0.7797 0.00 2 * 9 0.1050 0.39 4 ns MxSg/Lat 10 0.1414 0.28 5 ns 9 0.5761 0.01 8 * 9 0.1544 0.29 5 ns WAO/Lat - - - - - - 8 0.5055 0.04 8 * LAO/Lat - - - - - - 8 0.2830 0.17 4 ns Chapter 2: Morphology 32 Pronotum length * 0 1 2 3 4 5 6 7 0 50 100 150 200 250 300 350 P r (m m ) Metafemur length * 0 2 4 6 8 10 12 14 16 0 50 100 150 200 250 300 350 M f ( m m ) Head length 0 1 2 3 4 5 6 7 8 9 10 0 50 100 150 200 250 300 350 H l ( m m ) Head width 0 1 2 3 4 5 6 7 0 50 100 150 200 250 300 350 H w (m m ) Body length * 0 2 4 6 8 10 12 14 0 50 100 150 200 250 300 350 B l ( m m ) Fastigium width 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 50 100 150 200 250 300 350 Fa (m m ) Length of subgenital plate 0 0.5 1 1.5 2 2.5 3 3.5 0 50 100 150 200 250 300 350 S g (m m ) Maximum width of Subgenital plate 0 0.5 1 1.5 2 2.5 3 3.5 0 50 100 150 200 250 300 350 M xS g (m m ) Length of accessory organ 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 300 350 Latitude LA O (m m ) Width of accessory organ * 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 50 100 150 200 250 300 350 Latitude W A O (m m ) Figure 2.5. Relationships between the means of morphometric characters and latitude in populations of H. pallitarsis from North Island, New Zealand. Females, ●; males, ○. * indicates characters significantly correlated with latitude in female populations (Pearson r; p < 0.05). All regressions are linear. Latitude is expressed in minutes from Great Barrier Island locality (0 ′) as this is the northern most location. Chapter 2: Morphology 33 2.3.6 Variation in accessory organ shape Cuvier Island Middle Island Taiharuru Coromandel Town Palmerston North Figure 2.6. Images of dissected female accessory organs showing variability in shape from Cuvier Island, Middle Island, Tai haruru, Coromandel Town and Palmerston North. Table 2.7. Pearson correlation r of length of the female accessory organ (LAO) with length of pronotum, metafemur and body length. Pronotum Metafemur Body length LAO Pearson correlation r 0.38 4 0. 51 8 0.37 4 p value 0.002 0.00 0 0.00 4 Chapter 2: Morphology 34 2.4 Discussion Hemiandrus pallitarsis was identified from several new locations within the distributional range described by Johns (2001). This study was able to identify H. pallitarsis from Cuvier Island and on e of the Alderman Islands, previously only identified to the genus level. Significant sexual size dimorphism, and population morphometric variation was found in this study. However, the level of morphological variation found provides no indication of multiple ‘morpho- species’ at different populations in North Island, New Zealand, but is consistent with levels of intraspecific variation. 2.4.1 Sexual size dimorphism There was marked sexual dimorphism in H. pallitarsis. In all morphometric characters, females were significantly larger than males. This trend of female- biased sexual size dimorphism (SSD) is common throughout insects (Teder & Tammaru, 2005) and is also found within many Orthopteran families (e.g. Acridoidea, Bidau & Marti, 2007; Tettigoniidae, Gwynne, 1984; Podisminae, Tatsuta & Akimoto, 1998). Large body size in females is positively correlated with female fecundity (number of eggs) in a wide range of insects (Blanckenhorn, 2000; Fairbairn, 1997; Honek, 1993). Therefore, a linear increase in fecundity with body size may provide evidence for sexual selection where males prefer larger, more fecund mates (Andersen, 1994; Svensson & Petterson, 1987). On the other hand, Issac (2005 ), who studied mammals not insects, suggested decreased male- male competition could result in female- biased SSD, and that males are actually smaller rather than the females being larger. In H. pallitarsis , however, there is some evidence for sexual selection on some female traits. Gwynne (2005 ), experimentally rem oved the accessory organ from nine females, and found that in all cases copulation was prevented. Additionally, in observations of a mating pair in the field, a male was seen to abdominally probe a females mid - abdomen (approximately where her accessory organ is located), then after several minutes he rejected her and moved away (Gwynne, 2005). So it appears that males may be choosing mates based on this female accessory organ. It has also been suggested that the length of this female Chapter 2: Morphology 35 accessory organ may be positively correlated with fecundity (number of eggs in a brood), and therefore, may be under sexual selection to acquire spermatophylaxes from males (Gwynne, 2005). Although Gwynne did not find a correlation between accessory organ length and body siz e (based on length of the pronotum), a positive correlation was found in this study with three measures of body size (length of the pronotum, length of the metafemur, and an approximation of body length; p < 0.005) (Table 2.7). As a result, the length of the accessory organ does appear to increase linearly with body size. Interestingly though, the length of the accessory organ was not correlated with latitude, but the pronotum, metafemur and body length characters were. Due to the lack of general biolog ical knowledge of H. pallitarsis it is unclear how the difference in size between males and females is achieved. Three mechanisms have been hypothesized to lead to sexual size dimorphism: 1) one sex is larger at birth/hatching; 2) one sex grows at a faste r rate; or 3) one sex grows for a longer time period, in insects for example, the larger sex may have additional instars (Esperk & Tammaru, 2006). Therefore, future research examining the growth patterns and basic biology of ground weta will help to elucidate the mechanism leading to sexual size dimorphism in H. pallitarsis. 2.4.2 Geographical morphometric variation Overall male body size appears to separate Palmerston North from the Coromandel populations and Wellington, but there is no significant se paration in female body size between these populations. Male weta are significantly smaller in Palmerston North, a trend also seen in female weta. Female body size appears to be significantly correlated with latitude - smaller weta occur at higher latitudes. Male body size does not appear to be correlated with latitude until Wellington is removed from the data set. Overall, H. pallitarsis does not follow Bergmann’s rule like previous insect examples (Cushman et al., 1993; De Oliveira et al., 2004; Hawki ns & Lawton, 1995), but there is evidence that females follow the converse to Bergmann’s rule, (Atkinson, 1994; Masaki, 1978; Mousseau & Roff, 1989; Mousseau, 1997; Partridge & Coyne, 1997; Telfer & Hassall, 1999). It is important however, to point out that the pattern of decreasing body size with increasing latitude is likely influenced by sampling error. Female weta from the Chapter 2: Morphology 36 Wellington area were not measured, if they had been, we may have seen a similar pattern to what was shown in the male weta – no ge ographical cline in body size. The observed difference in weta body size in the Palmerston North population may relate to environmental conditions, such as season length (Blanckenhorn & Demont, 2004). Shorter seasons at high latitudes restrict the length of time for foraging, growth and development, and insects end up smaller in overall size (Blanckenhorn & Demont, 2004). Previous research on ground weta (from the Christchurch area) suggests they over- winter at approximately the sixth and seventh instars (Wahid, 1978 ), and growth and development mostly occurs during the warmer months (November – April). Therefore, if Palmerston North has a considerably shorter summer season, there would be less development time available, resulting in smaller weta. The lack of correlation between body size and latitude, but the obvious difference in body size of individuals from the Palmerston North population, suggests sexual selection, rather than natural selection, is acting on body size. Evidence for this may be found through examining an aspect of the mating behaviour of H. pallitarsis. Like many insects, males provide a nuptial meal to the female as a form of paternal investment (Thornhill & Alcock, 1983 ). In some cases larger males transfer larger nuptial gifts to females (Eisner & Meinwald, 1995; Fedorka & Mousseau, 2002b). Moreover, Fedorka and Mousseau (2002a) found that “…direct fecundity selection for larger gifts placed indirect positive selection on male body size due to the body size/gift correlation…” (p. 594). Consequently, it is possible that sexual selection is indirectly acting on male body size in H. pallitarsis as a result of female preference for larger nuptial gifts. The fact that Palmerston North males are smaller may be due to population density and competition. For example, Gwynne (1984) studied a species of Mormon cricket in populations of varying densities, and found in high density populations males selected larger females. Gwynne suggested this may be due to food limitation. Of course, it is difficult to imagine that all the populations sampled in the present study except for Palmerston North are at high densities and food limited given the suite of animals which predate upon them (Johns, 2001). In conclusion, this study was able to show significant sexual size dimorphism in H. pallitarsis , but further study is needed to understand how these size differences are ontogenetically achieved. Additional research is also required to fully understand the level of morphological variation within H. pallitarsis , through increased sampling of populations between the Coromandel Peninsula and Wellington regions, and also across Chapter 2: Morphology 37 North Island, from the east coast to the west coast. This information, in conjunction with information on environmental effects such as temperature and season length, may provide a clearer picture of whether H. pallitaris morphometrically varies clinally across the landscape. Chapter 2: Morphology 38 2.5 References Alexander, R. D., Hoogland, J. L., Howard, R. D., Noonan, K. M. and Sherman, P. W. (1 979) Sexual dimorphisms and breeding systems in pinnipeds, ungulates, primates, and humans. In: Evolutionary biology and human social behavior: An anthropological perspective (eds. N. A. Chagnon and W. Irons), pp. 402- 435. Duxbury, North Scituate, MA. Andersen, N. M. (1994) The evolution of sexual size dimorphism and mating systems in water striders (Hemiptera: Gerridae): a phylogenetic approach. Ecoscience, 1: 208- 214. Atkinson, D. (1994) Temperature and organism size - a biological role for ectotherms? . Advances in Ecological Research, 25: 1- 58. Bergmann, C. (1847) Ueber die Verhaltnisse der Warmeokonomie der Thiere zu ihrer Grosse. Goettinger Studien: 595- 708. Bidau, C. J. and Marti, D. A. (2007) Clinal variation of body size in Dichroplus pratensis (Ort hoptera: Acrididae): inversion of Bergmann’s and Rensch’s Rules. Annals of the Entomological Society of America, 100: 850- 8 60. Blackith, R. E. (1957) Polymorphism in some Australian locusts and grasshoppers. Biometrics, 13: 183- 196. Blanckenhorn, W. U. (2000) Temperature effects on egg size and their fitness consequences in the yellow dung fly Scatophaga stercoraria. Evolutionary Ecology, 14: 627- 643. Blanckenhorn, W. U. and Demont, M. (2004) Bergmann and Converse Bergmann latitudinal clines in arthropods: two endsof a continuum? Integrative and Comparative Biology, 44: 413- 424. Chapter 2: Morphology 39 Cushman, J. H., Lawton, J. H. and Manly, B. F. J. (1993) Latitudinal patterns in European ant assemblages: variation in species richness and body size. Oecologia, 95: 30- 37. Darwin, C. (1859) On the origin of species by means of natural selection. Murray, London. De Oliveira, C. M., Lopes, J. R. S., Dias, C. T. D. S. and Nault, L. R. (2004) Influence of latitude and elevation on polymorphism among populations of the corn leafhopper, Dalbulus maidis (DeLong and Wolcott) (Hemiptera: Cicadellidae), in Brazil. Environmental Entomology, 33: 1192- 1199. Eisner, T. and Meinwald, J. (1995) The chemistry of sexual selection. Proceedings of the National Academy of Sciences of the United States of America, 92: 50- 55. Esperk, T. and Tammaru, T. (2006) Determination of female - biased sexual size dimorphism in moths with a variable instar number: The role of additional instars. European Journal of Entomology, 103: 575- 586. Fairbairn, D. J. (1990) Factors influencing sexual size dimorphism in temperate water striders. The American Naturalist, 130: 61- 86. Fairbairn, D. J. (1997) Allometry for sexual size dimorphism: patterns and processes in the coevolution of body size in males and females. Annual Review of Ecological Systems, 28: 659- 687. Fedorka, K. M. and Mousseau, T. A. (2002a) Nuptial gifts and the evolution of male body size. Evolution, 56: 590- 596. Fedorka, K. M. and Mousseau, T. A. (2002b) Tibial spur feeding in ground crickets: larger males contribute larger gifts (Orthoptera: Gryllidae). The Florida Entomologist, 85: 317- 323. Gwynne, D. T. (1984) Sexual selection and sexual differences in mormon crickets (Orthoptera: Tettigoniidae, Anabrus simplex). Evolution, 38: 1011- 1022. Chapter 2: Morphology 40 Gwynne, D. T. (2002) A secondary copulatory structure in a female insect: a clasp for a nuptial meal? Naturwissenschaften, 89: 125- 127. Gwynne, D. T. (2005) The secondary copulatory organ in female ground weta Hemiandrus pallitarsis , Orthoptera: Anostostomatidae): a sexually s elected device in females? Biological Journal of the Linnaean Society, 85: 463- 46 9. Hawkins, B. A. and Lawton, J. H. (1995) Latitudinal gradients in butterfly body sizes: is there a general pattern? Oecologia, 102: 31- 36. Honek, A. (1993) Intraspecific var iation in body size in insects: a general relationship. Oikos, 66: 483- 492. Hurlbutt, B. (1987 ) Sexual size dimorphism in parasitoid wasps. Biological Journal of the Linnaean Society, 30: 63- 89. Isaac, J. L. (2005) Potential causes and life - history consequences of sexual size dimorphism in mammals. Mammal Review, 35: 101- 115. Johannesson, J. M. (1972) Insects of Red Mercury Island. Tane, 18: 81 - 86. Johns, P. M. (2001) Distribution and conservation status of ground weta, Hemiandrus species (Orthoptera: Anost ostomatidae). Science for Conservation Report 180. Department of Conservation, Wellington, New Zealand. Koehl, M. A. R. (1996 ) When does morphology matter? Annual Review of Ecologicol Systems, 27: 501- 542. Krasnov, B., Ward, D. and Shenbrot, G. (1996) Body size and leg length variation in several species of darkling beetles (Coleoptera: Tenebrionidae) along a rainfall and altitudinal gradient in the Negev Desert (Israel). Journal of Arid Environments, 34: 477- 4 89. Chapter 2: Morphology 41 Masaki, S. (1978) Seasonal and latitudinal adaptations in the life cycles of crickets. In: Evolution of insect migration and diapause (ed. H. Dingle), pp. 72- 100. Springer, New York. Mayr, E. (1996) What is a species, and what is not? . Philosophy of Science, 63: 262- 277. Moeed, A. and Meads, M. J. (1987) Invertebrate survey of offshore islands in relation to potential food sources for the little spotted kiwi, Apteryx oweni (Aves: Apterygidae). New Zealand Entomologist, 10: 50- 64. Mousseau, T. A. and Roff, D. A. (1989) Adaptation to seasonality in a cricket: patterns of phenotypic and genotypic variation in body size and diapause expression along a cline in season length. Evolution, 43: 1483- 1496. Mousseau, T. A. (1997) Ectotherms follow the converse to Bergmann’s rule. Evolution, 51: 630- 632. Park, O. (1949) Application of the converse Bergmann principle to the carabid beetle, Dicaelus purpuratus. Physiological Zoology, 22: 359- 372. Partridge, L. and Coyne, J. A. (1997) Bergmann’s rule in ectotherms: is it adaptive? Evolution, 51: 632- 635. Orthithol. Monogr. 33 (1984) Sexual selection, lek and arena behavior, and sexual size dimorphism in birds. American Ornithogists Union. Ralls, K. (1977) Sexual dimorphism in mammals: avian models and unanswered questions. American Naturalist, 111: 917- 938. Salmon, J. T. (1950) Revision of the New Zealand wetas - Anostostomainae (Orthoptera: Stenopelmatidae). Dominion Museum Records in Entomology, 1: 121- 177. Smith, R. J., Hines, A., Richmond, S., Merrick, M., Drew, A. and Fargo, R. (2000) Altitudinal variation in body size and population density of Nicrophorus investigator (Coleoptera: Silphidae). Environmental Entomology, 29: 290- 298. Chapter 2: Morphology 42 Svensson, B. G. and Petterson, E. (1987) Sex - role reversed courtship behaviour, sexual dimorphism and nuptial gifts in the dancefly, Empis borealis (L.). Annales Zoologica Fennici, 24: 323- 324. Tatsuta, H. and Akimoto, S. (1998) Sexual differences in the pattern of spatial variation in the brachypterous grasshopper Podisma sapporensis (Orthoptera: Podisminae). Canadian Journal of Zoology, 76: 1450- 1455. Teder, T. and Tammaru, T. (2005) Sexual size dimorphism within species increases with body size in insects. Oikos, 108: 321- 334. Telfer, M. G. and Hassall, M. (1999) Ecotypic differentiation in the grasshopper Chorthippus brunneus : Life h istory varies in relation to climate. Oecologia, 121: 245- 254. Thornhill, R. and Alcock, J. (1983) The evolution of insect mating systems. Harvard University Press, Cambridge, MA. Towns, D. R. (1985) The status and prospects of the rare New Zealand lizards Leiolopisma grande (Gray), Cyclodina whitakeri Hardy and Leiolopisma otagense McCann (Lacertilia: Scincidae). In: Biology of Australasian frogs and reptiles (eds. C. Grigg, R. Shine and H. Ehmann), pp. 481- 189. Royal Zoological Society of New South Wales. Tseng, M. and Rowe, L. (1999) Sexual dimorphism and allometry in the giant water strider Gigantometra gigas. Canadian Journal of Zoology, 77: 923- 929. Van Voorhies, W. A. (1997) On the adaptive nature of Bergmann size clines: a reply to Mousseau, Partridg e and Coyne. Evolution, 51: 635- 640. Vollrath, F. and Parker, G. A. (1992) Sexual dimorphism and distorted sex ratios in spiders. Nature, 360: 156- 159. Chapter 2: Morphology 43 Wahid, M. B. (1978) The biology and economic impact of the weta, Hemiandrus sp. (Orthoptera: Stenopelmatidae) in an apricot orchard, Horotane Valley. M.Sc. Thesis, University of Canterbury, Christchurch, New Zealand. Chapter 3: Phylogeography Male Hemiandrus. pallitarsis ready to jump from a branch in a Coromandel forest. Photo: K. Rabarts 42 Chapter 3: Phylogeography of a ground weta, Hemiandrus pallitarsis 3.1 Introduction Over the past five million years the New Zealand landscape has undergone mountain building in the South Island (S I), land uplift in the North Island (N I) , and periodic expansion and contraction of glaciers and vegetation zones due to climatic changes. Research into how these past geological and climatic events might have affected the present- day geographic structure of populations has been comprehensively studied in recent years. F or example, in the last ten years over 30 phylogeographic studies of New Zealand taxa have been published. The majority of these studies have focused on phylogeographic patterns in the SI, due its dynamic geological history, and implicate Pliocene mountain- building as drivers of the distribution and structure of many plant taxa (Heenan et al., 2002; Lockhart et al. , 2001; Winkworth et al. , 2002 ), and invertebrate taxa (Buckley & Simon, 2007; Chinn & Gemmell, 2004; Trewick et al. , 2000; Trewick & Morgan- Ric hards, 2005) . In addition to the uplift of the Southern Alps, Pleistocene glaciation and climate change appear to have influenced contemporary phylogeographic patterns in many SI taxa (e.g. moa, Baker et al. , 2005; cicadas, Buckley & Simon, 2007; snails, Neiman & Lively, 2004; weta, Trewick, 2001). Although considerable research has been devoted to understanding how Plio- Pleistocene geological and climatic events may have influenced distribution patterns among many SI organisms, rather less attention has been paid to how these past events may have impacted NI organisms. Of the 30 plus papers that have been published in the last 10 years on phylogeography in New Zealand, approximately half included some representation of NI taxa (e.g. greenshell mussel, Apt e & Gardner, 2002; sea- star, Ayers & Waters, 2005; cicada, Buckley et al., 20 01; freshwater fish, Burridge et al. , 2006; kokako, Murphy et al., Chapter 3: Phylogeography 43 2006; blue duck, Robertson et al., 2007) . However, several studies failed to sample across the entire distributional range of a species (Smith et al. , 2006), or instead focused on species restricted to the northern half of the NI (e.g. Spencer et al., 20 06; Hare et al., 2008; Gleason et al., 1999). Few studies have explicitly examined how patterns of genetic divergence within and among species correspond to Plio- Pleistocene geological and/or climatic changes in the NI (Apte et al., 2007; Morgan- Richards et al. , 2001; Stevens & Hogg, 2004). Those that have analysed intraspecific genetic variation across NI have fou nd apparently incongruent results. For example, Morgan- Richards et al., (2001) used mitochondrial DNA sequence data to investigate the date and pattern of divergence within the Auckland tree weta ( Hemideina thoracica), and to determine if the eight distinct chromosomal races were fixed by Pleistocene or Pliocene isolation events. They found mtDNA lineages and chromosomal races were closely correlated to Pliocene islands in Northland, and that H. thoracica probably expanded its range south into the lower N I after volcanic activity in the central NI and climate fluctuations during the Pleistocene. In contrast to this ‘north to south’ pattern, Apte et al., (2007 ) found strong evidence for mid to late Pliocene ‘south to north’ movement in freshwater crayfishe s (koura) ( Paranephrops planifrons ). Results from the cytochrome oxidase I (COI) gene revealed several patterns: firstly, koura appear to have dispersed from northern SI and Wairarapa into the east and west coasts of NI, from there they moved into norther n NI, and then south again back to Wairarapa. This south to north pattern of dispersal in koura contradicts the author’s prediction that P. planifrons survived in the northern archipelago during the Pliocene marine transgression, and subsequently disperse d south as the land uplifted. Given the current knowledge of the geological and climatic history of NI, both these patterns of lineage evolution could be predicted. Alternatively, it is also possible that the phylogeographic patterns observed have been masked by more recent events such as volcanism or habitat changes. Therefore, a detailed knowledge of the geophysical history of the NI is useful to establish the most likely explanation for contemporary phylogeographic patterns in NI taxa. Chapter 3: Phylogeography 44 3.1.1 North Island Plio-Pleistocene geophysical history Several authors have produced palaeogeographic maps (with varying degrees of similarity) showing the approximate maximum extent of marine transgression in the lower NI at the early Pliocene – an area known as t he ‘proto- strait’ (Fleming, 1979; Kamp, 1982; Kamp & Vucetich, 1982; Lewis & Carter, 1994; Stevens, 1974; Suggate et al., 1978). In general, all authors agree that land existed in what is now the upper half of NI with the southern coastline expanding from north of Taranaki (between Waitara and the Mokau River), east across the central NI to approximately 40 km from the east coast, and then up to East Cape. Rogers (1989) however, highlights some of the variation between authors in the areas suggested to be covered by the Pliocene marine transgression, such as the Kaimanawa Mountains and the Kaweka Ranges (Fleming, 1979; Grindley, 1960; Kamp, 1982; Suggate et al. , 19 78). Additionally, it is agreed that there was a Pliocene land connection between Wairarapa in NI, and the Marlborough Sounds in SI (see Rogers, 1989) , but the estimated position and size of that land connection is unclear. For example, Suggate et al ., (1978) and Lewis and Carter (1994) (and references therein) suggest during the Opoitian (lower Pliocene, 5.28 - 3.6 Ma) there was land from the Kapiti Coast to just south of Wanganui - termed the ‘Wanganui - Marlborough Shield’ (Fig. 3.1A). Stevens (1974), puts forward a rather different outline of the geological history in the central New Zealand r egion. He suggests the land connection between the North and South Islands during the lower Pliocene was more in the south Wellington area and that land was continuous along the southern Tararua Ranges, veering northeast before continuing north. He also illustrates the extensive marine transgression during the middle and late Pliocene which formed three straits: the ‘Kuripapango Strait’, between the rising Ruahine Range and the Kaweka Range - Kaimanawa Mountains; the ‘Manawatu Strait’, between the rising Ru ahine and Tararua Ranges (Fig. 3.1B); and according to Stevens, the Cook Strait was formed in the upper Pliocene (~ 3 Ma) by deep scouring apparently along the ‘Cook Strait Fault’. Stevens (1974) also suggested the Cook Strait was open until the Pleistocene when the sea level dropped 100 to 200 m. During this period (1.2 Ma – 500,000 years ago) evidence for land bridges between the Manawatu- Rangitikei Chapter 3: Phylogeography 45 and Marlborough Sounds regions come from the presence of sedimentary rocks currently found in the d’Urvil le Island - Marlborough Sounds region. Paradoxically (and more recently), Lewis & Carter (1994) suggest that the formation of the Cook Strait occurred in the mid Pleistocene when the sea level rose causing tidal scour of an area termed ‘the Narrows Basin’. Based on three unconformities representing erosion during interglacial periods and the presence of tidal flow during the last glacial maximum, Lewis and Carter suggest the Cook Strait first opened approximately 450,000 years ago; this is in contrast to Stevens’ earlier (and probably incorrect) estimate of 3 million years. During the late Pliocene - early Pleistocene, the NI shoreline in the Wanganui region changed rapidly, an area known globally as the Wanganui Basin because it contains a continuous cyclostratigraphy since ca. 2.6 Ma (Naish, 1997). Uplifting of the Wanganui Basin has exposed foraminiferal and molluscan fossils at each of approximately 22 cyclotherms, which have been used to determine the paleobathymetry of the area (Naish & Kamp, 1997). As the northern part of the Wanganui Basin went through this neotectonic upwarping, the ‘proto- strait’ narrowed to form the ‘Manawatu Strait’, and then as land emerged from the sea to the east in the early- mid Pleistocene, the ‘Manawatu Strait’ was blocke d and a ‘proto- Manawatu River’ formed (Fig. 3.1B - C) (Anderton, 1981; Cotton, 1958; Fleming, 1962). Figure 3.1. Palaeogeographic maps of the lower North Island (outline), showing land above sea level (grey) in the early Pliocene (A), end of the Pliocene (B), and in the early- mid Pleistocene (C) (modified from Lewis & Carter, 1994). Chapter 3: Phylogeography 46 The emergence date of the lower NI axial ranges is estimated at 1 M to ~ 500,000 years (Beu et al., 1981; Ghani, 1978; Walcott, 1978). Beu et al . (1981), e xamined Nukumaruan (2.4 – 1.63 Ma) marine rocks, Mesozoic clasts, and the faulting patterns of Castlecliffian (1.63 – 0.34 Ma) rocks and inferred tectonic uplift of the Ruahine Ranges occurred mostly in the last 1 Ma. Uplift of the Tararua Ranges was more recent than the Ruahines with Ghani (1978), suggesting tectonic uplift around 500,000 – 200,000 years ago. In conjunction with sea level changes and tectonism, volcanism also changed the landscape of the NI. Volcanism in the Taranaki region began in the early Pleistocene with the first volcano, Paritutu, radiometrically dated at around 1.75 Ma (Neall, 1974). Prior to this volcanic activity, the Taranaki region was below the sea surface and lahars produced from Paritutu flowed into the sea (Neall, 1982). The sequence of volcanic activity is referred to as the ‘Taranaki Volcanic Succession’ and the age of each volcano is as follows: Paritutu, last active around 1.76 Ma; Kaitake, last active 575, 000 years ago; Pouakai, last erupted between 216,000 and 250,000 years ago; and finally Mt. Taranaki, last erupting only about 250 years ago (Neall, 1982). At the end of the Last Glacial Maximum (LGM ) around 23,000 years ago, Mt. Taranaki erupted, causing a huge lahar avalanche which spread over 200 km 2 beyond the present coastline. This deposit (the Pungarehu Formation) joined the Farewell Rise, a spit formed by sand and gravel drifting northwards from the west coast of the South Island, resulting in a continuous landbridge between the North and South Island (Lewis & Carter, 1994). At the other end of the NI, islands formed by volcanic eruptions were repeatedly connected to, and separated from, the mainland as a result of sea level changes during the glacial- interglacial cycles of the Pleistocene. For example, volc anic eruptions 5 - 6 Ma to the east of the Coromandel Peninsula produced the Mercury Islands and at least five separate domes formed in the vicinity of the current Alderman Islands; three were centred on Middle Island and remnants of two other domes are now known as Ruamahuanui Island and Hongiora Island (Homer & Moore, 1992). At the LGM (Fig. 3.2), the sea level was approximately 120 m lower than present (Carter et al., 1983). All the current New Zealand offshore islands were connected to the mainland ex cept for the Three Kings Chapter 3: Phylogeography 47 Figure 3.2. Map showing glaciers and vegetation zones in New Zealand at the Last Glacial Maximum (modified from Barrell et al., 2005). Chapter 3: Phylogeography 48 and the Poor Knights Islands (Hayward, 1986). The sea began to rise at a rate of about 10 m per 1000 years and by around 12,000 years ago, the Hen and Chickens, Great and Little Barriers, Cuvier and the Alderman Islands had be come separated from the mainland. The majority of other islands, including the Mercury Islands, were separated between 10,000 – 7,000 years ago and the sea reached its present level 6,500 years ago (Gibb, 1983). The Mercury Islands separated from each ot her over several thousand years, first a ‘super island’ separated from the mainland Coromandel Peninsula and this became fragmented into Great Mercury, Red Mercury, Double, and Stanley Island with Middle, Green and Korapuki still connected to each other until around 6000 years ago (Towns et al., 1990; Towns, 1994) . Not only has the NI undergone a changing coastline over the past 5 Ma as a result of volcanic eruptions and sea level changes, but volcanism in the central NI region has greatly affected the landscape. The Taupo Volcanic Zone (Healy, 1962) covers a distance of 300 km, from Ohakune to White Island in the Bay of Plenty. Over the past 5 Ma, almost 17,000 cubic kilometres of rhyolitic material has erupted within the zone (Healy, 1962). The last maj or eruption was only ~1850 years ago, which deposited pyroclastic rock over 20,000 km 2 of central NI and ash over 30,000 km 2 of eastern NI (Fig. 3.3A) (Wilson & Walker, 1985). Chapter 3: Phylogeography 49 A B Figure 3.3. (A) Map showing the area of North Island covered by p yroclastic rock (dark grey) and ash (light grey) following the Taupo eruption ~1850 years ago. (B) Approximate numbers of endemic vascular plants according to Wardle 1963, and percentage of endemic insect species from Craw, 1988. In addition to the ge ophysical history of NI, it is also interesting to note the different levels of biodiversity and endemicity of extant taxa throughout New Zealand. Based on the number of endemic vascular plants (Wardle, 1963) and endemic insect species (Craw, 1988), New Z ealand has been divided into five regions of high and low endemicity (see Fig. 3.3 B). There are three areas rich in endemic species – northern NI, northern SI, and southern SI, and two endemic poor areas - central SI and lower NI. The endemic rich northern NI region contains a range of invertebrates which are apparently restricted to the area above 39° latitude: for example, two species of stick insects, Spinotechtrachus acornutus and Asteliaphasma jacunda (Trewick & Morgan - Richards, 20 05; Trewick et al., 2008); a species of velvet worm, Peripatoides aurorbis (Trewick, 2000 ); three tusked weta, Motuweta isolata, M. riparia, and Anisoura micobarica (Gibbs, 2001); three giant Chapter 3: Phylogeography 50 weta, Deinacrida fallai, D. heteracantha and D. mahoenui (Gibbs, 2001); and at leas t two ground weta, Hemiandrus “otekauri” and H. “elegans” (Johns, 2001). Likewise, there is greater species richness in the northern SI, for example five of the six short ovipositor ground weta species can be found in the northern half of SI (Johns, 2001) . The classic explanation for these differing degrees of endemism is that areas of high endemism were glacial refugia (Dumbleton, 1970; Wardle, 1963) and the low endemicity in lower NI can be explained by marine transgression in the Pliocene (Rogers, 1989). 3.1.2 Objectives Accordingly, the goal of this research is to determine the phylogeographical structure of endemic ground weta H. pallitarsis populations using mitochondrial cytochrome c oxidase subunit I (COI) sequences. Based on what is known a bout the geophysical history, patterns of endemicity, and current distribution of ground weta species in NI, several predictions can be made concerning the phylogeography of H. pallitarsis : • H. pallitarsis has moved in a ‘north to south’ pattern. Weta would have been restricted to the northern NI, then dispersed south as land emerged above sea level in what is now the lower NI during the Plio - Pleistocene. Genetic diversity should be lower in the more recently colonized populations in the southern half of NI (e.g. Morgan- Richards et al., 2001). • H. pallitarsis has moved in a ‘south to north’ pattern due to the land connection between the (now) Wellington area and upper SI in the early Pliocene. Range expansion north would have occurred as land uplifted i n the lower NI during late Pliocene – mid Pleistocene. • Alternatively, any signal of the impact of Pliocene land formation on H. pallitarsis has been masked by more recent events such as habitat shifts during the Chapter 3: Phylogeography 51 glacial- interglacial cycles in the Pleistocene, or the Taupo eruptions and subsequent ash and rock cover. Overall, the main objective of this chapter is to evaluate how historical geophysical processes have influenced the phylogeographic patterns of H. pallitarsis. However, this chapter also aims to confirm the monophyly of weta putatively identified as H. pallitarsis from sites across the NI including on offshore islands, and also to assess the phylogeographic relationships and evaluate levels of population genetic partitioning within and among populations across NI, New Zealand. Chapter 3: Phylogeography 52 3.2 Methods 3.2.1 Taxonomy of ground weta Johns (1997) lists seven valid species within Hemiandrus , and Jewell (2007) described another two species from the Fiordland area. Johns (2001) added a further 28 undescribed species of Hemiandrus , the names of which he disclaimed as “not available” as per Article 8.3 of the International Code of Zoological Nomenclature (1999). Johns (2001) has given these undescribed species tag - names (in quotation marks), therefore I r efer to the same tag- names. 3.2.2 Sampling sites and collections A total of 176 Hemiandrus pallitarsis individuals were collected between November 2006 and April 2007 from 24 locations in NI, New Zealand (Table 3.1, Fig. 3.4). Specimens from seventeen mainland locations were available in the collection of Steve Trewick and Mary Morgan - Richards at the Allan Wilson Centre for Molecular Ecology and Evolution, Massey University, Palmerston North. Specimens from Great Barrier Island were provided by John Early at the Auckland Museum, and weta were collected from Little Barrier Island by the Department of Conservation (DoC) and donated to the Allan Wilson Centre. In total, nineteen mainland sites and five islands were sampled across NI. Weta were hand col lected, identified and preserved as outlined in Chapter two. Specimens of H. bilobatus , H. “vicinus”, H. “promontorius”, H. “onokis” and H. “horomaka” were obtained as reference samples for phylogenetic out - grouping. Chapter 3: Phylogeography 53 Table. 3.1. Sampling locations for Hemiandrus specimens. * Individuals used in Bayesian analysis of the entire COI gene fragment. † Locations above sea level during the early Pliocene (Fleming, 1979; Lewis & Carter, 1994) (see Fig. 3.7). Species Code Locality Coordinates Accession No . H. pallitarsis G B Great Barrier Island 36°1 8 ´S, 17 5 °3 2 ´E H. pallitarsis LW 2 2 6 * LW 2 2 7 * Little Barrier Island 36°1 1 ´S, 17 5 °3 2 ´E H. pallitarsis CU76 CU77 * CU79 CU81 CU82 * CU83 CU84 CU85 CU86 Cuvier Island † 36°2 6 ´S, 17 5 °4 6 ´E H. pallitarsis MI 9 5 * MI9 6 MI9 7 MI9 8 * MI9 9 MI1 0 0 MI1 0 1 Middle Island, Mercury Islands † 36°3 8 ´S, 17 5 °5 1 ´E Chapter 3: Phylogeography 54 MI1 0 2 MI1 0 3 H. pallitarsis AL104 AL105 * AL106 * AL107 * AL108 AL110 AL111 AL112 Ruamahuanui Island, Alderman Islands † 36°5 7 ´S, 17 6 °0 5 ´E H. pallitarsis CO71 CO72 CO73 CO74 CO75 CO144 CO145 * CO148 CO149 * CO151 CO161 Mt Moehau, Stony Bay † 36°3 0 ´S, 17 5 °2 5 ´E H. pallitarsis CO53 CO61 CO62 Taiharuru Bay † 36°3 6 ´S, 17 5 °3 3 ´E Chapter 3: Phylogeography 55 CO66 CO67 H. pallitarsis CO37 CO41 CO121 CO127 CO128 CO129 CO130 CO141 Coromandel Town † 3 6°4 4 ´S, 17 5 °3 0 ´E H. pallitarsis KA87 KA88 * KA89 KA90 * KA91 KA94 KA119 KA120 Kauraeranga Valley † 37°0 7 ´S, 17 5 °3 7 ´E H. pallitarsis GW1 9 9 GW 2 0 0 Opepe Memorial Reserve, Taupo † 38°4 6 ´S, 17 6 °1 3 ´E H. pallitarsis LW1 8 9 GW 2 3 0 GW 2 3 1 GW 6 6 * Lake Waikarem oana † 38°4 5 ´S, 17 7 °0 9 ´E EU6 7 6 74 0 Chapter 3: Phylogeography 56 H. pallitarsis G W 9 1 * Mt Holdsworth, Wairarapa 40°5 5 ´S, 17 5 °2 0 ´E EU6 7 6 76 8 H. pallitarsis PN22 PN23 PN178 Parata Street, Palmerston North 40°2 1 ´S, 17 5 °3 7 ´E H. pallitarsis G W 9 2A Victoria Esplanade, Palmerston No rth 40°2 2 ´S, 17 5 °3 6 ´E H. pallitarsis PN174 * PN175 * PN176 * PN177 * Bledisloe Park, Palmerston North 40°2 2 ´S, 17 5 °3 7 ´E H. pallitarsis G W 2 3 2 Turitea Road, Palmerston North 40°2 1 ´S, 17 5 °3 7 ´E H. pallitarsis G W 8 7 * Totara Reserve, Pohangina Valley 4 0°0 8 ´S, 17 5 °5 0 ´E H. pallitarsis G W 1 9 0 Gordon’s Bush, Wanganui 39°5 4 ´S, 17 5 °0 5 ´E H. pallitarsis G W 1 7 1 Cambridge † 37°5 3 ´S, 17 5 °2 8 ´E H. pallitarsis TK187 * GW 2 2 8 GW 2 3 5 * GW 2 3 6 Oakura Beach, Taranaki 39°0 7 ´S, 17 3 °5 7 ´E H. pallitarsis GW8 9A GW8 9 B Khandallah, Wellington † 41°1 4 ´S, 17 4 °4 7 ´E H. pallitarsis GW8 0 GW 8 2 GW 8 4 Newlands, Wellington † 41°1 3 ´S, 17 4 °4 9 ´E Chapter 3: Phylogeography 57 GW 1 7 6 * GW 9 7 * EU6 7 6 79 7 H. pallitarsis WN1 8 1 WN1 8 2 WN1 8 3 Otari - Wilton Bush, Wellington † 41°1 6 ´S, 17 4 °4 5 ´E H. bilobatus G W 2 4 0 * G W 2 4 1 * Botanical Gardens, Wellington † 41°1 6 ´S, 17 4 °4 6 ´E H. bilobatus G W 2 5 * Newtown, Wellington † 41°1 8 ´S, 17 4 °4 6 ´E EU6 7 6 79 4 H. “onokis” G W 1 2 7 * GW 1 2 9 * Springfield, Porters Pass † 43°2 0 ´S, 17 1 °5 5 ´E EU6 7 6 77 8 EU6 7 6 77 9 H. “horomaka” G W 1 2 0 * Akaroa, Ban ks Peninsula † 43°4 8 ´S, 17 2 °5 8 ´E EU6 7 6 77 1 H. “promontorius” G W 5 5 * GW 1 2 2 * Marfell's Beach, Cape Campbell, South Marlborough † 41°4 3 ´S, 17 4 °1 2 ´E EU6 7 6 78 9 EU6 7 6 77 7 H. “promontorius” G W 1 9 3 * Muritai Reserve, Seaview, South Marlborough † 41°3 8 ´S, 17 4 °0 8 ´ E H. “vicinus” G W 5 4 * Whites Bay Track, Blenheim † 41°3 1 ´S, 17 3 °5 7 ´E EU6 7 6 78 8 Chapter 3: Phylogeography 58 Figure 3.4. Collection locations of Hemiandrus pallitarsis ( ●), and outgroup species (■) in New Zealand. Inset shows collection locations in the Coromandel. 3.2.3 DNA extraction Muscle tissue from the hind femur of one leg of each specimen was dissected out and stored at - 20°C. Total genomic DNA was extracted using a salting - out extraction method (Sunnucks & Hales, 1996). Tissue was crushed in 10µ L of 10mg/mL proteinase K and incubated in 600µ L of TNES buffer [20mM EDTA Chapter 3: Phylogeography 59 (ethylenediaminetetraacetic acid), 50mM TrisHCl, 400mM Sodium Chloride (NaCl), 0.5% Sodium D odecyl Sulphate (SDS ] at 55°C for 1- 3 hours. 170 µ L of 5M NaCL was added and the tube was mixed well for 15 seconds. The tube was then centrifuged at 13000 revolutions per minute (rpm) for 5 min in a Heraeus benchtop centrifuge. 780µ L of supernatant was removed and mixed with 780µ L of ice cold 100% ethanol and allowed to stand for approximately one hour or overnight at - 20°C. The samples were then centrifuged for a further 5 mins at 13000 rpm. Ethanol was removed from the tube and the DNA was further washed with 400µ L of 70% ethanol and then centrifuged for 5 mins at 13000 rpm. The ethanol was then remove d and the DNA allowed to air dry before it was re- suspended in 50µL of Milli - Q water. Extractions were quantified using NanoDrop (3.3.0) and agarose gel electrophoresis. 3.2.4 PCR amplification A range of universal invertebrate primer combinations were tried in order to sequence the COI and COII genes of the mitochondrial genome. Where possible, a ~1400 base pair (bp) fragment comprising most of the COI gene was amplified using the following primers (Folmer et al., 1 994) (Fig. 3.5): - LCO1490: 5’ - GGTC AACAAATCATAAACATATTGG - L2 - N - 3014: 5’ - TCCAATGCACTAATCTGCCATATTA A smaller ~700 bp region of the COI gene was amplified using the primer pair (Folmer et al., 1994) (Fig. 3.5): - LCO1490: 5’ - GGTCAACAAATCATAAACATATTGG - HCO2198: 5’ - TAAACTTCAGGGTGACCAAAAAAT CA Figure 3.5. Diagram of the cytochrome oxidase I gene showing positions and sequencing directions of the primers used in this study. Chapter 3: Phylogeography 60 In addition, fragments from five outgroup species ( 731 - 847 bp) were used in the analyses (made available by R. Pratt). These fragments overlapped at the 3’ end of the COI gene with the ~1400 bp fragments sequenced in this study, forming overall fragments of length 143 2 bp (missing nucleotides at 3’ end of some sequences and 5’ end of outgroup sequences). Polymerase chain reaction (PCR) used the following conditions: 94°C for 2 min (initial denaturation); 38 cycles (35 cycles for the smaller fragment) of 94°C for 30s (denaturation), 52°C for 30s (annealing), and 72°C for 1m 30s (primer extension); 72°C for 8 min (final extension). Amplification reactions consisted of 1.0µl of the extracted DNA, 0.08µl of ABGene Taq DNA polymerase (500U), 0.8µl of 25mM MgCl 2 , 0.2 µl of each 2mM dNTP, 0.4µl of each 10µM primer, 1.0µl of 10x PCR buffer, and 5.82µl of Milli- Q water in 10 µ l total volume. Agarose gel electrophoresis was used to assess the quantity and quality of the amplified DNA products. 3.2.5 DNA sequencing PCR products of the expected size were purified for sequencing using the SAP/EXO1 digest protocol (USB Corp., C leveland, OH, U.S.A.). Incubation of PCR products and SAP/EXO1 enzymes was at 37ºC for 30 min followed by a denaturation step of 80ºC for 15 min. Purified PCR products were sequenced with one or more of the primers listed above, using BigDye TM Terminator chemistry (Perkin - Elmer Applied Biosystems) (thermal cycling conditions were: 26 cycles of 96ºC for 10 s, annealing at 50ºC for 5 s and 60ºC for 4 mins). Sequencing products were cleaned following the CleanSEQ® Dye Terminator Removal procedure (Agencour t Bioscience Corp.). 10µl of CleanSEQ® magnetic beads and 62µl of 85% ethanol were added to the samples and mixed. Samples were incubated on a magnetic plate for 3 - 5 min before the supernatant was aspirated, then 100µl of 85% ethanol was added and aspira ted. After 10 min of air - drying off the magnetic plate, samples were returned to the plate, 40µl of elution buffer (0.1mM EDTA) was added and the plate was incubated for 5 min to separate the magnetic beads from the solution. 20µl of solution from each s ample was then submitted to the Allan Wilson Centre Genome Service, Massey University, Palmerston North for capillary separation on an AB13730 genetic analyser (Applied Biosystems Inc.). DNA sequence reads were aligned with SEQUENCHER v. 4.2 (Gene Codes Chapter 3: Phylogeography 61 Corp., Michigan). Ambiguous base calls were corrected manually and ambiguous end regions were trimmed so that all individuals were analysed over the same sequence length (638 bp for the shorter fragment and 1432 bp for the longer fragment). Unique haplotypes were identified using the program MacClade (v. 4.0). 3.2.6 Nuclear copies Nuclear copies of mitochondrial genes, often referred to as pseudogenes or Numts , occur when DNA, homologous to mtDNA is integrated into the nuclear genome (Du Buy & Riley, 1967). In such situations, PCR can sometimes amplify the Numt in addition to, or in preference to, the mitochondrial gene region of interest which then leads to incorrect genetic phylogenies. Numts have been found to have derived from all parts of the mtDNA genome, vary in size and vary in the levels of similarity to their mitochondrial counterparts (Zhang & Hewitt, 1996) . As of 2001, Numts have been documented in over 82 different species (for a list of all species see http://www.pseudogene.net), and a ppear to be especially common in Orthoptera (Bensasson et al., 2001) . Previous research on ground weta incurred problems with amplifying nuclear copies within the COI gene of ground weta and in particular H. pallitarsis (pers. comm. Renae Pratt). Consequently, in order to avoid the inclusion of nuclear sequences into the dataset a range of precautions were enforced: firstly, PCR amplifications showing more than one band were discarded; secondly, stock solutions of extracted DNA were serially diluted in or der to reduce the ratio of nuclear DNA to mtDNA, and therefore, favour the amplification of mitochondrial targets as opposed to nuclear targets (Kidd & Friesen, 1998); thirdly, nucleotide sequences sequenced from both ends of the gene fragment but yielded contig ambiguities were discarded; fourthly, sequences containing unexpected insertions or deletions, stop codons or frameshifts were also discarded (Kidd & Friesen, 1998); and finally, a primer sequence was developed from a suspected Numt sequence (PCR pr oduct clearly showing double bands; see Appendix E) and used to amplify possible Numts from several locations. The resulting sequences were analysed with several sequences not suspected of being nuclear copies using Neighbour - joining (NJ ) method in PAUP* v.4.0b10 (Swofford, 2002). The NJ tree shows the suspected Numts form a clade which is sister to the rest of Chapter 3: Phylogeography 62 the sequences (Fig. 3.6). This information was used to identify putative Numt sequences. Figure 3.6. Neighbour Joining (NJ ) tree with H. bilobatus as outgroup. Symbols indicate alternative mtDNA COI sequences from the same individuals: ● Cuvier Island,  Middle Island. Arrow shows position of sequenced Numts . 3.2.7 Phylogenetic analyses Phylogenetic analyses were performed using neighbour- joining (NJ ), maximum parsimony (MP) and maximum likelihood (ML) methods implemented by PAUP* v.4.0b10 (Swofford, 200 2). MP reconstructions were unweighted and run using the default settings. To determine the most appropriate model of DNA evolution, log - likelihood scores were generated and used to conduct an Akaike information criterion Chapter 3: Phylogeography 63 (AIC) test in MODE LTEST 3.06 (Posada & Crandall, 1998). NJ and ML reconstructions were run with the model and parameters obtained using MODE LTEST 3.06. Bootstrap resampling with 1000 replicates was used to determine the degree of support for each node in MP and NJ trees. Due to computing - time constraints only 100 replicates were used for the ML tree in conjunction with a cut down taxon list, which included only one representative sequence for each clade. Bayesian analysis was implemented with MR BAYES v.3.1.2 (Ronquist & Huelsenbeck, 2005 ) using HKY (nst=2) or GTR (nst=6) models, and were run for 6 million generations with two independent simultaneous runs with three heated chains that were sampled every 1000 generations. Consensus tree and credibility values for each node were obtained from sampled trees after burning the first 20% of the chain. To investigate the monophyly of H. pallitarsis , larger COI fragments (14 32 bp) were analysed, with ten individuals from five of the closest known relatives of H. pallitarsis included as outgroups (based on ovipositor length and modification of the 6 th abdominal sternite; Johns 2001). Seven outgroup sequences and three H. pallitarsis sequences, from the 3’ end of COI gene, were provided by R. Pratt. Intraspecific phylogenetic analyses were performed using the smaller COI fragment (63 8 bp). 3.2.8 Population genetic analyses Nucleotide diversity (π, Tajima, 1983), haplotype diversity (h, Nei, 1987), estimate of effective population size (θ(S), Tajima, 1989), and Tajima’s test of selective neutrality (Tajima’s D: Tajima, 1989) were calculated for each location and across all samples in the program ARLEQU IN 2.0 (Excoffier et al. , 2005). The partition of genetic variability among populations and among groups of populations was defined by analysis of molecular variance (AMOVA, Excoffier et al., 1992) using ARLEQU IN 2.0. In order to determine the possible origin if diversification of H. pallitarsis, thirteen populations were partitioned into those regions above sea level (R egion 1 in northern NI and Region 3 in the Wellington area) and below sea level (Region 2) during the Pliocene (Fig. 3.7). Greater genetic diversity occurs in older populations due to in situ survival and only a subset of this diversity is carried into founding populations from the source population (Avise, 2000; Hewitt, 1999) . Therefore, we would expect to find greater genetic diversity in Region 1 based on the prediction of ‘north to south’ Chapter 3: Phylogeography 64 expansion, or in Region 3 based on the ‘south to north’ predict ion (see Introduction - 3.1.2). Figure 3.7. Map showing hypothesized distribution of land in the North Island at the early Pliocene (grey areas) (modified from Morgan - Richards et al., 2001). Sampling locations within each region are indicated by white circles. Pairwise estimates of genetic differentiation between populations were examined using ΦST , an analogue of F ST that incorporates haplotype frequency and similarity (Excoffier et al. 1992). Genetic distances used were uncorrected p distances. Euclidean geographical distances among all population pairs were measured in Le Progiciel. Pairwise matrices of geographical distances and ΦST were compared using Mantel Tests for matrix correlation (Mantel, 1967), with significance assessed by 10,000 randomizations of the genetic distance matrix. Separate Mantel Tests were performed at three different geographic scales: the first and smallest geographic area encompassed the Coromandel Peninsula and nearby offshore islands, including Great Barrier and Little Barrier I slands (Fig. 3.8A); the next scale included the previous area plus the Chapter 3: Phylogeography 65 central and eastern NI (i.e. Cambridge, Taupo and Lake Waikaremoana) (Fig. 3.8B); and the last Mantel Test covered the entire sampling area (Fig. 3.8C). These isolation - by- distance (I BD) analyses were performed using ARLEQ U IN 2.0 (Excoffier et al. , 2005). Figure 3.8. Map of NI, New Zealand showing populations sampled at three geographical scales: A) Coromandel Peninsula and offshore islands (dark grey); B) Taupo, Lake Waikaremoa na and populations in A (medium grey); C) Taranaki, Palmerston North, Wellington and populations in A and B (light grey). 3.2.9 Molecular dating A molecular clock calibration rate of between 1.4 – 2.6% sequence divergence per million years was used to estimate possible divergence times of lineages within H. pallitarsis. This range of calibration rates has been derived from comparisons between geological data and arthropod mitochondrial data (Brower, 1994; Brown et al. , 1996; Brown et al., 1979; Folme r et al. , 1994; Juan et al. , 1995; Knowlton et al. , 1993; Knowlton & Weigt, 1998 ). Chapter 3: Phylogeography 66 3.3 Results Two datasets of cytochrome oxidase DNA sequences were produced for the ground weta. The first dataset contained sequences from 24 individuals of H. pallitars is from 13 locations, and five outgroup species: two H. “onokis” individuals, one specimen of H. “horomaka”, three H. bilobatus individuals, three H. “promontorius” individuals, and one specimen of H. “vicinus” (Table 3.1). The dataset contained 1432 bp, of which 406 (28.4%) were variable, and 353 (24.7%) were parsimony - informative. Approximately 84% of all substitutions were at third - codon positions, 13% of the variable sites occurred at the first- codon position, and 2% were found at the second - codon position. The second dataset contained sequences from a further 64 H. pallitarsis individuals from 6 additional locations. In total, the second dataset contained sequences from 88 H. pallitarsis individuals from 18 locations, each 638 bp in length. The 638 bp COI alignment contained 131 variable sites (20% ) , of which 117 (18%) were parsimony informative. Approximately 89% of all substitutions were at third - codon positions, and only 10% of the variable sites occurred at the first - codon position; no substitut ions were found at the second- codon position. There was no evidence of frameshifts or stop codons in the amino acid translation. All but two substitutions were synonymous mutations resulting in no amino acid change: two individuals from Middle Island and one from Cuvier Island contained Isoleucine amino acid instead of Valine; and all the northern Coromandel Peninsula individuals (i.e. Moehau, Taiharuru and Coromandel Town) contained Threonine amino acid instead of Alanine. Fifty - five unique haplotypes were detected, with no haplotype sharing among sampling sites (Appendix F). The largest number of nucleotide substitutions between haplotypes was 57 (8.9% sequence divergence), between a haplotype found in Taupo and one found in Wanganui, and only a single base pair substitution separates the nearest haplotypes (0.16% sequence divergence). 3.3.1 Phylogenetic analyses Phylogenetic analysis of the long DNA fragment (1432 bp) using Bayesian search method produced the consensus tree shown in Figure 3.9. The 24 H. pallitarsis Chapter 3: Phylogeography 67 individuals formed a monophyletic group (posterior probability = 0.98 ), which was sister to the Canterbury and Banks Peninsula species ( H. “onokis” and H. “horomaka”, respectively). H. bilobatus from Wellington forms a clade with H. “p romontorius” from Cape Campbell in the north east South Island. The dominant split within the H. “promontorius” clade equates to two separate collection sites, Marfell’s Beach and Muritai Reserve, which are approximately 26km from each other. The Marlbor ough sounds species ( H. “vicinus”) is sister to Wellington and Cape Campbell species ( H. bilobatus and H. “promontorius”). Although each species appears to form a monophyletic clade (despite the paucity of sampling of the outgroup taxa), there is greater diversity within H. pallitarsis than there is between H. bilobatus and H. “promontorius” and between H. “onokis” and H. “horomaka”. Figure 3.9. Bayesian consensus network showing the evolutionary relationships among short - ovipositor ground weta of New Zealand: Hemiandrus pallitarsis (large grey box), H. bilobatus, H. “onokis”, H. “horomaka”, H. “promontorius” and H. “vicinus”. Bayesian analysis based on 1432 bp of the COI mitochondrial gene, implemented with MrBayes v3.1.2. (Ronquist & Huelsenbeck, 2005) using HKY (nst=2) model, consisting of 6 million generations with two independent, simultaneous runs with three heated chains and a sample frequency of 1000 generations, and a 10% burn in. Node stability is indicated by posterior probabilities. Chapter 3: Phylogeography 68 All NJ , MP (consensus of 635 trees), ML (single optimal tree, - ln L = 2619.5477) and Bayesian analyses of the 55 haplotypes with the short dataset (638 bp) resulted in trees with similar topologies (Fig. 3.10, Fig. 3.11) The topology found was similar to that indicated by the Bayesian analysis of the long dataset (Fig. 3.9). All analyses revealed three major clades within H. pallitarsis : a predominantly northern clade (Clade I) (with the inclusion of Palmerston North individuals, Pohangina Valley and Wanganui), a central clade containing some Palmerston North, Taranaki, Taupo and Cambridge individuals (Clade II), and a southern (Wellington) clade (Clade III). Bootstrap resampling with NJ, MP and ML, and the Bayesian credibility values all indicated strong support for the three clades. Palmerston North was the only population which has haplotypes that occur in two clades; individuals from the northern side of the Manawatu River yielded Clade I while individuals from the south side of the river yielded Clade II h aplotypes (there was no correlation with collector or time of sequencing). These two populations are separated by < 7 kms but are genetically separated by between 6.3 – 7.7% sequence divergence (see Results - 3.3.4). Internal edges within Clade I are short implying little variation at this level; these short edges correspond to areas of conflict within the Bayesian consensus network (Fig. 3.11 ). Within Clade I, genetic structure is concordant with geographic distribution; populations from the northern half of the Coromandel Peninsula (i.e. Moehau, Taiharuru and Coromandel Town) are sister to Kauraeranga Valley, at the base of the peninsula, Middle Island and Ruamahuanui Island. However, Cuvier Island is not part of this Coromandel clade, bootstrap resampling supports the placement of an individual from Wanganui as sister to weta from Cuvier Island. There is also strong support for the separation of populations within Wellington (Clade III) - individuals from the suburb Newlands are sister to individuals from the suburbs Khandallah and Wilton and differ from each other by < 5.3%, although they are only ~ 7 km apart. Chapter 3: Phylogeography 69 Figure 3.10. Maximum likelihood tree for Hemiandrus pallitarsis based on 638 bp of the COI mitochondrial gene. Tree is midpoint rooted due to long branch lengths of outgroup species. An arrow indicates position of root in analyses that include outgroup Hemiandrus spp. Modeltest v3.06 selected TVM + I + G model ( - ln L = 3057.3407) , with a gamma distribution shape parameter of 0.5777, 0.5548 proportion of invariable sites, and base frequencies of A = 0.26410, C = 0.25660, G = 0.16600, and T = 0.31330. Three measures of branch support are indicated with 1000 NJ and MP bootstrap replicates shown above edges, respectively, and 100 ML bootstrap replicates shown below edges. Terminal labels indicate sampling location and specimen code. Chapter 3: Phylogeography 70 Figure 3.11. Unrooted consensus network for Hemiandrus pallitarsis COI mtDNA sequences. Bayesian analysis implemented with MrBayes v3.1.2. (Ronquist & Huelsenb eck, 2005) using HKY (nst=2) model, consisting of 6 million generations with two independent, simultaneous runs with three heated chains and a sample frequency of 1000 generations, and a 20% burn in. Values on branches are Bayesian credibility values. Te rminal labels indicate individual codes and sampling locations are indicated by ellipses. Chapter 3: Phylogeography 71 3.3.2 Population genetic analyses Of the 55 COI haplotypes sequenced, 51 were used for the population genetic analyses. Four of the populations were removed as the y each had a sample size of one (Great Barrier Island, Cambridge, Wanganui and Pohangina Valley). Tajima’s D was not significantly different from zero in any population (Table 3.2) indicating no evidence for selection on this locus (Tajima, 1989). Nucleotide diversity (π) ranged from 0 – 0.0429 (mean = 0.0086 , SD = 0.0054) with Palmerston North having the highest measure of nucleotide diversity (Table 3.2). Haplotype diversity ( h) ranged from zero, where each individual in the population has the same haplotype, to one, where each individual in the population has a unique haplotype (mean = 0.6316, SD = 0.1393). There were three locations in which only a single haplotype was found at each site, resulting in no haplotypic variation ( h = 0): Taiharuru (one haplotype/five individuals), Ruamahuanui Island (one haplotype/eight individuals), and Taupo (one haplotype / two individuals). Two populations had a haplotype diversity of one: Little Barrier Island (two haplotypes / two individuals) and Lake Waikaremoana (three haplotypes / three individuals). Overall, there is a pattern of low nucleotide diversity and high haplotype diversity for this COI gene fragment in H. pallitarsis . The 13 populations were partitioned into those regions above sea level in the early Pliocene (Regions 1 in the northern half of NI and Region 3 in the Wellington area) and the area below sea level (Region 2) (Fig. 3.7) to determine the possible origin of diversification of H. pallitarsis. Higher diversity was found in Region 3 (Wellington), followed by Region 2 (below sea level) and populations now occupying areas of substantial land area during the Pliocene in the northern half of the NI had the lowest haplotype and nucleotide diversity (Table 3.2). Overall genetic structure was hig h. AMOVA revealed that 73 of 78 (9 3.6%) pairwise population ΦST comparisons were statistically significant (P ≤ 0.05), and Φ ST values were relatively high, with a mean of 0.829 (range 0.42 – 1.00) (Table 3.3). AMOVA also revealed 79.4% of the molecular variance was partitioned among populations while only 20.6% (P ≤ 0.001) was partitioned within populations (Table 3.4). Collection Location (no. of populations sampled) N No. of haplotypes (K) No. of polymorphic sites S Haplotype diversity h (SD) Nucleotide diversity π (SD) Theta S θ (SD) Tajima's D Haplotype ID (no. of individuals) Great Barrier Island* (1) 1 1 0 NA NA NA NA A1 Little Barrier Island (1) 2 2 2 1.00 0 (0.500 ) 0.00 32 (0.00 39 ) 2.00 0 (1.732 0 ) 0.00 00 A2, A3 Moehau (2) 8 5 7 0.74 24 (0.11 58 ) 0.00 36 (0.00 24 ) 2.31 80 (1.19 98 ) - 0.07 64 A4(2), A5(2), A6(2), A7, A8 Cuvier Island (1) 9 4 5 0.58 33 (0.18 33 ) 0.00 23 (0.00 17 ) 1.83 97 (1.06 81 ) - 0.91 00 A9(6), A10, A11, A12 Taiharuru (1) 5 1 0 0.00 00 (0.00 0 ) 0.00 0 (0.000 ) 0.00 0 (0.000 ) 0.00 00 A13(5 ) Middle Island (1) 9 4 3 0.58 33 (0.18 33 ) 0.00 13 (0.00 12 ) 1.10 38 (0.73 85 ) - 0.93 61 A14(6 ), A15, A16, A17 Coromandel Town (2) 8 5 9 0.89 29 (0.08 58 ) 0.00 63 (0.00 40 ) 3.47 11 (1.81 79 ) 0.79 68 A18(2 ), A19, A20, A21(2 ), A22(2 ) Ruamahuanui Island (1) 8 1 0 0.00 0 (0.000 ) 0.00 0 (0.000 ) 0.00 0 (0.000 ) 0.00 00 A23 K auraeranga Valley (2) 8 7 14 0.96 43 (0.07 72 ) 0.00 82 (0.00 51 ) 5.39 94 (2.65 56 ) - 0.14 09 A24, A25, A26, A27, A28, A29, A30(2 ) Cambridge* (1) 1 1 0 NA NA NA NA B40 Taupo (1) 2 1 0 0.00 0 (0.000 ) 0.00 0 (0.000 ) 0.00 0 (0.000 ) 0.00 00 B4 1 (2) Lake Waikaremoana (1) 3 3 10 1.00 0 (0.272 2 ) 0.01 04 (0.00 85 ) 6.66 67 (4.32 78 ) - 0.00 02 A31, A32, A33 Taranaki (1) 4 2 6 0.50 0 (0.265 2 ) 0.00 47 (0.00 37 ) 3.27 27 (2.08 75 ) - 0.80 86 B4 2 , B4 3 (3 ) Wanganui* (1) 1 1 0 NA NA NA NA A34 Palmerston North (4) 9 8 56 0.97 22 (0.06 40 ) 0.04 29 (0.0 2 36 ) 20.604 5 (8.889 7 ) 1.67 63 A35, A36, A37, A38, B4 4, B45 (2), B4 6, B47 Pohangina Valley* (1) 1 1 0 NA NA NA NA A39 Wellington (3) 9 8 37 0.97 22 (0.06 40 ) 0.02 92 (0.01 63 ) 13.613 7 (5.992 2 ) 1.69 03 C48(2), C49, C50, C51, C52, C53, C54, C55 Table 3.2. Collection locations, number of individuals collected, molecular indices and haplotype codes. * removed from population- level analyses due to sample size of one. Table 3.3. Pairwise ΦST values among populations of Hemiandrus pallitarsis. Comparisons that were significant at the P < 0.05 level are indicated by an asterisk. Region 1 64 35 50 0.5766 (0.1418) 0.0036 (0.0027) 2.2798 (1.3538) Region 2 15 12 62 0.7361 (0.1646) 0.0238 (0.0137) 11.9386 (5.4886) Region 3 9 8 37 0.9722 (0.0640) 0.0292 (0.0163) 13.6137 (5.9922) Total (25) 88 55 149 0.6316 (0.1393) 0.0086 (0.0054) 4.6377 (2.3470) 1.2912 Population 1 2 3 4 5 6 7 8 9 10 11 12 1 Little Barrier Is. 2 Moehau 0.930 * 3 Cuvier Is. 0.958 * 0.93 0 * 4 Taiharuru 0.987 * 0.87 6 * 0.97 0 * 5 Middle Is. 0.970 * 0.92 5 * 0.96 3 * 0.97 4 * 6 Coromandel Town 0.869 * 0.69 6 * 0.90 3 * 0.74 8 * 0.88 8 * 7 Ruamahuanui Is. 0.992 * 0.93 2 * 0.97 4 * 1.00 0 * 0.97 2 * 0.88 0 * 8 Kauraeranga Valley 0.833 * 0.82 9 * 0.88 4 * 0.84 4 * 0.79 7 * 0.73 0 * 0.71 3 * 9 Taupo 0.982 0.95 5 * 0.97 5 * 1.00 0 * 0.98 1 * 0.93 1 * 1.00 0 * 0.90 5 * 10 Lake Waikaremoana 0.856 0.91 0 * 0.92 2 * 0.93 6 * 0.93 9 * 0.85 4 * 0.95 5 * 0.82 8 * 0.90 4 11 Taranaki 0.9 38 0.94 3 * 0.96 1 * 0.97 0 * 0.96 1 * 0.91 7 * 0.98 1 * 0.89 7 * 0.92 2 0.90 5 * 12 Palmerston North 0.458 * 0.65 1 * 0.60 9 * 0.56 5 * 0.60 8 * 0.57 3 * 0.58 3 * 0.53 5 * 0.42 4 * 0.47 1 * 0.53 1 * 13 Wellington 0.671 * 0.78 7 * 0.75 2 * 0.75 0 * 0.78 6 * 0.75 9 * 0.79 4 * 0.73 8 * 0.63 4 * 0.66 1 * 0.6 6 2 * 0.49 0 * Chapter 3: Phylogeography 71 Table 3.4. Analysis of molecular variance (AMOVA) using ARLEQU IN (Excoffier et al. 2005) among ground weta from 13 populations in NI, New Zealand. Source of variation d.f Sum of squares Variance components Percentage of variation p Among populations 12 1.691 0.02 03 9 Va 79.42 < 0.001 Within populations 75 0.396 0.00 52 8 Vb 20.58 Total 87 2.087 0.02 56 7 Fixation Index Φ ST : 0.794 1 8 Chapter 3: Phylogeography 72 3.3.3 Isolation-by-distance Genetic differentiation and geographical distance were not s ignificantly correlated at any of the three geographical scales (Fig. 3.12) , indicating that genetic distances among populations varied independently of the geographic proximity between populations. Figure 3.12. Isolation- by- distance plots for (A) Li ttle Barrier Island, the Coromandel Peninsula and surrounding islands - Z score = 13648, R 2 = 0.137551, r = - 0.370878, p = > 0.05, (B) the above plus Taupo and Lake Waikaremoana - Z score = 5040, R 2 = 0.062945, r = 0.250889, p = > 0.05, and (C) the above plus Taranaki, Palmerston North and Wellington - Z score = 1269, R 2 = 0.036622, r = 0.191369, p = > 0.05 (see Fig. 3.8). Chapter 3: Phylogeography 73 3.3.4 Genetic distances and molecular clock For the long dataset containing the outgroup species, uncorrected p genetic distances had a maximum of 16.4%. Most (96%) of pairwise distances between species were greater than 0.105 (10.5%) (Fig. 3.13); nine pairwise distances between H. bilobatus and H. “promontorius”, and two between H. “horomaka” and H. “onokis”, were less than 0.045 (4.5%). Substantial genetic structuring was evident within H. pallitarsis, with the highest uncorrected pairwise genetic distance (8.9%) found between weta from Taupo (GW200) and Wanganui (GW 190) (see Appendix G). The pairwise genetic distance within clade I reached 5.9% (mean = 3.8%), clade II reached 6.3% (mean = 3.7%), and clade III reached 5.3% (mean = 2.9%). Between clades I and II the highest genetic distance was 8.9% (mean = 7.0%), clades I and III got to 8.2% (mean = 7.3%) and between clades II and III the greatest genetic distance was 8.2% (mean = 7.0% ) (Appendix G). Estimates of genetic differentiation using the more complex nucleotide substitution model (TVM + I + G) were higher than the observed pairwise distances. For example, the TVM + I + G distance within H. pallitarsis reached 17.0% (mean = 8.0%) compared to the observed distance of 8.9% (mean = 5.3%). Using molecular clock calibration rates of 1.4 – 2.6% sequence divergence per million years, the three clades within H. pallitarsis appear to have diverged during the Pliocene (~2.0 – 6.3 Ma). Both northern clades (I and II) diverged from the Wellington clade (III) approximately 2.0 - 5.9 Ma and have been separated from each other for about the same amount of time. The two populations on either side of the Manawatu River appear to have diverged between 2.5 and 5.5 Ma. In the Coromandel, the populations on islands east of the peninsula diverged from the Coromandel Peninsula populations during the mid Pleistocene to mid Pliocene (Cuvier Island: 1.46 – 4.86 My; Middle Island: 0.66 – 3.82 My; Ruamahuanui Island: 0.42 – 3.14 My). Chapter 3: Phylogeography 74 0.150.120.090.060.030.00 110 100 90 80 70 60 50 40 30 20 10 0 Genetic distance (uncorrected p) Fr eq ue nc y Between species H. pallitarsis H. "promontorius" H. bilobatus H. "onokis" Figure 3.13. Frequency distribution of pairwise uncorrected p genetic distances from COI mtDNA sequences from four Hemiandrus species, showing pairwise distances between (dark grey bars) and within species. Chapter 3: Phylogeography 75 3.4 Discussion 3.4.1 Phylogenetics Hemiandrus pallitarsis is found in the North Island of New Zealand, from Wellington to Auckland (and is sympatric with several long ovipositor species and one short ovipositor species, H. bilobatus (Johns, 2001)). Phylogenetic analysis supports the monophyly of H. pallitarsis sampled through this range and on offshore islands, with respect to five of the six short ovipositor spe cies in New Zealand. Phylogenetic structure among the short ovipositor species confirmed the sister relationship predicted by Johns (2001) between H. “promontorius” and H. bilobatus , and also found these two species form a clade sister to H. “vicinus” fro m the Marlborough Sounds. This study also found H. “horomaka” from Banks Peninsula is sister to H. “onokis” from the Canterbury region, but was unable to resolve the relationship between H. pallitarsis and these other short ovipositor species. 3.4.2 Genetic distances In a global context, the high level of genetic divergence found within H. pallitarsis at this locus (up to 8.9%) is more typical of interspecific divergence in insects studied in the Northern Hemisphere (Funk et al. , 1995; Funk, 1999; La ngor & Sperling, 1997) , but is consistent with other studies of New Zealand invertebrates. Within the weta family Anostostomatidae, COI intraspecific uncorrected genetic distances have been reported up to 9.5% in Auckland tree weta Hemideina thoracica (Mo rgan- Richards et al. , 2001), up to 7.6% in alpine scree weta Deinacrida connectens (Trewick et al. , 2000) and between 5.3% and 7.1% for several other tree and giant weta species (Trewick & Morgan- Richards, 2005). The highest uncorrected distance among populations of H. pallitarsis was 8.9% which, although it falls within the range found in other Anostostomatids, is considerably smaller than what has been found in a New Zealand mite havestman, Aoraki denticulata (e.g. 14.4%, Boyer et al. , 2007). These larg e genetic distances, in conjunction with the molecular phylogeny which shows branch Chapter 3: Phylogeography 76 lengths within H. pallitarsis are longer than branches between the other five short ovipositor species, may suggest a faster rate of molecular evolution in H. pallitarsis compared to other short ovipositor ground weta (Boyer et al. , 2007). 3.4.3 Population structure and gene flow This study of mtDNA variation found that populations of H. pallitarsis were highly spatially structured. Despite the large genetic distances found within some populations (e.g. Wellington - 5.3%, Palmerston North – 7.7%), AMOVA indicated that the majority of genetic diversity was partitioned among populations rather than within populations (Table 3.4). High haplotype diversity was found in H. pallitarsis , with every location possessing a unique haplotype and in several locations every individual sequenced yielded a unique haplotype (Table 3.2). There was absolutely no haplotype sharing between localities, even those separated by less than 20 kms. In addition, low nucleotide diversity (π = 0.0086) and high ΦST values were found. All these results show substantial population structuring and suggest little recent gene flow between populations of H. pallitarsis. This may imply that ground weta, in particular female weta due to the inheritance of mitochondrial DNA, do not disperse far and may also have very small home ranges. Although there is a paucity of information on the dispersal ability and general ecology of ground weta, a recent study looked at insect colonization of progressively isolated peat bog plants ( Sporadanthus ferrugineus ) from a source population (Watts & Didham, 2006). They found H. pallitarsis at the source population but not in plants 30 m or further from the source populati on. These results support the idea that H. pallitarsis individuals do not travel far, and are unlikely to colonize isolated habitats. It is difficult to imagine, however, that habitat fragmentation has restricted dispersal and that weta are completely confined to isolated patches of vegetation for two reasons: firstly, H. pallitarsis occupies in a wide range of habitat types from mown lawn in urban gardens, to dense patches of Wandering Jew ( Tradescantia fluminensis ), to native forest (E. Chappell, pers. Obs.), and seems not to be reliant on native vegetation; secondly, juvenile ground weta are often observed “pinging” or jumping on the ground at night, often along open walking tracks, and can cover many metres when disturbed. Therefore, it seems likely t hat H. pallitarsis are capable of travelling across areas of open habitat, however further research is required Chapter 3: Phylogeography 77 to fully understand the dispersal abilities of all stages of the life cycle. The finding of high haplotype diversity, coupled with low nucleoti de diversity, suggests each population was formed by an independent founder event and over time genetic drift has resulted in each population becoming genetically distinct (Holland & Cowie, 2007). 3.4.4 Phylogeography Based on the geophysical history a nd patterns of endemicity and biodiversity in NI, several predictions were made regarding the phylogeography of H. pallitarsis : firstly, that population expansion would occur in a ‘north to south’ direction with the lowest genetic diversity occurring in lower NI; secondly, range expansion would occur in a ‘south to north’ direction; or thirdly, any effect of Pliocene impacts on the phylogeographic signal would be masked due to more recent geophysical events. Population genetic analysis revealed Region 3 ( populations from Wellington) had the highest level of genetic diversity and the northern NI region (Region 1) contained the lowest genetic diversity (Table 3.2). Bayesian analysis of the longer data set found H. pallitarsis from Wellington were sister to all the other individuals sequenced (Fig. 3.9). These results support the prediction of range expansion in a ‘south to north’ direction possibly following the pattern of land connection and uplift in the lower NI during the late Pliocene – mid Pleistocene (see Fig. 3.1). The Bayesian analysis also suggests the other short ovipositor NI species, H. bilobatus , may have moved north as it and H. “promontorius” are sister to the southern species, H. “vicinus”. This south to north pattern of population expansion was also found in New Zealand freshwater crayfishes (koura) which appear to have moved from north east South Island into central NI, and then from there into northern NI and south again to Wellington (Apte et al., 2007). Apte et al., also found a clear lineage division within the NI clade, separating northern NI koura from those found on the east and west coasts and from the lower NI. They suggest this northern lineage formed as a result of marine inundation during the Pliocene. No obvious northern lineage was found in this ground weta study, in fact, sequences from lower NI populations (Wanganui, Pohangina Valley and Palmerston North) grouped with all of the northern populations (Fig. 3.10 ). Additionally, it is unlikely that H. pallitarsis has re- colonized the northern half of NI in a stepwise manner through leading- edge re- colonization (Hewitt, 1999; Phillips, 1994), Chapter 3: Phylogeography 78 as no effect of genetic isolation by distance was found, even at the smallest geographical scale (Fig. 3.12A - C). Instead of continuous active dispersal across the landscape where new populations are constantly established ahead of the main population (Hewitt, 199 9) , passive dispersal, mediated by vectors such as flooding events and possibly animals, may explain the present - day phylogeographical patterns. Although there have been no observations of passive dispersal in ground weta, eggs and early instar nymphs live in burrows in the ground and so may be amenable to passive dispersal through flooding events or land slides where the top layer of soil is displaced. As an aside, in species which exhibit low dispersal distances, phylogeographic breaks can arise as a result of genetic drift and not due to geographic barriers (Irwin, 2002). Of course geographic barriers can play an important role in preventing gene flow, and can result in genetically divergent populations over time. Land mass in the NI of New Zealand during the Pliocene was less than that seen today and was fragmented into a series of islands (Fleming, 1979). As stated earlier, much of the lower NI seen today was below sea level and formed a ‘proto - strait’ which later narrowed to become the Manawatu Strait, separating land in the Wellington region from the northern half of the NI. It is tempting to suggest these ‘straits’ (and n ow the Manawatu River) have prevented gene flow between northern and southern populations – Palmerston North populations on the northern side of the Manawatu River are genetically divergent from those on the southern side of the river (Fig. 3.10). However the phylogeographical pattern found in H. pallitarsis is not that straightforward. Several northern NI populations (Cambridge, Taupo and Taranaki) form a clade with one of the Palmerston North populations (from the south side of the Manawatu River); indi cating weta may be capable of dispersing across this geographic barrier. It is possible that this phylogeographic pattern is a result of stochastic population extinction and expansion following habitat shifts during the glacial- interglacial cycles in the Pleistocene. Alternatively, the current phylogeographical pattern may have arisen due to more recent volcanic activity in the central NI. A large area of land surrounding Lake Taupo was either covered in pyroclastic rock or ash (See Fig. 3.3A), effective ly destroying the habitat and so it is unlikely weta were able to survive in this area. Therefore, it is possible that secondary contact of allopatrically differentiated populations has occurred (Morgan - Richards et al., 2000). Chapter 3: Phylogeography 79 3.4.5 Molecular clock and its caveats Using a molecular clock calibration range of between 1.4 and 2.6% sequence divergence per million years (Knowlton & Weigt, 1998; Knowlton et al. , 1993), mtDNA lineages within H. pallitarsis apparently diverged during the Pliocene (~2 - 6.3 M a). Several geographically proximate populations, such as the two Palmerston North sites, were highly genetically divergent. For example, the two reciprocally monophyletic clades in the Wellington region are separated by only ~7 kms, with no obvious barrier in between, but using the molecular clock calibration rate they have apparently diverged 3.8 – 1.7 Ma (mid Pliocene – early Pleistocene). Uncorrected genetic distance data suggests the island populations of H. pallitarsis off the east coast of the Coromandel Peninsula diverged from the mainland between 0.4 and 4.9 Ma. Considering these islands were connected to the Coromandel Peninsula at the last glacial maxima (~22,00 0 years ago) and, due to the cyclical pattern of sea level fluctuations during the Pleistocene, have had the opportunity for intermittent gene exchange until around 12,000 years ago, it is surprising to find such a high level of genetic divergence. Of course if this species does have a faster rate of mtDNA COI evolution, as suggested ea rlier, these divergence times will be overestimates, and the actual timing may be much more recent. However, research based on pedigree mtDNA data in humans and Adélie penguins have shown extremely high mutation rates, exceeding the tradition rate of 1% per million years (Howell et al., 2003; Lambert et al. , 2002). Contrasting substitution rates between population- level studies and phylogenetic studies led Ho et al. (2005), to examine the differences between short - term mutation rates and long- term substitution rates in avian and primate data. They found a reverse ‘J’ curve relationship between calibration age and the rate of change, that is, molecular rates decrease exponentially with increasing time. Likewise, Burridge et al. (2008) also found evidence f or “time dependency of molecular rates” in New Zealand freshwater fishes. Overall, it appears that molecular rates vary across different evolutionary timescales and so using a rate based on phylogenetic data for a population- level study will give misleading inferences. Based on this information, it is likely that H. pallitarsis does not have a faster rate of mtDNA COI evolution, and that population divergence has probably occurred some time during the Pleistocene. Chapter 3: Phylogeography 80 3.4.6 DNA barcoding and morphology DNA barcoding is a taxonomic method aimed at reliably and cost - effectively delimiting species and detecting cryptic species (Hebert et al. , 2003). Usually the 5’ end of the mitochondrial gene, cytochrome c oxidase I, has been used to determine levels of genetic diversity. Despite the initial attraction of this method, subsequent applications of DNA barcoding have found it is not universal and may be misleading in cases where ecological and morphological information is missing (Trewick, 2008). Meyer and Pauley (2005 ), highlight the importance of greater nucleotide diversity between species than within species. In this study pairwise genetic distances between four short ovipositor species of ground weta were greater than those found within H. pallitarsis (Fig . 3.13). These results, in conjunction with the morphometric analyses in Chapter 1, support the current recognized taxonomy for H. pallitarsis. Several recommendations have been put forward, however, to help solve issues regarding the use of DNA characters for species identification and delimitation, including the use of multiple characters, such as morphology and behaviour, and sequencing longer DNA fragments, around 1000 bp (with modern sequencing methods fragments of this size can usually be sequenced using a single set of primers) (Roe & Sperling, 2007 ). Overall this study found substantial genetic structuring within H. pallitarsis , with large genetic distances, high haplotype diversity, low nucleotide diversity and no haplotype sharing among populations. Analysis of mtDNA COI data revealed phylogeographic patterns consistent with a south to north pattern of movement, possibly related to land uplift in the lower North Island during the Pliocene. Phylogeographic patterns among populations of H. pallitarsis may have also been affected by Pleistocene climate change and volcanic activity. Genetic analyses, in conjunction with morphometric analyses give no indication of a species complex within H. pallitarsis, but suggest it is a highly genetically variable species. Chapter 3: Phylogeography 81 3.5 References Anderton, P. W. (1981) Structural evolution of the South Wanganui Bight. New Zealand Journal of Geology and Geophysics, 24: 39- 63. Apte, S. and Gardner, J. P. A. (2002) Population genetic subdivision in the New Zealand greenshell mussel ( Perna canaliculus ) inferred from single - strand conformation polymorphism analysis of mitochondrial DNA. Molecular Ecology, 11 : 1617- 1628. Apte, S., Smith, P. J. and Wallis, G. P. (2007) Mitochondrial phylogeography of New Zealand freshwater cr ayfishes, Paranephrops spp. Molecular Ecology, 16: 1897- 1908. Astrin, J. J., Huber, B. A., Misof, B. and Klutsch, C. F. C. (2006) Molecular taxonomy in pholcid spiders (Pholcidae, Araneae): evaluation of species identification methods using CO1 and 16S rRN A. Zoologica Scripta, 35: 441- 457. Avise, J. C. (2000) Phylogeography: the history and formation of species. Harvard University Press, Cambridge, MA. Ayers, K. L. and Waters, J. M. (2005) Marine biogeographic disjunction in central New Zealand. Marine Biol ogy, 147: 1045- 1052. Baker, A. J., Huynen, L. J., Haddrath, O., Millar, C. D. and Lambert, D. M. (2005 ) Reconstructing the tempo and mode of evolution in an extinct clade of birds with ancient DNA: the giant moas of New Zealand. Proceedings of the National Academy of Sciences, 10 2: 8257- 8262. Barrell, D. J. A., Alloway, B. V., Shulmeister, J. and Newnham, R. M. (eds) (2005) Towards a climate event stratigraphy for New Zealand over the past 30,000 years. Chapter 3: Phylogeography 82 Institute of Geological and Nuclear Sciences, Science Report, SR2005/ 07 (includes poster). Bensasson, D., Zhang, D. X., Hartl, D. L. and Hewitt, G. M. (2001) Mitochondrial pseudogenes: evolution's misplaced witnesses. Trends in Ecology and Evolution, 16: 314- 321. Beu, A. G., Browne, G. H. and Grant - Taylor, T. L. (1981) New Chamys delicata localities in the central North Island and uplift of the Ruahine Range. New Zealand Journal of Geology and Geophysics, 24: 127- 132. Boyer, S. L., Baker, J. M. and Giribet, G. (2007) Deep genetic divergences in Aoraki denticulata (Arachnida, Opiliones, Cyphophthalmi): a widespread 'mite harvestman' defies DNA taxonomy. Molecular Ecology, 16: 4999- 5016. Brower, A. V. Z. (1994) Rapid morphological radiation and convergence among races of the butterfly Heliconius erato inferred from patterns of mitochondrial DNA evolution. Proceedings of the National Academy of Sciences of the United States of America, 91: 6491- 64 95. Brown, W. D., Wideman, J., Andrade, M. C. B., Mason, A. C. and Gwynne, D. T. (1996 ) Female choice for an indicator o f male size in the song of the black- horned tree cricket, Oecanthus nigricornis (Orthoptera: Gryllidae: Oecanthinae). Evolution, 50: 2400- 2411. Brown, W. M., George, M. and Wilson, A. C. (1979) Rapid evolution of animal mitochondrial DNA. Proceedings of the National Academy of Sciences of the United States of America, 76: 1967- 1971. Buckley, T. R., Simon, C. and Chambers, G. K. (2001) Phylogeography of the New Zealand Cicada Maoricicada campbelli Based on Mitochondrial DNA Sequences: Chapter 3: Phylogeography 83 Ancient Clades Associated with Cenozoic Environmental Change. Evolution, 55: 1395- 1407. Buckley, T. R. and Simon, C. (2007) Evolutionary radiation of the cicada genus Maoricicada Dugdale (Hemiptera: Cicadoidea) and the origins of the New Zealand alpine biota. Biological Journal of the Linnean Society, 91: 419- 43 5. Burridge, C. P., Craw, D. and Waters, J. M. (2006) River capture, range expansion, and cladogenesis: The genetic signature of freshwater vicariance. Evolution, 60 : 1038- 1049. Burridge, C. P., Craw, D., Fletcher, D. and Waters, J. M. (2008) Geological dates and molecular rates: fish DNA sheds light on time dependency. Molecular Biology and Evolution, 25: 624- 633. Chinn, W. G. and Gemmell, N. J. (2004) Adaptive radiation within New Zealand endemic species of the cockroach genus Celatoblatta Johns (Blattidae): a response to Plio- Pleistocene mountain building and climate change. Molecular Ecology, 13: 1507- 1518. Cotton, C. A. (1958) Geomorphology. In: An Introduction to the study of landforms pp. 503. Whitcombe & Tombs, Chris tchurch. Craw, R. (1988) Continuing the synthesis between panbiogeography, phylogenetic systematics and geology as illustrated by empirical studies on the biogeography of New Zealand and the Chatham Islands. Systematic Zoology, 37: 291- 310. Du Buy, H. G. a nd Riley, F. L. (1967 ) Hybridization between the nuclear and kinetoplast DNA's of Leishmania enriettii and between nuclear and mitochondrial DNA's of mouse liver. Proceedings of the National Academy of Sciences of the United States of America, 57: 790- 797. Chapter 3: Phylogeography 84 Dumbleton, L. J. (1970) Pleistocene climates and insect distributions. The New Zealand Entomologist, 4: 3- 23. Edmands, S. (2001) Phylogeography of the intertidal copepod Tigriopus californicus reveals substantially reduced population differentiation at northern latitudes. Molecular Ecology, 10: 1743- 1750. Excoffier, L., Smouse, P. E. and Quattro, J. M. (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics, 131 : 479- 491. Excoffier, L., Laval, G. and Schneider, S. (20 05) Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online, 1: 47- 50. Fleming, C. A. (1962) New Zealand biogeography - a palaeontologist's approach. Tuatara, 10: 53- 108. Fleming, C. A. (1979) The geological history of New Zealand and its life. Auckland University Press, Auckland. Folmer, O., Black, M., Hoeh, W., R, L. and Vrijenhoek, R. (1994) DNA primers for the amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology, 3: 294- 299. Funk, D. J., Futuyma, D. J., Orti, G. and Meyer, A. (1995) Mitochondrial DNA sequences and multiple data sets: a phy logenetic study of phytophagous beetles (Chrysomelidae: Ophraella). Molecular Biology and Evolution, 12: 627- 640. Chapter 3: Phylogeography 85 Funk, D. J. (1999) Molecular systematics of cytochrome oxidase I and 16S from Neochlamisus leaf beetles and the importance of sampling. Molecular Biology and Evolution, 16: 67- 82. Ghani, M. A. (1978) Late Cenozoic vertical crustal movements in the southern North Island, New Zealand. New Zealand Journal of Geology and Geophysics, 21: 117- 125. Gibbs, G. W. (2001) Habitats and biogeography of New Z ealand's Deinacridine and tusked weta species. In: The biology of wetas, king crickets and their allies (ed. L. H. Field), pp. 35- 56. CABI Publishing, Wallingford. Gleeson, D. M., Howitt, R. L. J. and Ling, N. (1999) Genetic variation, population structure and cryptic species within the black mudfish, Neochanna diversus , an endemic galaxiid from New Zealand. Molecular Ecology, 8: 47- 57. Grindley, G. W. (1960) Sheet 8, Taupo, Geological Map of New Zealand 1:250 000. Department of Scientific and Industrial Re search, Wellington, New Zealand. Hare, K. M., Daugherty, C. H. and Chapple, D. G. (2008) Comparative phylogeography of three skink species ( Oligosoma moco, O. smithi, O. suteri ; Reptilia: Scincidae) in northeastern New Zealand. Molecular Phylogenetics and Evolution, 46: 303- 315. Hayward, B. W. (1986) Origin of the offshore islands of northern New Zealand and their landform development. In: The offshore islands of northern New Zealand (eds. A. E. Wright and R. E. Beever), pp. 129- 138. New Zealand Department of Lands and Survey information series. Healy, J. (1962) Structure and volcanism in the Taupo Volcanic Zone, New Zealand. In: The crust of the Pacific Basin (eds. G. A. Macdonald and H. Kuno), pp. 151–157. American Geophysical Union Monograph 6, Washington, D.C. Chapter 3: Phylogeography 86 Hebert, P. D. N., Cywinska, A., Ball, S. L. and deWaard, J. R. (2003) Biological identifications through DNA barcodes. Proceedings of the Royal Society of London, B, 70: 313- 321. Heenan, P. B., Mitchell, A. D. and Koch, M. (2 002) Molecular systemati cs of the New Zealand Pachycladon (Brassicaceae) complex: genetic circumscription and relationships to Arabidoposis s. l. and Arabis s. l. New Zealand Journal of Botany, 40: 543- 562. Hewitt, G. M. (1999) Post - glacial re- colonization of European biota. Biol ogical Journal of the Linnean Society, 68: 87- 112. Ho, S. Y. W., Phillips, M. J., Cooper, A. and Drummond, A. J. (2005) Time dependency of molecular rate estimates and systematic overestimation of recent divergence times. Molecular Biology and Evolution, 22: 1561- 1568. Holland, B. S. and Cowie, R. H. (2007) A geographic mosaic of passive dispersal: population structure in the endemic Hawaiian amber snail Succinea caduca (Mighels, 1845). Molecular Ecology, 16: 2422- 24 35. Homer, L. and Moore, P. (1992) Vanishing volcanoes: a guide to the landforms and rock formations of Coromandel Peninsula. Landscape Publications, Wellington. Howell, N., Smejkal, C. B., Mackey, D. A., Chinnery, P. f., Turnbull, D. M. and Herrnstadt, C. (2003 ) The pedigree rate of sequence divergence in the human mitochondrial genome: there is a difference between phylogenetic and pedigree rates. American Journal of Human Genetics, 72: 659- 670. Irwin, D. E. (2002 ) Phylogeographic breaks without geographic barriers to gene flow. Evolution, 56: 2383- 2394. Chapter 3: Phylogeography 87 Jewell, T. (2007) Two new species of Hemiandrus (Orthoptera: Anostostomatidae) from Fiordland National Park, New Zealand. Zootaxa, 1542: 49- 57. Johns, P. M. (1997) The Gondwanaland weta: family Anostostomatidae (formerly in Stenopelmatidae, Henicidae or Mimnermidae): nomenclatural problems, world checklist, new genera and species. Journal of Orthoptera Research, 6: 125 - 138. Johns, P. M. (2001) Distribution and conservation status of ground weta, Hemiandrus species (Orthoptera: Anostostomatidae). S cience for Conservation Report 180. Department of Conservation, Wellington, New Zealand. Juan, C., Oromi, P. and Hewitt, G. M. (1995) Mitochondrial DNA phylogeny and sequential colonization of Canary Islands by darkling beetles of the genus Pimelia (Tenebr ionidae). Proceedings: Biological Sciences, 261: 173- 180. Kamp, P. J. J. (1982) Landforms of Hawke's Bay and their origin: a plate tectonic interpretation. In: Landforms of New Zealand (eds. J. N. Soons and M. J. Selby), pp. 233- 254. Longman Paul, Auckland. Kamp, P. J. J. and Vucetich, C. G. (1982 ) Landforms of Wairarapa in a geological context. In: Landforms of New Zealand (eds. J. N. Soons and M. J. Selby), pp. 255- 268. Longman Paul, Auckland. Kidd, M. G. and Friesen, V. L. (1998) Sequence variation in the guillemot (Alcidae: Cepphus ) mitochondrial control region and its nuclear homolog. Molecular Biology and Evolution, 15: 61- 70. Knowlton, N., Weigt, L. A., Solorzano, L. A., Mills, D. K. and Bermingham, E. (1993) Divergence in proteins, mitochondrial DNA, and reproductive compatibility across the isthmus of Panama. Science, 260: 1629- 1632. Chapter 3: Phylogeography 88 Knowlton, N. and Weigt, L. A. (1998 ) New dates and new rates for divergence across the Isthmus of Panama. Proceedings: Biological Sciences, 265: 2257- 2263. Lambert, D. M ., Ritchie, P. A., Millar, C. D., Holland, B., Drummond, A. J. and Baroni, C. (2002) Rates of evolution in ancient DNA from Adelie penguins. Science, 295: 2270- 2273. Langor, D. W. and Sperling, F. A. H. (1997) Mitochondrial DNA sequence divergence in weevils of the Pissodes strobi species complex (Coleoptera: Curculionidae). Insect Molecular Biology, 6: 255- 265. Lewis, K. and Carter, L. (1994) When and how did the Cook Strait form? In: Evolution of the Tasman Sea Basin. Proceedings of the Tasman Sea Conference, Christchurch, N.Z. (eds. G. J. Van der Lingen, K. M. Swanson and R. J. Muir), pp. 119- 137. Balkema, Rotterdam, the Netherlands. Lockhart, P. J., McLenachan, P. A., Havell, D., Glenny, D., Huson, D. and Jensen, U. (2001) Phylogeny, radiation, and transoceanic dispersal of New Zealand alpine buttercups: molecular evidence under split decomposition. Annals of the Missouri Botanical Garden, 88: 458- 477. Mantel, N. (1967) The detection of disease clustering and a generalized regression approach. Cancer Research, 27: 209- 220. Meyer, C. P. and Paulay, G. (2005) DNA barcoding: error rates based on comprehensive sampling. PLoS Biology, 3: 422. Morgan- Richards, M., Trewick, S. A. and Wallis, G. P. (2000) Characterization of a hybrid zone between two chromosomal races of the weta Hemideina thoracica following a geologically recent volcanic eruption. Heredity, 85: 586- 592. Chapter 3: Phylogeography 89 Morgan- Richards, M., Trewick, S. A. and Wallis, G. P. (2001) Chromosome races with Pliocene origins: evidence from mtDNA. Heredity, 86: 303- 312. Murphy, S. A., Flux, I. A. and Double, M. C. (2006) Recent evolutionary history of New Zealand's North and South Island kokako ( Callaeas cinerea ) inferred from mitochondrial DNA sequences. Emu, 106: 41 - 48. Naish, T. R. (1997) Constraints on the amplitude of Late Pliocene eustatic sea- level fluctuations: new evidence from the New Zealand shallow - marine sediment record. Geology, 25: 113 9- 1142. Naish, T. R. and Kamp, P. J. J. (1997) Sequence stratigraphy of sixth- order (41 k.y.) Pliocene- Pleistocene cyclothems, Wanganui Basin, New Zealand; a case for the regressive systems tract. Geological Society of America Bulletin, 109: 978- 999. Neall, V. E. (1974 ) The volcanic history of Taranaki. Egmont National Park Board, New Plymouth. Neall, V. E. (1982) Landforms of Taranaki and the Wanganui lowlands. In: Landforms of New Zealand (eds. J. M. Soons and M. J. Selby), pp. 193- 212. Longman Paul, Auckland. Nei, M. (1987) Molecular Evolutionary Genetics. Columbia University Press, New York. Neiman, M. and Lively, C. M. (2004 ) Pleistocene glaciation is implicated in the phylogeographical structure of Potamopyrgus antipodarum, a New Zealand snail. Molecular Ecology, 13: 3085- 3098. Posada, D. and Crandall, K. A. (1998) Modeltest: testing the model of DNA substitution. Bioinformatics, 14: 817- 818. Chapter 3: Phylogeography 90 Robertson, B. C., Steeves, T. E., McBride, K. P., Goldstien, S. J., Williams, M. and Gemmell, N. J. (2007) Phylogeography of the New Zealand blue duck ( Hymenolaimus malacorhynchos ): implications for translocation and species recovery. Conservation Genetics, 8: 1431- 1440. Roe, A. D. and Sperling, F. A. H. (2007) Patterns of evolution of mitochondrial cytochrome c oxidase I and II DNA and implications for DNA barcoding. Molecular Phylogenetics and Evolution, 44: 325- 345. Rogers, G. M. (1989) The nature of the lower North Island floristic gap. New Zealand Journal of Botany, 27: 221- 241. Ronquist, F. and Huelsenbeck, J. P. (2005) Bayesian analysis of molecular evolution using MrBayes. In: Statistical methods in molecular evolution (ed. R. Nei lsen). Springer, New York. Smith, P. J., Margaret Mcveagh, S. and Collier, K. J. (2006) Genetic diversity and historical population structure in the New Zealand mayfly Acanthophlebia cruentata. Freshwater Biology, 51: 12- 24. Spencer, H. G., Brook, F. J. and Kennedy, M. (20 06) Phylogeography of Kauri snails and their allies from Northland, New Zealand (Mollusca : Gastropoda : Rhytididae : Paryphantinae). Molecular Phylogenetics and Evolution, 38: 835- 842. Stevens, G. R. (1974) A tramper's geology of the Tararuas. Government Printer, Wellington. Stevens, M. I. and Hogg, I. D. (2004) Population genetic structure of New Zealand's endemic corophiid amphipods: evidence for allopatric speciation. Biological Journal of the Linnean Society, 81: 119- 133. Chapter 3: Phylogeography 91 Suggate, R. P ., Stevens, G. R. and Te Punga, M. T. (1978) The Geology of New Zealand. Government Printer, Wellington. Sunnucks, P. and Hales, D. F. (1996) Numerous transposed sequences of mitochondrial cytochrome oxidase I - II in aphids of the genus Sitobian (Hemiptera: Aphididae). Molecular Biology and Evolution, 23: 510- 524. Swofford, D. (2002) PAUP*: Phylogenetic analysis using parsimony, version 4.0b10 for Macintosh, Sinauer Associates Inc. Tajima, F. (1983) Evolutionary relationships of DNA sequences in finite popul ations. Genetics, 105: 437- 460. Tajima, F. (1989) The effect of change in population size on DNA polymorphism. Genetics, 105: 437- 460. Towns, D. R., Atkinson, I. A. E. and Daugherty, C. H. (1990) The potential for ecological restoration in the Mercury Isla nds. In: Ecological restoration of New Zealand islands (eds. D. R. Towns, C. H. Daugherty and I. A. E. Atkinson), pp. 91- 108. Conservation sciences publication. Towns, D. R. (1994) The role of ecological restoration in the conservation of Whitaker's skink ( Cyclodina whitakeri ), a rare New Zealand lizard (Lacertilia: Scincidae). New Zealand Journal of Zoology, 21: 457- 471. Trewick, S. A. (2000) Mitochondrial DNA sequences support allozyme evidence for cryptic radiation of New Zealand Peripatoides (Onychophor a). Molecular Ecology, 9: 269- 281. Trewick, S. A., Wallis, G. P. and Morgan- Richards, M. (2000) Phylogeographical pattern correlates with Pliocene mountain building in the alpine scree weta (Orthoptera, Anostostomatidae). Molecular Ecology, 9: 657- 66 6. Chapter 3: Phylogeography 92 Trewick, S. A. (2001) Scree weta phylogeography: surviving glaciation and implications for Pleistocene biogeography in New Zealand. New Zealand Journal of Zoology, 28: 291- 298. Trewick, S. A. and Morgan- Richards, M. (2005) New Zealand Stick Insects. Reed Publishing, Auckland. Trewick, S. A., Morgan - Richards, M. and Collins, L. J. (2008) Are you my mother? Phylogenetic analysis reveals orphan hybrid stick insect genus is part of a monophyletic New Zealand clade. Molecular Phylogenetics and Evolution, in press. Walcott, R. I. (1978 ) Geodetic strains and large earthquakes in the Axial Tectonic Belt of North Island, New Zealand. Journal of Geophysical Research, 83: 4419- 44 29. Wardle, P. (1963) Evolution and distribution of the New Zealand flora, as affected by Quar tenary climates. New Zealand Journal of Botany, 1: 3- 17. Watts, C. H. and Didham, R. K. (2006) Influences of habitat isolation on invertebrate colonization of Sporadanthus ferrugineus in a mined peat bog. Restoration Ecology, 14: 412- 419. Wilson, C. J. N. and Walker, G. P. L. (1985 ) The Taupo Eruption, New Zealand I. General Aspects. Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, 31 4: 199- 228. Winkworth, R. C., Grau, J., Robertson, A. W. and Lockhart , P. J. (2002) The origin and evolution of the genus Myosotis L. (Boraginaceae). Molecular Phylogenetics and Evolution, 24: 180- 193. Zhang, D. X. and Hewitt, G. M. (1996 ) Nuclear integrations: challenges for mitochondrial DNA markers. TRENDS in Ecology and Evolution, 11: 247- 251. Chapter 4: Behavioural Drumming Final stage of copulation in Hemiandrus pallitarsis, with male (below) delivering the spermatophylax to the mid abdominal region of the female (above). Photo: E.Chappell 93 Chapter 4: Vibratory drumming variation in three populations of Hemiandrus pallitarsis 4.1 Introduction 4.1.1 Vibrational communication Insects use a range of mechanisms for communication, including vibration (and tremulation), percussion, stridulation, click mechanisms and air expul sion (Ewing, 1989 ), and in some cases a complex combination of these mechanisms is employed (Claridge, 2006). “Silent songs”, produced through substrate vibrations, are a widespread method of communication among insects (Cocroft & Rodriguez, 2005; Henry, 1994; Virant - Doberlet & Cokl, 2004). There are different levels of vibrational communication: near - field air pulsing (Henry, 1994), where individuals in close proximity use their wings to vibrate the air (Bennet - Clark, 1971); drumming, where an insect dir ectly vibrates a body part against the substrate (Stewart et al. , 1991); and tremulation, where an insect vibrates a part of its body without striking the substrate (indirect vibration) (Morris, 1980). Vibrational communication, particularly in associatio n with mate finding and courtship, is thought to be advantageous over the use of acoustic signals because it takes away the potential problems of attracting predators (Bell, 1980; Henry, 1994; Markl, 1983), despite this Cocroft and Rodriguez (2005), list a range of predators capable of eavesdropping on insect vibratory signals (e.g. spiders, Barth, 1997; Morris et al., 19 94; Hemiptera, Pfannenstiel et al., 199 5; and parasitoid wasps, Casas et al., 1998). It can also be disadvantageous due to the short transmission distance (e.g. connectivity of plants) (Henry, 1994 ). S ubstrate transmitted signals, like vibratory signals, are used in species recognition and can serve as an effective pre- mating barrier (Henry, 1993 ; Henry et al., 1999), but are also used for sexual advertisement and portraying aspects of mate quality. Chapter 4: Vibratory Drumming 94 4.1.2 Species recognition signals In the past, sexual signals such as songs, body posturing and displays, were thought to primarily function in species recognition and reproductive isolation (Darwin, 1859, 1871; Henry, 1994) . For example, extensive research on bird song has found birds can discriminate between conspecific and heterospecific songs (Baker & Baker, 1990; Nelson & Marler, 1993; Ratcliffe & Grant, 1985; Searcy et al. , 1995). The process where songs or signals are used by organisms to discriminate conspecifics was originally described by Paterson (1985), which he termed the specific - mate recognition system (SMRS ). This is where an individual produces a signal, such as a song, that another individual of the same species receives, recognizes and responds to. A communication system like this is important for successful fertilization between genetically compatible conspecifics (Wilczynski & Ryan, 19 99), and in cases where species rang es overlap (i.e. sympatric species) differences in sexual communication signals can be accentuated. Seddon (2005), for example, analysed acoustic calls of 21 species of Thamnophilid antbirds in the Neotropics. For each species the author compared call st ructure between a close relative in sympatry and one in allopatry and found that the songs of closely related sympatric species were more divergent that those of closely related allopatric species. Traditional theory suggests greater divergence between species signals in sympatry results from increased selection to limit producing infertile hybrids (Brown & Wilson, 1956; Dobzhansky, 1951), and so courtship signals that reliably reflect species identities are favoured. Calls used as “species recognition s ignals” are mutually recognized by conspecifics of each sex, and so it has been suggested that these signals should not vary greatly across a species geographical range due to stabilizing selection (Butlin, 1995; Paterson, 1985; Wilczynski & Ryan, 1999). In other words, as long as members of a species occupy their normal habitat, there should be little geographic variation in specific- mate recognition characters. In spite of this, species recognition signals can substantially vary across populations (Proh l et al. , 2007; Ryan et al. , 1996a; Ryan & Wilczynski, 1991; Ryan et al. , 1996b; Tregenza et al. , 2000) , suggesting only certain Chapter 4: Vibratory Drumming 95 aspects of these signals may be used for species recognition. As an example, Wilczynski et al., (1995), manipulated the advert isement call of the Neotropical frog ( Physalaemus pustulosus ) and found signal frequencies anywhere between 900 – 560 Hz , followed 50 ms later by a signal between 640 – 500 Hz was all that was necessary for species recognition, much of the natural call var iation was associated with intraspecific mate choice. 4.1.3 Factors shaping the evolution of sexual signals There are a myriad of factors that can shape the evolution of sexual communication systems and that can account for geographic variation. These factors can be divided into two categories: ecological and social factors. Ecological factors: Habitat structure and composition, can directly affect transmission of calls and signals (e.g. Zimmerman, 1983; Slabbekoorn & Smith, 2002; Tubaro & Lijtmaer, 2006), and in the case of vibrational signals, the distance of transmission is limited to the characteristics of the plant and connectivity to other plants (Cokl & Virant - Doberlet, 2003) . Environmental factors such as altitude and temperature can also affe ct advertisement calls (Narins & Smith, 1986). It has long been recognized that temperature affects the calling song of many Orthopteran groups, e.g. Gryllidae (Brooks, 1882; Dolbear, 1897; Prestwich & Walker, 19 81; Toms et al. , 1993), Tettigoniidae (Frin gs & Frings, 1957; Gwynne & Bailey, 1988 ; Walker et al. , 1973), and Acrididae (Skovmand & Pedersen, 1983; von Helverson & von Helverson, 1994). Generally, as temperature increases the pulse rate of the song (pulses per second) increases linearly. Such an effect is not surprising when you take into account the fact that these insect groups use muscle and the nervous system to produce sound, and hence Sanborn (2006) emphasized the importance of measuring body temperature, or ambient air temperature when rec ording calling insects. Interestingly, ‘silent’ songs are also linearly affected by temperature – increased temperature results in an increased drumming rate (drums per second) (Weissman, 2001). Chapter 4: Vibratory Drumming 96 Temperature can also indirectly affect communication systems through selection on body size. Ectotherms generally tend to be smaller in colder temperatures (Mousseau, 1997; Park, 1949; Partridge & Coyne, 1997; Van Voorhies, 1997), and it has been extensively shown in insects that larger males tend to have larger sound producing structures that produce lower frequency calls (Bennet - Clark, 1998; Gerhardt & Huber, 2002; Ryan & Brenowitz, 1985), which can travel over longer distances (Bradbury & Vehrencamp, 1998). Social factors: Sexual selection and character displacement can also result in intraspecific population variation in sexual signals. The process of sexual selection can be loosely defined as competition for fertilization opportunities with the opposite sex. Competition is usually in the form of male- male competition but cryptic female choice can also result in sperm competition (Eberhard, 1996). Character displacement is where historically separated species now have overlapping ranges but which produce infertile hybrids if mated (Butlin, 1987). Sexual sign als can be species specific, but mate quality information can also be included in the calls. For example, many studies have examined the effect of female choice on conspecific male signals and found females are attracted to some signals over others (Orci, 2007; Parri et al., 2002; Ryan & Keddy - Hector, 1992; Tokarz, 1995; Wilczynski et al., 1995; Wilczynski & Ryan, 1999). Moreover, in many cases sexual selection can act on body size with females preferring larger bodied males (Bateman et al., 2001; Brown et al., 1996; De Luca & Morris, 1998; Gwynne, 1982; Simmons, 1992), which as mentioned earlier, can produce lower frequency calls that can travel further. Where there are no barriers to migration and there is gene flow between geographically close populations, variation in sexually selected signals among populations can follow a gradient (cline) across a landscape (Lande, 1982). For example, Ryan et al., (1996) analysed patterns of advertisement call variation in the túngara frog ( Physalaemus pustulosus ) an d found that for some of the call variables there was a clinal pattern of variation. However, they found that the majority of the call variables did not show clinal variation and instead varied between two major allozyme groups. Therefore, sexual selecti on can also generate a non- linear relationship between calling signal and Chapter 4: Vibratory Drumming 97 geographic distance (i.e. no isolation- by- distance effect). In extreme cases, variation in sexually selected signals among populations can lead to speciation through population divergence (West - Eberhard, 1983). 4.1.4 The New Zealand ground weta: Hemiandrus pallitarsis In this chapter I describe the variation in male vibratory signals between three populations of Hemiandrus pallitarsis. Previous research has found evidence that vibratory signals are used for mate attraction in Hemiandrus (formerly Zealandosandrus ) subantarcticus (Butts, 1983) and H. (formerly Z.) maculifrons (Cary, 1981). More recently, Gwynne (2004) , analysed video recordings of vibrational drumming from four species of ground weta in New Zealand: seven H. pallitarsis weta from two locations in the Manawatu area, three H. “onokis” and three H. “vicinus” weta, and a single H. “promontorius” individual. All recordings were made inside a garden hut in Palmerston North. He found the length of the drumming bout and the number of pulses in each bout was species specific. He also made detailed observations of the vibratory drumming behaviour of H. pallitarsis. Gwynne observed adult male weta drumming their abdomens on the substrate, usually a plant surface such as a leaf or stem, and females drumming in apparent response to the male signal. The female drumming pattern appears to differ from that of the male; this however, is based on a recording from a single femal e. Hence, the complex sexual signalling behaviour of H. pallitarsis makes this a useful and interesting species in which to investigate intraspecific population variation in vibratory calls. Gwynne’s 2004 study highlighted the species specific nature of vibratory drumming behaviour in several species of Hemiandrus , but did not touch on the level of intraspecific variation or, intra - and inter- male variation in these vibrational signals. No other studies have examined in detail the vibrational mating behaviour of New Zealand weta, despite its possible importance in reproductive isolation and speciation. Chapter 4: Vibratory Drumming 98 4.1.5 Hypotheses Morphometric and molecular analysis of H. pallitarsis populations in previous chapters lead to two predictions in the pattern of drumming behaviour expected in this species: my first prediction based on morphometric data (Chapter 2) and geographical proximity is that populations on either side of the Manawatu River would exhibit more similar drumming behaviour patterns compared to Coromandel Peninsula weta (Fig. 4.1a); alternatively, based on genetic data (Chapter 3) I predict weta on the north side of the Manawatu River would exhibit more similar drumming behaviour patterns to Coromandel Peninsula weta than they would to weta south of the river due to genetic similarities (Fig. 4.1b). Figure 4.1. Trees illustrate predicted patterns of drumming behaviour in Hemiandrus pallitarsis: a) Morphometric/geographic hypothesis – smaller weta from geographically proximate populations will show behavioural similarities; b) Genetic hypothesis – genetically similar populations (Coromandel and Palmerston North, north of the river) will have drumming patterns more similar to each other than to Palmerston North, south of the river. Chapter 4: Vibratory Drumming 99 4.2 Methods 4.2.1 Study populations Three populations of H. pallitarsis were used for this study (Fig. 4.2): Coromandel Town (36°44 ´ S, 175°30 ´ E); a single location on the northern side of the Manawatu River - Manawatu - North (40°22 ´ S, 175°35 ´ E); and a single location at Massey University, Palmerston North, on the southern side of the Manawatu River - Manawatu - South (40°23 ´ S, 175°3 6 ´ E). The two most separated locations, Coromandel and Manawatu - South, are approximately 400 km apart, while Manawatu- North and M anawatu- South are less than 2 km apart but are separated by the Manawatu River. All individuals at each location were collected within approximately a 100 m 2 area. Figure 4.2. Sampling and recording locations of Hemiandrus pallitarsis, in North Isl and, New Zealand. Chapter 4: Vibratory Drumming 100 4.2.2 Recording techniques and drumming measurements Adult male H. pallitarsis were recorded from mid February through to early March, 2008. A pilot study showed that weta did not drum consistently after several hours or over an ex tended period (days to weeks) in captivity. Therefore, drumming behaviour was recorded in the field within 10 metres of where the weta was caught and within minutes of capture. Weta were hand collected at each field site after dark (from approximately 9: 00pm - 12:00am), and placed in individual plastic containers (4.4 x 5.8 mm) with holes in the lid for ventilation. The lid was removed and the container inverted onto the recording device so that the weta was left sitting directly on the recording device but could not move beyond the confines of the container (see Fig. 4.3). The recording device consisted of a Piezo film (3/8 ″ x 1″) contact pickup (obtained from www.windworld.com/index.htm) – a device which detects vibratory movement – taped down to a small portable surface and connected to an Olympus model VN - 2100PC digital voice recorder (Fig. 4.3). Each weta was left on the recording device for approximately 30 minutes (17 - 209 minutes), after which time the weta was marked on the pronotum with white correction fluid and released back to where it was found - this prevented the same individuals from being collected twice, and did not appear to be harmful as males with correction fluid were seen on following nights mating. Because temperature has been found to linearly affect the calling- song drumming rate in other Orthopterans (e.g. Jerusalem crickets, Weissman, 2001) , the ambient air temperature was measured using a digital thermometer placed at the base of the recording device at the beginning of each recording session. Figure 4.3. The recording setup showing a male weta inside a plastic container sitting on top of the Piezo film (beneath black electrical tape). Photos – E.M. Chappell. Chapter 4: Vibratory Drumming 101 Recordings were analysed using the sound analysis sof tware, Audacity 1.2.6 ( http://audacity.sourceforge.net ). Each drumming bout was amplified so each pulse within the bout, and each oscillation within a pulse could be clearly seen (Fig. 4.4). The following drumming parameters were measured directly from the oscilloscope: 1) Total drumming bout duration in seconds (Bout length), 2) Number of pulses within a drumming bout (Pulses), 3) Mean number of oscillations within one pulse (Oscillations), 4) Inter - pulse interval (seconds) (Inter - pulse), and (5) the number of pulses per second (Pulse rate) (Fig. 4.4). Figure 4.4. Oscillograph of a single drumming bout produced by a H. pallitarsis male showing measurements taken: top trace shows bout length, inter - pulse interval and number of pulses; bottom trace shows five pulses from within the drumming bout, indicating the number of oscillations within each pulse. Chapter 4: Vibratory Drumming 102 4.2.3 Statistical analyses SPSS software package (SPSS ver. 13.0, SPSS, Inc., Chicago, IL) wa s used to obtain descriptive statistics (mean and standard deviation) of drumming bouts for each individual weta in which more than one drumming bout was recorded, and for each population. Pearson r correlation coefficients were used to evaluate the relationship between each of the drumming parameters measured. MLwiN version 2.02 statistical software (Rasbash et al., 2005) was employed for testing intra - and inter- male variation and population variation. To test the assumption of normality, standardized residuals for each drumming parameter were plotted against the predicted values and the normal score. Each plot was examined for evidence of structure in the form of outliers, curved relationships or heteroscedasticity. In all cases residual plots reveal ed normally distributed data. A hierarchical model, with observations within specimens, was used to test whether drumming parameters differ significantly between the three populations sampled: Coromandel, Manawatu- North and Manawatu - South. For each drumm ing parameter (i.e. bout length, pulses), a forward selection method was employed whereby the starting model included only a constant and a population level term, then the effect of temperature was added. Two times log likelihood score, a measure of devia nce, was used to determine whether adding the effect of temperature improved the model (shown as a decrease in the log likelihood score). To test whether any decline was significant ( p ≤ 0.05), a Chi - square table of critical values was used with degrees of freedom equal to the change in the number of parameters. Initial model: Y ij = β 0 i + β 1 j + β 2 j Where: Y ij = each drumming parameter (e.g. bout length, pulses, etc.) i = individual level j = observations within an individual level β 0 i = reference population mean + an error value of mean observations within an individual β 1 j = difference in mean value between population 1 and reference population Chapter 4: Vibratory Drumming 103 β 2 j = difference in mean va lue between population 2 and reference population Second model: Y ij = β 0 i + β 1 j + β 2 j + β 3temp Where: β 3temp = mean temperature – population mean Parameters in the model were deemed significant if the sum of two times the standard error is less than the parameter estimate. The same method was employed to determine the significance of within individual variation. Significant dummy variables for the populations were taken as evidence of differences between the populations. Chapter 4: Vibratory Drumming 104 4.3 Results In to tal, 78 drumming bouts were recorded from 25 male weta. Of these, 9 weta drummed once while 16 weta drummed between 2 and 12 times whilst on the recording device (mean = 4.312, SD = 2.522 ). The number of individual weta recorded at each site ( N ) varied as did the temperature range: Coromandel, N = 12, 1 7.5 - 20.3°C; Manawatu - North, N = 5, 13.9 - 16.8°C; and Manawatu- South, N = 8, 13.9 - 18.8°C. Bout length was significantly correlated with the number of pulses ( r = 0.620, p = 0.001), inter - pulse interval ( r = 0.594, p < 0.01) and the pulse rate ( r = - 0.703, p < 0.001). The pulse rate was also significantly correlated with the inter - pulse interval ( r = - 0.886, p < 0.001). In all cases, there was as much variation within individuals (i.e. at the observat ions level) as there was between individuals in a population after adjusting for temperature. For example, the raw data on bout length within the Coromandel population ranged from 4.1 – 2.6 s, and the greatest variation in bout length within an individual was 3.9 – 2.6 s (Fig. 4.5). For four of the five drumming components measured, adding the effect of temperature to the model significantly decreased the log likelihood score (Chi - square critical value > 5.99, p = 0.05). The model for the number of oscil lations was not improved by adding the temperature effect and so only the initial model was used to compare population means. Chapter 4: Vibratory Drumming 105 Manawatu-SouthManawatu-NorthCoromandel 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 Location Bo ut le ng th ( se c) 10 11 12 13 14 15 16 17 18 19 1 20 21 22 23 24 25 2 3 4 5 6 7 8 9 code Weta Manawatu-SouthManawatu-NorthCoromandel 24 22 20 18 16 14 12 Location Pu ls es Manawatu-SouthManawatu-NorthCoromandel 4.5 4.0 3.5 3.0 2.5 Location O sc ill at io ns Chapter 4: Vibratory Drumming 106 Manawatu-SouthManawatu-NorthCoromandel 0.15 0.14 0.13 0.12 0.11 0.10 0.09 0.08 0.07 0.06 Location In te rp ul se Manawatu-SouthManawatu-NorthCoromandel 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 Location Pu ls er at e (p ul se s/ se c) Figure 4.5. Scatter plots using the raw data not adjusted for temperature, for each drumming component measured at each of the three populations. Symbols indicate individual weta; identical symbols indicate multiple drumming bouts recorded for a single individual. All of the drumming components revealed significant differences between Coromandel and Manawatu - North, and all components except for the inter - pulse interval showed significant differences between Coromandel and Manawatu - South (Table 4.1). Three out of the five drumming components measured revealed no significant difference between the two Manawatu populations. Drumming bouts in Coromandel were longer, contained more pulses, and subsequently had highe r pulse rates than either of the two Manawatu populations, but had fewer oscillations in each pulse and a shorter inter - pulse interval (Table 4.1). As the air temperature increased, the drumming bouts got shorter, Chapter 4: Vibratory Drumming 107 the number of pulses and oscillations decreased, and the inter - pulse interval also decreased. However, the shorter bout lengths and fewer pulses with increased temperature resulted in a faster pulse rate with increasing temperature (Table 4.1). Table 4.1. Mean values (±SE) for the drumming bout components of male ground weta H. pallitarsis from three different populations. Differences between populations and changes in drumming measurement with temperature shown on the right. * p ≤ 0.05; ns p > 0.05. Drumming component Mean (SE) Population comparisons (p values) Temperature increase by 1°C Coromandel (C) Manawatu-North (MN) Manawatu-South (MS) C vs. MN C vs. MS MN vs. MS Bout length (sec) 3.29 (0.087) 2.67 (0.260) 2.69 (0.155) * * ns - 0.278 Pulses 20.56 (0.346) 15.27 (1.037) 16.62 (0. 619) * * ns - 0.591 Oscillations 3.30 (0.054) 3.62 (0.084) 3.69 (0.057) * * ns - Inter - pulse (sec) 0.09 (0.002) 0.11 (0.006) 0.09 (0.004) * ns * - 0.003 Pulse rate (pulses/sec) 6.29 (0.068) 5.55 (0.204) 5.98 (0.122) * * * 0.211 Chapter 4: Vibratory Drumming 109 4.4 Discussion This study was the first to examine vibratory drumming patterns within a species of ground weta, Hemiandrus pallitarsis . Based on the degree of morphometric and genetic variation observed among populations of H. pallitarsis, two predictions were formulated: 1) smaller weta from the two geographically proximate Palmerston North populations would have more similar drumming patterns when compared to the larger weta from Coromandel (see Chapter 2); or 2) drumming patterns between Coromandel and populations north of the Manawatu River would have more similar drumming patterns due to genetic similarities (see Chapter 3). The pattern of population divergence followed that predicted by the morphometric/geographic proximity hypothesis (Fig. 4.6). Larger male weta from C oromandel had significantly longer drumming bouts with higher pulse rates than the smaller weta in both the Palmerston North populations. Three of five drumming components found no significant difference between the two geographically proximate Palmerston North populations. Similarity in drumming behaviour between the two Palmerston North populations contrasts with the genetic results found in Chapter 3 – Coromandel Peninsula and populations on the northern side of the Manawatu River formed a clade that w as sister to populations on the southern side of the river (see Chapter 3, Fig. 3.10 ). Analysis of genetic data also found no evidence for genetic isolation by distance (i.e. clinal variation) among all populations sampled across the North Island (see Cha pter 3, Fig. 3.12). In contrast, the pattern of vibrational drumming seemingly follows a cline, at least across part of the distributional range of H. pallitarsis (although only three populations were sampled in this study). Accordingly, the observed pat terns of gene flow between the Palmerston North and Coromandel populations suggest other Figure 4.6. Morphometric/geographic hypothesis. - smaller weta from geographically proximate populations show more similar drumming patterns. Chapter 4: Vibratory Drumming 110 factors, such as ecological or social factors, may have contributed to the geographical differences in drumming behaviour in H. pallitarsis. These results show that there is significant intraspecific variation in H. pallitarsis drumming calls, a pattern that disagrees with the expected consistency in specific- mate recognition characters across populations due to stabilizing selection (Butlin, 1995; Ferreira & Ferguson, 2002; Paterson, 1985). Finding intraspecific variation in communication signals is consistent with previous research examining population variation in insect songs. For example, Tregenza et al. (2000), found significant differences in the calling song of meadow grasshoppers ( Chorthippus parallelus ) between 8 populations across Europe and the United Kingdom. Gwynne (2004), first examined the vibratory drumming behaviour of ground weta, and found that the length of the drumming bout and mean number of pulses within a drumming bout varied between three species of short ovipositor ground weta. Specifically, H. pallitarsis had longer drumming bouts (4 – 6 sec) and more pulses per bout (18 - 21) than either H. “onokis” (2 – 3.1 s, and 8.7 – 12.5 pulses) or H. “vicinus” (1.1 – 1.8 s, and 7 – 8.6 pulses). Data from the present study suggest these characters also vary among populations within a species, and although the mean number of pulses in each population falls within the range found by Gwynne (15.3, 16.6 and 20.6), the mean duration of the drumming bouts do not (2.7, 2.7 and 3.3). It is possible then, that weta only use certain aspects of vibratory signals for species recognition (e.g. the number of pulses within a bout), and instead use a combination of me chanisms, such as pheromones, for species recognition. For example, Gwynne (2004 ) identified males depositing small brown droplets onto the surface of leaves where mating took place and suggests pheromones may also be involved in mate attraction. Explana tions for population variation in vibratory drumming signals include social factors, such as sexual selection through female preference, or ecological factors, such as habitat structure and composition, or climate differences. Firstly, population variati on as a result of sexual selection can be generated by a bias among females for particular male signal characteristics related to mate quality (Kotiaho et al., 1996; Orci, 2007; Ryan & Keddy - Hector, 1992; Tokarz, 1995; Wilczynski et al., 1995; Wilczynski & Ryan, 1999). In a study examining female preference for male wolf spider (Hygrolycosa rubrofasciata) substrate drumming using manipulated playback signals, female wolf spiders preferred drumming bouts of longer Chapter 4: Vibratory Drumming 111 duration (Parri et al. , 20 02). They sugges ted the length of the drumming bout may be indicative of mate quality in these spiders. Examining levels of intra - and inter- male variation may address the question of whether H. pallitarsis males portray aspects of mate quality in their drumming calls. F or all drumming components measured the variation within individual males was as great as that between males within a population. Since drumming bouts appear to be highly variable within individual males, it is unlikely that information about mate quality is being portrayed in these signals. However, due to the small sample sizes in this study, it is difficult to draw any conclusions about the role of female preference for male drumming signals. Future studies employing female choice experiments on varying male weta signals might reveal which components of the drumming may be involved in mate quality. In addition, it would be interesting to record H. pallitarsis drumming bouts from the Wellington region where it is sympatric with H. bilobatus , which also uses vibratory drumming in courtship displays. The theory of ‘character displacement’ predicts there will be increased selection where species with similar communication systems overlap in space (Brown & Wilson, 1956). However, sample sizes should be si gnificantly large enough to ensure that the variation within males and within the populations is detected. A second explanation to sexual selection that may account for the variation between populations found in this study is the effect of ecological factors, such as habitat characteristics and climate differences. One environmental aspect investigated in this study was ambient air temperature. The data revealed temperature significantly affected all drumming components measured, except for the mean numbe r of oscillations per pulse. Drumming bout duration, number of pulses and inter - pulse interval all decreased with increasing temperature. Pulse rate increased with increasing temperature, a result that parallels that found in other Orthopterans (e.g. Jer usalem crickets, Weissman, 200 1). It is well known that insects are cold blooded, and that their metabolism and activity is determined by their body temperature, which is in turn influenced by the surrounding air temperature (Mellanby, 1939). At low temperatures insect activity decreases. Palmerston North is approximately 400 km south of Coromandel and has a mean air temperature 2.1 °C lower than that recorded in Thames, a town 54 km south of Coromandel Town (New Zealand Meterological Service, 1983). Therefore, it is not surprising that drumming bouts from the colder Palmerston North populations were shorter, contained fewer pulses and subsequently had a lower pulse Chapter 4: Vibratory Drumming 112 rate than what was recorded in the northern Coromandel population. However, in order to accurately assess how temperature affects vibratory drumming in ground weta, future research should aim to record drumming bouts from geographically separated populations at the same temperature range and for a range of temperatures. In conclusion, the pattern of vibratory drumming found among populations of H. pallitarsis matched that predicted by previous morphometric data and geographic proximity - smaller weta from the two geographically proximate Palmerston North populations had more similar drumming patterns when compared to the larger weta from Coromandel. Understanding the possible impact of different selection pressures on this mode of communication requires further research, not only into aspects of the male signals, but also investigating how females receive these signals and the interaction between both senders and receivers when communication underlies mate choice. Further study will help us to understand the factors contributing to the evolution of vibratory communication in New Zealand wet a. Chapter 4: Vibratory Drumming 113 4.5 References Baker, M. C. and Baker, E. M. (1990) Reproductive behaviour of female buntings: isolation mechanisms in a hybridizing pair of species. Evolution, 44: 332- 338. Bateman, P. W., Gilson, L. N. and Ferguson, J. W. H. (2001) Male size and sequential mate preference in the cricket, Gryllus bimaculatus. Animal Behaviour, 6: 631- 637. Bell, P. D. (1980) Transmission of vibrations along plant stems: implications for insect communication. Journal of the New York Entomological Society, 88: 210- 2 16. Bennet - Clark, H. C. (197 1) Acoustics of insect song. Nature, 234: 255- 259. Bennet - Clark, H. C. (1998) Size and scale effects as constraints in insect sound communication. Philosophical Transactions of the Royal Society of London B, 353: 407- 419. Bradbury, J. W. and Vehrencamp, S. L. (1998 ) Principles of Animal Communication. Sinauer Associates, Massachusetts. Brooks, M. W. (1882) Influence of temperature on the chirp rate of the cricket. Popular Science Monthly, 20: 268. Brown, W. D., Wi deman, J., Andrade, M. C. B., Mason, A. C. and Gwynne, D. T. (1996 ) Female choice for an indicator of male size in the song of the black - horned tree cricket, Oecanthus nigricornis (Orthoptera: Gryllidae: Oecanthinae). Evolution, 50: 2400- 2411. Brown, W. L. and Wilson, E. O. (1956) Character displacement. Systematic Zoology, 5: 49- 64. Butlin, R. (1987) Speciation by reinforcement. Trends in Ecology & Evolution, 2: 8- 13. Chapter 4: Vibratory Drumming 114 Butlin, R. (1995) Genetic variation in mating signals and responses. In: Speciation and the recognition concept (eds. M. D. Lambert and G. H. Spencer), pp. 327- 366. The John Hopkins University Press, Baltimore, Maryland. Butts, C. A. (1983) The biologies of two species of weta endemic to the Snares island Zealandosandrus subantarcticus Salmon (Orthoptera: Stenopelmatidae) and Insulanoplectron spinosum Richards (Orthoptera: Rhaphidophoridae). B.Sc. Zoology Thesis, University of Canterbury, Christchurch, New Zealand. Cary, P. R. L. (1981) The biology of the weta Zealandosandrus gracilis (Orthoptera: Stenopelmatidae) from the Cass region. M.Sc. Thesis, University of Canterbury, Christchurch, New Zealand. Claridge, M. F. (2006) Insect sounds and communication - an introduction. In: Insect sounds and communication: physiology, behaviour, ecology and evolution (eds. S. Drosopoulos and M. F. Claridge), pp. 3- 10. Boca Ralton, Fla, CRC. Cocroft, R. B. and Rodriguez, R. L. (2005 ) The behavioural ecology of insect vibrational communication. Bioscience, 55: 323- 334. Cokl, A. and Virant - Doberlet, M. (2003) C ommunication with substrate- borne signals in small plant- dwelling insects. Annual Review of Entomology, 48: 29- 50. Darwin, C. (1859) On the origin of species by means of natural selection. Murray, London. Darwin, C. (1871) The descent of man and selection in relation to sex. Murray, London. De Luca, P. A. and Morris, G. K. (1998) Courtship communication in meadow katydids: female preference for large male vibrations. Behaviour, 135: 777- 793. Dobzhansky, T. (1951) Genetics and the origin of species. Columbia University Press, New York. Dolbear, A. E. (1897) The cricket as a thermometer. American Naturalist, 31: 970- 971. Chapter 4: Vibratory Drumming 115 Eberhard, W. G. (1996) Female control: sexual selection by cryptic female choice. Princeton University Press, Princeton, New Jersey. Ewing, A . W. (1989) Arthropod bioacoustics: neurobiology and behaviour. Edinburgh University Press, Edinburgh. Ferreira, M. and Ferguson, J. W. H. (2002) Geographic variation in the calling song of the field cricket Gryllus bimaculatus (Orthoptera: Gryllidae) and its relevance to mate recognition and mate choice. Journal of Zoology, 257: 163- 170. Frings, H. and Frings, M. (1957) The effects of temperature on chirp - rate of male conehead grasshoppers, Neoconocephalus ensiger . Journal of Experimental Zoology, 134: 411- 425. Gerhardt, H. C. and Huber, F. (2002) Acoustic communication in insects and anurans. Common problems and diverse solutions. University of Chicago Press, Chicago, IL. Gwynne, D. T. (1982) Mate selection by female katydids (Orthoptera: Tettigoniidae, Conocephalus nigropleurum). Animal Behaviour, 30: 734- 738. Gwynne, D. T. and Bailey, W. J. (1988) Mating system, mate choice and ultrasonic calling in a zaprochiline katydid (Orthoptera: Tettigoniidae). Behaviour, 105: 202- 223. Gwynne, D. T. (200 4) Reproduc tive behaviour of ground weta (Orthoptera: Anostostomatidae): drumming behaviour, nuptial feeding, post - copulatory guarding and maternal care. Journal of the Kansas Entomological Society, 77: 414- 428. Henry, C. S. (1993 ) Chrysoperla johnsoni (Neuroptera: C hrysopidae): acoustic evidence for full species status. Annals of the Entomological Society of America, 86: 14- 25. Henry, C. S. (1994) Singing and cryptic speciation in insects. Trends in Ecology and Evolution, 9: 388- 392. Chapter 4: Vibratory Drumming 116 Henry, C. S., Brooks, S. J., Johnson, J. B. and Duelli, P. (1999) Revised concept of Chrysoperia mediterranea (Holzel), a green lacewing associated with conifers: courtship songs across 2800 kilometers of Europe (Neuroptera: Chrysopidae). Systematic Entomology, 24: 335- 350. Kotiaho, J., A latalo, R. V., Mappes, J. and Parri, S. (1996) Sexual selection in a wolf spider: male drumming activity, body size, and viability. Evolution, 50: 1977- 1981. Lande, R. (1982) Rapid origin of sexual isolation and character divergence in a cline. Evolution, 36: 213- 223. Markl, H. (1983) Vibrational communication. In: Neuroethology and Behavioural Physiology. (eds. F. Huber and H. Markl), pp. 332 - 353. Springer - Verlag, Berlin. Mellanby, K. (1939 ) Low temperature and insect activity. Proceedings of the Royal Society of London. Series B, Biological Sciences, 127: 473- 487. Morris, G. K. (1980) Calling display and mating behaviour of Copiphora rhinoceros Pictet (Orthoptera: Tettigoniidae). Animal Behaviour, 28: 42- 51. Morris, G. K., Mason, A. C. and Wall, P. (1994) High ultrasonic and tremulation signals in neotropical katydids (Orthoptera: Tettigoniidae). Journal of Zoology (London), 233: 129- 163. Mousseau, T. A. (1997) Ectotherms follow the converse to Bergmann’s rule. Evolution, 51: 630- 632. Narins, P. M. and Smit h, S. L. (1986) Clinal variation in anuran advertisement calls: basis for acoustic isolation? Behavioral Ecology and Sociobiology, 19: 135- 141. Nelson, D. A. and Marler, P. (1993) Innate recognition of song in white - crowned sparrows: A role in selective vo cal learning? Animal Behaviour, 46: 806- 808. Chapter 4: Vibratory Drumming 117 New Zealand Meteorological Service (1983) Summaries of climatological observations to 1980. New Zealand Meteorological Service Miscellaneous Publication, 177. Orci, K. M. (2007 ) Female preferences for male song characters in the bush- cricket Isophya camptoxypha ( Orthoptera, Tettigonioidea). Journal of Insect Behaviour, 20: 503- 513. Park, O. (1949) Application of the converse Bergmann principle to the carabid beetle, Dicaelus purpuratus. Physiological Zoology, 22: 359- 372. Parri, S., Alatalo, R. V., Kotiaho, J. S., Mappes, J. and Rivero, A. (2002) Sexual selection in the wolf spider Hygrolycosa rubrofasciata: female preference for drum duration and pulse rate. Behavioral Ecology, 13: 615- 621. Partridge, L. and Coyn e, J. A. (1997) Bergmann’s rule in ectotherms: is it adaptive? Evolution, 51: 632- 635. Paterson, H. E. H. (1985 ) The recognition concept of species. In: Species and speciation (ed. E. S. Vrba), pp. 21- 29. Transvaal Museum Monograph No. 4, Pretoria: Transvall Museum. Pfannenstiel, R. S., Hunt, R. E. and Yeargan, K. V. (1995) Orientation of a Hemipteran predator to vibrations produced by feeding caterpillars. Journal of Insect Behaviour, 8: 1- 9. Prestwich, K. N. and Walker, T. J. (1981) Energetics of singing in crickets: effect of temperature in three trilling species (Orthoptera: Gryllidae). Journal of Comparative Physiology, 143B: 199- 212. Prohl, H., Hagemann, S., Karsch, J. and Hobel, G. (2007) Geographic variation in male sexual signals in strawberry poison frogs (Dendrobates pumilio). Ethology, 113: 825- 837. Rasbash, J., Browne, W., Healy, M., Cameron, B. and Charlton, C. (2005) M LwiN, 2.02. Multilevel Models Project, Institute of Education. Chapter 4: Vibratory Drumming 118 Ratcliffe, L. M. and Grant, P. R. (1985) Species recognition in D arwin's Finches ( Geospiza, Gould). III. Male response to play - back of different song types, dialects and heterospecific songs. Animal Behaviour, 33: 290- 307. Ryan, M. A., Cokl, A. and Walter, G. H. (1 996a) Differences in vibratory sound communication between a Slovenian and an Australian population of Nezara viridula (L.) (Heteroptera: Pentatomidae). Behavioural Processes, 36: 183- 193. Ryan, M. J. and Brenowitz, E. A. (1985) The role of body size, phylogeny and ambient noise in the evolution of bird song. American Naturalist, 126: 87- 100. Ryan, M. J. and Wilczynski, W. (1991) Evolution of intraspecific variation in the advertisement call of a cricket frog ( Acris crepitans , Hylidae). Biological Journal of the Linnean Society, 44: 249- 271. Ryan, M. J. and Ke ddy- Hector, A. (1992) Directional patterns of female mate choice and the role of sensory biases. The American Naturalist, 139: S4. Ryan, M. J., Rand, A. S. and Weigt, L. A. (19 96b) Allozyme and advertisement call variation in the tungara frog, Physalaemus pustulosus . Evolution, 50: 2435- 2453. Sanborn, A. F. (2006) Acoustic signals and temperature. In: Insect sounds and communication: physiology, behaviour, ecology and evolution (eds. S. Drosopoulos and M. F. Clardge). Boca Ralton, Fla, CRC. Searcy, W. A., P odos, J., Peters, S. and Nowicki, S. (1995) Discrimination of song types and variants in song sparrows. Animal Behaviour, 49: 1219- 1226. Seddon, N. (2005) Ecological adaptation and species recognition drives vocal evolution in Neotropical suboscine birds. Evolution, 59: 200- 215. Simmons, L. W. (1992) Sexual selection and body size in a natural population of the field cricket, Gryllus campestris (L.). Journal of Orthoptera Research, 1: 12- 13. Chapter 4: Vibratory Drumming 119 Skovmand, D. and Pedersen, S. B. (1983) Song recognition and song pattern in a short - horned grasshopper: a parameter carrying specific behavioral information. Journal of Comparative Physiology, 153A: 393- 4 01. Stewart, K. W., Bottorff, R. L., Knight, A. W. and Moring, J. B. (1991 ) Drumming of four North American euhologna than stonefly species, and a new complex signal pattern in Nemoura spiniloba jewett (Plecoptera : Nemouridae). Annals of the Entomological Society of America, 84: 201- 206. Tokarz, R. R. (1995) Mate choice in lizards: a review. Herpetological Monographs, 9: 17- 40. Toms, R. B., Ferguson, J. W. H. and Becker, S. (1993) Relationship between body temperature in stridulating male crickets (Orthoptera, Gryllidae). South African Journal of Zoology, 28: 71- 73. Tregenza, T., Pritchard, V. L. and Butlin, R. K. (2000) Patterns of trait divergence between populations of the meadow grasshopper, Chorthippus parallelus . Evolution, 54: 574- 585. Tubaro, P. L. and Lijtmaer, D. A. (2006) Environmental correlates of song structure in forest grosbeaks and saltators. The Condor, 108: 120- 129. Van Voorhies, W. A. (1997) On the adaptive nature of Bergmann size clines: a reply to Mousseau, Partridge and Coyne. Evolution, 51: 635- 640. Virant - Doberlet, M. and Cokl, A. (2004) Vibrational communication in insects. Neotropical Entomology, 33: 121- 134. von Helverson, O. and von Helverson, D. (1994) Forces driving coevolution of song and song recognition in grasshoppers. In: Neural Basis of Behavioural Adaptations, Progress in Zoology (eds. K. Schildberger and N. Elsner), pp. 253 - 284. Gustav Fischer Verlag, Stuttgart. Chapter 4: Vibratory Drumming 120 Walker, T. J., Whitesell, J. J. and Alexander, R. D. (1973) The robust conehead: two widespread sibling species (Orthoptera: Tettigoniidae: Neoconocephalus «robustus»). Ohio Journal of Science, 73: 321- 33 0. Weissman, D. B. (2001) Communication and reproductive behaviour in North American Jerusalem Crickets ( Stenopelmatus ) (Orthoptera: Stenopelmatidae). In: The biology of wetas, king crickets and their allies (ed. L. H. Field), pp. 351–373. CABI, Wallingford, UK. West - Eberhard, M. J. (1983) Sexual selection, social competition, and speciation. Quarterly Review of Biology, 58: 155- 183. Wilczynski, W., Rand, S. A. and Ryan, M. J. (1995) The processing of spectral cues by the call analysis system of the túngara frog, Physalaemus pustulosus. Animal Behaviour, 49: 911- 929. Wilczynski, W. and Ryan, M. J. (1999) Geographic variation in animal communication systems. In: Geographic variation in behaviour: perspectives on evolutionary mechanisms (eds. S. A. Foster and J. A. Endler), pp. 234- 261. Oxford University Press, New York. Zimmerman, B. L. (1983 ) A comparison of structural features of calls of open and forest habitat frog species in the central Amazon. Herpetologica, 39: 235- 246. Thesis Conclusions Adult female Hemiandrus pallitarsis on the forest floor on Cuvier Island. Photo: E.Chappell 121 Thesis Conclusions For the last fifty years there has been substantial debate concerning the defini tion of a species, and there have been at least 24 different species concepts named (Mayden, 1997). Much of argument seems to surround the difference between the concept of a species and actually inferring boundaries and numbers of species, that is, species delimitation (De Queiroz, 2007). Although there are attempts to form a single unifying species concept, several authors have stressed the importance of examining multiple different characters in order to gain evidence for the existence of separate species (De Queiroz, 2007; Roe & Sperling, 2007). Accordingly, the aim of this thesis was to document the degree of morphological, genetic and behavioural variation within an endemic species of ground weta, Hemiandrus pallitarsis , in North Island, New Zealand. The results from all three characters provide no evidence of a species complex within H. pallitarsis, but suggest that it is morphologically, genetically and behaviourally variable across the North Island. In documenting the degree of variation within this species some interesting patterns were found. Specifically, it was identified that female weta are significantly larger than males, and weta in Palmerston North tend to be smaller than weta in northern Coromandel populations, possibly due to differences in temperature and season length between the two locations. Phylogenetic and phylogeographic analysis of H. pallitarsis populations across the North Island revealed substantial genetic structuring, large genetic distances, and a possible southern origin for this species. In fact, the results suggest H. pallitarsis populations have undergone range expansion in a south to north pattern, possibly following the pattern of land connection and uplift in the lower North Island during the late Pliocene – mid Pleistocene. Additionally, the application of the commonly used molecular clock calibration rate (1.4 – 2.6% sequence divergence per million years) to estimate possible divergence times of lineages within H. pallitarsis , was found to be problematic. This study highlighted the discrepancy in using a rate based on phylogenetic studies for a population- level study and found that divergence times tend to be overestimated when the invertebrate molecular clock is applied to H. pallitarsis 122 populations. The pre- copulatory drumming behaviour of male H. pallitarsis wetas was recorded in two adjacent populations and one distant population. Variation in the patterns of drumming supported the morphology/geographic proximity hypothesis – smaller weta in the two geographically proximate Palmerston North populations had more similar drumming behaviour patterns than larger weta from Coromandel – and did not match the patterns of genetic isolation previously found. The different topics examined in this thesis illustrate the importance of using multiple characters coupled with geographic information when investigating the species status of an organism. Moreover, this thesis was able to provide some baseline data on the level of morphometric, genetic and behavioural variation that can be found within a single terrestrial invertebrate species in North Island, New Zealand. 123 References De Queiroz, K. (2007) Species concepts and species delimitation. Systematic Biology, 56: 879- 886. Mayden, R. L. (1997) A hierarchy of species concepts: the denouement in the saga of the species problem. In: Species: The units of biodiversity (eds. M. F. Claridge, H. A. Dawah and M. R. Wilson), pp. 381- 424. Chapman and Hall, London. Roe, A. D. and Sperling, F. A. H. (2007 ) Patterns of evolution of mitochondrial cytochrome c oxidase I and II DNA and implications for DNA barcoding . Molecular Phylogenetics and Evolution, 44: 325- 345. 124 Appendix A. Correlation matrix for nine morphometric variables produced using 91 weta measurements. Values in bold show strongly correlated characters ( > 90% ). Pronotum length Protibia length Mes otibia length Metatibia length Metafemur length Head width Head length Fastigium width Body length Pronotum length 1 Protibia length 0.8213 1 Mesotibia length 0.8099 0.9574 1 Metatibia length 0.8201 0.9264 0.9316 1 Metafemur length 0.8888 0.9229 0.9220 0.9297 1 Head width 0.8954 0.73 44 0.72 48 0.73 92 0.82 45 1 Head length 0.8996 0.71 01 0.70 45 0.72 09 0.79 87 0.9386 1 Fastigium width 0.7425 0.52 96 0.51 17 0.49 41 0.61 87 0.84 18 0.81 19 1 Body length 0.7962 0.59 45 0.60 92 0.61 89 0.69 98 0.81 32 0.82 68 0.72 22 1 125 Appendix B. One sample t - test results produced using SPSS version 4.0 (SPSS, Chicago, Ill., USA) for eleven morphometric traits. Significant values at the p < 0.05 level are marked * Test Value = 0 95% Confidence Interval of the Difference Characters t df Sig. (2 - tailed) Mean Difference Lower Upper Pr 1.541 14 .146 .026 00 - .01 0 2 .062 2 Mf 1.636 14 .124 .040 67 - .01 2 6 .094 0 Fa - .800 14 .437 - .01 6 0 0 - .05 8 9 .026 9 BL .341 14 .738 .032 67 - .17 2 9 .23 8 2 Hw .602 14 .557 .018 00 - .04 6 2 .082 2 Hl .359 14 .725 .012 00 - .05 9 7 .083 7 Sg - .516 14 .614 - .01 3 3 3 - .06 8 8 .042 1 MxSg 1.024 14 .323 .033 33 - .03 6 5 .103 2 MnSg - 3.400 7 .011 * - .05 2 5 0 - .08 9 0 - .01 6 0 LAO - 2.16 9 5 .082 - .04 0 0 0 - .08 7 4 .007 4 WAO - 2.28 2 5 .07 1 - .06 5 0 0 - .13 8 2 .008 2 Appendix C. Principal components analysis results for male H. pallitarsis. PCA axis 1 2 3 4 Eigenvalue 5.4636 0.8931 0.5682 0.3959 Percent of total 0.683 0.112 0.071 0.049 Cumulative percent 0.683 0.795 0.866 0.915 Eigenvectors (component loadings) Pronotum 0.393 0.001 - 0.261 - 0.130 Metafemur 0.365 0.170 - 0.386 - 0.067 Fastigium 0.267 - 0.612 0.643 - 0.161 Body length 0.379 - 0.072 - 0.015 0.338 Head width 0.384 - 0.248 - 0.252 - 0.136 Head length 0.405 - 0.150 - 0.129 - 0.098 Subgenital plate 0.275 0.628 0.421 - 0.556 Width subgenital plate 0.333 0.336 0.334 0.708 126 Appendix D. Principal components analysis results for female H. pallitarsis. Appendix E. Agarose gel showing amplified DNA products. Arrow indicates product with two bands, where one is potentially a Numt sequence. PCA ax is 1 2 3 4 Eigenvalue 5.1772 1.0752 0.6577 0.3901 Percent of total 0.647 0.134 0.082 0.049 Cumulative percent 0.647 0.782 0.864 0.913 Eigenvectors (component loadings) Pronotum 0.403 - 0.252 - 0.033 - 0.180 Metafemur 0.402 - 0.046 0.144 - 0.289 Fas tigium 0.339 - 0.006 - 0.563 0.580 Body length 0.291 - 0.292 0.749 0.375 Head width 0.407 - 0.141 - 0.224 - 0.087 Head length 0.410 - 0.177 - 0.113 - 0.167 Subgenital plate 0.256 0.658 0.178 0.423 Width subgenital plate 0.278 0.604 0.073 - 0.439 Appendix F. Number of individuals sequenced, number of haplotypes found, and number of each haplotype found at each locat ion. Location No. of individuals No. of haplotypes Clade I Haplotype ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Great Barrier Island 1 1 1 Little Barrier Island 2 2 1 1 Moehau 12 5 2 2 6 1 1 Cuvier Island 9 5 6 1 1 1 Taiharuru 5 1 5 Middle Island 9 4 6 1 1 1 Coromandel Town 8 5 2 1 1 2 2 Ruamahuanui Island 8 1 1 Kauaeranga Valley 8 7 1 1 1 1 1 1 Cambridge 1 1 Taupo 2 1 Lake Waikaremoana 3 3 Taranaki 4 2 Wanganui 1 1 Palmerston North 9 8 Pohangina Valley 1 1 Wellington 9 8 Total 92 56 Appendix F. cont… Clade I cont . Clade II Clade III Haplotype ID 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 Great Barrier Island Little Barrier Island Moehau Cuvier Island Taiharuru Middle Island Coromandel Town Ruamahuanui Island Kauaeranga Valley 2 Cambridge 1 Taupo 2 Lake Waikaremoana 1 1 1 Taranaki 1 3 Wanganui 1 Palmerston North 1 1 1 1 1 2 1 1 Pohangina Valley 1 Wellington 2 1 1 1 1 1 1 1 Appendix G. Pairwise genetic distances based on COI sequence data. TVM + I + G above diagonal, observed distanc e below diagonal. Individual code names, sampling locations and haplotype codes are shown. Bold coloured boxes indicate clades in ML analysis (Fig. 3.10 ). Code Location Haplotype 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 GB Great Barrier Island 1 I 0.05 0.06 0.04 0.04 0.04 0.04 0.05 0.07 0.08 0.08 0.07 0.04 0.05 0.05 0.05 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.05 0.04 0.04 0.05 GW226 Little Barrier Island 2 0.04 0.00 0.06 0.06 0.07 0.07 0.08 0.08 0.08 0.08 0.07 0.06 0.07 0.08 0.07 0.07 0.06 0.06 0.06 0.06 0.06 0.06 0.05 0.05 0.05 0.06 GW227 Little Barrier Island 3 0.04 0.00 0.07 0.06 0.07 0.07 0.08 0.08 0.08 0.08 0.07 0.06 0.08 0.08 0.07 0.08 0.06 0.06 0.06 0.06 0.06 0.06 0.05 0.05 0.05 0.06 CO144 Moehau 4 0.03 0.05 0.05 0.00 0.01 0.01 0.01 0.06 0.07 0.06 0.06 0.02 0.04 0.05 0.04 0.04 0.02 0.01 0.01 0.02 0.01 0.04 0.04 0.03 0.03 0.04 CO145 Moehau 5 0.03 0.04 0.05 0.00 0.01 0.01 0.01 0.06 0.06 0.06 0.06 0.02 0.04 0.05 0.04 0.05 0.02 0.01 0.01 0.02 0.02 0.04 0.04 0.03 0.03 0.04 CO148 Moehau 6 0.03 0.05 0.05 0.01 0.01 0.00 0.00 0.06 0.07 0.06 0.06 0.02 0.05 0.05 0.05 0.05 0.02 0.02 0.02 0.02 0.02 0.04 0.04 0.04 0.04 0.05 CO71 Moehau 7 0.04 0.05 0.05 0.01 0.01 0.00 0.00 0.06 0.07 0.06 0.06 0.02 0.05 0.05 0.04 0.05 0.02 0.01 0.02 0.02 0.02 0.04 0.04 0.04 0.04 0.04 CO73 Moehau 8 0.04 0.05 0.05 0.01 0.01 0.00 0.00 0.06 0.07 0.06 0.06 0.03 0.05 0.05 0.05 0.05 0.02 0.02 0.02 0.02 0.02 0.04 0.04 0.04 0.04 0.05 CU76 Cuvier Island 9 0.05 0.06 0.06 0.04 0.04 0.04 0.04 0.04 0.01 0.00 0.00 0.07 0.07 0.07 0.06 0.07 0.06 0.05 0.05 0.05 0.05 0.06 0.05 0.05 0.05 0.06 CU77 Cuvier Island 10 0.06 0.06 0.06 0.05 0.05 0.05 0.05 0.05 0.01 0.01 0.00 0.07 0.08 0.08 0.07 0.08 0.07 0.06 0.06 0.06 0.06 0.07 0.06 0.06 0.06 0.07 CU82 Cuvier Island 11 0.06 0.06 0.06 0.05 0.05 0.05 0.05 0.05 0.00 0.01 0.00 0.07 0.07 0.07 0.07 0.07 0.06 0.06 0.06 0.06 0.06 0.07 0.06 0.06 0.06 0.06 CU86 Cuvier Island 12 0.05 0.05 0.05 0.04 0.04 0.05 0.04 0.05 0.00 0.00 0.00 0.06 0.07 0.07 0.07 0.07 0.06 0.05 0.05 0.06 0.05 0.06 0.06 0.06 0.06 0.06 CO67 Taiharuru 13 0.03 0.05 0.05 0.02 0.02 0.02 0.02 0.02 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.02 0.01 0.01 0.02 0.02 0.04 0.04 0.04 0.04 0.04 MI100 Middle Island 14 0.04 0.05 0.06 0.03 0.03 0.04 0.04 0.04 0.05 0.05 0.05 0.05 0.04 0.00 0.00 0.00 0.05 0.04 0.04 0.04 0.04 0.03 0.03 0.02 0.03 0.03 MI101 Middle Island 15 0.04 0.06 0.06 0.04 0.04 0.04 0.04 0.04 0.05 0.05 0.05 0.05 0.04 0.00 0.00 0.00 0.05 0.04 0.05 0.05 0.05 0.04 0.03 0.03 0.03 0.03 MI98 Middle Island 16 0.04 0.05 0.05 0.03 0.03 0.04 0.03 0.04 0.05 0.05 0.05 0.05 0.04 0.00 0.00 0.00 0.04 0.03 0.04 0.04 0.04 0.03 0.03 0.02 0.02 0.03 MI99 Middle Island 17 0.04 0.05 0.06 0.03 0.04 0.04 0.04 0.04 0.05 0.06 0.05 0.05 0.04 0.00 0.00 0.00 0.05 0.04 0.04 0.05 0.04 0.03 0.03 0.02 0.03 0.03 CO121 Coromandel Town 18 0.04 0.05 0.05 0.02 0.02 0.02 0.02 0.02 0.04 0.05 0.05 0.05 0.02 0.04 0.04 0.03 0.04 0.01 0.01 0.00 0.01 0.03 0.04 0.04 0.03 0.04 CO127 Coromandel Town 19 0.03 0.04 0.04 0.01 0.01 0.01 0.01 0.01 0.04 0.05 0.04 0.04 0.01 0.03 0.03 0.03 0.03 0.01 0.01 0.00 0.00 0.03 0.03 0.03 0.02 0.03 CO129 Coromandel Town 20 0.03 0.04 0.04 0.01 0.01 0.02 0.02 0.02 0.04 0.04 0.04 0.04 0.01 0.03 0.04 0.03 0.03 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.03 0.03 CO141 Coromandel Town 21 0.03 0.04 0.04 0.01 0.02 0.02 0.02 0.02 0.04 0.05 0.05 0.04 0.02 0.03 0.04 0.03 0.04 0.00 0.00 0.01 0.00 0.03 0.03 0.03 0.03 0.04 CO37 Coromandel Town 22 0.04 0.05 0.05 0.01 0.01 0.02 0.02 0.02 0.04 0.05 0.04 0.04 0.02 0.03 0.04 0.03 0.03 0.01 0.00 0.01 0.00 0.03 0.03 0.03 0.03 0.03 AL112 Ruamahuanui Island 23 0.03 0.04 0.04 0.03 0.03 0.03 0.03 0.03 0.05 0.05 0.05 0.05 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.02 0.02 0.01 0.02 KA119 Kauaeranga Valley 24 0.04 0.04 0.04 0.03 0.03 0.03 0.03 0.03 0.04 0.05 0.05 0.04 0.03 0.02 0.03 0.02 0.03 0.03 0.02 0.03 0.03 0.03 0.02 0.00 0.01 0.01 KA120 Kauaeranga Valley 25 0.04 0.04 0.04 0.03 0.03 0.03 0.03 0.03 0.04 0.05 0.05 0.04 0.03 0.02 0.02 0.02 0.02 0.03 0.02 0.03 0.03 0.03 0.01 0.00 0.01 0.00 KA87 Kauaeranga Valley 26 0.03 0.04 0.04 0.03 0.03 0.03 0.03 0.03 0.04 0.05 0.05 0.04 0.03 0.02 0.03 0.02 0.02 0.03 0.02 0.02 0.03 0.02 0.01 0.01 0.01 0.01 KA88 Kauaeranga Valley 27 0.04 0.04 0.04 0.03 0.03 0.04 0.03 0.04 0.04 0.05 0.05 0.05 0.03 0.02 0.03 0.02 0.03 0.03 0.03 0.03 0.03 0.03 0.02 0.01 0.00 0.01 KA89 Kauaeranga Valley 28 0.03 0.05 0.05 0.03 0.03 0.03 0.03 0.03 0.04 0.04 0.04 0.04 0.03 0.02 0.02 0.02 0.02 0.03 0.03 0.03 0.03 0.03 0.02 0.01 0.01 0.01 0.01 KA90 Kauaeranga Valley 29 0.03 0.05 0.05 0.03 0.03 0.03 0.03 0.03 0.05 0.05 0.05 0.05 0.03 0.03 0.03 0.02 0.03 0.03 0.02 0.03 0.03 0.03 0.01 0.01 0.01 0.00 0.01 KA91 Kauaeranga Valley 30 0.04 0.04 0.04 0.03 0.03 0.03 0.03 0.03 0.04 0.05 0.05 0.04 0.03 0.02 0.03 0.02 0.02 0.03 0.02 0.03 0.03 0.03 0.01 0.00 0.00 0.01 0.00 GW230 Lake Waikaremoana 31 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.06 0.05 0.06 0.05 0.05 0.05 0.04 0.05 0.05 0.05 0.05 0.05 0.05 GW231 Lake Waikaremoana 32 0.05 0.05 0.06 0.05 0.05 0.05 0.06 0.06 0.05 0.05 0.05 0.05 0.06 0.05 0.06 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 LW189 Lake Waikaremoana 33 0.05 0.05 0.06 0.05 0.05 0.05 0.06 0.06 0.05 0.05 0.05 0.05 0.06 0.05 0.06 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 GW190 Wanganui 34 0.05 0.05 0.05 0.05 0.06 0.05 0.05 0.06 0.03 0.04 0.03 0.03 0.06 0.06 0.06 0.05 0.06 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 GW92A Palmerston North 35 0.04 0.05 0.05 0.04 0.04 0.05 0.05 0.05 0.04 0.04 0.04 0.04 0.05 0.04 0.04 0.04 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 PN178 Palmerston North 36 0.04 0.05 0.05 0.05 0.04 0.05 0.05 0.05 0.04 0.05 0.05 0.04 0.05 0.05 0.05 0.04 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 PN22 Palmerston North 37 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04 0.05 0.05 0.04 0.05 0.05 0.05 0.05 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 PN23 Palmerston North 38 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04 0.05 0.05 0.05 0.04 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 GW87 Pohangina Valley 39 0.04 0.05 0.06 0.04 0.04 0.05 0.05 0.05 0.04 0.05 0.05 0.04 0.05 0.05 0.05 0.05 0.06 0.05 0.05 0.04 0.05 0.05 0.05 0.05 0.05 0.05 0.05 GW171 Cambridge 40 0.06 0.08 0.08 0.07 0.07 0.07 0.07 0.07 0.08 0.08 0.08 0.07 0.07 0.06 0.07 0.06 0.06 0.08 0.07 0.07 0.08 0.08 0.08 0.07 0.07 0.07 0.07 GW200 Taupo 41 0.06 0.08 0.09 0.07 0.07 0.07 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.07 0.07 0.07 0.07 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.07 0.08 GW235 Taranaki 42 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.06 0.07 0.07 0.07 0.07 0.06 0.05 0.05 0.05 0.05 0.07 0.06 0.06 0.06 0.07 0.07 0.06 0.06 0.06 0.06 TK187 Taranaki 43 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.08 0.08 0.08 0.07 0.07 0.06 0.06 0.06 0.06 0.07 0.07 0.07 0.07 0.07 0.08 0.07 0.07 0.07 0.07 GW232 Palmerston North 44 0.07 0.08 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.08 0.07 0.07 0.08 0.08 0.07 0.07 0.07 0.07 0.07 PN174 Palmerston North 45 0.07 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.08 0.07 0.07 0.08 0.08 0.07 0.07 0.07 0.07 0.07 PN176 Palmerston North 46 0.07 0.07 0.08 0.07 0.07 0.08 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.08 0.08 0.07 0.08 0.08 0.07 0.07 0.07 0.07 0.07 PN177 Palmerston North 47 0.06 0.07 0.07 0.07 0.06 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.07 0.06 0.06 0.06 0.06 0.07 0.07 0.07 0.07 0.07 0.06 0.07 0.07 0.07 0.07 GW80 Wellington 48 0.08 0.08 0.08 0.07 0.07 0.07 0.08 0.08 0.07 0.06 0.07 0.07 0.08 0.08 0.08 0.07 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.07 0.08 GW82 Wellington 49 0.07 0.08 0.08 0.07 0.07 0.08 0.08 0.08 0.07 0.07 0.07 0.07 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.08 GW89A Wellington 50 0.08 0.08 0.07 0.07 0.08 0.07 0.07 0.07 0.06 0.06 0.06 0.06 0.08 0.07 0.07 0.07 0.07 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.07 0.07 0.08 GW89B Wellington 51 0.08 0.08 0.07 0.07 0.07 0.06 0.06 0.07 0.06 0.06 0.06 0.06 0.08 0.07 0.07 0.07 0.07 0.08 0.07 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.08 WN181 Wellington 52 0.07 0.07 0.07 0.07 0.07 0.06 0.07 0.07 0.06 0.06 0.07 0.06 0.08 0.07 0.07 0.07 0.07 0.08 0.08 0.07 0.08 0.08 0.08 0.08 0.08 0.07 0.08 WN182 Wellington 53 0.08 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.06 0.07 0.06 0.08 0.07 0.07 0.07 0.08 0.08 0.08 0.07 0.08 0.08 0.08 0.08 0.08 0.08 0.08 WN183 Wellington 54 0.08 0.08 0.07 0.08 0.07 0.07 0.07 0.07 0.06 0.06 0.07 0.06 0.08 0.08 0.08 0.07 0.08 0.08 0.08 0.07 0.08 0.08 0.08 0.08 0.08 0.08 0.08 GW176 Wellington 55 0.07 0.08 0.08 0.07 0.07 0.07 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.08 0.07 0.07 0.08 Appendix G. cont… Code Location Haplotype 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 GB Great Barrier Island 1 0.04 0.04 0.04 0.07 0.07 0.07 0.07 0.05 0.05 0.06 0.06 0.06 0.10 0.10 0.08 0.10 0.11 0.11 0.11 0.10 0.12 0.12 0.12 0.12 0.12 0.12 0.12 0.12 GW226 Little Barrier Island 2 0.06 0.06 0.05 0.07 0.08 0.07 0.07 0.06 0.07 0.07 0.07 0.07 0.14 0.16 0.09 0.11 0.14 0.12 0.13 0.11 0.13 0.13 0.12 0.12 0.12 0.12 0.13 0.13 GW227 Little Barrier Island 3 0.06 0.06 0.05 0.07 0.08 0.08 0.07 0.06 0.06 0.06 0.07 0.08 0.15 0.16 0.10 0.11 0.15 0.13 0.14 0.12 0.14 0.14 0.11 0.11 0.11 0.12 0.12 0.13 CO144 Moehau 4 0.04 0.04 0.04 0.07 0.08 0.08 0.08 0.06 0.06 0.06 0.06 0.06 0.11 0.12 0.09 0.11 0.11 0.11 0.11 0.10 0.11 0.11 0.12 0.11 0.11 0.12 0.12 0.11 CO145 Moehau 5 0.04 0.04 0.04 0.07 0.07 0.07 0.08 0.05 0.06 0.06 0.06 0.06 0.11 0.12 0.09 0.10 0.11 0.11 0.11 0.10 0.11 0.11 0.12 0.11 0.11 0.11 0.12 0.10 CO148 Moehau 6 0.04 0.04 0.04 0.07 0.08 0.08 0.07 0.06 0.06 0.06 0.07 0.06 0.11 0.13 0.09 0.11 0.12 0.12 0.13 0.11 0.11 0.12 0.10 0.09 0.10 0.10 0.11 0.11 CO71 Moehau 7 0.04 0.04 0.04 0.07 0.08 0.08 0.08 0.06 0.06 0.06 0.06 0.07 0.12 0.13 0.09 0.11 0.12 0.12 0.12 0.11 0.12 0.12 0.10 0.10 0.10 0.10 0.11 0.11 CO73 Moehau 8 0.04 0.04 0.04 0.07 0.09 0.09 0.08 0.06 0.06 0.06 0.07 0.07 0.12 0.13 0.10 0.12 0.12 0.12 0.13 0.11 0.12 0.12 0.11 0.10 0.10 0.11 0.11 0.12 CU76 Cuvier Island 9 0.05 0.06 0.05 0.07 0.07 0.07 0.04 0.05 0.05 0.06 0.06 0.06 0.13 0.14 0.11 0.13 0.11 0.11 0.12 0.11 0.10 0.11 0.09 0.09 0.10 0.10 0.10 0.11 CU77 Cuvier Island 10 0.06 0.07 0.06 0.07 0.08 0.08 0.04 0.06 0.06 0.06 0.06 0.06 0.14 0.15 0.11 0.13 0.12 0.12 0.13 0.11 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 CU82 Cuvier Island 11 0.05 0.06 0.06 0.07 0.08 0.08 0.04 0.06 0.06 0.06 0.06 0.06 0.14 0.15 0.11 0.13 0.12 0.12 0.13 0.11 0.11 0.11 0.10 0.10 0.10 0.11 0.11 0.11 CU86 Cuvier Island 12 0.05 0.06 0.06 0.06 0.07 0.07 0.04 0.05 0.05 0.05 0.06 0.06 0.13 0.14 0.10 0.12 0.11 0.11 0.11 0.10 0.10 0.10 0.10 0.10 0.09 0.10 0.10 0.10 CO67 Taiharuru 13 0.04 0.04 0.04 0.07 0.08 0.08 0.08 0.07 0.06 0.06 0.06 0.07 0.11 0.13 0.09 0.11 0.12 0.12 0.13 0.11 0.12 0.12 0.13 0.12 0.12 0.13 0.13 0.12 MI100 Middle Island 14 0.02 0.03 0.03 0.08 0.07 0.07 0.08 0.06 0.06 0.06 0.06 0.08 0.10 0.11 0.07 0.09 0.11 0.11 0.11 0.10 0.12 0.12 0.11 0.11 0.11 0.12 0.12 0.12 MI101 Middle Island 15 0.02 0.03 0.03 0.08 0.08 0.08 0.08 0.06 0.06 0.06 0.06 0.08 0.11 0.12 0.08 0.10 0.11 0.11 0.11 0.10 0.12 0.12 0.11 0.11 0.11 0.12 0.12 0.12 MI98 Middle Island 16 0.02 0.03 0.02 0.07 0.07 0.07 0.08 0.05 0.06 0.06 0.06 0.08 0.10 0.11 0.08 0.09 0.11 0.11 0.11 0.10 0.12 0.12 0.10 0.10 0.11 0.11 0.12 0.11 MI99 Middle Island 17 0.03 0.03 0.03 0.08 0.08 0.08 0.08 0.06 0.06 0.06 0.06 0.08 0.11 0.12 0.08 0.09 0.11 0.11 0.12 0.11 0.12 0.13 0.11 0.11 0.12 0.12 0.13 0.12 CO121 Coromandel Town 18 0.04 0.04 0.04 0.06 0.07 0.07 0.07 0.06 0.05 0.05 0.05 0.07 0.14 0.15 0.10 0.12 0.13 0.13 0.14 0.12 0.13 0.14 0.13 0.13 0.14 0.14 0.14 0.14 CO127 Coromandel Town 19 0.03 0.03 0.03 0.06 0.07 0.07 0.07 0.06 0.05 0.05 0.05 0.06 0.12 0.14 0.10 0.11 0.12 0.12 0.13 0.12 0.13 0.13 0.13 0.12 0.12 0.13 0.13 0.12 CO129 Coromandel Town 20 0.03 0.03 0.03 0.07 0.08 0.08 0.07 0.06 0.05 0.05 0.05 0.06 0.12 0.13 0.09 0.11 0.12 0.12 0.12 0.11 0.12 0.12 0.12 0.11 0.11 0.11 0.12 0.12 CO141 Coromandel Town 21 0.04 0.03 0.03 0.06 0.06 0.07 0.06 0.05 0.05 0.05 0.05 0.06 0.14 0.14 0.10 0.12 0.13 0.13 0.13 0.12 0.13 0.13 0.13 0.13 0.13 0.14 0.14 0.13 CO37 Coromandel Town 22 0.03 0.03 0.03 0.06 0.07 0.07 0.07 0.05 0.05 0.05 0.05 0.06 0.13 0.15 0.10 0.12 0.13 0.13 0.14 0.12 0.12 0.12 0.13 0.12 0.13 0.13 0.12 0.12 AL112 Ruamahuanui Island 23 0.02 0.02 0.02 0.07 0.07 0.08 0.08 0.05 0.05 0.05 0.05 0.07 0.13 0.14 0.11 0.12 0.12 0.11 0.11 0.10 0.13 0.13 0.12 0.12 0.13 0.13 0.14 0.12 KA119 Kauaeranga Valley 24 0.01 0.02 0.00 0.07 0.07 0.07 0.07 0.05 0.05 0.05 0.05 0.07 0.12 0.13 0.08 0.10 0.12 0.12 0.12 0.11 0.12 0.12 0.11 0.11 0.12 0.12 0.13 0.12 KA120 Kauaeranga Valley 25 0.01 0.01 0.00 0.07 0.06 0.06 0.07 0.05 0.05 0.05 0.05 0.07 0.11 0.13 0.09 0.10 0.11 0.11 0.12 0.11 0.12 0.12 0.11 0.11 0.12 0.13 0.13 0.11 KA87 Kauaeranga Valley 26 0.01 0.00 0.01 0.06 0.07 0.07 0.07 0.05 0.05 0.05 0.05 0.07 0.11 0.13 0.09 0.11 0.11 0.11 0.12 0.11 0.11 0.12 0.11 0.11 0.12 0.12 0.13 0.11 KA88 Kauaeranga Valley 27 0.01 0.01 0.00 0.07 0.07 0.07 0.07 0.05 0.05 0.05 0.05 0.08 0.12 0.13 0.10 0.12 0.12 0.12 0.12 0.11 0.13 0.13 0.12 0.12 0.13 0.13 0.14 0.12 KA89 Kauaeranga Valley 28 0.01 0.01 0.06 0.06 0.06 0.06 0.05 0.05 0.05 0.05 0.06 0.11 0.12 0.09 0.11 0.10 0.10 0.11 0.10 0.11 0.11 0.11 0.10 0.11 0.12 0.12 0.11 KA90 Kauaeranga Valley 29 0.01 0.01 0.07 0.08 0.08 0.08 0.06 0.05 0.06 0.05 0.07 0.12 0.13 0.10 0.12 0.12 0.12 0.12 0.11 0.12 0.12 0.12 0.11 0.12 0.13 0.13 0.12 KA91 Kauaeranga Valley 30 0.01 0.01 0.07 0.07 0.07 0.07 0.05 0.05 0.05 0.05 0.07 0.11 0.13 0.09 0.11 0.11 0.11 0.12 0.11 0.12 0.12 0.11 0.11 0.12 0.13 0.13 0.12 GW230 Lake Waikaremoana 31 0.05 0.05 0.05 0.02 0.02 0.07 0.05 0.06 0.06 0.05 0.07 0.11 0.12 0.10 0.12 0.14 0.14 0.15 0.13 0.11 0.11 0.12 0.12 0.11 0.12 0.12 0.11 GW231 Lake Waikaremoana 32 0.05 0.05 0.05 0.01 0.00 0.07 0.05 0.06 0.06 0.06 0.06 0.12 0.12 0.11 0.12 0.12 0.13 0.13 0.12 0.12 0.12 0.11 0.12 0.12 0.13 0.12 0.12 LW189 Lake Waikaremoana 33 0.05 0.05 0.05 0.01 0.00 0.07 0.05 0.07 0.07 0.07 0.06 0.11 0.12 0.10 0.11 0.13 0.13 0.14 0.12 0.12 0.12 0.11 0.11 0.11 0.12 0.12 0.11 GW190 Wanganui 34 0.05 0.06 0.05 0.05 0.05 0.05 0.06 0.06 0.06 0.06 0.06 0.15 0.17 0.10 0.12 0.14 0.13 0.13 0.13 0.13 0.13 0.10 0.11 0.11 0.12 0.11 0.14 GW92A Palmerston North 35 0.04 0.04 0.04 0.04 0.04 0.04 0.05 0.02 0.02 0.02 0.05 0.11 0.11 0.10 0.11 0.11 0.11 0.11 0.10 0.09 0.09 0.11 0.10 0.11 0.10 0.10 0.09 PN178 Palmerston North 36 0.04 0.04 0.04 0.05 0.05 0.05 0.05 0.01 0.00 0.01 0.05 0.12 0.12 0.11 0.12 0.13 0.12 0.12 0.11 0.10 0.10 0.11 0.10 0.10 0.10 0.10 0.10 PN22 Palmerston North 37 0.04 0.04 0.04 0.05 0.05 0.05 0.05 0.02 0.00 0.01 0.06 0.12 0.12 0.11 0.12 0.13 0.12 0.13 0.12 0.10 0.10 0.11 0.10 0.10 0.11 0.11 0.10 PN23 Palmerston North 38 0.04 0.04 0.04 0.04 0.05 0.05 0.05 0.02 0.01 0.01 0.06 0.11 0.12 0.10 0.12 0.12 0.11 0.12 0.11 0.08 0.09 0.11 0.10 0.10 0.10 0.10 0.09 GW87 Pohangina Valley 39 0.05 0.05 0.05 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.13 0.14 0.11 0.13 0.12 0.11 0.12 0.11 0.11 0.11 0.10 0.10 0.10 0.10 0.10 0.11 GW171 Cambridge 40 0.07 0.07 0.07 0.07 0.07 0.07 0.08 0.07 0.07 0.07 0.07 0.07 II 0.00 0.04 0.05 0.06 0.06 0.07 0.05 0.09 0.10 0.09 0.09 0.10 0.10 0.10 0.09 GW200 Taupo 41 0.07 0.08 0.07 0.07 0.07 0.07 0.09 0.07 0.07 0.07 0.07 0.08 0.00 0.05 0.06 0.07 0.07 0.07 0.06 0.11 0.11 0.11 0.11 0.11 0.11 0.12 0.10 GW235 Taranaki 42 0.06 0.07 0.06 0.07 0.07 0.07 0.07 0.06 0.07 0.07 0.07 0.07 0.03 0.04 0.01 0.08 0.07 0.08 0.06 0.09 0.09 0.07 0.08 0.07 0.08 0.08 0.08 TK187 Taranaki 43 0.07 0.07 0.07 0.07 0.08 0.07 0.08 0.07 0.08 0.08 0.08 0.08 0.04 0.04 0.01 0.09 0.08 0.08 0.07 0.11 0.11 0.08 0.09 0.09 0.09 0.09 0.10 GW232 Palmerston North 44 0.06 0.07 0.07 0.08 0.07 0.08 0.08 0.07 0.08 0.08 0.07 0.07 0.05 0.05 0.06 0.06 0.01 0.01 0.01 0.13 0.13 0.13 0.12 0.13 0.13 0.14 0.11 PN174 Palmerston North 45 0.06 0.07 0.07 0.08 0.07 0.08 0.08 0.07 0.07 0.07 0.07 0.07 0.05 0.05 0.05 0.06 0.01 0.00 0.01 0.12 0.12 0.13 0.13 0.13 0.13 0.14 0.11 PN176 Palmerston North 46 0.07 0.07 0.07 0.08 0.08 0.08 0.08 0.07 0.07 0.08 0.07 0.07 0.05 0.05 0.05 0.06 0.01 0.00 0.01 0.12 0.13 0.13 0.13 0.12 0.13 0.13 0.11 PN177 Palmerston North 47 0.06 0.07 0.07 0.08 0.07 0.07 0.08 0.06 0.07 0.07 0.07 0.07 0.04 0.04 0.05 0.05 0.01 0.01 0.01 0.11 0.12 0.11 0.11 0.11 0.12 0.12 0.10 GW80 Wellington 48 0.07 0.08 0.08 0.07 0.08 0.08 0.08 0.06 0.06 0.07 0.06 0.07 0.06 0.07 0.06 0.07 0.08 0.08 0.08 0.07 III 0.00 0.07 0.06 0.07 0.07 0.07 0.00 GW82 Wellington 49 0.07 0.08 0.08 0.07 0.08 0.08 0.08 0.06 0.07 0.07 0.06 0.07 0.06 0.07 0.06 0.07 0.08 0.08 0.08 0.07 0.00 0.07 0.07 0.07 0.07 0.07 0.01 GW89A Wellington 50 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.07 0.05 0.06 0.08 0.08 0.08 0.07 0.05 0.05 0.00 0.01 0.01 0.01 0.06 GW89B Wellington 51 0.07 0.07 0.07 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.06 0.07 0.06 0.06 0.08 0.08 0.08 0.07 0.05 0.05 0.00 0.01 0.00 0.01 0.06 WN181 Wellington 52 0.07 0.08 0.08 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.06 0.07 0.05 0.06 0.08 0.08 0.08 0.07 0.05 0.05 0.01 0.01 0.00 0.00 0.06 WN182 Wellington 53 0.07 0.08 0.08 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.07 0.06 0.06 0.08 0.08 0.08 0.07 0.05 0.05 0.01 0.00 0.00 0.00 0.06 WN183 Wellington 54 0.08 0.08 0.08 0.07 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.06 0.07 0.07 0.06 0.06 0.08 0.08 0.08 0.08 0.05 0.05 0.01 0.01 0.00 0.00 0.07 GW176 Wellington 55 0.07 0.07 0.07 0.07 0.07 0.07 0.08 0.06 0.07 0.07 0.06 0.07 0.06 0.06 0.06 0.07 0.07 0.07 0.07 0.07 0.00 0.01 0.05 0.04 0.05 0.05 0.05