1 Simon Hodge and Rob Cruickshank Editorial: A Hard Day’s Night - What’s to be gained from fumbling around in the dark?

Simon Hodge and Rob Cruickshank

In many cases daily patterns in insect activity are so well defined between night and day there is little doubt about when, or how, an investigator should go about studying them. However, even for some relatively well-studied groups (e.g. Diptera; spiders; beneficial insects; herbivores; pollinators), collecting during the daytime can produce very different results compared with collecting at night (Lewis & Taylor 1964; Janzen 1973; Green 1999; Brown & Schmitt 2001; Devoto 2011; Suter & Benson 2014).

If insect activity is concentrated at a particular time period (e.g. dawn; dusk; noon), sampling can be made more efficient by focussing on a narrower time window. For example, when studying non-native pollinators in Fiji, Prasad & Hodge (2013a) found that the exotic allodapine bee, Braunsapis puangensis, showed a clear peak in foraging activity in the middle of the day (although not when it was raining!). In a subsequent biogeographic survey all collecting was performed between 11am and 2pm in order to optimise catches, and also so that any absences were less likely to be explained by the bees being inactive (Prasad & Hodge 2013b).

In the previous issue of The Weta (vol. 51), two papers highlighted how knowledge of daily behaviour patterns could help researchers avoid performing surveys at the wrong time of day. Watts et al. (2017) described a survey of Cook Strait giant wētā, Deinacrida rugosa, on Matiu/Somes Island. Because these wētā leave their retreats around dusk, and are nocturnal foragers, the authors performed their searches at night. For this species, previous knowledge regarding its natural history and feeding behaviour meant the researchers already knew daytime searches were unlikely to produce meaningful data. In the same issue of The Weta, Vink et al. (2017) described new observations of the marine spider Desis marina. The authors reported very few specimens were located using a previously- described search technique that involved lifting seaweed holdfasts. However, during night-time searching they observed approximately 50 specimens in just 2.5 hours. Based on these findings, one could envisage any The Weta 52: 1-6 2 future field research on Desis marina would profit by including some nocturnal survey work.

Some insect-collecting methods, such as pitfall traps, flight interception traps, and Malaise traps, are usually in place for a number of days before the contents are retrieved, and the catch contains both day and night collections. Additional information on insect activity would be obtained if the traps were emptied more often, separating crepuscular, day, and night-time catches (e.g. Chatzimanolis et al. 2004). Obviously visiting and emptying traps two or three times a day would require extra effort for the researcher [although some elaborate ‘clockwork’ mechanisms have been developed for pitfall traps so that catches from different time periods are kept separate; e.g. Blumberg & Crossley 1988; Buchholz 2009]. But even if the separation of night and day catches was carried out over a short time period, or as a pilot study, this may still produce valuable data indicating that at least some species were active at different times of day.

Some collecting methods are, on the face of it, closely associated with a particular part of the daily cycle. For example, moths represent the largely nocturnal Lepidoptera and are predominantly collected at night using an attractant (e.g. light; bait; ‘sugaring’). On the other hand, the day-flying Lepidoptera represented by butterflies can, if the researcher has the required identification skills, be monitored by sight, using linear transects or fixed time searches to give a standardized sample unit. Just as light trapping (for moths, caddisflies, Diptera, etc.) is associated with night-collecting, using variably-coloured sticky traps and pan traps is associated with sampling day- flying insects. However, because light traps are rarely run during the day, the propensity of these traps to attract day-active insects is generally not known. Similarly, it is often assumed that insects respond differently to light reflected off different coloured sticky traps. However, it is rarely (if ever?) checked whether any discrepancies among catches obtained by different coloured traps would also occur during darkness, which would be an interesting phenomenon to report (and then try to explain!).

The local abundance of nocturnally-active insects can be estimated by examining occupation of natural shelters, or by using ‘artificial cover objects’ (ACOs). Daytime occupation of ‘motels’, corrugated metal shelters, and baited bathroom tiles has been used to study populations of wētā, katipō and woodlice respectively, and allows estimates of abundance by using 3 Simon Hodge and Rob Cruickshank animal counts or the proportions of shelters that are occupied (Bowie et al. 2006; Hodge & Standen 2006; Costall & Death 2010). However, it is important to recognize that the data obtained in this way reflect not only animal abundance, but also their inclination to take shelter in the ACO, and the availability of natural shelters as alternative resting places.

It can generally be assumed that passive collecting methods, such as pitfalls and interception traps, may function with more or less equal efficiency during night and day. However, active (or ‘hand’) searching for insects at night is clearly not the same process as searching for insects during daytime. Moving around field sites in darkness is inherently more difficult; sampling may be slowed, the ‘agility’ of the collector may decrease, and taking notes, writing labels, transferring specimens from net or pooter to storage vessels, and pouring preservatives, will all tend to be at least slightly more problematic at night than during the day. The collector’s field of view will be limited to that offered by artificial lighting (lanterns, lamps, head torches) and although this may serve to focus the collector’s viewpoint, also means animals moving in the peripheral vision may be missed. Frustratingly, the light from a head torch can attract some species that would not appear during daytime samples, and may mean the catch is no longer representative of the immediate sampling area. Conversely, exposure to bright lights can supress typical behaviour of nocturnal species and make them more difficult to locate (Shimoda & Honda 2013).

If differences in species abundance do occur between night and day, then this raises questions about the mechanisms causing these differences. If animals are recorded at any time, does this mean they are likely to have been in the general vicinity all of the time? If so, differences in their recorded abundance must be due to changes in activity and/or the success of the collecting methods used. If sampling is more or less efficient at different times of day, how and when can we decide that some animals are strictly nocturnal and others totally diurnal? For example, if animals are caught at night but not during the day, how many ‘empty’ day time samples are required before you can state with a given degree of confidence that the animals are active only at night? (see Hodge & Vink 2017).

We encourage researchers to obtain more data on daily patterns of New Zealand insects. Often, an initial phase of a field entomology research project is used to compare and modify collecting techniques, decide on The Weta 52: 1-6 4 minimum sample sizes, and so on, and this development phase could be adapted to include a comparison of day and night-time catches. Although some (or much) of the information obtained might be self-evident, and reinforce knowledge already reported in the literature, with luck some new data or unusual findings may also be gained, and could result in the collecting periods for the main study species becoming more focussed.

Examples of questions that could be easily tested during a pilot study include:

 What differences occur in the observed abundance of target species if hand searching or sweep netting is performed at night compared with day time collecting?  What do light-traps catch during the day? What time do night-time target species first tend to appear?  How do the catches of sticky traps and pan traps differ between night and day? Does colour influence night time catches of sticky traps and pan traps?  What happens to the occupancy rates of ACOs if checked at night compared to being assessed during the day?

In New Zealand, as in many parts of the world, apart from some species deemed important to agriculture, and some other conspicuous species vaunted for conservation status, little tends to be known about the ecology and behaviour of many of the insect species we have in our collections. Knowing an insect’s name, where it occurs, and the time of year it was collected, is an obvious first step. With just a little more effort during the collection phase, additional statistics on daily patterns in activity could be obtained, which might just provide an extra line of valuable information on the specimen label.

References

Bowie, M.H., Hodge, S., Banks, J. & Vink, C. (2006) An appraisal of simple tree-mounted shelters for non-lethal monitoring of weta (Orthoptera: Anostostomatidae and Rhaphidophoridae) in New Zealand nature reserves. J. Insect Cons. 10, 261-268.

5 Simon Hodge and Rob Cruickshank

Blumberg AY, Crossley DA (1988) Diurnal activity of soil-surface arthropods in agroecosystems: design for an inexpensive time-sorting pitfall trap. Agric. Ecosys. Environ. 20: 159-164.

Buchholz S (2009) Design of a time-sorting pitfall trap for surface-active arthropods. Entomologia Experimentalis et Applicata 133: 100-103.

Brown MW, Schmitt JJ (2001) Seasonal and diurnal dynamics of beneficial insect populations in apple orchards under different management intensity. Environ. Entomol. 30: 415-424.

Chatzimanolis S, Ashe JS, Hanley RS (2004) Diurnal/nocturnal activity of rove beetles (Coleoptera: Staphylinidae) on Barro Colorado Island, Panama assayed by flight intercept trap. The Coleopterists Bull. 58:569- 577.

Costall JA, Death RG. 2010. Population monitoring of the endangered New Zealand spider, Latrodectus katipo, with artificial cover objects. New Zeal J Ecol. 34:253–258.

Devoto M, Bailey S, Memmott J (2011) The ‘night shift’: nocturnal pollen- transport networks in a boreal pine forest. Ecol. Ent. 36: 25-35.

Green J (1999) Sampling method and time determines composition of spider collections. J. Arachnol 27: 176–182.

Hodge S, Standen V (2006) The use of cryptozoa boards' to examine the distribution of woodlice (Isopoda) and millipedes (Diplopoda) in a disused limestone quarry. Entomologists Monthly Magazine 142: 55-61.

Hodge S, Vink CJ (2017) Evidence of absence is not proof of absence: the case of the New Brighton katipō? New Zeal. J. Zool. 44: 14-24.

Janzen DH (1973) Sweep samples of tropical foliage insects: effects of seasons, vegetation types, elevation, time of day, and insularity. Ecology 54: 687-708.

Lewis T, Taylor LR (1964) Diurnal periodicity of flight by insects. Trans. Royal Entomol. Soc. Lond. 116: 393-479. The Weta 52: 1-6 6

Prasad AV, Hodge S (2013a) Factors influencing the foraging activity of the allodapine bee Braunsapis puangensis on creeping daisy (Sphagneticola trilobata) in Fiji. J. Hymenoptera Res. 35: 59-69.

Prasad AV, Hodge S (2013b) The diversity of arthropods associated with the exotic creeping daisy Sphagneticola trilobata in Suva, Fiji Islands. Entomologists Monthly Magazine 149: 155-161.

Shimoda M, Honda K (2013) Insect reactions to light and its applications to pest management. Appl. Entomol. Zool. 48:413-421.

Suter RB, Benson K (2014) Nocturnal, diurnal, crepuscular: activity assessments of Pisauridae and Lycosidae. J Arachnol. 42:178-191.

Vink CJ, McQuillan BN, Simpson AH, Correa-Garhwal SM (2017) The marine spider, Desis marina (Araneae: Desidae): new observations and localities. The Weta 51: 71-79.

Watts C, Thornburrow D, Stringer I (2017) Ecological observations of Cook Strait giant wētā, Deinacrida rugosa (Orthoptera: Anostostomatidae), on Matiu/Somes Island. The Weta 51: 11-19. 7 Corinne Watts et al. Food of Cook Strait giant wētā, Deinacrida rugosa on Matiu/Somes Island: do plant nutrient levels influence wētā distribution?

Corinne Watts1, Danny Thornburrow1, Ian Stringer2 and Vanessa Cave3

1Landcare Research, Private Bag 3127, Hamilton, New Zealand 2Department of Conservation, PO BOX 10420, Wellington, New Zealand 3AgResearch Ltd, Private Bag 3115, Hamilton 3240, New Zealand

Introduction

Cook Strait giant wētā, Deinacrida rugosa Buller, 1871 (Orthoptera: Anostostomatidae) has been one of the most frequently translocated insects for conservation purposes in New Zealand with 538 individuals being involved in seven translocations between 1977 and 2010 (Watts et al. 2008; Sherley et al. 2010 and references within).The second translocation was made in 1996 when 62 individuals were released at the north end of Matiu/Somes Island, Wellington Harbour (Gascoigne 1996). We subsequently studied the giant wētā population on Matiu/Somes Island between 2007 and 2016, and found that its geographic distribution changed with time. In 2008, most wētā were found at the north of the island and were rarely encountered at the south end because they were still becoming established there (Watts et al. 2009). Conversely, by 2013 most wētā were encountered in the southwest of the island and few were found in the north. This remained the distribution in 2015 and 2106 (Watts et al. in prep).

Dr George Gibbs (unpublished data) previously found that the Wellington tree wētā Hemideina crassidens (Blanchard, 1851) also underwent a similar distributional shift after this insect was released at the north of Matiu/Somes Island in 1996 and 1997. He found that adult males in the north were smaller than those in the south, and that those in the south were of similar size to adult males from Mana Island, from where the insects had been originally taken. He speculated that these changes in size and distribution might be related to some nutritional factor that is perhaps present in higher concentrations in the south (Dr George Gibbs, pers. comm.). This follows The Weta 52: 7-16 8 because Dr Gibbs found that male H. crassidens develop into larger adults when fed more animal protein (unpublished results).

In this study we have recorded what we observed D. rugosa feeding on in the field and obtained basic nutrient compositions of the three most commonly eaten foods plus the leaves of a tree eaten by both H. crassidens and D. rugosa. In addition, we measured the sizes of adults found throughout the island to determine if the size of adult D. rugosa varied geographically. The aims of the research were (1) investigate whether the size of adult D. rugosa varied over Matiu/Somes Island, (2) record what plants adult D. rugosa were observed eating, and (3) analyse the nutrient levels in the three main food items eaten by D. rugosa together with the leaves of taupata (Coprosma repens) which is frequently eaten by both Wellington tree wētā and D. rugosa to determine if there was a relationship between nutrient levels and geographic distribution of these wētā.

Methods

Visual searches

A visual search using spotlights was made each night over four nights from 2120 to 0130 hours along six transects on Matiu/Somes Island in February 2013, 2015 and 2016. The path and up to ca 1 m on both sides, together with vegetation up to ca 2 m high, were systematically and thoroughly searched by three people without disturbing the vegetation. When an adult wētā was found, a position was recorded using a GPS and the wētā was marked with small individually numbered labels. The length of the pronotum (a surrogate for wētā size) was measured in the dorsal mid-line using digital callipers.

Feeding

Each adult wētā found was recorded as not eating or eating, and if it was eating then what it was feeding on was also noted. Wētā were scored as eating if they had vegetation or animal material in their mandibles or if their jaws were close to a leaf or part of an animal showing fresh visible damage. The latter were considered to have stopped eating when illuminated and become motionless.

9 Corinne Watts et al.

Plant nutrient analysis

Chemical analyses were performed on leaves of taupata and on three of the plants most frequently observed eaten by D. rugosa (clover, Trifolium repens; plantain, Plantago coronopus; stinking camomile, Anthemis copula). Samples of fresh leaves were collected from each species at three different locations in the northern and southern end of Matiu/Somes Island (six collection locations, 24 samples). The locations were >5 m apart but were otherwise the first places where all four plant species were found within a 1-m radius while walking either south along the North transect or walking east along the Southwest transect. At each location, clover leaves from numerous plants were collected together in a brown paper bag whereas leaves from three different individual plants of the other species were collected together and kept segregated in brown paper bags. Bags were labelled with the GPS position for the collection location and later oven dried for 24 h at 60°C. Nitrogen, phosphorus, potassium, S, calcium, magnesium, sodium, iron, manganese, zinc, copper, and boron concentrations were determined on 10 g of each sample by Hill Laboratories, Hamilton, New Zealand. We used nitrogen content as a proxy for protein content.

Analysis

The relationship between wētā size (pronotum length) and geographic location (north-south or east-west) was explored using linear regression in R (version 3.0.1) (R Core Team 2014). Two-way ANOVAs, blocked by collection location, was used to compare the average nutrient content of leaves between the four species on the north and south of Matiu/Somes Island. The ANOVA were fitted using GenStat 17 (VSN International 2014).

Results

Geographical variation in size of adult wētā

Overall, the pronotum lengths of female adult wētā varied from 10.7 mm to 15.6 mm (mean ± SE, 13.09 ± 0.10 mm) and 10.2 mm to 13.4 mm in males (11.58 ± 0.14 mm). The pronotum length of males did not vary along either north-to-south (slope of linear regression: P = 0.400) or east-to-west (P = The Weta 52: 7-16 10

0.129) of Matiu/Somes Island, whereas the pronota of females captured in the west were slightly longer than from those in the east (slope – 0.000842±0.000339 mm m–1, P = 0.0138) but there was no change from north to south (slope P = 0.963; Fig. 1).

Food eaten by adult wētā

When illuminated by torch light some wētā continued to feed but most froze and ceased feeding with their mandibles adjacent to or touching damaged leaves. In some cases, leaf material was still in their mouths. Of a total of 324 females and 147 males examined, we were unable to determine whether seven females (4.3%) had been eating. A higher proportion of females (48.1%) were feeding or had apparently just stopped feeding when observed, compared with males (22.5%). Those that were not feeding included two pairs of males and females observed in copula and five females observed ovipositing. Of wētā that were eating, most (74.4% F; 78.1% M) were consuming clover, plantain, stinking camomile or grass (Table 4).The remainder mostly ate a variety of other plants including buttercup (Ranunculus repens), convolvulus (Calystegia tuguriorum), daisy (Bellis perennis), dandelion (Taraxacum officinale), dry macrocarpa foliage (Cupressus macrocarpa), empty flax seed case (Phormium tenax), native spinach (Tetragonia tetragonioides) and moss. Six females and one male were eating material of animal origin consisting of a feather, bird faeces (two observations), a dry dead Mimopeus opaculus beetle, a dead skink, a tree wētā, and a ground wētā (Hemiandrus sp.); both wētā appeared to have died recently (Table 1).

Chemical composition of food plants The average chemical content of leaves differed among the four species sampled (clover, stinking camomile, plantain and taupata) but no differences were detected between leaves from the north and south of Matiu/Somes Island, although sample sizes were small. Calcium and sodium levels came close to having different concentrations between north and south and their interactions between location and plant species came close to significance. Thus the concentrations of calcium tended to be higher in the south in three of the plant species whereas in plantain it tended to be lower in the south. Sodium, in contrast, tended to be higher in north except in taupata where it tended to be higher in the south (Appendices 1 and 2). In addition, since no differences were detected in nitrogen levels then it is likely that there were 11 Corinne Watts et al. no detectable differences in overall protein concentrations between plants in the north and south (Appendix 1 and 2).

Figure 1. Variation in pronotum length of adult D. rugosa in relation to location on Matiu/Somes Island. Position is given as GPS co-ordinates (New Zealand Transverse Mercator 2000) of northing and easting. Only females showed a significant change in size from east to west (shown as a regression line; R2 = 0.032). Solid circles = females; open circles = males.

15 15

14 14 y = 4586.7-0.0008x

13 13

12 12

11 11

Pronotum length (mm)

10 10

9 9 1756000 1756100 1756200 1756300 1756400 1756500 5430400 5430600 5430800 5431000 5431200 5431400 Northing Easting

Table 1. Proportions (%) of adult D. rugosa wētā feeding when found and the proportion of different foods they were eating.

Food item Female (%) Male (%) Feeding 48.2 22.5 Clover 11.1 2.8 Plantain 9.6 6.3 Stinking camomile 9.0 4.9 Grass 6.2 3.5 Taupata 2.8 2.8 Dandelion 2.5 0 Other plants 5.2 1.4 Animal 1.9 0.7 Unsure if feeding 2.2 0 The Weta 52: 7-16 12

Discussion

Chemical composition of plants in relation to wētā size and distribution

We found no evidence that nutrient levels differed from north to south amongst the three main plant species we observed D. rugosa eating. We also detected no north to south difference in nutrient levels in leaves of taupata bushes which are eaten by both D. rugosa and H. crassidens. Our investigation was a preliminary one and further samples both from the same plant species and from additional food species may show that nutrient levels in the plants eaten do vary geographically. The latter seems likely because adult male tree wētā had slightly longer metatibiae than those found on Mana Island (Watts et al. 2009) and Dr George Gibbs (unpublished data), as noted above, found that adult Wellington tree wētā matured at later instars when fed more animal protein, thereby becoming larger. In addition, adult female D. rugosa in the west were slightly larger than those in the east (estimated mean ± SEM for adult female pronotum length: maximum observation west 13.10±0.05 mm; maximum observation east, 13.15±0.6 mm; these observations were 643 m apart) indicating a possible difference in available nutrient levels but this was not investigated further. Species of Deinacrida are not known to develop through variable numbers of instars so their adults are expected to vary less in size than adult males of Hemideina species (Spencer 1995; Stringer and Cary, 2001). Both D. rugosa and H. crassidens will eat animals when they can although they are primarily herbivorous, so availability of animal food may be more important than plant nutrition but we did not investigate this (Ramsay 1955; Barrett and Ramsay 1991).

Conclusions

It is still not clear why numbers of both D. rugosa and H. crassidens diminished in the north following their releases there and remained high in the southwest once they reached that location. One possibility is that nocturnally active geckos, which are insectivorous, were abundant where few wētā were found so their presence may have depressed the numbers of both tree and giant wētā Other causes suggested for D. rugosa are changes in the availability of suitable habitat (the lawn alongside paths that this wētā frequents at night diminished between 2008 and 2016 but the reduction was not measured), and the harvesting of D. rugosa from Matiu/Somes Island 13 Corinne Watts et al. for translocation elsewhere (Watts et al. in prep). However, we have not demonstrated that the changes in distribution were due to differences in chemical composition of food plants. We acknowledge that there are many other components in food which we did not investigate, and that some of these may well affect the geographic distribution of these species.

Acknowledgements

This research was supported by CoRE funding for Crown Research Institutes from the Ministry of Business, Innovation and Employment’s Science and Innovation Group, under CO9X0503 and Department of Conservation (DOC) investigation No. 4091. We thank Jo Greenwood and Emma Dunning (DOC Rangers, Matiu/Somes Island), and Wellington Tenths Trust for their support. Thanks to George Gibbs for providing unpublished results and discussing our results.

References

Barrett P, Ramsay GW. 1991. Keeping Wētās in Captivity. Wellington Zoological Gardens, Wellington. 60 p.

Gascoigne B. 1996. First transfer of Cook Strait wētā (Deinacrida rugosa) from Mana Island to Somes Island (Matiu). Unpublished report G11-803. Department of Conservation, Wellington. 6 p.

R Core Team. 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. www.R-project.org/.

Ramsay GW. 1955. The exoskeleton and musculature of the head and the life-cycle of Deinacrida rugosa Buller. MSc thesis, Victoria University of Wellington, New Zealand. 163 p.

Sherley GH, Stringer I, Parrish GR. 2010. Summary of native bat, reptile, amphibian and terrestrial invertebrate translocations in New Zealand. Science for Conservation No. 303. Department of Conservation, Wellington. 39 p.

The Weta 52: 7-16 14

Spencer AM. 1995. Sexual selection in the male tree wētā Hemideina crassidens (Orthoptera: Stenopelmatidae). MSc, Victoria University of Wellington, Wellington. 85 p.

Stringer I, Cary P. 2001. Postembryonic development and related changes. In: The Biology of Wētās, King Crickets and their Allies (ed. Field LH), pp. 395–426. CAB International, Wallingford, UK.

VSN International. 2014. GenStat for Windows, 17th Edn. VSN International, Hemel Hempstead, UK. www.GenStat.co.uk.

Watts C, Emson R, Thornburrow D, Maheswaran R. 2012. Movements, behaviour and survival of adult Cook Strait giant wētā (Deinacrida rugosa; Anostostomatidae: Orthoptera) immediately after translocation as revealed by radiotracking. Journal of Insect Conservation 16: 763– 776.

Watts C, Stringer I, Sherley G, Gibbs G, Green C. 2008. History of wētā (Orthoptera: Anostostomatidae) translocation in New Zealand: lessons learned, islands as sanctuaries and the future. Journal of Insect Conservation 12: 359–370.

Watts C, Stringer I, Thornburrow D, Sherley G, Empson R. 2009. Morphometric change, distribution, and habitat use of Cook Strait giant wētā (Deinacrida rugosa Orthoptera: Anostostomatidae) after translocation. New Zealand Entomologist 32: 59–66.

15 Corinne Watts et al.

Appendix 1. Nutrient concentrations of plant leaves collected in 2013 from northern and southern ends of Matiu/Somes Island. (mean ± se)

Plant species Nutrient North South Mean SE Mean SE Clover B (mg/kg) 24.6 0.7 22.3 1.5 Plantain B (mg/kg) 23.3 0.3 25.7 0.9 Stinking camomile B (mg/kg) 48.3 1.7 66.7 11.9 Taupata B (mg/kg) 27.0 3.6 36.3 4.4 Clover Ca (%) 1.33 0.07 0.80 0.15 Plantain Ca (%) 2.20 0.04 1.80 0.23 Stinking camomile Ca (%) 1.19 0.07 0.98 0.13 Taupata Ca (%) 1.01 0.08 1.32 0.20 Clover Cu (mg/kg) 11.0 1.5 9.3 0.7 Plantain Cu (mg/kg) 14.7 1.2 16.7 0.67 Stinking camomile Cu (mg/kg) 14.3 1.2 13.7 0.7 Taupata Cu (mg/kg) 10.0 0.6 10.0 2.6 Clover Fe (mg.kg) 200.0 60.3 206.0 56.9 Plantain Fe (mg.kg) 128.0 30.0 116.7 29.6 Stinking camomile Fe (mg.kg) 73.0 2.5 140.0 36.7 Taupata Fe (mg.kg) 66.7 10.0 68.0 5.1 Clover Mg (%) 0.39 0.02 0.35 0.02 Plantain Mg (%) 0.55 0.05 0.49 0.05 Stinking camomile Mg (%) 0.42 0.03 0.42 0.02 Taupata Mg (%) 0.29 0.06 0.30 0.06 Clover Mn (mg/kg) 66.0 9.1 70.0 10.5 Plantain Mn (mg/kg) 61.3 4.4 109.0 24.5 Stinking camomile Mn (mg/kg) 216.7 6.7 347.0 83.2 Taupata Mn (mg/kg) 160.7 75.5 240.0 41.6 Clover N (%) 3.93 0.21 4.76 0.27 Plantain N (%) 2.80 0.12 3.00 0.10 Stinking camomile N (%) 2.93 0.35 2.47 0.27 Taupata N (%) 1.87 0.09 1.93 0.24 Clover P (%) 0.25 0.02 0.35 0.02 Plantain P (%) 0.43 0.02 0.41 0.02 Stinking camomile P (%) 0.32 0.05 0.35 0.06 Taupata P (%) 0.30 0.05 0.35 0.03 Clover K (%) 2.63 0.27 2.46 0.16 Plantain K (%) 2.57 0.41 2.97 0.27 Stinking camomile K (%) 3.73 0.03 3.27 0.72 Taupata K (%) 3.10 0.40 3.03 0.52 Clover Na (%) 0.261 0.040 0.520 0.055 Plantain Na (%) 0.808 0.136 0.699 0.036 Stinking camomile Na (%) 0.777 0.044 1.026 0.109 Taupata Na (%) 0.151 0.019 0.199 0.046 Clover S (%) 0.19 0.01 0.23 0.01 Plantain S (%) 0.37 0.05 0.35 0.05 Stinking camomile S (%) 0.18 0.02 0.17 0.02 Taupata S (%) 0.27 0.01 0.23 0.02 Clover Zn (mg/kg) 43.3 3.0 68.6 22.7 Plantain Zn (mg/kg) 51.7 6.6 108.0 46.8 Stinking camomile Zn (mg/kg) 61.3 1.9 126.7 38.1 Taupata Zn (mg/kg) 46.3 8.8 86.3 12.9 The Weta 52: 7-16 16

Appendix 2. Resulting probabilities from two-way ANOVAs comparing mean chemical content of leaves from four plant species (clover, plantain, stinking camomile, taupata) blocked by collection location (north, south).

Plant Plant species x Element Location species Location Boron <0.001 0.177 0.136 Calcium <0.001 0.083 0.052 Copper <0.001 0.943 0.476 Iron 0.012 0.609 0.670 Magnesium 0.004 0.438 0.855 Manganese <0.001 0.139 0.500 Nitrogen <0.001 0.374 0.083 Phosphorus 0.031 0.130 0.513 Potassium 0.055 0.866 0.602 Sodium <0.001 0.098 0.067 Sulphur <0.001 0.720 0.405 Zinc 0.332 0.097 0.781 17 Mike Winterbourn Unexpected records of spear-winged fly larvae (Diptera: Lonchopteridae) from New Zealand streams

Mike Winterbourn

School of Biological Sciences, University of Canterbury, Christchurch Email: [email protected]

Introduction

The purpose of this note is to report the unexpected finding of dipteran larvae belonging to the family Lonchopteridae in aquatic samples taken from South Island streams. Because lonchopterid larvae are not included in New Zealand keys to either terrestrial or aquatic insects I provide a brief description of a larva and summarise aspects of the larval biology.

The Lonchopteridae, spear-winged or pointed-wing flies, occur in many parts of the world, but no species are endemic to New Zealand. The adult flies are small, brownish or yellowish insects 2-5 mm long with pointed wings that are folded flat above the abdomen at rest (Harrison 1950; Bartak 2008). Although several genera have been erected in the family, Klymko and Marshall (2008) considered there was little justification for dividing the group into multiple genera and all extant lonchopterids are generally treated as belonging to a single genus, Lonchoptera Meigen.

The cosmopolitan L. bifurcata (Fallén) is the only species found in New Zealand. This species was initially reported by Harrison (1950) from the Auckland area as Lonchoptera dubia Curran, which along with L. furcata Fallén (a name also found in the New Zealand literature) are now considered to be synonyms of L. bifurcata. Subsequently, adults were taken in sweep samples from pastures at numerous localities throughout the North Island (Cumber and Harrison 1959) and records posted on the NatureWatch New Zealand website (http://naturewatch.org.nz) between 2013 and 2017 indicate it has a distribution extending from at least Whangarei to Dunedin. Unlike other species of Lonchoptera, L. bifurcata is parthenogenetic, with several ‘chromosomal races’ showing small differences in wing venation being recognised in North America by Stalker (1956). More recently, The Weta 52: 17-24 18

Ochman et al. (1980) demonstrated clonal variation within L. bifurcata using starch gel electrophoresis.

The main habitat of larvae is generally considered to be amongst dead leaves and other decaying vegetation (Smith 1996; Bartak 2008). Ferrar (1987) stated they preferred to live amongst leaves in a humid environment, and Brindle and Smith (1978) also emphasised their need for moisture. The larvae of L. nigrociliata Duda have been found under small rocks on shingle shorelines (Drake 1996) and at least two species have been found in springs and seeps (Klymko and Marshall 2008; Omelkova et al. 2013). Additionally, Smith (1996) commented that L. bifurcata had been found on brussels sprouts and turnips and speculated that transportation with vegetables combined with its parthenogenetic reproduction might explain its cosmopolitan distribution.

Little is known about the biology or ecology of lonchopterid larvae. However, Baud (1970) reported that L. bifurcata and L. lutea Panzer had six larval instars and that larval + pupal development took 34-48 days at 20±2oC. Given their preferred habitat appears to be amongst decaying plant material it is not surprising that larvae are thought to be saprophagous, ingesting fine organic particles and associated microorganisms such as bacteria and fungi (Stalker 1956; Baud 1970; Jones 1979; Bartak 2008). Ferrar (1987) considered they scraped microbes from fallen leaves and noted that known gut contents included fungal spores, coccoid algal cells, pollen, fine humic materials, fungal hyphae, testate amoebae and bacteria. The gut contents of the L. bifurcata larva shown in Figure 1 included fine organic and inorganic particles, algal filaments, fungal hyphae and spores.

Larval morphology

The anatomy and external morphology of the larva of Lonchoptera was described in detail (in German) by de Meijere (1900) and subsequent descriptions have been provided by Whitten (1956) and Peterson (1987). The cephalopharyngeal skeleton was described by Ferrar (1987). I follow the terminology of Peterson (1987) in defining the thoracic segment with the serrated anterior margin as the metathorax, posterior to which the first two abdominal segments are fused together. In contrast, earlier authors considered the metathorax of Peterson to be the mesothorax. Larvae are ventrally flattened and dorsally rounded, about twice as long as wide and up 19 Mike Winterbourn to 4 mm long (Foote 1991). Pairs of long filaments are present on the thorax and at the posterior end of the abdomen.

The mobile head is not sclerotised. The sclerotised metanotum has a prominently serrated anterior margin. The first two abdominal segments are fused, as are the last two abdominal segments. The abdomen bears seven notal plates, which have a finely mottled appearance produced by large numbers of very small, slightly raised, hexagonal protrusions. In life, larvae have a yellowish grey-green colour, but alcohol-preserved material I have seen is white or yellowish-white. Colourless, striated plates occur laterally on most body segments. Larvae lack eyes, legs or prolegs and are amphineustic with functional spiracles on the first thoracic and last abdominal segments (Whitten 1956).

New Zealand larvae (Lonchoptera bifurcata)

Material examined

Four alcohol-preserved larvae collected in stream samples and referred to me for identification.

1 larva, Tent Burn, Beachcroft Road, Southbridge, 8 January 2009, Michelle Greenwood, University of Canterbury.

1 larva, Waipahi River near The Cairns, SE of Gore. 7 February 2010, Amber Sinton, EOS Ecology, Christchurch.

1 larva, small spring-fed stream in pasture, Ashburton Forks, 18 March 2015, Haley Devlin, University of Canterbury (mounted on slide).

1 larva, tributary of Lower Pahaoa River, Wairarapa, 23 March 2017, Nick Hempston, EOS Ecology, Christchurch.

Description

A larva (length 3.1 mm) mounted in lactophenol-PVA is shown in Figures 1 and 2. A 5th instar larva of L. bifurcata figured by Bode (1973) was 2.8 mm long, suggesting that the New Zealand specimens are final (6th) instar larvae. The Weta 52: 17-24 20

Morphological features that can be seen are as follows. The head bears two pairs of short appendages: maxillary palps that are 1-segmented and have a broad base and 2-segmented antennae, which are three times longer than the maxillary palps. The cephalopharyngeal skeleton is prominent and bears a pair of curved mouth hooks. Three pairs of long filamentous processes are present on the thorax. The first pair (length 0.46 mm) arises from the pronotum, whereas the other two pairs extend from the serrated anterior margin of the metanotum.

The more central filaments are the longest (0.63 mm) and the more lateral pair, the shortest (0.36 mm). A complex ‘spine’ with a bifid distal section is present at the base of each lateral metanotal filament. Similar spines of variable length arise from each of the lateral striated plates, which also bear short, pointed spines anterior to the more complex, distally bifid spines. Anterior spiracles with very short stalks are prominent on the pronotum and black posterior spiracles are present on longer stalks antero-lateral to the long (0.55 mm) posterior spines. The raised protrusions on the dorsal surface can be seen as dotted hexagons in Figure 2.

Comments

Most descriptions of lonchopterid larvae state there are two pairs of long anterior filaments and one pair of long posterior filaments. In contrast, L. bifurcata clearly has three pairs of long anterior filaments as seen in Figure 2 and in the figure provided by Baud (1973). The reason for this discrepancy is that the lateral metanotal filament is much shorter in other species that have been illustrated in the literature (i.e. L. fallax de Meijere, L. lutea, L. tristis Meigen) and in books and monographs in which unidentified Lonchoptera larvae are figured (e.g., Brindle and Smith 1978, Peterson 1987, Foote 1991, Courtney et al. 2000). Because many described species of Lonchoptera are not known as larvae it is unclear whether long lateral metanotal filaments occur in other species.

It is also unknown whether larvae of L. bifurcata occur naturally in streams or whether individuals present in aquatic samples were taken fortuitously or accidently. Streams where they have been found are in pastoral land with a variety of shrubs, grasses and emergent macrophytes at their margins. Given that they require access to air for gas exchange and require a moist or humid microenvironment it is likely that suitable larval habitat in streams or rivers 21 Mike Winterbourn would be at the channel margin or amongst debris or stones which break the surface of the water.

Figure 1. A slide-mounted larva of Lonchoptera bifurcata collected at the Ashburton Forks showing the general body form, anterior and posterior filaments, lateral spines posterior spiracles and gut contents dominated by detritus.

The Weta 52: 17-24 22

Figure 2. Anterior end of larva showing antennae, maxillary palps, cephalopharyngeal skeleton, anterior spiracles and long anterior filaments. A lateral abdominal plate with associated spines can be seen at the upper right of the figure.

Acknowledgements

I thank Michelle Greenwood, Haley Devlin, Nick Hempston and Amber Sinton who sent specimens to me and provided information on larval habitats and collection sites.

References

Bartak M. 2008. Family Lonchopteridae. In: Global diversity of dipteran families (Insecta Diptera) in freshwater (excluding Simuliidae, 23 Mike Winterbourn

Culicidae, Chironomidae, Tipulidae and Tabanidae) (eds Wagner R and 13 others). Hydrobiologia, 595: 494.

Baud F. 1970. Le développement post-embryonnaire de deux Dipteres Musidoridés: Musidora furcata Fall. et M. lutea Panz. Revue Suisse de Zoologie, 77: 647-650.

Baud FJ. 1973. Biologie et Cytologie de cinq espèces du genre Lonchoptera Meig. (Dipt.) don’t l’une est parthénogénétique et les autres bissexuées, avec quelques remarques d’ordre taxonomique. Revue Suisse de Zoologie, 80: 473-515.

Brindle A, Smith KGV. 1978. The immature stages of flies. The Amateur Entomologist, 15: 38-64.

Courtney GW, Sinclair BJ, Meier R. 2000. Morphology and terminology of Diptera larvae. In: Contributions to a Manual of Palaearctic Diptera1: General and Applied Dipterology (eds Papp L, Daras B) pp. 85-161. Science Herald, Budapest.

Cumber RA, Harrison RA. 1959. The insect complex of sown pastures in the North Island III The Diptera as revealed by summer sweep-sampling. New Zealand Journal of Agricultural Research, 2: 741-762. de Meijere JCH. 1900. Ueber die larve von Lonchoptera. Ein Beitrug zur Kenntniss der cyclorrhaphen Dipteranlarven. Zoologissche Jahrbucher Abteilung für Systematik, Geographie und Biologie der Tiere, 14: 87- 132.

Drake CM. 1996. The larva and habitat of Lonchoptera nigrociliata (Diptera: Lonchopteridae). Dipterists Digest, 3: 28-31.

Ferrar P. 1987. A guide to the breeding habits and immature stages of Diptera Cyclorrhapha. Entomograph Vol. 8. Brill, Leidon.

Foote BA. 1991. Lonchopteridae (Lonchopteroidea) Spur-winged flies. In: Immature Insects Vol. 2 (ed. Stehr FW) pp. 787-788. Kendall/Hunt, Dubuque, Iowa.

The Weta 52: 17-24 24

Harrison RA. 1950. Occurrence of Lonchoptera dubia Curran in New Zealand (Diptera: Lonchopteridae). Transactions of the Royal Society of New Zealand, 78: 449-450.

Jones MG. 1979. Observations on the biology of Lonchoptera lutea Panzer (Diptera: Lonchopteridae) from cereal crops. Bulletin of Entomological Research, 69: 637-643.

Klymko J, Marshall SA. 2008. Review of the Nearctic Lonchopteridae (Diptera), including descriptions of three new species. The Canadian Entomologist, 140: 649-673.

Ochman H, Stille B, Niklasson M, Selander RK. 1980. Evolution of clonal diversity in the parthenogenetic fly Lonchoptera dubia. Evolution, 34: 539-547.

Omelkova M, Syrovatka V, Kroupalova V, Radkova V, Bojkova J, Horsak M, Zhai M, Helesic J. 2013. Dipteran assemblages of spring fens closely follow the gradient of groundwater mineral richness. Canadian Journal of Fisheries and Aquatic Sciences, 70: 689-700.

Peterson BV. 1987. Lonchopteridae. In: Manual of Nearctic Diptera Vol. 2. (eds McAlpine JF, Peterson BV, Shewell GE, Teskey HJ, Vockeroth JR, Wood DM). Monograph/Agriculture Canada, 28: 675-680.

Smith KGV. 1996. Family Lonchopteridae. In: Evenhuis NL (ed.) Catalog of the Diptera of the Australasian and Oceanian Regions (online version) www.hbs.bishopmuseum.org/aocat/aocathome (last viewed 10 August 2017).

Stalker HD. 1956. On the evolution of parthenogenesis in Lonchoptera (Diptera). Evolution, 10: 345-359.

Whitten JM. 1956. The tracheal system of the larva of Lonchoptera lutea Panzer (Diptera: Lonchopteridae). Proceedings of the Royal Entomological Society of London, 31: 105-108. 25 Nicholas Martin Interesting insects and other invertebrates: a guide to the Landcare Research internet factsheets

Nicholas A. Martin

Research Associate, Landcare Research, Private Bag 92170, Auckland 11072. Email: [email protected]

Where in New Zealand would you go to find pictures and information about Brown shield bugs (Pentatomidae) that look like the Brown marmorated stinkbug, or to find out if the strange looking ladybird is the very variable Harlequin ladybird that has just arrived in your area, or is it a local species that you had not seen before?

Answer:  go to the internet factsheets, “Interesting Insects and other Invertebrates” http://nzacfactsheets.landcareresearch.co.nz/Index.html,  click on FACTSHEETS  scroll down to Brown shield bug - Dictyotus caenosus, or Harlequin ladybird - Harmonia axyridis  click on ‘Recognition’.

There you will find a brief description of the distinctive features of the taxa and similar looking species, followed by pictures of the relevant life stages of these taxa. With luck there will be a factsheet on the species of Shield bug (7 species) or ladybird (21 species) that you have found.

There are now over 100 of these factsheets on-line with more being added each month. Maybe soon you will be able to make your own contribution? In this brief article I will summarise the features of the series, its current contents and how to find your way around them, followed by a brief history of this project.

The idea behind the factsheets is to make available basic information with appropriate illustrations about our invertebrate fauna to entomologists and non-entomologists including students. The format has certain mandatory fields, but is also flexible (Table 1). It will be possible to add new fields if The Weta 52: 25-32 26 required. Each heading and field only shows up if it is used in a factsheet. Because the factsheets are for non-entomologists I have minimised the use of technical terms and explained them when first used. There is a glossary with fuller explanations of some terms. I have also avoided the standard method of citing references by using a suitable phrase.

All Factsheets have two illustrations at the beginning. I use them to show the adult and juvenile or distinctive plant damage such as leaf mines or galls. There is a provision for a distribution map in the ‘Biostatus and Distribution’ field and an illustration of the life cycle or annual cycle at the start of the ‘Annual Cycle and Life Stages’ field (see Pohutukawa gall fly - Fergusonina metrosiderosi). All other illustrations come at the end of the relevant field. All images are enlarged by clicking on them. You can then scroll through the enlarged images.

Because I wanted to make the factsheets ‘user friendly’ I decided to put a common name at the start of the title. I appreciate that this may make them difficult to find when scrolling through the list of titles under the ‘FACTSHEETS’ tab. However, we have provided a SEARCH function that has turned out to be much more useful than I initially expected. You can search on the current name of an organism or its synonym, its classification (technical and common names), its plant hosts (scientific and common names, plant families), its biostatus, main feeding behaviour. The SEARCH window has the available filters on the left and the results of the search on the right. For each species found in a search, the name is accompanied by information from the ‘Biostatus and Distribution’ field and the first image which is always an illustration of the organism, usually the adult (Fig. 1).

The Factsheets are available for all invertebrates, though only insects and mites have been entered so far. If you enter ‘Arthropoda’ in the search box, you will get a summary of the contents of the factsheets in the ‘Available filters’ on the left, for example, the number of factsheets for each Arthropod Order (Fig. 2), Feeding type and Biostatus (Fig. 3).

An example of the flexibility of these factsheets is the uses to which the field ‘Additional Information’ can be put. If you search on ‘Research Project’ you will get a list of factsheets where this heading has been used. Open the Factsheet, click on Additional Information, and you should find a suggestion of a topic that could provide an interesting project for someone. Other 27 Nicholas Martin headings used in the ‘Additional Information’ field include ‘Biological Control’, ‘Behaviour of moths’, ‘Why Stink bugs’, ‘Effect of giant willow aphid honeydew on bee honey’ and ‘Why are some nymphs hairy and others smooth?’.

One of the big advantages of the internet over printed publications is that it is easy to make changes and keep a factsheet up-to-date. All these factsheets show when they were last revised. However, it is also important to know what has been changed. For example, is it the addition of a photograph, correction of an error or the addition of new information? Every change to these factsheets is documented in ‘Update History’ field. I have just completed the change of scientific name of a moth, Poroporo fruit borer. Luckily the common name stays the same so non-entomologists will still be able to find it easily.

Table 1. Fields available in the factsheets are listed in order in which they are seen. Fields are only shown if there is content in the factsheet being read.

* Mandatory fields

Unique Identifier* Two Images* Display Title* Recognition* Author* Life Stages and Annual Cycle* Date* Biostatus and Distribution* Classification* Natural Enemies* (up to 9 fields, 4 mandatory) Taxonomic Name* Host Plants Common Names* Honeydew feeding Suggested Citation* Prey – Hosts Synonyms Control Taxonomic Notes Other Images Browse* (for searching) Bug Signs Information Sources Acknowledgements Update History

The Weta 52: 25-32 28

Figure 1. Two results from a search for ladybirds. The text is from the ‘Biostatus and Distribution’ field’, and the figure is the first image of the Factsheet.

Figure 2. The number of factsheets for species in each Arthropod order (as of 9 September 2017).

29 Nicholas Martin

Figure 3. The number of factsheets for two types of biological characters, Feeding type and Biostatus (as of 9 September 2017).

Origins of the Factsheets

The ABOUT tab gives the official origins of the factsheet series as follows: ‘The development of these factsheets was made possible by funding from the TFBIS (Terrestrial and Freshwater Biodiversity Information Systems) Fund, administered by the Department of Conservation. Landcare Research developed the software to create, store and generate the factsheets and Nicholas Martin of Plant & Food Research, in conjunction with Landcare Research, developed the content specifications. Feedback from people representing end-user groups was an important part of the process, and we thank them for their valuable input.’ However, my wish to produce factsheets on insects and mites developed from an idea I had about 20 years ago.

In the 1990s there was a lot of discussion about biodiversity and how to measure it. All methods then - as now - were based on an arbitrary selection of taxa and required expensive expertise (e.g. Holdaway et al. 2017). In the late 1990s, while working on pest control with the vegetable industry, I read about a UK Super Market scheme to have a superior brand based on The Weta 52: 25-32 30

Environmentally Friendly methods of production and requiring plans to protect the local wild life (e.g. rabbits, hares and foxes). Such schemes require independent monitoring. How could we develop such a scheme in New Zealand that included protection and enhancement in indigenous habitats that could be assessed by a consultant? I felt that it needed to go beyond plants and birds.

At that time, I was interested in leaf miners and realised that if a consultant was assessing the species of plants present in a habitat, they could also identify and record the leaf miners on the plants if they had a pictorial guide for the required plant species. This concept expanded to include plant galls and other forms of unique plant damage. I created a database to handle the information on invertebrate-host plant associations and the photographs. After about 2 years funding, the active work on the project ceased until there was an opportunity through TFBIS to put the database, named Plant- SyNZ™, on line. I then thought that if people were finding galls and leaf mines, they would want to know what the invertebrate causing the damage looked like: hence the idea for simple Factsheets.

The original idea for plant species-based pictorial identification guides was for 1-2 pages of laminated A4 paper. However, the advent of relatively cheap tablet computers has brought the concept into the 21st Century. Tablet computers enable the pdf pictorial identification guides to have many more photos and to include fungi and other micro-organisms as well as predators and parasitoids of the herbivores found on the plant. Tablet computers are also good for scrolling through the photos, it is easy to enlarge the images up to three times, and see much more detail. Each of the new identification guides are still for a single plant species. The original new Plant-SyNZ guides for New Zealand Flax and Cabbage tree have just been updated and four more plants have been added. With the current concern about protecting the environment from the excesses of some farmers and growers, maybe the time has come to implement the original idea behind the internet factsheets?

The Future

IT experts at Landcare are currently working to improve the entry of information into the Factsheet Database. Once that has happened, it will be possible to make the database available outside of the Landcare Research 31 Nicholas Martin

Tamaki Campus. The Entomology Group of Landcare Research will continue to have a moderating role.

Acknowledgements

TFBIS (Terrestrial and Freshwater Biodiversity Information Systems) Fund, administered by the Department of Conservation. Landcare Research, and The Strategic Science Investment Fund to Landcare Research from the Ministry of Business, Innovation and Employment’s Science and Innovation Group have provided funding for this project. The New Zealand Institute for Plant & Food Research Limited continue to provide resources for taking photographs. Thomas Buckley provided helpful comments on the manuscript.

References

Holdaway RJ, Wood JR, Dickie IA, Orwin KH, Bellingham PJ, Richardson SJ, Lyver PO, Timoti P, Buckley TR. 2017. Using DNA metabarcoding to assess New Zealand’s terrestrial biodiversity. New Zealand Journal of Ecology 41 (2): 218-225.

Internet Factsheets: Interesting Insects and other Invertebrates nzacfactsheets.landcareresearch.co.nz/Index.html

Plant-SyNZ™ internet database: herbivore - host plant associations plant-synz.landcareresearch.co.nz/SearchForm.aspx

Plant-SyNZ™: an invertebrate herbivore biodiversity assessment tool plant-synz.landcareresearch.co.nz/index.asp The Weta 52: 25-32 32

33 Simon Hodge & John Marris Colour preferences in sand dune insects found on ice plant flowers

Simon Hodge1* & John Marris2

1 - Faculty of Agriculture and Life Sciences, Lincoln University, Canterbury, New Zealand. 2 - Bio-Protection Research Centre, Lincoln University, Canterbury, New Zealand.

*Email: [email protected]

Introduction

The ice plant (Carpobrotus edulis (L.), family Aizoaceae) is a succulent perennial native to South Africa that has been introduced widely around the world to prevent erosion of sandy or loose soils. At New Brighton sand dunes near Christchurch, dense mats of C. edulis now form part of a flora dominated by exotic plants, including marram grass (Ammophila arenaria L.), tree lupin (Lupinus arboreus Sims) and purple ragwort (Senecio elegans L.).

Exotic plants frequently provide acceptable habitats for diverse assemblages of insects and spiders (e.g. Prasad & Hodge 2013). The abundance and diversity of nectar- and pollen-feeding insects can show positive responses to exotic plants that provide extended flowering periods to native plant species (Showler 1989; Stary and Tkalcu 1998; Neinhuis et al. 2009; Vila et al. 2009). Other insects, such as ground beetles, have exhibited negative responses to the presence of exotic plant species in their habitat (e.g. Topp et al. 2008).

Flower-visiting insects frequently exhibit preferences for certain flower characteristics, such as shape, scent and colour, and it is often assumed there is some adaptive role to flower selection in terms of improved insect fitness. Flower colour is one of a range of signals plants use to attract pollinators and insects can learn to associate flower colour with a high quality resource (Raine & Chittka 2007). The Weta 52: 33-39 34

At New Brighton, a number of species of arthropods have been observed visiting ice plant flowers, including endemic Coleoptera (Lagrioida brouni Pascoe; Inophloeus rubidus Broun) and native arachnids (Cheiracanthium stratioticum Koch; Oxyopes gracilipes (White, 1849)) (Hodge et al. 2017). At least two colour forms of ice plant flowers occur at New Brighton: a common yellow form and a less common pink/purple form (Figure 1). This study compared overall arthropod occupation rates of the two flower colours and assessed whether individual insect species exhibited a preference for either colour form.

Figure 1. Two colour forms of the iceplant, Carpobrotus edulis, at New Brighton sand dunes

Methods

A total of 3360 yellow flowers and 240 purple flowers of Carpobrotus edulis were inspected for arthropods at New Brighton sand dunes (-43.522, 172.736) between 9 November and 10 December 2014 (Hodge et al. 2017). This highly unbalanced sampling was undesirable, but reflects the unequal abundance of the two colour forms at this site. Insects and spiders were collected from flowers using a battery operated aspirator, with the exception of bumblebees which were identified to species on site.

The native New Zealand ice plant Disphyma crassifolium, which has smaller (2–4 cm diameter) white or pink flowers and a more slender (4 mm), cylindrical stem compared to C. edulis, also occurs at New Brighton 35 Simon Hodge & John Marris

(Chinnock 1972). Carpobotus edulis has flowers of 8–10 cm diameter and much thicker (12 mm) and angular stems than the native species. We are confident that no ‘pure’ D. crassifolium were sampled but cannot rule out that some hybrids with intermediate properties of the two species may have been included in the survey.

Overall occupancy rates and occupancy by the three most commonly encountered insect species, on yellow or purple flowers were compared using Fisher’s Exact test.

Results

A total of 478 individual arthropods belonging to 32 species were recorded on C. edulis flowers (Hodge et al. 2017: Appendix). The overall occupancy rate of flowers was 10%, but yellow flowers (10.4%) were over twice as likely to have arthropods compared to purple flowers (4.6%) (Fisher’s exact test, P = 0.002) (Table 1).

The most commonly-occurring species was a pollen beetle in the family Melyridae, close to the genus Dasytes sp., which made up 58% of the total invertebrate numbers. The bumble bee Bombus terrestris (15% of records) and a long-horned grasshopper, Conocephalus bilineatus (8% of records) were also relatively common in the collections.

Dasytes sp. displayed a strong (i.e. fourteen-fold) preference for yellow flowers over purple in terms of presence-absence occupancy (Table 1). Conversely, Bombus terrestris exhibited a less-dramatic (≈ 2-fold) but statistically significant preference for purple flowers. The equally substantial colour preference (2-fold) displayed by C. bilineatus for yellow flowers was not identified as being statistically significant (Table 1). The ambiguity of this last result reflects the smaller sample size obtained for this species, and more data are required to clarify whether this species exhibits a real colour preference or not.

The Weta 52: 33-39 36

Table 1. Levels of occupancy (%) of yellow (n = 3360) and purple (n = 240) ice plant flowers at New Brighton sand dunes. P-value obtained from Fisher’s Exact Test based on raw count data of occupied and unoccupied flowers of each colour form.

Occupation (%) Total Total Yellow Purple P Abundance Total 478 10.00 10.39 4.59 0.002 arthropods ‘Dasytes’ sp. 276 5.22 5.56 0.42 < 0.001 Bombus 70 1.94 1.82 3.75 0.049 terrestris Conocephalus 39 1.03 1.07 0.42 0.513 bilineatus

Discussion

The strongest flower colour preference by a single insect species was that of Dasytes sp. for yellow ice plant flowers, which is similar to that described for other species of pollen beetles (e.g. Meligethes in Doring et al. 2012). The preference for purple flowers by Bombus terrestris is also similar to previous studies where a bias towards violet or blue flowers has been described for this species (Raine & Chittka 2007).

Exhibiting a colour preference suggests that visiting flowers of one colour are perceived as more profitable to the insect than visiting flowers of the other. We did not perform any analysis to distinguish whether the purple or yellow flowers contained different amounts of resources such as pollen or nectar. Indeed, as no consistent colour preference was identified across species, these benefits appear to vary among different insect species.

Assemblages of pollinators and other plant-visiting insects are often sampled using pan traps, the colour of which can influence both the abundance and species composition of the animals obtained (Campbell & Hanula 2007; Vrdoljak & Samway 2012). The results of our study suggest that using any one colour of pan trap to study the flower-visiting insects at New Brighton dunes would lead to significant under-collecting of some 37 Simon Hodge & John Marris species, and a range of trap colours is required to obtain satisfactory coverage of all the insect species present.

We focussed on the arthropods visiting one species of flower; there are a number of other species of colourful flowers present at New Brighton dunes and potentially hundreds of species of insects that may choose to visit them (MacFarlane 2005). Further study is required to build a more complete picture of the networks of flower-visiting insects that occur at this site and also to provide more information on the use of exotic plant species by endemic New Zealand insects.

Acknowledgements

The authors wish to thank Jason Roberts and Antony Shadbolt of Christchurch City Council, for permission to carry out the survey at New Brighton and providing a copy of the report by Rod MacFarlane.

References

Campbell JW, Hanula JL (2007) Efficiency of Malaise traps and colored pan traps for collecting flower visiting insects from three forested ecosystems. J Insect Conserv 11: 399-408.

Chinnock RJ (1972) Natural hybrids between Disphyma and Carpobrotus (Aizoaceae) in New Zealand. New Zeal J Bot 10: 615-625.

Doring T, Skellern M, Watts N, Cook SM (2012) Colour choice behaviour in the pollen beetle Meligethes aeneus (Coleoptera: Nitidulidae). Physiological Entomology 37: 360–368.

Hodge S, Curtis N, Vink CJ, Marris J, Brown SDJ (2017) Native arthropods on exotic sand dune flowers: consideration of sample size and number for investigating rare species and sparse communities. Arthropod-Plant Interactions 11, 691-701.

Macfarlane RP (2005) New Brighton Sand Dune Invertebrates. A report prepared for Christchurch City Council. Christchurch, New Zealand.

The Weta 52: 33-39 38

Neinhuis CM, Dietzsch AC, Stout CJ (2009) The impacts of an invasive alien plant and its removal on native bees. Apidologie 40: 450-463.

Prasad AV, Hodge S (2013) The diversity of arthropods associated with the exotic creeping daisy Sphagneticola trilobata in Suva, Fiji Islands. Entomol Month Mag 149: 155- 161.

Raine NW, Chittka L (2007) The Adaptive Significance of Sensory Bias in a Foraging Context: Floral Colour Preferences in the Bumblebee Bombus terrestris. PLoS ONE 2(6): e556. doi:10.1371/journal.pone.0000556.

Showler K (1989) The Himalayan balsam in Britain - an undervalued source of nectar. Bee World 70: 130-131.

Stary P, Tkalcu B (1998) Bumble-bees (Hym. Bombidae) associated with the expansive touch-me-not, Impatiens glandulifera in wetland biocorridors. Anzeiger fur Schadlingskunde Pflanzenschutz Umweltschutz 71: 85-87.

Topp W, Kappes H, Rogers F (2008) Response of ground-dwelling beetle (Coleoptera) assemblages to giant hogweed (Reynoutria spp.) invasion. Biol Invasions 10: 381-390.

Vila M, Bartomeus I, Dietzsch AC, Petanidou T, Steffan-Dewenter I, Stout JC, Tscheulin T (2009) Invasive plant integration into native plant- pollinator networks across Europe. Proc Roy Soc B 276: 3887-3893.

Vrdoljak SM, Samways MJ (2012) Optimising coloured pan traps to survey flower visiting insects. J Insect Conserv 16: 345–354.

39 Simon Hodge & John Marris

APPENDIX

List of arthropod species observed on ice plant flowers at New Brighton (see Hodge et al 2017 for further details).

En – endemic; Na – native; In –introduced.

COLEOPTERA: Anthicidae Lagrioida brouni Pascoe, 1876 En; Chrysomelidae Bruchidius villosus (Fabricius, 1792) In; Coccinellidae Coccinella undecimpunctata (Linnaeus, 1758) In; Curculionidae Inophloeus rubidus Broun, 1881 En; Melyridae ‘Dasytes’ sp. En; Scarabaeidae Pyronota sp. En. DIPTERA Agromyzidae Cerodontha australis Malloch, 1925 Na; Bibionidae Dilophus nigrostigma (Walker, 1848) Na; Calliphoridae Calliphora vicina Robineau-Desvoidy, 1830 In; Canacidae Tethinosoma fulvifrons (Hutton, 1901) Na; Chironomidae Chironomus sp. Meigen, 1803 Na; Chloropidae Apotropina tonnoiri (Sabrosky, 1955) Na; Chloropidae Aphanotrigonum huttoni (Malloch, 1931) Na; Dolichopodidae Parentia sp. Hardy, 1935 Na; Drosophilidae Drosophila pseudoobscura Frol. & Ast., 1929 In; Drosophila simulans Sturtevant, 1919 Na; Drosophilidae Scaptomyza sp. Hardy, 1950 In; Empidae Thinempis sp. Bickel, 1996 Na; Ephydridae Psilopa metallica (Hutton, 1901) Na; Hecamede granifera Thompson, 1869 In; Ephydridae Hydrellia tritici Coquillett, 1903 Na; Syrphidae Platycheirus sp. Le Peletier & Serville, 1828 Na; Teratomyzidae Teratomyza neozelandica Malloch, 1933 En. HEMIPTERA Psyllidae. HYMENOPTERA Apidae Bombus terrestris (Linnaeus, 1758) In; Braconidae; Eulophidae. Lepidoptera Tortricidae. ORTHOPTERA Tettigoniidae Conocephalus bilineatus (Erichson, 1842) Na. ARACHNIDA Eutichuridae Cheiracanthium stratioticum Koch, 1873 Na; Lyniphiidae Microctenonyx subitaneus (Pickard-Cambridge, 1875) In; Oxyopidae Oxyopes gracilipes (White, 1849) Na. The Weta 52: 40-54 40 Population of the ngaio weevil (Anagotus stephenensis) on Stephens Island/Takapourewa

Mark Anderson

Marlborough Boys College, 5, Stephenson Street, Blenheim, 7201 Email: [email protected]

Introduction

Stephens Island is a remote island in western Cook Strait, three kilometres north of D’urville Island. It was heavily deforested by fire and grazing following the building of a lighthouse in 1894, removing around 90% of the forest cover. Presently, the island consists of small patches of original forest, which cover approximately 12% of the island area (East et al., 1995). The remainder of the island’s vegetation consists of vines, tall and short grass, scrub and areas of regenerating forest, where trees have been planted since 1990.

The island has a rich diversity of reptiles, with the largest population of tuatara in the world, three species of gecko, and four species of skink. The presence of these reptiles in unusually large numbers is attributed to the presence of over one million burrowing fairy prions, whose nutrient enrichment of the soils provides minerals for enhanced vegetation growth and an increased density of ground dwelling invertebrates (Brown, 2001; Mulder et al., 2001). Stephens Island has never had mainland mammalian predators, other than some cats which were removed in 1925 (Brown, 2001). The ngaio weevil is a large, nocturnal, flightless weevil ranging in length from 20 to 31mm. The species was discovered in 1916 on Stephens Island and is considered a relict population (Kuschel et al., 1996). It has a historic range as far as South Canterbury and was once common, having been found in reasonable numbers (39) in cave deposits produced by the extinct laughing owl (Kuschel et al 1996). Intensive searches in 1971 and 1995 saw only one or two specimens, indicating it had become rarer than when discovered in 1916 when 15 specimens were collected from Stephens Island (Kuschel et al., 1996). It has previously been seen on ngaio (Myoporum laetum) and feeding notches on leaves are a sign of presence (McGuiness, 2001). 41 Mark Anderson

Methods

A search area was determined following two visits to Stephens Island in July and September 2011. Because of seabird burrows, the search area was limited to the tracks, fence lines and roads on the island and one area of retired grass paddock that was not burrowed. Searching occurred over 31 nights. This began at least one hour after sunset and involved scanning the leaves and stems of ngaio with a head torch from the ground upwards.

The search area used to ascertain the presence or absence of ngaio weevils consisted of a core area which was searched on 28 nights and other areas which were searched as time allowed. The core area was determined from observations of four ngaio weevils in September 2011. The other areas were determined by the presence of ngaio trees larger than 1 metre. These other areas were each searched on 3-8 occasions.

Each ngaio weevil was individually marked on their dorsal side, using up to 6 dots of brown/black/white nail varnish to produce a code. They were photographed to determine sex (see Figure 2). The specimen’s mass (accuracy +/- 0.01g) was measured using an electronic balance. Length (accuracy +/- 1mm) was measured at its longest point, from the end of the rostrum to the end of the thorax.

The height above the ground that the weevil was found was recorded with a tape measure. The temperature at the start of the search and towards the end of the search was recorded using a Pasco data logger. When a new site was found for a ngaio weevil, the tree was marked with flagging tape and later measured. The GPS position of each sighting was recorded, along with tree height (estimate) and circumference at the lowest point possible. The host plant species was also recorded.

Weevil Sexing

The sex of Helm’s beech weevil (Anagotus helmsi) is determined by observing the shape of the terminal sternite (McBurney, 1976) and this method was also used to determine the sex of the ngaio weevil. In Anagotus helmsi, McBurney distinguishes the sexes by ‘the male having a terminal sternite approximately half as long as wide, with a concave area apically’. The Weta 52: 40-54 42

The female is described as ‘the length and width are subequal, the surface is uniformly convex’ (Figure 2).

On the third search night, a male and female were seen on the same branch and photographed together (Figure 3) to determine differences in morphology and confirm the gender differences. Following this discovery, the ventral side of each weevil was photographed using a macro feature on a Panasonic digital camera using a head torch for illumination. This photograph was then analysed the following day to determine the sex of each weevil. Data were only entered for individuals where there was a clear photograph and unambiguous shape of the terminal sternite.

Results

Population, distribution and habitat

There were 528 ngaio weevils sighted over 28 nights of searching: this number included 259 marked individuals. Photographs of 248 individuals were analysed to determine sex and the sample consisted of 173 males and 75 females, giving a male:female sex ratio of 2.3:1.

Figure 1. Sex determination in Helms beech weevil (after McBurney, 1976)

43 Mark Anderson

Figure 2. Female (left) and male (right) ngaio weevils

Figure 3. Ngaio weevil in defence posture

The Weta 52: 40-54 44

The population of the ngaio weevil is not evenly distributed across Stephens Island: 92% of individuals were found in one area of relatively recently planted (c. 25 years) ngaio with low numbers seen in ngaio elsewhere. Weevils were not seen in the mature forest areas in Keepers Bush around the two ex-lighthouse keeper’s houses or in mature forest areas on the summit track.

All ngaio weevils were seen either on ngaio trees (85% of sightings) or in the proximity of ngaio. Other invertebrates seen on ngaio included tree weta, Cook Strait giant weta, chafer beetles (seen browsing) and small Lyprobates sp. The ngaio weevils were quite frequently found on grass under ngaio branches (10.6% of sightings) and occasionally on vines growing under or next to branches of ngaio. Sightings occurred on trees of average height of 2.67m (range 1.29-4m) and average circumference of 51cm (range 20- 88cm). These trees were mostly planted trees (82%), the remainder on more mature trees with low branches that overhung fences or paths. All of the ngaio weevils observed, were on trees that had leaf litter under the canopy and low branches that touched the grass or vines growing underneath. The ground cover in mature bush was devoid of leaf litter due to the presence of petrels and their burrows.

Behaviour

Ngaio weevils are nocturnal and were observed on Stephens Island from 81 minutes after sunset at 9:15pm (Nelson data) until 3.00am. One individual was kept inside a large dark coloured bucket with a lid in a brightly lit room for further study. It retained a pattern of emergence at various times of night and crawling into the leaf litter at the bottom of the container before daylight, where it remained until dusk. It did not emerge from the leaf litter during the daylight when the bucket followed the sunrise/sunset patterns. When the bucket was moved to a darkened room, the pattern was disrupted and the weevil was seen actively feeding at 2pm and 7.11pm. Despite checking every 2-4 hours, in one instance, it was not seen to emerge for 47 hours.

When the weevil was found during daytime searches of the leaf litter in the bucket, it was found 20-30mm deep in the leaf litter, motionless, clinging to leaves or bark. In the field, one individual was found on two occasions during the daytime (10.40am and 4.00pm) following searches of the grass 45 Mark Anderson underneath low-hanging ngaio branches. It was clinging to grass stems on both occasions. During evening searches, it was common (56 sightings) to see ngaio weevils on grass underneath ngaio branches. Some of these (three sightings) were very close to the ground (0.02-0.04m). Ngaio weevils were seen at an average height above the ground of 0.59m (range 0.02-1.70m), despite good visibility on most trees up to 3m. They are most frequently seen at the lower end of this height range with 90% of sightings occurring below 1m.

A variety of behaviours was recorded, including moving up and down plant stems, feeding, playing dead, dropping off a leaf onto the ground and standing still on leaves. Most observations were of ngaio weevils standing motionless, when even a small amount of light (including red light) illuminated them. Occasionally, they would continue their behaviour of moving or feeding when illuminated. When handled, they would usually remain completely motionless with legs tucked underneath their body (Figure 3). This behaviour was also observed in the weevils found hiding in leaf litter or grass during the day. Occasionally they would remain active and walk away from the electronic balance or hand. On a few occasions, they would release their grip of the leaf when an attempt to collect them was made and fall into the leaf litter or grass below and remain motionless. On other occasions, they would grip onto leaves or stems tightly and would not let go.

Three pairs of ngaio weevils were observed mating, with the male weevil on top of the female (Figure 4). Male and female weevils were frequently seen close together on the same area of a tree.

Marked individuals were frequently seen on consecutive nights, although some re-sightings did not occur for up to three weeks. On a follow-up visit to the island in December 2014, one marked weevil was observed that was marked 3 years and 2 months previously. Weevils were observed to move between adjacent ngaio trees, but never to move distances further than 5m. Most of the repeat sightings of individuals occurred on the same tree.

Feeding

Ngaio Weevils have been observed feeding on ngaio leaves and leaf stalks. The feeding sign is a large semi-circular notch in foliage which has a deeper The Weta 52: 40-54 46 indentation on the rostrum side (Figure 5). The weevils feed by continuously browsing on the same area, making deeper and deeper incisions. The captive weevil appeared to be chewing leaves off at the stalk and allowing them to drop onto the floor of the bucket. A small number of leaves were chewed along the leaf stalk, but not enough to cause them to drop. A secondary feeding sign is the production of dark brown, fairly liquid, faecal matter which has been observed on leaves with ngaio weevils and being produced by the weevils.

Sex differences

There was a large difference in the numbers of sightings of male and female weevils. Males were sighted on 361 occasions (71% of sightings) and females on 144 occasions. Of the sample of individuals that were sexed, 173 were males and 75 were females, and therefore the sample (70 % males) appeared to be representative of the sex ratio seen in the wild population. Length and mass showed a linear relationship for both males (rP = 0.75, P < 0.001) and females (rP = 0.79, P < 0.001) (Figure7). There were significant differences in the length and in the mass of male and female Ngaio Weevils. Male average length was 24.9mm (range 20-29mm); female average length was 27.8mm (range 23-31mm) (t = 12.5, P < 0.001). The average male mass was 0.68g (range 0.36-1.05g); female average mass was 0.96g (range 0.48- 1.30g) (t = 12.0, P < 0.001).

Discussion

The development of new habitat for ngaio weevils has allowed their numbers to increase dramatically over the last 20 years. Habitat loss due to deforestation, which began in 1894, would likely have caused a population decline that continued after 15 specimens were collected when the species was discovered in 1916. In 1989, there had not been any sightings of the ngaio weevil for 18 years (unpublished note to J Dugdale, 1989). The actual population count of 259 individuals over 28 nights is the highest ever recorded. The population has benefitted from the reforestation of the island and specifically by the planting of ngaio trees. The distribution of the weevils is not uniform, possibly due to the preference for feeding on trees with low branches that allow easy access from their daytime refuges in the leaf litter/dense grass (Figure 8). Many areas of the island were not searched 47 Mark Anderson due to the fragile nature of the ecosystem and further surveying is therefore required.

The area where 92% of individual weevils were found, the Ruston Spur, is an area of planted ngaio (in approx. 1990), which is separate, but close to the mature Ruston Bush. Some movement of ngaio weevils has occurred along the ground in this area, as the trees are planted in groups. Within 300m is a similar sized area of planted ngaio where this species is currently absent, providing potential for further dispersal and population increase.

Tuatara are an obvious predator of the ngaio weevil. Adult tuatara have been observed on top of muhlenbekia and grass next to ngaio branches that host weevils. Tuatara are ambush predators, that rely on movement of prey for detection and have sight perception capabilities comparable to birds and mammals (Woo, 2004), and can prey upon the movement of ngaio weevils as they move onto low branches. In the area of highest weevil population density, the population of tuatara and fairy prion appears to be much lower than the other core search areas on the island. Fairy prion densities are linked to tuatara densities (Markwell, 1998) due to the reliance on the seabird as a food source for the reptiles. The very low density of fairy prion burrows in the core search area could explain the higher density of ngaio weevils in this area. Conversely, in other areas with suitable ngaio but with higher fairy prion/tuatara populations, the weevils occurred in considerably lower numbers.

Tuatara densities are highest in mature forest (18 sightings per person per hour), low in vines (7 sightings per person per hour) and much lower in tall grass (3 sightings per person per hour) (East et al, 1995). The tall grass in the core area provides a protective screen for access for weevils onto the tree, which is absent in the mature forest. The canopy forest is almost devoid of an understory with seedlings negatively affected and leaf litter removed by burrowing fairy prions (Mulder, 2001). This could explain the lack of sightings in the mature forest where visibility is increased for predators and the preference for forest margins, where grass and vines can grow and allow access for the species onto their host plant and screening from predators.

The lack of dead ngaio on Stephens Island was thought to be a possible limiting factor for weevil population growth due to the larval stage requirements (McGuinness, 2001). In the area where most sightings were The Weta 52: 40-54 48 made, there are very small amounts of dead ngaio and distances of nearly 50m to mature bush, where there is a greater amount. Large holes have been observed in dead ngaio (Figure 6) close to ngaio weevil sightings, however, the size of emergence holes was smaller (5.5mm) than those of the smaller weevil Anagotus helmsi, which have been observed to ‘not exceed 10mm’ (McBurney 1976). Ngaio weevils have only been observed eating ngaio on Stephens Island. Their diet may include small amounts of karaka, a food source of its sister species Anagotus turbotti, although it is unlikely that this would be in large enough amounts to host a large population. The ability of nagio weevil to use karaka should be tested, as this may provide opportunities for their translocation.

Whilst feeding, this species was observed to produce faecal matter, which is deposited on the leaves and may attract the opposite sex or other individuals to aggregate for mating, as other species have been observed to aggregate on production of pheromones in their faecal matter (Barnes, 1989).

The nocturnal ngaio weevil was observed to emerge from the leaf litter/grass, where it stays motionless during the day, and their daytime behaviour provides a safe place for them to hide away from predatory birds. The weevils climb up the surrounding plants, which are grasses, muhlenbekia or spinach and access ngaio in this manner, and then use the stem and adjoining leaves to move around the ngaio, having been frequently observed on the stem and moving between leaves. The weevils usually stop activity when illuminated by torchlight and bury themselves in leaf litter to move away from light during the day, and their emergence patterns are strongly affected by light levels, as shown by the individual that began to emerge during the day in a darkened environment. However, sightings of this species were unaffected by the presence of moonlight.

Ngaio weevils did not appear to be a mobile species during this study. On rare occasions, weevils were observed to have moved to adjacent trees and on one occasion approximately 5m away from the original site. They were not observed in transit on grass between ngaio trees, although this was not deliberately searched. Generally, each individual would be seen on the same aspect of the same bush on each occasion. There was not a single sighting of a ngaio weevil in two different locations greater than 5m apart. It is possible that there is a seasonal movement that occurs outside of this study period, when females disperse to lay eggs on suitable ngaio. A minimal 49 Mark Anderson movement between trees for this large flightless weevil may be a behavioural adaptation, as the chances of surviving predation by tuatara would be minimised (Kuschel, 1996).

There was a bias towards males over females in the ngaio weevil population on Stephens Island (male:female =2.3:1). It is possible that the energy requirements of the larger female larvae are greater than male specimens and thus they mature more slowly or fail to reach maturity. Alternatively the sex ratio could be caused by higher levels of predation on the larger and possibly slower-moving female.

There was a large mass and length difference between male and female ngaio weevils and the data collected on mass and length should allow easy identification of sex by someone collecting weevils for translocation. Of the weevils that were >0.9g mass, 94% were female. Of the weevils that were 28mm in length or greater, 91% were female. Identification using these parameters would allow over 85% of weevils to be correctly sexed in this study without studying the terminal sternite.

There was also a large range of sizes within each gender. Males ranged from 20-29mm and females from 23-31mm. These large size ranges may be caused by differences in larval nutrition (Trumble et al., 1978) or temperature (Guppy, 1974) during development and may reflect a lack of optimum conditions throughout the study area for this species. There may be an imbalance of smaller sizes during the study period due to smaller individuals emerging earlier due to lower fat reserves. Emergence of adults may have continued throughout the study period, as sightings of new individuals occurred on most days, even in areas that had very good visibility of trees. This species was thought to live for two years (Dugdale, 1993) and so would have a cycle of mating and feeding through two seasons of activity, with fertilised eggs deposited throughout this season. A sighting of a marked weevil three years and two months after being marked records a new longevity for this species.

Mating behaviour was been observed in this study in October and November and in December. This has important impacts on any translocation proposal. If a small number of this species were to be moved to a tuatara-free island in the Marlborough Sounds, it should be before mating and egg deposition has occurred, so that this process can occur in the new habitat. This would The Weta 52: 40-54 50 allow successful breeding in case this species has less than a two year life span as an adult.

Figure 4. Ngaio weevils mating

Figure 5. Ngaio weevils feeding

51 Mark Anderson

Figure 6. Wood boring in ngaio

References

Anon, 1989. Notes to JS Dugdale about Anagotus stephenensis and Mecodema punctellum on Victoria University headed notepaper, Stephens Island Records, Department of Conservation, Sounds Area Office, Picton.

Barnes BN, Capatos D, 1989. Evidence for an aggregation pheromone in adult frass of banded fruit weevil, Phlyctinus callosus (Schoenherr) (Col., Curculionidae). Journal of Applied Entomology 108: 512–518.

Brown D, 2001. Stephens Island; ark of the light. Cloudy Bay Publishing, Blenheim, NZ.

Dugdale, JS, 1993. Letter to I Millar, Stephens Island Records, Department of Conservation, Sounds Area Office, Picton.

East TK, East M, Daugherty CH, 1995. Ecological restoration and habitat relationships of Reptiles on Stephens Island, New Zealand. New Zealand Journal of Zoology 22: 249-261. The Weta 52: 40-54 52

Guppy JC, Mukerji MK, 1974. Effects of temperature on developmental rate of the immature stages of the Alfalfa Weevil, Hypera postica (Coleoptera:Cuculionidae). Canadian Entomologist 106: 93-100.

Kuschel G, Worthy TH, 1996. Past distribution of large weevils Coleoptera:Curculionidae) in the South Island, New Zealand, based on Holocene fossil remains. New Zealand Entomologist 19:15-19.

Markwell TJ, 1988. Relationship between Tuatara Sphenodon punctatus and Fairy Prion Pachyptila turtur densities in different habitats on Takapourewa (Stephens Island), Cook Strait, New Zealand. Marine Orithology 26: 81-83.

Marris J, 2001.Beetles of Conservation Interest from the Three Kings Islands, A report submitted to the Department of Conservation, Northland Conservancy Ecology and Entomology Group, Lincoln University.

May BM, 1987.Immature stages of Curculionoidea (Coleoptera):Rearing records 1964-1986. New Zealand Entomologist 9:44-56.

McBurney GR,1976.Notes on the Life History and Distribution of Anagotus helmsi (Coleoptera:Curculionidae). New Zealand Entomologist 6: 177- 181.

McGuinness CA, 2001. The conservation requirements of New Zealand's nationally threatened invertebrates. Threatened Species Occasional Publication 20: 658.

Mulder CPH, Keall SN, 2001. Burrowing seabirds and reptiles: impacts on seeds, seedlings and soils in an island forest in New Zealand.Oecologia 127: 350-360.

Pawson S, 2002. Survey of threatened Coleoptera on Stephens Island, Stephens Island Records, Department of Conservation, Sounds Area Office, Picton.

Sibul I, Voolma K, 2004. The Abundance, seasonal dynamics and Sex Ratio of the large Pine Weevil, Hylobiusabietis (L.), in reforestation areas: 53 Mark Anderson

Assessment by ground pitfalls and baited pitfall traps. Transactions of the Faculty of Forestry, Estonian Agricultural University 37: 56–62.

Trumble JT, Kok LT, 1978. Laboratory propagation of Ceuthorrhynchidius horridus (Coleoptera: Curculionidae), An introduced Weevil for Biocontrol of Caduus Thistles, Canadian Entomologist 110:1091-1094.

Ure G, 1994. Stephens Island Weevil, Summary of A Survey Carried Out From April1993-March 1994, Report compiled for The Department of Conservation, Stephens Island Records, Sounds Area Office, Picton.

Woo KL, 2004. Acquisition of a Learned Operant and Critical Flicker- Fusion Rate in the Tuatara (Sphenodon spp.) Unpublished Msc Thesis, VUW, New Zealand.

Figure 7. Relationship between mass and length of male and female ngaio weevils

The Weta 52: 40-54 54

Figure 8. Tall grass providing access onto ngaio.

55 Chand et al. Egg deposition by Spiralling whiteflies (Aleurodicus dispersus) reduces the stomatal conductance of cassava (Manihot esculenta)

RR Chand1*, AD Jokhan2, R Prakash2

1 - School of Science and Technology, University of Fiji, Lautoka, Fiji islands. 2 - Faculty of Science, Technology and Environment, University of the South Pacific, Fiji islands. *Corresponding author: [email protected]

Introduction

Whiteflies are considered a most damaging pest in all cassava-producing regions, as they are responsible for transmitting plant viruses and directly damaging plants via heavy infestation (Reddy, 2015). Aleurodicus dispersus Russell (Hemiptera: Aleyrodidae) commonly known as Spiralling whitefly, a native to the Caribbean region and Central America (Reddy, 2015; Waterhouse & Norris, 1987). Over 300 plant species from approximately 77 families have been recorded as hosts of Aleurodicus dispersus Russell, worldwide (Lambkin, 1999), and the species is known to have widely spread over North America, South America, Asia, Africa, Australia and in several Pacific Island Countries (PIC). In the South Pacific it is known from Majuro (1986) (Marshall Is), Cook Islands (1984), Fiji (1985), Nauru (1987), Papua New Guinea (1987), Kiribati (June 1988), Tokelau (late 1988) and Tonga (November 1988) (Waterhouse & Norris, 1987).

Whiteflies are considered to be a major agricultural pest, causing damage to crops by altering the growth, photosynthesis, chemical and phenological processes (Boopathi et al., 2015; Nabity et al., 2009). Whiteflies secrete sticky honeydew which at many times results in the formation of dark sooty moulds on leaves. Nymphs also secrete white, waxy flocculent materials which can spread widely. The secretion of honeydews causes premature shedding of leaves, and the sooty moulds hinder photosynthesis by blocking the entry of carbon dioxide into the leaf cells through the The Weta 52: 55-60 56 stomata, which greatly reduces the photosynthetic product values (Henneberry et al., 2007; McAuslane et al., 2004; University of Florida, 2015).

Stomatal apertures control both the water loss from plant leaves and the uptake of CO2 for photosynthesis. Measurements of stomatal apertures are important indicators of plant water status and gas exchange, giving an insight of the plant’s ability to grow and adapt to the changing environmental conditions. Stomatal conductance gives a numerical measure of the water vapour or carbon dioxide passage rates through the stomata or small pores of the plant. This short study aimed to investigate whether stomatal conductance of cassava leaves was affected by infested of Spiralling whitefly eggs.

Materials and Methods

The Spiralling whiteflies (Aleurodicus dispersus) were introduced and bred (3 months after planting) on 6 cassava plants (Manihot esculenta (Crantz)) under glasshouse conditions without any exposure to insecticides. The plants were maintained in the glasshouse for appropriately 6-7 months prior to the study. The temperature for the experimental period (September-December) ranged from 25-30°C with the relative humidity of around 80%. Similarly, 6 plants without whiteflies were used as control plants.

The stomatal conductance in mmol m−2 s−1 was measured on the upper canopy leaves (height =135-150 cm) of the cassava plants using an AP4 Leaf Porometer (Model: AP4 Delta-T Device- Cambridge- UK) between 1100 hours and 1200 hours daily for 7 days. The stomatal conductance was measured as described by Schymanski et al. (2013), as high leaf temperature is best captured during sun flecks around 1100 hours. The mature leaves (not too young and too old) were randomly selected for each day taking leaf size and canopy height into account.

To statistically test the difference between the infested and non-infested leaves, an independent sample t-test was performed at each time point.

57 Chand et al.

Results

The diurnal stomatal conductance of cassava leaves was significantly affected by A. Dispersus egg deposition. Lower conductances were recorded for the infested leaves for all days. The infested leaves had lower mean values of conductance (M= 11.90 mmol m−2 s−1) while the non- infested leaves were associated with higher mean values (M= 17.80 mmol m−2 s−1), (p<0.05).Within the first 3 days of measurements, the conductance trend was similar for both infested and non-infested leaves showing no significant difference. After 3 days the conductances for the infested leaves were significantly low (Figure 1).

The stomata usually maintain the internal partial CO2 pressure in relation to the external pressure (Neves et al., 2006; Ramos & Grace, 1990). Figure 2 depicts how the eggs laid by female whiteflies on cassava leaves affect the stomatal pores. It was roughly estimated that an average whitefly egg blocks 36 stomatal pores in cassava leaves.

Discussion

The findings clearly showed a significant decrease in the stomatal conductance of whitefly infested leaves when compared to non-infested leaves. The significant decrease in the stomatal conductance is due to the eggs being laid on the lower surface of the leaves by the female whiteflies (Figure 2); egg deposition covered the stomata thus blocking its access to light and carbon dioxide. Egg deposition hence reduces the effective surface area of the leaf by lowering the overall productivity.

Other studies have also reported a reduction in the rate of photosynthesis by the infestation of whiteflies. For instance, Lin et al. (1999) reported that the photosynthesis activity is reduced by 50% after 60 days of the introduction of whiteflies on cotton leaves. It was also noted that Silverleaf whitefly infestation reduced cotton foliar photosynthesis rates, where physiological damage occurred with infestation levels of 10-20 adult whiteflies per leaf (Yee et al., 1996). Touhidul Islam and Shunxiang (2009) reported similar findings in eggplant leaves where they revealed that the whitefly infestation lowered the effective leaf area which significantly reduced photosynthesis. Likewise, the photosynthesis activity The Weta 52: 55-60 58 was found to be reduced by 30% in rice plants infested with the planthopper Nilaparvata lugens (Watanabe & Kitagawa, 2000). Overall, this short investigation shows that whitefly infestation can significantly reduce the stomatal conductance of cassava leaves by blocking stomata by egg deposition. Further research is required to examine the subsequent effect of this stomatal blocking on plant performance and crop yield.

Acknowledgements

The authors would like to acknowledge the Ministry of Agriculture, Fiji for identification of the whitefly species and providing relevant information. The authors are also thankful to Ms Prianshika Sen and Mr Shiva Padiyachi for their support in terms collecting results for this study.

References

Boopathi, T., Karuppuchamy, P., Kalyanasundaram, M. P., Mohankumar, S., Ravi, M., & Singh, S. B. (2015). Microbial control of the exotic spiralling whitefly, Aleurodicus dispersus (Hemiptera: Aleyrodidae) on eggplant using entomopathogenic fungi. African Journal of Microbiology Research 9: 39-46.

Henneberry, T., Naranjo, S., Forer, G., & Horowitz, A. (2007). Chapter 5- Biology, Ecology, and Management of Sweetpotato Whiteflies on Cotton. In E. Hequet (Ed.), Sticky Cotton: Causes, Effects, and Prevention (pp. 51-67). United States

Lambkin, T. A. (1999). A host list for Aleurodicus dispersus Russell (Hemiptera: Aleyrodidae) in Australia. Australian Journal of Entomology 38: 373-376.

Lin, T.-B., Schwartz, A., & Saranga, Y. (1999). Photosynthesis and Productivity of Cotton under Silverleaf Whitefly Stress. Crop Science 39: 174-184.

McAuslane, H. J., Chen, J., Carle, R. B., & Schmalstig, J. (2004). Influence of Bemisia argentifolii (Homoptera: Aleyrodidae) infestation 59 Chand et al.

and squash silverleaf disorder on zucchini seedling growth. Journal of Economic Entomology 97: 1096-1105.

Nabity, P. D., Zavala, J. A., & DeLucia, E. H. (2009). Indirect suppression of photosynthesis on individual leaves by arthropod herbivory. Annals of Botany 103: 655-663.

Neves, A. D., Oliveira, R. F., & Parra, J. R. P. (2006). A new concept for insect damage evaluation based on plant physiological variables. Anais da Academia Brasileira de Ciências 78: 821-835.

Ramos, J., & Grace, J. (1990). The Effects of Shade on the Gas Exchange of Seedlings of Four Tropical Trees from Mexico. Functional Ecology 4: 667-677.

Reddy, P. P. (2015). Cassava, Manihot esculenta. In P. P. Reddy (Ed.), Plant Protection in Tropical Root and Tuber Crops (pp. 17-81). New Delhi: Springer India.

Schymanski, S. J., Or, D., & Zwieniecki, M. (2013). Stomatal control and leaf thermal and hydraulic capacitances under rapid environmental fluctuations. PLoS ONE 8: 1-16.

Touhidul Islam, M., & Shunxiang, R. (2009). Effect of sweetpotato whitefly, Bemisia tabaci (Homoptera: Aleyrodidae) infestation on eggplant (Solanum melongena L.) leaf. J Pest Science 82: 211-215.

University of Florida. (2015). Florida Whitefly. Retrieved 19/09/2015, entomology.ifas.ufl.edu/hodges/white_website/webpages/FAQs.html

Watanabe, T., & Kitagawa, H. (2000). Photosynthesis and Translocation of Assimilates in Rice Plants Following Phloem Feeding by the Planthopper Nilaparvata lugens (Homoptera: Delphacidae). Journal of Economic Entomology, 93: 1192-1198.

Waterhouse, D. F., & Norris, K. R. (1987). Biological Control: Pacific Prospects: Inkata Press Melbourne.

The Weta 52: 55-60 60

Yee, W. L., Toscano, N. C., Chu, C.-C., Henneberry, T. J., & Nichols, R. L. (1996). Bemisia argentifolii (Homoptera:Aleyrodidae) Action Thresholds and Cotton Photosynthesis. Environmental Entomology 25: 1267-1273.

Figure 1. Diurnal stomatal conductance (mean ±SE; n=3) of M. esculenta leaves infested or not infested with Aleurodicus dispersus. The mean of 3 replicate measurements between 1100 hours and 1200 hours is shown.

Figure 2. (A) - the stomatal underneath the normal leaf surface and (B) - the egg attached on the stomata of infested leaf at 400X Magnification. Clear nail polish was used to strip off stomatal aperture from cassava leaves.

61 John Grehan & Clinton Care Recent observations on the entry of trees by larvae of the puriri moth, Aenetus virescens (Lepidoptera: Hepialidae)

John R. Grehan1, Clinton Care2

1 – Research Associate, Section of Invertebrate Zoology, Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh PA 15213, USA. Email: [email protected] 2 - 903 Queen Street, Thames 3500, New Zealand. Email: [email protected]

Development of Aenetus virescens larvae involves two distinct feeding stages – mycophagy (fungal feeding) and phytophagy (plant feeding) (Grehan 1987a). The mycophagous stage occurs on the forest floor where larvae graze the fruiting bodies of polypore fungi, usually those that grow as encrustations on the underside of logs or other dead wood (Grehan 1984). After about 2-3 months the larvae then migrate to host trees where they bore into stems or branches and graze on callus tissue at the tunnel entrance for the next 1-4 years or more (Grehan 1988a, b). These ‘tree phase’ larvae complete development in at least 17 New Zealand native and two exotic angiosperm species (Grehan 1984). Larval growth and survival has been the subject of recent ecological investigations (Yule & Burns 2015, Yule & Burns 2017a, b).

Variations in the size of larvae moving to host trees indicate that the transfer can occur at more than one particular instar. The transferring instar is called a ‘transfer phase’ and is characterized by an increase in the size of the sclerotized regions (pinnacula) that also become almost black in color (Grehan 1981). The dorsal pinnacula also fuse across the midline, giving the appearance of dark stripes. Although the transfer morph was illustrated by Grehan (1981) it has never been recorded in colour or shown within the new tunnel.

Examples of the transfer phase larvae were recently extracted from host plants by Clinton Care. In addition to the dark pinnacula, the cuticle of the The Weta 52: 61-64 62 transfer larvae has an olive tint, whereas the pinnacula and cuticle of the thoracic segments are paler (Fig. 1). The transfer phase morphology appears to be lost within the first moult following establishment in the host tree. The functional significance of the temporary morph has never been evaluated. It is possible that the darker coloration has a greater camouflage effect compared with the pale coloration of larvae in the mycophagous stage that remain secluded under a webbing of silk and fecal pellets. As described by Grehan (1983), the larva excavates a short tunnel and a minimal amount of bark tissue is removed around the tunnel entrance (Fig. 2). The tunnel itself turns sharply at a right angle from horizontal to vertical (Fig. 3).

The transfer morph has been recorded in A. virescens and in A. cohici of New Caledonia (Grehan 1988b), which raises the possibility that transfer morphs are also present in other Aenetus species. The phylogenetically important question that remains to be answered is whether the transfer morph occurs in the other specialist groups of stem boring Hepialidae, such as Endoclita of Asia and Phassus of Mexico and Central-South America.

Figure 1. Dorsal view of Aenetus virescens transfer larva.

63 John Grehan & Clinton Care

Figure 2. Exposed tunnel entrance of newly established larva of Aenetus virescens in Ligustrum lucidum showing minimal removal of bark.

Figure 3. Transfer phase larvae of Aenetus virescens within new tunnel in Ligustrum lucidum.

The Weta 52: 61-64 64

References

Grehan, J. R. 1983. Larval establishment behaviour of the borer Aenetus virescens (Lepidoptera: Hepialidae) in live trees. New Zealand Entomologist 7: 413-417.

Grehan, J. R. 1981. Morphological changes in the three-phase development of Aenetus virescens larvae (Lepidoptera: Hepialidae). New Zealand Journal of Zoology 8: 505-514.

Grehan, J. R. 1984. The host range of Aenetus virescens (Lepidoptera: Hepialidae) and its evolution. New Zealand Entomologist 8: 52-61.

Grehan, J. R. 1988a. Life cycle of the wood-borer Aenetus virescens (Lepidoptera: Hepialidae). New Zealand Journal of Zoology 14: 209- 217.

Grehan, J. R. 1988b. Evolution of arboreal tunneling by the larvae of Aenetus (Lepidoptera: Hepialidae). New Zealand Journal of Zoology 14: 441-462.

Yule, K. J. and Burns, K. C. 2015. Drivers of aggregation in a novel arboreal parasite: the influence of host size and infra-populations. International Journal of Parasitology 45: 195-202.

Yule, K. J. and Burns, K.C. 2017a. Adaptive advantages of appearance: predation, thermoregulation,and color of webbing built by New Zealand’s largest moth. Ecology 98: 1324-1333.

Yule, K. J. and Burns, K.C. 2017b. Host defence predicts host specificity in a long-lived arboreal parasite. Evolutionary Ecology 31: 37-50. 65 Michael Wakelin Drinking by Certonotus fractinervis (Hymenoptera: Ichnuemonidae) at a fungal fruiting body

Michael Wakelin

47 Hunt Street, Anderson’s Bay, Dunedin. [email protected]

An adult male Certonotus fractinervis (Vollenhoven, 1873) was observed drinking from a droplet on the surface of a stroma fruiting body of the fungus Cyttaria nigra Rawlings, 1956 (Figure 1). The site was near Black Gully in the Blue Mountains, Southland (45.902oS, 169.370oE), at 800m altitude in a bushline stand of silver beech (Lophozonia menziesii) where galls caused by the fungus were common and in all stages of fruiting. It was about 13:30 NZST on 11 January 2017, with calm, cool, overcast conditions without rain. The position as in the photograph lasted 2 to 3 minutes during a period of 10 minutes when the wasp was seen flying and walking within an area of approximately 2 m3 in the sub-canopy (about 4 m above ground). The wasp was actively waving its antennae in an apparent search, with attention paid to immature stroma and the occasional droplets seen on them. It did not alight on any other part of the tree.

This ichneumon wasp is a parasite of the larvae of the elephant weevil (Rhynchodes ursus White, 1846), which bore into live trees including beech species throughout New Zealand. Cyttaria (Cyttariaceae) are Ascomycete fungi and the three species in New Zealand (Cyttaria nigra, C. pallida and the misnamed C. gunni; see Peterson and Pfister 2010) are all restricted to silver beech, occurring across its range (Rawlings 1956). Silver beech is found from Auckland to Southland but is absent from Taranaki, central Westland and Stewart Island, and is sparsely distributed in eastern South Island.

Fruiting of each Cyttaria species seen in 2017 on the eastern slopes of Maungatua (45.875oS, 170.150oE) lasted about one month, starting with C. pallida in early October, C. ‘gunni’ in early November and C. nigra in early December. Cyttaria nigra seemed to be mainly found on the higher slopes. Droplets were again observed on C. nigra at Maungatua but were not noticeably sticky or sweet tasting. There are no reports of excretions The Weta 52: 65-67 66 from Cyttaria stroma, although they have structures such as mucus- containing apotheca and papillae venting to the surface (Rawlings 1956) and they contain sugar (Toledo et al. 2016). Cyttaria stromata are eaten by people in Australia and South America (Schmeda-Hirschmann et al. 1999) and in New Zealand by kereru and possums (Rawlings 1956).

Many invertebrates are known to utilise fungi for feeding or breeding and they in turn attract predators and parasites (Hodge et al. 2010). It seems that this male parasitic wasp was seeking food, as consuming sugars can improve ichneumon survival (Khatri 2011). No other sources of nectar or water were seen at the time, and honeydew produced by scale insects (Ultracoelostoma spp.) is rare in silver beech (Beggs & Wardle 2006). The possibility that Cyttaria fungi may be a valuable resource for some invertebrates in silver beech forest warrants further investigation.

References

Beggs JR, Wardle DA. 2006. Keystone Species: competition for honeydew among exotic and indigenous species. In: Biological Invasions in New Zealand (eds RB Allen & WG Lee) pp. 281–294. Springer-Verlag, Berlin Heidelberg.

Hodge S, Marshall SA, Oliver H, Berry J, Marris J, Andrews I. 2010. A preliminary survey of the insects collected using mushroom baits in native and exotic New Zealand woodlands. New Zealand Entomologist 33: 43-54.

Khatri D. 2011. Reproductive biology of Diadegma semiclausum Hellen (Hymenoptera: Ichneumonidae). (Unpublished MSc thesis). Massey University, Palmerston North, New Zealand.

Peterson KR, Pfister DH. 2010. Phylogeny of Cyttaria inferred from nuclear and mitochondrial sequence and morphological data. Mycologia 102: 1398-416.

Rawlings GB. 1956. Australasian Cyttariaceae. Transactions of the Royal Society of New Zealand 84: 19–28.

67 Michael Wakelin

Schmeda-Hirschmann G, Razmilic I, Reyes S, I Gutierrez M, I Loyola J. 1999. Biological Activity and Food Analysis of Cyttaria spp. (Discomycetes). Economic Botany 53: 30-40.

Toledo CV, Barroetaveña C, Fernandes Â, Barros L, Ferreira ICFR. 2016. Chemical and Antioxident Properties of Wild Edible Mushrooms from Native Nothofagus spp. Forest, Argentina. Molecules 21: 1201.

Figure 1. Adult male Certonotus fractinervis drinking from a droplet on a Cyttaria nigra stroma.

The Weta 52: 68-70 68

Book Review: Palma, R.L. 2017. Phthiraptera (Insecta). A catalogue of parasitic lice from New Zealand. Lincoln, New Zealand, Landcare Research. Fauna of New Zealand Ko te Aitanga Pepeke o Aotearoa. Number 76. 400 pp. http:doi.org/10.793/J2FNZ.76.

Terry D. Galloway

Department of Entomology, University of Manitoba, Winnipeg, Manitoba, Canada, R3T 2N2. [email protected]

The day Ricardo Palma began working together with Prof. Robert (Bob) Pilgrim in 1974 was an auspicious one for the taxonomy of lice of New Zealand. Ricardo accepted the position of curator at Te Papa Tongarewa two years later and, at about the time Bob retired in 1983, they divided their attentions, with Bob assuming primary responsibility for fleas and Ricardo taking on lice. They continued to collaborate on several taxonomic revisions (e.g., Palma and Pilgrim 1983, 1984, 1988, 2002) and published a list of the lice infesting birds in New Zealand (Pilgrim and Palma 1982), but it was clear from this latter work that in the land of birds, there was a great deal still to do. The present catalogue, which includes 424 species and subspecies of lice in 101 genera, is the culmination of over 40 years of intensive study of this important group of ectoparasites.

The opening checklist allows the reader to go immediately in the text to any taxon of interest, and the following 10-page introduction is packed with useful information. There is a table with a list of genera of lice and the number of species and subspecies in each. The table also includes descriptive categories so the reader can quickly and easily determine numbers of endemics, native species, introduced species, taxa of uncertain status and taxa newly reported for New Zealand in this catalogue. In Palma’s history of louse research in New Zealand, along with tabulated numbers of taxa over time, the reader is provided with the basis for the current status of knowledge of lice in the country. The style and format for the catalogue are clearly specified to help the reader interpret taxonomic treatments for each taxon. Repositories for types are provided for all taxa where known, and there is a complete list of abbreviations for museums.

69 Terry Galloway

The largest component of the catalogue is taken up by the list of taxa reported from New Zealand. Each taxon follows the same descriptive sequence. The original name is followed by a complete nomenclatural history and accompanying references, where relevant for New Zealand. The type host, hosts in New Zealand and other known hosts are listed. A general geographic distribution is provided, along with a detailed list of occurrences in New Zealand according to standard Fauna of New Zealand area codes (Crosby et al. 1976; a map appears on p. 394). Of particular value is a list of all references to New Zealand records for each taxon, along with a number of additional significant references. This allows the reader easy access to the scattered literature to the order. The author provides concise remarks on nomenclatural status of each species of louse and in many cases on hosts. This is particularly important where there are differences from the most recent world checklist (Price et al. 2003). Lice and parasites in general, are not usually at the forefront of concern for policies on species at risk. Here, Palma identifies conservation status for lice where warranted based on at-risk status of specific hosts.

There is an easily accessed host-parasite list in which current scientific and common names are accompanied by lists of lice known to infest each host. The author uses a series of symbols by which the reader can identify lice endemic to New Zealand, new host records for lice, hosts which breed only in the New Zealand subregion and hosts which have been introduced into New Zealand. Of particular interest, is a list of the 18 species of birds that breed in the New Zealand subregion, from which no lice have been collected regionally. Some of these birds are rare endemics or are native species that are found in other biogeographic regions; some are introduced, with lice known to infest them elsewhere in the world. This list provides a clear focus for collection of lice in the future, should the opportunity arise. Following an extensive 32-page list of references, the author has provided excellent habitus photos for males and females (where both are available; some specimens have been chemically stained) of all genera of lice recorded in New Zealand. A scale bar is provided for each photo, which is sufficiently detailed to assist in generic identification.

I don’t believe there has ever been a time when there has been so much worldwide attention on taxonomy and ecology of parasitic lice. As the author points out in his introduction, there is much research to be done using molecular tools as an aid to sort out complex taxonomic issues for The Weta 52: 68-70 70 many groups if lice. This catalogue of lice from New Zealand is the benchmark for future work in this biogeographic subregion. Louse workers from all parts of the world will benefit from having a copy.

References

Crosby TK, Dugdale JS, Watt JC. 1976: Recording specimen localities in the New Zealand subregion. NZ J Zool, 25: 175-183.

Palma RL, Pilgrim RLC. 1983: The genus Bedfordiella (Mallophaga: Philopteridae) and a note on the lice from the Kerguelen petrel (Pterodroma brevirostris). National Museum NZ Records, 2: 145-150.

Palma RL, Pilgrim RLC. 1984: A revision of the genus Harrisoniella (Mallophaga: Philopteridae). N Z J Zool, 11: 145-166.

Palma RL, Pilgrim RLC. 1988 (1987): A revision of the genus Perineus (Phthiraptera: Philopteridae). NZ J Zool, 14: 563-586.

Palma RL, Pilgrim RLC. 2002: A revision of the genus Naubates (Insecta: Phthiraptera: Philopteridae). J Royal Soc NZ, 32: 7-60.

Pilgrim RLC, Palma RL. 1982: A list of the chewing lice (Insecta: Mallophaga) from birds in New Zealand. Notornis, 29 (Supp.): 1-32 (also as National Museum of New Zealand Miscellaneous Series 6).

Price RD, Hellenthal RA, Palma RL. 2003: World checklist of chewing lice with host associations and keys to families and genera. pp. 1-448. In: Price RD, Hellenthal RA, Palma RL, Johnson KP and Clayton DH. The chewing lice: world checklist and biological overview. Illinois Natural History Survey Special Publication 24. x + 501 pp.

71 Erratum Erratum

Iwasaki JM, Miller C, Donovan BJ (2017) New records for Bombus terrestris (Hymenoptera: Apoidea) on flowers of native plants in New Zealand. The Weta 51: 47-50.

An error occurred in Table 1 of the original version of this paper, in that the location of the Bombus terrestris record of 19/12/07 on Peraxilla sp was given as The Remarkables, Otago. This record was actually from Lake Ohau. The correct version of Table 1 is given below.

The Weta 52: 71-72 72

19.12.2007 Date initialof observation 6.1.2016 summer 2014 14.12.2014 5.1.2016 23.11.2014 summer 2014

, ,

Remarkables, Remarkables, Remarkables, Remarkables, Remarkables, Remarkables, Remarkables,

Site CO LakeOhau Village The CO The CO The The CO The CO The CO MK

nectar nectar nectar nectar nectar

Forage

sightings.

queen Bee Caste worker queen queen, worker queen, worker

Bombus terrestris terrestris Bombus

rosmarinifolium

sp.

Mackenzie

sp.

-

Plant species Plant Peraxilla Celmisia gracilenta Lobelia Dracophyllum Gaultheria nubicola Leucopogon fraseri Gentianella corymbifera

lant species and details of of details and species lant

List of p of List

Central Otago; MK Otago; Central

-

Plant Family Plant Loranthaceae Asteraceae Campanulaceae Ericaceae Ericaceae Ericaceae Gentianaceae

Table 1. Table CO