Biology and conservation of the rare New Zealand land snail Paryphanta busbyi watti (Mollusca, Pulmonata)

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/45403646 Biology and conservation of the rare New Zealand land snail Paryphanta busbyi watti (Mollusca, Pulmonata) ARTICLE in INVERTEBRATE BIOLOGY SEPTEMBER 2005 Impact Factor: 1.2 DOI: 10.1111/j.1744-7410.2003.tb00088.x Source: OAI CITATIONS 17 READS 63 5 AUTHORS, INCLUDING: Ian Stringer New Zealand Department of Conservation 60 PUBLICATIONS 452 CITATIONS Jay McCartney Massey University 23 PUBLICATIONS 169 CITATIONS SEE PROFILE SEE PROFILE Available from: Jay McCartney Retrieved on: 09 April 2016

Invertebrate Biology 122(3): 24 1-25 1. 0 2003 American Microscopical Society, Inc. Biology and conservation of the rare New Zealand land snail Paryphanta busbyi watti (Mollusca, Pulmonata) Ian A.N. Stringer,',a Suzanne M. Bassett,2 Megan J. M~Lean,~ Jay McCartney,2 and G. Richard Parrish' I Department of Conservation, PO. Box 10420, Wellington, New Zealand Institute of Natural Resources, Massey University, Private Bag 11222, Palmerston North, New Zealand ' The Ecology Centre, School of Life Sciences, University of Queensland, Qld. 4072, Australia Abstract. The biology of Paryphanta busbyi watti, an endangered carnivorous land snail, was studied mostly by following large juvenile and adult snails with harmonic radar. The snails are nocturnally active and most (79%) hide during the day under leaf litter or in dense vegetation. Fecal analysis showed that the diet is primarily earthworms, but some cannibalism of smaller snails occurs. Empty shells appear to be an additional source of dietary calcium. Mating occurred most frequently between April and July. Mating snails stayed together for 4-7 days, and each pair reversed their positions at least twice. Four snails were first found mating 151-1240 d after they acquired adult shells, and 7 snails were observed mating a second time after 66-298 d. We found 8 nests and observed 6 snails ovipositing; 5 snails laid eggs in holes they dug and one laid eggs in a crevice between rocks. In 2 instances, oviposition was recorded 52 and 140 d after mating. Snails were estimated to lay on average -17.5 eggs per year in 3-5 clutches. Most oviposition was observed in August/September, but some occurred between November and February. Of the snails that died, pigs killed 13.6% and humans inadvertently killed another 13.6%. Other snails died from unknown causes mostly during the drier and warmer months, from November to April. This large land snail survives in the presence of introduced predators, but some life history traits could predispose it to a rapid decline in numbers if new predators arrive. Additional ke.y words: behavior, harmonic radar Paryphanta busbyi watti Powell (Pulmonata, Rhytididae) is a rare snail that occurs within a total area of about 15 km2 at the end of the Aupouri Peninsula, New Zealand (Fig. 1) (Stringer & Montefiore 2000). It is one of two subspecies; the other, P. b. busbyi Gray, is relatively common in forest and scrub in the upper part of Northland between Kaitaia and Warkworth (Fig. 1) (Parrish et al. 1995). Snails of both subspecies are large-maximum diameter of P. b. watti is up to 62 mm and P. b. busbyi to 79 mm-and both are carnivorous, feeding on earthworms and possibly other ground-dwelling invertebrates (Powell 1979). Both subspecies are fully protected (Wildlife Act 1953). Effective conservation of P. b. watti requires knowledge of its basic biology to help conservation managers make informed decisions. In this paper we provide additional details of this snail's life history that a Author for correspondence. E-mail: istringer@doc.govt.nz have not already been published. Our results were obtained by repeatedly recapturing marked snails with transponders attached to their shells, using harmonic radar. Before we started working on this snail in 1994, information was restricted to taxonomy (literature included in Powell 1979), distribution (Parrish 1992; Goulstone et al. 1993; Sherley 1993), and a description of 1 egg (Powell 1946). Stringer & Montefiore (2000) then provided detailed data on distribution, habitat, and morphological changes in the shell during growth, together with some preliminary data on growth, eggs, site fidelity, mortality, and general biology. Most of their information was obtained from unmarked snails. Stringer et al. (2002) reported that the snails lay 2-8 eggs at a rate of 1-2 eggs per day over 2-5 d, that incubation takes 152-221 d, and that, after hatching, the snails remain underground for a minimum of 32-85 d. They showed from recaptures of marked snails that this species has determinate growth; shell growth ceases after 3-4.3 yr, but the snails continue to live for up to 4.1 yr more (Stringer et al. 2002). Little in-

242 formation is available on the growth, eggs, and biology of P. b. busbyi (Hutton 1881; Powell 1930; O'Connor 1945; Ohms 1948; Dell 1955; Ballance 1985; Meads 1990; Parrish et al. 1995; Montefiore 1996; Coad 1998). Paryphantu b. watti is rare largely because humans reduced its original forest habitat on the Aupouri Peninsula to a few remnants over the last 1000 yr, and some forest remnants were further degraded by livestock when the area began to be farmed in the 1870s (Gardner & Bartlett 1980; Millener 1981 ; Goulstone et al. 1993). Introduced predatory mammals, particularly feral pigs (Sus scrofa), also substantially reduced snail numbers (Parrish et al. 1995). Much of the area where P. b. watti occurs became protected for conservation in 1966 and the natural vegetation is recovering slowly. Pigs are hunted in some areas where P. b. wutti occurs but, despite this, numbers of this snail remain low. r -'- tr ~ I Aupouri 13 165" 170' 17yE -2 _L ~ 35" 40"s 1qo. 45" Stringer et al. NoHh Cape I Study area Methods Three sites (A,B,C in Fig. 1) where the snails occurred in relatively high densities (>lo0 ha ') were chosen after a preliminary survey (Stringer & Montefiore 2000). We limited our activities to less than 2000 m2 within each site because the area is legally protected and under the control of the New Zealand Department of Conservation. Care was taken to restrict potential damage to both the habitat and the snails. Temporary quadrats varying from 150 to 400 m2 were established in each of the 3 study sites (Stringer & Montefiore 2000). These were searched in October/ November each year by sorting through all the litter. Many of the snails were first found during these searches, but others were initially found up to 20 m away while we were searching for snails with harmonic radar transponders attached to them, as described below. We made 26 field trips to the study sites between August 1994 and November 2000. These were spread throughout the year to ensure that samples were taken in all seasons. The snails were checked on 1-7 different days during each field trip. Marking, measurements, and relocation Each time a snail was found, the maximum diameter of its shell was measured and the shell was recorded as adult or juvenile. In adult shells, the periostracum at the edge of the aperture is rolled tightly inward to form a hard lip. In contrast, the periostracum of juvenile snails projects from the edge of the aperture, giving these shells a soft lip (Stringer et al. 2002). The Fig. 1. Maps of New Zealand showing study sites (A,B,C). Paryphanta busbyi watti is restricted to the end of the Aupouri Peninsula, whereas P. b. busbyi occurs between Kaitaia and wx-kworth. terms juvenile and adult are used in an operational sense because the snail probably becomes sexually mature well before the shell forms a lip and ceases to grow (Stringer & Montefiore 2000). Snails found in November 1999 and November 2000, at the end of the study, were released unmarked. Before November 1999, each snail with a maximum shell diameter >20 mm had a harmonic radar transponder fitted to its shell. The transponder consisted of a thin C-shaped disc of annealed copper with a 23040 diode attached across the ends. It was glued onto the upper surface of the shell (using "Liquid Nails," Selleys Ltd.) after the shell was wiped clean, then lightly abraided with fine sand-paper. Full details are given in Lovei et al. (1997) and Stringer & Montefiore (2000). An identification number was engraved through the periostracum of any shell >20 mm in diameter using a portable engraver ("Arlec," Dick Smith Electronics). The identification number was positioned near the aperture and on the upper surface where it would not be covered by subsequent shell growth. All snails were released exactly where they were found. Snails with harmonic radar transponders were relocated using a portable harmonic radar transceiver

Land-snail biology and conservation 243-40 35 u).iii 30 C u) 25 W.- - > 20 c C 15 W $ 10 a 5 0 Depth of leaf litter(cm) Fig. 2. Depth of leaf litter covering individuals of Paryphanta busbyi watti that were found using harmonic radar (256 observations from a total of 93 snails). The black bar marks a break in the depth scale. (Recco rescue system, Recco AB, Sweden; see Lovei et al. 1997 and Stringer & Montefiore 2000 for use of the harmonic radar). Recaptures of snails with transponders were treated as representative samples of all snails with shells >20 mm in diameter for the purposes of estimating proportions of the population engaged in a particular behavior. The numbers of snails with transponders found at each study site during each field trip were too low to analyse separately, so data from all 3 sites were combined. Fecal collection Fresh fecal strings found under snails were collected in small sealable plastic bags containing a few drops of formalin. Small portions of the feces were later macerated in water for a few minutes and examined under a dissecting microscope equipped with an ocular micrometer. Each fecal string was examined until >SO earthworm chaetae were found and measured, but entire feces were examined when <SO were found. Night observations Observations of snails at night were made at Site C using an infrared sensitive video camera (Panasonic WV-BP312 with auto focus lens WV-LA908C3). A custom-made transmitter sent the image to a building 4.7 km away, where it was watched directly and recorded on a time-lapse video recorder (Panasonic AG6040). The video camera was set up on a bank overlooking an arena of ground 0.80 m X 1.35 m, bounded by 4 boards (rough-sawn untreated pine; 150 inm wide, 25 mm thick) set on edge and pegged in place. Observations were made all night on a pair of snails 11-15 June 1996 and on another pair 26-30 August 1996. One snail moved out of the arena 5 h after nightfall on 14 June 1996. Results Overall, 126 live specimens of Puiyphuntu busbyi wutti were found at the 3 study sites between August 1994 and November 2000, and harmonic radar transponders were attached to the shells of all the adults (62) and 31 of the juveniles. These snails were found again 1-36 times (median 3 recaptures; mean and S.E. 5.4 2 0.6 recaptures). Stringer & Montefiore (2000) and Stringer et al. (2002) give information on snails that hatched from eggs, but none were found again once they left their nests. Resting places When active at night, individuals of P. b. watti generally travel over the top of leaf litter, but during the day marked individuals were found hidden under leaf litter on 256 occasions or on the ground hidden under dense vegetation on 134 occasions. Of those found in leaf litter, 63.2% were covered by up to 2.5 cm of leaves and 5.2% were found under piles of leaves and sticks that were 10 cm or more deep (Fig. 2). Three of the snails under leaf litter had the front of their shells partly inserted into the soil. A further 20 snails were found under rocks, 9 were in holes in soil, and 6 were under logs. Snails that were found clearly visible on the surface (67 occasions) were most often seen during April to July and October to December (Fig. 3B). Diet No snail fed when pairs of snails were observed all night for 4 nights on 2 separate occasions. The only records of feeding were from 2 adult snails that were found during the day with their heads inserted into the shells of smaller conspecific individuals. The smaller snails appeared to have just died but were partly consumed. Fecal strings were found underneath 15 snails during the daytime; the proportions of these found at various times of the year are shown in Fig. 3A. Of these feces, 12 that appeared to have been freshly deposited were collected. Each was 2-3 mm wide and up to 60 mm long. They consisted largely of brownish material enclosed by a fragile clear membrane, with a region of white material up to about 6 mm long at one end. The latter consisted of clear spheres mostly -5 pm in diameter (range 4.5-10 pm) densely packed within a clear tough membrane. A partial analysis (by Grasslands Research Centre, Palmerston North, New Zealand) of the clear spheres from the only fecal string

244 Stringer et al. 15 A Withfeces m Rasping shell Table 1. Lengths of radular teeth from Paryphanta busbyi watti found in feces of P. 6. watti together with the maximum diameters of the shells of the snails that produced the feces. Radular teeth v,.- ::I 5- ((1 U 20 5 Diameter of shell mean? S.E. Range 52.7 (adult) 1 395-52.8 (adult) 1 395-53.8 (adult) 5 392 t- 17 329-42 1 54.6 (juvenile) 5 346 t 19 310-420 54.8 (adult) 4 393 t- 21 330-420 55.6 (juvenile) 2 390 -L 0-55.8 (adult) 32 406 t- 6 316-447 55.8 (adult) 23 432 2 4 380-470 56.6 (adult) 1 382-0 Time of year Fig. 3. Relationship between the time of year and the proportion of individuals of Paryphanta bushyi watti found using harmonic radar. Percentages are given in relation to the number of snails with harmonic radar transponders that were alive at each time. A. Snails with fecal strings (n=15) or with the anterior portion of their bodies inserted into empty shells (n=7). B. Snails visible on the surface during the day (n = 67). that was uncontaminated by soil showed that the white material contained calcium (1 1.75% WW) and phosphorus (17.19%), together with trace elements (Mn 4.72%, K 1.33%, Na 0.37%, Zn 0.27%, S 0.21%; Sr, Al, Fe, and Co <O.l%). We were unable to analyse phosphate, but if all the phosphorus was in this form then it would account for 52% of the white material. One fecal string had a portion of undigested earthworm - 10 mm long and -5 mm in diameter between the brown and white regions. The brownish material in all feces consisted mostly of fine, irregularly shaped or amorphous particles together with sand grains, earthworm chaetae, and a few radular teeth of conspecific snails scattered through it (Table 1). We distinguished 3 categories of chaetae of slightly different shapes (Fig. 4), but we were unable to identify the earthworm species from which they came. The only other identifiable components were the remains of treefern scales and sporangia, an occasional pollen grain, rare pieces of leaf cuticle or plant hairs, the distal end of an amphipod limb, and a single unidentified insect mandible. In daytime observations during August to September, and November to December (Fig. 3A), 8 snails were found with their heads extended into old empty shells. Of the empty shells, 5 were P. b. watti, and the others were Ainborhytida duplicatu (SUTER) (Rhytididae) and Placostylus ambugiosus SUTER (Bulimulidae). All had some calcareous material missing from inside the last whorl. One adult snail was also found on top of a small conspecific juvenile. The adult snail had a portion of the ventral part of its foot extended into the opening of the small juvenile's shell. The juvenile was situated about one quarter of the way back from the anterior end of the adult snail's foot and when the snails were separated, the adult's foot appeared to have occupied about half of the last whorl of the juvenile shell. Reproductive behavior Mating. The 2 pairs of snails observed at night (each pair during 4 nights) were not seen to copulate, but 22 pairs were found during the day with one snail on top of the other and another 5 pairs were found during the day lying within 3 cm of each other. Most of these observations were during the colder and wetter months (Fig. 5A). Most pairs (13) were first found with the foot of the upper snail adhering to the upper surface of the lower snail's shell and with the lower snail withdrawn into its shell (see plate If, Stringer & Montefiore 2000). In 7 pairs, the head of the upper snail was extended and bent back into the aperture of the lower snail's shell (Fig. 6), whereas in 6 pairs, the upper snail's foot was adhering to the lower shell, but its head was withdrawn into its own shell. A snail with its head inserted into the lower shell usually withdrew within a few seconds of being disturbed, although it

Land-snail biology and conservation 245 Fig. 4. Shapes of earthworm chaetae (of unidentified earthworm species) found in feces of Paryphanta busbyi watti. Laying eggs often continued to adhere to the lower shell until picked up. In the 9 remaining pairs, both snails were fully retracted, with one shell on top of the other. Both snails of all pairs were measured the first time they were found, then replaced as they were found. Two pairs of snails were examined over 7 consecutive days (Table 2). They remained together for a minimum of 4-7 days and reversed their positions at least twice. Each snail remained on top of its mate for 1 or 2 days with the front of its body inserted into the aperture of its mate's shell (Table 2). Four snails were found mating after their development was followed from juvenile to adult, based on shell form, and these provided the only information on when this species first mates (Table 3). A further 18 snails were observed mating 8-913 d (median 433 d; mean and S.E. 406.2 t 63.8 d) after they were first found with adult shells and 7 of these were found mating a second time after 66-298 d (median 127 d; mean and S.E. 136.1 t 31.8 d). Oviposition. We have information from only 2 snails on the period between mating and egg laying. One laid eggs 140 d after it was observed mating, and the other laid eggs 52 d after it had been found within 3 cm of another snail. During November to February and during August to September, in the course of a total of 312 observations of adult snails with transponders, we found 6 snails that were ovipositing (Fig. 5B). These snails were monitored for 1-5 days (a total of 14 observations), and they laid a total of 15 eggs on 11 of these days. The nest holes were 9-20 mm in diameter and up to -20 mm deep. The sixth snail laid an egg in a crevice -10 mm wide between 2 rocks but this nest was not checked subsequently. Four of the holes were dug in soft soil and one was in deep humus beneath bracken. One hole went into a bank horizontally, 2 were angled -45" into sloping ground, and 2 were vertical. In most cases the snail completely covered the eggs with soil 5-10 mm deep, leaving a slight de- 2o i 10 i 5 0 Time of year Dead Fig. 5. Time of year when reproductive behavior and mortality of Paryphanta busbyi watti were observed. Percentages are given in relation to the number of snails with harmonic radar transponders that were alive at each time. A. Proportion of snails adhering to the shells of other snails (n=45), or found lying within 3 cm of other snails (close together) (n=5). B. Proportion of snails found laying eggs (n=6). C. Proportion found dead (n= 13). pression, but in one case the topmost eggs were covered by only -1 mm of soil. Two other clutches of eggs were found after the snails had covered them and left; both appeared to have been laid in vertical holes. Stringer et al. (2002) give the clutch sizes, measurements of the eggs, and incubation periods for all eggs that were found. Mortality Of the 22 snails with transponders that died, pigs killed 3, another 2 appeared to have been crushed by

246 Stringer et al. vehicles, and one was crushed by humans. Others that died were found intact but empty. Most of the latter were found between November and February when it is warm, whereas few died from April to July (late autumn and early winter), and none died in spring (Fig. 5C). Discussion So far we have only sketched out the biology of Paryphanta busbyi watti from fragmentary observations. Many of our data are preliminary because we found and followed few of these rare snails and because of the long periods between observations. Diurnal resting behavior Hiding and burying under leaf litter protects this snail from visual predators such as birds, but it also undoubtedly offers some protection from desiccation. This is especially important because P. b. watti does not form a protective epiphragm. However, the snails frequent habit of resting with the shell aperture in contact with the ground, or occasionally burying the front of the shell in soil, must reduce water loss. Diet All snail feces examined contained earthworm chaetae. Very little else could be identified, and it is possible that the bulk of fecal material was the gut contents of the earthworm prey. This suggests that earthworms are a main food of these snails, but it is difficult to classify worm species from their chaetae alone (Brockie 1959), and other structures used for identifying earthworms in feces, such as gizzards (Bradbury 1977), were not found. The most likely prey are megascolicids because these are the only large earthworms present in indigenous forest and scrub in New Zealand (Lee 1959), and large piles of their fecal casts were common where P. b. watti was found. Megascolicid earthworms comprise the largest component of the faunal biomass in New Zealand forests (Brockie 1992), so they would appear to be an ideal food source for such a large carnivorous snail. Both megascolicid and lumbricid earthworms seem to be the main food of P. b. busbyi (Suter 1899; Dell 1949; Vause 1977), and this subspecies occasionally regurgitates complete earthworms when picked up (Coad 1998). All radular teeth found in feces were from adult snails, and all were of similar size (Table 1). This suggests that they probably originated from the snails themselves rather than from feeding on smaller conspecifics. They were present in too many feces to have - Fig. 6. Typical mating position of Paiyphunta busbyi watti. The upper snail has the front of its head inserted into the aperture of the lower snail s shell. originated from the snails having fed on other large individuals that had died. The 2 instances in which we saw large snails feeding on smaller ones certainly appeared to be cannibalism rather than feeding on carrion because the remaining flesh of the small snails was fresh. Paryphanta b. busbyi is cannibalistic (Parrish et al. 1995), but has not been reported to feed on carrion, and did not feed on meat or crushed specimens of Cantarus aspersa (Muller) (Helicidae) in captivity (Vause 1977). Other rhytidids, however, probably do eat carrion (Herbert 1991; Efford 1998). Finally, it is possible that P. b. watti preys on cooccurring snails of other species-amhorhytida duplicata and the young of the endangered snail Placostylus ambagiosus. Paryphanta b. busbyi is known to eat individuals of Amborhytida elsewhere (Parrish et al. 1995). Snails and earthworms are commonly eaten by carnivorous snails throughout the world (Efford 1998), and both are included in the diet of many other rhytidids in New Zealand and elsewhere (Dell 1949; Smith 1971 ; Powell 1979; Herbert 1991 ; Devine 1997). Other prey reported for New Zealand rhytidids includes the eggs of other rhytidid species, snails, slugs, millipedes, and amphipods (Powell 1979; Efford & Bokeloh 1991; Efford 1998, 2000). Removal of calcium from empty shells The removal of calcareous material from empty shells or the consumption of eggshell by land snails, including those that are not predatory, is well known (Wolda 1970; Nisbet 1974; Owiny 1974; Baur 1988), and calcium shortage has even been suggested as the cause for some cannibalism (e.g., Wolda & Kreulen 1973). For species of Rhytida in New Zealand, Efford (1998) reported that these snails were sometimes found inside the empty shells of individuals of Powelliphan-

Land-snail biology and conservation 247 Table 2. Relative positions of mating pairs of Paryphanta busbyi watti over 7 days in June 1996. Upperllower Date Description of upper snail Description of lower snail AIB A/B AIB B/A A/B CID DIC C/D CID C&D 09 10 11 12 15 09 10 11 12 15 Adhering to B, head withdrawn Adhering to B, head withdrawn Head inserted into B Head inserted into A Head inserted into B Adhering to D, head withdrawn Head inserted into C Head inserted into D Head inserted into D Moved mart Retracted Retracted Retracted Foot out, head retracted Foot out, head retracted Retracted Foot out, head retracted Foot out, head retracted Retracted ta, and Dell (1949) noted that individuals of Rhytida removed almost all the calcareous material from the shells of snails they feed on. Herbert (1991) reported that the rhytidid Natalina caffra FCrussace introduces the posterior end of its foot into empty shells and appears to remove calcareous material through the sole of its foot, and this may have been what the adult of P. b. watti was doing when we observed it with part of its foot inserted into the shell of a smaller conspecific snail. Other rhytidids, including P. b. busbyi, are reported to transport prey and other items on the back of their foot (Coad 1998; Efford 1998; Appleton & Heeg 1999), but P. b. watti was not observed to do this. Reproductive behavior When we found one snail on top of another, we assumed that this position was associated with mating. Certainly Achatina rnarginata SWAINSON (Achatinidae) copulates in this position, but large everted penes are visible coiled together (Plummer 1975). We saw no evidence of an intromittent organ, even when we observed a snail in the act of withdrawing its head from the shell of the snail beneath it, so this position is possibly precopulatory as reported for Achatina achatina LINNAEUS by Mead (1950). Vause (1977) found that captive individuals of P. b. husbyi changed positions over several days when mating, as does P. b. watti, but reported that captive individuals of P. b. busbyi formed clumps of 3 or 4 individuals that stayed together for some hours in early summer. However, we only found pairs of P. b. watti on the Aupouri Peninsula and most often in autumn and early winter. This is the beginning of the wettest time of year (New Zealand Meteorological Service 1983; Stringer & Montefiore 2000), and mating in other pulmonates is often associated with moist conditions (e.g., Runham & Laryea 1968; Heatwole & Heatwole 1978; Solem & Christensen 1984). The inference of a possible delay between mating and egg laying by P. b. watti must be treated with caution because it is based on only 2 data (52 and 140 d) and because of the long intervals when the snails went unobserved between our visits to the Aupouri Peninsula. However, a delay of such duration would not be unusually long, because in achatinids, for example, delays can vary from a few days to years (see references in Plummer 1975). The 2 individuals of P. b. watti that were observed mating and later ovipositing were paired with different snails, neither of which was subsequently observed to lay eggs. Although it is quite likely that we simply missed egg laying by the 2 snails partners, McLauchlan ( 1 95 1) reported that only one member of each mating pair of the rhytidid Stt-angesta capillacea FBrrusac lays eggs and the other never does. Oviposition by P. 6. watti is probably spread over many months, although it is most frequent during the wettest months, presumably because the soil is moist and nest holes are easily dug by the snails. The limited information on other New Zealand rhytidids indicates that oviposition is also spread over several months, usually October to early December, and eggs are only occasionally laid at other times (O Connor 1945; Meads et al. 1984). Overall, snails with transponders were found mating >4 times more frequently than they were found ovipositing. This is likely to reflect a true difference in the frequencies of the 2 behaviors (Fisher exact test p = 3.5.10-5) because both are probably equally likely to be observed as both probably take place over a similar number of days. Although all snails observed mating and ovipositing had adult shells, it is still possible that this snail could be reproductively active before shell growth ceases, as reported for at least 3 other snails with determinate growth (Chatfield 1968; Plummer 1975; Barker 1999). Oviposition. Burying eggs is not unusual in pul-

248 Stringer et al. monates (e.g., Solem & Christensen 1984; Baur 1988), and this is certainly done by the rhytidids P. b. busbyi, Powelliphanta traversi traversi (POWELL), and Natalinu species (Herbert 1991; Coad 1998) in a manner similar to that of P. b. watti. Some New Zealand rhytidids, however, appear to lay their eggs in leaf litter or leaf mould without digging (O Connor 1945; Efford 1998). The holes made by P. b. watti are much narrower than the shell, so the snail must dig with the foot, rather than use the shell like a ploughshare, as does Achatina marginata (Plummer 1975). Burying eggs undoubtedly protects them from desiccation during their long incubation (Stringer & Montefiore 2000; Stringer et al. 2002). In addition, it may protect them from egg predators, including the rhytidids Delos cf. coresia GRAY, Deiouagapia cordelia HUTTON, and Amborhytida duplicata, which also occur on the Aupouri Peninsula (Goulstone et al. 1993). We do not know how many times an individual of P. 6. watti oviposits each year, because our sampling was intermittent. However, we can make a crude estimate that each snail lays -17.5 eggs per year if we assume that our observations (15 eggdl1 days) are representative of the entire year: 11 observations of snails ovipositing (including same snails observed ovipositing on different days) / 312 observations of adult snails X 365 days X 1.36 eggs laid per day. The mean clutch size for P. b. watti is 24.5 eggs (Stringer et al. 2002), so each snail probably lays 3-5 clutches per year. This seems reasonable because each clutch represents about 23% of the live body mass of a snail (Stringer et al. 2002) and as such must represent a considerable investment in energy and a drain on calcium reserves. Finally, Stringer et al. (2002) reported an overall observed hatching rate of 83%, so a crude estimate of reproductive rate is -14.5 young per year. This, although low for pulmonates, is by no means the lowest; for some tiny land snails, the lifetime average is only 2-6 eggs (Baur 1989). Mortality We could not determine the cause of most mortality because only empty intact shells were left. It seems likely that death was related to desiccation, especially in small juveniles, because most empty shells were found in the warmest and driest part of the year (Fig. 5C). Predation by invertebrates, as discussed below, and cannibalism, as discussed above, are also possible causes. However, these snails are not always cannibalistic as reported by Vause (1977), because some clutches of up to 6 young snails stay underground for some months after hatching from eggs (Stringer et al. 2002). Table 3. Time (days) when 4 individuals of Paryphanta hushyi watti were first found mating after developing adult shells. The minimum time is after a snail first found with an adult shell and the maximum time is after the snail was last observed as a juvenile. Time (d) Date snail found mating Minimum Maximum 16 Apr 98 1165 1240 09 Jun 97 931 1020 03 Jul 95 224 313 01 Dec 95 151 375 Biology and conservation Fossils of Paryphanta busbyi watti have been reported only from sand deposits, but the species clearly once had a wider distribution over the end of the Aupouri Peninsula (Powell 1946; Brook 1999). The restricted distribution today is due mostly to past habitat destruction, although the forest is now slowly regenerating in some areas protected by the New Zealand Department of Conservation. Feral pigs restrict the distribution of P. 6. watti because they can find and eat most large snails in areas they visit. However, at least one population of snails seems to have expanded where pigs are hunted, as mentioned above (Stringer & Montefiore 2000). Rodents, probably Rattus rattus, accounted for only 2.1% of snail deaths, but rodentdamaged shells were more frequent at low elevations where relatively few snails occur, so these predators have a proportionally larger effect there (Stringer & Montefiore 2000). Other potential mammalian predators present are mice (Mus musculus), Norway rats (Rattus nowegicus), brushtailed possums (Trichosurus vulpecula), European hedgehogs (Erinaceus europpaeus), stoats (Mustela erminea), and weasels (Mustela nivalis). Most are serious predators of large land snails elsewhere in New Zealand (Powell 1946; Meads et al. 1984; King 1990; Coad 1998; Sherley et al. 1998). However, in the shells we found, we could link the damage only to pigs and rodents. It is likely that P. b. watti had to contend with a variety of other predators before introduced mammals arrived, but many of these no longer occur on the Aupouri Peninsula. Those that are still present include invertebrates such as large centipedes, mygalomorph spiders (Laing 1982), carabid beetles, and wcta (Orthoptera: Anostostomatidae), together with a variety of birds. Species that are known to eat snails, but that no longer occur there, include some larger lizards (Gekkonidae and Skincidae), tuataras (Sphenodon punctatus) (Southey 1985; Ussher 1995), at least 7 extant and 8 extinct birds (Millener 1981; Heather & Robertson

Land-snail biology and conservation 249 1996; Brook 2000). Brook (2000) lists some of these birds as the probable cause of damage to fossil shells of Placostylus ambagiosus, the other large land snail that occurs with P. b. watti, although no predatordamaged fossils of P. b. watti are reported. At least one introduced omnivorous snail, Oxychilus cellarius MULLER (Zonitidae), which is associated with the decline of indigenous snails elsewhere in New Zealand, is now present at the end of the Aupouri Peninsula (Barker 1999), but we have not yet found it where P. b. watti occurs. As predatory species of Oxychilus already coexist with P. b. busbyi in the Waipu area of Northland, they are also likely to coexist with P. b. watti. All predatory species of Oxychilus in New Zealand are small (shell diameter 10-15 mm), so they could probably eat only the smallest specimens of P. b. watti (G.M. Barker, pers. comm.) and could themselves be eaten by large individuals. P. b. busbyi eats other smaller predatory snails (Parrish et al. 1995). Serious future threats to P. 6. watti are species of predatory ants in the genera Linepithema, Iridomyrmex (Dolichoderinae), and Pheidole (Myrmecinae), which are well established in Auckland and elsewhere, and are spreading (Green 1992; Berry et al. 1997; Harris 2001). It seems only a matter of time before they arrive on the Aupouri Peninsula, but we may have the opportunity to predict how they affect P. b. watti when they reach populations of P. b. busbyi, which is also classified as threatened (Molloy et al. 1994). Our data indicate that P. b. watti probably has a low reproductive rate, and this, together with its large size and slow development (Stringer & Montefiore 2000; Stringer et al. 2002), are among the traits that Daugherty et al. (1993) considered to be associated with the long period that the New Zealand fauna evolved in the absence of mammalian predators. However, P. b. watti still survives on the Aupouri Peninsula in the presence of introduced mammalian predators, although its distribution is limited and it is rare over most of the areas where it occurs. The low reproductive rate of this snail together with its long developmental period (Stringer et al. 2002) indicates a low rate of population growth. In addition, the young of P. b. watti appear to experience high mortality once they emerge from the ground (Stringer & Montefiore 2000), so the rate of recruitment into the adult population is probably low. This may be compensated for, to some extent, by the relatively long adult life (Stringer et al. 2002). This snail is therefore likely to require extra protection if it suffers a serious setback, such as loss of habitat by fire, or if an additional threat emerges, such as the arrival of a new predator. Acknowledgments. We thank Trevor and Gail Bullock, Francis Fitzpatrick, Simon Job, Whiti Abraham, D.J. Niho, and Andrew Abraham (Department of Conservation, Te Paki Field Centre) for support and assistance in the field during this study. We are most appreciative of the help and assistance with the field work given by Richard Montefiore, James Bower, Nick Singers, Liz Grant, Stephanie Prince, Chris Devine, Jens Jorgenson, Paul Barrett, Phil Battley, Rachel Standish, Edgardo Moreno, Christine Bayler, Emma Barroclough, Sam Bradshaw (all Massey University, Ecology Group); Greg Sherley and Ian Flux (Department of Conservation, Science and Research); Natasha Coad (University of Auckland, Biological Sciences); Jenny Lee, Astrid Mariacher, Kim Carter, Len Doel, Alina Arkins (volunteers), Michel Salas (C.I.R.A.D., New Caledonia), and Rori Renwick (Department of Conservation, Kaitaia). GAbor Lovei (HortResearch, Palmerston North) generously provided the harmonic radar, and plasma emission spectrometry analysis was done by Grasslands Research Centre, Palmerston North. This research was funded by Department of Conservation investigation No. 1939 and Massey University. References Appleton CC & Heeg J 1999. Removal of calcium by Natalina cufra (Pulmonata: Rhytidae) from the shells of its prey. J. Molluscan Stud. 65(2): 271-273. Ballance AP 1985. Paryphanta at Pawakatutu. Tane 3 I : 13-18. Barker GM 1999. Naturalised terrestrial Stylommatophora (Mollusca: Gastropoda). Fauna of New Zealand. Manaaki Whenua Press, Lincoln, New Zealand. 253 pp. Baur B 1988. Do the risks of egg cannibalism and desiccation influence the choice of oviposition sites in the land snail Arianta arbustorum? J. Zool. (Lond.) 216: 495-502. 1989. Growth and reproduction of the minute land snail Punctum pygmaeum (Draparnaud). J. Molluscan Stud. 55: 383-387. Berry JA, Green OR, & Shattuck SO 1997. Species of Pheidole Westwood (Hymenoptera: Formicidae) established in New Zealand. N.Z. J. Zool. 24: 25-33. Bradbury K 1977. Identification of earthworms in mammalian scats. J. Zool. 35: 553-555. Brockie RE 1959. Observations on the food of the hedgehog (Erinaceus europeus L.) in New Zealand. N.Z. J. Sci. 2: 121-1 36. 1992. A living New Zealand Forest. David Bateman Auckland. 172 pp. Brook FJ 1999. Stratigraphy, land snail faunas, and palaeoenvironmental history of coastal dunefields at Te Werahi, northernmost New Zealand. J. R. Soc. N.Z. 29: 361-393. 2000. Prehistoric pedation of the land snail Placostylus arnbagiosus Suter (Stylommatophora: Bulimulidae), and evidence for the timing of establishment of rats in northernmost New Zealand. J. R. SOC. N.Z. 30: 227-241. Chatfield JE 1968. The life history of the helicid snail Monacha cantiana (Montagu), with reference also to M. cartusiana (Muller). Proc. Malacol. Soc. Lond. 38: 233-245. Coad N 1998. The kauri snail (Paryphanta busbyi busbyi).

250 Stringer et al. Its ecology and the impact of introduced predators. M.Sc. thesis, University of Auckland, New Zealand. 160 pp. Daugherty CH, Gibbs GW, & Hitchmough RA 1993. Megaisland or micro-continent? New Zealand and its fauna. Trends Ecol. Evol. 8: 437-442. Dell RK 1949. A note on the feeding habits of the Paryphantidae (Mollusca) with special reference to cannibalism in Rhytida. N.Z. Sci. Rev. 7: 104. 1955. A contribution to the study of rates of growth in Paqphanta busbyi (Gray) (Mollusca, Pulmonata). Rec. Dom. Mus. (Wellington) 2: 145-146. Devine CD 1997. Some aspects of behaviour and ecology of the land snail Powelliphanta traversi traversi (Rhytididae: Rhytidinae). M.Sc. thesis, Massey University, New Zealand. 137 pp. Efford M 1998. Distribution and status of native carnivorous land snails in the genera Wainuia and Rhytida. Science for Conservation 101. 48 pp. 2000. Consumption of amphipods by the New Zealand land snail Wainuia urnula (Pulmonata: Rhytididae). J. Molluscan Stud. 66(1): 45-52. Efford M & Bokeloh D 1991. Some results from a study of the land snail Wainuia urnula (Pulmonata: Rhytididae) and their implications for snail conservation. DSIR Land Resources contract report 91/1. 22 pp. Gardner RO & Bartlett JK 1980. Forest flora of the North Cape region. Tane 26: 223-234. Green OR 1992. New Zealand ants (Hymenoptera: Formicidae). Distribution and effects. Proc. 41st annual conference of the Entomological Society of New Zealand, Heretaunga, Wellington, Entomological Society of New Zealand, pp. 67-72. Goulstone JF, Mayhill PC, & Parrish GR 1993. An illustrated guide to the land Mollusca of the Te Paki Ecological Region, Northland, New Zealand. Tane 34: 1-32. Harris RJ 2001. Argentine ant (Linepithema humile) and other adventive ants in New Zealand. DOC Science Internal Series No. 7. Department of Conservation, Wellington. 20 pp. Heather BD & Robertson HA 1996. The Field Guide to the Birds of New Zealand. Viking, Auckland. 432 pp. Heatwole H & Heatwole A 1978. Ecology of the Puerto Rican camaenid tree snails. Malacologia 17: 241-315. Herbert DG 1991. South Africa s carnivorous snails. Afr. Wild]. 45: 6-1 1. Hutton FW 1881. Notes on some pulmonate Mollusca. Trans. Proc. N.Z. Inst. 14: 150-158 + 2 plates. King CM 1990. The Handbook of New Zealand Mammals. Oxford University Press, Auckland. 600 pp. Laing DJ 1982. Snail-eating behaviour of the tunnel web spider Porrhothele antipodiana Mygalomorphae. Tuatara 25(2): 74-81. Lee KE 1959. The Earthworm Fauna of New Zealand. Wellington, R.E. Owen, Government Printer. 486 pp. Lovei GL, Stringer IAN, Devine CD, & Cartellieri M 1997. The harmonic radar-a method using inexpensive tags to study invertebrate movement on land. N.Z. J. Ecol. 21(2): 187-193. McLauchlan CF 1951. Basic work on the life cycle of some Australian snails. Proc. R. Zool. Soc. N.S.W. 1949-50: 26-36. Mead AR 1950. Comparative genital anatomy of some African Achatinidae (Pulmonata). Bull. Mus. Comp. Zool. Harv. Univ. 105: 219-291. Meads M 1990. Forgotten Fauna. Wellington, DSIR Publishing. 95 pp. Meads MJ, Walker KJ, & Elliott GP 1984. Status, conservation, and management of the land snails of the genus Powelliphanta (Mollusca: Pulmonata). N.Z. J. Zool. 1 1 : 277-306. Millener PR 1981. The quaternary avifauna of the North Island, New Zealand. PhD thesis, University of Auckland, New Zealand. 897 pp. Molloy J, Davis A, & Tisdall C 1994. Setting Priorities for the Conservation of New Zealand s Threatened Plants and Animals. Wellington, Department of Conservation. 64 pp. Montefiore R 1996. Report on kauri snails, Paryphanta busbyi busbyi, in the Waitakere Ranges. Unpublished report to Department of Conservation and Auckland City Council. 17 pp. Nisbet RH 1974. The life of Achatinidae in London. Proc. Malacol. Soc. Lond. 41: 172-283. New Zealand Meteorological Service 1983. Summaries of Climatological Observations to 1980. NZ Meteorological Service Misc. Publ. 177. Ministry of Transport, Wellington. 177 pp. O Connor AC 1945. Notes on the eggs of New Zealand Paryphantidae, with description of a new subgenus. Trans. R. Soc. N.Z. 2001. 75(1): 54-56. Ohms PG 1948. Some aspects of the anatomy of Pavyphanta busbyi. M.Sc. thesis. University of Auckland. 65 pp. + 20 plates. Owiny AM 1974. Some aspects of the breeding biology of the equitorial land snail Limicolaria martensiana (Achatinidae: Pulmonata). J. Zool. (Lond.) 172: 191-206. Parrish R 1992. Paiyphunta busbyi watti survey-te Paki 18-19 March 1992. Unpublished report, Department of Conservation. 4 pp. Parrish R, Sherley G, & Aviss M 1995. Giant land snail recovery plan Placostylus spp., Paryphanta spp. Wellington, Threatened Species Unit, Department of Conservation. 13 pp. Plummer JM 1975. Observations on the reproduction, growth, and longevity of a laboratory colony of Achatinu (Calachatina) marginata (Swainson) subspecies ovum. Proc. Malacol. Soc. Lond. 41: 395-413. Powell AWB 1930. The Paryphantidae of New Zealand: their hypothetical ancestry, with descriptions of new species and a new genus. Rec. Auckl. Inst. Mus. l(1): 17-56. 1946. The Paryphantidae of New Zealand. No. V. Further new species of Paryphanta, Wainuia, and Rhytidu. Rec. Auckl. Inst. Mus. 3(2): 99-136 + plates 8-10, 1979. New Zealand Mollusca. Marine, Land, and Freshwater Shells. Collins, Auckland. 500 pp. Runham NW & Laryea AA 1968. Studies on the maturation

Land-snail biology and conservation 25 1 of the reproductive system of Agrolimax reticulatus (Pulmonata: Limacidae). Malacologia 7( l): 93-108. Sherley G 1993. Research and management progress report on Placostylus flax snails at Te Paki, Far North, and Whangaruru North Head Scenic Reserve: October 1993. Unpublished report, Department of Conservation, Wellington. 24 pp. Sherley GH, Stringer IAN, Parrish GR, & Flux I 1998. Demography of two land snail populations (Placostylus ambagiosus, Pulmonata: Bulimulidae) in relation to predator control in the far north of New Zealand. Biol. Conserv. 84: 83-88. Smith BJ 1971. Carnivorous snails of the family Paryphantidae. Aust. Nat. Hist. 17: 55-58. Solem A & Christensen CC 1984. Camaenid land snail reproductive cycle and growth patterns in semiarid areas of north-western Australia. Aust. J. Zool. 32: 471-491. Southey IC 1985. The ecology of three rare skinks. M.Sc. thesis. University of Auckland. 99 pp. Stringer IAN & Montefiore R 2000. Distribution and biology of the endangered kauri snail, Paryphanta busbyi watti. Science for Conservation 163. 42 pp. Stringer IAN, McLean MJ, Arnold GC, Bassett SM, & Montefiore R 2002. Growth and development of the rare land snail Paryphanta busbyi watti (Pulmonata: Rhytididae). Moll. Res. 22: 203-220. Suter H 1899. Description of a new variety and five new species of New Zealand land Mollusca. Proc. Malacol. SOC. Lond. 3: 286-291. Ussher G 1995. Feeding ecology and dietary interactions of tuatara and kiore on the Chickens Islands. M.Sc. thesis, University of Auckland, Auckland. 186 pp. Vause P 1977. Some observations on Puryphanta busbyi busbyi (Gray 1840). Poirieria 9: 48-51. Wildlife Act 1953. 7th Schedule, Sections 2, 7B. Government Printer, Wellington. Wolda H 1970. Variation in growth rate in the land snail Cepaea nernovnlis. Res. Popul. Ecol. (Tokyo) 12: 185-204. Wolda H & Kreulen DA 1973. Ecology of some experimental populations of the land snail Cepaea nernoralis L. 11. Production and survival of eggs and juveniles. Neth. J. Zool. 23: 168-188.