The effects of introduced predators and the invasive weed Tradescantia fluminensis ( Vell.) (Commelinaceae) on the land snail Powelliphanta traversi traversi (Powell) (Gastropoda: Pulmonata: Rhytididae : a thesis presented in partial fulfillment of the requirements for the degree of Master of Science in Ecology at Massey University
Powelliphanta traversi traversi (Powell) was studied at two forest remnants in the Horowhenua District. The effects of introduced predators, predator control, the invasive weed Tradescantia fluminensis (Vell.), and Grazon® herbicide on these snails were investigated. Brodifacoum poison was used in two areas of Lake Papaitonga Scenic Reserve to determine the effect of rodent control on P. t. traversi. Mouse abundance (inferred from tracking tunnel indices) was reduced in both poisoned areas below levels observed in two other areas that were not poisoned. Rat abundance was reduced below pre-poisoning levels but only to levels below one of the non-poison areas. In each poison and non-poison area, four 100m2 quadrats were searched for P. t. traversi snails immediately before poisoning, and 12 and 19 months after poisoning commenced. After 19 months, only one poisoned area showed an overall increase in the number of snails, with significantly more live snails found (45) than at either of the two previous searches (22 before poisoning and 28 after 12 months of poisoning) (P<0.05). Numbers of empty P. t. traversi shells found in each area decreased at each search suggesting that shell accumulation is constant rather than seasonal. Rats were the greatest identified predator of P. t. traversi at Lake Papaitonga (17.87% of all empty shells), but the proportion of shells damaged by blackbirds and song thrushes was also high (11.91% of all empty shells) and increased from pre-poisoning numbers in three of the areas. Overall, there was no conclusive evidence to suggest that the numbers of live P. t. traversi increased as a result of rodent poisoning during the time period of this study. The effect of T. fluminensis on the movements of P. t. traversi at Prouse Bush was determined using harmonic radar. There was large variation in the movements and a highly significant difference between individual snails (P<0.01), with some snails regularly moving between areas of T. fluminensis and leaf litter. There was no significant difference in the mean daily displacement of movements by snails in leaf litter and T. fluminensis, but T. fluminensis did appear to affect home range size. Snails always found under T. fluminensis had significantly smaller mean 90% home range estimates (43.91 m2) than snails that were only ever found in leaf litter or those that moved between litter and T. fluminensis (171.35 m2 and 610.14 m2 respectively) (P<0.05). Snails in T. fluminensis had a significantly wider size-frequency distribution than those in leaf litter (P<0.05) and no live snails <35mm were found in leaf litter. There was no significant difference between the size-frequency distributions of empty shells found in both habitats, but their density was significantly greater in leaf litter (P<0.05). Powelliphanta traversi traversi regularly use T. fluminensis as a habitat and any control measures affecting this weed in native bush remnants need to be considered with regard to their possible effects on these snails. The toxicity of a 1.4% Grazon® solution (active ingredient triclopyr) to P. t. traversi was investigated by first using three life history stages of the brown garden snail (Cantareus aspersus Müller). After 149 days, there was significantly greater mean mortality of C. aspersus exposed to a direct spray and to a sprayed environment (82.34% and 78.40% respectively) than in a control treatment (36.95%) (P<0.05). Cantareus aspersus egg mortality (86.00%) was significantly greater than adult and juvenile snail mortality (66.86 and 62.20% respectively) (P<0.05). Five P. t. traversi snails were also exposed to a single environmental spray of a 1.4% Grazon solution but no mortality or detrimental effects were observed after 149 days. A 1.4% Grazon solution does not appear to be toxic to P. t. traversi snails when sprayed on leaf litter where the snails live so Grazon appears to be a suitable herbicide for controlling T. fluminensis in forest remnants containing P. t. traversi.