Results from a legume oversowing and phosphorus topdressing experiment are reported. The depleted tussock grassland site at 700 m altitude with 1000 mm rainfall, soil pH of 5.25 and low available P plus Hieracium weed invasion was typical of much South Island montane grazing land. 'Grasslands Maku' lotus (Lotus pedunculatus Cav.) and 'Grasslands Huia' white clover (Trilolium repens L.) established rapidly and yields in the second year were similar but lotus gave higher yields subsequently. White clover had lower P% in leaf dry matter and higher scarab populations than lotus. Trijolium umbiguum cv. Prairie establishment was sparse and yields were low at first but after five years it had spread by rhizome growth to form a moderately productive sward. All legume species responded to increasing rates of P at establishment and to greater frequency of maintenance superphosphate but the most efficient superphosphate regime was for low to moderate rates (123 to 300 kg/ha) at sowing reinforced by annual topdressings at 125 kg/ha. Legume oversowing with fertilizer substantially reduced Hieracium pillosella cover.

No abstract

This study describes an assessment of the medium-term effects of selection logging on the reptiles and frogs in the montane forests near Queanbeyan, New South Wales. A total of 15 reptile and two frog species was detected across 20 fixed-area plots that were selected according to broad forest type and logging history. Reptile species richness was not significantly different among the different habitat treatments. Total reptile abundance was significantly higher on logged plots, but was not influenced by forest moisture type. Individual analyses for five lizard species (accounting for 84% of all data) showed that several species (Eulamprus tympanum, E. heatwolei and Pseudemoia spenceri) had significantly greater abundances on the logged plots than on the unlogged plots, while two species (Nannoscincus maccoyi and Niveoscincus coventryi) did not differ significantly in their abundances across treatments. Thus, none of these species (which included arboreal, terrestrial and fossorial species) appeared to be adversely affected by past logging practices. Insufficient data on frogs were obtained for analysis. An additional 26 species (13 reptile and 13 frog species) were recorded during a regional survey involving 57 sites. Only two frog species were detected at more than 10 sites, illustrating the difficulty in obtaining data to determine whether logging has any impact on frog populations. Snakes are similarly difficult to assess: 13 snakes from six species were detected in this study. It is concluded that (i) the most abundant reptile species were not adversely affected by past logging, and (ii) many species of herpetofauna were patchy in their distribution across the study area, preventing resolution of the influence of habitat type or disturbance. The difficulty in collecting quantitative data on frogs and snakes confirms the importance of surveys that target endangered frog and snake species so that protective buffer zones can be delineated. Further research is required on these groups to assess the impact of forest management practices.

The pattern of distribution of the invasive vine Clematis vitalba in a typical New Zealand native forest remnant and the influence of chilling duration (chilling), nitrate (NO3-), and light on germination of C. vitalba were investigated. In the forest remnant, C. vitalba was not established in undisturbed forest where light levels were <3% full sunlight and rarely established where light levels were <5%. Clematis vitalba was, however, established in forest gaps and margins where light levels exceeded 5%, particularly when these areas were associated with recent soil disturbance. Under controlled environment conditions, germination in the absence of chilling, NO3-, or light was low (3%). When applied alone, chilling and NO 3-increased germination. Light, however, only increased germination when applied with one or both other factors. In the pres- B97006Received 27 January 1997; accepted 27 May 1997 ence of light, germination increased up to c. 80% with increased chilling (0-12 weeks), and increased up to c. 50% with increased applied NO 3-in the range 0.1-2.5 mol/m 3 but then changed little with additional NO3-up to 50.0 mol/m 3. In general, germination in the presence of two or three factors was greater than when either factor was applied alone.In the field, seeds remaining on vines over winter retained a high degree of dormancy and viability. We suggest that sporadic release of this seed effectively produces a seed bank. By comparing the levels of chilling, NO3-, and light that can be expected in the field with the levels that influence germination under controlled environment conditions, and considering a seed bank, we argue that C. vitalba is likely to rapidly establish following soil disturbance. We suggest that rapid establishment may account for the distribution of C. vitalba in recently disturbed, high light sites and, to some extent, the success of C. vitalba in New Zealand native forest remnants.

Captive bred Green and Golden Bell Frog Litoria aurea tadpoles were introduced to a coastal wetland near Pambula on the far south coast of New South Wales. The reintroduction involved the release of approximately 5000 captive-bred tadpoles and subsequent monitoring. Before the reintroduction could take place several requirements had to be satisfied. A re-introduction proposal was prepared for the Department of Environment and Conservation (DEC) and independently reviewed by two referees, pre-release surveys of frogs were undertaken to determine if L. aurea existed on the site and to ascertain if the pathogen Bactrachochytrium dendrobatiodis (chytrid) was present within the endemic frog population. The captive bred tadpoles were tested for the presence of chytrid prior to release. The water quality was tested at the proposed release site and potential predators were removed. The project required maintenance and breeding of the captive population, and post release surveys. The project took 4.5 years from submission of the proposal to the first release of tadpoles and to date has cost approximately $190K, of which only $37K was funded. It is anticipated that ongoing costs will be in the order of $25K per year for the next four years . Details of the project costs and chronology of actions are given in order to assist others who wish to undertake similar projects. So far two adults have been detected 13 months post-release.