David J. Saville

Multiple Comparison Procedures: The Practical Solution

Abstract A practicing statistician looks at the multiple comparison controversy and related issues through the eyes of the users. The concept of consistency is introduced and discussed in relation to five of the more common multiple comparison procedures. All of the procedures are found to be inconsistent except the simplest procedure, the unrestricted least significant difference (LSD) procedure (or multiple t test). For this and other reasons the unrestricted LSD procedure is recommended for general use, with the proviso that it should be viewed as a hypothesis generator rather than as a method for simultaneous hypothesis generation and testing. The implications for Scheffes test for general contrasts are also discussed, and a new recommendation is made.

Effect of herbage species and feeding level on internal parasites and production performance of grazing lambs

Abstract The effect of internal parasites on the performance of lambs grazing different forage species was examined in two experiments. In Experiment 1, 300 Coopworth lambs grazed stands of Nui ryegrass, Wana cocksfoot, Au Triumph tall fescue, Puna chicory, or Otaio lucerne for 64 days between February and April 1992. Lambs were either infected with 10000 mixed species L3 larvae at the start of the grazing or maintained free of parasites by fortnightly drenching with ivermectin. Herbage was offered at either 2 or 4 kg DM/head per day (green). In Experiment 2, Nui ryegrass was replaced with low‐endophyte Ruanui ryegrass and lambs were infected with 19500 L3 larvae 21 days before grazing which commenced on 2 February 1993 and continued for 100 days. In Experiment 1, few herbage larvae were recovered during grazing and differences among forage species were not significant. In Experiment 2, more larvae per kg DM were recovered from the grass swards than from either chicory or lucerne (P < 0.05), differences a...

Demographic and Biomass Production Consequences of Inundative Treatment of Cirsium arvense with Sclerotinia sclerotiorum

The adventitious shoots in three populations of Cirsium arvense in sheep-grazed pastures were treated in October (spring) 1991 with a mycelium/wheat formulation of Sclerotinia sclerotiorum and the fates of mapped shoots were followed over the growing season. In untreated plots, deaths through natural causes were compensated for by births (emergence of new shoots above the soil) throughout the growing season, but, on plots treated with S. sclerotiorum, deaths from the induced disease exceeded births for 35 days following treatment, causing the shoot population to decline markedly. Disease-induced deaths occurred only among shoots present at the time of treatment; there was no evidence of transfer of the pathogen to shoots emerging after the treatment was applied. A life-table analysis showed that only 8% of the adventitious shoots emerging during the growing season survived to seeding on treated plots, compared with 28% on the untreated plots; most mortalities occurred in shoots at the vegetative stage of ...

Effect of Application Time on the Efficacy of Sclerotinia sclerotiorum as a Mycoherbicide for Cirsium arvense Control in Pasture

An experiment was conducted in sheep-grazed pasture in three regions of New Zealand over three years to evaluate the effect of application time on the efficacy of a dry granule myceliumon-wheat formulation of Sclerotinia sclerotiorum for the biological control of Cirsium arvense . At each site, the experimental mycoherbicide was applied to a previously untreated plot in each month of the year for three years at a dose of 50 g m -2 . Applications made during the spring and early summer months of October, November and December significantly reduced the ground cover of C. arvense for 67, 67 and 44%, respectively of these applications. Reduced ground covers ranged from 38 to 81% of the cover on untreated plots. Applications in late summer and autumn were less effective. Correlations of ground cover by C. arvense with climate parameters suggested that free moisture promoted treatment efficacy, but that intense rainfall after treatment reduced efficacy through wash off. The importance of leaf wetness for the efficacy of the mycoherbicide was confirmed by comparing disease development and mortality in C. arvense shoots with and without enhanced moisture levels under field conditions. A water miscible formulation applied as a slurry was less dependent upon leaf wetness than the dry granule formulation.

Dairy production revenue losses in New Zealand due to giant buttercup (Ranunculus acris)

Abstract Giant buttercup (Ranunculus acris L.), a species of European origin, is widespread throughout New Zealand, and is an intractable weed of dairy pastures in parts of the south Auckland, Hawkes Bay, and Taranaki regions and in the districts of Southern Wairarapa, Horowhenua, and Tasman. In this study, the seasonal progression of the weeds ground cover in dairy pastures in the Tasman District was measured during 1989–90 and used, in conjunction with monthly weed‐free pasture dry matter production values, to estimate the annual percentage loss in pasture production due to a typical infestation of the weed subjected to typical control measures. Extrapolation to other infested regions and districts provided a national estimate of milk solids revenue loss due to giant buttercup in New Zealand dairy pastures of

Does transmission of the rust pathogen, Puccinia punctiformis, require stem mining vectors?

156 million in the 2001–02 season.

Risk analysis for biological weed control - simulating dispersal of Sclerotinia sclerotiorum (Lib.) de Bary ascospores from a pasture after biological control of Cirsium arvense (L.) Scop.

Abstract Recent research in Europe has suggested that stem mining insects may be important for vectoring the pathogen Puccinia punctiformis, and largely responsible for its systemic infection in the weed, Cirsium arvense. However, here we present comparative survey data showing that the level of systemic disease is the same in Europe and New Zealand, with and without stem miners, respectively, casting doubt on the idea that these insects are necessary for transmission of the fungus.

Chondrostereum purpureum and Fusarium tumidum independently reduce regrowth in gorse (Ulex europaeus)

Biological control of Cirsium arvense(L.) Scop. in pasture by the plurivorous plantpathogenic fungus Sclerotiniasclerotiorum (Lib.) de Bary mayresult in the formation, escape and aerialdispersal of ascospores, creating an additionaldisease risk in down-wind market garden crops. To determine the width of a safety zone for apasture subjected to this form of weed control,we simulated the spatial pattern in the ratioof added (due to biocontrol) to naturallyoccurring airborne ascospores (due to marketgarden crops) around a 1ha virtual biocontrolpasture under either sheep or dairy cattlegazing over a 91-day emission period in 1996 inCanterbury, New Zealand. This was achievedusing a unique combination of two computermodels; SPORESIM-1D (for spore escape from avegetation source) and PC-STACKS (a modernGaussian plume model for dispersal beyond asource). Plumes of dispersing ascospores weremodelled for each hour of the emission periodfor both the virtual market garden andbiocontrol sites, and the aerial density of theascospores was averaged over the period. Assuming that a 1:1 ratio of added to naturallypresent spores is acceptable, no safety zonewas necessary for either of the modeledpastures. A ten-fold ratio (1:10 added tonatural) necessitated safety zones of 300 and150 m for the sheep and dairy pasturerespectively. Uncertainties associated withextrapolation of this conclusion to individualpasture management scenarios, and to otheryears and climatically different regions arediscussed.

Survival and germination of shallow‐buried sclerotia of Sclerotinia sclerotiorum in pastures in Canterbury

Abstract An experiment was conducted in two gorse populations (Ulex europaeus) in which Chondrostereum purpureum was applied each month as mycelial-agar cultures to the wounds of decapitated stems of previously untreated gorse plants to determine its potential as a mycoherbicide. Summer–autumn (Feb–May) or late winter–early spring (Aug–Sept) applications were effective in both populations, halving stem stump survival (from an average of 56 to 29%). Another experiment in the same gorse populations evaluated the combined effects of C. purpureum applied in May to decapitated stems, and Fusarium tumidum applied as spores in an invert emulsion to regenerative shoots 5–6 months later. There was no evidence of synergism between the two fungi; each pathogen independently reduced the density of regenerative shoots on the decapitated stems by 39–63% averaged over the 12 months following their respective applications. It is concluded that both pathogens have potential as mycoherbicides for gorse regenerating after stem decapitation.

Effects of pasture competition and specialist herbivory on the performance of Cirsium arvense

Abstract Laboratory‐produced Sclerotinia sclerotiorum sclerotia of two size ranges ( 4 mm diam.) were buried in mesh bags in sheltered or exposed positions in the A horizon under two pastures in Canterbury in June. This emulated sclerotia entering a heavy and a light soil after mycoherbicide application to Californian thistle in pasture. The bags, each containing 20 sclerotia were lifted 3,5, and 7 months after burial. Large sclerotia survived better than small sclerotia. The percentage of sclerotia producing mycelium in culture was less after 5 months burial than after 3 months, but was then greater after 7 months than after 5 months. Carpogenic germination was greater in exposed than in sheltered locations especially at the site with heavy soil, and was not influenced by sclerotium size. The number of intact exhumed sclerotia yielding micro‐organisms was greater after 5 months than after 3 months, but less after 7 months than after 5 months, which contrasted with the changes in ability of exhum...