1978). Groups are usually defined a priori by some criterion external to the set of variables measured, e.g., rootstocks (Cruz-Castillo et al., 1992), carbohydrate status (Prins et al., 1990), supermarkets selling flowers (Behe et al., 1992), tree species (Majer et al., 1992), or swede (Brassica napus L., Napobrassica Group) cultivars (Cole and Phelps, 1979). The objective of this paper is to present a description of CDA and demonstrate its applicability and potential in horticultural research.

Extant research in evaluation of accounting harmonisation has mainly focused on measurement of harmonisation and evaluation of its impact on share market indicators. Within these themes, however, researchers have been studying the impact of formal harmonisation on material harmonisation, and share prices and returns. Given the important role formal harmonisation plays in accounting harmonisation, this study introduces a methodology for the measurement of formal accounting harmony between countries. The methodology introduced in this paper can be used to identify areas of harmony or disharmony for policy-making purposes. It also provides means of gaining valuable insights into the nature of harmony or disharmony between different countries' accounting rules. Finally, it creates new grounds for examining various other aspects of accounting harmonisation.

Levels of nitrogen, phosphorus and potassium were measured for the foliage of two co-dominant eucalypts at each of two sites, one in eastern Australia and the other in Western Australia. In eastern Australia, foliage was sampled in the canopy and subcanopy for narrowleaved ironbark Eucalyptus crebra and grey box E. moUucana and in Western Australia, for jarrah E. marginata and marri E. calophylla. The Western Australian trees were also sampled for 'young' and 'old' leaves. Both eucalypts in eastern Australia had greater nitrogen and phosphorus levels, but lower potassium, than E. marginata or E. calophylla. Eucalyptus calophylla foliage had greater levels of all three nutrients than E. marginata foliage as did E. crebra relative to E. mollucana. At both sites, foliar nutrient levels were greater in the canopy than subcanopy foliage, and, at least in Western Australia, the younger leaves had greater nutrient levels than the older leaves. The observed differences in foliar nutrient levels are consistent with observed trends in the abundance and diversity of foliage arthropods and the use of the trees as foraging substrates by birds.

No abstract

Linear discriminant analysis between two populations is considered in this paper. Error rate is reviewed as a criterion for selection of variables, and a stepwise procedure is outlined that selects variables on the basis of empirical estimates of error. Problems with assessment of the selected variables are highlighted. A leave-one-out method is proposed for estimating the true error rate of the selected variables, or alternatively of the selection procedure itself. Monte Carlo simulations, of multivariate binary as well as multivariate normal data, demonstrate the feasibility of the proposed method and indicate its much greater accuracy relative to that of other available methods.