Since the purpose of an index is to retrieve information rapidly, the index must be closely correlated with the objectives of its users (2). An index may deal with general or specialized information. Examples within the field of agriculture range from the computerized Catalog and Index (CAIN, AGRICOLA) for the Bibliography of Agriculture (3) to the table of contents of the Pisum Newsletter. The scope of subject matter is the most important determinant of the method and format of indexing.
The hypothesis that carbon balance is the basis for differences in responses by lightly and normally cropped apple trees to European red mite (ERM) [Panonychus ulmi (Koch)] damage was tested. Mature `Starkrimson Delicious' (Malus domestica Borkh.)/M.26 apple trees were hand-thinned to light (125 fruit/tree, about 20 t/ha) or normal (300 fruit/tree, about 40 t/ha) target crop levels and infested with low [<100 cumulative mite-days (CMD)], medium (400 to 1000 CMD) or high (>1000 CMD) target levels of ERM. A range of crop loads and CMD was obtained. Mite population density, fruit growth, leaf and whole-canopy net CO2 exchange rates (NCER) were measured throughout the growing season of 1994. Leaf area and vegetative growth per tree were also measured. Yield and final mean fruit size were determined at harvest. Return bloom and fruiting were determined the following year. Total shoot length per tree was not affected by crop load or mite damage. ERM reduced leaf and whole-canopy NCER. Normally cropped trees showed fruit weight reduction earlier and more severely than lightly cropped trees with high mite injury. Variation in final fruit weight, return bloom and return fruiting was much better related to whole-canopy NCER per fruit than to CMD.
Fruit maturity, quality, calcium concentration and economic value of `Starkrimson Delicious' (Malus domestica Borkh.) apples, under a range of crop levels and European red mite [Panonychus ulmi (Koch)] cumulative mite-days (CMD), were best explained by local surface regression models involving CMD and crop load. Fruit from trees with low CMD and a light crop (125 fruit/tree, about 20 t/ha) were the most mature at harvest. Those fruit had higher ethylene concentrations, starch pattern indices, soluble solids concentrations, and watercore incidence at harvest than fruit from trees with low CMD and a normal crop (300 fruit/tree, about 40 t/ha), or with high CMD at any crop level. Those fruit also had higher incidences of watercore and internal breakdown after 4 months of cold storage. Calcium concentrations in fruit increased as crop load and CMD increased. Whole-canopy net CO2 exchange rate per fruit related better to fruit quality and calcium concentrations than either crop load or CMD alone, but was always a much worse predictor than local surface regressions. Low CMD and normally cropped trees had the highest crop value; lightly cropped trees had an intermediate crop value; while high CMD and normally cropped trees had the lowest crop economic value. Crop load should be considered when defining action thresholds for mites, and harvest schedules for apples should reflect crop load and mite populations on apple trees.