The goal of understanding in limnology

Part of the charm of science is its unpredictability R. MacArthur Limnology has enjoyed a long tradition of progress based on thoughtful interpretation of empirical findings. New understanding has grown in part from refining the details of known relationships but more often from uncovering interactions and processes that were previously unknown. The field is divided at present into camps which pursue the study of integral properties like biomass, productivity, and nutrient fluxes, and those which study the biological entities at the level of populations or communities. The conceptual dichotomy is a split between primary attention to the laws of ther-modynamics or to the law of evolution by natural selection. Current challenges require steps to bridge this unfortunate gap and to broaden the conceptual bases of all ecological studies. In recent years ecologists have been presenting introspective examinations of their discipline at an alarming rate, as though practicing scientists need philosophical Andrewartha 1984). Readers are variously charged to follow prescribed methods of problem solving, to heed the apparently strong dichotomy between holism and re-ductionism, and sometimes to seek " pre-dictive power " rather than causality and derivation. The issues have reached prominent proportions in ecology within the past decade, but limnology was never immune from the debate. Evidently G. E. Hutchin-son unknowingly was stung by it during his tenure review when C. Juday argued that his theories about lake processes lacked the weight of real data tonnage (Hutchinson 1979, p. 246-247). In the later years of their collaboration, Juday and E. A. Birge had developed an approach based on graphical correlation and rigid empiricism. They and their colleagues surveyed hundreds of lakes for properties like transparency, organic matter, and phosphorus. They then plotted them, either as frequency distributions or mean trend lines (e.g. Juday and Birge 193 1, 1932; Juday et al. 1935) and voila-ex data venit veritas. The approach persists for practical reasons and it continues to be the most sensible way to categorize masses of data. What apparently got Hutchinson into hot water with some of the midwestern lim-nologists was his propensity to speculate and generalize beyond the data at hand. Years later the drive to comprehend empirical results in a broad conceptual context prompted one of his students to write " Scientists are perennially aware that it is best not to trust theory until it is confirmed by evidence. It is equally true. .. that it is best …