Plant community succession is one of the most ubiquitous of ecological processes. The change in structure and species composition of assemblages of plants after physical disturbances, or after release from agricultural management, has provided much grist for ecology. Indeed, the founding of the science of ecology in the United States is closely associated with studies of succession (Cowles 1899, Clements 1916, Cooper 1926). The differences in conditions and interactions that develop through succession provide major contrasts that plants and animals exploit. Thus, succession is one of the major sources of diversity in the living world. Life history and evolutionary contrasts between species, physiological and morphological strategies, assembly rules, and ecosystem processes are among the various ecological processes that assort along successional gradients. Differences in the successional status of different patches in a landscape are among the major sources of biological diversity.

Because succession perfuses so much of ecology, and because the change in communities is crucial to management and conservation, it has been important to learn how the process occurs. Fundamental to the understanding of succession is the need to know what the patterns of community change through time actually are. All else -- the understanding of mechanisms, the prediction of trends, the use of succession by managers -- depends on a sound knowledge of the patterns of change. In the early days of ecology, the only method available to discover the patterns of community change through time was to compare sites of different ages since disturbance or abandonment. This method, called either space-for-time substitution or chronosequence, assumes that the different sites are subject to the same conditions and have the same species available to them. If this crucial assumption is not met, the patterns may reflect permanent differences between the sites or other ecological processes rather than successional change. It was this assumption that the Buell's and John Small wished to test.

The Piedmont of New Jersey, where the BSS is located, shares with many other sites in that geomorphic province in the eastern U.S., soils that are problematic in some ways for agriculture, or location in the path of urban spread. Very intense use, erodability, or droughtiness characterize many Piedmont farms. Therefore, many Piedmont sites were abandoned from agriculture as a result of opening more hospitable soils in the Midwest and changes in the economic and social situation for agriculture in the east. Many fields in the Piedmont of central New Jersey were abandoned in the 1940's. These had been the subject of a study using space-for-time substitution (Bard 1950). The fields surrounding Mettler's Woods, held by descendants of Mynheer Cornelius VanLiew, the original Dutch settler who established the farm in 1701, were well cared for and reasonably productive. But the changing situation in the 1950's led to the selling of the farm and the subsequent abandonment of agriculture. The Buell's and Small wished to determine whether the results obtained by the chronosequence that Bard constructed in abandoned fields near HMF held when one examined specific fields through time.

The Buell's therefore decided to study succession in specific fields through time. Indeed, not only would the same fields be traced through time, but the very same study plots would be used. Based on consultations with a statistician in the early years of the study, the Buell's decided that 48 plots would be permanently marked in each of the 10 fields. These plots were 1 m2, and were rectangular to capture heterogeneity in the herbaceous and shrubby communities expected to dominate in the first decades of succession.