Methods:
Historic Flora
In order to determine the species that were documented to have occurred at Harvard Forest prior to our recent field investigations, we conducted a comprehensive review of all relevant herbarium specimens and published and unpublished manuscripts. The primary sources of information were: (1) the Harvard Forest Herbarium. Prior to our study, the HF Herbarium included more than 2,300 specimens collected in or near Harvard Forest, more than 85% of which were collected in the first half of the 20th century, primarily during three periods: 1908-1913 (~ 10% of specimens), 1928-34 (~35%), and 1947-1949 (~45%). All specimens in the Harvard Forest herbarium were verified to correct taxonomic errors and update nomenclature; and (2) previous studies with detailed information on the early flora of Harvard Forest, especially Jack (1911) and Smith (1948a, 1949). Because Raup’s (1938) checklist of the flora of Petersham does not distinguish species found at Harvard Forest from those found elsewhere in Petersham, we have accepted only those species from Raup (1938) for which we have specimens from Harvard Forest.
We also searched regional herbaria (HUH, MASS) for specimens from HF, and located ~ 180 C. E. Smith Jr. specimens collected at HF in 1947 in the Brandeis University Archives. This collection has since been generously donated to HF and incorporated into the Harvard Forest Herbarium. In August 2007, when we were nearing completion of this project, we located an additional ~1400 unmounted specimens collected at or near Harvard Forest prior to 1950 that were apparently held in the personal collection of Hugh Raup and stored at HF for an unknown period of years or decades. This collection included: ~ 350 "new" specimens (i.e., non-duplicates of specimens in the HFH) from the collectors that were active in 1908-1913 and 1928-1934, ~ 450 specimens collected by H. M. Raup and others that were duplicates of specimens in the HF herbarium, ~ 170 "new" H. M. Raup specimens collected in 1937-1938, and more than 400 "new" specimens collected by K. A. Raup in 1947. Specimens of species that were otherwise undocumented from Harvard Forest, as well as specimens that represent the earliest records for a species from Harvard Forest, were mounted and entered into the HF herbarium, and these records were included in all data analyses. Unmounted duplicates of specimens in the HFH (identified by specimen number) and specimens of common species for which we had other herbarium specimens were discarded. Nomenclature follows Sorrie and Somers (1999) or Flora of North America, except where indicated in the text.
Modern Flora
We conducted extensive botanical surveys across Harvard Forest in 2004-2005; limited additional field work in 2006 and 2007 targeted species that we suspected to occur at HF but that had not been located or adequately documented in our earlier surveys. Specimens were collected for laboratory determination when field identification was uncertain. We also attempted to collect for the HF Herbarium at least one specimen of every species that occurs at HF for which we found no earlier specimens.
For organizational purposes, surveys were conducted within the 33 administrative ‘compartments’ that comprise the 4 main tracts of Harvard Forest (Prospect Hill, Tom Swamp, Slab City, and Simes); Harvard Pond was designated as Tom Swamp 11. Several small HF tracts were excluded from our field surveys because of time constraints (e.g., Schwarz, Pisgah, Hamilton, Tall Timbers). Several recent additions to Harvard Forest property were also excluded (e.g. Bryant, Wilson, Flint, and Gould lots) so as to ensure that our data are comparable with those from the early 20th century. Although the Simes Tract was not acquired by Harvard Forest until 1977, we included this 125-ha area in our survey because of numerous historical collections from "Camel’s Hump" or other areas (e.g., portions of Dugway Rd.) that have since been incorporated into the Simes Tract.
During field surveys, one or more of the authors conducted thorough searches of each habitat or "plant community" within a compartment, recording all vascular plant species that were naturally occurring. Species that were planted at Harvard Forest were included only in cases where they had clearly spread from the original plantings and become naturalized. We noted cases where we were uncertain of whether a species has truly spread and become naturalized. Most compartments were visited several times to account for seasonal variation and to search for species that were likely to be present but were not previously recorded. Although field inventories covered all major habitats within each compartment, our primary objective was to document the flora of Harvard Forest as a whole, not to develop comprehensive compartment lists. Thus, species that we recorded frequently are likely to occur in at least some compartments where we did not record them, and species that we recorded rarely may occasionally be found elsewhere. Roadsides at the boundaries of Harvard Forest were included in our field surveys, as was the ~ 15 m-wide cut boundary of the Quabbin Reservation on the western border of the Simes Tract.