In June of 1962, it was my pleasure to present at the Third Annual Meeting of The Society for Economic Botany a plan for an unusual approach to the search for useful plants (1). It consisted of making greater use of major herbaria by extracting from them all unpublished label data which might be potentially valuable to man from the standpoint of health and medicine. I would like to report today on the realization of this plan at one major institution and on what appears to be a large windfall of possible food plants among our findings. The search to which I refer was carried out through the sponsorship of the Botanical Museum of Harvard University and consisted of a sheet by sheet examination of the estimated 2,500,000 specimens of flowering plants in the combined collections of the Arnold Arboretum and Gray Herbarium of that University.3 Two persons trained in herbarium techniques and working less than full time were able to complete the search in five years. The work was supported in turn by Smith, Kline & French Laboratories; the National Institute of Mental Health; and the Eli Lilly Research Laboratories. The information sought and recorded covered a broad range of data, including not only specific folk or native remedies but any notes allowing possible physiological activity. The search was undertaken in the belief that herbaria represent rich and untapped reservoirs of such information, which, in some instances, might provide the only remaining clews to medical materials of peoples the cultures of which already have been absorbed by civilization. We were eager also to recognize species which may not have been used by