Herbaria are a major frontier for species discovery

Abstract

Despite the importance of species discovery, the processes including collecting, recognizing, and describing new species are poorly understood. Data are presented for flowering plants, measuring quantitatively the lag between the date a specimen of a new species was collected for the first time and when it was subsequently described and published. The data from our sample of new species published between 1970 and 2010 show that only 16% were described within five years of being collected for the first time. The description of the remaining 84% involved much older specimens, with nearly one-quarter of new species descriptions involving specimens >50 y old. Extrapolation of these results suggest that, of the estimated 70,000 species still to be described, more than half already have been collected and are stored in herbaria. Effort, funding, and research focus should, therefore, be directed as much to examining extant herbarium material as collecting new material in the field.

Fig. 1.

Herbarium specimen of Strobilanthes frondosa first collected (C) in 1924 from Burma (Cooper 5943A), published 70 y later (D) in 1994 (24). In this example, I equals 70 y. The specimen is from the Royal Botanic Garden Edinburgh (photo courtesy of Prashant Awale).

Fig. 2.

Fraction of specimens remaining to be discovered against discovery time (I), for data from Kew Bulletin (blue lines) and Monographs (red lines). For each series, the central line shows the mean and the upper and lower lines show the 95% confidence limits on the mean. (A) All data: Mean I (and 95% confidence limits) to describe half the species is 23 y (22–25 y) for Kew and 28 y (25–34 y) for Monographs. (B) Specimens collected 1956 (the mean of C), fitted using Cox proportional hazards models: Mean I (and 95% confidence limits) to describe half the species is 32 y (31–33 y) for Kew Bulletin and 31 y (29–35 y) for Monographs. Note that the scales on the abscissa in A and B differ.