Given limited resources, many researchers advocate focusing conservation ef
forts on hotspots, geographical areas with high numbers of species (i.e., r
ichness), endemic species, rare or threatened species, and/or high levels o
f threat to species survival. The hotspot approach is an efficient and simp
le way to conserve species diversity, assuming that hotspots do not change
over space or time. We tested whether hotspots change across space and time
using a database of endangered and threatened species listed by the U.S. g
overnment from 1967 to 1999. We determined hotspots based on the cumulative
set of species listed for three overlapping and successively longer time p
eriods: 1967-1979, 1967-1989, and 1967-1999. We used minimum area complimen
tarity analysis, which selected the smallest set of areas (in our study, U.
S. counties) needed to represent a chosen set of species. Over time, the nu
mber of endangered and threatened species in the United States increased fr
om 76 in 1967 to 1123 in 1999. As the number of species increased over time
, hotspots changed in two ways: the number of hotspots increased and the ra
nk of hotspots shifted. Hotspots increased from 84 in 1979, to 166 in 1989,
to 217 in 1999. Only 63 of these counties were designated as hotspots in a
ll three periods. The remaining changes resulted from addition and deletion
of counties as hotspots over time. Some counties were removed from the lis
t or changed in relative work from one time period to the next regardless o
f their rank. Counties added as hotspots could rank anywhere on the list, a
nd they were not merely low-ranking counties added to represent one or a fe
w species. Therefore, hotspots serve as a useful tool for guiding conservat
ion efforts but, given their spatiotemporal variability, do not represent a
final solution.