The survival time of a white spruce tree (Picea glauca (Moench) Voss) with
respect to the attack by the white pine weevil (Pissodes strobi (Peck)) was
defined as the number of years until an attack occurred. In this context,
resistant trees are those capable of delaying attacks rather than preventin
g attacks. This study investigated the patterns of survival times of resist
ant and susceptible trees in a progeny trial of white spruce in the interio
r of British Columbia, Canada. By doing so, we demonstrated that survival t
ime analysis, a technique specifically developed for dealing with clinical
medicine and industrial reliability tests, was a very useful approach in an
alyzing resistance of hosts to herbivore attack. The results suggested that
survivorship of resistant trees was significantly different from that of s
usceptible trees, as expected. The median survival time of susceptible tree
s was similar to 12 yr after seeding, while it was 21 yr for resistant tree
s. The survival rimes of white spruce trees were identified adequately foll
owing a lognormal distribution. Based on the estimated hazard functions of
the lognormal distribution, the most severe weevil infestation was predicte
d to occur at 19 yr for susceptible trees, and at 26 yr for resistant trees
. infestation level was affected by several factors, such as tree height, s
easonal temperature, and precipitation. An accelerated failure rime model w
as used to evaluate the effect of these factors on survivorship. The identi
fication of resistant or susceptible trees by survival times is to be incor
porated into an integrated pest management program for the white pine weevi
l in British Columbia.