5.2.1 Theories of how bark beetles find suitable host trees
There are two theories on how bark beetles find suitable host trees
(McMullen and Atkins, 1962). The first is that they locate such
trees by orienting over several meters to volatile chemicals
usually released by damaged or diseased trees (called "primary
attraction"). The second theory is that beetles fly about and
encounter suitable host trees at random, whereupon they land and
test them by short-range olfaction or gustation. The two theories
are not mutually exclusive, and one or the other may primarily
operate in a particular species. In California, host finding by the
important pests D. brevicomis and I. paraconfusus is thought to be
a random process. Ponderosa pines that were killed by freezing with
dry ice and screened to prohibit bark beetle attack, did not have
higher landing rates for the prevalent D. brevicomis and I.
paraconfusus bark beetles (among other species) than did living
trees. Landing rates on diseased and healthy trees also were
similar; it was estimated that each tree in the forest was visited
by about one D. brevicomis beetle each day (Moeck et al., 1981;
D.L. Wood, 1982). Logs of freshly cut ponderosa pine placed in
sticky screen traps did not catch beetles of these species, while
at the same time high numbers were attracted to synthetic pheromone
or infested logs (Moeck et al., 1981).
In addition to I. paraconfusus and D. brevicomis, many species
probably visit trees at random whereupon they are "tested for
resistance" by beetles during an attack. For example, Scolytus
quadrispinosus was caught equally on traps placed in host shagbark
hickory, Carya ovata, and nonhost white oak, Quercus alba (Goeden
and Norris, 1965). Berryman and Ashraf (1970) found attacks by
Scolytus ventralis in the basal section of 74% of grand fir
examined, while only 3.5% of these trees were colonized. Most
unsuccessful attacks were abandoned before beginning the gallery.
The attacks on grand fir appeared random during the early part of
the flight period before aggregations resulted. Hynum and Berryman
(1980) caught D. ponderosae in traps on 96% of the lodgepole pines
(P. contorta) sampled, but only 66% of these pines were killed.
Also, they found no differences in landing rates between killed and
surviving lodgepole pines or between host and nonhost trees. A
direct relationship between the number of D. ponderosae caught on
unattacked trees and the number of trees upon which beetles landed
was found in a study of 123 lodgepole pines (Raffa and Berryman,
1979). I. grandicollis landed equally on sticky traps on trees
judged resistant or susceptible based on crown area (Witanachchi
and Morgan, 1981). However, Schroeder (1987) found an average of 35
T. piniperda landing on lower vigor Scots pine, P. sylvestris (as
judged by less crown area), than on higher vigor trees (mean of
22). These differences could be due to secondary release of
monoterpenes by beetles boring in the low vigor trees that were
less able to resist attack.
There is some evidence that I. typographus is weakly attracted
to host volatiles (Austarå et al., 1986; Lindelöw et al., 1992) or
monoterpenes such as alpha-pinene (Rudinsky et al., 1971), but other
studies have not observed any attraction to host volatiles or
synergism of pheromone and host volatiles (Schlyter et al., 1987a).
A computer model by Gries et al. (1989), in which "beetles" must
take a series of flights between trees in a grid (each flight to
one of eight neighboring trees) and test each tree for suitability,
showed that few beetles would find the widely scattered hosts
designated as susceptible. Thus, they concluded that a mechanism of
long-range primary attraction would be required for maintenance of
the population. However, a more recent computer model, in which
beetles "fly" more naturally among randomly dispersed susceptible
trees can be used to show that a significant proportion of the
population will find the susceptible trees (of actual trunk
diameter) by chance interception (Byers, 1993a). If beetles then
test the defenses of the potential host (although this rarely has
been observed, see 5.4) then weaker, more susceptible, trees will
not exude adequate resin and allow the beetle to produce
aggregation pheromone. According to the later model, this will in
effect greatly increase the effective "radius" of the tree so that
the remainder of the population can quickly find and colonize these
trees (Byers, 1993a).
In addition to the random and host volatile theories, some
bark beetles may find weakened and susceptible host trees by
orienting to volatiles produced by competing species during
colonization. The volatiles can be host compounds that virtually
any bark beetle would release upon attack (e.g., monoterpenes) or
pheromone components of these other species. For example, D.
brevicomis responds to pheromone components of I. paraconfusus
(Byers and Wood, 1981a); I. typographus responds to exo-brevicomin
(from D. micans and Dryocoetes spp., Borden et al., 1987) when
combined with its pheromone components (Tommerås et al., 1984); and
several sympatric species of Ips in the southeastern United States
are cross-attracted to infested pine logs (Birch et al., 1980b).
Byers, J.A. 1995. Host tree chemistry affecting colonization in bark
beetles, in R.T. Cardé and W.J. Bell (eds.). Chemical Ecology of
Insects 2. Chapman and Hall, New York, pp. 154-213.