Download many factors influence the evolution of herbivore

896
SPECIAL FEATURE-INSECT
Ecology, Vol. 69, No. 4
HOST RANGE
Ecology, 69(4), 1988, pp. 896-897
() 1988 by the Ecological Society of America
MANY FACTORS INFLUENCE THE EVOLUTION OF HERBIVORE
DIETS, BUT PLANT CHEMISTRY IS CENTRAL
JACK C. SCHULTZ
Pesticide Research Laboratory, Penn State University, University Park, Pennsylvania 16802 USA
There is little question that predators, as well as
many other factors, influence dietary evolution in phytophagous arthropods (Southwood 1973). Brower
(1958) outlined elegantly the role of generalist predators almost 30 yr ago. Although Bernays and Graham
cite several recent theoretical discussions, there really
has been little experimental work. As a consequence,
Bernays and Graham decry an "overemphasis" on host
plant chemistry.
It is important to remember that in 1964 Ehrlich
and Raven correctly identified an underemphasis on
plant chemistry. Recognition of this underemphasis
created a subdiscipline and allowed ecologists to appreciate interactions and phenomena that were missed
entirely before their chemical basis was clear.
Although it is true that the ensuing enthusiasm for
chemical ecology has exceeded that for other influences
on dietary evolution, I cannot agree that plant chemistry should receive reduced attention, or that it is less
important than we thought. Its importance remains
under-appreciated and demands increased study. In
this essay I outline why this is so, and suggest that this
view is fully consistent with giving natural enemies
more attention.
WHY PLANT CHEMISTRY NEEDS MORE,
NOT LESS, STUDY
I see at least three reasons for expanding studies of
plant chemistry's impact on herbivores.
1) Chemical evidence is presently inadequate. Bernays and Graham assert that plant chemistry is relatively "unimportant" on the basis of an "absence of
evidence," rather than "evidence of absence." Here
are two examples.
a) Diet breadth cannot be defined usefully in
terms of plant taxa. As Janzen (1979) pointed out,
herbivores are unlikely to select food on the basis of
Latin binomials. The relevant chemistry of related
plants is known for a pitifully small number of species
or families; when it is known, the suggestion that it
structures herbivore diets is generally supported (Fox
and Morrow 1981). We simply don't know enough
about other plants to conclude that their chemistry
is unimportant.
b) The potential selective impact of herbivory
on naturally occurring plants has been measured very
few times, all within the past 5 yr (e.g., Marquis
1984). Because the quantitative genetics of the plant
species must be known to infer selective impact, even
these studies are preliminary; nonetheless, each suggests that even small amounts of damage can have
significant fitness effects.
In each case we have an absence of evidence, not
evidence of absence. This cannot be remedied without
a better understanding of plant chemistry, its fitness
cost to plants, its effectiveness, its heritable basis, and
its interaction with other factors influencing herbivory.
2) Discrepancies between herbivore behavioral responses and physiological adaptation may not be real.
Cues need not be toxic themselves. However, the central (and old) question of what comes first in colonizing
new host plants, recognition or adaptation (Dethier
1947), remains unanswered. Frequent encounters with
a novel host could select for greater survivorship in
offspring, for discrimination and preference by females,
or both. These evolutionary alternatives cannot be distinguished without knowing (a) relevant host traits, (b)
sensory responses of the insect, (c) the insect's physiological capacity for surviving on the host, and (d)
heritable variation in items a-c. The difficulty in doing
this cannot be resolved by switching hypotheses (e.g.,
to predation), although other factors must be integrated
at some point.
3) I believe that the most important influence of
plant chemistry on insect herbivory lies in mediating
interactions between herbivores and other selective
factors (Schultz 1983). Here are two examples.
a) Oak tannins can reduce gypsy moth growth
and fecundity by 30% or perhaps more (Rossiter et
al. 1988), despite larval traits that reduce tannin activity (Schultz and Lechowicz 1986; J. C. Schultz,
personal observation). Host plant choice, which is
made by larvae, involves avoiding nonphenolic toxins (Barbosa and Krischik 1987), and high tannin
concentrations (J. C. Schultz, personal observation).
Although nutrient concentrations also influence gypsy moth growth and fecundity (J. C. Schultz and M.
C. Rossiter, personal observation), a significant part
This content downloaded from 128.206.166.171 on Thu, 07 Jan 2016 18:23:48 UTC
All use subject to JSTOR Terms and Conditions
August 1988
SPECIAL FEATURE-INSECT
of the gypsy moth's definition of "leaf quality" involves tannins and other phenolics.
At the same time, parasitoids and pathogens can
be the most important sources of mortality (USDA
1981). The effectiveness of several parasitoid species
depends on host larval growth rates. Parasitism increases when host growth is slowed (Weseloh et al.
1983) to the degree that we observe as a function of
decreasing leaf quality (M. C. Rossiter et al. 1988
and personal observation). Plant phenolics can have
both direct and indirect negative impacts on larvae.
Because leaf tannins also dramatically reduce larval susceptibility to nuclear polyhedrosis virus
(Keating and Yendol 1987, Keating et al. 1988), the
effect on the gypsy moth of "decreasing" host plant
quality could also be very positive. In this example,
herbivore feeding preferences and diet breadth are
correlated with plant chemistry. There are also important natural enemies presumably acting as selective agents in the system, but their impact may be
mediated by plant chemistry.
b) The grasshopper, Astroma riojanum (Proscopiidae) feeds on creosote bush (Larrea cuneifolia:
Zygophyllaceae) in Argentine deserts. It is monophagous and highly cryptic on its host. Female antipredator traits change from crypsis to repellency
above a size threshold, apparently owing to the presence of deterrent plant materials in the larger insects'
guts (Schultz 1981). Moreover, the rate at which females grow to the "refuge" size is determined by the
highly variable concentration of resins in host plant
tissues consumed (Rhoades 1977). In this case, the
interaction between insect and predator is mediated
by the appearance of the host plant, its chemistry,
and choices made by feeding insects.
How TO PROCEED
I agree that the study of nonchemical influences on
herbivore diet evolution has suffered while attention
focused on plant chemistry. However, we cannot afford
to take an either/or, dialectical approach to the problem. Although this approach can energize research, I
believe that it is more likely to retard our understand
of complex interactions.
Ecologists must accept chemistry as a part of natural
history and integrate new kinds of information creatively. Ecology should be a synthetic science, based
on complete understanding of the parts but with a full
appreciation of their interactions. Arguments that plant
chemistry is not "the predominant" influence on the
HOST RANGE
897
evolution of herbivore diets are not compelling, interesting, or useful. Let's not polarize the issues; let's see
how these systems work.
LITERATURE CITED
Barbosa, P., and V. A. Krischik. 1987. Influence of alkaloids
on feeding preference of eastern deciduous forest trees by
the gypsy moth, Lymantria dispar. American Naturalist
130:53-69.
Brower, L. P. 1958. Bird predation and food plant specificity
in closely related procryptic insects. American Naturalist
92:183-187.
Dethier, V. G. 1947. Chemical insect attractants and repellents. McGraw-Hill, New York, New York, USA.
Ehrlich, P. R., and P. H. Raven. 1964. Butterflies and plants:
a study in coevolution. Evolution 18:586-608.
Fox, L. R., and P. A. Morrow. 1981. Specialization: species
property or local phenomenon? Science 211:887-893.
Janzen, D. H. 1979. New horizons in the biology of plant
defenses. Pages 331-350 in G. H. Rosenthal and D. H.
Janzen, editors. Herbivores: their interactions with secondary plant metabolites. Academic Press, New York, New
York, USA.
Keating, S. T., and W. G. Yendol. 1987. Influence of selected
host plants on gypsy moth (Lepidoptera: Lymantriidae) larval mortality caused by a baculovirus. Environmental Entomology 16:459-462.
Keating, S. T., W. G. Yendol, and J. C. Schultz. 1988.
Relationship between susceptibility of gypsy moth larvae
(Lepidoptera: Lymantriidae) to a baculovirus and host plant
foliage constituents. Environmental Entomology, in press.
Marquis, R. J. 1984. Leaf herbivores decrease fitness of a
tropical plant. Science 226:537-539.
Rhoades, D. F. 1977. Integrated antiherbivore, antidesiccant and ultraviolet screening properties of creosote bush
resin. Biochemical Systematics and Ecology 5:281-290.
Rossiter, M. C., J. C. Schultz, and I. T. Baldwin. 1988. Relationships among defoliation, red oak phenolics, and gypsy
moth growth and reproduction. Ecology 69:267-277.
Schultz, J. C. 1981. Adaptive changes in the antipredator
behavior of a grasshopper during development. Evolution
35:175-179.
1983. Impact of variable plant chemical defenses
on insect susceptibility to parasites, predators, and diseases.
Symposia of the American Chemical Society 208:37-55.
Schultz, J. C., and M. J. Lechowicz. 1986. Host plant, larval
age and feeding behavior influence midgut pH in the gypsy
moth (Lymantria dispar L.). Oecologia (Berlin) 71:133137.
Southwood, T. R. E. 1973. The insect/plant relationship:
an evolutionary perspective. Symposia of the Royal Entomological Society of London 6:3-30.
USDA. 1981. The gypsy moth: toward integrated pest management. C. C. Doane and M. L. McManus, editors. United
States Forest Service Technical Bulletin 1584.
Weseloh, R. M., T. G. Andreadis, R. E. B. Moore, J. F. Anderrson, N. R. Dubois, and F. B. Lewis. 1983. Field confirmation of a mechanism causing synergism between Bacillus thuringiensis and the gypsy moth parasitoid, Apanteles
melanoscelus. Journal of Invertebrate Pathology 41:99-103.
This content downloaded from 128.206.166.171 on Thu, 07 Jan 2016 18:23:48 UTC
All use subject to JSTOR Terms and Conditions