Quantitative Defense Theory and Patterns of Feeding by Oak Insects Author(S): Stanley H

Home , Chemical defense, Condensed tannin, Hydrolysable tannin, Stilbosis

International Association for Ecology

Quantitative Defense Theory and Patterns of Feeding by Oak Insects Author(s): Stanley H. Faeth Source: Oecologia, Vol. 68, No. 1 (1985), pp. 34-40 Published by: Springer in cooperation with International Association for Ecology Stable URL: http://www.jstor.org/stable/4217793 Accessed: 07/05/2010 21:20

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=springer.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Springer and International Association for Ecology are collaborating with JSTOR to digitize, preserve and extend access to Oecologia.

http://www.jstor.org Oecologia(Berlin) (1985) 68:34-40 Oecologia © Springer-Verlag1985

Quantitative defense theory and patterns of feeding by oak insects

Stanley H. Faeth Department of Zoology, Arizona State University, Tempe, AZ 85287, USA

Summary. Patterns of herbivory over a two year period These theories were ascendant for almost a decade: on Quercusemoryi (Fagaceae) were correlatedwith seasonal "predictable", "apparent" plants should respond evolu- and yearly changes in tannin and protein content. Quantita- tionarily to herbivore pressureswith quantitative defenses. tive defense theory predicts that tannin and protein content Tannins were considered as the ideal quantitative defense. in apparent plants should be negatively and positively cor- Most of the evidence rested upon the fact that tannins are related, respectively,with degree of herbivory. Most herbi- ubiquitous and abundant among apparent, woody plants. vory occurred early in the growing season, but the pattern Experimentalevidence for the defensive nature and mode varied between the two years. Tannin and protein content of action of tannis was exiguous or nonexistent (Zucker sometimes varied negatively and sometimes positively with 1983). Recently, the idea that tannis are generalized de- degree of herbivory; they did not consistently covary with fenses and are always detrimental to herbivorous insects herbivory. Protein content was positively correlated with has been challenged (Bernays 1978, 1981; Bernays et al. herbivory in 1981-1982 but not in 1982-1983. Condensed 1980; Bernaysand Woodhead 1982; Fox 1981; Berenbaum tannin content was negatively correlated with herbivory in 1983; Coley 1983; Zucker 1983). 1981-1982 but not in 1982-1983. Hydrolysable tannin con- Here, patternsof herbivoryand seasonal changes in tan- tent was positively correlatedwith herbivoryactivity. Multi- nins and protein content of Quercusemoryi are examined ple regression analyses indicated these phytochemical vari- relative to the theory of quantitativedefense in "apparent" ables explained either no significant variation in herbivory plants. (1982-1983) or did so in a fashion opposite (1981-1982) to the predictions of the theory of quantitative defense. Material and methods Feeding by oak insects was not solely a function of seasonal changes in quantitativedefenses and nutrients.Ob- Sampling viously, population dynamics of the insects are sensitive to factors other than of the trees and I Six trees (- 3 m in height) of Quercus emoryi (Fagaceae) phytochemistry selected from 50 discuss other factors that can influence of herbi- were randomly approximately similarly- patterns sized trees that were located within 150 m radius of each vory. other in a riparian zone surrounding Parker Creek in the Sierra Ancha ExperimentalForest (USDA), Gila, Co. AZ. Leaves of Q. emoryi persist for one year from budbreak in mid-April to early May. The six trees were monitored for number of leaves that most insects feed study Feeny (1970) proposed lepidopterous insects, and seasonal in hydrolysable on robur in the because leaf damaged by changes Quercus early spring quality and condensed tannin, and water content. Samples declines as the season He concluded protein, growing progresses. of 100-200 leaves were randomly collected from each tree that both number of and abundances decline in a species monthly from April 1981 (budbreak) to November 1981, season because tannin content and leaf increase, toughness and from April 1982 to November 1982. Leaves were also while content decreases as leaves mature (Feeny protein sampled in early April 1981 (before budbreak)and in Feb- and Bostock 1968; 1970). Based 1968; Feeny Feeny primar- ruary and April 1983 (before budbreak). Leaf samples thus ily on this work, Feeny (1975, 1976) and others (Rhoades the entire seasons of 1981-1982 and that linked represent growing and Cates 1976) postulated a theory general 1982-1983. types of chemical defense with "apparency" of plants. plants, such as trees, which are likely to be dis- Long-lived Estimates of damaged leaves covered by herbivores, evolved "quantitative" defenses, while short-lived, transitory plants evolved "qualitative" Sampled leaves were transported to the laboratory where defenses. Quantitative defenses act in a stoichiometric fash- they were segregated into insect-damagedand undamaged ion. For example, tannins supposedly bind plant proteins categories. Leaves were considered damaged if > 10% of and insect proteolytic enzymes and thus increasingamounts leaf area was consumed (determined by a Licor leaf area render protein less available to insect herbivores. meter). This level of damage was chosen because damage 35 ,t** less than this amount was not always discernibleas resulting from insects (e.g., small necrotic areas were caused by fungi, 4 ~~~~~~~~~~~4, bacteria, and Mean cumulative physical damage). percent 4 l4 l 4 of leaves damaged for all trees was calculatedfor 1981-1982 UJ and 1982-1983. 40- I ~ ~1 9 Using percent of leaves damaged as an indicator of in- a * 4 sect herbivory assumes that once leaves were no l iil4'I - A i1 ; damaged, 11 20- further occurred on these leaves. In other 4 _ herbivory words, o 30- herbivory late in the season occurred on leaves previously a 10-o undamaged by earlier herbivores. To test this assumption, -j 0 I regressed average mass (leaf mass and area are highly I I I I I I I I I I I I I I I I I I I I I I I I u) correlatedin Q. emoryi, r2=0.965, P<0.001, Bultman and MAY SEPT JAN MAY SEPT JAN MAY Faeth, unpublished data) of damaged leaves for each tree 1981 1982 1983 in each season with date of leaf sample. None of the 12 Fig. 1. Seasonal changes in cumulative percentage of leaves dam- regressions(6 trees x 2 seasons) had slopes that were signifi- aged by insects in 1981-1982 and 1982-1983. Each point represents cantly different from zero (P>0.30). Since average damage the mean cumulative percent of leaves damaged for 6 study trees, did not increase with time, this suggests that insect herbi- with bars representingstandard errors of means at each date vores did not repeatedly consume leaves that were already damaged by earlier insects. Protein and hydrolysableand condensed tannin content Phytochemicalanalyses were regressed separatelywith cumulative increases in her- 1981-82 and 1982-83 to test the Insect-damagedand intact leaves were analyzed separately bivory during predictions of and Rhoades and Cates that for each tree at each sample date for protein and hydrolys- Feeny (1975, 1976) (1976) able and condensed tannin content. Condensed tannins herbivory should be positively correlated with protein content and correlated with tannin content. analyses included segregation into monomeric fractions of negatively Sample dates when no occurred on trees were not used phenols, polymeric fractions (condensed tannins), and total herbivory in since the focus of this is rela- vanillin positive fractions (monomers plus polymers). Only regression analyses, study of of and More- analyses of condensed tannins are presentedhere; total van- tionship timing herbivory phytochemistry. over, inclusion of zero does not illin positive shows similar trends because monomeric frac- herbivory points qualita- tions were low and varied little. leaves tively alter results. Finally, multiple regression (separately Randomly sampled for each was used to determine the relative amount (same leaves used in Materials and methods: year) Sampling) of variance in of the inde- from each tree were immediately frozen in dry ice in the patterns herbivory explained by variables of and tannin content. All as- field, later lyophilized in the laboratory, and then ground pendent protein into a fine Ground were stored in sumptions of normality and homoscedasticity of variances powder. samples jars were tested. at 5° C with a dessicant until analyses could be performed. Each analysis for each tree from each date was done in triplicate. Details of protein and tannin analyses are found Results in Faeth (1986). Brief summariesare provided here. Protein content of ground, lyophilized leaf samples was Seasonalpatterns of herbivory determined the BioRad Method by (Richmond, CA). Most occurred in the season Monomeric condensed tannins, and total vanillin herbivory early growing phenols, and the of varied between positive were determined by a modified acidified vanillin (Fig. 1), patterns herbivory 1981-1982 and 1982-1983. In the 1981-1982 sea- method (Broadhurst and Jones 1978). Monomeric phenols growing son, all leaf occurred from late and total vanillin positive were first determined, and then virtually chewing quickly, to late In 1982-1983, was condensed tannins were determined by subtracting April May (Fig. 1). herbivory monomers from total vanillin for each trial. prolonged, occurring from late April to October (Fig. 1); positive Hydro- the two of were in lysable tannins were determinedby a modified iodate tech- major episodes herbivory spring (late and fall to late nique (Bate-Smith 1977) and expressed as percent mass April-lateMay) (early September November). dry In both the herbivores were tannic acid equivalents as compared to a tannic acid stan- years, primary Lepidoptera (Arctiidae, Noctuidae, Geome- dard curve prepared for each run. For protein and tannin Lasiocampidae, Phycitidae, tridae, analyses, all extraction solutions, reagent concentrations, Stenomidae, Tortricidae), Coleoptera (Chrysomeli- dae and and Tetti- temperatures,and time of reactions were empiricallydeter- Cucurlionidae), Orthoptera (Acrididae mined to standardizeand goniidae), and Hymenoptera (Tenthredinidae).Sucking in- optimize reactivity. sects Water content of leaves was measuredin 1981-1982 and were not included in this study because the original of defense was from 1982-1983 for newly-emergedleaves, 1 month, and 1 year concept quantitative theory developed old leaves. seasonal patterns of leaf chewers (Feeny 1970, 1975). How- ever, like leaf chewers, sucking insects are also most numer- ous early in the growing season. Herbivory by vertebrates Statistical analyses did not occur on the study trees. Repeated measures ANOVA (Winer 1971) was used to test for differences in and and condensed protein hydrolysable Seasonal and in tannin content between intact and herbivore-damaged yearly changes phytochemistry leaves and for effects of seasonality on phytochemistry for In both years, protein content of leaves on the 6 study study trees in 1981-82 and 1982-83. trees varied significantly with time during the growing sea- 36

8' 8-

6- 6

a- MY SEPT\I JA MA 4. 4 2. Seasonal in relative z X Fig. changes percent -p protein content of undamaged and insect __.s V damaged leaves in 1981-1982 and 1982- 2 2 1983. Points representmeans of 6 trees. Standarderrors for each point are in v . . . . . I . . v . . . . '. '. . ' '. Appendix I MAY' 'SEPT JAN MAY MAY SEPT' JAN MAY 1981 1982 1982 1983

0 - 0 UNDAMAGED 3 30- l- *---- - INSECT DAMAGED 30-

I 20- 20- I Fig. 3. Seasonal changes in percent hydrolysabletannin content of undamaged Wg_: < =------10 10- e-, ." _ .» * . - - - -_ - _ _. _ _ and insect damaged leaves in 1981-1982 and 1982-1983. Points representmeans of 6 ° trees. Standard errors for each point are in 0 . . . . . Appendix I MAY 'SEPT JAN ' MAY MAY SEPT JAN 'MAY 1981 1982 1982 1983

----- UNDAMAGED _ 6- *.----- INSECT DAMAGED 6- 3-

en 4- 4-

z -- . . -_ * ---- _ -- 4. Seasonal in condensed ,)sl Fig. changes percent I ------w 2- O* 2- tannin content of undamaged and insect ,/~~~~~~~~~ damaged in 1981-1982 and 1982-1983. Points 0 - ,Y 0 representmeans of 6 trees. Standard errors for

at0 ! I i I I I II I I I 0 T i i I I I i i i i T I I each point are in Appendix I MAY' ' SEPT JAN' MAY MAY SEPT' ' 'JAN MAY 1981 1982 1982 1983 son (Fig. 2; 1981-1982, F=22.89, df=6, 30, P<0.001; crease in hydrolysable tannin on leaves consumed by herbi- 1982-1983, F=13.43, df=8, 40, P<0.001). In 1981-1982, vores (Fig. 3). This trend was evident in both years (Fig. 3) protein content was initially high at budbreak, declined rap- but only significant in 1982-83 (1981-1982, F= 1.01, df= 1, idly, and then increased gradually over the growing season. 5, P>0.30; 1982-1983, F=25.87, df=l, 5, P<0.01). De- However, in 1982-83, protein levels were more variable cline in hydrolysabletannin content may be part and parcel through the season (Fig. 2). In both years, protein content of either increases in wound-repair compounds (Yang and of leaves damaged by herbivores was significantly lower Pratt 1978) or inducible defenses in damaged leaves, since than that of intact leaves (Fig. 2, 1981-1982, F= 17.03, df= hydrolysable tannin content did not vary with time once 1, 5, P<0.01; 1982-1983, F=25.15, df=1, 5, P<0.01); leaves were expanded. herbivory by insects was associated with a decline in nutri- Condensed tannin content varied significantlywith time tional content of leaves. in each growing season (Fig. 4; 1981-1982, F= 3.36, df=6, Hydrolysable tannin content of leaves was initially high 30, P<0.05; 1982-1983, F=14.84, df=8, 40, P<0.001). in both years at the onset of budbreak, comprising> 18% Condensed tannin content of insect-damaged leaves was of dry weight of leaves (Fig. 3). Hydrolysable tannin con- greater than in intact leaves in both years (Fig. 4; tent then declined rapidly (within 30 days) and remained 1981-1982, F=6.67, df=l, 5, P<0.05; 1982-1983, F= at relatively constant levels for the remainder of the growing 16.35, df= 1, 5, P<0.01). The differencesin condensed tan- season. Hydrolysable tannin content varied significantly nin content were maintained between insect-damaged and with time (1981-1982, F=3.48, df=6, 30, P<0.01; intact leaves throughout each growing season, despite the 1982-1983, F=9.51, df=8, 40, P<0.001) only when the fact little herbivory occurredafter the first half of the grow- first sample date was included. If this sample was deleted, ing season (Fig. 1). the effect of time became nonsignificant (P>0.05). This Water content of undamaged leaves in 1981-1982 and early decline may simply be due to a dilution effect; hydro- 1982-1983, respectively, was as follows: budbreak 67%, lysable tannins were probably synthesized before or at bud- 68%, 1 month after budbreak 54%, 55%; 6 months after break, and concentrations declined only because other leaf budbreak45%, 46%, and just before abscission 44%, 45%. constituents (e.g. lignin, cellulose) were added during leaf Water content of insect-damaged leaves just before abscis- expansion. sion was similar to undamaged leaves: 41.6% and 42.7% This phenomenon may also explain the apparent de- in the growing seasons, respectively. 37

100 - 90- Fig. 5. Relationship of percent of leaves damaged v 80- *0 * by insects in sample periods with relative percent 2 70- 60- protein content of leaves in 1981-1982 and 1982- 03 > 50- 1983. Each point representspercent of leaves U 40- damaged for an individual tree and corresponding - 30- protein content at a sample date. See text for S, 20- * * @g - estimates, significance,correlation coefficients, and 10- . *. V-*a .,* R2 of regressions 0o % PROTEIN (DRY WT 1981 a0 s O 2 4 6 8 10 % PROTEIN (DRY WT.) 1981 % PROTEIN (DRY WT) 1982

100- Q 90- Fig. 6. Relationship of percent of leaves damaged C 80- in : 70- by insects sample periods with hydrolysable C 60- tannin content of leaves in 1981-1982 and 1982- : 50- 1983. Each point representspercent of leaves w 40- damaged for an individual tree and corresponding o- 30- * hydrolysabletannin content at a sample date. See aS 20- · 10- 6 * text for estimates, significance,correlation 0o I · . I I I and R2 of l; Ir coefficients, regressions I% HYDROLYZABLE TANNINSl (DRY1 W rb 981ne 30w 0 5 10 15 20 25 30 % HYDROLYZABLETANNINS (DRY WT.) 1981 % HYDROLYZABLETANNINS (DRY WT.) 1982

100 I 90- - Fig. 7. Relationship of percent of leaves damaged w 80- insects in with condensed tannin E 70- by sample periods a 60. content of leaves in 1981-1982 and 1982-1983. u3 50. Each point representspercent of leaves damaged * . 40- *. for an individual tree and correspondingcondensed 30. * . tannin content at a sample date. See text for 20- S estimates, significance,correlation coefficients, and 10- * * 0. 6 * *·6¶ . 0 _ 0. 1 R2 of regressions u% C D T S (Y W 1 2 3 4

% CONDENSED TANNINS (DRY WT.) 1981 % CONDENSED TANNINS (DRY WT.) I982

Relationship of leaf phytochemistry with herbivory tiple regression because it is evident from the individual regressions that some independent variables are likely to Protein content. Amount herbivory (cumulative percent of be collinear. damaged leaves) by leaf chewers was significantlyand posi- Multiple regression for herbivory in 1981-1982 showed tively correlated(P< 0.001) with seasonal mean protein lev- hydrolysable tannin content was entered on the first step, els in leaves in 1981-82 (Fig. 5, regression equation: y= and this variable explained 88.2% of the variation in herbi- 17.67X-26.75, R2=0.632). However, in 1982-1983, vory (F= 104.46 d.f. = 1, 14, P < 0.01). However, this rela- amount of herbivory was not significantly correlated (P> tionship was positive. Other phytochemical variables en- 0.20) with protein content (Fig. 5). tered after hydrolysabletannin content did not significantly increase amount of variation Examination of re- Hydrolysabletannin content. Amount of herbivory was sig- explained. residual and residual nificantly and positively correlated (P< 0.001) with sea- siduals, plots, statistics showed no violation of used in sonal hydrolysable tannin content in 1981-82 (Fig. 6, re- assumptions multiple regression analyses. gression equation: y = 3.89X-27.84, R2 = 0.882). How- for 1982-83 indicated con- ever, in 1982-1983, this relationship of herbivory with hy- Multiple regressionanalyses densed tannin content entered the drolysable tannin content (Fig. 6) was not significant (P> regressionequation first, but did not 0.20). significantly account for variation in herbivory (F=2.35, d.f.=1, 23, P>0.10). Other phytochemical vari- Condensedtannin content. Amount of herbivorywas signifi- ables entering the equation after condensed tannin content cantly and negatively correlated (P<0.05) with condensed did not significantlyincrease amount of variation explained. tannin content in 1981-1982 (Fig. 7, regression equation: Examination of residuals revealed no violation of assump- y= -24.95 X+ 57.67, R2 = 0.213). However, because of the tions. large amount of variation about the regression line, condensed tannin content little of the varia- explained relatively Discussion tion (21.3%) in number of leaves damaged. In 1982-1983, this was no relationship(Fig. 7) longer significant(P> 0.05) Herbivory patterns were variable from year to year. About but was still negative (r, correlation coefficient= -0.27). 30% of available leaves were damaged by herbivores, with differentseasonal patternsbetween the two growing seasons Multiple regression analyses (Fig. 1). Composition of the insect fauna did not noticeably Stepwise multiple regression analyses were performed for vary between years. However, little leaf area was removed each year, with amount of herbivory during each sample (<5% of total area). Edwards and Wratten (1983) found period as the dependent variable, and protein, hydrolysable similarpatterns in several plant species, i.e., few leaves com- and condensed tannin as independentvariables. I used mul- pletely consumed but a large proportion of leaves with some 38 damage. Virtually all herbivory in 1981-82 occurs within hydrolysable tannin content or with increased condensed 30 days of budbreak, while in 1982-83, the majority of tannin content. However, other experiments (Faeth 1986) herbivory occurs between budbreak in late April October. have clearly shown that these same chemical changes occur In fact, the largest increase occurred between late August after insect or mechanicaldamage to leaves. and mid-October. Induced response to herbivory has been documented Seasonal trends in phytochemistry do not conform to in other plants (Haukioja and Niemela 1979; Ryan 1979; Feeny's (1970) results that protein content increases and Rhoades 1979; Carroll and Hoffman 1980; Bryant 1981; tannin content declines over the growing season. Feeny and Schultz and Baldwin 1982; Baldwin and Schultz 1983). Bostock (1968) reported that hydrolysable tannin content Schultz and Baldwin (1983) reportedhigher levels of hydro- remains fairly constant throughout the growing season. lysable and condensed tannins in leaves of red oak trees Hydrolysable tannin content in Q. emoryi was highest at that were previouslydefoliated or currentlybeing defoliated budbreak (Fig. 3), and then rapidly declines to remain at by gypsy moths. This response was short-lived, peaking constant levels for the remainder of the growing season. about 20 days after defoliation. Contrary to these results, Some tropical tree species also have higher concentrations hydrolysable tannins did not increase in damaged leaves, of phenolic compounds in new leaves than old leaves (Mil- although sampling periods may have been too infrequent ton 1979; Oates et al. 1980). The decline in Q. emoryi was to detect short term increases.Also, induction in condensed likely a dilution effect resulting from addition of other leaf tannin content in damaged leaves persisted throughout the constituents as the leaf constituents as the leaf expands; growing season in both years (Fig. 4). This persistence of nevertheless, insects feeding just after budbreak must still elevated condensed tannins in damaged leaves occurredde- contend with a relative high hydrolysable tannin content spite the fact that the same leaves were not repeatedlydam- per unit of leaf ingested. aged throughout the year. Condensed tannins do not increase in a monotonic fashion demonstrated by Feeny (1970) for Q. robur and for Herbivoryand chemicaldefense other perennials (Lawton 1976; Dement and Mooney 1974). Condensed tannins increased only in the first 2 or Regression analyses of amount of herbivory with phyto- 3 months, then leveled off for the remainderof the season chemical changes in Q. emoryi do not generally support in both years. the predictions of the theory of quantitative defense in ap- Feeny (1970) showed protein levels decline monotoni- parent plants (Feeny 1975, 1976; Rhoades and Cates 1975). cally through the growing season, and felt that protein lev- While protein content was positively correlated with herbi- els are at least partly responsible for early feeding by oak vory in 1981-1982 as predicted by the theory, this relation- insects. Protein levels of Q. emoryi were high at budbreak ship was not significant in 1982-83. Similarly, the theory in 1981-1982, declined rapidly and then increased mono- of quantitative defense predicts a negative correlation in tonically for the rest of growing season until protein content timing of herbivory with tannin levels. This prediction is at the latter part of the season approached levels at bud- fulfilled in 1981-1982 for condensed tannin content, but break (Fig. 2). In 1982-1983, changes in protein content not in 1982-1983. Moreover, while the regression of con- were much more variable, with highest levels found at mid densed tannin and amount of herbivory was significantly and late season (Fig. 2). Discrepancies in seasonal trends negative in 1981-1982, it explained only a small amount in protein between Q. robus and Q. emoryi may partially of variability (21%). Finally, hydrolysable tannin content result from different assay methods. Kjeldahl (Feeny 1970) was significantly and positively correlated with herbivory is an indirect measurementof protein, since protein is esti- in 1981-1982, a result directly contrary to the predictions mated from organic nitrogen content which includes non- of quantitative defense theory. proteinaceous organic nitrogen (Bradstreet 1965; Feeny In general, overall changes in leaf chemistry were poor 1970). BioRad (this study) is a direct measure of number predictors of seasonal patterns in herbivory. In 1981-1982, of peptide bonds. When BioRad is compared to Kjeldahl hydrolysabletannin content explained most of the variation for Emory oak, there are large discrepanciesin relative pro- (88%) in herbivory, but this was a positive correlation and tein, especially early in the growing season (Faeth unpub- contradicts predictions of the theory. Other phytochemical lished data). Clearly, however, protein content, which is variablesdid not significantlyincrease the amount of varia- considered a key nutrient for insects (McNeill and South- tion explained. In 1982-83, no single chemical variable or wood 1978; Mattson 1980; Scriber and Slansky 1981), did combination of variables explained significant variation in not decrease monotonically through the season, and was herbivory intensity. not consistently related to seasonal herbivory patterns as The generality of quantitative defensive theory must be predicted by the theory of quantitative defense. seriously questioned. Traditionally-held quantitative defenses such as tannins are not consistently good indicators of deterrence to herbivores: in fact, tannin content Inducible chemical in leaves may changes herbivore-damaged vary with herbivoryin a fashion opposite to their purported Condensed tannin content was markedly and significantly role as defensive chemicals. Karban and Ricklefs (1984) greater in damaged leaves than intact leaves in both years found that lepidopterandiversities and abundances are not (Fig. 4). Protein and hydrolysable tannin content were gen- correlated with either tannin or nutrient content of various erally lower in damaged than in intact leaves, but it is not tree species. Coley (1983) also did not find much support clear if these decreases were simply caused by a relative for the theory of quantitative defense theory in examining increase in condensed tannin. The increase in condensed the relationship of herbivory to phenolic and tannin levels tannins suggests they are induced by Q. emoryi in response in tropical trees. Coley (1983) did find that leaf toughness to herbivory. It could be argued that insects on oaks are and fiber content were negatively correlated, while water simply selectively feeding on leaves with lower protein or content was positively related with herbivory. Leaf tough- 39 ness was low and water content of Q. emoryi was high the spring to avoid leaves previously damaged by other (67%) in 1981-1982 when most herbivory occurred (Fig. 1). herbivores. Since herbivoresseem to be generally unrespon- But in 1982-1983, a large fraction of herbivory occurred sive to phytochemical changes that occur seasonally, it is in the fall when water content had declined to near lowest doubtful that damaged leaves are avoided because of direct levels and leaf toughness should have been high. toxic effects of increased tannins. However, predators Tannins can be harmful or deterrent to individual insect (Heinrich and Collins 1983) and parasitoids (Vinson 1975; species on specific plants. Zucker (1983) argued various Faeth 1985, 1986) use physical or chemical changes in dam- tannins are likely very specific in their mode of action aged leaves to locate insect hosts. Risks from intensified against herbivores and pathogens, based on their stereo- attack by predators and parasites associated with damaged chemical heterogeneity. Bernays (1981) argued against the leaves may have selected for early feeding by insects. idea of tannins as quantitative, generalized defenses based Certainly, other factors may also play a role in patterns on conflicting effects of tannins on insects in bioassay stu- of insect herbivoryon long-lived plants (e.g., leaf toughness, dies. The original notions of tannins as quantitative de- water content, or variable predator and parasite attack). fenses in apparent plants is also suspect based solely on I am not claiming that seasonal changes in phytochemistry seasonal and yearly patterns of tannins and herbivory, upon and nutrientsin long-lived plants have no effect on seasonal which the theory was originally based. Phytochemical chan- patterns of herbivory, only that these changes do not, in ges alone are inadequate to explain herbivory patterns, themselves, explain herbivory patterns. Responses of each especially if variability in herbivory and chemistry from insect species to specific chemical and phenological changes year to year is considered. in "apparent" plants may account for some variation in herbivory. However, sufficient evidence has accumulated such that defenses alone cannot be considered Alternativehypotheses quantitative the only explanation for these patterns. Despite some year-to-year variability in herbivory patterns (Fig. 1), it is clear that insect herbivores consume a large Acknowledgments.I thank M. Axelrod, G.C. Faeth, T.G. Faeth, fraction of leaves in the spring and sometimes in the fall. T.J. Faeth, N.J. Tiller, and W.T. Tiller for assistance in the field. My results indicate phytochemical changes alone cannot M. Axelrod, A. Gieschen, and C. Schilling diligently performed explain this pattern. What then are possible explanations many of the phytochemical analyses in the laboratory. I am to W.V. Zuckerfor his in for the pattern of spring and occasional fall feeding of in- especially grateful help developing tannin and protein M. T.L. Bultman, P.C. sects on oak that was shown assay procedures. Auerbach, originally by Feeny (1970) Coley, J. Schultz, D. Simberloff, D.R. Strong, Jr. and M. Taper and documented here? Holmes et al. (1979) suggested inten- made many helpful comments on the manuscript.The U.S. Forest sity of bird predation in midsummer may force lepidopter- Serviceand the Center for EnvironmentalStudies (ASU) generous- ous insects to feed early in spring or in fall on northeastern ly provided access and accomodations at the Sierra Ancha Experi- trees. Two other hypotheses may explain the patterns of mental Station. This researchwas supported by a Faculty Grant- feeding by leaf-chewing insects on Q. emoryi: In-Aid from ASU and NSF grants BSR-8110832 and BSR- 8415616. 1. Early feeding may be partially linked with temperature requirements and life histories of chewing insects and Q. emoryi. In the study area, temperatures increased rapidly AppendixI. Standard errors of means for Figs. 2, 3, and 4. Stan- after May to highs of - 38° C, which may prohibit develop- dard errors are not shown on graphs because of some overlap ment or decrease fecundity of leaf-chewing insects. In of S.E. bars 1981-1983, most feeding was concentrated just after bud- burst from May to June, but before hot, dry conditions Date Protein Hydrolysable Condensed tannin tannin in June through mid-July (Fig. 1). In 1982-1983, most feed- occurred in the same and also from ing again interval, Sep- intact dam- intact dam- intact dam- tember to when November, temperatures subsided. This aged aged aged second interval of feeding may reflect feeding by diapaused larvae or adults, or successful later generations of spring 4-30-81 0.75 0.81 2.57 - 0.26 - feeders. It is interesting to note that temperatures were 5-31-81 0.42 0.78 1.30 1.41 0.43 0.36 above normal and precipitation below normal in the sum- 7-9-81 0.93 1.01 2.36 2.03 1.32 1.24 mer of 1981, while in 1982 both temperatures and precipita- 8-2-81 0.65 0.63 2.31 1.74 1.37 1.39 tion were at normal levels (National Weather Service). Tem- 9-5-81 0.52 0.59 0.98 1.47 1.42 1.63 perature and precipitation may partially explain the lack 10-10-81 0.62 0.70 1.37 2.57 1.19 1.45 11-7-81 0.71 0.87 2.40 1.84 of late summer and fall feeding in 1981 if these factors 1.25 1.48 4-11-82 0.24 0.31 1.21 3.01 1.23 1.43 caused low survivorship during the early summer months. Additional for this support hypothesis is that leaf-mining 4-30-82 0.62 0.38 - - - - insects on Q. emoryi, which are buffered from external tem- 5-19-82 0.86 0.73 4.73 6.30 0.40 0.45 peratures and changes in humidity since they are enclosed 6-16-82 1.36 0.74 1.85 1.76 0.75 0.54 in leaf tissues (Hering 1951; Faeth et al. 1981), begin feeding 7-14-82 0.82 0.50 1.95 1.59 0.75 0.86 in late June and continue through fall and winter months 8-18-82 0.77 0.96 1.67 2.43 0.81 1.76 (Faeth 1985, 1986). 9-11-82 0.59 0.51 1.38 1.56 0.86 1.42 10-7-82 1.05 0.83 1.73 1.57 0.57 0.98 2. Patterns of herbivory on Q. emoryi may be related to 11-19-82 0.76 0.96 1.83 1.99 0.87 0.59 2-26-82 1.05 1.44 1.31 1.31 0.66 chemical and physical changes that occur in damaged leaves 1.11 11-26-83 0.56 0.37 1.49 2.00 0.76 1.51 (Figs. 2-4). Many herbivores may concentrate feeding in 40

References Feeny PP (1976) Plant apparency and chemical defense. In: JW Wallace, RL Mansell (eds) Biochemical interactions between plants and insects. Recent Adv Phytochem 10:1-40 Baldwin IT, Schultz JC (1983) Rapid changes in tree leaf chemistry Feeny PP, Bostock H (1968) Seasonalchanges in the tannincontent induced by damage: evidence for communication between of oak leaves. Phytochem 7:871-880 plants. Science 221:277-279 Fox LR (1981) Defense and dynamics in plant-herbivoresystems. Bate-Smith EC (1977) Astringent tannins of Acer species. Phyto- Amer Zool 21:853-864 chemistry 16:1421-1426 Haukioja E, Niemela PE (1979) Birch leaves as a resource for BerenbaumMR (1983) Effects of tannins on growth and digestion herbivores: seasonal occurrence of increased resistance in fo- in two species of papilionids. Entomol Exper et Appi liage after mechanical damage of adjacent leaves. Oecologia 34:245-250 (Berlin) 39:151-159 Bernays EA (1978) Tannins: an alternative viewpoint. Entomol Heinrich B, Collins SL (1983) Caterpillar leaf damage and the Exper et Appl 24:244-253 game of hide-and-seekwith birds. Ecology 64:592-602 BernaysEA (1981) Plant tannins and insect herbivores:an apprais- Hering EM (1951) Biology of the leaf miners. Junk, Gravenhages, al. Ecol Entomol 6:353-360 The Netherlands Bernays EA, ChamberlainDJ, McCarthyP (1980) The differential Holmes RT, Schultz JC, Nothnagle PJ (1979) Bird predation on effects of ingested tannic acid on differentspecies of Acridoidea. forest insects: an exclosure experiment.Science 206:462-463 Entomol Exper et Appi 28:158-166 Karban R, Ricklefs RE (1984) Leaf traits and species richness Bernays EA, Woodhead S (1982) Plant phenols utilizedas nutrients and abundance of Lepidopteranlarvae on deciduous trees in by a phytophagous insect. Science 216:201-202 southern Ontario. Oikos 43:165-170 Bradstreet RB (1965) The Kjeldahl method for organic nitrogen. Lawton JH (1976) The structureof the arthropod community on Academic Press, NY bracken. Botan J Linn Soc 73:187-216 Broadhurst RB, Jones WT (1978) Analysis of cindensed tannins Mattson WJ (1980) Herbivoryin relationto plant nitrogencontent. using acidified vanillin. Sci Food Agric 29:788-794 Ann Rev Ecol System 11:119-161 Bryant JP (1981) Phytochemical deterrence of snowshoe hare McNeill S, Southwood TRE (1978) Role of nitrogenin the develop- browsing adventitious shoots of four Alaskan trees. Science ment of insect-plant relationships. In: JB Harborne (ed) Bio- 213:889-890 chemical aspects of plant and animal coevolution. Academic Carroll CR, Hoffman CA (1980) Chemicalfeeding deterrentmobi- Press, NY, pp 77-98 lized in response to insect herbivory and counteradaption by Milton K (1979) Factors influencing leaf choice by howler mon- Epilachnatredecimnota. Science 209:414-416 keys: a test of some hypotheses of food selection by generalist Coley PD (1983) Herbivory and defensive characteristicsof tree herbivores.Amer Nat 114:362-378 species in a lowland tropical forest. Ecol Mono 53:209-233 Oates JF, Waterman PG, Choo GM (1980) Food selection by Dement WA, Mooney HA (1974) Seasonal variation in production the south Indian leaf-monkey, Presbytisjohnii, in relation to of tannins and cyanogenic glycosides in the chaparral shrub, leaf chemistry.Oecologia (Berlin) 45:45-56 Heteromelesarbutifolia. Oecologia (Berlin) 15:65-76 Rhoades DF (1979) Evolution of chemical defense against herbi- EdwardsPJ, Wratten SD (1983) Wound induced defences in plants vores. In: GA Rosenthal, DH Janzen (eds) Herbivores.Their and their consequencesfor patternsof insect grazing.Oecologia interactionswith secondaryplant metabolites.Academic Press, (Berlin) 59:88-93 NY, pp 3-54 Faeth SH (1985) Host leaf selection by leaf-mininginsect, Stilbosis Rhoades DF, Cates RG (1976) Toward a general theory of plant juvantis: interactions among three trophic levels. Ecology antiherbivorechemistry. Rec Adv Phytochem 10:168-213 66:870-875 Ryan CA (1979) Proteinase inhibitors. In: GA Rosenthal, DH Faeth SH (1986) Indirect interactions between temporally-sepa- Janzen (eds) Herbivores. Their interactions with secondary rated herbivoresmediated by changes in the host plant. Ecology plant metabolites. Academic Press, NY, pp 599-618 (in press) Schultz JC, BaldwinIT (1982) Oak leaf quality declinesin response Faeth SH, Mopper S, Simberloff D (1981) Abundances and diver- to defoliation by gypsy moth larvae. Science 217:149-151 siy of leaf-mining insects on three oak host species: effects of Scriber M, Slansky F (1981) The nutritionalecology of immature host-plant phenology and nitrogen content of leaves. Oikos insects. Ann Rev Entomol 26:183-211 37:238-251 Vinson SB (1975) Biochemicalcoevolution betweenparasitoids and Feeny PP (1968) Effect of oak leaf tannins on larval growth of their hosts. In: PW Price(ed) Evolutionarystrategies of parasit- the winter moth Operophtera brumata. Insect Physiol ic insects and mites. Plenum Press, NY, pp 14-48 14:805-817 Winer BJ (1971) Statistical principles in experimental design. Feeny PP (1970) Seasonal changes in oak leaf tannins and nutrients McGraw-Hill,NY as a cause of spring feeding by wintermoth caterpillars.Ecology Yang SF, Pratt HK (1978) The physiology of ethylene in wounded 51:565-581 plant tissues. In: G Kahl (ed) Biochemistryof wounded plant Feeny PP (1975) Biochemicalcoevolution between plants and their tissues. de Gruyter, Berlin,pp 595-622 insect herbivores. In: LE Gilbert, PH Raven (eds) Coevolution Zucker WV (1983) Tannins: does structure determine function? of animals and plants: Symposium V. First InternationalCon- An ecological perspective.Amer Nat 121:335-365 gress System. and Evol Biol. University of Texas Press, Austin, Texas pp 3-19 Received March 20, 1985