*Corresponding author. Tel.:#1-617-627-3543; fax:#1-617-627-3805. E-mail address:[email protected] (C.M. Orians)
1Current address: Department of Biology, Loyola College, Baltimore, MD 21210, USA.
Phenolic glycosides and condensed tannins in
Salix sericea, S. eriocephala
and their F1 hybrids:
not all hybrids are created equal
Colin M. Orians
!
,
*
, Megan E. Gri
$
ths
!
, Bernadette M. Roche
"
,1
,
Robert S. Fritz"
!Department of Biology, Tufts University, Medford, MA 02155, USA
"Department of Biology, Vassar College, Poughkeepsie, NY 12601, USA Received 26 May 1999; accepted 26 July 1999
Abstract
The performance of hybrids depends upon the inheritance and expression of resistance traits. Secondary chemicals are one such resistance trait. In this study, we measured the concentra-tions of phenolic glycosides and condensed tannins in parental and F1 hybrid willows to examine the sources of chemical variation among hybrids. S. sericea produces phenolic glycosides, salicortin and 2@-cinnamoylsalicortin, and low concentrations of condensed tannin in its leaves. In contrast,S. eriocephalaproduces no phenolic glycosides but high concentrations of condensed tannins in its leaves. These traits are inherited quantitatively in hybrids. On average, F1 hybrids are intermediate for condensed tannins, suggesting predominantly additive inheritance or balanced ambidirectional dominance of this defensive chemical from the parental species. In contrast, the concentration of phenolic glycosides is lower than the parental midpoint, indicating directional dominance. However, there is extensive variation among F1 hybrids. The concentration of tannin and phenolic glycosides in F1 hybrid families is either (1) lower than the midpoint, (2) higher than the midpoint, or (3) indistinguishable from the midpoint of the two parental taxa. It appears that the production of the phenolic glycosides, especially 2@-cinnamoylsalicortin, is controlled by one or more recessive alleles. We also observed a two-fold or greater di!erence in concentration between some hybrid families. We discuss how chemical variation may e!ect the relative susceptibility of hybrid willows to herbivores. ( 2000 Elsevier Science Ltd. All rights reserved.
Keywords:Salicaceae; Willows; Hybridization; F1 hybrids; Phenolic glycosides; Condensed tannins
1. Introduction
Hybridization is now regarded as a common feature of many plant species, and can lead to speciation and introgression (Rieseberg, 1991, 1995; Rieseberg and Brunsfeld, 1992; Rieseberg and Ellstrand, 1993; Ellstrand et al., 1996; Arnold, 1997). In the last 10 yr, there have been numerous studies suggesting that hybridization, via changes in
resistance, also a!ects the ecology and evolution of plant-herbivore interactions (e.g.,
Whitham, 1989; Boecklen and Spellenberg, 1990; Whitham et al., 1991, 1994, 1999; Aguilar and Boecklen, 1992; Floate and Whitham, 1993; Paige and Capman, 1993; Fritz et al., 1994, 1996; Strauss, 1994; Messina et al., 1996; Orians et al., 1997; Fritz, 1999; Pilson, 1999). We know that the relative resistance of hybrids and parental taxa to herbivores is highly variable (reviewed by Strauss, 1994; Fritz, 1999), but what determines resistance is not well understood. Although multiple traits are involved, secondary chemicals, such as phenolics, terpenoids, alkaloids and glucosinolates, are often key components of the resistance of plants to herbivores (Rosenthal and Janzen,
1979; Roitberg and Isman, 1992; Bernays and Chapman, 1994; CardeH and Bell, 1995;
Chew and Renwick, 1995). Given the general importance of secondary chemistry to
plant}herbivore interactions, it is surprising that relatively few researchers have
studied the secondary chemistry of hybrids in relation to plant}herbivore interactions
(Huesing et al., 1989; Weber et al., 1994; Orians et al., 1997). We believe that knowledge of how secondary chemicals are expressed in hybrids will help researchers understand patterns of hybrid resistance.
The concentration of secondary chemicals in hybrids typically di!ers qualitatively
and quantitatively from their parents (Connor and Purdie, 1976; Harborne and Turner, 1984; Meier et al., 1989; Altman et al., 1990; Levy and Milo, 1991; Rieseberg and Ellstrand, 1993; Orians and Fritz, 1995). Qualitatively, novel chemicals may be found or parental chemicals may be lost. Quantitatively, patterns of expression depend upon the genetic control of production. Parental chemicals may be overex-pressed (over-dominance), underexoverex-pressed (incomplete dominance), intermediate (no dominance), similar to one of the parental species (dominance), or similar to both parents (co-dominance) (Van Brederode et al. 1974; Falconer, 1989).
Quantitat-ive chemical di!erences can a!ect the resistance of a plant (Huesing et al., 1989;
Orians et al., 1997). What is lacking is an understanding of how secondary
chemistry varies among hybrids in ways that could a!ect patterns of resistance to
herbivores.
We have been studying the consequences of hybridization in willow for a number of years. Willows and other Salicaceous plant species produce two main secondary chemicals: phenolic glycosides and condensed tannins, both of which are known to
in#uence the susceptibility of plants to insect and mammalian herbivores (Zucker,
In our system,S. sericeaproduces high concentrations of phenolic glycosides and
low concentrations of condensed tannin, whileS. eriocephalaproduces high
concen-trations of condensed tannin but no phenolic glycosides (Orians and Fritz, 1995). The concentrations of both phenolic glycosides and condensed tannins are intermediate,
not signi"cantly di!erent than the midpoint, in naturally occurring hybrids of these
two species (Orians and Fritz, 1995). However, these hybrids were of unknown pedigree. What is the chemistry of known F1 hybrids? Is there chemical variation
among F1 hybrids? How does the chemistry of experimental hybrids di!er from
natural hybrids? These are all questions that will help us understand how secondary chemistry might determine patterns of hybrid survival. Here we report on the concentrations of condensed tannin and phenolic glycosides in two year old seedlings
ofSalix sericea,Salix eriocephalaand their F1 hybrids.
2. Materials and methods
2.1. Plants
Both Salix sericea Marshall and Salix eriocephala Michx. are abundant in the
northeastern United States and eastern Canada (Argus, 1986). The species often co-occur and hybridization between the two is common (Argus, 1986; Mosseler and Papadopol, 1989; Fritz et al., 1994). Willows are dioecious and in this system the
initial hybridization event is unidirectional (S. eriocephalais the paternal parent and
S. sericeais the maternal parent), due to stigma-pollen incompatibility in femaleS.
eriocephala(Mosseler, 1990).
2.2. Generation of seedlings(parental and F1 hybrid)
We performed intraspeci"c and interspeci"c crosses using known pure parents,
based on RAPD analysis, to generate pure parental and F1 hybrid progenies. From a larger set of crosses (Fritz et al., 1998), we selected seven groups of progenies where
the parents of each F1 hybrid cross were used in the matched intraspeci"c cross (Table
1). Our goal was to avoid using a parental clone twice. However, oneS. sericeaparent
(s17) was involved in two crosses.
To ensure purity of each family, the female catkins were covered with mesh bags prior to stigma emergence to prevent visitation by insect pollinators. Bags were removed only while making crosses. The resulting seeds were mass planted in trays of
MetroMix 360TMin late May, and, covered with white, shear seed cloth to maintain
a high moisture level and to prevent accidental colonization of the trays by naturally dispersing willow seeds. The cloth was removed after about two weeks, when natural seed dispersal had ended. Four-week-old seedlings were transplanted into 3.7 l pots
Table 1
Crosses performed to generate related parental and F1 hybrid progeny
Group S. sericea F1 Hybrids S. eriocephala
$]# $]# $]# Note: F1 hybrids within each group have a female in common with theSalix sericeaprogeny and male in common with theS. eriocephalaprogeny.
mixture mixed with 13 g of slow-release fertilizer (10 : 10 : 10 NPK). Plants were randomized within blocks.
Leaf samples for chemical analyses were collected on 18 July. Twelve fully expanded
leaves were collected from each plant (the"rst two fully expanded leaves from each of
six shoots). Leaves were cut at the petiole, placed on ice, transported to the lab, and vacuum-dried for at least 24 h. Vacuum-drying leaves has been shown to prevent the loss of both condensed tannin and phenolic glycosides (Orians, 1995). Once dry, leaves
were ground to a"ne powder using a Wiley Mill equipped with a size 30 mesh and
then stored in a!203C freezer.
2.3. Chemical analyses
The phenolic glycosides were assayed using standard techniques (Orians, 1995).
Brie#y, leaf powder (30$3 mg) was extracted in cold MeOH (10 mg leaf powder/1 ml
MeOH) with sonication for 10 min. Cold water was constantly#ushed through the
sonicator to prevent the heating of samples. We centrifuged and "ltered (0.45lm
"lter) each sample before placing extracts in crimp-top vials. Extracts were kept in the
freezer until analysis (48 h or less). We quanti"ed the concentration of glycosides with
an HPLC equipped with an autosampler and a UV detector set at 274 nm (Hewlett-Packard). A reverse-phase NOVA-PAK C
18 (4lm) column (Waters) and a gradient
system of distilled water and MeOH was used. 1,3-dimethoxybenzene was used as the
internal standard. Standard curves were determined for salicortin and 2@
-cinnamoyl-salicortin.
The condensed tannins also were extracted using standard techniques (Orians,
1995). Brie#y, approximately 300 mg leaf powder was washed with ether and extracted
with 70% acetone for 3 h in a 403C water bath. Acetone was removed under reduced
pressure and all extracts were diluted to 8 ml with distilled water. The butanol/HCl method was used to quantify tannin concentrations (Hagerman and Butler, 1989). The
concentration of the condensed tannin (mg/g dry leaf) was determined using puri"ed
2.4. Statistical analyses
A one way analysis of variance was used to determine the overall e!ects of taxon on
condensed tannins (using the means of each of the 7 families/taxon). The natural log of condensed tannin concentration was used in the analysis because variance increased
with the mean. For the phenolic glycosides onlyS. sericeaand hybrids were included
in the model, becauseS. eriocephalagenerally does not contain phenolic glycosides.
Contrasts were performed to determine if the hybrids were signi"cantly di!erent
from the midpoint of the two parental taxa (non-transformed data were used for the
contrasts), that is, if one averages the parental values do the hybrids di!er from that
average. This was done using the means comparison contrasts in SuperANOVA (Abacus Concepts, 1989). These comparisons were performed with all taxa included in the model.
Separate analyses were done for each group of families (Table 1). As above, an
analysis of variance was used to determine the e!ects of family (or taxon) on
chemistry, and contrasts, as described above, were performed to determine if the
concentrations of the chemicals in the hybrid family was signi"cantly di!erent from
the average of the two parental families.
Regression analysis was used to determine the relationship between parental and hybrid chemistry, and between the production of phenolic glycosides and condensed tannins.
3. Results
In general, the concentrations of condensed tannins is high inS. eriocephala, low in
S. sericeaand intermediate in hybrids, while the concentration of the two phenolic
glycosides shows the opposite pattern (Table 2). The concentrations measured in these
plants are similar to that measured in"eld plants (unpublished data), and the patterns
are similar to those reported previously (Orians and Fritz, 1995). In general, S.
eriocephaladoes not produce phenolic glycosides in its leaves (see also Orians and
Fritz, 1995). However, two of the four siblings of oneS. eriocephalacross (e60]hhe12)
produced phenolic glycosides (data not shown in Table 2), and the four s78]hhe12
siblings had levels of phenolic glycosides and tannins similar toS. sericea. Although
these results suggest there is something unusual about hhe12, other crosses involving
hhe12 (e64]hhe12 and s57]hhe12) were not unusual (unpublished data). Currently,
we are unable to explain the anomalous results obtained with the hhe12 crosses. Because of these anomalous results, we excluded crosses involving hhe12 before
testing for taxon e!ects. Overall, there was a signi"cant e!ect of taxon on chemistry
and all three taxa were signi"cantly di!erent from one another (ANOVA,F*125.6,
p)0.0001) (Table 2). Contrasts revealed signi"cant deviation from the average (or
midpoint) of the two parental taxa for salicortin (p(0.01) and 2@-cinnamoylsalicortin
(p(0.001), but not for condensed tannins (p"0.99). For both phenolic glycosides,
Table 2
Mean concentration (MG/G dry leaf weight), standard errors (in parentheses), and coe$cients of variation (cv) for condensed tannin and the phenolic glycosides inS. sericea, F1 hybrids andS. eriocephalaleaves
Taxon Chemical type
Condensed tannin Salicortin 2@-cinnslctn
S. sericea 27.8 (1.8) a 104.8 (4.1) a 18.5 (1.4) a
cv"7.6 cv"10.2 cv"20.0
F1 hybrid 71.7 (11.9) b 48.0 (10.7) b 4.5 (2.4) b
cv"19.7 cv"33.5 cv"44.1
S. eriocephala 126.6 (4.2) c None None
cv"10.6
Note: 2@-cinnslctn"2@-cinnamoylsalicortin. Di!erent letters denote signi"cant di!erences among taxa (p(0.001).
In addition, there was more among-family variation for hybrids than for the
parental taxa (see Coe$cients of Variation in Table 2). This was especially evident for
the phenolic glycosides. Some hybrid families produced very low concentrations of
salicortin and 2@-cinnamoylsalicortin (26.3 and 0.6 mg/g dry leaf weight, respectively
for Group IV) while other families produced relatively high concentrations (60.6 and 2.1 mg/g dry leaf weight, respectively for Group VI). Thus some families produce over
twice the concentration of other families. If we include the s78]hhe12 cross salicortin
and 2@-cinnamoylsalicortin concentrations were even higher 106.7 and 18.4 mg/g dry
leaf weight, respectively.
For each group of crosses, there was a highly signi"cant di!erence among families
(or taxa) (F*19.5;p(0.001) (Table 3). Based on our previous results, we expected
that F1 chemistry would be indistinguishable from the midpoint of the parental taxa. However, contrasts revealed that some F1 hybrids were lower than the midpoint, while others were higher than the midpoint (Fig. 1). Excluding the cross involving
hhe12 (Group VII), condensed tannin concentration was signi"cantly di!erent than
the parental midpoint for 2 of 6 hybrid families, one higher and one lower. Salicortin
concentration was signi"cantly lower than the midpoint for 3 of 6 families. In contrast,
2@-cinnamoylsalicortin concentrations were signi"cantly lower for all 6 hybrid
fami-lies. For Group VII, condensed tannin was lower and the phenolic glycosides were higher than the parental midpoint (Fig. 1). Generally, the production of phenolic
glycosides, especially 2@-cinnamoylsalicortin, showed directional dominance or
reces-sive deviations.
There were signi"cant positive correlations between parental and hybrid chemistry.
The concentration of condensed tannins in the hybrids was positively correlated with
the concentration in the S. sericea parent (R2"0.79, p"0.02), but not with the
S. eriocephala parent (R2"0.08, p"0.59). There was a positive correlation for
2@-cinnamoylsalicortin (R2"0.64,p"0.05). However, there was no correlation for
Table 3
Family mean (standard error) concentration, MG/G dry leaf weight, for condensed tannin and phenolic glycosides inS. sericea(SS), F1 hybrids andS. eriocephala(SE)
Group Family Tannin Salicortin 2@cinnslct
I SS s17]s22 22.8(3.3) a 110.8 (7.2) a 18.8 (2.4) a
F1 s17]e18 66.8 (10.3) b 43.9 (6.7) b 2.2 (0.5) b
SE e15]e18 122.4(4.7) c NONE NONE
II SS s17]s16 20.8(0.6) a 121.2 (4.4) a 13.4 (1.0) a
F1 s17]ne36 48.6 (10.1) b 39.3 (8.1) b 1.8 (0.5) b
SE e23]ne36 106.4(7.0) c NONE NONE
III SS s28]s27 24.7(2.2) a 98.7 (6.2) a 17.9 (1.9) a
F1 s28]ne35 87.4(7.1) b 31.0 (9.5) b 2.4 (1.0) b
SE e23]ne35 113.0(3.0) c NONE NONE
IV SS s37]s55 25.3(4.6) a 97.4 (6.3) a 14.2 (1.0) a
F1 s37]ne33 86.1(8.0) b 26.3 (3.9) b 1.1 (0.2) b
SE e29]e115 122.4(5.5) c NONE NONE
V SS s83]s14 24.7(3.5) a 88.9 (8.9) a 19.9 (0.9) a
F1 s83]ne37 70.5 (10.8) b 28.6 (4.5) b 1.7 (0.5) b
SE e3]ne37 117.5(3.3) c NONE NONE
VI SS s1]s95 22.6(1.6) a 105.4 (7.6) a 23.3 (2.5) a
F1 s1]hhe46 72.8(4.4) b 60.6 (6.6) b 3.9 (1.0) b
SE e64]hhe46 143.9(4.8) c NONE NONE
VII SS s78]S60 21.4(0.9) a 111.2(7.9) a 22.0 (2.1) a
F1 s78]hhe12 15.8(1.1) a 106.7(4.2) a 18.5 (1.1) a
SE e60]hhe12 116.1 (13.5) b 24.9 (17.3) b 0.2 (0.2) b
Note: 2@cinnslct"2@-cinnamoylsalicortin. Di!erent letters denote signi"cant di!erences among families within each group (p(0.001).
Among hybrids we found an inverse correlation between the concentration of
tannin and salicortin (Fig. 2) and between tannin and 2@-cinnamoylsalicortin
(>"78.5!3.3X, R2"0.41, p(0.001). These negative correlations suggest that
hybrids may not be able to produce high concentrations of both phenolic glycosides and condensed tannins (see also Orians and Fritz, 1995). Interestingly, many F1 hybrids contained low concentrations of both condensed tannin and salicortin (Fig. 2).
4. Discussion
Fig. 1. The proportional deviation from the parental midpoint. (A) Values above the midpoint indicate that condensed tannin concentrations deviate towardS. eriocephala. (B) and (C) Values above the midpoint indicate salicortin and 2@-cinnamoylsalicortin concentrations deviate towardS. sericea. NS: not signi"cant (*)p(0.05; (**)p(0.01.
2@-cinnamoylsalicortin (Orians et al., 1996). As a consequence most parental chemicals
are found in hybrids (Table 2; Rieseberg and Ellstrand, 1993). However, the concen-tration in the hybrids can be quite variable (Harborne and Turner, 1984). Our results show that on average F1 hybrids contain intermediate concentrations of condensed tannin (equal to the parental midpoint), suggesting predominantly additive inherit-ance or a lack of domininherit-ance of this defensive chemical from the parental species. In
contrast, the concentration of the phenolic glycosides, salicortin and 2@
-cinnamoyl-salicortin, is lower than the parental midpoint, indicating incomplete dominance over production. This incomplete dominance suggests that one or more alleles that control
phenolic glycoside production, especially 2@-cinnamoylsalicortin, are recessive in the
F1 hybrid. Altman et al. (1990) also found recessive control of the production of the terpenoid raimondal in cotton hybrids; compared to the parental taxon, the
concen-tration in hybrids was 9}12%.
Fig. 2. The relationship between condensed tannin and salicortin concentration in F1 hybrids ofS. sericea andS. eriocephala.
patterns of secondary chemical expression is variable. Of the chemicals present in both parents and hybrids, the concentration of 38% were similar to one or both of the two parental taxa (dominance or co-dominance), 29% were intermediate to the two parental taxa (incomplete dominance or no dominance), 19% were greater than either parent (over-dominance), and 14% were less than either parent (underexpression) (Fahselt and Ownbey, 1968; Belzer and Ownbey, 1971; McMillan et al., 1975; Har-borne and Turner, 1984; Spring and Schilling, 1990; Buschmann and Spring, 1995; Orians and Fritz, 1995). Parental chemicals are not always expressed in the F1 hybrids. Current estimates suggest that parental chemicals are found only 68% of the time, lost 27% of the time and novel 5% of the time in the hybrids (Rieseberg and Ellstrand, 1993).
However, none of the above-cited studies looked at variation among F1 hybrids. We found extensive variation among F1 hybrid families (Table 2). Some hybrid families produced concentrations higher or lower than the midpoint (incomplete dominance), concentrations similar to one of the parental taxa (dominance, Group VII), or levels indistinguishable from the parental midpoint (no dominance). Furthermore, there was extensive variation among hybrid families. For example, the concentration of salicortin in one family was over twice that of a second family (Group
VI vs. IV, Table 3). Parental production explained some of the di!erences in
produc-tion of condensed tannin and 2@-cinnamoylsalicortin by hybrids families; there was
concentration of condensed tannin in hybrids was not correlated withS. eriocephala,
but withS. sericea, the parental taxon that produces low concentrations. It is not clear
why this occurred. No correlation was observed for related parents and hybrids for
salicortin. Although genetic di!erences among parental individuals are the most likely
explanation for the variation in concentration, non-genetic maternal factors could be
important as well. Further work is required to di!erentiate between genetic and
maternal e!ects.
The negative correlation between condensed tannin and phenolic glycosides in
hybrids suggests that F1 hybrid o!spring are not able to produce high concentrations
of both condensed tannin and phenolic glycosides (Fig. 2). We reported a similar
trade-o! previously (Orians and Fritz, 1995). The basis of this trade-o! between
tannins and phenolic glycosides remains unknown. There are two possible
explana-tions. First, there may be a resource allocation trade-o! such that only a certain
amount of carbon is allocated to these carbon-based defenses. Production of one would limit the production of the other. However, this explanation does not explain the low production of both chemicals in some hybrids. Second, there may be a genetic
tradeo!. If speci"c alleles control the production of both chemicals, one chemical
may be produced at the expense of the other ("negative pleiotropy). Again,
this explanation does not explain the reduced production of both chemical types in some individuals. Perhaps the reduced production is due to heterozygosity in some parents. This could lead to low production of one or both chemicals in some hybrids.
Additional work is required to determine if this trade-o! is due to resource
allocation constraints or to genetic constraints. The analysis of later generation hybrids could provide critical information. Unless there is strong genomic selection preventing reorganization (Rieseberg et al., 1995, 1996), F2 recombinants could contain all the alleles responsible for the production of both chemicals. These individuals would be able to produce high concentrations of both chemicals, unless there is a resource allocation constraint. We are currently designing experiments with F2 recombinants to test these alternative hypotheses. Furthermore, if we do identify F2 recombinants that produce high concentrations of both, we could examine the costs of chemical production (defense). Reduced growth of these individuals
relative to those with lower concentrations would suggest a trade-o!between growth
and defense.
4.1. Implications
This study clearly documents extensive variation among F1 hybrids, and this
variation may a!ect the hybrid survival. Phenolic glycosides, for example, are
Unfortunately, few studies have evaluated the importance of secondary chemistry. Previously, we showed the phenolic glycoside chemistry of naturally occurring
hybrids was not signi"cantly di!erent than the midpoint of the two parental taxa
(Orians and Fritz, 1995; Orians, unpublished data). In contrast, in this study, we show that the concentration of phenolic glycosides in F1 hybrids was lower on average than the midpoint. Also, the concentrations reported here are lower than in our previous studies of naturally occurring hybrids (Orians and Fritz, 1995; Orians, unpublished data). Taken together, these results suggests that recombinant hybrids that produce higher concentrations of phenolic glycosides may be more likely to survive. If true, it
would be the "rst example of selection on hybrid chemistry. We are currently
experimentally testing the role of chemistry in the resistance of hybrids.
Are herbivores generally selective"lters on the"tness of hybrids? Unfortunately,
there is little empirical work on this question. Our work and that of others has demonstrated extensive variation in the secondary chemistry of hybrid plants, and we
believe that such variation can a!ect hybrid"tness.
Overall, our understanding of the e!ects of hybridization on plant}herbivore
interactions is in its infancy. There is debate over the role of herbivores in plant
hybridization, and the e!ects of hybridization on community structure, host range
expansion, and evolution of herbivore virulence (Whitham, 1989; Floate and Whitham, 1993; Fritz, 1999; Pilson, 1999; Whitham et al., 1999). Secondary chemistry
is unquestionably an important determinant of plant}herbivore interactions, and we
believe more careful attention to chemistry will help predict the ecological and evolutionary consequences of hybridization.
Acknowledgements
We give thanks to Len and Ellie Sosnowski for permitting us to conduct experi-ments on their property, Rachel Samberg for help with the chemical analyses, and Diana Pilson for her thoughtful discussion of the results. This research was supported by the National Science Foundation (Grant No. DEB 92-07363), a Science Education Grant from the Howard Hughes Medical Institute, the Faculty Research Awards Committee (Tufts University). This work was completed while CMO was a Mellon Grant Research Fellow at Tufts University.
References
Abacus Concepts, 1989. SuperANOVA. Abacus Concepts, Berkeley, CA.
Aguilar, J.M., Boecklen, W.J., 1992. Patterns of herbivory in theQuercus grisea]Quercus gambeliispecies
complex. Oikos 64, 498}504.
Altman, D.W., Stipanovic, R.D., Bell, A.A., 1990. Terpenoids in foliar pigment glands of A, D, and AD genome cottons: introgression potential for pest resistance. J. Heredity 81, 447}454.
Argus, G.W., 1986. The genusSalix(Salicaceae) in the southeastern United States. Syst. Bot. Monogr. 9, 1}170.
Arnold, M.L., 1997. Natural Hybridization and Evolution. Oxford University Press, Oxford.
Belzer, N.F., Ownbey, M., 1971. Chromatographic comparisons ofTragopogonspecies and hybrids. Am. J. Bot. 58, 791}802.
Bernays, E.A., Chapman, R.F., 1994. Host-Plant Selection by Phytophagous Insects. Chapman & Hall, New York.
Berenbaum, M. R., Zangerl, A. R., 1992. Genetics of secondary metabolism and herbivore resistance in plants. In: Rosenthal, G. A., Berenbaum, M. R. (Eds.), Herbivores, Their Interactions with Plant Secondary Metabolites, Vol. 2, Ecological and Evolutionary Processes, 2nd edition. Academic Press, San Diego, pp. 415}438.
Boecklen, W.J., Spellenberg, R., 1990. Structure of herbivore communities in two oak (Quercusspp.) hybrid zones. Oecologia 85, 92}100.
Buschmann, H., Spring, O., 1995. Sesquiterpene lactones as a result of interspeci"c hybridization in Helianthusspecies. Phytochemistry 39, 367}371.
CardeH, R.T., Bell, W.J. (Eds.), 1995. Chemical Ecology of Insects, Vol. 2. Chapman & Hall, New York. Chew, F.S., Renwick, J.A.A., 1995. Host-plant choice inPierisbutter#ies. In: CardeH, R.T., Bell, W.J. (Eds.),
Chemical Ecology of Insects, Vol. 2. Chapman and Hall, New York, pp. 214}238.
Clausen, T.P., Reichardt, P.B., Bryant, J.P., Werner, R.A., Post, K., Frisby, K., 1989. Chemical model for short-term induction in quacking aspen (Populus tremuloides) foliage against herbivores. J. Chem. Ecol. 15, 2335}2346.
Connor, H.E., Purdie, A.W., 1976. Inheritance of triterpene methyl ethers inCortaderia (Gramineae). Phytochemistry 15, 1937}1939.
Denno, R.F., Larsson, S., Olmstead, K.L., 1990. Role of enemy-free space and plant quality in host-plant selection by willow beetles. Ecology 71, 124}137.
Ellstrand, N.C., Whitkus, R., Rieseberg, L.H., 1996. Distribution of spontaneous plant hybrids. Proc. Natl. Acad. Sci. 93, 5090}5093.
Fahselt, D., Ownbey, M., 1968. Chromatographic comparison ofDicentraspecies and hybrids. Am. J. Bot. 55, 334}345.
Falconer, D.S., 1989. Introduction to Quantitative Genetics, 3rd edition. Wiley, Somerset, NJ.
Floate, K.D., Whitham, T.G., 1993. The`Hybrid Bridgeahypothesis: host shifting via plant hybrid swarms. Am. Nat. 141, 651}662.
Fritz, R.S., 1999. Resistance of hybrid plants to herbivores: genes, environment, or both? Ecology 80, 382}391.
Fritz, R.S., Nichols-Orians, C.M., Brunsfeld, S.J., 1994. Interspeci"c hybridization of plants and resistance to herbivores: hypotheses, genetics, and variable responses in a diverse herbivore community. Oecologia 97, 106}117.
Fritz, R.S., Roche, B.M., Brunsfeld, S.J., Orians, C.M., 1996. Interspeci"c and temporal variation in herbivore responses to hybrid willows. Oecologia 108, 121}129.
Fritz, R.S., Roche, B.M., Brunsfeld, S.J., 1998. Genetic variation in resistance of hybrid willows to herbivores. Oikos 83, 117}128.
Hagerman, A.E., Butler, L.G., 1989. Choosing appropriate methods and standards for assaying tannin. J. Chem. Ecol. 15, 1795}1810.
Harborne, J.B., Turner, B.L., 1984. Plant Chemosystematics. Academic Press, San Diego.
Huesing, J., Jones, D., Deverna, J., Myers, J., Collins, G., Severson, R., Sisson, V., 1989. Biochemical investigations of antibiosis material in leaf exudate of wild Nicotiana species and interspeci"c hybrids. J. Chem. Ecol. 15, 1203}1217.
Julkunen-Tiitto, R., 1986. A chemotaxonomic survey of phenolics in leaves of Northern Salicaceae species. Phytochemistry 25, 663}667.
Julkunen-Tiitto, R., 1989. Phenolic constituents ofSalix: a chemotaxonomic survey of further Finnish species. Phytochemistry 28, 2115}2125.
Kolehmainen, J., Roininen, H., Julkunen-Tiitto, R., Tahvanainen, J., 1994. Importance of phenolic gluco-sides in host selection of shoot galling saw#y,Euura amerinae, onSalix pentandra. J. Chem. Ecol. 20, 2455}2466.
Lindroth, R.L., Peterson, S.S., 1988. E!ects of plant phenols on performance of southern armyworm larvae. Oecologia 75, 185}189.
Lindroth, R.L., Scriber, J.M., Hsia, M.T.S., 1988. Chemical ecology of the tiger swallowtail: mediation of host use by phenolic glycosides. Ecology 69, 814}822.
McMillan, C., Chavez, P.I., Mabry, T.J., 1975. Sesquiterpene lactones ofXanthium strumariumin a Texas population and in experimental hybrids. Biochem. Syst. Ecol. 3, 137}141.
Meier, B., Bettschart, A., Shao, Y., Lautenschlager, E., 1989. Einsatz der modernen HPLC fur chemotaxonomische Untersuchungen morphologisch schwer zu di!erenzierender Salix-Hybriden. Planta Medica 55, 213}214.
Messina, F.J., Richards, J.H., McArthur, E.D., 1996. Variable responses of insects to hybrid versus parental sagebrush in common gardens. Oecologia 107, 513}521.
Mosseler, A., 1990. Hybrid performance and species crossability relationships in willows (Salix). Can. J. Bot. 68, 2329}2338.
Mosseler, A., Papadopol, C.S., 1989. Seasonal isolation as a reproductive barrier among sympatricSalix species. Can. J. Bot. 67, 2563}2570.
Orians, C.M., 1995. Preserving leaves for tannin and phenolic glycoside analyses: a comparison of methods using three willow taxa. J. Chem. Ecol. 21, 1235}1243.
Orians, C.M., Fritz, R.S., 1995. The e!ects of hybridization between twoSalixspecies on foliar chemistry. J. Chem. Ecol. 21, 1245}1253.
Orians, C.M., Huang, C., Wild, A., Dorfman, K.A., Zee, P., Dao, M.T.T., Fritz, R.S., 1997. Willow hybridization di!erentially a!ects preference and performance of herbivorous beetles. Entomol. Exp. Appl. 83, 285}294.
Orians, C.M., Roche, B.M., Fritz, R.S., 1996. The genetic basis for variation in the concentrations of phenolic glycosides inSalix sericea: An analysis of heritability. Biochem. Syst. Ecol. 24, 719}724. Paige, K.N., Capman, W.C., 1993. The e!ects of host-plant genotype, hybridization, and environment on
gall-aphid attack and survival in cottonwood: the importance of genetic studies and the utility of RFLPS. Evolution 47, 36}45.
Pilson, D., 1999. Plant hybrid zones and insect host range expansion. Ecology 80, 407}415.
Rieseberg, L.H., 1991. Homoploid reticulate evolution inHelianthus(Asteraceae): Evidence from ribosomal genes. Am. J. Bot. 78, 1218}1237.
Rieseberg, L.H., 1995. The role of hybridization in evolution: old wine in new skins. Amer. J. Bot. 82, 944}953.
Rieseberg, L.H., Brunsfeld, S.J., 1992. Molecular evidence and plant introgression. In: Soltis, P.S., Soltis, D.E., Doyle, J.D. (Eds.), Molecular Systematics of Plants. Chapman & Hall, New York, pp. 151}176.
Rieseberg, L.H., Ellstrand, N.C., 1993. What can molecular and morphological markers tell us about plant hybridization. Crit. Rev. Pl. Sci. 12, 213}241.
Rieseberg, L.H., Sinervo, B., Linder, C.R., Ungerer, M.C., Arias, D.M., 1996. Hybrid speciation accom-panied by genomic reorganization in wild sun#owers. Science 272, 741}745.
Rieseberg, L.H., Van Fossen, C., Desrochers, A.M., 1995. Hybrid speciation accompanied by genomic reorganization in wild sun#owers. Nature 375, 313}316.
Roitberg, B.D., Isman, M.B. (Eds.), 1992. Insect Chemical Ecology. Chapman Hall, New York. Rosenthal, G.A., Janzen, D.H. (Eds.), 1979. Herbivores: Their Interactions with Secondary Plant
Metab-olites. Academic Press, New York.
Rowell-Rahier, M., 1984. The presence or absence of phenolglycosides inSalix(Salicaceae) leaves and the level of dietary specialization of some of their herbivorous insects. Oecologia 62, 26}30.
Schultz, J.C., 1989. Tannin-insect interactions. In: Hemingway, R.W., Karchesy, J.J. (Eds.), Chemistry and Signi"cance of Condensed Tannins. Plenum Press, New York, pp. 417}433.
Spring, O., Schilling, E.E., 1990. The origin of Helianthu multiyorus and H.]laetiyorus (Asteraceae). Biochem. Syst. Ecol. 18, 19}23.
Strauss, S.Y., 1994. Levels of herbivory and parasitism in host hybrid zones. Trends Ecol. Evol. 9, 209}214. Tahvanainen, J., Julkunen-Tiitto, R., Kettunen, J., 1985a. Phenolic glycosides govern the food selection
Tahvanainen, J., Helle, E., Julkunen-Tiitto, R., Lavola, A., 1985b. Phenolic compounds of willow bark as deterrents against feeding by mountain hare. Oecologia 65, 319}323.
Van Brederode, J., Van Wielink-Hillebrands, G.H., Van Nigtevecht, G., 1974. Dominance relations between isovitexin: 7-O-glycosyltransferase alleles inMelandrium. Molecular gen. Genet. 130, 307}314. Weber, D.J., Gang, D.R., Halls, S.C., Smith, B.N., McArthur, E.D., 1994. Inheritance of hydrocarbons in
subspeci"c big sagebrush (Artemisia tridentata) hybrids. Biochem. Syst. Ecol. 22, 689}697. Whitham, T.G., 1989. Plant hybrid zones as sinks for pests. Science 244, 1490}1493.
Whitham, T.G., Martinsen, G.D., Floate, K.D., Dungey, H.S., Potts, B.M., Keim, P., 1999. Plant hybrid zones a!ect biodiversity: tools for a genetic-based understanding of community structure. Ecology 80, 416}428.
Whitham, T.G., Morrow, P.A., Potts, B.M., 1991. Conservation of hybrids. Science 254, 779}780. Whitham, T.G., Morrow, P.A., Potts, B.M., 1994. Plant hybrid zones as centers of biodiversity: the
herbivore community of two endemic Tasmanian eucalypts. Oecologia 97, 481}490.
