| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ecology |
University of California, Life Sciences Building, Santa Barbara, California 93106 USA
Received for publication July 16, 2005. Accepted for publication October 13, 2006.
ABSTRACT
The ecological causation hypothesis for secondary sexual dimorphism was tested in Salix glauca, a dioecious willow shrub. Plants growing in a Colorado Rocky Mountain (USA) krummholz mosaic of mesic and xeric patches were monitored for four consecutive years. Ecological causation is predicated on unique resource demands associated with sexual function. In S. glauca, seeds have twofold higher N and P concentrations compared to pollen. P, but not N, allocation costs differed between sexes at plant and flower scales. Ecological causation also predicts spatial segregation of sexes along underlying habitat gradients. In five populations of S. glauca, sexes displayed significant spatial segregation. The theory also predicts that sexes differ in performance across gradients of environmental stress or resource availability. Consistent with this hypothesis, females had lower drought tolerance than males under years of extreme aridity. Furthermore, over 10 years, annual shoot growth for females was greatest in mesic habitat patches, while males grew at a consistent rate regardless of habitat aridity. Because current shoot growth is correlated with future catkin production, habitat specialization likely provides a fitness payoff in females. Overall, this long-term study provides some of the strongest evidence to date for ecological causation of secondary sexual dimorphism in plants.
Key Words: Colorado Rocky Mountains • dioecy • ecological causation hypothesis • Salicaceae • Salix • secondary sexual dimorphism • water relations • willows
The separation of male and female sexual functions into discrete individuals is a common animal phenomenon and occurs in a large number of plants as well. Dioecy is found in 157 flowering plant families and in 7.5% of flowering plant genera (Renner and Ricklefs, 1995
). Once dioecy has arisen, this does not end natural selection on gender dimorphism (see reviews of the evolution of dioecy in Bawa, 1980
; Charlesworth, 1999
), but rather initiates the evolution of secondary sexual dimorphism. Secondary sexual dimorphism is the separation of male and female characters that are not related directly to gamete production (i.e., the androecium and gynoecium in plants). These characters could be morphological as in body size differences in many animal species (Darwin, 1874
), physiological involving differentiation in photosynthesis or water use (Dawson and Geber, 1999
; Obeso, 2002
), or phenological as in the timing or age of onset for reproduction (Delph, 1999
). The theory of ecological causation explains differentiation in secondary characteristics between sexes by postulating that males and females have adapted to different ecological niches (Shine, 1989
). Here, I address evidence for niche divergence in accord with the ecological causation hypothesis, noting that other major hypotheses for secondary sexual divergence are not mutually exclusive and that sexual selection and fecundity selection could also act on niche constructing traits.
Ecological causation may proceed via any of three general mechanisms: dimorphic niches (Ralls, 1976
; Hedrick and Temeles, 1989
), bimodal niches (Slatkin, 1984
; Shine, 1989
), or competitive displacement (Slatkin, 1984
; Eppley, 2006
). Although these mechanisms may not be mutually exclusive and may piggyback on one another (Shine, 1989
), this study focuses on the dimorphic niche scenario of ecological causation as the mechanism giving rise to secondary sexual divergence in Salix glauca (Salicaceae). The dimorphic niche mechanism is based on the assumption that there are intrinsic differences between males and females due to their reproductive roles (Ralls, 1976
; Hedrick and Temeles, 1989
).
A complete assessment of the dimorphic niche hypothesis requires several lines of evidence. First, the hypothesis is based on sexual allocation theory and assumes that allocation to reproduction differs between sexes (Charlesworth and Charlesworth, 1978
, 1987
; Charnov, 1982
; Stanton and Galloway, 1990
; Charlesworth, 1999
). Under resource limitation, niche differentiation will be favored if resources limiting reproduction differ between males and females, reflecting the nutrient makeup of pollen vs. seeds and fruit as well as the timing of allocation to reproduction (Bullock and Bawa, 1981
; Ågren, 1988
; Stanton and Galloway, 1990
; Ashman, 1992
; Ashman and Baker, 1992
). Consistent with this view, on a per flower basis, nutrient allocation to reproduction differs between the sexes in several gynodioecious and dioecious plant species (Ågren, 1988
; Ashman, 1992
, 1994b
; Ashman and Baker, 1992
; Gehring and Monson, 1994
). However, it is unclear whether these differences scale up to the whole plant level or are balanced by sexual divergence in flower number (Carroll and Delph, 1996).
A second prediction of the dimorphic niche hypothesis is that spatial segregation of the sexes will accompany secondary dimorphism if resources are distributed in a heterogeneous manner across a population's range (Cox, 1981
; Geber, 1999
). In the dwarf willow Salix arctica, female plants are common in mesic, nutrient-rich habitats while males are common in xeric, nutrient-poor sites (Dawson and Bliss, 1989b
). If spatial segregation is not evident in dioecious populations, it may be that the appropriate niche dimensions have not been explored. For example, sexes could possess different resource efficiencies or morphologies (e.g., rooting depth) allowing them to co-occur and subdivide seemingly homogenous habitats.
Third, the dimorphic niche hypothesis predicts that males and females should differ in performance across gradients of abiotic stress or resource availability. One of the most important resources to terrestrial organisms is water. Several studies support the idea that the sexes of dioecious plants differ in water relations (Dawson and Bliss, 1989a
, b
, 1993
; Dawson and Ehleringer, 1993
; Laporte and Delph, 1996
; Dawson and Geber, 1999
; Espirito-Santo et al., 2003
). In a classic study, Dawson and Bliss (1989b)
showed that males and females of S. arctica differ in drought tolerance and water use efficiency (WUE). Females exhibited greater stomata conductance in mesic habitats, but males had greater WUE than females under xeric conditions. These findings indicate that males and females have diverged genetically to exploit relatively dry vs. wet conditions.
Last, if ecological niches of sex morphs diverge, then growth rates of the two sexes should be maximized under unique environmental conditions. Most comparisons of performance between sexes of dioecious plants have focused on short-term physiological measurements rather than long-term growth. Comparisons of short-term measures of performance alone can be misleading because correlations of fitness and growth to physiological traits varies widely among species, environments, and life stages (Poorter, 1989
; Geber and Griffen, 2003
). The one study to date comparing growth rates of males and females was conducted in a common garden environment, potentially inflating the magnitude of genetically based differentiation in growth rate between sexes (Wheelwright and Logan, 2004
).
Here, I comprehensively address the four predictions of the dimorphic niche hypothesis in Salix, a model system for the study of secondary sexual divergence. Through the use of experiments and observations in natural populations of Salix glauca over 4 years, I focused on the following questions: (1) Do sexes exhibit contrasting patterns of nutrient investment in reproduction? (2) Do spatial distributions of the sexes differ, as predicted under divergent selection for resource uptake? (3) Do sexes exhibit specialization in physiological traits related to resource gradients in the natural environment? (4) Do unique combinations of environmental conditions favor growth in each sex? While these questions have been addressed separately in other studies, here I attend to all four predictions in the same plant species.
MATERIALS AND METHODS
Study system
Salix glauca L. (Salicaceae) is a dominant species within the alpine krummholz habitat that forms the interface between true alpine vegetation and subalpine forest in the Rocky Mountains of western North America. Salix glauca has a circumpolar distribution (Argus, 1973
). Willows hybridize broadly and at the field site, S. glauca may form cryptic hybrids with the co-occurring species S. brachycarpa (Dorn, 1997
). Hybridization has the potential to decouple genetic linkage with sex, making it more difficult to detect adaptive differences between sexes. Salix glauca is dioecious, and over 3 years of observations, I have not observed a single incidence of sex switching in a sample of 100 males and 100 females that I have surveyed repeatedly. Plants have a mixed mode of pollination, relying on both insects and wind for pollen transfer (P. Kevan, University of Guelph, unpublished data). For males and females, shoots produce catkins (clusters of unisexual flowers) solely on the previous year's growth (L. Dudley, personal observation). Males and females did not differ in levels of foliar N (pooled mean ± 1 SE, 2.9 ± 0.08%, N = 55, P > 0.9) and specific leaf area (
= 0.0117 ± 0.0003 g·cm–2, N = 149, P > 0.5 for the difference between morphs). As with other willow species (Keoleian and Volk, 2005
), S. glauca propagates readily through woody rhizomes. In this study, I selected individuals that could be distinguished from one another based on central axis identity and spatial discontinuities among neighboring plants.
The main study population is located at 3620–3890 m a.s.l. on Pennsylvania Mountain (Park County 39°15' N 106°7' W) in the Mosquito Range (Pike National Forest) of central Colorado, USA. At this site, willows are one of the first species to leaf out and flower in early June and one of the last to senesce in late August to early September. Four other sites within the Pike National Forest were used for a survey of sex ratio and spatial segregation: Boreas Pass (39°24' N 105'58° W, 3461–3521 m a.s.l.), Hoosier Ridge (39°22' N 106°3' W, 3515–3624 m a.s.l.), Kite Lake (39°19' N 106°6' W, 3515–3610 m a.s.l.), and Weston Pass (39°8' N 106°11' W, 3572–3641 m a.s.l.).
The krummholz comprises a heterogeneous environment, having standing water, snow banks, rocky outcrops, and fine silty soil within meters of one another. In the present study, mesic habitat patches were designated as those having standing water during most of the growing season with hydrophilic plants such as primrose, marsh marigolds, and moss typically present. Xeric habitat patches were characterized by bare soil intermixed with lichen covered rocks, cushion plants, grasses, and drought-tolerant herbs. Soil cores show a sixfold difference in relative water content between soils of mesic sites (0.6 ± 0.1% w/w) and xeric sites in one sampling during 2002 ( = 0.1 ± 0.02% w/w; t = 2.13, P < 0.01, N = 6; here and elsewhere error brackets show standard errors unless otherwise noted). Over the course of this study, the average predawn leaf water potentials (
l) of willows in xeric sites ( = –0.32 ± 0.01 MPa) were more negative compared to those of plants in mesic sites ( = –0.26 ± 0.01 MPa; F1,380 = 8.43; P < 0.004;). Additionally, ammonium concentration is higher in soil of mesic habitat patches ( = 32.3 ± 1.9 g·g–1) than soil of xeric patches ( = 22.9 ± 1.9 g·g–1; F1,8.49 = 11.42; P < 0.009). Mesic habitat patches also tend to have more acidic soils than xeric patches (water as solvent: respectively, soil pH averages 4.30 ± 0.23, N = 11 and 4.79 ± 0.21, N = 13; F1,22 = 2.49; P = 0.13; with 1 M KCl as solvent, respectively, soil pH averages 3.73 ± 0.08, and 3.96 ± 0.08; N = 7; F1,12 = 4.65; P < 0.05).
This study was conducted from June to September 2001–2003. Precipitation varied dramatically over the course of the experiment with 2002 being the driest of the past 100 years. In 2001 the average precipitation trends in this region were mirrored except for an earlier snowmelt. Although not as dry as 2002, 2003 also was characterized by lower than average precipitation and earlier snowmelt (USDA, 2004
).
Sex-specific resource allocation to reproduction
In 2000, ten 100-m transects were laid out to sample most of the krummholz area on the east slope of Pennsylvania Mountain. In each transect, 10 random points were selected, and the nearest pair of male and female plants to each point was located. Pairs were chosen such that plants of the two sexes were either touching or within 1 m of each other and of approximately the same size based on visual inspection. Plants were permanently tagged. In 2002, catkin nutrient content (N and P) was compared between paired male and female plants. Male catkins in full flower were collected before pollen dehiscence (16–18 June). Female catkins were collected after fruit maturity and before dispersal (21–24 August). Three flowering catkins from each male and five fruiting catkins from each female were randomly chosen, collected in small envelopes, and air-dried with silica gel. Fewer catkins were collected from males to minimize possible effects of clipping early in the growing season on subsequent growth. A random subset of plants from which catkins were collected was included in the analysis of nutrient content (for N, female N = 23, male N = 24; for P female N = 15, male N = 18). Male catkins were ground using a mortar and pestle. For female catkins, seeds were separated from the rest of the catkin material by forceful rubbing through a sieve. The ground and separated material was dried again at 50°C for several days, reweighed, and put into a concentrated acid (H2SO4) for digestion. The digestion was then diluted using DI water to 1 mL from which samples were taken for nitrogen (N) and phosphorous (P) analysis following Ashman and Baker (1992)
. To estimate the average whole plant nutrient budgets for reproduction in the different sexes, catkin nutrient content was scaled up to the whole plant level using mean estimates for catkin mass and shoot number per plant in the population [(g nutrient/g seed or pollen) x (g seed or pollen/catkin) x (catkins/shoot) x (shoots/plant)] with shoot number based on plant circumference (regression, r2 = 0.98 for the relationship between shoot number and circumference).
One-way analysis of variance (ANOVA) with sex as the explanatory variable was used to compare reproductive allocation with respect to N and P at the catkin level (Proc GLM, Statistical Analysis System, version 9; SAS Institute, Cary, North Carolina, USA). P content was angular-transformed to meet model assumptions.
Spatial distribution of the sexes
Frequency and distribution of male and female willows were surveyed in 2004 in five populations (Boreas Pass, Hoosier Ridge, Kite Lake, Pennsylvania Mountain, and Weston Pass). The distance between populations ranged from 9.2 km to 36.2 km. Ten 50-m transects, running north–south, were laid out at 100-m intervals in each population. Ten random points were chosen on each transect, and the sex closest to each point was recorded. I also recorded the sex of the closest neighbor one or more meters away from this plant in a randomly chosen compass direction.
A 
2 goodness of fit test was used to compare the ratio of females to males in each population with an overall expectation of 50 : 50. Sex was compared between nearest neighbors along sampling transects in a frequency analysis so that spatial segregation (Si) of the individual sexes in each of the five populations could be tested (Dixon, 1994
). Si is similar to a log-odds ratio, based on the frequency ratio of each sex. If Si > 0 for a sex, then individuals of that sex are more segregated than expected by chance. If Si < 0, then individuals of the other sex occur more frequently as neighbors than expected at random. If sexes segregate according to microhabitat, then Si should be greater than zero. A Z-score (significance test criterion value) was computed for each population to test whether the observed Si value differs significantly from the value predicted with random intersex spacing.
Sex-specific divergence in physiological traits
On Pennsylvania Mountain, four sites, each consisting of a pair of adjacent mesic and xeric krummholz habitat patches, were chosen in 2001 based upon accessibility and adequate density of S. glauca. A transect was laid through the willows in each habitat patch, and the closest male or female shoot (branch, N = 6 each) to each of 12 random points was located, permanently tagged, and sampled in the summers of 2001 to 2003 (one site was dropped due to loss of permanent tags at the end of the 2001 growing season). When original branches were broken or died, I sampled adjacent branches from the same individual (N = 5 branches replaced).
Gas exchange measures were taken on intact leaves of each shoot using a portable photosynthesis system (Li-6200; Li-Cor, Inc., Lincoln, Nebraska, USA): photosynthesis (A) and conductance (g) were measured, and instantaneous water use efficiency (WUEi) was calculated as the ratio of A/g. Conductance should reflect transpiration, provided that the vapor pressure gradient does not drop excessively during the measurement; to minimize this drop, the measurement was completed quickly. Measurements were taken on fully expanded healthy leaves between 0900 and 1400 MST, during the time of highest incident radiation. For 2001–2003, measurements were made respectively from 30 July to 18 August, 27 to 28 June, and 28 July to 8 August). Only four plants per sex·per patch were sampled in 2002 and 2003 due to time constraints.
Leaf water status was examined for the same plants during the growing seasons of 2001–2003. On 12 occasions, plants did not have any acceptable leaves or could not be located during predawn measurement; this resulted in a slightly reduced sample size (N = 84 total). Leaf water potentials (l) were measured at predawn (0300–0600 hours MST) and mid-day (1100–1300 hours MST), using a plant water status console (3000 series; Soil Moisture Equipment Corp., Santa Barbara, California, USA). In each year, measurements were made respectively from 6 July to 20 August, 22 June to 21 August, and 25 July to 2 August. Leaves chosen for l had petioles of sufficient length to accommodate the chamber gasket and appeared healthy. Leaves with mite domatia, other evidence of herbivory, or discoloration were not used. For 2001–2002, leaves were cut, placed into a moistened paper towel inside of a plastic sandwich bag, placed on ice in a cooler, and then transported to the water status console. In 2003, the pressure chamber was moved to each habitat patch and cut leaves were placed immediately into the chamber. The change in sampling reduced transit effort, but likely had little impact on measurements (see Fig. 2).
|
). Based on these values (consistent for all analyses), a compound symmetry structure was used for g, first-order autoregressive for photosynthetic residuals, and a heterogeneous first-order autoregressive structure for WUEi. The Satterthwaite method was used to estimate degrees of freedom. All subsequent analyses using Proc Mixed followed the same criteria.
Leaf water potentials (l, Mpa) were also analyzed using SAS, Proc Mixed with sex, habitat type, time of day (predawn or mid-day), and year (2001–2003) as fixed effects and site as a random blocking effect. Measurements were repeated on subjects (plants) nested within the site by sex by habitat type interaction. A heterogeneous first-order autoregressive covariance matrix was found as the best model for the variance matrix based on AIC, AICC, and BIC criteria.
Sexual dimorphism in growth rate
To test whether males and females had growth-specific optima in relation to microhabitat differences, permanently tagged shoots (64 original, three deaths, four lost tags, male N = 29, female N = 28) were collected in 2003, and annual shoot growth rates for the preceding 10 years were estimated as incremental distance between terminal bud scale scars along the main axis of the shoot (Fox and Stevens, 1991
). Although some shoots had scars that indicated ages of 30 or more years, most could be aged reliably only to 10 years, so measurements were confined to that interval, reducing variance due to measurement error.
The relationship of flowering catkin production per plant in 2003 to annual shoot growth increment in 2002 and shoot basal diameter in 2003 was examined to address whether factors affecting growth rate and plant size influenced reproductive success for plants of each sex. Annual shoot growth in 2002 provided the structural template for 2003 catkin production. Shoot basal diameter reflects the cumulative resource status of the shoot as influenced by its history of past growth (Ashman, 1994a
); a thicker shoot should have greater access to nutrients through more xylem and phloem conduits. Measurements were made on shoots of a random subsample of the plants that were permanently tagged in 2000 and used for the catkin nutrient analysis. Shoot growth increment over the growing season, defined as the distance between the most distal pair of terminal bud scars, was measured after leaf senescence in 2003 (24 September) for 5% of the shoots per plant. Basal diameter was measured in 2003 just above the ground.
Sources of variation in annual growth rate from 1993 to 2003 were analyzed using Proc Mixed. Sex, habitat type, and year (1993–2003) were fixed effects, site was used as a random blocking effect, and plants represented repeated subjects nested within the site by sex by habitat type interaction. A first-order autoregressive covariance structure was chosen based on AIC, AICC, and BIC criteria.
To ascertain the relationship between growth and reproductive output, 2003 basal diameter, 2002 shoot growth, and sex were used in an additive linear model to explain variation in catkin production in 2003 (SAS, Proc REG). Catkin production was log10-transformed to meet assumptions of regression, and sex was entered as a dummy variable (female = 0, male = 1).
RESULTS
Sex-specific resource allocation to reproduction
N and P concentrations were greater in seeds than pollen. Seeds averaged 0.0020 ± 0.0002 g P·g–1 dry mass, and pollen averaged 0.0008 ± 0.0002 g P·g–1 dry mass (F1,31 = 16.77; P < 0.0003). Similarly, seeds averaged 0.029 ± 0.002 g N·g–1, and pollen averaged 0.015 ± 0.002 g N·g–1 (F1,45 = 18.92; P < 0.001). When scaled to the whole plant, P budgets to reproduction were four times greater for females (0.20 g·plant–1) than males (0.05 g·plant–1). This result is consistent with a higher nutrient cost of reproduction for females vs. males. However, whole plant N budgets did not differ strongly between the sexes; N allocation to reproduction averaged 0.36 g·plant–1 in females and 0.33 g·plant–1 in males.
Spatial distribution of sex
At all five localities, willow populations were strongly female biased (2 = 139.4; P < 0.001; Boreas = 72%, Hoosier = 76%, Kite = 84%, Pennsylvania = 71%, and Weston = 79% female). Males and females had significant spatial segregation in 90% of cases (Table 1). Across populations, sexes had a similar degree of segregation with Si averaging 0.55 ± 0.14 in females and 0.54 ± 0.49 in males. The pronounced spatial segregation of the sexes supports the prediction that the sexes are sorting out in a heterogeneous environment.
|
|
|
Instantaneous WUE showed a significant sex by year interaction (F2,143 = 7.13; P < 0.001; Table 2). In the less extreme years of 2001 and 2003, females and males had similar water use efficiencies (Fig. 1D, F). In 2001, WUEi averaged 57 ± 2 µmol·mol–1 for females and 53 ± 2 µmol·mol–1 for males. In 2003, WUEi averaged 46 ± 4 µmol·mol–1 in females and 45 ± 2 µmol·mol–1 in males. However in the drought year of 2002, females had significantly lower average WUEi than males (respectively, 54 ± 4 µmol·mol–1 and 68 ± 2 µmol·mol–1; P < 0.001) as expected given females' reduced photosynthesis (Fig. 1E).
Leaf water status (l) varied temporally and spatially as indicated by a significant three-way interaction between habitat type, year, and time of day (F2,213 = 7.89; P < 0.0005; Table 3). The difference in mid-day l between habitats grew more pronounced over time from 2001 to 2003 (Fig. 2D–F). There was a trend for plants in xeric sites to have more negative l than plants in mesic sites in 2002 (P = 0.09), but a clear difference was seen in 2003 with xeric sites having a much lower mid-day l (P < 0.0001). For both sexes, l responded similarly to diurnal, spatial, and annual variation in aridity with no significant main or interaction effects for sex (P > 0.1; Table 3).
|
|
DISCUSSION
In this study, I addressed the dimorphic niche hypothesis of ecological causation as an adaptive explanation for secondary sexual dimorphism. I did this by first examining the main assumption of sexual allocation theory—females invest more resources into reproduction than males—in a comparison of nutrient costs for male and female organs. Results show that the validity of this assumption for Salix glauca depends on how reproductive allocation is assessed. Females possessed greater P concentrations in their seeds than males contained in their pollen at both the catkin and whole plant level; however, N concentrations, although greater in seed than pollen, were equivalent between the sexes at the whole plant level. Because nutrients are broadly limiting in the alpine habitat of S. glauca, results support gender-specific reproductive costs. Second, spatial segregation of sexes in five geographically discrete populations is consistent with the prediction of the dimorphic niche hypothesis that sexes should segregate due to environmental heterogeneity. Third, an organism's fundamental niche is defined by its resource use and/or environmental conditions under which it persists in the absence of competitors and natural enemies (Silvertown, 2004
). Females and males of S. glauca have clearly diverged in their fundamental niche with respect to water use. Notably, this discrepancy was only apparent in extreme-drought years when females had reduced photosynthesis compared to males. The fourth prediction asserts that if ecological niches of sex morphs diverge, then growth rates of the two sexes should be maximized under unique environmental conditions. In accordance, there is divergence in incremental growth rates of male and female willows growing in mesic and xeric microhabitats. Female annual growth is 85% higher in mesic sites, while males have a nonsignificant 8% increase in incremental growth over the same moisture gradient. This pattern indicates that female willows have a narrower niche breadth than males over a water gradient.
Unique reproductive ecologies may arise through temporal divergence between sexes in allocation to reproduction. For example, Delph (1990)
found that females of Hebe subalpina were able to grow as much as males despite having greater reproductive investment because females produced more leaves than males early in the growing season before fruit maturation. In S. glauca, seasonal variation in allocation to reproduction between sexes may exacerbate nutrient limitation for females due to temporal flux of nutrients in alpine soils. In the alpine, the timing of nutrient availability, especially in dry habitats, is correlated with snowmelt and consequently with male, but not female reproductive allocation (Bowman et al., 1993). More severe nutrient competition between growth and reproductive organs likely characterizes females, even when, as with N, whole plant allocation to reproduction is similar between sexes. As the soil matrix dries, whether from wet to dry sites or with late summer drought, water is less available to carry nutrients into plant roots, and competition between growth and reproduction may become more severe. For females that allocate resources to maturing fruit throughout the season, nutrient trade-offs between growth and reproduction are likely to be more intense than in males that allocate resources to maturing flowers and pollen only early in the growing season. In S. glauca, growth discrepancies between male and female willows in wet and dry sites conform to this view.
Divergence in allocation between the sexes should be considered when predicting changes in population structure for dioecious species under global change. For example, as anthropogenic activity increases in habitats above tree line, a shift from plant growth limitation by N to P is predicted (Körner, 1999
). My results suggest that this change may have a larger impact on females compared to males because of the greater P allocation to reproduction. For female-biased populations like those of S. glauca, as well as numerous other dioecious plant species (see compilation in Delph, 1999
), such anthropogenic influences could reach beyond the individual plant level to alter population sex ratios, population stability, and even species' distribution.
Studied populations were all highly female-biased. Female dominance may in part arise from other compensatory mechanisms not directly measured in this study. For example, photosynthate demand in females may be partially offset by the photosynthetic bracts on maturing catkins (Bazzaz et al., 1979) or by sink-driven photosynthesis of maturing female catkins (Laporte and Delph, 1996
). Reproductive costs may also be partly offset by translocation of resources from nonreproductive shoots to reproductive shoots within a plant (Banuelos and Obeso, 2004
). If resource translocation occurs among shoots to support fruit set, then physiological, growth, or demographic costs of reproduction may be more evident in herbaceous plants or smaller shrubs than in the relatively large, mature individuals included in this study (e.g., Donovan and Ehleringer, 1991
; Mediavilla and Escudero, 2004
). Measurement of drought tolerance and carbon assimilation in small vegetative plants of S. glauca reveals marked differences in water relations between sexes, suggesting that some or all of these compensatory mechanisms occur at maturity, masking (or reducing) fecundity costs (Dudley, 2006
).
Biotic factors such as herbivory may also play a role in determining sex frequency in dioecious species (Boecklen and Hoffman, 1993
). Damage from insects and ungulate herbivores is prevalent in the alpine krummholz habitat and may favor female willows (Danell et al., 1991
). However, several non-ecological mechanisms may also contribute to sex ratio bias, including prezygotic selection via gamete competition, irregular meiosis, or sibling competition (Stehlik and Barrett, 2005
). From 13 crosses in Salix viminalis, sibships grown under optimum conditions were largely female biased (Alstrom-Rapaport et al., 1997
). Sex ratio asymmetry could also be reinforced by restricted seed dispersal. Because seed dispersal is leptokurtic away from maternal plants, a clumping of offspring around seed parents is expected (Heilbuth et al., 2001
). Hence most seeds should fall into microsites favoring females, promoting a female-biased sex ratio.
Male and female willows have a spatial segregation consistent with their different niche requirements (Fox and Harrison, 1981
; Bierzychudek and Eckhart, 1988
). While spatial segregation may alternatively be due to intermorph competition (Freeman et al., 1976
), few studies have looked at this idea directly, and the two that do argue instead for the dimorphic niche hypothesis (Fox and Harrison, 1981
; Waser, 1984
). For clonal plants such as willows, spatial segregation may also arise from vegetative propagation. Single genets in other Salix species can cover hundreds of meters of stream bank in riparian habitat (Douhovnikoff et al., 2004
). The extent of clonal spread in S. glauca is not clear, but I took precautions to insure that shrubs sampled in my survey represented discrete individuals. The consistent pattern of spatial segregation for both morphs in the five geographically isolated populations sampled in the present study is strong evidence that natural selection has acted on sexes of S. glauca to promote segregation in different microhabitats.
Physiological specialization between the sexes in S. glauca was only evident in the extreme drought environment of 2002, unlike other studies that show upregulation in carbon gain in females under benign conditions (Wheelwright and Logan, 2004
). In 2002, males also had greater instantaneous WUE than females, while maintaining their photosynthetic rates. This finding suggests that males may better tolerate drier environments and extreme drought years, supporting Dawson and Bliss's (1993)
conclusion that male willows have a larger fundamental niche than females. It is noteworthy that 3 years of sampling were necessary to capture the conditions under which sexes of S. glauca diverge in drought tolerance. Studies that report physiological trends based on only 1 year of sampling may miss extreme years, which nonetheless have significant effects on fitness and survival, especially for long-lived, perennial species such as willows.
Smaller, establishing plants are more sensitive to the environment than larger, mature individuals (De Soyza et al., 1996
). Growth rates for female S. glauca were greatest under mesic conditions, but the trend weakened as shoots aged, indicating that selection on physiology in dioecious plants may occur at smaller sizes prior to maturity. In Acer negundo, biased sex ratios are thought to arise prior to the development of large root systems that access groundwater and buffer mature plants from dry periods (Dawson and Ehleringer, 1993
).
Previous shoot growth and current shoot diameter were positively related to catkin production, suggesting underlying growth constraints on fitness for each sex. The influence of the previous year's growth on catkin number probably reflects architectural constraints (Diggle, 1997
). Males and females limit catkin initiation to shoot length laid down in the previous year; a longer shoot has more potential bud sites than a shorter one. This architectural feature provides direct feedback between previous and current reproductive allocation (but see Fox and Stevens, 1991
; Nicotra, 1999
). The relationship between catkin production and shoot basal diameter may reflect the capacity of large shoots, with greater phloem inputs to sustain current reproductive investment (Banuelos and Obeso, 2004
). Although I cannot exclude the hypothesis that growth rate and reproduction are spuriously correlated due to underlying environmental variation (Rausher, 1992
), the finding of a positive relationship between the two processes, coupled with habitat-specific growth optima for male and female S. glauca, supports the idea that spatial segregation of the sexes is adaptive.
Willows are used in a variety of ecosystem management projects: as biomass crops (Volk et al., 2004
), in soil stabilization (DLWC, 2005
), and in riparian restoration (Rood et al., 2003
). Management practices informed by morph differences in niche dimensions could be beneficial. For example, planting females in wet areas and males in drier ones could provide better yield or coverage than a haphazard distribution of the sexes. Overall, this study demonstrates that alpine Salix glauca possesses secondary sexual dimorphism likely as a result of natural selection for exploitation of dimorphic niches in a relatively species sparse community. Whether sexes of riparian willows in lowland habitats have reduced niche partitioning awaits further study.
FOOTNOTES
1 The author thanks C. Galen, D. Bowers, J. Carrell, R. Crocroft, S. Pallardy, R. Semlitsch, M. Brock, A. Dona, J. Geib, and three anonymous reviewers for helpful comments on the manuscript; and M. Carter, J. Huddle, D. Patillo, and J. Uren for assistance in the field. This project was funded by University of Missouri GAANN to L.S.D. and by the National Science Foundation grant DEB-0316110 to C. Galen. 
2 E-mail: dudley{at}lifesci.ucsb.edu
LITERATURE CITED
Ågren J.. 1988. Sexual differences in biomass and nutrient allocation in the dioecious Rubus chamaemorus. Ecology 69: 962-973.[CrossRef][ISI]
Alstrom-Rapaport C. Lascoux M. Gullberg U.. 1997. Sex determination and sex ratio in the dioecious shrub Salix viminalis L. Theoretical Applied Genetics 94: 493-497.[CrossRef]
Argus G. W.. 1973. The genus Salix in Alaska and the Yukon National Museum of Natural Sciences, Ottawa, Canada.
Ashman T.-L.. 1992. Indirect costs of seed production within and between seasons in a gynodioecious species. Oecologia 92: 266-272.[CrossRef][ISI]
Ashman T.-L.. 1994a. A dynamic perspective on the physiological cost of reproduction in plants. American Naturalist 144: 300-316.[CrossRef][ISI]
Ashman T.-L.. 1994b. Reproductive allocation in hermaphrodite and female plants of Sidalcea oregana ssp. spicata (Malvaceae) using four currencies. American Journal of Botany 81: 433-438.[CrossRef][ISI]
Ashman T. L.. 2003. Constraints on the evolution of males and sexual dimorphism: field estimates of genetic architecture of reproductive traits in three populations of gynodioecious Fragaria virginiana. Evolution 57: 2012-2025.[CrossRef][ISI][Medline]
Ashman T.-L. Baker I.. 1992. Variation in floral sex allocation with time of season and currency. Ecology 73: 1237-1243.[CrossRef][ISI]
Banuelos M.-J. Obeso J.-R.. 2004. Resource allocation in the dioecious shrub Rhamnus alpinus: the hidden costs of reproduction. Evolutionary Ecology Research 6: 397-413.
Bawa K. S.. 1980. Evolution of dioecy in flowering plants. Annual Review of Ecology and Systematics 11: 15-40.[CrossRef][ISI]
Bierzychudek P. Eckhart V.. 1988. Spatial segregation of the sexes of dioecious plants. American Naturalist 132: 34-43.[CrossRef][ISI]
Boecklen W. J. Hoffman M. T.. 1993. Sex-biased herbivory in Ephedra trifurca: the importance of sex-by-environment interactions. Oecologia 96: 49-55.[CrossRef][ISI]
Bullock S. H. Bawa K. S.. 1981. Sexual dimorphism and the annual flowering pattern in Jacaratia dolichaula (D. Smith) Woodson (Caricaceae) in a Costa Rican rain forest. Ecology 62: 1494-1504.[CrossRef][ISI]
Charlesworth B. Charlesworth D.. 1978. A model for the evolution of dioecy and gynodioecy. American Naturalist 112: 975-997.[CrossRef][ISI]
Charlesworth D.. 1999. Theories of the evolution of dioecy. In M. A. Geber, T. E. Dawson, L. F. Delph [eds.], Gender and sexual dimorphism in flowering plants 33-60 Springer-Verlag, New York, New York, USA.
Charlesworth D. Charlesworth B.. 1987. The effect of investment in attractive structures on allocation to male and female functions in plants. Evolution 41: 948-968.[CrossRef][ISI]
Charnov E. L.. 1982. The theory of sex allocation Princeton University Press, Princeton, New Jersey, USA.
Cox P. A.. 1981. Niche partitioning between sexes of dioecious plants. American Naturalist 117: 295-307.[CrossRef][ISI]
Danell K. Hjältén J. Ericson L. Elmqvist T.. 1991. Vole feeding on male and female willow shoots along a gradient of plant productivity. Oikos 62: 145-152.[CrossRef][ISI]
Darwin C.. 1871. The descent of man and selection in relation to sex John Murray, London, UK.
Darwin C.. 1874. The descent of man and selection in relation to sex Murray, London, UK.
Dawson T. E. Bliss L. C.. 1989a. Intraspecific variation in the water relations of Salix arctica, an arctic-alpine dwarf willow. Oecologia 79: 322-331.[CrossRef][ISI]
Dawson T. E. Bliss L. C.. 1989b. Patterns of water use and the tissue water relations in the dioecious shrub Salix arctica: the physiological basis for habitat partitioning between the sexes. Oecologia, Berlin 79: 332-343.[CrossRef]
Dawson T. E. Bliss L. C.. 1993. Plants as mosaics: leaf-, ramet-, and gender-level variation in the physiology of the dwarf willow, Salix arctica. Functional Ecology 7: 293-304.[CrossRef][ISI]
Dawson T. E. Ehleringer J. R.. 1993. Gender-specific physiology, carbon isotope discrimination, and habitat distribution in box elder, Acer negundo. Ecology 74: 798-814.[CrossRef][ISI]
Dawson T. E. Geber M. A.. 1999. Sexual dimorphism in physiology and morphology. In M. A. Geber, T. E. Dawson, L. F. Delph [eds.], Gender and sexual dimorphism in flowering plants 175-215 Springer-Verlag, New York, New York, USA.
De Soyza A. Franco A. C. Virginia R. A. Reynolds J. F. Whitford W. G.. 1996. Effects of plant size on photosynthesis and water relations in the desert shrub Prosopis glandulosa (Fabaceae). American Journal of Botany 83: 99-105.[CrossRef][ISI]
Delph L. F.. 1990. Sex-differential resource allocation patterns in the subdioecous shrub Hebe subalpina. Ecology 71: 1342-1351.[CrossRef][ISI]
Delph L. F.. 1999. Sexual dimorphism in life history. In M. A. Geber, T. E. Dawson, L. F. Delph [eds.], Gender and sexual dimorphism in flowering plants 149-173 Springer-Verlag, Berlin, Germany.
Delph L. F. Gehring J. L. Arntz A. M. Levri M. Frey F. M.. 2005. Genetic correlations with floral display lead to sexual dimorphism in the cost of reproduction. American Naturalist 166: S31-S41.
Delph L. F. Gehring J. L. Frey F. M. Arntz A. M. Levri M.. 2004. Genetic constraints on floral evolution in a sexually dimorphic plant revealed by artificial selection. Evolution 58: 1936-1946.[CrossRef][ISI][Medline]
Diggle P. K.. 1997. Ontogenetic contingency and floral morphology: The effects of architecture and resource limitation. International Journal of Plant Sciences 158: S99-S107.[CrossRef][ISI]
Dixon P. M.. 1994. Testing spatial segregation using a nearest-neighbor contingency table. Ecology 75: 1940-1948.[CrossRef][ISI]
DLWC.. 2005. Willows in the hunter: a guide to identification and management New South Wales Department of Land and Water Conservation, Newcastle, Australia.
Donovan L. A. Ehleringer J. R.. 1991. Ecophysiological differences among juvenile and reproductive plants of several woody species. Oecologia 86: 594-597.[CrossRef][ISI]
Dorn R. D.. 1997. Rocky Mountain region willow identification guide. Renewable Resources: R2-RR-97-01.
Douhovnikoff V. McBride J. R. Dodda R. S.. 2004. Salix exigua clonal growth and population dynamics in relation to disturbance regime variation. Ecology 86: 446-452.[ISI]
Dudley L. S.. 2006. Ecological conditions of secondary sexual dimorphism in Salix glauca: fundamental and realized dimorphic niche Ph.D. dissertation, University of Missouri, Columbia, Missouri, USA.
Eppley S. M.. 2006. Females make tough neighbors: sex-specific competitive effects in seedlings of a dioecious grass. Oecologia 146: 549-554.[CrossRef][ISI][Medline]
Espirito-Santo M. M. Madeira B. G. Neves F. S. Faria M. L. Fagundes M. Fernandes G. W.. 2003. Sexual differences in reproductive phenology and their consequences for the demography of Baccharis dracunculifolia (Asteraceae), a dioecious tropical shrub. Annals of Botany (London) 91: 13-19.[CrossRef]
Fox J. F. Harrison A. T.. 1981. Habitat assortment of sexes and water balance in a dioecious grass. Oecologia 49: 233-235.[CrossRef][ISI]
Fox J. F. Stevens G. C.. 1991. Costs of reproduction in a willow: experimental responses vs. natural variation. Ecology 72: 1013-1023.[CrossRef][ISI]
Freeman D. C. Klikoff L. G. Harper K. T.. 1976. Differential resource utilization by the sexes of dioecious plants. Science 193: 597-599.
Geber M. A.. 1999. Theories of the evolution of sexual dimorphism. In M. A. Geber, T. E. Dawson, L. F. Delph [eds.], Gender and sexual dimorphism in flowering plants 95-122 Springer, New York, New York, USA.
Geber M. A. Griffen L. R.. 2003. Inheritance and natural selection on functional traits. International Journal of Plant Sciences 164: S21-S42.[CrossRef]
Gehring J. L. Monson R. K.. 1994. Sexual differences in gas exchange and response to environmental stress in dioecious Silene latifolia (Caryophyllaceae). American Journal of Botany 81: 166-174.[CrossRef][ISI]
Hedrick A. V. Temeles E. J.. 1989. The evolution of sexual dimorphism in animals: hypotheses and tests. Trends in Ecology & Evolution 4: 136-138.[CrossRef][ISI]
Heilbuth J. C. Ilves K. L. Otto S. P.. 2001. The consequences of dioecy for seed dispersal: modeling the seed-shadow handicap. Evolution 55: 880-888.[CrossRef][ISI][Medline]
Keoleian G. A. Volk T. A.. 2005. Renewable energy from willow biomass crops: life cycle energy, environmental, and economic performance. Critical Reviews in Plant Sciences 24: 385-406.[CrossRef][ISI]
Körner C.. 1999. Alpine plant life: functional plant ecology of high mountain ecosystems Springer-Verlag, New York, New York, USA.
Laporte M. M. Delph L. F.. 1996. Sex-specific physiology and source-sink relations in the dioecious plant Silene latifolia. Oecologia 106: 63-72.[ISI]
Littell R. C. Milliken G. A. Stroup W. W. Wolfinger R. D.. 1996. SAS system for mixed models SAS Institute, Cary, North Carolina, USA.
Mediavilla S. Escudero A.. 2004. Stomatal responses to drought of mature trees and seedlings of two co-occurring Mediterranean oaks. Forest Ecology and Management 187: 281-294.[CrossRef][ISI]
Nicotra A. B.. 1999. Reproductive allocation and the long-term costs of reproduction in Siparuna grandiflora, a dioecious neotropical shrub. Journal of Ecology 87: 138-149.[CrossRef]
Obeso J. R.. 2002. The costs of reproduction in plants. New Phytologist 155: 321-348.[CrossRef][ISI]
Poorter H.. 1989. Interspecific variation in relative growth rate on ecological causes and physiological consequences. In H. Lambers, M. L. Cambridge, H. Konings, T. L. Pons [eds.], Causes and consequences of variation in growth rate and productivity of higher plants SPB Academic Publishing, The Hague, Netherlands.
Ralls K.. 1976. Mammals in which females are larger than males. Quarterly Review of Biology 51: 245-276.[CrossRef][Medline]
Rausher M. D.. 1992. The measurement of selection on quantitative traits biases due to environmental covariances between traits and fitness. Evolution 46: 616-626.[CrossRef][ISI]
Renner S. S. Ricklefs R. E.. 1995. Dioecy and its correlates in the flowering plants. American Journal of Botany 82: 596-606.[CrossRef][ISI]
Rood S. B. Gourley C. R. Ammon E. M. Heki L. G. Klotz J. R. Morrison M. L. Mosley D. Scoppettone G. G. Swanson S. Wagner P. L.. 2003. Flows for floodplain forests: a successful riparian restoration. Bioscience 53: 647-656.[CrossRef][ISI]
Shine R.. 1989. Ecological causes for the evolution of sexual dimorphism: a review of the evidence. Quarterly Review of Biology 64: 419-461.[CrossRef][Medline]
Silvertown J.. 2004. Plant coexistence and the niche. Trends in Ecology & Evolution 19: 605-611.[CrossRef][ISI]
Slatkin M.. 1984. Ecological causes of sexual dimorphism. Evolution 38: 622-630.[CrossRef][ISI]
Stanton M. L. Galloway L. F.. 1990. Natural selection and allocation to reproduction in flowering plants. In M. Mangel [ed.], Sex allocation and sex change: experiments and models 1-50 American Mathematical Society, Providence, Rhode Island, USA.
Stehlik I. Barrett S. C. H.. 2005. Mechanisms governing sex-ratio variation in dioecious Rumex nivalis. Evolution 59: 814-825.[ISI][Medline]
USDA.. 2004. Hoosier Pass SNOTEL water year Natural Resources Conservation Service National Water and Climate Center. Website http://www.wcc.nrcs.usda.gov/cgibin/wygraph-multi.pl? [accessed 8 December 2005].
Volk T. A. Verwijst T. Tharakan P. J. Abrahamson L. P. White E. H.. 2004. Growing fuel: a sustainability assessment of willow biomass crops. Frontiers in Ecology and Environment 2: 411-418.
Waser N. M.. 1984. Sex ratio variation in populations of a dioecious desert perennial, Simmondsia chinensis. Oikos 42: 343-348.[CrossRef][ISI]
Wheelwright N. T. Logan B. A.. 2004. Previous-year reproduction reduces photosynthetic capacity and slows lifetime growth in females of a neotropical tree. Proceedings of the National Academy of Sciences, USA 101: 8051-8055.
Willson M. F.. 1990. Sexual selection in plants and animals. Trends in Ecology & Evolution 5: 210-214.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
= 0.05
