| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
0 Department of Ecology, Evolution, and Natural Resources, Rutgers University, New Brunswick, New Jersey 08901-1582 USA
Received for publication September 9, 1999. Accepted for publication February 1, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Amaranthaceae • Amaranthus cannabinus • dioecy • flowering phenology • freshwater marsh • salt marsh • sex expression • sex-specific selection
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) and as high as 10% (Lloyd, 1982
). A more recent survey of 240 000 angiosperm species by Renner and Ricklefs (1995)
found that 6% were dioecious. Many ecological (e.g., Bawa and Opler, 1975
; Bawa, 1980
; Freeman, Harper, and Ostler, 1980
; Muenchow, 1987
; Bawa, 1994
; Renner and Ricklefs, 1995
, and references therein) and genetic (Bawa and Opler, 1975
; Thomson and Barrett, 1981
; Lloyd, 1982
; Charlesworth and Charlesworth, 1987
) factors have been suggested as likely to lead to the development of the dioecious condition. Since dioecy has evolved in such a variety of plant groups and under such a wide range of habitats, Baker (1984)
and Anderson and Stebbins (1984)
have recommended that each case of dioecy be investigated individually.
Many of the current ideas on the role of specialization in the evolution of dioecy can be traced back to Darwin (1877)
. Dioecy results in a division of labor between the sexes and may eventually lead to greater reproductive efficiency (Lloyd, 1982
). Male and female reproductive functions may be optimized through different selective pressures (Lloyd and Webb, 1977
; Willson, 1979
). This sex-specific selection may lead to phenological differences and dimorphism in male and female plants (Cox, 1981
; Meagher, 1984
). Sex-specific selection may also lead to differential resource use by the sexes, sometimes called "sexual niche partitioning." Differential resource utilization may be spatial, where the sexes use different microhabitats, or temporal, where the sexes use the same resources but at different times.
Spatial segregation of the sexes can be a result of (1) differential germination or survival of the sexes or (2) sex expression lability, whereby an individual alters its sex expression in relation to its environment (Freeman, Klikoff, and Harper, 1976
; Freeman, Harper, and Charnov, 1980
; Bierzychudek and Eckhart, 1988
). There are many reported cases of spatial segregation of males and females of dioecious species along environmental gradients (Freeman, Klikoff, and Harper, 1976
; Grant and Mitton, 1979
; Freeman, Harper, and Charnov, 1980
; Bierzychudek and Eckhart, 1988
; Sakai and Weller, 1991
; Dawson and Ehleringer, 1993
). The general trend arising from these studies is that males are typically more abundant than females in more stressful environments, such as areas with lower soil moisture. An unequal sex ratio in a population may result from sexual niche partitioning and the differential availability of microsites for males and females (Putwain and Harper, 1972
; Meagher, 1981
). Deviations from a 1:1 sex ratio may also be caused by differences between the sexes in germination requirements (Purrington, 1993
; Lyons, Shah-Mahoney, and Lombard, 1995
), survival (Lloyd and Webb, 1977
; Krischik and Denno, 1990a
), competitive abilities (Freeman, Klikoff, and Harper, 1976
; Cox, 1981
; Ågren, 1988
), and flowering phenology (Conn, 1981
; Conn and Blum, 1981
; Purrington, 1993
).
Amaranthus cannabinus (L.) Sauer (water hemp) is a dioecious annual that grows in freshwater, brackish, and saltwater marshes along the eastern coast of the United States (Sauer, 1955
). As a rapidly growing, wind-pollinated annual, it is an ideal species to investigate sex specific selection, sex ratios, and stability of sex expression. Furthermore, because A. cannabinus grows in fresh and salt marsh habitats, the effects of salinity on sex expression and sex-specific selection can be investigated. Our first objective was to investigate the sex ratios, stability of sex expression, spatial distribution of the sexes, allocation strategies, and phenologies of the sexes in freshwater and saltwater populations of A. cannabinus.
A second objective was to examine the effects of salinity on vegetative and reproductive development in the sexes by growing plants from both fresh and salt marsh habitats in the greenhouse at zero, low, and high levels of salinity. With respect to flowering phenology, we had two basic hypotheses. The first hypothesis was that phenology of the sexes in the greenhouse would be the same as that observed in the field for each population, regardless of salinity level. If this null hypothesis (H0) were not rejected, it would indicate that there is a strong genetic component for phenology in the populations of A. cannabinus. Alternatively, flowering phenology may be influenced by salinity level for both populations. Failure to reject this hypothesis (HA) may be evidence of physiological differences between the sexes. Phenological differences between the sexes could also be a result of sex-specific germination rates. To explore this possibility, germination rates for males and females were examined in the greenhouse experiment.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; M. R. Bram, personal observation). Its habitat range extends along the eastern coast of the United States from Maine to Florida. The genus Amaranthus contains 
90 species (Clemants, 1992
), most of which are monoecious. Ten species, including A. cannabinus, are dioecious (Sauer, 1955, 1972
). Water hemp is wind-pollinated, and its seeds are dispersed by wind and water. Apomixis and clonal growth do not occur in A. cannabinus (Bram, 1998
).
Field observations
Study sites
Three populations in southern New Jersey were studied during the summers of 1994 and 1995. The Hamilton/Trenton Marsh (Mercer County) is a freshwater tidal wetland located between Trenton, New Jersey and Bordentown, New Jersey. This site will be referred to as the Freshwater-1 (FW-1) site. The second freshwater site (FW-2) was located in Mount Holly, New Jersey at the Rancocas Nature Center (Burlington County). The study population grew along the banks of the Rancocas Creek, a tidal tributary of the Delaware River. A diverse group of annuals and perennials occurred at both freshwater sites. The third site was a tidal saltwater marsh in Eldora, New Jersey (Cape May County). This saltwater marsh is along New Jersey Route 47, and the creek running through the marsh is the East Creek. This site is dominated by Spartina alterniflora, a perennial grass, and will be referred to as the Saltmarsh (SM) site.
Flowering sex ratios and spatial distribution of the sexes
In the summers of 1994 and 1995, transects were established at three elevations in the marsh at FW-1 and FW-2 and at four elevations at the SM site. Each transect ran parallel to the water's edge. Transects were established in the low (closest to the creek), middle, and high marsh at each site; there were two middle-marsh transects at SM. At the FW-2 and the SM sites, transects were 15 m in length, while at the FW-1 site transects were 7.5 m long due to marsh topography. The transects were evenly spaced at each site in order to obtain representative samples of each population; there were 6.5 m between transects at the saltmarsh and 1–1.5 m between transects at the two freshwater sites. Due to the meandering nature of the creek at the FW-1 site, low, middle, and high marsh transects were established at each of three different areas in the marsh to provide a comparable sample size (22.5 m for each elevation).
Along each transect, a 50 x 50 cm quadrat was used to census plants every 2 m. The number of flowering male and female plants within each quadrat was recorded. In 1994, FW-1 and SM were each sampled twice, and FW-2 was sampled once. In 1995, FW-1 was sampled four times, and FW-2 and SM were each sampled three times.
Each population was analyzed separately. Deviations from a 1:1 ratio of male and female plants for each sampling date were tested using chi-square. To test for homogeneity of the distribution of males and females at each site, quadrat counts were pooled along each transect (low, middle, and high marsh transects). Next, the PROC FREQ procedure of the 6.12 version of SAS for Windows 95 (SAS, 1996
) was used to calculate either the chi-square statistic or Fisher's exact test statistic using a two-way cross-tabulation table for transect x sex (SAS, 1990
). Fisher's exact test was used when any expected value of the chi-square table was less than five.
Sexual dimorphism and stability of sex expression
In order to investigate sex-specific differences in vegetative and reproductive characters, plants were tagged early in the season (pre-reproduction) at both FW-1 (80 plants) and SM (75 plants) in 1994. These tagged plants were examined throughout the season to determine the period of flower initiation, sex expression, and constancy of sex expression. In addition, the following traits were quantified: plant height (measured three times during the season for FW-1 plants and four times for SM plants), the number of branches (measured twice during the season), and the relative number of leaves (measured twice during the season). Actual leaf counts for each plant were not practical due to the large numbers of leaves per plant. Plants were instead given a ranked score of 1 (fewest leaves), 2 (intermediate), or 3 (most leaves). In 1995, 125 plants were tagged at each of the three sites. These plants were followed throughout the season to determine the period of flower initiation and of mortality of the sexes.
Each population was analyzed separately. Repeated-measures analysis of variance (PROC GLM, SAS) was used to test for differences between males and females for plant height and number of branches. Differences in the relative number of leaves between the sexes were tested with the Wilcoxon rank-sum test (PROC NPAR1WAY, SAS). The PROC FREQ procedure of SAS was used to calculate the chi-square statistic or Fisher's exact test to analyze differences in flower initiation and mortality rates between males and females.
Salinity experiment
Seed source
Seeds were collected in the late summer and early autumn of 1995 from two of the New Jersey populations that were studied during the summers of 1994 and 1995, Freshwater-1 (FW-1) and Saltmarsh (SM).
Cold treatment and germination
Seeds were placed in closed, plastic sandwich boxes (11 x 11 x 3.5 cm) on moist, Rochester blue-gray blotter paper; each box contained ten or fewer seeds. All sandwich boxes were placed in a cold room (4°C) on 3 March 1996 for 2 mo. On 10 May 1996, the sandwich boxes were placed in a germinator with alternating temperatures of 20°C (1800 to 0900) and 30°C (0900 to 1800) and alternating periods of light (0600 to 1800) and dark (1800 to 0600). Seeds were checked daily until the majority of seeds had germinated in each population (2 wk). Seeds with both a radicle and cotyledons were considered fully germinated.
Greenhouse conditions
All plants were grown in the greenhouses at Nelson Biological Laboratories, at Rutgers, in Piscataway, New Jersey. Fully germinated seedlings were planted in standard 10-cm plastic pots filled with a 1:1 mixture of ProMix and New Jersey Piedmont loam. The pots were placed in plastic seedling flats that were kept filled with water or saline solution (see below) so that the soil within the pots was always saturated. Flats of plants were arranged on the greenhouse benches so that each population by treatment combination (see below) was equally represented within each block of the greenhouse benches. On 20 June 1996, 4–5 wk after planting, all plants were placed in flats containing water and given 10 mL of 25–10–10 Miracle-Gro fertilizer.
Salinity treatments
Within each population, plants were randomly assigned to one of three treatment groups: ZERO salinity (0 ppt), LOW salinity (10 ppt), or HIGH salinity (20 ppt). These salinity levels are within the range of salinity experienced by A. cannabinus at the study populations (Bram, 1998
). Instant Ocean Synthetic Sea Salt (Aquarium Systems, Mentor, Ohio, USA), a nitrate-free and phosphate-free synthetic sea salt, was used to make the LOW and HIGH saline solutions. Plants assigned to the ZERO salinity group received water without Instant Ocean. Salinity treatments were begun 2 d after the last seedlings were planted. To reduce osmotic shock, salinity levels were gradually increased over 3 d for the LOW salinity group and over 6 d for the HIGH salinity group. Water or saline solutions were added to the plastic flats that contained the pots of plants (i.e., the water and solutions were not added to the soil surface). All flats were emptied each week and filled with fresh saline solution or water.
There were 50 seedlings for each treatment (25 from FW-1 and 25 from SM) at the start of the experiment. Due to the intense heat (sometimes as high as 32°C) and light (no shading compound on the glass) in the greenhouse, the plants in the LOW and HIGH salinity treatments showed signs of salinity stress (wilting and leaves drying from the tips) by the first week in June, 2–3 wk after planting. Therefore, to minimize severe salinity stress, the treatments were modified to the following: (1) all plants were flushed with water (water was added to the soil surface) and water was added to all plastic flats for 3 d, (2) on the fourth day, all flats were emptied and one-half strength saline solutions were added to the flats of the LOW and HIGH treatments; fresh water was added to the ZERO treatment flats, (3) for the next 3 d, plants received full-strength saline solutions in the LOW and HIGH treatments; water was added to the ZERO treatment flats, and (4) steps 1–3 were repeated each week. The experiment was terminated at senescence on 9 August 1996.
Characters measured/calculated
The following characters were studied: the number of days to germinate, the number of days from germination to flower initiation, constancy of sex expression, plant height (measured five times during the experiment), the number of leaves (counted three times during the experiment), the length and width of the largest leaf (measured once, approximately midway through the experiment), and aboveground biomass at harvest. Harvested plants were dried in an oven for 48 h at 80°C, and aboveground biomass was measured using a Mettler PE 600 balance.
All analyses were done using the 6.12 version of SAS for Windows 95 (SAS, 1996
). Differences between the sexes in the number of days to germinate were tested with the Wilcoxon rank-sum test (PROC NPAR1WAY, SAS) because this variable was not normally distributed. Repeated-measures analysis of variance (PROC GLM, SAS) was used to analyze plant height and the number of leaves. The number of days to initiate flowering, the length and width of the largest leaf, and final biomass were analyzed by multiple analysis of variance (MANOVA, PROC GLM, SAS). Tukey's studentized range (HSD) test (PROC GLM, SAS) was used for post hoc multiple comparisons when appropriate. T tests were performed to test for within-population differences between the sexes in the timing of flower initiation (PROC TTEST, SAS).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) and, therefore, are not presented here. As in the previous year, the sex ratio at FW-1 did not differ significantly from 1:1 until the end of the season when there were significantly more females flowering than males (P < 0.001; Table 1). No significant deviations from a 1:1 sex ratio were found in 1995 at FW-2. At SM, there was a significant male bias among flowering plants in the beginning of August 1995 (P < 0.001), as there had been in the previous summer. On 24 August 1995, however, the sex ratio at SM was almost exactly 1:1. By 3 September 1995, there was a significant female bias at SM (P < 0.001).
|
|
= 0.7888, P = 0.0054; Fig. 1A). The growth rate was nearly identical for males and females between 21 July 1994 and 4 August 1994. After 4 August 1994, however, the increase in height was greater in females than in males (P = 0.0021 for difference between second and third measurements, repeated-measures ANOVA of contrast variables).
|
= 0.8057, P = 0.0061; Fig. 1B). Between 19 July 1994 and 1 August 1994, growth rate was greater in males than females (P = 0.0393 for difference between first and second measurements, repeated-measures ANOVA of contrast variables). After 1 August 1994, females continued to grow at approximately the same rate until the third measurement, while male growth rate began to decline (Fig. 1B). Both sexes had reached their maximum height by 18 August 1994.
Females had significantly more branches than males at FW-1 (Wilks' = 0.7280, P = 0.0002; Table 3). The mean number of branches did not change significantly for plants between 29 July 1994 and 23 August 1994, as indicated by the nonsignificant effect of time in the repeated-measures ANOVA (Wilks' = 0.9929, P = 0.5778). Females at FW-1 also had more leaves on average than males on both 29 July 1994 and 23 August 1994 (Z = 3.945, P = 0.0001, and Z = -5.287, P = 0.0001, respectively).
|
= 0.8446, P = 0.0017). By 1 August 1994, males had stopped producing branches while females continued to develop new ones (Table 3). Females at SM also had more leaves on average than males, but, again, this difference was not evident until the end of the season (Z = -1.565, P = 0.1176 for 1 August 1994, and Z = -3.675, P = 0.0002 for 2 September 1994). Figure 2 shows the cumulative percentage of tagged plants that had begun flowering by each sampling date in 1994. At the FW-1 site, 67 of the 80 plants survived to flower; out of the 67 survivors, 35 were males and 32 were females. Initiation of flowering was nearly identical for males and females at FW-1 in 1994 (P = 1.000, Fisher's exact test; Fig. 2A). At the SM site, 72 of the 75 plants survived to flower; there were 35 males and 37 females. By 1 August 1994, 80% of the males had begun flowering while only 60% of the females had initiated flowering (Fig. 2B). Fisher's exact test on the number of males and females that had begun flowering by each sampling date was not significant (P = 0.175), possibly due to the relatively small sample size.
|
). Out of 125 plants tagged at each site in 1995, the number of plants that survived to flower was 33 at FW-1, 73 at FW-2, and 88 at SM. The drought in New Jersey during this summer probably contributed to the high mortality. The FW-1 site was not used to investigate flowering phenology or time of mortality of males and females due to the small sample size at this site. The period of flower initiation did not differ between males and females at FW-2 (P = 0.745, Fisher's exact test) or at SM in 1995 (P = 0.568, Fisher's exact test). Hurricane Felix hit the New Jersey coast in mid-August, preventing data collection on flowering phenology at the SM site on 19 August 1995. As a result, the third sampling period extended from 3 August 1995 to 24 August 1995, possibly preventing detection of differences between the sexes in flowering phenology.
Figure 3 shows the cumulative percentage of tagged plants that had died by each sampling date at SM. The mortality rate of males was significantly higher than that of females (P < 0.0001, Fisher's exact test). By 22 September 1995, 80% of the tagged males had died, while only 27% of the tagged females had died. Mortality rates did not differ for males and females at FW-2 (P = 0.418, Fisher's exact test).
|
Salinity experiment
Germination rate, sex expression, and leaf number
Germination rates did not differ between the sexes (Z = 0.5287, P = 0.5955). Sex expression remained constant throughout the experiment for the plants that survived to flower.
Sex (SEX) had an overall significant effect on the mean number of leaves (Wilks' = 0.8448, P = 0.0001). For the first two sample dates (3 and 6 wk after planting), the average number of leaves did not differ significantly between the sexes. By the third sampling date (8 wk after planting), however, females had more leaves on average than males (P = 0.0001; female mean = 47.10 leaves, SD = 19.18; male mean = 33.13 leaves, SD = 14.65).
There was a significant treatment by population (TRT x POP) interaction for the mean number of leaves (Wilks' = 0.9303, P = 0.0440; Fig. 4A). In FW-1 plants, leaf number was decreased by both LOW and HIGH salinity treatments with respect to the ZERO treatment. In contrast, leaf number in SM plants was similar for the ZERO and LOW treatments but was dramatically decreased by the HIGH treatment. There was also a significant interaction between population and sex (POP x SEX) (Wilks' = 0.9239, P = 0.0048; Fig. 4B). Males from both populations peaked in leaf number by the second sampling date (24 June), while females continued to produce more leaves (Fig. 4B). The mean number of leaves increased at a greater rate for FW-1 females than for SM females (Fig. 4B).
|
= 0.5038, P = 0.0001; Wilks' = 0.3175, P = 0.0001; and Wilks' = 0.5055, P = 0.0001, respectively). Additionally, all possible interactions of the three main effects were significant: TRT x POP (Wilks' = 0.8813, P = 0.0302), TRT x SEX (Wilks' = 0.8864, P = 0.0393), POP x SEX (Wilks' = 0.6715, P = 0.0001), and TRT x POP x SEX (Wilks' = 0.8117, P = 0.0005). Figure 5 shows mean plant height for both sexes at all TRT x POP combinations. As salinity increased, plant height decreased within a particular sex and population, and plants from SM were taller than plants from FW-1 by the fourth height measurement (3 July; Fig. 5). SM plants showed a greater plasticity in plant height across treatments than did FW-1 plants (Fig. 5). By the third height measurement (18 June), males were taller than females. Between the fourth and final height measurements, the growth rate in height had almost stopped for FW-1 males (solid lines, closed symbols), while for FW-1 females (solid lines, open symbols) and both sexes of SM plants (dashed lines), height was still increasing (Fig. 5). For both populations, female growth rate in height was greater than that of males after the fourth measurement.
|
= 0.6003, P = 0.0001; Table 4). As salinity increased, leaf length, leaf width, and biomass decreased; all three treatments were statistically different from each other. The number of days to initiate flowering increased as salinity increased, but only the ZERO and HIGH treatments were statistically different (Table 4). Population also significantly affected leaf size, timing of flower initiation, and biomass (Wilks' = 0.0909, P = 0.0001; Table 5). SM plants had longer, thinner leaves and greater aboveground biomass than did FW-1 plants (Table 5). FW-1 plants began flowering earlier than those from the SM population (Table 5). There was an overall significant effect of SEX (Wilks' = 0.6896, P = 0.0001), which was primarily due to leaf width and final biomass; females had wider leaves than males (female mean = 2.31 cm, SD = 0.71; male mean = 2.09 cm, SD = 0.60) and greater final biomass than males (female mean = 3.05 g, SD = 0.97; male mean = 2.40 g, SD = 0.75). Leaf width in females was more negatively affected by increasing salinity than was leaf width in males, as evidenced by a significant TRT x SEX interaction (Wilks' = 0.8875, P = 0.0418).
|
|
= 0.9234, P = 0.0305) was solely due to the timing of flower initiation (Fig. 6). Males and females did not differ significantly in flower initiation for the FW-1 population (P = 0.6912, t test; Fig. 6). However, the SM males flowered significantly earlier than the SM females (P = 0.0043, t test; Fig. 6). Females of both populations varied more in the timing of flower initiation than did males (SM: female s2 = 78.85, male s2 = 31.18; FW-1: female s2 = 44.66, male s2 = 21.78). Table 6 contains the means and variances for flower initiation for males and females of each TRT x POP combination. In the SM plants, the variance for flower initiation in females increased as salinity increased, while male variance in flower initiation changed little across treatments (Table 6). In the FW-1 plants, the females varied less in flower initiation across treatments than the males did; male variance actually decreased as salinity increased (Table 6).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Adult sex ratios of A. cannabinus did not differ significantly from 1:1 in the field. Bram (1998)
also found adult sex ratios of water hemp to be 1:1 in the greenhouse. The observed temporal deviations from a 1:1 sex ratio in the field can be attributed to differences in flowering phenology and differences in the timing of mortality between the sexes. At the salt marsh population (SM), the majority of male plants began flowering before most female plants, leading to an early male-biased sex ratio among flowering individuals. By the middle of the flowering period, however, the majority of females had initiated flowering, and the sex ratio was almost exactly 1:1. The sex ratio became female-biased at the end of the season as male plants senesced. At FW-1, one of the freshwater populations, the flowering sex ratio did not deviate from 1:1 until the middle or end of the season when it became female-biased. Similar to the SM site, the female bias was probably due to the earlier death of males. No deviations from a 1:1 sex ratio were observed at the second freshwater population, FW-2, by the final sampling date, 25 August 1995. However, the female bias among flowering individuals at both FW-1 and SM was not observed until the first week in September 1995. Therefore, if the sampling period had been extended at FW-2, it is possible that a similar, significant female bias among flowering plants would have been observed at the end of the season.
These population-specific patterns of flowering phenology for SM and FW-1 were also observed in the salinity experiment. Males and females from the freshwater population initiated flowering at the same time while males from the salt marsh population began flowering significantly earlier than females from the salt marsh, regardless of salinity level. Salinity did delay the onset of flowering in the greenhouse, but increases in salinity did not lead to a statistically significant delay in female flowering. Therefore, the population-specific patterns of flowering phenology are due to genetic differences between the populations. Because germination rates did not differ between the sexes in the greenhouse, phenological differences between males and females are due to differential growth and development and are not simply the result of sex-specific germination rates.
The timing of flower initiation was more variable in females than in males. Furthermore, as salinity increased, the variance in flower initiation for SM females increased by an order of magnitude. This greater variability in females may explain the differences in flowering phenology between the sexes at SM. In the salt marsh, males may be less variable in flower initiation, so that most males begin flowering at approximately the same time, leading to a male bias among flowering individuals. Some females may flower early, but it is not until almost two weeks after most males have initiated flowering that enough females have begun flowering to yield a 1:1 sex ratio among flowering plants. Salt marsh females may show greater plasticity in the timing of flower initiation because plants growing in a tidal salt marsh experience fluctuations in salinity levels throughout the season, month, and even the day. Historically, plants at the SM site would have experienced more intense selection for plasticity in response to salinity level than plants at the freshwater marshes. In the greenhouse, plants from the SM population also exhibited greater plasticity in plant height across salinity treatments than did plants from the FW-1 population.
Perhaps the general lack of phenological differences between the sexes at the freshwater sites is partly due to habitat structure and species composition. While tidal inundation is common to both the salt marsh and the freshwater marshes, the two habitats are substantially different. Freshwater marshes of the eastern coast of the United States support a large number of annuals (Simpson et al., 1983
; Parker and Leck, 1985
), while salt marshes of this area are typically dominated by a few perennials. Spartina alterniflora, a perennial grass, is the most abundant plant species in the SM site. Most water hemp plants at SM reach a height of 1–1.5 m, which is about the same height as the S. alterniflora. In sharp contrast to this, A. cannabinus in the freshwater marshes (FW-1 and FW-2) typically reaches a height of 2–2.5 m. Surrounding annual vegetation in both freshwater marshes, such as Zizania aquatica and Ambrosia trifida, can reach heights of over 3 m. Intense competition among the many fast-growing annuals at the freshwater marshes may select against delayed female growth into the canopy; a delayed growth spurt of even 2 wk could be detrimental.
Many investigators have reported that males of dioecious species flower significantly earlier than females (Putwain and Harper, 1972
; Lloyd and Webb, 1977
; Conn and Blum, 1981
; Meagher, 1981
; Miller and Lovett Doust, 1987
; Krischik and Denno, 1990b
; García and Antor, 1995
; Delph, 1999
, and references therein). This difference in phenology may be closely tied to differences in resource allocation between the sexes (Lloyd and Webb, 1977
; Delph, 1999
). Male reproductive function ends with pollen dispersal, while female reproductive activities continue through the maturation of fruit. Due to this disparity, females may flower only after accumulating enough resources through vegetative growth to support the prolonged reproductive period (Lloyd and Webb, 1977
; Gross and Soule, 1981
; Purrington and Schmitt, 1998
; Delph, 1999
). Indeed, females of many dioecious species are larger or produce more leaves than males (e.g., Gross and Soule, 1981
, and references therein; Lovett Doust and Laporte, 1991
; Korpelainen, 1992, 1994
; Freeman et al., 1993
; García and Antor, 1995
). In A. cannabinus, the resource demand during seed maturation is probably very high. Each female is capable of producing thousands of seeds; females grown in the greenhouse produced an average of 3500 seeds per plant (J. A. Quinn, M. R. Bram, and T. E. Taylor, Rutgers University, New Brunswick, New Jersey, unpublished data).
In the field, A. cannabinus females produced more leaves and branches than males. These vegetative differences between males and females were evident very early in the season at FW-1 and were statistically significant at SM by the end of the season. Additionally, at both of these sites, male growth in height slowed by the beginning of August while females continued to grow taller. These general vegetative differences between the sexes were also observed in the greenhouse salinity experiment. Female plants in the greenhouse had more leaves, wider leaves, and greater final biomass than male plants. A prolonged female growth period was also observed in the greenhouse for leaf production and plant height. Lemen (1980)
studied three dioecious amaranths (not including A. cannabinus) and also found that females were consistently larger than males.
Males, however, were significantly taller than females in the field at the salt marsh and in the greenhouse. Other investigators have reported taller males in dioecious plants (Lloyd and Webb, 1977
; Quinn, 1991
; Quinn et al., 1994
). Many investigators (e.g., Lloyd and Webb, 1977
; Freeman et al., 1997
, and references therein; Dawson and Geber, 1999
) have suggested that increased height in males may aid in pollen dispersal in wind-pollinated species. The different reproductive roles of males and females may also explain the numerous observations that males often become senescent or die before females (Lloyd and Webb, 1977
; Conn and Blum, 1981
; Meagher and Antonovics, 1982
; Miller and Lovett Doust, 1987
). In the field, the mortality rate of tagged males at the SM site was significantly greater than that of females. At the FW-2 population, no significant difference in mortality rates between the sexes was found, although, again, this may simply be due to the shorter sample period at this site (see above).
There was no evidence for differential spatial distribution of the sexes at any of the three field sites. Detection of spatial segregation of the sexes may depend upon the scale employed experimentally in the field. Meagher (1980)
found different patterns of spatial segregation of males and females to be detectable at different scales in Chamaelirium luteum. Due to the very high densities of A. cannabinus in the field, it was impractical to do a nearest neighbor analysis to look for spatial segregation on a smaller scale. Since sex expression in A. cannabinus was not observed to be labile in the field or in the salinity greenhouse experiment (but see Bram, 1998
), any differential spatial distribution of the sexes would have to be a result of either differential germination or differential mortality. In the current study, germination rates did not differ between the sexes in the greenhouse experiment, and there was no evidence of differential mortality between prereproductive males and females.
The findings of this study of A. cannabinus are strikingly similar to the studies of the annual Rumex hastatulus by Conn (1981)
and Conn and Blum (1981)
. In R. hastatulus, males were initially taller, flowered earlier, and senesced earlier than females (Conn, 1981
; Conn and Blum, 1981
). Some authors (e.g., Putwain and Harper, 1972
; Cox, 1981
) have interpreted similar phenological differences between the sexes of dioecious species as evidence of temporal niche partitioning between the sexes. Conn (1981)
and Conn and Blum (1981)
, however, rejected this interpretation because the canopy dominance by males in R. hastatulus was very transitory, lasting only 1–2 wk. They suggested that the phenological differences between the sexes were most likely a result of disruptive selection due to the different reproductive roles of males and females. Given the similarly transitory nature of the phenological differences observed in A. cannabinus plants in the salt marsh, it seems unlikely that the observed phenological differences are evidence of temporal niche partitioning between the sexes. Instead, the phenological differences between males and females are more likely due to sex-specific selection for optimal reproductive success.
Increased reproductive efficiency through sex-specific growth patterns may have been an important selective factor involved in the evolution of dioecy in A. cannabinus. Taller males with fewer branches and leaves may be more successful at pollen dispersal, and females that grow over a longer period and accumulate more resources may be better seed parents. While the optimization of male and female reproductive functions is an obvious advantage of dioecy, it is not certain whether it was a selective drive for the evolution of dioecy or whether sex-specific differences developed later, as a result of the dioecious habit (Meagher, 1980
; Sakai and Weller, 1991
). Another factor that may have contributed to the evolution of dioecy in A. cannabinus is the need to reduce inbreeding (Bram, 1998
). In all likelihood, ecological and genetic factors have worked in concert to promote the evolution of dioecy (Willson, 1979
; Anderson and Stebbins, 1984
; Charlesworth and Charlesworth, 1987
; Quinn, 1991
).
|
FOOTNOTES |
|---|
2 Author for reprint requests, current address: Patrick Center for Environmental Research, Academy of Natural Sciences of Philadelphia, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania 19103-1195 USA.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Anderson, G. J., and G. L. Stebbins. 1984 Dioecy versus gametophytic self-incompatability: a test. American Naturalist 124: 423–428.[CrossRef][ISI]
Baker, H. G. 1984 Some functions of dioecy in seed plants. American Naturalist 124: 149–158.[CrossRef][ISI]
Bawa, K. S. 1980 Evolution of dioecy in flowering plants. Annual Review of Ecology and Systematics 11: 15–39.
———. 1994 Pollinators of tropical dioecious angiosperms: a reassessment? No, not yet. American Journal of Botany 81: 456–460.[CrossRef][ISI]
———, and P. A. Opler. 1975 Dioecism in tropical forest trees. Evolution 29: 167–179.[CrossRef][ISI]
Bierzychudek, P., and V. Eckhart. 1988 Spatial segregation of the sexes of dioecious plants. American Naturalist 132: 34–43.[CrossRef][ISI]
Bram, M. R. 1998 Sex expression, sex-specific traits, and inbreeding depression in freshwater and salt marsh populations of Amaranthus cannabinus (L.) Sauer, a dioecious annual. Ph.D. dissertation, Rutgers University, New Brunswick, New Jersey, USA.
Charlesworth, D., and B. Charlesworth. 1987 Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18: 237–268.[CrossRef][ISI]
Clemants, S. E. 1992 Chenopodiaceae and Amaranthaceae of New York State. In R. S. Mitchell [ed.], Flora of New York State X, Bulletin 485. New York State Museum, Albany, New York, USA.
Conn, J. S. 1981 Phenological differentiation between the sexes of Rumex hastalulus: niche partitioning or different optimal reproductive strategies? Bulletin of the Torrey Botanical Club 108: 374–378.[CrossRef][ISI]
———, and U. Blum. 1981 Differentiation between the sexes of Rumex hastatulus in net energy allocation, flowering and height. Bulletin of the Torrey Botanical Club 108: 446–455.[CrossRef][ISI]
Cox, P. A. 1981 Niche partitioning between sexes of dioecious plants. American Naturalist 117: 295–307.[CrossRef]
Darwin, C. 1877 The different forms of flowers on plants of the same species. Appleton, New York, New York, USA.
Dawson, T. E., and J. R. Ehleringer. 1993 Gender-specific physiology, carbon isotope discrimination, and habitat distribution in boxelder, Acer negundo. Ecology 74: 798–815.[CrossRef][ISI]
———, and M. A. Geber. 1999 Sexual dimorphism in physiology and morphology. In M. A. Geber, T. E. Dawson, and L. F. Delph [eds.], Gender and sexual dimorphism in flowering plants, 175–215. Springer-Verlag, New York, New York, USA.
Delph, L. F. 1999 Sexual dimorphism in life history. In M. A. Geber, T. E. Dawson, and L. F. Delph [eds.], Gender and sexual dimorphism in flowering plants, 149–173. Springer-Verlag, New York, New York, USA.
Freeman, D. C., J. L. Doust, A. El-Keblawy, K. Miglia, and E. D. McArthur. 1997 Sexual specialization and inbreeding avoidance in the evolution of dioecy. Botanical Review 63: 65–92.
———, K. T. Harper, and E. L. Charnov. 1980 Sex change in plants: old and new hypotheses. Oecologia 47: 222–232.[CrossRef][ISI]
———, ———, and W. K. Ostler. 1980 Ecology of plant dioecy in the intermountain region of western North America and California. Oecologia 44: 410–417.[CrossRef][ISI]
———, L. G. Klikoff, and K. T. Harper. 1976 Differential resource utilization by the sexes of dioecious plants. Science 193: 597–599.
———, E. D. McArthur, S. C. Sanderson, and A. R. Tiedemann. 1993 The influence of topography on male and female fitness components of Atriplex canescens. Oecologia 93: 538–547.[CrossRef][ISI]
García, M. B., and R. J. Antor. 1995 Sex ratio and sexual dimorphism in the dioecious Borderea pyrenaica (Dioscoreaceae). Oecologia 101: 59–67.
Grant, M. C., and J. B. Mitton. 1979 Elevational gradients in adult sex ratios and sexual differentiation in vegetative growth rates of Populus tremuloides Michx. Evolution 33: 914–918.[CrossRef][ISI]
Gross, K. L., and J. D. Soule. 1981 Differences in biomass allocation to reproductive and vegetative structures of male and female plants of a dioecious, perennial herb, Silene alba (Miller) Krause. American Journal of Botany 68: 801–807.[CrossRef][ISI]
Johnson, T. 1997 "A warning of enviro-doom at global warming conference." The Star Ledger, 7 October 1997: 17.
Korpelainen, H. 1992 Patterns of resource allocation in male and female plants of Rumex acetosa and R. acetosella. Oecologia 89: 133–139.[CrossRef][ISI]
———. 1994 Sex ratios and resource allocation among sexually reproducing plants of Rubus chamaemorus. Annals of Botany 74: 627–632.
Krischik, V. A., and R. F. Denno. 1990a Differences in environmental response between the sexes of the dioecious shrub, Baccharis halimifolia (Compositae). Oecologia 83: 176–181.[CrossRef][ISI]
———, and ———. 1990b Patterns of growth, reproduction, defense, and herbivory in the dioecious shrub Baccharis halimifolia (Compositae). Oecologia 83: 182–190.[CrossRef][ISI]
Lemen, C. 1980 Allocation of reproductive effort to the male and female strategies in wind-pollinated plants. Oecologia 45: 156–159.[CrossRef][ISI]
Lloyd, D. G. 1982 Selection of combined versus separate sexes in seed plants. American Naturalist 120: 571–585.[CrossRef][ISI]
———, and C. J. Webb. 1977 Secondary sex characteristics in plants. Botanical Review 43: 177–216.
Lovett Doust, J., and G. Laporte. 1991 Population sex ratios, population mixtures and fecundity in a clonal dioecious macrophyte, Vallisneria americana. Journal of Ecology 79: 477–489.[CrossRef]
Lyons, E. E., N. Shah-Mahoney, and L. A. Lombard. 1995 Evolutionary dynamics of sex ratio and gender dimorphism in Silene latifolia: II. Sex ratio and flowering status in a potentially male-biased population. Journal of Heredity 86: 107–113.
Meagher, T. R. 1980 Population biology of Chamaelirium luteum, a dioecious lily. I. Spatial distributions of males and females. Evolution 34: 1127–1137.[CrossRef][ISI]
———. 1981 Population biology of Chamaelirium luteum, a dioecious lily. II. Mechanisms governing sex ratios. Evolution 35: 557–567.[CrossRef][ISI]
———. 1984 Sexual dimorphism and ecological differentiation of male and female plants. Annals of the Missouri Botanical Gardens 71: 254–264.[CrossRef][ISI]
———, and J. Antonovics. 1982 The population biology of Chamaelirium luteum, a dioecious member of the lily family: life history studies. Ecology 63: 1690–1700.[CrossRef][ISI]
Miller, J., and J. Lovett Doust. 1987 The effects of plant density and snail grazing on female and male spinach plants. New Phytologist 107: 613–621.[CrossRef][ISI]
Muenchow, G. E. 1987 Is dioecy associated with fleshy fruit? American Journal of Botany 74: 287–293.[CrossRef]
Parker, V. T., and M. A. Leck. 1985 Relationships of seed banks to plant distribution patterns in a freshwater tidal wetland. American Journal of Botany 72: 161–174.[CrossRef][ISI]
Purrington, C. B. 1993 Parental effects on progeny sex ratio, emergence, and flowering in Silene latifolia (Caryophyllaceae). Journal of Ecology 81: 807–811.[CrossRef]
———, and J. Schmitt. 1998 Consequences of sexually dimorphic timing of emergence and flowering in Silene latifolia. Journal of Ecology 86: 397–404.[CrossRef]
Putwain, P. D., and J. L. Harper. 1972 Studies in the dynamics of plant populations. V. Mechanisms governing the sex ratio in Rumex acetosa and R. acetosella. Journal of Ecology 60: 113–129.[CrossRef]
Quinn, J. A. 1991 Evolution of dioecy in Buchloe dactyloides (Gramineae): tests for sex-specific vegetative characters, ecological differences, and sexual niche-partitioning. American Journal of Botany 78: 481–488.[CrossRef][ISI]
———, D. P. Mowrey, S. M. Emanuele, and R. D. B. Whalley. 1994 The "Foliage is the Fruit" hypothesis: Buchloe dactyloides (Poaceae) and the shortgrass prairie of North America. American Journal of Botany 81: 1545–1554.[CrossRef][ISI]
Renner, S. S., and R. E. Ricklefs. 1995 Dioecy and its correlates in the flowering plants. American Journal of Botany 82: 596–606.[CrossRef][ISI]
Richards, A. J. 1986 Plant breeding systems. George Allen and Unwin Ltd., Boston, Massachusetts, USA.
Sakai, A. K., and S. G. Weller. 1991 Ecological aspects of sex expression in subdioecious Schiedea globosa (Caryophyllaceae). American Journal of Botany 78: 1280–1288.[CrossRef][ISI]
SAS. 1990 SAS/STAT user's guide, version 6, 4th ed. SAS Institute Inc., Cary, North Carolina, USA.
———. 1996 The SAS System on Microsoft Windows 95. SAS Institute Inc., Cary, North Carolina, USA.
Sauer, J. 1955 Revision of the dioecious amaranths. Madroño 13: 5–46.
———. 1972 The dioecious amaranths: a new species name and major range extensions. Madroño 21: 426–434.
Simpson, R. L., R. E. Good, M. A. Leck, and D. F. Whigham. 1983 The ecology of freshwater tidal wetlands. BioScience 33: 255–259.[CrossRef][ISI]
Thomson, J. D., and S. C. H. Barrett. 1981 Selection for outcrossing, sexual selection, and the evolution of dioecy in plants. American Naturalist 118: 443–449.[CrossRef][ISI]
Willson, M. F. 1979 Sexual selection in plants. American Naturalist 113: 777–791.[CrossRef][ISI]
This article has been cited by other articles:
|
|
 |
|
  S. Matsuyama and M. Sakimoto Allocation to Reproduction and Relative Reproductive Costs in Two Species of Dioecious Anacardiaceae with Contrasting Phenology Ann. Bot., June 1, 2008; 101(9): 1391 - 1400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. V. Zandt and S. Mopper Delayed and carryover effects of salinity on flowering in Iris hexagona (Iridaceae) Am. J. Botany, November 1, 2002; 89(11): 1847 - 1851. [Abstract] [Full Text] [PDF]
|
|
| |||||
