Printer Friendly

Effects of time-since-fire and microhabitat on the occurrence and density of the endemic Paronychia chartacea ssp. chartacea in Florida scrub and along roadsides.


The environmental conditions affecting the occurrence and density of plant species must be understood to promote successful conservation (Clubbe et al., 2004; Jacquemyn et al., 2009; Sankaran, 2009). In many habitats, natural disturbances such as fire, wind, flooding, drought and animal activity affect the local densities of plant species (White, 1979; Sousa, 1984). Fire, for instance, creates open spaces, which provide opportunities for colonization and persistence of many species (Menges and Hawkes, 1998). Anthropogenic disturbances such as roads create habitats that may mimic the effects of natural disturbances (Forman, 1995; Petru and Menges, 2004), thereby creating artificial habitats that may be beneficial for some imperiled species. Therefore, assessment of the occurrence and density of an imperiled plant species in both natural and anthropogenic habitats is necessary to develop a conservation plan.

Fire is a natural disturbance in many shrubland ecosystems (Little, 1979; Abrahamson, 1984; Christensen, 1985; Keeley and Keeley, 1988; Moreno and Oechel, 1994; Bradstock et al., 2001). Plant species in pyrogenic habitats have evolved a variety of mechanisms that allow them to persist and recover after fire (Sousa, 1984; Christensen, 1985; Whelan, 1995). Some species are resilient to fires and resprout after burning, some are killed by fire and recolonize via seedling recruitment, and others both resprout and recruit seedlings after fire (Keeley, 1977; Keeley and Zedler, 1978; Myers, 1990; Menges and Kohfeldt, 1995; Whelan, 1995; Weekley and Menges, 2003). Fire removes woody dominant species and creates open areas that are favorable for seedling recruitment (Sousa, 1984), which is important for obligate seeding species that cannot become established in densely vegetated areas (Connell and Slatyer, 1977).

Open patches within the dominant vegetation, often called gaps, occur in many ecosystems (Pickett and White, 1985). Gaps provide areas for seedling recruitment and growth of both woody and herbaceous species (Platt and Weis, 1977; Goldberg and Gross, 1988; Canham, 1989; Rebertus and Burns, 1997; Quintana-Ascencio and Menges, 2000; Kwit and Platt, 2003; Menges et al., 2008). Plants in gaps appear to experience reduced competition (Morgan, 1997; McGuire et al., 2001; Suding and Goldberg, 2001; Petru and Menges, 2003). Within gaps, spatial variation (e.g., center vs. edge) in resources can affect plant establishment, growth and survival (Brokaw and Busing, 2000). Furthermore, the amount of open space in a habitat affects the distribution and density of plant species (Went, 1942; Wu and Levin, 1994; Hawkes and Menges, 1995, Menges et al., 2008).

Disturbances such as fire can create gaps by consuming vegetation, while other mechanisms likely contribute to gap persistence. For example, gaps may be maintained by allelopathic chemicals released by dominant shrubs (Hunter and Menges, 2002; Hewitt and Menges, 2008), particularly in ecosystems with areas of bare soil and low plant densities (Muller, 1966). Allelopathic chemicals can inhibit germination, limit growth of herbaceous species (Muller, 1966) and reduce the effective gap size for plants that require open space, which influences species occurrences and densities.

The occurrence and density of a particular plant species is affected by human disturbances such as roads. When roads and road edges are similar to natural areas, they may provide habitat for native plants (Petru and Menges, 2004; Quintana-Ascencio et al., 2007), and roadsides may act as refugia and connections between natural habitats (Andrews, 1990). Furthermore, seed dispersal can be facilitated by wind in roadside habitats (Forman et al., 2003). On the other hand, roadsides can be detrimental to native plants due to alteration of water movement, erosion processes and invasion by exotic species (Forman et al., 2003). Orientation of roadsides affects habitat quality, with south-facing edges having the highest light availability and temperature and lowest moisture (Matlack, 1993). Furthermore, the distance from the edge of vegetation patches into roadsides may affect species interactions. The ability of roads to provide adequate habitat may affect the occurrence and density of plant species.

Our study took place in Florida scrub habitats on the Lake Wales Ridge in central peninsular Florida, where fire maintains ecosystems including xeric upland habitats such as shrub-dominated Florida scrub (Abrahamson et al., 1984; Myers, 1990; Menges, 1999). Gaps in the shrub matrix are created by fire, but also occur in long-unburned rosemary scrub, perhaps due to allelopathic properties of Florida rosemary (Ceratiola ericoides) (Hunter and Menges, 2002; Hewitt and Menges, 2008). Many species are gap specialists, occurring mainly in recently burned areas or in sites with larger gaps (Menges and Kohfeldt, 1995; Menges et al., 2008). Previous studies of Lake Wales Ridge endemic species have shown the importance of fire, open space and microhabitat (e.g., presence/absence of leaf litter) in determining population occurrences and densities (Hawkes and Menges, 1995; Menges and Kimmich, 1996; Hunter and Menges, 2002; Menges and Quintana-Ascencio, 2004) and that roadsides can support higher population densities than natural scrub habitats (Menges et al., 2006; Quintana-Ascencio et al., 2007; E. Menges, pers. obs.). Paronychia chartacea ssp. chartacea (hereafter Paronychia) is a scrub endemic that is smaller in stature and shorter-lived than other rare scrub herbs. It is one of the most abundant species in the rosemary scrub seed bank (E. Menges and N. Kohfeldt, pers. obs.; J. Navarra, pers. comm.) and occurs along the sides of sand roads. Paronychia recruits from seed (Mengen and Kohfeldt, 1995) and increases in cover after fire (Johnson and Abrahamson, 1990).

In this study, we investigated the occurrence and density of Paronychia in Florida rosemary scrub and roadside populations in relation to time-since-fire and microhabitat. In rosemary scrub, we focused on gap size and location within a gap, whereas in roadsides, we evaluated distance from rosemary scrub and roadside aspect. Our study addressed three main questions: (1) How do time-since-fire and microhabitat affect the occurrence and density of Paronychia in rosemary scrub? (2) How do microhabitat and time-since-fire affect the occurrence and density of Paronychia in roadsides? (3) How does density of Paronychia plants in scrub populations compare to roadside populations? Based on previous studies of other rosemary scrub specialists (Menges and Kohfeldt, 1995; Menges and Kimmich 1996; Quintana-Ascencio et al., 2007), we hypothesized that Paronychia gap occupancy and density in rosemary scrub would be greater in recently burned than in long unburned sites and greater in large gap centers than in large gap edges and small gaps. We hypothesized that Paronychia occurrence and density would be lowest in south facing roadsides and would increase with distance from rosemary scrub. Furthermore, we hypothesized that Paronychia density would be greater in roadsides than in rosemary scrub.



This study was conducted in rosemary scrub and roadsides at Archbold Biological Station (ABS) in Highlands County, Florida, USA (27[degrees]10'50"N, 81[degrees]21'0"W). ABS typically has warm wet summers and cool dry winters (Abrahamson et al., 1984). Mean annual precipitation is 136.5 cm (ABS weather records, 1932-2004), and mean annual temperature is 22.3 C (ABS weather records, 1952-2004). ABS comprises a mosaic of plant communities including seasonal ponds, flatwoods, scrubby flatwoods, oak-hickory scrub and sand pine scrub. Rosemary scrub (also known as the rosemary phase of sand pine scrub) is characterized by xeric white sand and even-aged stands of Florida rosemary (Ceratiola ericoides Michx.), with areas of bare sand colonized by herbaceous and suffrutescent species and ground lichens (Abrahamson et al., 1984). In addition, numerous unpaved sand roads, which serve as fire breaks, traverse ABS property, creating open habitats that are colonized by many scrub endemics (Quintana-Ascencio et al., 2007).

Paronychia chartacea Fern. ssp. chartacea L.C. Anderson (Caryophyllaceae) is endemic to central peninsular Florida, state endangered (Coile and Garland, 2003) and federally threatened (USFWS, 1999). Paronychia is herbaceous and has been described as both an annual (Christman and Judd, 1990) and a short-lived perennial (Anderson, 1991; Menges and Kohfeldt, 1995). Plants above ground for at least 3 mo vary in length (0.1 to 33.7 cm), width (0.1 to 29.9 cm) and height (0.4 to 9.8 cm) (J. Schafer, pers. obs.). Paronychia is considered a gap specialist because it occurs at greater densities in areas of open bare sand (Hawkes and Menges, 1996). Paronychia recovers after fire through germination from seed (Menges and Kohfeldt, 1995), colonizes recently burned areas more quickly than other species endemic to the scrub (Christman and Judd, 1990) and appears in areas postfire where it was scarce or absent prefire (Johnson and Abrahamson, 1990). Paronychia occurs primarily in Florida rosemary scrub and along the edges of associated sand roads.


We studied Paronychia populations in 16 rosemary scrub sites and 16 roadside sites at ABS. In rosemary scrub, we randomly selected four sites from each of four time-since-fire classes (2, 4-6 (hereafter 5), 10 and >15 y since fire) (Table 1). Within each rosemary scrub site, we marked all gaps, defined as areas of open sand within the shrub matrix with two perpendicular axes [greater than or equal to] 1 m in length (thus, a gap is [greater than or equal to] 1 [m.sup.2] in area) (Menges et al., 2008). Dominant shrubs include Florida rosemary, clonal oaks (Quercus inopina Ashe, Quercus geminata Small, Quercus chapmanii Sarg.), ericads (Lyonia spp.) and palmettos (Serenoa repens (W. Bartram) Small and Sabal etonia Swingle ex Nash); gap boundaries were defined by shrubs [greater than or equal to] 50 cm tall. Gaps were grouped into large and small size classes; small gaps had their entire area within 1 m of the gap edge, and thus were 1 to 4 [m.sup.2] in area, whereas large gaps had some area greater than 1 m from the gap edge, and thus were greater than 4 [m.sup.2] in area. Open areas among dominant shrubs can include bare sand, litter, ground lichens, mosses and herbaceous and suffrutescent plants. In Nov. and Dec. 2002, each gap (n = 2159) was checked for the presence of Paronychia.

To assess the distribution and density of Paronychia, four small and four large gaps with Paronychia were randomly selected in each of the 16 rosemary scrub sites. We randomly placed three circular plots (50 cm diameter) in each small gap, in each large gap edge (within 1 m of the gap edge) and in each large gap center ([greater than or equal to] 1 m from the gap edge). In three rosemary scrub sites fewer than four small gaps or four large gaps were occupied by Paronychia, and in one rosemary scrub site several large gap centers were too small to fit three plots. Thus, the number of plots per rosemary scrub site varied between 30 and 36. Overall, we established 186 plots in small gaps, 189 plots in large gap edges and 185 plots in large gap centers. In Feb. 2003, we checked each plot for the presence of Paronychia and counted all Paronychia plants present. Although plots were established only in gaps with Paronychia, some gaps had, by chance, no plots with Paronychia present.

In Apr. 2003, we measured the area of all gaps where Paronychia occupancy and density were assessed (n = 119). To determine the area of large gaps, we mapped the edge of the gap using a Trimble Pro X-R Global Positioning System (GPS) with submeter accuracy. Because the GPS is less accurate for smaller areas, we measured eight distances from the gap center to the gap edge and calculated the area of the resulting triangles to determine the area of small gaps.

We located roadside sites adjacent to rosemary scrub vegetation (different sites than those described above) and randomly selected four sites for each of four roadside aspects (North, South, East and West). We established a macroplot in each of the 16 roadside sites, and we determined the width of a macroplot as the distance from the rosemary scrub vegetation to the road as indicated by the presence of tire tracks. Macroplots ranged from 0.5-3 m in width, from 4-12 m in length, and varied in area from 6-12 [m.sup.2]. Rosemary scrub vegetation bordering roadside macroplots varied from 2 to 39 y since fire. Within each macroplot, 24 circular plots (50 cm diameter) were randomly established: 12 near Florida rosemary ([less than or equal to] 1 m away from the edge of Florida rosemary scrub) and 12 far from Florida rosemary (1-3 m away), if possible. When a roadside macroplot was [less than or equal to] 1 m wide, all plots were located near Florida rosemary. Overall, we established 240 plots near rosemary and 144 far from rosemary. In Feb. and Mar. 2003, we checked each plot for the presence of Paronychia and counted all Paronychia plants present.


We used Kruskal-Wallis tests to evaluate the effect of time-since-fire on the number and size of gaps in rosemary scrub sites. We used a one-way ANOVA to analyze the relationship between the percentage of gaps occupied by Paronychia and time-since-fire. The percentage of gaps occupied by Paronychia across rosemary scrub sites met the assumptions of normality and homogeneity of variances.

We evaluated logistic regression models of Paronychia occurrence in rosemary scrub with time-since-fire (as a continuous variable) and microhabitat (small gap, large gap center or large gap edge). For presence of plants in roadsides, we evaluated logistic regression models with distance from vegetation (near or far), time-since-fire (as a continuous variable) and roadside aspect (North, South, East, or West). We included site as a random factor to account for pseudoreplication due to multiple plots within rosemary scrub patches and within roadsides (Crawley, 2007). We used the Akaike Information Criterion corrected for small sample size ([AIC.sub.C]) to determine the relative support for each model (Burnham and Anderson, 2002). Models were ranked based on their weights. The Akaike weights ([w.sub.i]) give an estimate of the differences in relative information among the models (Burnham and Anderson, 2002). We analyzed logistic regression models of the odds of presence/absence of Paronychia as the dependent variable. Each model evaluated a set of contrasts for the categorical variables using a reference. Small gaps were the reference for the model of Paronychia in rosemary scrub. North aspect was the reference for the model of Paronychia in roadsides. To evaluate the alternative models, we used R (CRAN 2006, R Development Core Team).

We used linear regression to analyze the relationship between time-since-fire and Paronychia density in rosemary scrub patches. We calculated Paronychia density per m") for each gap and then calculated the mean gap density for each rosemary scrub patch. Mean densities were log transformed to meet the assumptions of the analysis. Time-since-fire was considered as a continuous variable.

Considering that microhabitat can affect the occurrence of Paronychia, we also wanted to determine if there were differences in density among microhabitats with Paronychia present. Thus, for rosemary scrub and roadsides, only plots with Paronychia were included in the analysis (n = 49, 32, and 78 for small gaps, large gap edges and large gap centers in rosemary scrub, respectively; n = 145 and 81 for near and far from rosemary in roadsides, respectively; n = 61, 57, 72, and 36, for North, South, East and West roadside aspects, respectively). Densities of Paronychia (per [m.sup.2]) in rosemary scrub and roadsides could not be transformed to normality, so we used Kruskal-Wallis tests. Significant differences were determined using Bonferroni adjusted significance values (Sokal and Rohlf, 1995).


We used a Kruskal-Wallis test to assess the relationship between habitat (rosemary scrub vs. roadside) and Paronychia density. For this analysis, we included plots with and without Paronychia (n = 384, 134, 144, 144, and 138 for roadsides and sites 2, 5, 10 and >15 y since fire, respectively). Significant differences were determined using Bonferroni adjusted significance values (Sokal and Rohlf, 1995). All statistical analyses other than the logistic regressions were conducted in SPSS version 11.5 (SPSS, 2000).


The number of rosemary scrub gaps did not vary significantly with time-since-fire (Kruskal-Wallis [chi square] = 3.83, df = 3, P = 0.280), but the size of rosemary scrub gaps tended to increase with time-since-fire (Kruskal-Wallis [chi square] = 6.27, df = 3, P = 0.099). The percentage of rosemary scrub gaps occupied by Paronychia tended to decrease with time-since-fire, but the trend was only marginally significant (F = 2.83, df = 3,12, P = 0.083; Fig. 1). Gap occupancy was highest 2 y (mean [+ or -] SE = 47.4% [+ or -] 3.9%) and 5 y postfire (mean [+ or -] SE = 49.7% [+ or -] 6.9%) and lowest > 15 y postfire (mean [+ or -] SE = 28.5% [+ or -] 5.8%).

There was a significant association between the presence of Paronychia in rosemary scrub and time-since-fire, microhabitat and their interaction (Tables 2 and 3). Overall, the percentage of plots occupied by Paronychia changed with time-since-fire (40.3%, 38.9%, 22.9% and 11.6% for 2, 5, 10 and > 15 y since fire, respectively); however, this pattern varied among microhabitats (Fig. 2). The variation in plot occupancy with time-since-fire in small gaps was similar to large gap edges, but different from large gap centers (Table 3). Plot occupancy was highest in large gap centers in all sites except those > 15 y since fire, in which small gaps had the highest plot occupancy (Fig. 2).

In rosemary scrub, Paronychia density in gaps decreased with time-since-fire ([R.sup.2] = 0.618, P < 0.001; Fig. 3). For plots occupied by Paronychia, density (mean # of individuals per [m.sup.2] [+ or -] SE) differed among small gaps (8.4 [+ or -] 1.2; n = 49), large gap edges (17.2 [+ or -] 3.7; n = 32) and large gap centers (18.5 [+ or -] 2.4; n = 78) (Kruskal-Wallis [chi square] = 17.07, df = 2, P < 0.001).

Presence of Paronychia in roadsides was best explained by a model including roadside aspect, distance from rosemary, time-since-fire and their interactions (Tables 4 and 5). Plot occupancy in roadsides with North aspects was similar to that in East and West aspects, but was significantly different from the South aspects (Table 5). Overall, there is weak evidence for an effect of distance from rosemary scrub on plot occupancy (Table 5). The effect of distance from rosemary scrub on plot occupancy varied among roadside aspects (Fig. 4). Plot occupancy was similar near and far from rosemary for the North, East and West aspects, but far from rosemary in the South aspect had lower plot occupancy (8.3%) than near to rosemary (68.5%). There was a significant interaction effect of time-since-fire and distance from rosemary scrub on Paronychia plot occupancy (Table 5). For plots near rosemary scrub, plot occupancy increased with time-since-fire, while for plots far from rosemary scrub, plot occupancy decreased with time-since-fire. Overall, 53.1% of roadside plots were occupied by Paronychia, which is greater than plot occupancy in rosemary scrub patches at any time-since-fire class.


Within roadsides, Paronychia density tended to differ with roadside aspect, though the trend was only marginally significant (Kruskal-Wallis [chi square] = 6.47, df = 3, P = 0.091). Paronychia density (mean # of individuals per [m.sup.2] [+ or -] SE) was highest in roadsides with an East aspect (23.8 [+ or -] 2.3; n = 72) and lowest in roadsides with a West aspect (16.4 [+ or -] 3.4; n = 36). Paronychia density increased significantly with distance from rosemary shrubs (Kruskal-Wallis [chi square] 8.78, df = 1, P = 0.003). Mean density ([+ or -] SE) was 17.7 [+ or -] 1.5 near rosemary (n = 145) and 28.3 [+ or -] 3.2 far from rosemary (n = 81). Paronychia density was higher in roadsides than in rosemary scrub sites regardless of time-since-fire of rosemary scrub sites (Kruskal-Wallis [chi square] = 135.83, df = 4, P < 0.001; Fig. 5).


In rosemary scrub, Paronychia occurrence and density decreased with time-since-fire, and in both rosemary scrub and roadsides, Paronychia occurrence and density were higher in microhabitats associated with open space (i.e., large gap centers and far from rosemary scrub). Thus, patterns of Paronychia occurrence and density are similar to herbaceous species in other pyrogenic habitats, where fire removes litter, decreases shrub cover and creates open space. Although Paronychia density was higher in roadsides than in rosemary scrub, and roadsides may act as refugia among rosemary scrub patches, roadsides may alter demography (Menges et al., 2006; Quintana-Ascencio et al., 2007) or evolution (Schlaepfer et al., 2002), providing inadequate habitat over the long-term.


In our study, Paronychia gap occupancy in rosemary scrub declined with time-since-fire, tending to be lower in long unburned sites. In contrast, Menges et al. (2008) found no effect of time-since-fire on Paronychia gap occupancy. We sampled 2159 gaps, every gap in the 16 rosemary scrub patches studied, whereas Menges et al. (2008) sampled 805 gaps, a sub-sample of gaps in 28 rosemary scrub patches. The apparent discrepancy between the two studies may reflect spatial variation in Paronychia gap occupancy within rosemary scrub patches or the ability of our larger data set to detect patterns not evident in the earlier study.

Within rosemary scrub gaps, Paronychia occurrence declined with time-since-fire. Paronychia requires open space, which is available in recently burned rosemary scrub. Ground lichens take 10-12 y to recover to preburn levels (Johnson and Abrahamson, 1990) and approach 100% cover in long unburned areas. Plant litter accumulates over time, which leads to a decrease in the amount of open space. Changes in plant and litter cover cause the quality of gap habitats to decrease over time, which could contribute to the decline in Paronychia occupancy with time-since-fire.

As time-since-fire increased, density of Paronychia within rosemary scrub gaps also decreased. Similarly, Menges and Kohfeldt (1995) found that abundance of Paronychia significantly decreased with time-since-fire in rosemary scrub sites, and Johnson and Abrahamson (1990) found that Paronychia abundance was greater 2 y after fire than prefire. Fire has comparable effects on other species in fire-adapted habitats. For example, densities of annual and perennial herbs increased after fire in Mediterranean chaparral (Tyler, 1995), and densities of the perennial herb Schwalbea americana increased after fire in longleaf pine forests (Kirkman et al., 1998). In Florida, densities of the endemic vine Bonamia grandifloria were greater in a burned scrub site compared to adjacent unburned habitat (Hartnett and Richardson, 1989), and mortality of the rare herb Eryngium cuneifolium was positively associated with time-since-fire (Menges and Kimmich, 1996; Menges and Quintana-Ascencio, 2004). Hawkes and Menges (1996) found no relationship between time-since-fire and density of Paronychia, but density of Paronychia was measured in randomly located plots across the upland landscape, including through the shrub matrix, so their measurements do not adequately represent density of Paronychia in gaps.


Within rosemary scrub gaps, the occurrence of Paronychia differed among microhabitats defined by gap size and location within the gap. Up to 10 y postfire, Paronychia was more likely to be found in the centers of large gaps, indicating Paronychia's preference for bare sand microsites. Surprisingly, however, greater than 15 y postfire, Paronychia was more likely to be found in small gaps. Species richness of gap specialists increases as gap area increases (Menges et al., 2008), suggesting that there are more species competing for resources in large gaps than in small gaps. Among gap specialists, Paronychia may be a weak competitor, but better able to recruit and persist in the limited open space in small gaps.

Regardless of time-since-fire, Paronychia occupancy was lower in the edges of large gaps than in either the centers of large gaps or small gaps (Fig. 2). In recently burned rosemary scrub, gap boundaries are defined by resprouting species such as oaks, ericads and palmettos because rosemary is killed by fire (Johnson and Abrahamson, 1990). Rosemary takes 10-12 y to recover to its preburn cover (Johnson and Abrahamson, 1990), so in longer unburned rosemary scrub, gap boundaries are more often defined by rosemary. Thus, our results suggest that shrubby hardwoods and palmettos as well as rosemary may limit Paronychia colonization. Other scrub endemics are similarly affected by shrubs. For example, the majority of Hypericum cumulicola individuals, another rosemary gap specialist, occur greater than 0.5 m away from rosemary and oak shrubs (Quintana-Ascencio and Morales-Hernandez, 1997), and Eryngium cuneifolium survival increased with distance from shrubs, and especially with distance from Florida rosemary shrubs (Menges and Kimmich, 1996).


Several factors may contribute to the decrease in Paronychia occupancy of large gap edges with time-since-fire (Fig. 2). First, litter depth is greater near gap edges in rosemary scrub (J. Schafer, pers. obs.), and litter depth has been shown to affect seedling recruitment in a fire-prone Mediterranean shrubland (Lloret, 1998). Second, density of rosemary roots is greater directly under rosemary individuals than 2 m away (Hunter and Menges, 2002; Hewitt and Menges, 2008). Petru and Menges (2003) experimentally created complete (aboveground and belowground) gaps in rosemary scrub, and these complete gaps had higher colonization and seedling numbers than natural gaps, possibly due to decreased belowground competition. Finally, rosemary litter leachates have allelopathic effects on several herbaceous scrub species, including Paronychia (Hunter and Menges, 2002). Since cover of rosemary increases with time, rosemary litter also increases with time. Thus, the allelopathic effects of rosemary concomitant with increased litter depth and root density may contribute to the decline in Paronychia occupancy of large gap edges as time-since-fire increases.

Paronychia density was also influenced by gap size and location within a gap. Paronychia density was greatest in the centers of large gaps, which are the areas with the most open space and furthest from rosemary and shrub species. Paronychia density was intermediate in the edges of large gaps. Though these areas are near larger shrub species, they are also connected to a larger open area, unlike small gaps, which had the lowest density of Paronychia. Previous work in rosemary scrub has found that densities of both Paronychia and PolygoneUa basiramia, another gap specialist, increase with open space (Hawkes and Menges, 1995, 1996). Comparably, in old field habitats, seedling emergence and survival are greater in bare ground than in vegetated areas (Gross and Werner, 1982), and abundance of Silene douglasii var. oraria (Carophyllaceae), a grassland endemic, decreases with increasing vegetation height and cover, suggesting that competition may play a role in affecting its density (Kephart and Paladino, 1997). The centers of large gaps provide space for seedling recruitment and escape from shrub species, leading to greater densities of Paronychia.

Within roadside macroplots, the occurrence and density of Paronychia were influenced by distance from rosemary scrub. Paronychia occurrence near or far from rosemary scrub was not consistent among roads with different aspects, but across all roadsides, Paronychia density was greater further from the rosemary scrub edge. Although roadsides experience more vehicle disturbance further from rosemary scrub, allelopathic effects of rosemary (Hunter and Menges, 2002), above- and belowground competition for resources, and plant litter accumulation are greater near rosemary.

Roadside aspect also significantly affected Paronychia occupancy, but had minimal impact on density. Differences in Paronychia occurrence may be related to variation in sun intensity and soil temperature. In oak-chestnut forest patches, South-facing edges had the highest light levels, temperature, and the lowest litter moisture, while North-facing edges had low light levels and temperature, and high litter moisture (Matlack, 1993). Along roads through pastureland and forest in Germany, soil temperature was highest in West-facing roadsides and road shoulders (Ellenberg et al., 1981; as cited in Forman et al., 2003). We did not measure light levels, soil temperature or soil moisture in our study, but scrub roadsides with North, East and West aspects are shaded by vegetation at different times during the day. Although roadsides with a South aspect have sun present throughout the day, they had intermediate Paronychia occupancy and density.

Time-since-fire of the rosemary scrub vegetation bordering roads also contributed to variation in Paronychia occupancy of roadside habitats. Sand movement by wind can determine patterns of sand erosion and accretion, and vegetation along a roadside can protect the soil surface from wind erosion and act as a windbreak (Forman et al., 2003). In rosemary scrub, the height and density of vegetation bordering roads increases with time-since-fire, which may lead to sand accumulation near rosemary scrub vegetation. Percent germination of Paronychia seeds has been found to decrease with burial depth (Petru and Menges, 2004). Furthermore, taller and denser vegetation may cause increased shading of roadsides and increased competition. Nonetheless, Paronychia occurrence increased near rosemary scrub with time-since-fire, suggesting that the protective effects of vegetation outweigh negative interactions in roadsides.

Both Paronychia occupancy and density tended to be greater in roadsides than in rosemary scrub. The scrub endemic Hypericum cumulicola showed more variable temporal and spatial density in roadsides than in scrub, with some years and some roadsides having much higher densities than nearby scrub (Quintana-Ascencio et al., 2007). Differences in microclimate and wind (Forman et al., 2003) between roads and rosemary scrub habitats may explain differences in the density of Paronychia. First, during dry months, roadsides have higher soil moisture than rosemary scrub (Quintana-Ascencio et al., pers. obs.). Thus, roads have greater water availability when water is most limiting. Second, sand movement is more variable along roads than in scrub gaps. Although sand accretion, which inhibits germination of Paronychia, is evident over short time intervals in roads, over longer intervals, sand erosion is greater (Petru and Menges, 2004). The erosion of sand along roads may contribute to higher germination, and thus, higher densities of Paronychia in roadsides.

Roads may serve as connections between rosemary scrub patches and provide an important seed source for colonization of recently burned areas, but roadsides cannot replace rosemary scrub habitats. The scrub endemic Dicerandra frutescens had positive growth rates in oak-hickory scrub and negative growth rates in roadsides (Menges et al., 2006), indicating that roadsides are not as stable as natural habitats. Sand roads and firelanes may provide refugia for Paronychia in fire suppressed habitats, but roads may act as an evolutionary trap (Schlaepfer et al., 2002), providing open habitat that is of lower quality than open habitat in rosemary scrub. Effective fire management is preferable to the proliferation of roads for many reasons, including the danger of roads as conduits for exotic species in upland Florida ecosystems (Gordon et al., 2005) and the possibility that environmental conditions of roads can alter extinction risks of threatened species (Quintana-Ascencio et al., 2007).

This study did not investigate variation in the soil seed bank of Paronychia, though this could affect Paronychia occurrence and density because Paronychia recruits from seed after fire (Menges and Kohfeldt, 1995). Paronychia is one of the most abundant species in the seed bank of rosemary scrub (E.S. Menges and N. Kohfeldt, pers. obs.; J. Navarra, pers. comm.), and Paronychia recruits in areas where it was not present prefire (Johnson and Abrahamson, 1984). Paronychia seeds are primarily dispersed by gravity, but ants may move Paronychia seeds over short distances (L. Sullivan, pers. comm.), and wind and water may also be secondary dispersal agents. Postfire recruitment from seed dispersing from outside a burned area may occur on the edges of rosemary scrub patches, but it is unlikely that Paronychia seeds are dispersed over long distances. It is most likely that postfire recruitment occurs from a persistent belowground seed bank. Paronychia seeds have been found throughout rosemary scrub patches, both near and far from shrubs, but seeds are rare in disturbed scrub where introduced grasses are replacing scrub vegetation (J. Navarra, pers. comm.).

Paronychia has been characterized as both an annual (Christman and Judd, 1990) and a short-lived perennial (Anderson, 1991; Menges and Kohfeldt, 1995); many Paronychia individuals in rosemary scrub live for more than a year and some live for more than 2 y (J. Schafer, pers. obs.) In chaparral and sage scrub, many seeding herbaceous perennials have peak recruitment in the second and fifth years after fire (Keeley et al., 2006). Annuals in these ecosystems exhibited a variety of postfire densities, and many species had the highest densities in the first 2 y after fire and remained present during the following years (Keeley et al., 2006). Thus, patterns of Paronychia density with time-since-fire are comparable to densities of both perennials and annuals in other fire prone shrublands.

Patterns of Paronychia occurrence and density across gradients of time-since-fire and openness and among habitats suggest management strategies to protect this species. Both fire and open space are beneficial to Paronychia populations. Since fire and microhabitat quality (i.e., open space) are not independent of one another, managing one component of rosemary scrub habitats will affect the other. The recommended fire return interval for rosemary scrub is 15-30 y (Menges, 2007), which should allow postfire recovery and maintain populations of Paronychia chartacea ssp. chartacea. Although roadsides supported the highest densities of Paronychia, individuals in roadsides appear to be shorter-lived than individuals in rosemary scrub (J. Schafer, pers. obs.), suggesting that roadsides may not provide adequate habitat for Paronychia over the long term.

Acknowledgments.--We thank Amanda Brothers and Marcia Rickey for help in the field, Marcia Rickey for statistical advice and Roberta Pickert for help with GIS work. This manuscript was improved by comments from the associate editor and two anonymous reviewers. We also thank Kevin Main and others who have contributed to effective fire management at Archbold Biological Station. This work was supported by Archbold Biological Station.




ABRAHAMSON, W. G. 1984. Species responses to fire on the Florida Lake Wales Ridge. Am. J. Bot., 71:35-43.

--, A. F. JOHNSON, J. N. LAYNE AND P. A. PERONI. 1984. Vegetation of the Archbold Biological Station, Florida: an example of the southern Lake Wales Ridge. Fla. Sci., 47:209-250.

ANDERSON, L. C. 1991. Paronychia chartacea ssp. minima (Caryopbyllaceae): a new subspecies of a rare Florida endemic. Sida, 14:435-441.

ANDREWS, A. 1990. Fragmentation of habitat by roads and utility corridors: a review. Aust. Zool., 26:130-141.

BRADSTOCK, R. A., J. E. WILLIAMS AND A. M. GILL. 2001. Flammable Australia: the fire regimes and biodiversity of a continent. Cambridge University Press, Cambridge, New York. 486 p.

BROKAW, N. AND R. T. BUSING. 2000. Niche versus chance and tree diversity in forest gaps. Trends Ecol. Evol., 15:183-188.

BURNHAM, K. P. AND D. R. ANDERSON. 2002. Model selection and multimodel inference: a practical information-theoretic approach. Springer, New York. 488 p.

CANHAM, C. D. 1989. Different responses to gaps among shade-tolerant tree species. Ecology, 70:548-550.

CHRISTENSEN, N. L. 1985. Shrubland fire regimes and their evolutionary consequences, p. 85-100. In: S. T. A. Pickett and P. S. White (eds.). The ecology of natural disturbance and patch dynamics. Academic Press, New York.

CHRISTMAN, S. P. AND W. S. JUDD. 1990. Notes on plant endemic to Florida scrub. Fla. Sci., 53:52-73.

CLUBBE, C., M. GILLMAN, P. ACEVEDO-RODRIGUEZ AND R. WALKER, 2004. Abundance, distribution, and conservation significance of regionally endemic plant species on Anegado, British Virgin Islands. Oryx, 38:342-346.

COILE, N. C. AND M. A. GARLAND. 2003. Notes on Florida's endangered and threatened plants. Fourth Edition. Florida Department of Agriculture and Consumer Services, Gainesville, Florida. 127 p.

CONNELL, J. H. AND R. O. SLATYER. 1977. Mechanisms of succession in natural communities and their role in community stability and organization. Am. Nat., 111:1119-1144.

CRAN 2006, R Development Core Team.

CRAWLEY, M.J. 2007. The R book. John Wiley & Sons, Ltd, England.

ELLENBERG, H., K. MULLER AND T. STOTTELLE 1981. Strassen-Okologie: Auswirkungen von Autobahnen und Strasse auf Okosysteme deutscher Landschaften, p. 19-122. In: Okologie und Strasse, Broschurenreihe der Deutschen Strassenliga, Ausgabe 3, Bonn, Germany.

FORMAN, R. T. T. 1995. Land Mosaics: the Ecology of Landscapes and Regions. Cambridge University Press, Cambridge, New York. 632 p.

--, D. SPERLING, J. A. BISSONETTE, A. P. CLEVENGER, C. D. CUTSHALL, V. H. DALE, L. FAHRIG, R. FRANCE, C. R. GOLDMAN, K. HEANUE, J. A. JONES, F.J. SWANSON, T. TURRENTINE AND T. C. WINTER. 2003. Road ecology: science and solutions. Island Press, Washington, D.C. 481 p.

GOLDBERG, D. E. AND K. L. GROSS. 1988. Disturbance regimes of midsuccessional old fields. Ecology, 69:1677-1688.

GORDON, D. R., C. H. GREENBERG, S. U. CROSS, OVER AND O. L. SLAPCINSKY. 2005. Effects of unpaved road soils on persistence of three non-native grass species. Nat. Area. J., 25:257-262.

GROSS, K. L. AND P. A. WERNER. 1982. Colonizing abilities of "biennial" plant species in relation to ground cover: implications for their distributions in a successional sere. Ecology, 63:921-931.

HARTNETT, D. C. AND D. R. RICHARDSON. 1989. Population biology of Bonamia grandiflora (Concolculaceae): effects of fire on plant and seed bank dynamics. Am. J. Bot., 76:361-369.

HAWKES, C. V. AND E. S. MENGES. 1995. Density and seed production of a Florida endemic, Polygonella basiramia, in relation to time since fire and open sand. Am. Midl. Nat., 133:138-148.

--AND--. 1996. The relationship between open space and fire for species in a xeric Florida shrubland. Bull. Torrey Bot. Club., 123:81-92.

HEWITT, R. E. AND E. S. MENGES. 2008. Allelopathic effects of Ceratiola ericoides (Empetraceae) on germination and survival of six Florida scrub species. Plant Ecol., 198:47-59.

HUNTER, M. E. AND E. S. MENGES. 2002. Allelopathic effects and root distribution of Ceratiola ericoides (Empetraceae) on seven rosemary scrub species. Am. J. Bot., 89:1113-1118.

JACQUEMYN, H., R. BRVS, O. HONNAY AND M. J. HUTCHINGS. 2009. Biological flora of the British Isles: Orchis mascula (L.) L. J. Ecol., 97:360-377.

JOHNSON, A. F. AND W. G. ABRAHAMSON. 1990. A note on the fire responses of species in rosemary scrubs on the southern Lake Wales Ridge. Fla. Sci., 53:138-143.

KEELEY, J. E. 1977. Seed production, seed populations in the soil, and seedling production after fire for two congeneric pairs of sprouting and non-sprouting chaparral shrubs. Ecology, 58:820-829.

--AND P. H. ZEDLER. 1978. Reproduction of chaparral shrubs after fire: A comparison of sprouting and seeding strategies. Am. Midl. Nat., 99:142-161.

--AND S. C. KEELEY. 1988. Chaparral, p. 166-207. In: M. G. Barbour and W. D. Billings (eds.). North American terrestrial vegetation. Cambridge University Press, Cambridge, New York.

--, C.J. FOTHERINGHAM AND M. BAER-KEELEY. 2006. Demographic patterns of postfire regeneration in Mediterranean-climate shrublands of California. Ecol. Monogr., 76:235-255.

KEPHART, S. R. AND C. PALADINO. 1997. Demographic change and microhabitat variability in a grassland endemic, Silene douglasii var. oraria (Caryophyllaceae). Am. J. Bot., 84:179-189.

KIREMAN, L. K., M. B. DREW AND D. EDWARDS. 1998. Effects of experimental fire regimes on the population dynamics of Schwalbea americana L. Plant Ecol., 137:115-137.

KWIT, C. AND W.J. PLATT. 2003. Disturbance history influences regeneration of non-pioneer trees. Ecology, 84:2575-2581.

LITTLE, S. 1979. Fire and plant succession in the New Jersey pine barrens, p. 297-314. In: R. T. T. Forman (ed.). Pine barrens: ecosystem and landscape. Academic Press, New York.

LLORET, F. 1998. Fire, canopy cover and seedling dynamics in Mediterranean shrubland of northeastern Spain. J. Veg. Sci., 9:417-430.

MATLACK, G. R. 1993. Microenvironment variation within and among forest edge sites in the eastern United States. Biol. Conserv., 66:185-194.

MCGUIRE, J. P., R.J. MITCHELL, E. B. MOSER, S. D. PECOT, D. H. GJERSTAD AND C. W. HEDMAN. 2001. Gaps in a gappy forest: plant resources, longleaf pine regeneration, and understory response to tree removal in longleaf pine savannas. Can. J. Forest Res., 31:765-778.

MENGES, E. S. 1999. Ecology and conservation of Florida scrub, p. 7-22. In: R. C. Anderson, J. S. Fralish and J. M. Baskin (eds.). Savannas, barrens, and rock outcrop plant communities of North America. Cambridge University Press, Cambridge, New York.

--. 2007. Integrating demography and fire management: an example from Florida scrub. Aust. J. Bot., 55:261-272.

-- AND N. KOHFELDT. 1995. Life history strategies of Florida scrub plants in relation to fire. Bull. Torrey Bot. Club, 122:282-297.

--AND J. KIMMICH. 1996. Microhabitat and time-since-fire: effects on demography of Eryngium cuneifolium (Apiaceae), a Florida scrub endemic plant. Am. J. Bot., 83:185-191.

--AND C. V. HAWKS. 1998. Interactive effects of fire and microhabitat on plants of Florida scrub. Ecol. Appl., 8:935-946.

--AND P. F. QUINTANA-ASCENCIO. 2004. Population viability with fire in Eryngium cuneifolium: deciphering a decade of demographic data. Ecol. Monogr., 74:79-99.

--, P. F. QUINTANA ASCENCIO, C. W. WEEKLEY AND O. G. GAOUE. 2006. Population viability analysis and fire return intervals for an endemic Florida scrub mint. Biol. Conserv., 127:115-127.

--, A. CRADDOCK, J. SALO, R. ZINTHEFER AND C. W. WEEKLEY. 2008. Gap ecology in Florida scrub: species occurrence, diversity, and gap properties. J. Veg. Sci., 19:503-514.

MORENO, J. M. AND W. C. OECHEL. 1994. The role of fire in Mediterranean-type ecosystems. Springer-Verlag, Berlin. 201 p.

MORGAN, J. W. 1997. The effect of grassland gap size on establishment, growth, and flowering of the endangered Rutidosis leptarrhynchoides (Asteraceae). J. Appl. Ecol., 34:566-576.

MULLER, C. H. 1966. The role of chemical inhibition (allelopathy) in vegetational composition. Bull. Torrey Bot. Club, 5:332-351.

MYERS, R. L. 1990. Scrub and high pine, p. 151-193. In: R. Myers and J. J. Ewel (eds.). Ecosystems of Florida. University of Florida Press, Orlando.

PETRU, M. AND E. S. MENGES. 2003. Seedling establishment in natural and experimental Florida scrub gaps. J. Torrey Bot. Soc., 130:89-100.

--AND--. 2004. Shifting sands in Florida scrub gaps and roadsides: dynamic microsites for herbs. Am. Midl. Nat., 151:101-113.

PICKETT, S. T. A. AND P. S. WHITE. 1985. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York. 472 p.

PLATT, W.J. AND I. M. WEIS. 1977. Resource partitioning and competition within a guild of fugitive prairie species. Am. Nat., 111:479-513.

QUINTANA-ASCENCIO, P. F. AND M. MORALES-HERNANDEZ. 1997. Fire-mediated effects of shrubs, lichens and herbs on the demography of Hypericum cumulicola in patchy Florida scrub. Oceologia, 112:263-271.

--AND E. S. MENGES. 2000. Competitive abilities of three narrowly endemic plant species in experimental neighborhoods along a fire gradient. Am. J. Bot., 87:690-699.

--, C. W. WEEKLEY AND E. S. MENGES. 2007. Comparative demography of a rare species in Florida scrub and road habitats. Biol. Conserv., 137:263-270.

REBERTUS, A. J. AND B. R. BURNS. 1997. The importance of gap processes in the development and maintenance of oak savannas and dry forests. J. Ecol., 85:635-645.

SANKARAN, M. 2009. Diversity patterns in savanna grassland communities: implications for conservation strategies in a biodiversity hotspot. Biodiversity and Conserv., 18:1099-1115.

SCHLAEPFER, M. A., M. C. RUNGE AND P. W. SHERMAN. 2002. Ecological and evolutionary traps. Trends in Ecology and Evolution, 17:474-480.

SOKAL, R. R. AND F.J. ROHLF. 1995. Biometry. 3rd ed. Freeman, New York.

SOUSA, W. P. 1984. The role of disturbance in natural communities. Ann. Rev. Ecol. Syst., 15:353-391.

SPSS. 2000. SPSS, Inc, Chicago, Illinois.

SUDING, K. N. AND D. GOLDBERG. 2001. Do disturbances alter competitive hierarchies? Mechanisms of change following gap creation. Ecology, 82:2133-2149.

TYLER, C. M. 1995. Factors contributing to postfire seedling establishment in chaparral: direct and indirect effects. J. Ecol., 83:1009-1020.

US FISH AND WILDLIFE SERVICE. 1999. Multi-species recovery plan for the threatened and endangered species of south Florida. US Fish and Wildlife Service, Atlanta, Georgia. 2179 p.

WEEKLEY, C. W. AND E. S. MENGES. 2003. Species and vegetation responses to prescribed fire in a long-unburned, endemic-rich Lake Wales Ridge scrub. J. Torrey Bot. Sot., 130:265-282.

WENT, F. W. 1942. The dependence of certain annual plants on shrubs in southern California deserts. Bull. Torrey Bot. Club., 69:100-114.

WHELAN, R.J. 1995. The ecology of fire. Cambridge University Press, Cambridge, New York. 346 p.

WHITE, P. S. 1979. Pattern, process, and natural disturbance in vegetation. Bot. Rev., 45:229-299.

WU, J. AND S. A. LEVIN. 1994. A spatial patch dynamic modeling approach to pattern and process in an annual grassland. Ecol. Monogr., 64:447-464.


Department of Biology, University of Florida, Gainesville 32611


Archbold Biological Station, P.O. Box 2057, Lake Placid, Florida 33862


Department of Biology, University of Central Florida, Orlando 32816



Archbold Biological Station, P.O. Box 2057, Lake Placid, Florida 33862

(1) Corresponding author: FAX: (352) 392-3993; e-mail:
TABLE 1.--Summary of rosemary scrub sites. Small gaps were 1
to 4 [m.sup.2] in area and large gaps were greater than 4
[m.sup.2] in area

Years                 % of gaps      # of gaps        Mean gap
since fire   # of    occupied by     with plots       area (SE)
(fire class) gaps    Paronychia    (small, large)   ([m.sup.2])

 2 (2)        196      55.6            4.4           30.9 (15.2)
 2 (2)         53      41.5            4.1            9.3 (5.8)
 2 (2)        113      39.8            4.4            9.4 (3.0)
 2 (2)         59      52.5            4.2            8.4 (3.6)
 4 (5)        531      34.3            4.4           23.9 (8.4)
 6 (5)         83      67.5            4.4           15.5 (3.8)
 6 (5)        119      50.4            4.4           26.6 (12.0)
 6 (5)        265      46.4            4.4           13.3 (4.6)
10 (10)        46      37.0            4.4           15.9 (3.2)
10 (10)       203      58.6            4.4           23.6 (8.2)
10 (10)        73      35.6            4.4           17.3 (6.2)
10 (10)        44      36.4            4.4           16.0 (6.4)
17 (>15)       46      23.9            2.4           26.0 (9.4)
17 (>15)      179      16.2            4.4           23.9 (8.8)
31 (>15)      119      43.7            4.4           37.3 (13.3)
39 (>15)       30      30.0            4.2           14.7 (6.3)

TABLE 2.--[AIC.sub.c] results of logistic regression models of odds
of presence-absence of Paronychia in plots in rosemary scrub
patches as the dependent variable. MH = microhabitat; TSF =
time-since-fire; B=rosemary  scrub patch as a random effect;
K = number of parameters in the model; -2 log likelihood =
test  parameter from the logistic regression model; [AIC.sub.c] =
corrected Akaike Information Criterion; [[DELTA].sub.i] = difference
between [AIC.sub.c], and [AIC.sub.min], the minimum
[AIC.sub.c] of all models; [w.sub.I] = Akaike weight
indicating the degree of support for each model. n = 560 for
all models

                    -2 log
Model          K   likelihood   [AIC.sub.c]   [[DELTA].sub.i]

MH X TSF + B   7      599.1         585.2          0
MH + TSF + B   5      599.6         589.7          4.6
MH + B         4      608.0         600.0         14.9
TSF + B        3      630.5         624.5         39.4

Model          [w.sub.i]

MH X TSF + B       0.907
MH + TSF + B       0.092
MH + B             0.0005
TSF + B           <0.001

TABLE 3.--Coefficients of the best logistic regression model
of the odds of presence-absence of Paronychia in rosemary
scrub patches as the dependent variable. TSF = time-since-fire.
The reference model is small gap. n = 560. Random effect
variance = 0.30147

Fixed effect             Estimate    SE    Z value   Pr(>[absolute
                                                     value of z])

(Intercept)                 -0.77   0.32     -2.43      0.02
Large gap edge              -0.08   0.40     -0.20      0.84
Large gap center             1.35   0.35      3.85      0.0001
TSF                         -0.03   0.02     -1.45      0.15
Large gap edge * TSF        -0.07   0.04     -1.54      0.12
Large gap center * TSF      -0.05   0.03     -1.91      0.06

TABLE 4.--[AIC.sub.c] results of logistic regression models
of odds of presence-absence of Paronychia in roadside plots
as the dependent variable. D = distance from rosemary; A =
aspect; TSF = time-since fire; R = roadside as a random
effect. n = 384 for all models. Column headings are the same
as in Table 2

                                -2 log
Model               [KAPPA]   likelihood   [AIC.sub.c]

D x A x TSF + R        17        450.3         452.0
D + A + TSF + R         7        472.8         473.1
TSF + R                 3        488.9         489.0
A + R                   6        507.0         507.3
D + R                   4        523.7         523.8

Model               [[DELTA].sub.i]   [w.sub.i]

D x A x TSF + R           0            0.999
D + A + TSF + R          21.1         <0.001
TSF + R                  36.9         <0.001
A + R                    55.2         <0.001
D + R                    71.8         <0.001

TABLE 5.--Coefficients of the best logistic regression model
of the odds of presence-absence of  Paronychia in roadside
plots as the dependent variable. A = aspect; D = distance
from rosemary; TSF = time-since-fire. The reference model is
North aspect. n = 384. Random effect variance = 0.31396

Fixed effect        Estimate    SE    Z value    Pr (>[absolute
                                                 value of z])

(Intercept)            -1.10   0.78     -1.41       0.16
A (South)               2.04   0.95      2.15       0.03
A (East)               18.46   9.70      1.90       0.06
A (West)                1.08   1.09      0.99       0.32
D                       1.37   0.76      1.81       0.07
TSF                     0.09   0.04      2.43       0.02
A(South) D             -3.27   1.21     -2.70       0.01
A(East) D              -7.16   5.24     -1.37       0.17
A(West) D              -1.48   0.85     -1.74       0.08
A(South) TSF           -0.09   0.06     -1.59       0.11
A(East) TSF            -2.05   1.12     -1.82       0.07
A(West) TSF            -0.08   0.06     -1.44       0.15
D * TSF                -0.06   0.03     -2.26       0.02
A(South) D * TSF        0.14   0.07      2.07       0.04
A(East) D * TSF         0.78   0.60      1.29       0.20
A(West) D * TSF         0.04   0.04      1.00       0.32
COPYRIGHT 2010 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Schafer, Jennifer L.; Menges, Eric S.; Quintana-Ascencio, Pedro F.; Weekley, Carl W.
Publication:The American Midland Naturalist
Date:Apr 1, 2010
Previous Article:Re-examining the importance of fish in the diets of stream-dwelling crayfishes: implications for food web analyses and conservation.
Next Article:Ratsnakes and brush piles: intended and unintended consequences of improving habitat for wildlife?

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters