2011 SOUTHEASTERN NATURALIST 10(4):761–771 Susceptibility of Cultivated Native Wildflowers to Deer Damage Lucas W. DeGroote1,*, Holly K. Ober1, James H. Aldrich2, Jeff G. Norcini3, and Gary W. Knox2 Abstract - Foraging preference of Odocoileus virginianus (White-tailed Deer) at ornamental plantings was compared amongst 11 wildflower species native to north Florida and south Georgia. Deer exhibited strong preference for Coreopsis floridana (Florida Tickseed), C. gladiata (Coastalplain Tickseed), C. integrifolia (Fringeleaf Tickseed), and Rudbeckia fulgida (Orange Coneflower). Browsing significantly reduced the height of Florida, Coastalplain, and Fringeleaf Tickseeds, and reduced the number of Florida and Fringeleaf Tickseed flowers. Browsing pressure remained high throughout the growing season; therefore, temporary exclosures are unlikely to offer a viable solution to damage caused by deer. Information on variation in deer preference between species and across seasons should help private landowners and public land managers make strategic decisions regarding which species to establish at residences, food plots, or roadside beautification projects. Introduction Odocoileus virginianus Zimmermann (White-tailed Deer) are capable of causing extensive damage to ornamental plants and agricultural crops (Conover 1997, Conover and Kania 1988, Garrison and Lewis 1987, Stratton and Smathers 1996). Conversion of natural habitat to agricultural lands and housing developments coupled with restriction of hunting in developed areas has increased both the contact between humans and deer as well as the frequency and extent of damage inflicted on ornamentals and crops by deer throughout the southeastern US (Harden et al. 2005). Deer can be deterred from browsing ornamentals and crops through a variety of options; however, nearly all of these options are costly, unsightly, work for only a brief period of time, or are considered objectionable to some members of the public (Andelt et al. 1994, Conover 2001, Mulinas et al. 1994, Rosenberry et al. 2001). A more viable, long-term option for preventing deer damage to ornamentals and crops entails selecting species or varieties that deer find unpalatable for most of the calendar year (Conover and Kania 1988). Because deer preferences vary seasonally (Garrison and Gedir 2006, Godvik et al. 2009), an understanding of the changes in deer foraging habits throughout the growing season would allow homeowners to restrict efforts to protect ornamentals to those periods when the plants are most susceptible to deer browsing. 1Department of Wildlife Ecology and Conservation, North Florida Research and Education Center, University of Florida, 155 Research Road, Quincy, fl32351. 2Department of Environmental Horticulture, North Florida Research and Education Center, University of Florida, 155 Research Road, Quincy, fl32351. 3OecoHort, LLC, 726 Riggins Road, Tallahassee, fl32308. *Corresponding author - degroote.1@gmail.com. 762 Southeastern Naturalist Vol. 10, No. 4 Another potential concern regarding deer foraging on ornamentals involves the possible attraction of deer to roadside plantings. Many states have adopted roadside beautification projects such as wildflower planting right-of-way programs. The Florida Wildflower Foundation (2011) and the Georgia Department of Transportation (GDOT 2011) offer grants to promote highway beautification through landscape programs, with the hope that these programs will lower maintenance costs, reduce erosion, and increase driver alertness. While we are not aware of any research investigating the relationship between highway beautification projects and deer-vehicle collisions, alteration of roadside vegetation is recognized by wildlife biologists and department of transportation administrators as one of the more cost-effective options for mitigating deer-vehicle collisions (Sullivan and Messmer 2003). It therefore seems logical that planting native species found to be desirable or undesirable by deer could increase or decrease, respectively, the frequency of deer-vehicle collisions. Understanding which species of wildflowers are most susceptible to deer damage would be helpful in providing recommendations regarding which species should not be planted along roadsides. Many studies that have attempted to determine preferences of foraging deer have been conducted under artificial conditions with captive deer that may not feed in a manner typical of free-ranging individuals (Crouch 1966, Pepin et al. 2006, Radwan and Crouch 1974, Sauve and Cote 2006). Assessing the preferences of free-ranging deer whose behavior has not been altered in any way is a superior means of evaluating impacts of browsing pressure on ornamentals, particularly if conducted in areas with naturally high densities of deer. We developed a project to investigate foraging preference of wild Whitetailed Deer among annual and perennial Asteraceae native to north Florida and south Georgia. Several species of Coreopsis were chosen for our study because we had previously observed a difference in deer preference amongst species of this genus, and because Coreopsis (undesignated species) is Florida’s state wildflower and is therefore desirable in natural, roadside and ornamental plantings. The 11 species selected for this study, despite their relatedness, represent wildflowers with both upland and wetland habitat preferences; spring, summer, fall or continuous bloom periods; annual, short-lived perennial, and perennial characteristics; as well as rhizomatous, stoloniferous, and crown growth types. The objectives of our study were to determine (1) which wildflower species are most preferred by foraging deer in this region, (2) the effect of browsing on growth and flower production, and (3) if any species would benefit from temporary protection against foraging deer during a particular period during the year. Our results should ultimately reduce economic losses incurred by individuals interested in maintaining ornamental plantings and potentially decrease the likelihood of deer-vehicle collisions in areas with purposeful roadside plantings. Methods We initiated a study to evaluate foraging preference of free-ranging White-tailed Deer for native wildflowers at the North Florida Research and Education Center 2011 L.W. DeGroote, H.K. Ober, J. Aldrich, J.G. Norcini, and G.W. Knox 763 (NFREC) in Quincy, fl(30°32'42"N, 84°35'39"W) from April to November of 2008 and 2009. Although the exact density of deer at our study area is not known, deer density in northern Gadsden County is estimated to be among the highest in the state of Florida (QDMA 2011). Preference was tested amongst 11 species of native annuals and perennials (Asteraceae) selected for their availability (seeds collected <100 mi from study area), potential usefulness in landscape settings, and anecdotal observation of deer browsing preference. Species tested included Coreopsis basalis Blake (Goldenmane Tickseed), C. floridana E.B. Sm. (Florida Tickseed), C. gladiata Walt. (Coastalplain Tickseed), C. integrifolia Pior. (Fringeleaf Tickseed), C. lanceolata L. (Lanceleaf Tickseed), C. leavenworthii T. and G. (Leavenworth’s Tickseed), Gaillardia pulchella Foug. (Firewheel), Ratibida pinnata Vent. (Yellow Coneflower), Rudbeckia fulgida Ait. (Orange Coneflower), R. hirta L. (Black-eyed Susan), and R. mollis Ell. (Softhair Coneflower). Testing was performed at 2 experimental plots, 650 m2 and 850 m2 in size. Plot 1 was located near a busy roadway (50 m), pond (50 m), deciduous forest (50 m), and fields of peanuts and soybeans (<300 m), while plot 2 was located near private property with coniferous trees but no understory (50 m), mixed coniferous-deciduous forest (160 m), fields of winter grasses (50 m), and vegetable plantings (<400 m; pumpkins, squash, green peppers, and onions). Each plot was covered in landscape fabric, divided into 8 replicated blocks, and planted each year with 3 plants per species in each block. To reduce competition, facilitate growth, and provide equal opportunity for browsing by deer, container-raised seedlings were planted at least 0.6 m apart, fertilized when planted (5 g of Osmocote 15-9-12, 12–14 month southern formulation), irrigated during drought periods until established, and hand-weeded as needed. Three of the 8 blocks in each plot were protected from deer with a tall fence, and 5 were left unprotected. Plant height, number of flowers (0 – 1, 2 – 5, 6 – 20, or >20), and deer damage (presence or absence of browsing) were recorded every 2 weeks from 7 April (shortly after planting) through the end of their effective growing season, 17 November. We quantified deer preference and the effect of deer on wildflower species in 4 ways: percentage of plants browsed, effect of browsing on flower production, severity of browse damage, and timing of browse damage. All statistical analyses were conducted with the program R version 2.8.1 (R Development Core Team 2000). The percentage of plants browsed was compared amongst the 11 species using a general linear model (GLM) with quasibinomial errors (Crawely 2007). Unlike traditional contingency tables, the aforementioned GLM is more robust for datasets with an unbalanced design due to missing values (in our case, from plant death). Plants that died before browsing occurred were excluded from the analysis, and a plant was considered browsed if it was browsed at any time throughout the year. To determine if browsing differed between years and plots, we modeled the ratio of browsed to unbrowsed plants as a binomial denominator against species (a multilevel factor), year, plot, and all two-way interactions. We used an iterative process whereby whichever independent variable (plot, year, species, or the interaction variables) that 764 Southeastern Naturalist Vol. 10, No. 4 explained the least amount of variability was removed from the full model, and P-values were calculated from a likelihood ratio test comparing the reduced model to the full model. This process was repeated until the likelihood ratio was significantly different, indicating the most parsimonious model had been identified (Diggle et al. 2002). We then used the same model-reduction process to identify which species were browsed more frequently. Species were removed one at a time from the aforementioned most parsimonious model and compared to the prior model. Resulting P-values were bootstrapped to calculate q-values, the analogous Bayesian posterior P-value, to account for positive false discovery rates (pFDR) resulting from multiple comparisons (Dabney and Storey 2010, Storey 2003). We compared flower production between browsed and unbrowsed plants for the four most commonly browsed species using an analysis of variance (ANOVA) and Tukey’s honest significant differences (Tukey’s HSD). Because plants inside and outside of the deer enclosure provided useful information on flower performance, all plants were included in the analysis. We created a single ANOVA model to compare the maximum number of flowers per plant from any sampling period (using category midpoints of 0.5, 3.5, 12.5, and 25) by species, year, plot, browse damage (presence or absence), and their 2-way interaction variables. We compared the severity of damage amongst the four most commonly browsed species by quantifying height lost from browsing. Height loss was calculated by taking the height of a plant when browsing was observed and subtracting it from the height measured 2 weeks prior. If an individual plant was browsed more than once throughout the study, height loss was averaged to avoid pseudo-replication. To account for heterogeneity present across species and year, we used a general least squares model to determine average height loss for each species-year combination. Lastly, we used generalized linear mixed models with binomial errors to investigate browse timing for the four most commonly browsed species. The full model consisted of presence or absence of browse damage regressed by plot, year, species, and the first- through third-order polynomial of sampling week (mean centered). Each plant was included as a random effect to account for repeated measurements of the same plants over time. The most parsimonious model was identified with the same iterative process used in the percent browsed analysis. Browse timing was graphically represented by locally weighted least squares (Lowess) curves showing the percentage of browsed plants over the year for each species-year combination. Results The percentage of plants browsed by deer was significantly higher for Florida Tickseed, Coastalplain Tickseed, Fringeleaf Tickseed, and Orange Coneflower than for Lanceleaf Tickseed, Goldenmane Tickseed, Leavenworth’s Tickseed, Firewheel, Yellow Coneflower, Black-eyed Susan, and Softhair Coneflower (q-value ≤ 0.05, Table 1). The interaction between plot and species was nonsignifi cant (likelihood ratio test: P = 0.12), and the most parsimonious model 2011 L.W. DeGroote, H.K. Ober, J. Aldrich, J.G. Norcini, and G.W. Knox 765 indicated that the percentage of browsed plants was significantly greater at plot 1 than plot 2 (P < 0.01). Deer browsing had a significant effect on maximum flower production (P < 0.01, F1 , 317 = 28.65), where browsed plants produced fewer flowers per plant (mean = 6.4) than unbrowsed plants (mean = 11.7; Fig.1). Tukey’s HSD revealed Figure 1. Maximum number of flowers produced in the presence (white) or absence (gray) of browse damage for the four most frequently browsed species of wildflowers. Bars represent standard errors and significant differences are indicated by * (P ≤ 0.05). Table 1. Preference of White-tailed Deer for native wildflower species based on results of general linear model analyses. The q-value is the analogous P-value for multiple comparisons positive false discovery rates. Significant values and percentages are shown in bold. Species q-value Percent browsed Coreopsis lanceolata (Lanceleaf Tickseed) 0.42 3% Ratibida pinnata (Yellow Coneflower) 0.42 3% Coreopsis basalis (Goldenmane Tickseed) 0.42 5% Gaillardia pulchella (Firewheel) 0.16 5% Rudbeckia mollis (Softhair Coneflower) 0.18 17% Rudbeckia hirta (Black-eyed Susan) 0.16 23% Coreopsis leavenworthii (Leavenworth’s Tickseed) 0.16 27% Rudbeckia fulgida (Orange Coneflower) 0.04 42% Coreopsis gladiata (Coastalplain Tickseed) 0.03 48% Coreopsis floridana (Florida Tickseed) 0.01 60% Coreopsis integrifolia (Fringeleaf Tickseed) <0.01 67% 766 Southeastern Naturalist Vol. 10, No. 4 that browsed Florida Tickseeds produced 49% fewer flowers (P = 0.01) than unbrowsed plants (mean difference = 5.98), while browsed Fringeleaf Tickseeds produced 58% fewer flowers (P < 0.01) than unbrowsed plants (mean difference = 11.03). Browsing had no effect on Coastalplain Tickseed or Orange Coneflower flower production (P = 0.99). Because browsing at plot 2 was infrequent, the severity and timing of browse damage was investigated at plot 1 only. Height of Florida, Coastalplain, and Fringeleaf Tickseed was significantly reduced by browsing (P < 0.01 for 2008 and 2009; Fig. 2), while Orange Coneflower height increased from growth faster than height was reduced by browsing (P < 0.01 for 2008 and 2009). Browsing began 5 weeks earlier (P < 0.01) in 2009 (2 June) than 2008 (5 July). In both years, the percentage of browsed Florida and Coastalplain Tickseed remained high (>50%) from the initial date of browsing to the end of the growing season (Fig. 3). In 2008, Fringeleaf Tickseed also had a high percentage of plants browsed from the initial date of browsing onward, but the percentage of plants browsed in 2009 was only high from 2 June to 10 August (P < 0.01). The percentage of browsed Orange Coneflower was high (>50%) from 2 June to 10 September 2009 only (P < 0.01). Figure 2. Average height loss of the three species of wildflowers browsed most heavily by White-tailed Deer (height prior to damage – height after damage) at plot 1. Height loss in 2008 is represented by gray bars, 2009 is in white. Bars represent standard errors. 2011 L.W. DeGroote, H.K. Ober, J. Aldrich, J.G. Norcini, and G.W. Knox 767 Discussion Four of the 11 native wildflower species were more susceptible to browsing by White-tailed Deer: Florida, Coastalplain, and Fringeleaf Tickseed, as well as Orange Coneflower (Table 1). Research at other sites has shown that browsing by White-tailed Deer can inhibit wildflower growth and flower production, altering the structure and dynamics of plant populations in natural settings (Augustine and DeCalesta 2003, Barrett and Stiling 2006, Frankland and Nelson 2003, Kettenring et al. 2009, Rooney 2001, Rooney and Gross 2003). For example, browsing increased mortality, reduced plant height, inhibited flower production, and decreased fecundity of Trillium grandiflorum (Michx.) Salisb. (White Trillium) in North American forests (Frankland and Nelson 2003, Rooney and Gross 2003). Closer to our study area, an experimental study in central Florida demonstrated that Liatris ohlingerae (S.F. Blake) B.L. Rob. (Florida Blazing Star) were shorter, less likely to flower, and had fewer inflorescences in areas where deer were not excluded (Kettingring et al. 2009). We found that Orange Coneflower Figure 3. Lowess curves representing the percentage of browsed plants over time (April to November) at plot 1 for the four preferred species: (A) Florida Tickseed, (B) Coastalplain Tickseed, (C) Fingeleaf Tickseed, and (D) Orange Coneflower. Solid line and points correspond to browsing in 2008, dashed lines and open circles correspond to 2009. Tick marks on the x-axis indicate the first sampling period of each month. Three sampling periods occurred in July, two in every other month. 768 Southeastern Naturalist Vol. 10, No. 4 was the only species of the four most heavily browsed species (i.e., species for which >40% of plants were damaged) that was able to regrow and flower to sustain an attractive appearance (Figs. 1 and 2). Florida and Fringeleaf Tickseed plants browsed by deer produced fewer flowers and often perished (65% and 25% mortality, respectively). We did not observe any effect of browsing on the maximum number of flowers produced by Coastalplain Tickseed because most plants of this species did not survive to flower (17% survival for unbrowsed, 13% for browsed). Coastalplain Tickseed’s low survival rate is likely explained by the relatively dry conditions in our ornamental landscape setting compared with the species’ preference for wetter habitats (Norcini and Aldrich 2007). Of the few Coastalplain Tickseeds which survived to flower (n = 15), browsed plants produced fewer flowers (mean = 9.6) than unbrowsed individuals (mean = 18.0). Because of their overall susceptibility to browsing, cultivation of Florida, Coastalplain, and Fringeleaf Tickseeds for aesthetic purposes in north Florida and south Georgia may be extremely difficult without protection from deer (i.e., exclusion fences; Rosenberry et al. 2001). Unprotected, long-term ornamental plantings of these annual and short-lived perennial species would likely require re-seeding to compensate for reduced seed production brought about by browsing. Conversely, the relative unpalatability of Goldenmane Tickseed, Lanceleaf Tickseed, Leavenworth’s Tickseed, Firewheel, Yellow Coneflower, Black-eyed Susan, and Softhair Coneflower, and the ability of Orange Coneflower to withstand browsing, could make them desirable species for nurseries and ornamental gardens in areas with high deer densities. The strong preference White-tailed Deer exhibited for Florida, Coastalplain, and Fringeleaf Tickseeds may attract deer to areas with purposeful plantings of these species, an effect which could be desirable or undesirable. Planting these species in landscape beds could attract deer to properties for wildlife viewing or hunting. Food plots, plantings intentionally established by hunters to concentrate deer in particular areas, often consist of non-native plants or agriculture crops which may require annual planting, herbicide, insecticide, and fertilizer (i.e., Hehman and Fulbright 1997, Johnson and Dancak 1993). Using native wildflowers adapted to north Florida and south Georgia that are known to be favored by deer and also known to be capable of sustaining browsing could reduce costs and environmental impacts typically associated with foodplot establishment and maintenance. Browsing at our study site began late summer in 2008 (28 July) and early summer the following year (2 June 2009). The intensity of browsing over time also varied between years and across species (Fig. 2). For example, Florida, Coastalplain, and Fringeleaf Tickseed were intensely browsed from late July to mid-November in 2009, while the intensity of browsing on Florida and Fringeleaf Tickseeds was lower overall in 2008 and slacked for 6 weeks between late August and early October. The time of year in which browsing began and the intensity of browsing may have differed between years because food preferences of free-ranging deer can be influenced by the availability of local food resources (Garrison and Gedir 2006) and differential risk of predation (Brown and Kotler 2011 L.W. DeGroote, H.K. Ober, J. Aldrich, J.G. Norcini, and G.W. Knox 769 2004, Godvik et al. 2009). Likewise, food availability or predation risk could also explain why damage was significantly different between our two study plots. Specifically, plot 1 was located closer to agricultural fields with crops deer find highly palatable (peanuts and soybeans), while plot 2, being located farther away from roads, water, and deciduous cover, may have posed a higher risk for predation. Because of the variability we observed between plots and years, we recommend that private land owners and nurseries monitor local deer damage to Florida, Coastalplain, or Fringeleaf Tickseed to assess the cost versus benefit of protecting these species throughout the growing season. Our results indicate that White-tailed Deer preference for wildflowers changed over time and differed amongst species, which suggests that carefully planned highway beautification projects could reduce the risk of deer-vehicle collisions. Traditional roadside plantings in the southeastern US consist of clover and cover grass which can be palatable for deer year-round (Chapman et al. 2009, Murphy et al. 1985). Because deer in our study did not browse any wildflowers during the spring, replacing traditional roadside plantings with native wildflower plantings that flower during this time of year could potentially reduce deer-vehicle collisions. Moreover, planting other species which we found to be unpalatable for deer year-round could further reduce this risk. However, a multi-year, multi-location study comparing traditional roadside plantings to native grass-wildflower stands would be needed to elucidate the effect of plantings on deer-vehicle collisions. The four most commonly browsed species all flowered in the late summer or fall. None of the spring or early summer blooming wildflowers in our study were browsed. This pattern may have arisen because deer find wildflower plants in bloom less palatable or because other highly preferred foods were more available early in the year. In comparison to the other species tested, the four preferred species are naturally found in wetter habitats (Clewell 1985) and possess more succulent stems and leaves (Norcini and Aldrich 2007). Taken together, our results indicate that White-tailed Deer in north Florida may prefer wildflowers that favor wet habitats, possess succulent foliage, and/or bloom in the late summer or fall; however, further study is needed to determine if these preferences remain true across a wider diversity of habitats and native wildflower species. Acknowledgments We thank T. Batey, A. Brock, J. Crowell, and S. Wright for their assistance establishing and maintaining research plots, planting and irrigating wildflowers, collecting data, and entering data. Literature Cited Andelt, W.F., K.P. Burnham, and D.L. Baker. 1994. Effectiveness of capsacin and bitrex repellents for deterring browsing by captive Mule Deer. Journal of Wildlife Management 58:330–334. Augustine, D.J., and D. DeCalesta. 2003. Defining deer overabundance and threats to forest communities: From individual plants to landscape structure. Ecoscience 10:472–486. 770 Southeastern Naturalist Vol. 10, No. 4 Barrett, M.A., and P. Stiling. 2006. Effects of Key Deer herbivory on forest communities in the lower Florida Keys. Biological Conservation 129:100–108. Brown, J.S., and B.P. Kotler. 2004. Hazardous duty pay and the foraging cost of predation. Ecology Letters 7:999–1014. Chapman, G., E. Bork, N. Donkor, and R. Hudson. 2009. Yields, quality, and suitability of four annual forages for deer pasture in north central Alberta. The Open Agriculture Journal 3:26–31. Clewell, A.F. 1985. Guide to the Vascular Plants of the Florida Panhandle. Florida State University Press, Tallahassee, FL. 605 pp. Conover, M.R. 1997. Monetary and intangible valuation of deer in the United States. Wildlife Society Bulletin 25:298–305. Conover, M.R. 2001. Effects of hunting and trapping on wildlife damage. Wildlife Society Bulletin 29:521–532. Conover, M.R., and G.S. Kania. 1988. Browsing preference of White-tailed Deer for different ornamental species. Wildlife Society Bulletin 16:175–179. Crouch, G.L. 1966. Preferences of Black-tailed Deer for native forage and Douglas-fir seedlings. Journal of Wildlife Management 30:471–475. Crawely, M.J. 2007. Chapter 16: Proportion data. Pp. 569–591, In The R Book. John Wiley and Sons Ltd. West Sussex, UK. 942 pp. Dabney, A., and J.D. Storey. 2010. qvalue: Q-value estimation for false discovery rate control. R package version 1.22.0. Available online at http://CRAN.R-project.org/ package=qvalue. Accessed 7 June 2011. Diggle P.J., P. Heagerty, K.Y. Liang, and S.L. Zeger. 2002. The Analysis of Longitudinal Data. Second Edition. Oxford University Press. Oxford, UK. 396 pp. Florida Wildflower Foundation. 2011. Florida’s native wildflowers. Available online at http://www.flawildflowers.org. Accessed 7 February 2011. Frankland, F., and T. Nelson. 2003. Impacts of White-tailed Deer on spring wildflowers in Illinois, USA. Natural Areas Journal 23:341–348. Garrison, E., and J. Gedir. 2006. Ecology and management of White-tailed Deer in Florida. Florida Fish and Wildlife Conservation Commission. Tallahassee, FL. 49 pp. Garrison, R.L., and J.C. Lewis. 1987. Effects of browsing by White-tailed Deer on yields of soybeans. Wildlife Society Bulletin 15:555–559. Georgia Department of Transportation (GDOT). 2011. Landscape Program. Available online at http://www.dot.state.ga.us/informationcenter/programs/environment/landscapes/ Pages/default.aspx. Accessed 7 February 2011. Godvik, I.M.R., L.E. Loe, J.O. Vik, V. Veiberg, R. Langvatn, and A. Mysterud. 2009. Temporal scales, trade-offs, and functional responses in Red Deer habitat selection. Ecology 90:699–710. Harden, C.D., A. Woolf, and J. Roseberry. 2005. Influence of exurban development on hunting opportunity, hunter distribution, and harvest efficiency of White-tailed Deer. Wildlife Society Bulletin 33:233–242. Hehman, M.W., and T.E. Fulbright. 1997. Use of warm-season food plots by White-tailed Deer. Journal of Wildlife Management 61:1108–1115. Johnson, M.K., and K.D. Dancak. 1993. Effects of food plots on White-tailed Deer in Kisatchie National Forest. Journal of Range Management 46:110–114. Kettenring, K.M., C.W. Weekley, and E.S. Menges. 2009. Herbivory delays flowering and reduces fecundity of Liatris ohlingerae (Asteraceae), an endangered, endemic plant of the Florida scrub. Journal of the Torrey Botanical Society 136:350–362. 2011 L.W. DeGroote, H.K. Ober, J. Aldrich, J.G. Norcini, and G.W. Knox 771 Mulinas, M.C., A.F. Rhoads, and J.R. Mason. 1994. Effectiveness of odour repellents for protecting ornamental shrubs from browsing by White-tailed Deer. Crop Protection 13:393–397. Murphy, R.K., N.F. Payne, and R.K. Anderson. 1985. White-tailed Deer use of an irrigated agriculture grassland complex in central Wisconsin. Journal of Wildlife Management 49:125–128. Norcini, J.G., and J.H. Aldrich. 2007. Performance of native plants under north Florida conditions. Florida Cooperative Extension Service Publication ENH 1074. Gainesville, FL. 22 pp. Pepin, D., P.C. Renaud, Y. Boscardin, M. Goulard, C. Mallet, F. Anglard, and P. Ballon. 2006. Relative impact of browsing by Red Deer on mixed coniferous and broadleaved seedlings: An enclosure-based experiment. Forest Ecology and Management 222:302–313. Quality Deer Management Association (QDMA). 2011. Quality Deer Management Association’s Whitetail map guide. Available online at http://www.i-maps.com/Qdma. Accessed 18 February 2011. Radwan, M.A., and G.L. Crouch. 1974. Plant characteristics related to feeding preference by Black-tailed Deer. Journal of Wildlife Management 38:32–41. R Development Core Team. 2009. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Rooney, T.P. 2001. Deer impacts on forest ecosystems: A North American perspective. Forestry 74:201–208. Rooney, T.P., and K. Gross. 2003. A demographic study of deer-browsing impacts on Trillium grandiflorum. Plant Ecology 168:267–277. Rosenberry, C.S., L.I. Muller, and M.C. Conner. 2001. Movable, deer-proof fencing. Wildlife Society Bulletin 29:754–757. Sauve, D.G., and S.D. Cote. 2006. Winter forage selection in White-tailed Deer at high density: Balsam Fir is the best of a bad choice. Journal of Wildlife Management 71:911–914. Storey, J.D. 2003. The positive false discovery rate: A Bayesian interpretation and the q-value. Annals of Statistics 31:2013–2035. Stratton, G.R., and W.M. Smathers, Jr. 1996. Crop damage levels in South Carolina imply a changing role for White-tailed Deer hunters. Pp. 92–97, In R. Johnson (Ed.). A Symposium on the Economics of Wildlife Resources on Private Lands. 5–7 August 1996. Auburn University. Auburn, AL. Sullivan, T.L., and T.A. Messmer. 2003. Perceptions of deer-vehicle collision management by state wildlife agency and department of transportation administrators. Wildlife Society Bulletin 31:163–173.