Some Like it Hot: Fire and Legume Germination 
in the Longleaf Pine Ecosystem 
 
by 
 
Marion Scott Wiggers 
 
 
 
 
A thesis submitted to the Graduate Faculty of 
Auburn University 
in partial fulfillment of the 
requirements for the Degree of 
Master of Science 
 
Auburn, Alabama 
May 9, 2011 
 
 
 
 
Keywords: fire effects, germination, legumes, Fabaceae, 
longleaf pine ecosystem, fine-scale heterogeneity 
 
 
Copyright 2011 by Marion Scott Wiggers 
 
 
Approved by 
 
Robert S. Boyd, Co-chair, Professor of Biological Sciences 
L. Katherine Kirkman, Co-chair, Adjunct Professor of Forestry and Wildlife Sciences 
Dean Gjerstad, Professor Emeritus of Forestry and Wildlife Sciences 
 
 
ii 
 
 
 
 
 
Abstract 
 
 
Fire is a globally distributed disturbance and is important for structuring and maintaining 
the diversity of many frequently burned ecosystems. Competitive relationships and vegetation 
patterns may be affected by fire?s influence on survival, reproduction, and recruitment of plants. 
Each fire is unique and the effects of a given fire are not necessarily uniform throughout the area 
burned. Within the southeastern United States, the once-vast longleaf pine (Pinus palustris Mill.) 
ecosystem has been reduced to a mere fraction of its former range by overharvesting and land 
conversion. This fire-dependent system is also home to one of the most diverse plant 
assemblages known in North America. The effectiveness of restoration and management of this 
imperiled system may be improved by increased knowledge of fire-related germination and 
recruitment responses for functionally important groups of plant species. 
This project first investigated potential mechanisms of fire-induced germination and then 
sought to understand how fine-scale fire heterogeneity may influence germination of common 
native legume species. For the first part of the project, six legume species were used to examine 
the effects of exposure to various components of fire: 1) elevated temperatures (five levels with 
dry heat temperatures ranging from presumably non-lethal to lethal) over three durations; 2) 
duration (three intervals) of moist heat (i.e., steam); 3) mechanical scarification; and 4) smoke. 
Legume species used represented a range of seed hardness, varying from soft-seeded to 
extremely hard-seeded. Results indicate that response to both dry and moist heat was species-
specific, but moist heat had the greatest potential to stimulate germination. The legumes studied 
iii 
also exhibited high heat tolerance, which may be a function of their low seed moisture contents. 
Smoke had no detectable effect on germination of most species, but may have inhibited 
Centrosema virginianum. 
Two studies were used for the second part of the project to examine post-fire germination 
of nine legume species in response to fuel variation in laboratory and field settings. For both 
studies, fine fuel loads (longleaf pine needles) and woody fuels (longleaf pine cones) were 
manipulated. Germination and mortality were assessed for six species after experimental 
laboratory fires. Germination of seven species was assessed for four months in the field 
following experimental burns of small plots. In both studies, exposure to fire reduced seed 
germination compared to unburned controls. Furthermore, germination of seeds that were under 
burned pine cones was reduced more than treatments burned with pine needles alone. 
Manipulation of biomass of fine fuel loads did not affect germination in either study. 
Together, the results of all experiments indicate the potential influence of fire on 
germination and mortality of legume seeds. Such impacts may, in turn, affect recruitment of 
these and similar species, thereby influencing post-fire vegetation patterns. Additionally, a 
hypothetical model is presented that describes the potential influence of pine cones on post-fire 
seedling recruitment. 
iv 
 
 
 
 
 
Acknowledgments 
 
 
I could not have completed this research without the patience, thoughtfulness, 
enthusiasm, and encouragement of my committee members: Robert Boyd, Katherine Kirkman, 
and Dean Gjerstad. I am grateful to Kevin Hiers for his insight into and support of this research. I 
would also like to thank the Joseph W. Jones Ecological Research Center for providing me with 
a generous home and funding for my graduate research. Numerous scientific and support staff at 
the Jones Center provided valuable assistance over the course of my graduate career; however, 
some went above and beyond the call of duty, including: Drs. Robert Mitchell and Lindsay 
Boring contributed many helpful ideas throughout all phases of my research; Dr. Mike Conner 
who patiently answered endless statistical questions; Liz Cox who provided tireless library 
assistance; and Jean Brock who provided expert GIS assistance. Additionally, Melanie Kaeser, 
Jason McGee, Josh Warren, Stephen Pecot, Kim Coffey, Scott Taylor, and Jim Bradley provided 
much-needed logistical support. My research would not have been possible without the help of 
numerous field assistants and friends: Carla Atkinson, Amanda Bessler, Amanda Briant, Lee 
Camfield, Michelle Creech, Kim Cressman, Mary Ann (Hollenbeck) Devine, David Dyson, Matt 
Greene, Laura Holtrop, Gwen Iacona, Kirk McAlpin, Cristina McKernan, Tara Muenz, Caroline 
Polgar, Braden Ramage, Leighton Reid, Lindsey Sargent, Ryan Smith, Christen Steele, Stribling 
Stuber, Katie Stuble, Jacob Thompson, and Andy Whelan. Finally, I would like to thank my 
parents, Michael and Vicki Wiggers, for their love and support throughout my graduate program. 
v 
 
 
 
 
 
Table of Contents 
 
 
Abstract ......................................................................................................................................... ii 
Acknowledgments........................................................................................................................ iv 
List of Tables .............................................................................................................................. vii 
List of Figures ............................................................................................................................ viii 
List of Abbreviations ................................................................................................................... ix 
Chapter 1. Introduction ................................................................................................................. 1 
 Project Overview .............................................................................................................. 1 
 Literature Review.............................................................................................................. 2 
 Objectives ......................................................................................................................... 6 
 References ......................................................................................................................... 7 
Chapter 2. Mechanisms of Fire-Induced Germination and Survival of Six Legume Species in  
the Longleaf Pine Ecosystem .......................................................................................... 12 
 Abstract ........................................................................................................................... 13 
 Introduction ..................................................................................................................... 14 
 Materials and Methods .................................................................................................... 17 
 Results ............................................................................................................................. 22 
 Discussion ....................................................................................................................... 25 
 References ....................................................................................................................... 31 
Chapter 3. Impacts of Fine-Scale Variation in Surface Fire Environment on Legume  
Germination in the Longleaf Pine Ecosystem ................................................................ 45 
vi 
 Abstract ........................................................................................................................... 46 
 Introduction ..................................................................................................................... 47 
 Materials and Methods .................................................................................................... 49 
 Results ............................................................................................................................. 57 
 Discussion ....................................................................................................................... 59 
 References ....................................................................................................................... 64 
Chapter 4. Summary and Conclusions ........................................................................................ 79 
Appendix ..................................................................................................................................... 83 
 
vii 
 
 
 
 
 
List of Tables 
 
 
Table 2.1. Legume seed germination and mortality responses to untreated control,  
scarification, and smoke treatments ................................................................................ 37 
Table 2.2. Mortality responses of legume seeds to dry heat treatments ..................................... 38 
Table 2.3. Germination of legume seeds following dry heat treatments .................................... 39 
Table 2.4. Germination and mortality of legume seeds following moist heat treatments .......... 41 
Table 2.5. Temperatures of dry heat, steam, and smoke treatments ........................................... 42 
Table 3.1. Influence of fine fuel load variation on post-burn germination and mortality of  
legume seeds for the laboratory germination study ........................................................ 70 
Table 3.2. Effect of burning on germination of legume seeds (July ? November 2007) for  
the field germination study ............................................................................................. 71 
Table 3.3. Effect of fuel variation on post-burn germination of legume seeds for the field  
germination study............................................................................................................ 72 
viii 
 
 
 
 
 
List of Figures 
 
 
Figure 2.1. Mortality and germination responses of legume seeds to dry heat treatments ......... 43 
Figure 2.2. Mortality and germination responses of legume seeds to steam .............................. 44 
Figure 3.1. Example of plot layout (post-fire) ............................................................................ 73 
Figure 3.2. Thermal imagery of a typical burn during flaming combustion and during  
smoldering/glowing combustion with smoldering pine cones visible ............................ 74 
Figure 3.3. Effect of fuel treatments on mean maximum temperatures and mean durations  
of elevated temperatures for the laboratory study and field study .................................. 75 
Figure 3.4. Germination of untreated controls and burned treatments for the field  
germination study............................................................................................................ 76 
Figure 3.5. Effect of presence of pine cones during fires on post-burn germination of  
legume seeds for the field germination study ................................................................. 77 
Figure 3.6. Potential impacts of burned pine cones on survivorship and germination of seeds . 78 
ix 
 
 
 
 
 
List of Abbreviations 
 
 
ANOVA Analysis of Variance 
CEVI Centrosema virginianum 
DECI Desmodium ciliare 
DEFL Desmodium floridanum 
LEAN Lespedeza angustifolia 
MIQU Mimosa quadrivalvis 
PECA Pediomelum canescens 
RCB Randomized Complete Block 
SD Standard Deviation 
SE Standard Error 
TEVI Tephrosia virginiana 
TZ 2,3,5-Triphenyl Tetrazolium Chloride 
 
 
1 
 
 
 
 
 
CHAPTER 1 
INTRODUCTION 
 
PROJECT OVERVIEW 
Fire is an integral disturbance agent for many ecosystems around the world and a driver 
of vegetation distribution and diversity (Bond et al. 2005; Bond and Keeley 2005). Within these 
systems, fire?s effects range from altered survival and reproduction of individuals to altered 
successional patterns (Bond and van Wilgen 1996; DeBano et al. 1998). Furthermore, each fire is 
unique and variation in fuel loads, types, and arrangement coupled with species-specific 
tolerances to fire creates a heterogeneous suite of conditions and possible responses to a given 
fire (Whelan 1995; Bond and van Wilgen 1996).  
Frequent fire has played a critical role in the evolution, structure, and diversity of the 
longleaf pine (Pinus palustris Mill.) ecosystem of the southeastern United States (Frost 2006). 
Legumes are important components of this diverse ecosystem?s groundcover, contributing to the 
system?s overall floristic richness (Drew et al. 1998; Hainds et al. 1999), replenishing nitrogen 
lost during fires (Hendricks and Boring 1999; Hiers et al. 2003), and serving as valuable forage 
for a variety of game and non-game species (Yarrow and Yarrow 1999; Miller and Miller 2005). 
Fires have been demonstrated to stimulate germination of many leguminous species (reviewed in 
Baskin and Baskin 2004; Keeley and Fotheringham 2000) and, consequently, to influence 
legume germination and establishment in post-fire environments. 
 
2 
This project examined post-fire germination responses of a suite of native legumes 
common to longleaf pine stands in Georgia?s Gulf Coastal Plain. Potential mechanisms of fire-
related germination cues were investigated, including elevated temperatures, moist vs. dry heat, 
mechanical scarification, and smoke. Additionally, germination responses to fire heterogeneity 
were studied by manipulating fuels prior to fires. 
The following sections review literature pertinent to the role of fire on reproduction and 
recruitment of plants, physical seed dormancy in hard-seeded species, and heterogeneity of fuels 
and fires. Particular attention is paid to legumes within longleaf pine savannas. 
 
LITERATURE REVIEW 
Plant Populations 
Fires affect plant populations in complex ways (Whelan 1995; Bond and van Wilgen 
1996). Availability of resources (e.g., nutrients, water, light), individual survival, and 
reproduction may all be influenced by fires (Bond and van Wilgen 1996; DeBano et al. 1998); 
however, both direction (i.e., positive or negative) and magnitude of plant population responses 
to a given fire depend on numerous factors, such as season of burn, fire intensity, and species 
identity (Bond and van Wilgen 1996). 
Fires may kill certain species, while favoring others. For instance, Mallik and 
Gimingham (1985) found that several English heathland shrubs differed in their abilities to 
survive and resprout following experimental fires. Similarly, individuals within each species may 
exhibit variable responses to fire depending on age or size. Moreno and Oechel (1991b, 1993) 
found that while resprouting of chamise (Adenostoma fasciculatum Hook. & Arn.) shrubs in 
 
3 
California chaparral was reduced following fires, large individuals were better able to survive 
and resprout following high intensity fires than small individuals. 
Seedling recruitment may be influenced by seed availability, suitable microsite 
availability, or both (Eriksson and Ehrlen 1992; Setterfield 2002). Fires may increase the 
available seed pool following fires by increasing reproductive output. This phenomenon has been 
particularly well-documented within longleaf pine stands for numerous species including grasses 
(van Eerden 1997; Shepherd et al. In Press), legumes (Hiers et al. 2000), and other forb species 
(Platt et al. 1988a; Brewer and Platt 1994; Norden and Kirkman 2004). Seeds within the soil seed 
bank may also be stimulated to germinate by fires and, as previously noted for survival of mature 
individuals, seedling recruitment following fires may differ among species and within species 
(Bond and van Wilgen 1996). For example, Moreno and Oechel (1991a) used increased fuel 
loads to increase intensities of experimental fires in California chaparral. In that study, 
Adenostoma fasciculatum seedling recruitment decreased with increasing intensities, as did 
overall herbaceous recruitment; however, recruitment of the hard-seeded leguminous shrub Lotus 
strigosus (Nutt.) Greene increased with increasing intensities. Similarly, seedling recruitment of 
hard-seeded Acacia spp. was greater following high intensity fires than after low intensity fires 
in jarrah (Eucalyptus marginata Sm.) forests of Western Australia (Shea et al. 1979). 
Additionally, fires may create open sites suitable for seedling recruitment by killing potential 
competitors or removing litter (Bond and van Wilgen 1996). These open spaces also may have 
increased nutrients and water available for seedling use (Bond and van Wilgen 1996). 
Germination and Physical Dormancy 
Several factors associated with fire have been demonstrated to influence germination and 
survival of seeds. Fire-related germination cues that may operate in combination include, among 
 
4 
others, heat (Cushwa et al. 1968; Martin et al. 1975; Auld and O?Connell 1991), scarification of 
the seed coat (Teketay 1996), smoke (Clarke and French 2005; Lindon and Menges 2008), 
chemicals leached from charred plant material (Keeley 1987; Newton et al. 2006), and increased 
light (McPherson and Muller 1969; Christensen and Muller 1975; Keeley and Keeley 1987). 
Hard seed coats occur in at least 15 plant families, including, for example, the 
Anacardiaceae, Cannaceae, Convolvulaceae, Cucurbitaceae, Geranicaceae, Fabaceae, 
Malvaceae, Rhamnaceae, and Sapindaceae (Baskin and Baskin 1998; Baskin et al. 2000). While 
hard-seededness is common in several families, not all species within these families have hard 
seeds (Keeley and Fotheringham 2000). Hard seed coats impose physical dormancy on embryos 
in seeds by reducing permeability of the seed coat to the fluids and gases that are required for 
germination (Ballard 1973; Baskin and Baskin 1998). Often, once physical dormancy has been 
broken, seeds will readily germinate unless secondary physiological dormancy (e.g., requiring 
specific temperature or light regimes to germinate) is also present (Baskin and Baskin 1998; 
Baskin et al. 2000). Such secondary physiological dormancy occurs in at least seven families, 
including the Fabaceae (Baskin et al. 2000). Dormant seeds allow for the formation of persistent 
seed banks that can replenish aboveground plant populations following an environmental 
disturbance, such as fire (Baskin and Baskin 1998). Heat from fires is thought to stimulate 
germination of many hard-seeded species by scarifying seeds (i.e., disrupting the seed coat and 
making it permeable to water and gases), thus breaking this physical dormancy (Baskin and 
Baskin 1998). Both dry heat and moist heat have been found to increase germination rates of 
hard-seeded species (Martin et al. 1975). Furthermore, duration of exposure to elevated 
temperatures may influence seed survival and germination (Levitt 1980). 
 
5 
Smoke-induced germination has been found in a wide variety of plant species worldwide 
(Keeley and Fotheringham 2000; Van Staden et al. 2000). Keeley and Fotheringham (2000) 
assert that smoke-induced germination may be non-existent in the Fabaceae. Smoke may, 
however, play a secondary role in germination once the seed coat has been disrupted by fire 
because compounds in smoke that act as potential cues may persist in the soil or on plant tissues 
following a fire?s passage (Van Staden et al. 2000). Smoke is a complex substance and many 
attempts to isolate compounds that may influence germination have met with little success (Van 
Staden et al. 2000). Recently, however, two butenolide compounds have been isolated from 
smoke and found to have opposing effects on germination (Light et al. 2010). In their study of 
these butenolides, Light et al. (2010) found that 3-methyl-2H-furo[2,3-c]pyran-2-one (also called 
karrikin 1; So?s et al. 2010) promoted germination, while the related 3,4,5-trimethylfuran-2(5H)-
one inhibited germination. Furthermore, it is likely that a variety of other compounds in smoke 
may also influence germination (So?s et al. 2010). 
Fire Heterogeneity 
Variations in fuel loads, types, and arrangement are important determinants of fire 
intensity (Whelan 1995; Bond and van Wilgen 1996), which in turn influences the ecological 
effects of a given fire (Kennard and Outcalt 2006). Within longleaf pine stands, pine needles are 
an abundant fuel and may represent over half of the fuel load (Ottmar and Vihanek 2000; Ottmar 
et al. 2003). Furthermore, longleaf pine needles have higher energy contents than other potential 
fuels (e.g., live woody plants, other leaf litter, live and dead herbaceous species [Robertson 
2007]) possibly due to higher contents of terpenes, which are highly flammable (Orme?o et al. 
2009). Increased fire intensities may result from this combination of high energy content and 
abundance of pine needle litter under pine canopies. In the Florida sandhills, oaks and juvenile 
 
6 
pines located near large longleaf pine trees suffered greater mortality during fires than those 
located farther away (Williamson and Black 1981; Rebertus et al. 1989a,b), presumably due to 
increased fire intensity associated with increased fuel accumulation beneath the pine canopies. 
Other studies in longleaf pine stands have found similarly increased fire intensities associated 
with increased fuel loads (Thaxton and Platt 2006; Jack et al. 2010). 
In longleaf pine stands, the distribution of fuels is related to the distribution and density 
of pine trees (Platt et al. 1988b; Robertson 2005; Varner 2005; Mitchell et al. 2009), tending to 
increase concomitantly with stand basal area and proximity to mature trees. Fine fuels and coarse 
fuels from pine trees differ in their burning characteristics and may therefore differ in their 
ecological effects. In contrast to pine needles, which may burn for several minutes (Fonda 2001), 
pine cones may burn for well over half an hour (Fonda and Varner 2005). Furthermore, pine 
cones represent a potentially significant amount of fuel, with individual trees capable of 
producing >100 cones during a mast year (Boyer and White 1989; Mitchell et al. 2009), but the 
ecological effects of pine cones as fuel remain poorly understood (Fonda and Varner 2005). The 
impact of longleaf pine cones on fire effects is also likely greater following mast years, when 
pine cones are particularly abundant. 
 
OBJECTIVES 
The objectives of the research presented herein were to: 1) determine the species-specific 
effects of fire on legume seed germination and mortality using fire surrogates that represent 
environmental factors associated with exposure to fire (Chapter 2); and 2) examine the effects of 
variable fine fuel loads and the presence of longer burning fuels (pine cones) on the survivorship 
and germination of a suite of common legume species (Chapter 3). Specifically, in Chapter 2, six 
 
7 
legume species common to longleaf pine stands in southwestern Georgia were used to determine 
the effects of the following treatments: 1) elevated temperatures (five levels with dry heat 
temperatures ranging from presumably non-lethal to lethal) over various durations (three time 
intervals); 2) duration (three intervals) of moist heat (i.e., steam); 3) physical scarification; and 4) 
smoke. Chapter 3 presents two experiments: first, germination and survival of six legume species 
following experimental fires were assessed in the laboratory; and second, post-fire germination 
responses of seven legume species to fuel manipulations (both pine needles and pine cones) were 
determined in the field. 
 
REFERENCES 
Auld TD, O?Connell MA (1991) Predicting patterns of post-fire germination in 35 eastern 
Australian Fabaceae. Austral Ecol 16:53-70 
Ballard LAT (1973) Physical barriers to germination. Seed Sci Technol 1:285-303 
Baskin CC, Baskin JM (1998) Seeds: ecology, biogeography, and evolution of dormancy and 
germination. Academic Press, San Diego 
Baskin JM, Baskin CC (2004) A classification system for seed dormancy. Seed Sci Res 14:1-16 
Baskin JM, Baskin CC, Li X (2000) Taxonomy, anatomy and evolution of physical dormancy in 
seeds. Plant Species Biol 15:139-152 
Bond WJ, Keeley JE (2005) Fire as a global ?herbivore?: the ecology and evolution of flammable 
ecosystems. Trends Ecol Evol 20:387-394 
Bond WJ, van Wilgen BW (1996) Fire and plants. Chapman Hall, London 
Bond WJ, Woodward FI, Midgley GF (2005) The global distribution of ecosystems in a world 
without fire. New Phytol 165:525-538 
Boyer WD, White JB (1989) Natural regeneration of longleaf pine. In: Waldrop TA (ed) 
Proceedings of the symposium on the management of long leaf pine; 1989 April 4-6; 
Long Beach, MS. Gen Tech Rep SO-75. U.S. Department of Agriculture, Forest Service, 
Southern Forest Experiment Station, New Orleans, pp 94-113 
Brewer JS, Platt WJ (1994) Effects of fire season and herbivory on reproductive success in a 
clonal forb, Pityopsis graminifolia. J Ecol 82:665-675 
 
8 
Christensen NL, Muller CH (1975) Relative importance of factors controlling germination and 
seedling survival in Adenostoma chaparral. Am Mid Nat 93:71-78 
Clarke S, French K (2005) Germination response to heat and smoke of 22 Poaceae species from 
grassy woodlands. Aust J Bot 53:445-454 
Cushwa CT, Martin RE, Miller RL (1968) The effects of fire on seed germination. J Range 
Manag 21:250-254 
DeBano LF, Neary DG, Ffolliott PF (1998) Fire?s effects on ecosystems. John Wiley and Sons, 
New York 
Drew MB, Kirkman LK, Gholson AK (1998) The vascular flora of Ichauway, Baker County, 
Georgia: a remnant longleaf pine/wiregrass ecosystem. Castanea 63:1-24 
Eriksson O, Ehrlen J (1992) Seed and microsite limitation of recruitment in plant populations. 
Oecologia 91:360-364 
Fonda RW (2001) Burning characteristics of needles from eight pine species. For Sci 47:390-396 
Fonda RW, Varner JM (2005) Burning characteristics of cones from eight pine species. 
Northwest Sci 78:322-333 
Frost C (2006) History and future of the longleaf pine ecosystem. In: Jose S, Jokela EJ, Miller 
DL (eds) The longleaf pine ecosystem: ecology, silviculture, and restoration. Springer, 
New York, pp 9-42 
Hainds MJ, Mitchell RJ, Palik BJ, Boring LR, Gjerstad DH (1999) Distribution of native 
legumes (Leguminoseae) in frequently burned longleaf pine (Pinaceae)-wiregrass 
(Poaceae) ecosystems. Am J Bot 86:1606-1614 
Hendricks JJ, Boring LR (1999) N2-fixation by native herbaceous legumes in burned pine 
ecosystems of the southeastern United States. For Ecol Manag 113:167-177 
Hiers JK, Mitchell RJ, Boring LR, Hendricks JJ, Wyatt R (2003) Legumes native to longleaf 
pine savannas exhibit capacity for high N2-fixation rates and negligible impacts due to 
timing of fire. New Phytol 157:327-338 
Hiers JK, Wyatt R, Mitchell RJ (2000) The effects of fire regime on legume reproduction in 
longleaf pine savannas: is a season selective? Oecologia 125:521-530 
Jack SB, Hiers JK, Mitchell RJ, Gagnon JL (2010) Fuel loading and fire intensity?effects on 
longleaf pine seedling survival. In: Stanturf JA (ed) Proceedings of the 14th biennial 
southern silvicultural research conference. Gen. Tech. Rep. SRS-121. U.S. Department of 
Agriculture, Forest Service, Southern Research Station, Asheville, pp 275-279 
Keeley JE (1987) Role of fire in seed germination of woody taxa in California chaparral. 
Ecology 79:2320-2326 
 
9 
Keeley JE, Fotheringham CJ (2000) Role of fire in regeneration from seed. In: Fenner M (ed) 
Seeds: the ecology of regeneration in plant communities, 2nd edn. CAB International, 
Wallingford, pp 311-330 
Keeley JE, Keeley SC (1987) Role of fire in the germination of chaparral herbs and 
suffrutescents. Madro?o 34:240-249 
Kennard DL, Outcalt K (2006) Modeling spatial patterns of fuels and fire behavior in a longleaf 
pine forest in the southeastern USA. Fire Ecol 2:31-52 
Levitt J (1980) Responses of plants to environmental stresses. Vol. I: Chilling, freezing, and high 
temperature stresses, 2nd edn. Academic Press, New York 
Light ME, Burger BV, Staerk D, Kohout L, Van Staden J (2010) Butenolides from plant-derived 
smoke: natural plant-growth regulators with antagonistic actions on seed germination. J 
Nat Prod 73:267-269 
Lindon HL, Menges E (2008) Effects of smoke on seed germination of twenty species of fire-
prone habitats in Florida. Castanea 73:106-110 
Mallik AU, Gimingham CH (1985) Ecological effects of heather burning II. Effects on seed 
germination and vegetative regeneration. J Ecol 73:633-644 
Martin RE, Miller RL, Cushwa CT (1975) Germination response of legume seeds subjected to 
moist and dry heat. Ecology 56:1441-1445 
McPherson JK, Muller CH (1969) Allelopathic effects of Adenostoma fasciculatum, ?chamise?, 
in the California chaparral. Ecol Monog 39:177-198 
Miller JH, Miller KV (2005) Forest plants of the southeast and their wildlife uses. University of 
Georgia Press, Athens 
Mitchell RJ, Hiers JK, O?Brien JJ, Starr G (2009) Ecological forestry in the southeast: 
understanding the ecology of fuels. J For 107:391-397 
Moreno JM, Oechel WC (1991a) Fire intensity effects on germination of shrubs and herbs in 
southern California chaparral. Ecology 72:1993-2004 
Moreno JM, Oechel WC (1991b) Fire intensity and herbivory effects on postfire resprouting of 
Adenostoma fasciculatum in southern California chaparral. Oecologia 85:429-433 
Moreno JM, Oechel WC (1993) Demography of Adenostoma fasciculatum after fires of different 
intensities in southern California chaparral. Oecologia 96:95-101 
Newton RJ, Bond WJ, Farrrant JM (2006) Effects of seed storage and fire on germination in the 
nut-fruited Restionaceae species, Cannomois virgata. S Afr J Bot 72:177-180 
 
10 
Norden AH, Kirkman LK (2004) Factors controlling the fire-induced flowering response of the 
federally endangered Schwalbea americana L. (Scrophulariaceae). J Torrey Bot Soc 
131:16-22 
Orme?o E, C?spedes B, S?nchez IA, Velasco-Garc?a A, Moreno JM, Fernandez C, Baldy V 
(2009) The relationship between terpenes and flammability of leaf litter. For Ecol Manag 
257:471-482 
Ottmar RD, Vihnanek RE (2000) Stereo photo series for quantifying natural fuels. Volume VI: 
longleaf pine, pocosin, and marshgrass types in the Southeast United States. PMS 835. 
National Wildfire Coordinating Group, National Interagency Fire Center, Boise 
Ottmar RD, Vihnanek RE, Mathey JW (2003) Stereo photo series for quantifying natural fuels. 
Volume VIA: Sand hill, sand pine scrub, and hardwoods with white pine types in the 
south-east United States with supplemental sites for volume VI. PMS 838. National 
Wildfire Coordinating Group, National Interagency Fire Center, Boise 
Platt WJ, Evans GW, Davis MM (1988a) Effects of fire season on flowering of forbs and shrubs 
in longleaf pine forests. Oecologia 76:353-363 
Platt WJ, Evans GW, Rathbun SL (1988b) The population dynamics of a long-lived conifer 
(Pinus palustris). Am Nat 131:491-525 
Rebertus AJ, Williamson GB, Moser EB (1989a) Fire-induced changes in Quercus laevis spatial 
pattern in Florida sandhills. J Ecol 77:638-650 
Rebertus AJ, Williamson GB, Moser EB (1989b) Longleaf pine pyrogenicity and turkey oak 
mortality in Florida xeric sandhills. Ecology 70:60-70 
Robertson K (2005) Fire ecology: pineland fuel study. Tall Timbers Res Station Res Notes 
4(1):2 
Robertson K (2007) Fire ecology: energy levels of pineland fuels. Tall Timbers Res Station Res 
Notes 6(3):2 
Setterfield SA (2002) Seedling establishment in an Australian tropical savanna: effects of seed 
supply, soil disturbance and fire. J Appl Ecol 39: 949-959 
Shea SR, McCormick J, Portlock CC (1979) The effects of fires on regeneration of leguminous 
species in the northern jarrah (Eucalyptus marginata Sm.) forest of Western Australia. 
Aust J Ecol 4:195-205. 
Shepherd BJ, Miller DL, Thetford M (In Press) Fire season effects on flowering characteristics 
and germination of longleaf pine (Pinus palustris) savanna grasses. Restor Ecol 
So?s V, Sebesty?n E, Juh?sz A, Light MA, Kohout L, Szalai G, Tandori J, Van Staden J, Bal?zs 
E (2010) Transcriptome analysis of germinating maize kernels exposed to smoke-water 
and the active compound KAR1. BMC Plant Biol 10:236 
 
11 
Teketay D (1996) Germination ecology of twelve indigenous and eight exotic multipurpose 
leguminous species from Ethiopia. For Ecol Manag 80:209-223 
Thaxton JM, Platt WJ (2006) Small-scale fuel variation alters fire intensity and shrub abundance 
in a pine savanna. Ecology 87:1331-1337 
van Eerden BP (1997) Studies on the reproductive biology of wiregrass (Aristida stricta Michx.) 
in the Carolina sandhills. MSc Thesis, University of Georgia, Athens 
Van Staden J, Brown NAC, J?ger AK, Johnson TA (2000) Smoke as a germination cue. Plant 
Species Biol 15:167-178 
Varner JM (2005) Smoldering fire in long-unburned longleaf pine forests: linking fuels with fire 
effects. PhD Dissertation, University of Florida, Gainesville 
Whelan RJ (1995) The ecology of fire. Cambridge University Press, Cambridge 
Williamson GB, Black EM (1981) High temperature of forest fires under pines as a selective 
advantage over oaks. Nature 293:643-644 
Yarrow GK, Yarrow DT (1999) Managing wildlife. Sweetwater Press, Birmingham 
 
12 
 
 
 
 
 
CHAPTER 2 
MECHANISMS OF FIRE-INDUCED GERMINATION AND SURVIVAL OF SIX LEGUME 
SPECIES IN THE LONGLEAF PINE ECOSYSTEM 
 
13 
ABSTRACT 
Fire is an important driver of community diversity and structure for ecosystems 
worldwide. Fires affect competitive relationships between and within species by influencing 
individual survival, growth, reproduction, and recruitment. Physical seed dormancy via an 
impermeable seed coat is a characteristic of many species, especially those found within fire-
maintained ecosystems. In such fire prone sites, fire may enhance germination of these species. 
Understanding how fire-related factors affect germination can help predict fire regimes and 
predict how fire heterogeneity influences patterns of diversity within frequently burned systems. 
Six leguminous species common to the frequently burned longleaf pine ecosystem of 
southwestern Georgia, USA were used to examine the effects of: 1) elevated temperatures (five 
levels with dry heat temperatures ranging from presumably non-lethal to lethal) over three 
durations; 2) duration (three intervals) of moist heat (i.e., steam); 3) physical scarification; and 4) 
smoke. Legume species chosen represented a range of seed hardness, varying from soft-seeded 
to extremely hard-seeded. Response to both dry and moist heat was species-specific, but moist 
heat had the greatest potential to stimulate germination. The legumes studied also exhibited high 
heat tolerance, which may be a function of their low seed moisture contents. Smoke had no 
detectable effect on germination of most species, but may have inhibited germination of 
Centrosema virginianum. Potential ecological and managerial implications of these results are 
discussed. 
KEYWORDS: germination; fire; fire surrogates; longleaf pine ecosystem; legumes; Fabaceae 
 
14 
INTRODUCTION 
Fire is an important driver of plant diversity within many ecosystems worldwide (Whelan 
1995; Bond and van Wilgen 1996). However, each fire is unique and the effects of a given fire 
are not necessarily uniform throughout the area burned. The behavior, intensity, and severity of a 
given fire are strongly affected by a variety of factors, including, time since last burn, season of 
burn, fuel characteristics (e.g., type, arrangement, amount, and moisture), weather (e.g., 
temperature, humidity, and wind speed), and landscape position (Whelan 1995; DeBano et al. 
1998). Furthermore, within a given fire, fire intensities may exhibit marked heterogeneity at a 
variety of scales (Kennard and Outcalt 2006) resulting from patchy distribution of fuels (Hiers et 
al. 2009). 
Fires affect plant populations in complex ways (Whelan 1995). Availability of resources 
(e.g., nutrients, water, light), competitive relationships, survival, and reproduction may all be 
influenced by fires (Whelan 1995; DeBano et al. 1998). In particular, fire?s effects on plant 
reproduction and recruitment have been widely studied. Fires may promote or reduce post-fire 
vegetative reproduction (Hoffmann 1998; Liu and Menges 2005), fruit and seed production 
(Hiers et al. 2000; Norden and Kirkman 2004), and seedling recruitment (Setterfield 2002; Liu 
and Menges 2005). The magnitude and direction (i.e., positive or negative) of these effects, 
however, depends on species identity, season of burn, fire behavior, and location within the 
burned area (Knapp et al. 2009; Stephan et al. 2010). 
Leguminous species are an important component of many frequently burned systems, 
including, for example, North American prairies (Woodmansee et al. 1981; Leach and Givnish 
1996) and longleaf pine savannas (Drew et al. 1998; Hainds et al. 1999) as well as 
Mediterranean-type ecosystems worldwide (Arianoutsou and Thanos 1996). Because of their 
 
15 
ability to fix atmospheric nitrogen, legumes are important for replenishing nitrogen lost to 
volatilization during fires (Hendricks and Boring 1999; Hiers et al. 2003). Additionally, legumes 
provide forage for numerous game and non-game animal species (Yarrow and Yarrow 1999; 
Miller and Miller 2005). 
Legume abundance is often enhanced in frequently burned systems (e.g., Stoddard 1931; 
Chapman 1932; Lemon 1949; Towne and Knapp 1996). Increased light availability following 
fires may promote nitrogen-fixing species on nitrogen poor sites (Leach and Givnish 1996). In 
the absence of fire, these species may have a competitive disadvantage during seedling 
establishment because of inadequate light levels required for the energetically expensive 
symbiotic nitrogen fixation (Cathey 2005). Legume abundance may also be enhanced following 
fires if their soil-stored seeds require fire to germinate. Many members of the Fabaceae are hard-
seeded (Ballard 1973; Morrison et al. 1998). These hard seed coats impose physical dormancy on 
embryos of these species by reducing permeability of the seed coat to fluids and gases, thereby 
limiting hydration and subsequent germination (Ballard 1973; Baskin and Baskin 1998). Once 
the seed coat is scarified, however, these seeds often readily germinate (Baskin et al. 2000).  
Heat from fires is thought to stimulate germination of many hard-seeded species by 
scarifying seeds (i.e., disrupting the seed coat and making it permeable to water and gases), thus 
breaking this physical dormancy (Baskin and Baskin 1998). Heat may also directly stimulate 
embryos to germinate?as has been noted for several species in South Africa (Blommaert 1972; 
Musil and De Witt 1991; Van de Venter and Esterhuizen 1988). Both dry heat and moist heat 
have been found to increase germination rates of hard-seeded species (Martin et al. 1975). 
Furthermore, duration of exposure to elevated temperatures may influence seed survival and 
germination (Levitt 1980). In addition to heat, chemicals in smoke (Clarke and French 2005; 
 
16 
Lindon and Menges 2008), ash, and charate (Keeley 1987; Newton et al. 2006) have also been 
found to stimulate germination. 
Understanding how fire-related factors affect seed germination, vegetative recruitment, 
and seed production is a preliminary step for predicting how fire regimes and fire heterogeneity 
may influence patterns of diversity within frequently burned systems (Driscoll et al. 2010), such 
as the longleaf pine (Pinus palustris Mill.) ecosystem of the southeastern United States. The 
necessity of process-based restoration of the longleaf pine community is becoming increasingly 
recognized (e.g., Van Lear et al. 2005; Mitchell et al. 2005, 2006; Kirkman et al. 2006). Thus, a 
better understanding of germination and recruitment strategies of important groundcover species 
may improve the development of ecologically sustainable management tools. 
Longleaf pine stands once covered more than 37 million hectares (Ware et al. 1993), but 
have been reduced to less than 5% of their original extent (Noss 1989). Fire has played a critical 
role in the evolution, structure, and diversity of the longleaf pine ecosystem (Frost 2006). Natural 
longleaf pine communities have highly diverse groundcover layers, which may have >40 species 
per square meter (Peet and Allard 1993; Kirkman et al. 2001) and are home to numerous 
endemic and threatened species (Hardin and White 1989). Groundcover species account for the 
high plant species richness found within natural longleaf pine stands that are frequently burned 
(Drew et al. 1998; Kirkman et al. 2004). Knowledge of specific fire-related germination cues 
within the longleaf pine system is limited to only a small fraction of the system?s numerous 
species (e.g., Cushwa et al. 1968; Martin et al. 1975; Young et al. 2007; Bolin 2009). 
The goal of this study was to determine the species-specific effects of fire on legume seed 
germination and mortality, using fire surrogates that represent components of environmental 
factors associated with fire. Specifically, for six legume species common to longleaf pine stands 
 
17 
in southwestern Georgia, the effects of the following treatments were examined: 1) elevated 
temperatures (five levels with dry heat temperatures ranging from presumably non-lethal to 
lethal) over various durations (three); 2) duration (three) of moist heat (i.e., steam); 3) physical 
scarification; and 4) smoke. Additionally, the legume seeds chosen for this study represent a 
range of seed hardness, varying from relatively soft-seeded to extremely hard-seeded. 
 
MATERIALS AND METHODS 
Study Species 
Six legume species native to the southeastern United States were selected for this study: 
Centrosema virginianum (L.) Benth., Crotalaria rotundifolia Walter ex J.F. Gmel., Desmodium 
floridanum Chapm., Lespedeza angustifolia (Pursh) Elliot, Mimosa quadrivalvis L., and 
Rhynchosia reniformis DC; nomenclature follows Wunderlin and Hansen (2003). These species 
were selected based on their perennial life history, common occurrence throughout natural 
longleaf pine stands within the study site, and because they represent a range of seed hardness 
(based on scarification tests). Similar to other Desmodium species (Martin et al. 1975; Coffey 
and Kirkman 2006), D. floridanum is soft-seeded and does not typically require scarification for 
germination. As determined by earlier germination trials (L.K. Kirkman, unpublished data), Ce. 
virginianum is moderately hard-seeded, requiring some scarification to germinate (2 s in a 
Forsberg? tumble scarifier lined with 150 grit sandpaper; Forsberg, Inc., Thief River Falls, 
Minnesota, USA), while L. angustifolia, R. reniformis, M. quadrivalvis, and Cr. rotundifolia are 
extremely hard-seeded, requiring more intensive scarification (L.K. Kirkman, unpublished data: 
9, 9, 15, and 17 s, respectively) for maximum germination to occur. 
 
18 
All seeds, except those of R. reniformis, were harvested in 2005 or 2006 from a native 
seed production garden of the Joseph W. Jones Ecological Research Center at Ichauway 
(Ichauway) in Baker County, Georgia, USA. Rhynchosia reniformis seeds were harvested from 
natural longleaf pine stands at Ichauway in 2007. Seeds were stored in burlap or paper bags in a 
climate-controlled laboratory at 21?C prior to experimentation. 
Fire Effect Surrogate Treatments 
All species were subjected to heat treatments representing a range of fire conditions: 1) 
untreated control; 2) factorial combinations of dry heat at five temperatures (40, 60, 80, 100, and 
150?C) and three durations (1, 4, and 10 min); 3) moist heat (i.e., steam) at four durations (0.16, 
1, 4, and 10 min); 4) manual scarification (seed coat abrasion); and 5) exposure to pine needle 
smoke. Heat treatment durations were chosen to represent a range of potential exposure times 
experienced during actual fires (Heyward 1938; Beadle 1940; Robertson 2007). Four replicates 
of 20 seeds species-1 for each treatment and control (20 seeds ? 22 treatments ? 4 replicates = 
1,760 seeds species-1) were used. 
DRY HEAT ? Dry heat treatments were applied by placing seeds of all species in Pyrex? 
glass petri dishes and heating each replicate dish in one of four ovens. Timing of each dry heat 
treatment period began when appropriate treatment temperatures (? 3?C) were reached, as 
indicated by a chromel-alumel (K-type) thermocouple suspended in the center of each oven. 
Exposure to 60?C for short periods of time has been found to be lethal to plant tissue, although 
seeds often have higher lethal temperature thresholds than other plant tissues (Levitt 1980; 
DeBano et al. 1998). 
For the purposes of this study, it was assumed that 40?C for the durations used would not 
be lethal or induce germination; therefore, 40?C was used in lieu of untreated controls for the 
 
19 
statistical analyses described below. Temperatures near the soil surface may reach or exceed 
40?C on a hot summer day within longleaf pine stands in southwest Georgia (G. Iacona, 
unpublished data). Furthermore, after examining the effects of heat on 35 legume species, Auld 
and O?Connell (1991) concluded that temperatures of 40?C were not sufficient to induce 
germination of these species. 
MOIST HEAT ? Moist heat treatments were applied by placing seeds into a nylon mesh 
bag and suspending the bag 2.5 cm from the nozzle of an active commercial steam cleaner 
(Vapor Clean TR 4, Commital-Sami, Marostica, Italy) for the appropriate durations. A chromel-
alumel thermocouple was placed immediately behind each mesh bag to record initial and final 
temperatures of steam for each replicate. 
SCARIFICATION ? Seeds were manually scarified by nicking the seed coat opposite to each 
seed?s hilum with a razor. 
SMOKE ? A custom-built ?smoker? was used to expose seeds to smoke derived from pine 
needles. The smoker consisted of one upright and one inverted steel bucket (30 cm diameter by 
35 cm deep) attached 30 cm above the upright bucket?s base using a 6 cm ? 15 cm steel pipe. 
Air-dried longleaf pine needles were burned in the upright bucket and smoke was passively 
transferred to the inverted bucket via the connecting pipe after sealing the upright bucket with a 
steel lid. Seeds were then placed in Pyrex? glass petri dishes under the smoker and exposed to 
smoke for 10 min. To minimize heat transfer into the smoke chamber (i.e., the inverted bucket), 
water mist was sprayed on the exterior of the smoker to cool it. Temperatures associated with 
smoke treatments were monitored with a K-type thermocouple placed in the center of the smoke 
chamber and level with the seeds. Maximum temperatures were recorded for each replicate 
treatment. 
 
20 
Data Collection 
GERMINATION ? Immediately following the treatments, seeds of each species were placed 
in covered plastic petri dishes (20 seeds per dish) on filter paper (Whatman No. 1) moistened 
with deionized water. Seeds were then allowed to germinate for three weeks (January-February 
2008) in a heated (10?C minimum) greenhouse at Ichauway under ambient winter light 
conditions. Baskin and Baskin (1998) note that 2-4 weeks are typically adequate durations for 
germination experiments. Furthermore, previous experience with these and related species 
indicated that fungal attacks on seeds increase with the time seeds are exposed to damp filter 
paper (pers. obs.). Based on these personal observations and similar ones by Baskin and Baskin 
(1998), three weeks were considered adequate to observe most germination while minimizing 
fungal infestation. Seeds were monitored for germination (defined as protrusion of the radicle 
through the seed coat) three times per week. Rotten seeds (identified by extensive visible fungal 
infestation and/or being easily crushed under light pressure) were considered dead (Baskin and 
Baskin 1998). Following each round of data collection, germinated and rotten seeds were 
removed from petri dishes and discarded. Germination response for each replicate was 
determined as the proportion of seeds germinated. 
VIABILITY AND MORTALITY ASSESSMENTS ? Following the three week greenhouse 
germination period, viability of all remaining seeds was determined using a 2,3,5-triphenyl 
tetrazolium chloride (TZ) test (Peters 2000). To perform the TZ test, the seed coats of all 
remaining seeds were cut with a razor opposite the hilum and the seeds were allowed to imbibe 
water on filter paper moistened with deionized water in petri dishes for 12 to 24 h. Following 
imbibition of water, excess moisture was blotted from the filter paper, the paper was remoistened 
with 1.0% TZ, and then the covered petri dishes were wrapped in aluminum foil to eliminate 
 
21 
light contamination and stored in a laboratory at 21?C for 48 h. Seeds were considered viable if 
the radicle and cotyledons were both stained or if a radicle had emerged from a seed coat. 
Unstained seeds and seeds with only cotyledons or radicle stained were considered non-viable 
(Peters 2000). Seed mortality was calculated as the total percentage of dead (rotten and non-
viable) seeds for each replicate.  
SEED WEIGHT AND MOISTURE CONTENT ? Mean seed weights and percent moisture 
content by weight were determined for each species. Six non-scarified seeds of each species 
were weighed individually and then oven-dried at 80?C. Seed weights were recorded every other 
day for six days, until seed weights stabilized. The difference in seed weight before and after 
drying was then used to calculate percent moisture content. 
Statistical Analyses 
Data from each species were analyzed separately. All analyses were performed using the 
SAS software package (v9.1, SAS Institute, Cary, North Carolina). Germination and mortality 
percentages were normalized prior to analyses using arcsine square root transformations. A two-
way ANOVA (Littell et al. 2002) was used to analyze seed germination and mortality responses 
to dry heat exposure time and temperature. A one-way ANOVA (Littell et al. 2002) was used to 
determine if germination or mortality varied in response to steam duration. For this analysis, we 
included the control treatment as a zero minute duration. Where ANOVAs indicated significant 
differences between treatments, a Tukey?s test (? = 0.05; Littell et al. 2002) was used for post-
hoc pairwise means comparisons. Separate two-sample t-tests (Littell et al. 2002) were used to 
compare germination and mortality responses of untreated controls with smoke and manually 
scarified treatments. For the t-tests, Satterthwaite corrections were used to adjust the degrees of 
freedom when F-tests indicated that variances were unequal (Littell et al. 2002). Species 
 
22 
experiencing significantly enhanced germination following heat treatments were considered to 
have heat-induced germination, while those that merely survived were considered heat tolerant. 
 
RESULTS 
UNTREATED CONTROLS ? Germination of untreated seeds tended to follow patterns of 
relative seed hardness, with hard-seeded species typically exhibiting lower germination rates 
than soft-seeded species. Low mortality of controls (Table 2.1) indicated that initial viability of 
all species was high. Despite high viability, germination of the untreated hard-seeded species 
(Cr. rotundifolia, L. angustifolia, M. quadrivalvis, and R. reniformis) was low (i.e., <25%). In 
contrast, over half of the untreated seeds from the soft-seeded D. floridanum and the moderately 
hard-seeded Ce. virginianum germinated over the course of the study. 
DRY HEAT ? All species, regardless of relative seed hardness, exhibited the capacity to 
survive exposure to high temperatures. With the exception of the most extreme treatment (10 
min exposure to 150?C), which killed most or all seeds of all species, mortality did not differ 
among the combinations of temperature and durations tested for any species (Fig. 2.1, Table 2.2). 
Germination responses were more variable than mortality responses. Increasing 
temperatures above 40?C did not affect germination of M. quadrivalvis and R. reniformis (Fig. 
2.1, Table 2.3). In contrast, increasing temperatures did significantly affect germination of Ce. 
virginianum, Cr. rotundifolia, and D. floridanum. Duration of elevated temperatures also 
affected germination of Ce. virginianum, Cr. rotundifolia, and D. floridanum (Table 2.3). Post-
hoc analyses did not indicate any germination differences between durations for Cr. rotundifolia 
or D. floridanum and only indicated significant differences between 1 and 4 min durations for 
Ce. virginianum. Furthermore, the interaction between duration and temperature significantly 
 
23 
affected germination of all species except M. quadrivalvis and R. reniformis (Table 2.3). 
Germination of Ce. virginianum, Cr. rotundifolia, and D. floridanum was significantly reduced 
after exposure to 10 min of 150?C dry heat (Table 2.3; Fig. 2.1). Crotalaria rotundifolia tolerated 
high temperatures, but exhibited a more complex response than the other species studied (Table 
2.3; Fig. 2.1). Specifically, germination was enhanced following 4 and 10 min exposures to 80?C 
and 100?C and seeds survived brief (1 and 4 min) exposures to 150?C without significantly 
increased mortality. Germination of L. angustifolia also was increased by high temperatures, but 
only following exposure to 100?C for 10 min (Table 2.3; Fig 2.1). 
MOIST HEAT ? Steam influenced both germination and mortality of most species (Table 
2.4, Fig. 2.2). Only germination of R. reniformis, which was low regardless of treatment, did not 
significantly respond to variation in steam duration. Furthermore, all species exhibited similar 
response patterns to increased duration of steam, with brief exposure to steam (0.16 min) tending 
to increase germination without significantly affecting mortality of any species (Table 2.4; Fig. 
2.2). Similarly, although their germination remained relatively low, the hard-seeded species L. 
angustifolia and R. reniformis experienced 5.8- and 2.6-fold increased germination (respectively) 
following 0.16 min exposure to steam. Species with the hardest seed coats (M. quadrivalvis and 
Cr. rotundifolia) experienced greatest germination rates following 1 min exposure to steam. 
Exposure to 4 and 10 min of steam significantly increased mortality of seeds from all species 
regardless of seed hardness. 
SCARIFICATION ? Manual scarification increased germination rates of all species and had 
no effect on mortality (Table 2.1). Hard-seeded species experienced the greatest increases in 
germination, increasing by 3.8? to 19.5? that of untreated controls (Cr. rotundifolia and L. 
angustifolia, respectively). Scarification also significantly increased germination of the soft-
 
24 
seeded D. floridanum and moderately hard-seeded Ce. virginianum by 1.3? and 1.8?, 
respectively. 
SMOKE ? Exposure to pine needle smoke typically did not affect germination or seed 
mortality (Table 2.1). The only exception was Ce. virginianum, which experienced a 65% 
reduction in germination compared with controls (t = 3.106, P = 0.039). However, this reduced 
germination of Ce. virginianum was not accompanied by significantly increased mortality (t = -
1.003, P = 0.391). 
TREATMENT TEMPERATURES ? Temperatures of dry heat, steam, and smoke treatments 
are listed in Table 2.5. Actual temperatures of dry heat treatments averaged to within <1.5?C of 
their prescribed temperatures. Final temperatures of steam treatments typically were between 
about 80-90?C. Final smoke temperatures tended to be 20-25?C, but occasionally briefly rose to 
30-50?C during treatments; however, spraying the exterior of the smoke chamber with water mist 
rapidly cooled interior temperatures and limited these increases to less than 30 s. 
SEED WEIGHT AND MOISTURE CONTENT ? Of seeds used in the study, those of L. 
angustifolia were the smallest, weighing 1.99 ? 0.55 mg (mean ? SD dry weight), followed by 
Cr. rotundifolia (2.65 ? 0.47 mg), D. floridanum (3.97 ? 0.69 mg), M. quadrivalvis (8.18 ? 1.90 
mg), R. reniformis (11.77 ? 1.34 mg), and Ce. virginianum (12.33 ? 3.02 mg). Seed moisture 
content of species used in this study was relatively low, typically less than 10%. Mimosa 
quadrivalvis and R. reniformis had the lowest seed moisture contents (4.3 ? 1.5% and 5.3 ? 
1.1%, respectively). Highest moisture contents were found in seeds of Ce. virginianum (9.9 ? 
0.8%) and Cr. rotundifolia (9.0 ? 0.6%). Desmodium floridanum and L. angustifolia had 
moisture contents between these extremes (7.4 ? 0.6% and 6.7 ? 1.5%, respectively). 
 
 
25 
DISCUSSION 
Overall, heat treatments enhanced germination of five of the six species examined in the 
current study. The finding that heat exposure enhanced germination of nearly all of the legume 
species examined is consistent with fire-related germination cues reported for other legume 
species of the longleaf pine system in the southeastern United States (e.g., Cushwa et al. 1968; 
Martin et al. 1975; Young et al. 2007). Similar to these previous studies, the degree of 
germination responses to individual temperature, duration, and moisture conditions of the heat 
treatments varied by species. 
The observed increases in germination following mechanical scarification of the seed 
coat suggest that disruption of the seed coat is the likely mechanism by which heat-stimulated 
germination occurs within the species studied. This is consistent with observations of other 
authors (reviewed in Murdoch and Ellis 2000). Weak spots on the seed coat such as the lens 
(lens-shaped area of raised tissue near the hilum that is also called the strophiole by some 
authors) or hilum (scar where the seed was attached to the funiculus during development) may be 
cracked during heat (Murdoch and Ellis 2000; Baskin et al. 2000). Heat may increase pressure 
within seeds due to water vaporization (Hanna 1984) or cause differential swelling and shrinking 
of cells (Serrato-Valenti et al. 1995), thereby cracking weak spots or other areas of the seed coat. 
During the present study, cracked seed coats of individual seeds were occasionally noticed 
following application of the 100?C and 150?C dry heat treatments. Similarly, seeds subjected to 
moist heat treatments, particularly for durations of ?1 min, were often noticeably larger and 
softer post-treatment than pre-treatment. Steam appeared to have penetrated the seed coat and 
subsequently hydrated the seeds.  
 
26 
Only R. reniformis germination was not enhanced by either dry or moist heat treatments. 
One explanation may be the length of storage since seed harvest. Unlike the one- and two-year-
old seeds of other species obtained for this study, seeds of R. reniformis were harvested only a 
few months prior to the experiment. Baskin and Baskin (1998) noted that germination results of 
seeds requiring a period of after-ripening may differ if tested prior to or following after-ripening 
of the seeds. After-ripening may increase the ambient temperatures required for germination. It is 
not known if after-ripening of R. reniformis occurs; however, nearly 90% of manually scarified 
R. reniformis seeds germinated, so it is unlikely that after-ripening, if required, influenced the 
observed results. Another fire-related cue?such as ash, leachates, higher temperatures, longer 
durations of increased heat, or some combination thereof, may influence germination of this 
species. Alternatively, it is possible that fire simply does not enhance germination of R. 
reniformis. 
Differential germination responses to moist versus dry heat have also been reported 
previously (Cushwa et al. 1968; Martin et al. 1975; Kauffman and Martin 1991; Herranz et al. 
1998, 1999; Le Fer and Parker 2005). Of the species studied herein, dry heat enhanced 
germination of two species, whereas moist heat enhanced germination of five species. Of the two 
species enhanced by exposure to dry heat (Cr. rotundifolia and L. angustifolia), Cr. rotundifolia 
appears to require the most intense heat environment to elicit germination. Germination of the 
moderately hard-seeded Ce. virginianum and soft-seeded D. floridanum remained high at all but 
the most extreme combination of temperature and duration (i.e., 150?C for 10 min). Reduced 
germination following the most intense heat shock is similar to results of Rivas et al. (2006), who 
found that germination of six Australian legume species was reduced after exposures to 5 and 10 
min of 150?C. Interestingly, under similar temperatures and durations, the soft-seeded D. 
 
27 
floridanum and moderately hard-seeded Ce. virginianum had similar or higher germination rates 
than the hard-seeded Cr. rotundifolia and L. angustifolia.  
Brief exposure to moist heat (i.e., ?0.16 min) increased germination of all species except 
R. reniformis. Germination of D. floridanum and Ce. virginianum appears to be most sensitive to 
increasing duration of moist heat, while germination of Cr. rotundifolia and M. quadrivalvis 
appears to be most tolerant of increasing duration of moist heat. These observations generally 
follow patterns of relative seed hardness, where germination of the softer seeded species is 
reduced by shorter durations of moist heat than germination of the harder seeded species. In 
addition, the data indicate that some soft-seeded species, such as Desmodium spp., may benefit 
from scarification. 
Consistent with earlier studies of legumes within the longleaf pine ecosystem (e.g., 
Cushwa et al. 1968; Martin et al. 1975; Young et al. 2007), the legume species studied herein 
exhibited the capacity to survive exposures to temperatures that are generally considered fatal for 
other plant tissues (i.e., ?60?C). Furthermore, this study demonstrated that herbaceous legume 
seeds have the capacity to survive temperatures 40-50?C higher than previously noted (i.e., 150 
?C) for this system (Cushwa et al. 1968; Martin et al. 1975); however, seeds of both leguminous 
and non-leguminous species have been reported to survive brief (i.e., 1-5 min) exposures to 
200?C in Africa (Gashaw and Michelsen 2002). Indeed, with the exception of D. floridanum, a 
small proportion of seeds (up to 40%) of all species examined herein survived exposure to 150?C 
dry heat for 10 min, although mortality increased significantly at this extreme exposure. 
Various properties of seeds have been shown to be associated with the ability to survive 
high temperatures, including seed coat thickness (DeBano et al. 1998),  internal moisture content 
(Levitt 1980), and size (Williams et al. 2003; Overbeck et al. 2006). Little is known about the 
 
28 
physical properties (e.g., thickness) of the seed coats of the species studied herein. As noted 
above, relatively hard-seeded species tolerated greater exposure to moist heat; however, no 
similar pattern was evident for dry heat treatments. Species with small seeds may survive 
exposures to high temperatures better than larger seeded species (Williams et al. 2003; Overbeck 
et al. 2006), but no such pattern was apparent in the current study. My observations, however, 
are qualitative because statistical comparisons were not made between species. The relatively 
low seed moisture contents of the species in this study are similar to values reported for hard-
seeded species (Morrison et al. 1998) and this characteristic is typically associated with increased 
ability to tolerate elevated temperatures (Levitt 1980; Le Fer and Parker 2005). 
Heat-induced germination of hard seeds requires adequate heat to scarify seeds, but not 
so much heat that seeds are killed (Banda et al. 2006), suggesting that both temperature and 
duration are important. This study provides some evidence regarding upper threshold 
temperatures for seed survival or induced germination when exposed to dry heat. Additionally, 
the interaction of temperature and soil moisture was not examined in this study, although other 
studies indicate that this may be important because of moist soil?s ability to buffer elevated 
temperatures of brief fires (Cushwa et al. 1968; Martin et al. 1975; Le Fer and Parker 2005). As 
described here and elsewhere (e.g., Levitt 1980; Herranz et al. 1999; Banda et al. 2006), duration 
of elevated temperatures (dry or moist heat) appears to be critical for seed survival and 
germination of these species. Tolerance of high temperatures may allow seeds to persist within 
the soil seed bank until environmental conditions favor germination. Alternatively, species with 
heat-stimulated germination may find the post-fire environment favorable for seedlings (Bond 
and van Wilgen 1996) because of increased availability of nutrients, light, and open space. 
 
29 
In particular, hard-seeded species requiring scarification to break dormancy may be 
favored in fire-maintained systems (Whelan 1995; Towne and Knapp 1996). Dormant seeds may 
survive in the soil seed bank for years or even many decades (Baskin and Baskin 1998; Fenner 
and Thompson 2005) and, thus, provide a buffer against localized extirpation of their species 
(Thompson et al. 1998; Stocklin and Fisher 1999). Evidence from a buried seed bag study 
indicates that hard-seeded legumes may have the capacity to form a long-term persistent seed 
bank within the longleaf pine system (L.K. Kirkman, unpublished data). Persistent seed banks 
may maintain genetic diversity within frequently burned systems (Dolan et al. 2008) or promote 
species spread by colonizing open gaps following fires. 
Although smoke-induced germination has been found in a wide variety of plant species 
worldwide (Keeley and Fotheringham 2000; Van Staden et al. 2000), the present study found no 
evidence that longleaf pine needle smoke induces germination of the observed legume species. 
While it is possible that plant-derived smoke from sources other than longleaf pine needles may 
induce germination in these species, the results of this study are consistent with Keeley and 
Fotheringham?s (2000) assertion that smoke-induced germination may be non-existent in the 
Fabaceae. In addition, Van Staden et al. (2000) note that smoke may play a secondary role in 
germination once the seed coat has been disrupted by fire because compounds in smoke that act 
as potential cues may persist in the soil or on plant tissues following a fire?s passage. 
The only species apparently influenced by smoke was Ce. virginianum, which 
experienced reduced germination following smoke exposure. The observed reduction in 
germination was not accompanied by increased mortality and, therefore, may represent a 
potential inhibition response to a chemical (or chemicals) in longleaf pine needle smoke. Such an 
inhibitory effect of smoke has been reported for various species around the world (Keeley and 
 
30 
Fotheringham 2000), while Lindon and Menges (2008) found similarly decreased germination of 
the Florida native legume Crotalaria pallida (Aiton) following 10 and 30 min smoke exposures. 
The butenolide 3,4,5-trimethylfuran-2(5H)-one has recently been isolated from smoke and 
identified as a germination inhibitor (Light et al. 2010), but its presence in longleaf pine-derived 
smoke is presently unknown. Further investigation is required to determine if germination of Ce. 
virginianum is in fact inhibited by longleaf pine needle-derived smoke. 
Restoration efforts within the longleaf pine ecosystem may be enhanced by knowledge of 
germination requirements of a suite of functionally important species as well as their responses 
to particular fire regimes (Driscoll et al. 2010). Results presented herein provide additional 
information regarding potential mechanisms of germination and survival of legume seeds within 
this system. However, caution must be used when attempting to apply these results to responses 
of legume seeds in the field. First, seeds of these species may respond differently to fire-related 
cues in soil compared to cues established in the lab. For example, Christensen and Muller (1975) 
found higher germination of a variety of herbaceous chaparral species when seeds were heated in 
chaparral soils compared to seeds exposed to heat without soil in the lab. The authors attributed 
those differences to the likely presence of chaparral leachates in the soils. Second, the behavior 
and impact of each fire is unique (Whelan 1995). Variation in fuel features (load, type, 
distribution, and moisture), weather, and local topography all interact to create fires that are 
spatially and temporally diverse. Within longleaf pine stands distribution of fuel loads and types 
has been found to vary at fine scales (i.e., 0.1-10 m2; Hiers et al. 2009) which may in turn 
influence local intensities and residence times of potentially lethal temperatures (Thaxton and 
Platt 2006; Hiers et al. 2009). Companion studies (presented in Chapter 3) were performed to 
investigate how fuel-mediated variation in fire effects may alter recruitment of plant species via 
 
31 
differential germination and survival of seeds following fires. In particular, these companion 
studies indicate that fuels that increase durations of lethal temperatures (such as pine cones) may 
create a mosaic of open spaces available for recruitment of species from the soil seed bank that 
are able tolerate the increased intensity of the fine-scale fire environment. Thus, those species 
that survive and germinate within these open spaces may have an initial competitive advantage 
over neighboring individuals that are unable to take advantage of the increased resource 
availability within these open spaces. 
 
REFERENCES 
Arianoutsou M, Thanos CA (1996) Legumes in fire-prone Mediterranean regions: an example 
from Greece. Int J Wild Fire 6:77-82 
Auld TD, O?Connell MA (1991) Predicting patterns of post-fire germination in 35 eastern 
Australian Fabaceae. Austral Ecol 16:53-70 
Ballard LAT (1973) Physical barriers to germination. Seed Sci Technol 1:285-303 
Banda T, Schwartz MW, Caro T (2006) Effects of fire on germination of Pterocarpus 
angolensis. For Ecol Manag 233:116-120 
Baskin CC, Baskin JM (1998) Seeds: ecology, biogeography, and evolution of dormancy and 
germination. Academic Press, San Diego 
Baskin JM, Baskin CC, Li X (2000) Taxonomy, anatomy and evolution of physical dormancy in 
seeds. Plant Spec Biol 15:139-152 
Beadle NCW (1940) Soil temperatures during forest fires and their effect on the survival of 
vegetation. J Ecol 28:180-192 
Blommaert KLJ (1972) Buchu seed germination. J S African Bot 38:237-239. 
Bolin JF (2009) Heat shock germination responses of three eastern North American temperate 
species. Castanea 74:160-167 
Bond WJ, van Wilgen BW (1996) Fire and plants. Chapman Hall, London 
Cathey SE (2005) Growth and N2-fixation of legumes native to the longleaf-wiregrass 
ecosystem. MSc Thesis, University of Florida, Gainesville 
 
32 
Chapman HH (1932) Is the longleaf type a climax? Ecology 13:328-334 
Christensen NL, Muller CH (1975) Effects of fire on factors controlling plant growth in 
Adenostoma chaparral. Ecolog Monogr 45:29-55 
Clarke S, French K (2005) Germination response to heat and smoke of 22 Poaceae species from 
grassy woodlands. Aust J Bot 53:445-454 
Coffey KL, Kirkman LK (2006) Seed germination strategies of species with restoration potential 
in a fire-maintained pine savanna. Nat Areas J 26:289-299 
Cushwa CT, Martin RE, Miller RL (1968) The effects of fire on seed germination. J Range 
Manag 21:250-254 
DeBano LF, Neary DG, Ffolliott PF (1998) Fire?s effects on ecosystems. John Wiley and Sons, 
New York 
Dolan RW, Quintano-Ascencio PF, Menges ES (2008) Genetic change following fire in 
populations of a seed-banking perennial plant. Oecologia 158:355-360 
Drew MB, Kirkman LK, Gholson AK (1998) The vascular flora of Ichauway, Baker County, 
Georgia: a remnant longleaf pine/wiregrass ecosystem. Castanea 63:1-24 
Driscoll DA, Lindenmayer DB, Bennett AF, Bode M, Bradstock RA, Cary GJ, Clarke MF, 
Dexter N, Fensham R, Friend G, Gill M, James S, Kay G, Keith DA, MacGregor C, 
Russel-Smith J, Salt D, Watson JEM, Williams RJ, York A (2010) Fire management for 
biodiversity conservation: key research questions and our capacity to answer them. Biol 
Conserv 143:1928-1939 
Fenner M, Thompson K (2005) The ecology of seeds. Cambridge University Press, Cambridge 
Frost C (2006) History and future of the longleaf pine ecosystem. In: Jose S, Jokela EJ, Miller 
DL (eds) The longleaf pine ecosystem: ecology, silviculture, and restoration. Springer, 
New York, pp 9-42 
Gashaw M, Michelsen A (2002) Influence of heat shock on seed germination of plants from 
regularly burnt savanna woodlands and grasslands in Ethiopia. Plant Ecol 159:83-93 
Hainds MJ, Mitchell RJ, Palik BJ, Boring LR, Gjerstad DH (1999) Distribution of native 
legumes (Leguminosae) in frequently burned longleaf pine (Pinaceae)-wiregrass 
(Poaceae) ecosystems. Am J Bot 86:1606-1614 
Hanna PJ (1984) Anatomical features of the seed coat of Acacia kempeana (Mueller) which 
relate to increased germination rate induced by heat treatment. New Phytol 96:23-29 
Hardin ED, White DL (1989) Rare vascular plant taxa associated with wiregrass (Aristida 
stricta) in the southeastern United States. Nat Areas J 9:234-245 
 
33 
Hendricks JJ, Boring LR (1999) N2-fixation by native herbaceous legumes in burned pine 
ecosystems of the southeastern United States. For Ecol Manag 113:167-177 
Herranz JM, Ferrandis P, Martinez-Sanchez JJ (1998) Influence of heat on seed germination of 
seven Mediterranean Leguminosae species. Plant Ecol 136:95-103 
Herranz JM, Ferrandis P, Martinez-Sanchez JJ (1999) Influence of heat on seed germination of 
nine woody Cistaceae species. Int J Wildl Fire 9:173-182 
Heyward F (1938) Soil temperatures during fire in the longleaf pine region. J For 36:478-491 
Hiers JK, Wyatt R, Mitchell RJ (2000) The effects of fire regime on legume reproduction in 
longleaf pine savannas: is a season selective? Oecologia 125:521-530 
Hiers JK, Mitchell RJ, Boring LR, Hendricks JJ, Wyatt R (2003) Legumes native to longleaf 
pine savannas exhibit capacity for high N2-fixation rates and negligible impacts due to 
timing of fire. New Phytol 157:327-338 
Hiers JK, O?Brien JJ, Mitchell RJ, Grego JM, Loudermilk EL (2009) The wildland fuel cell 
concept: an approach to characterize fine-scale variation in fuels and fire in frequently 
burned longleaf pine forests. Int J Wildl Fire 18:315-325 
Hoffmann WA (1998) Post-burn reproduction of woody plants in a neotropical savanna: the 
relative importance of sexual and vegetative reproduction. J Appl Ecol 35:422-433 
Kauffman JB, Martin RE (1991) Factors influencing the scarification and germination of three 
montane Sierra Nevada shrubs. Northwest Sci 65:180-187 
Keeley JE (1987) Role of fire in seed germination of woody taxa in California chaparral. 
Ecology 79:2320-2326 
Keeley JE, Fotheringham CJ (2000) Role of fire in regeneration from seed. In: Fenner M (ed) 
Seeds: the ecology of regeneration in plant communities, 2nd edn. CAB International, 
Wallingford, pp 311-330 
Kennard DL, Outcalt K (2006) Modeling spatial patterns of fuels and fire behavior in a longleaf 
pine forest in the southeastern USA. Fire Ecol 2:31-52 
Kirkman LK, Coffey KL, Mitchell RJ, Moser EB (2004) Ground cover recovery patterns and 
life-history traits: implications for restoration obstacles and opportunities in a species-rich 
savanna. J Ecol 92:409-421 
Kirkman LK, Mitchell RJ, Helton RC, Drew MB (2001) Productivity and species richness across 
an environmental gradient in a fire-dependent ecosystem. Am J Bot 88:2119?2128 
Kirkman LK, Mitchell RJ, Kaeser MJ, Pecot SD, Coffey KL (2006) The perpetual forest: using 
undesirable species to bridge restoration. J Appl Ecol 44:604-614 
 
34 
Knapp EE, Estes BL, Skinner CN (2009) Ecological effects of prescribed fire season: a literature 
review and synthesis for managers. Gen Tech Rep PSW-GTR-224. U.S. Department of 
Agriculture, Forest Service, Pacific Southwest Research Station, Albany 
Leach MK, Givnish TJ (1996) Ecological determinants of species loss in remnant prairies. 
Science 273:1555-1558 
Le Fer D, Parker VT (2005) The effect of seasonality of burn on seed germination in chaparral: 
the role of soil moisture. Madro?o 52:166-174 
Lemon PC (1949) Successional responses of herbs in the longleaf-slash forest after fire. Ecology 
30:135-145 
Levitt J (1980) Responses of plants to environmental stresses. Vol. I: Chilling, freezing, and high 
temperature stresses, 2nd edn. Academic Press, New York 
Light ME, Burger BV, Staerk D, Kohout L, Van Staden J (2010) Butenolides from plant-derived 
smoke: natural plant-growth regulators with antagonistic actions on seed germination. J 
Nat Prod 73:267-269 
Lindon HL, Menges E (2008) Effects of smoke on seed germination of twenty species of fire-
prone habitats in Florida. Castanea 73:106-110 
Littell RC, StroupWW, Freund RJ (2002) SAS for linear models, 4th edn. SAS Institute, Cary 
Liu H, Menges ES (2005) Winter fires promote greater vital rates in the Florida Keys than 
summer fires. Ecology 86:1483-1495 
Martin RE, Miller RL, Cushwa CT (1975) Germination response of legume seeds subjected to 
moist and dry heat. Ecology 56:1441-1445 
Miller JH, Miller KV (2005) Forest plants of the southeast and their wildlife uses. University of 
Georgia Press, Athens 
Mitchell RJ, Franklin JF, Palik BJ, Kirkman LK, Smill LL, Engstrom RT, Hunter ML (2005) 
Natural disturbance-based silviculture for restoration and maintenance of biological 
diversity. National Commission on Science for Sustainable Forestry, Washington, DC. 
Available at http://ncseonline.org/ewebeditpro/items/O62F3299.pdf (Accessed 15 
November 2010) 
Mitchell RJ, Hiers JK, O?Brien JJ, Jack SB, Engstrom RT (2006) Silviculture that sustains: the 
nexus between silviculture, frequent prescribed fire, and conservation of biodiversity in 
longleaf pine forests of the southeastern United States. Can J For Res 36:2724-2736 
Morrison DA, McClay K, Porter C, Rish S (1998) The role of the lens in controlling heat-
induced breakdown of testa-imposed dormancy in native Australian legumes. Ann Bot 
82:35-40 
 
35 
Murdoch AJ, Ellis RH (2000) Dormancy, viability and longevity. In: Fenner M (ed) Seeds: the 
ecology of regeneration in plant communities, 2nd edn. CABI Publishing, New York, pp 
183-214 
Musil CF, De Witt DM (1991) Heat-stimulated germination in two Restionaceae species. S Afr J 
Bot 57:175-176 
Newton RJ, Bond WJ, Farrrant JM (2006) Effects of seed storage and fire on germination in the 
nut-fruited Restionaceae species, Cannomois virgata. S Afr J Bot 72:177-180 
Norden AH, Kirkman LK (2004) Factors controlling the fire-induced flowering response of the 
federally endangered Schwalbea americana L. (Scrophulariaceae). J Torrey Bot Soc 
131:16-22 
Noss RF (1989) Longleaf pine and wiregrass: keystone components of an endangered ecosystem. 
Nat Areas J 9:211-213 
Overbeck GE, Muller SC, Pillar VD, Pfadenhauer J (2006) No heat-stimulated germination 
found in herbaceous species from burned subtropical grassland. Plant Ecol 184:237-243 
Peet RK, Allard DJ (1993) Longleaf pine vegetation of the southern Atlantic and eastern Gulf 
Coast regions: a preliminary classification. Proc Tall Timbers Fire Ecol Conf 18:45-81 
Peters J (2000) Tetrazolium testing handbook. Contribution No. 29 to the handbook on seed 
testing. Association of Official Seed Analysts, Lincoln 
Rivas M, Reyes O, Casal M (2006) Influence of heat and smoke treatments on the germination of 
six leguminous shrubby species. Int J Wildl Fire 15:73-80 
Robertson K (2007) Fire ecology: fire behavior notes. Tall Timbers Research Notes 6:2 
Serrato-Valenti G, De Vries M, Cornara L (1995) The hilar region in Leucaena leucocephala 
Lam. (De Wit) seed: structure, histochemistry and role of the lens in germination. Ann 
Bot 75:569-574 
Setterfield SA (2002) Seedling establishment in an Australian tropical savanna: effect of seed 
supply, soil disturbance and fire. J Appl Ecol 39:949-959 
Stephan K, Miller M, Dickinson MB (2010) First-order fire effects on herbs and shrubs: present 
knowledge and process modeling needs. Fire Ecol 6:95-114 
Stocklin J, Fisher M (1999) Plants with longer-lived seeds have lower local extinction rates in 
grassland remnants 1950-1985. Oecologia 120:539-543 
Stoddard HL (1931) The bobwhite quail: its habits, preservation and increase. Charles Scribner?s 
Sons, New York 
 
36 
Thaxton JM, Platt WJ (2006) Small-scale fuel variation alters fire intensity and shrub abundance 
in a pine savanna. Ecology 87:1331-1337 
Thompson K, Bakker JP, Bekker RM, Hodgson JG (1998). Ecological correlates of seed 
persistence in soil in the northwest European flora. J Ecol 86:163-169 
Towne EG, Knapp AK (1996) Biomass and density responses in tallgrass prairie legumes to 
annual fire and topographic position. Am J Bot 83:175-179 
Van de Venter HA, Esterhuizen AD (1988) The effects of factors associated with fire on seed 
germination of Erica sessiliflora and E. hebecalyx (Ericaceae). S Afr J Bot 54:301-304 
Van Lear DH, Carroll WD, Kapeluck PR, Johnson R (2005) History and restoration of the 
longleaf pine-grassland ecosystem: implications for species at risk. For Ecol Manag 
211:150-165 
Van Staden J, Brown NAC, J?ger AK, Johnson TA (2000) Smoke as a germination cue. Plant 
Species Biol 15:167-178 
Ware S, Frost CC, Doerr PD (1993) Southern mixed hardwood forest: the former longleaf pine 
forest. In: Martin WH, Boyce SG, Echternacht AC (eds) Biodiversity of the Southern 
United States: Lowland Terrestrial Communities. John Wiley & Sons, New York, pp 
447-493 
Whelan RJ (1995) The ecology of fire. Cambridge University Press, Cambridge 
Williams PR, Congdon RA, Grice AC, Clarke PJ (2003) Fire-related cues break seed dormancy 
of six legumes of tropical eucalypt savannas in north-eastern Australia. Austral Ecol 
28:507-514 
Woodmansee RG, Vallis I, Mott JJ (1981) Grassland nitrogen: In: Clark FR, Rosswall T (eds) 
Terrestrial Nitrogen Cycles. Ecological Bulletin 33. Swedish Natural Science Research 
Council, Stockholm, pp 443-462 
Wunderlin RP, Hansen BF (2003) Guide to the vascular plants of Florida, 2nd edn. University 
Press of Florida, Gainesville 
Yarrow GK, Yarrow DT (1999) Managing wildlife. Sweetwater Press, Birmingham 
Young AS, Chang SM, Sharitz RR (2007) Reproductive ecology of a federally endangered 
legume, Baptisia arachnifera, and its more widespread congener, B. lanceolata 
(Fabaceae). Am J Bot 94:228-236 
 
37 
 
Species Treatment Germination (%)  Mortality (%) 
Ce. virginianum Control 50.6 ? 1.9 B a  0.0 ? 0.0 A a 
 Scarified 93.6 ? 3.9 A   6.4 ? 3.9 A  
 Smoke 33.0 ? 5.1  b  1.3 ? 1.3  a 
Cr. rotundifolia Control 21.8 ? 2.5 B a  21.9 ? 3.5 A a 
 Scarified 83.8 ? 2.4 A   16.3 ? 2.4 A  
 Smoke 21.4 ? 5.1  a  12.8 ? 4.5  a 
D. floridanum Control 63.8 ? 3.1 B a  15.0 ? 3.5 A a 
 Scarified 84.3 ? 3.2 A   15.7 ? 3.2 A  
 Smoke 61.9 ? 3.7  a  6.3 ? 2.4  a 
L. angustifolia Control 5.0 ? 3.5 B a  5.0 ? 2.0 A a 
 Scarified 97.5 ? 1.4 A   2.5 ? 1.4 A  
 Smoke 7.3 ? 4.2  a  3.8 ? 3.8  a 
M. quadrivalvis Control 11.3 ? 3.8 B a  4.8 ? 3.4 A a 
 Scarified 98.8 ? 1.3 A   0.0 ? 0.0 A  
 Smoke 6.3 ? 1.3  a  1.3 ? 1.3  a 
R. reniformis Control 5.1 ? 3.0 B a  8.8 ? 1.2 A a 
 Scarified 87.4 ? 2.5 A   12.6 ? 2.5 A  
  Smoke 3.8 ? 1.3  a  11.3 ? 2.4  a 
 
Table 2.1. Legume seed germination and mortality responses to untreated (control), scarification, 
and smoke treatments. Within each response, different uppercase letters indicate significant 
differences between control and scarified seeds of each species, while different lowercase letters 
indicate significant differences between controls and smoke treatments as determined by separate 
two-sample t-tests (? = 0.05). 
 
38 
 
Species Treatment df F P 
Ce. virginianum Temperature 2,45 4.97 <0.0001 
 Time 4,45 13.37 0.011 
 Temp. ? Time 8,45 6.62 <0.0001 
Cr. rotundifolia Temperature 2,45 14.53 <0.0001 
 Time 4,45 10.08 <0.0001 
 Temp. ? Time 8,45 8.54 <0.0001 
D. floridanum Temperature 2,45 6.20 0.004 
 Time 4,45 21.72 <0.0001 
 Temp. ? Time 8,45 8.75 <0.0001 
L. angustifolia Temperature 2,45 11.27 <0.001 
 Time 4,45 25.22 <0.0001 
 Temp. ? Time 8,45 13.37 <0.0001 
M. quadrivalvis Temperature 2,45 3.36 0.043 
 Time 4,45 2.40 0.064 
 Temp. ? Time 8,45 3.65 0.002 
R. reniformis Temperature 2,45 2.61 0.084 
 Time 4,45 11.63 <0.0001 
 Temp. ? Time 8,45 5.19 <0.001 
 
Table 2.2. Mortality responses of legume seeds to dry heat treatments. Where ANOVAs were 
significant, Tukey?s mean separation tests (? = 0.05) always indicated that mortality at 150?C, 10 
min, and their combination (i.e., 150?C for 10 min) were significantly greater than all others. 
Post-hoc analyses did not indicate any other significant differences. Means are illustrated in Fig. 
2.1. 
 
39 
 
Treatment Ce. virginianum Cr. rotundifolia D. floridanum 
  df F P df F P df F P 
Temperature 
(?C) 2,45 5.44 0.007 2,45 0.29 0.747 2,45 2 0.146 
40   a       
60   a       
80   a       
100   a       
150   b       
Time (min) 4,45 9.92 <0.0001 4,45 5.21 0.001 4,45 16.08 <0.0001 
1   a   a   a 
4   ab   a   a 
10   b   a   a 
Temp. ? Time 8,45 6.65 <0.0001 8,45 9.57 <0.0001 8,45 9.34 <0.0001 
40,1   a   bcd   a 
40,4   a   cd   a 
40,10   a   de   a 
60,1   a   bcd   a 
60,4   a   de   a 
60,10   a   cd   a 
80,1   a   cd   a 
80,4   a   cd   a 
80,10   a   ab   a 
100,1   a   cd   a 
100,4   a   abce   a 
100,10   a   a   a 
150,1   a   abce   a 
150,4   a   abc   a 
150,10     b     d     b 
 
Table 2.3. (Continued on next page.) Germination of legume seeds following dry heat 
treatments. Different letters under P-values indicate significant differences between response 
variables within a species using Tukey?s mean separation tests (? = 0.05). Means are illustrated 
in Fig. 2.1. 
 
40 
 
Treatment L. angustifolia M. quadrivalvis R. reniformis 
  df F P df F P df F P 
Temperature 
(?C) 2,41 0.32 0.731 2,45 2.3 0.112 2,45 1.21 0.307 
40          
60          
80          
100          
150          
Time (min) 4,41 1.68 0.173 4,45 0.29 0.882 4,45 2.4 0.063 
1          
4          
10          
Temp. ? Time 8,41 4.78 <0.001 8,45 1.7 0.125 8,45 0.41 0.906 
40,1   ab       
40,4   b       
40,10   b       
60,1   ab       
60,4   ab       
60,10   b       
80,1   ab       
80,4   ab       
80,10   ab       
100,1   b       
100,4   b       
100,10   a       
150,1   ab       
150,4   ab       
150,10     b             
 
Table 2.3. (Continued from previous page.) 
 
41 
 
Species Response df F P Duration (min) 
     0 0.16 1 4 10 
Ce. virginianum Germination 4,15 24.23 <0.0001 ab a ab b c 
 Mortality 4,15 38.19 <0.0001 c c c b a 
Cr. rotundifolia Germination 4,15 7.49 0.001 b a a a ab 
 Mortality 4,15 5.24 0.007 b b b ab a 
D. floridanum Germination 4,15 35.53 <0.0001 ab a bc c d 
 Mortality 4,15 26.23 <0.0001 cd d cb b a 
L. angustifolia Germination 4,15 5.88 0.004 b a ab ab b 
 Mortality 4,15 7.41 0.001 b b b ab a 
M. quadrivalvis Germination 4,15 7.17 0.002 b ab a a a 
 Mortality 4,15 7.19 0.001 b b b ab a 
R. reniformis Germination 4,15 2.43 0.093      
 Mortality 4,15 19.57 <0.0001 c c bc b a 
 
Table 2.4. Germination and mortality of legume seeds following moist heat treatments. Different 
letters under duration indicate significant differences between moist heat treatment times as 
determined by Tukey?s mean separation tests (? = 0.05). Means are illustrated in Fig. 2.2. 
 
42 
 
Treatment Duration (min) 
 0.16 1 4 10 
Dry Heat (?C)     
40  39.0 ? 0.4 38.7 ? 0.7 39.9 ? 0.6 
60  59.4 ? 0.6 59.7 ? 0.6 59.7 ? 0.6 
80  78.7 ? 0.6 79.0 ? 0.5 79.5 ? 0.7 
100  99.1 ? 0.7 99.3 ? 0.7 99.2 ? 0.6 
150  148.8 ? 0.8 148.8 ? 0.8 149.2 ? 0.8 
Steam 91.0 ? 0.7 89.2 ? 0.7 79.1 ? 3.9 82.5 ? 1.6 
Smoke    22.1 ? 1.1 
 
Table 2.5. Temperatures (mean ? SE) of dry heat, steam, and smoke treatments. Temperatures of 
dry heat treatments represent means of initial and final temperatures (n = 8). Values for steam 
and smoke represent mean final temperatures (n = 4). 
 
43 
 
Figure 2.1. Mortality and germination responses of legume seeds to dry heat treatments. Values 
represent mean germination (solid lines) or mortality (dashed lines). Error bars represent ? one 
SE. Results of statistical analyses for mortality and germination can be found in Tables 2.2 and 
2.3, respectively. 
Ce. virginianum Cr. rotundifolia 
Pe
rce
nt 
Ge
rmi
na
tion
 an
d Mor
tali
ty 
 
44 
 
 
Figure 2.2. Mortality and germination responses of legume seeds to steam. Values represent 
mean germination (solid lines) or mortality (dashed lines). Error bars represent ? one SE. 
ANOVA results and post-hoc Tukey?s means separations test results are presented in Table 2.4. 
Ce. virginianum Cr. rotundifolia 
Perc
en
t G
ermi
na
tion
 an
d M
ortal
ity
 
 
45 
 
 
 
 
 
CHAPTER 3 
IMPACTS OF FINE-SCALE VARIATION IN SURFACE FIRE ENVIRONMENT ON 
LEGUME GERMINATION IN THE LONGLEAF PINE ECOSYSTEM 
 
46 
ABSTRACT 
By influencing fire behavior, fine-scale fuel heterogeneity may influence patterns of plant 
recruitment and regeneration from soil seed banks. In particular, hard-seeded species, such as 
many members of the Fabaceae, may experience differential germination and survival as a result 
of heterogeneous fire intensities mediated by variable distribution of fine fuel. Post-fire 
germination was examined for nine legume species native to longleaf pine stands (Pinus 
palustris) of the southeastern United States in response to fuel variation using laboratory and 
field studies. For both studies, fine fuel loads (longleaf pine needles) and woody fuels (presence 
or absence of longleaf pine cones) were manipulated. Germination and mortality were assessed 
for six species after exposure to experimental laboratory fires. Germination of seven species was 
assessed for four months in the field following experimental burns of small plots. Burning 
reduced germination compared to unburned controls in both studies. Furthermore, legumes 
exposed to burning pine cones experienced greater reductions in germination than when exposed 
to burning pine needles alone. Manipulation of fine fuel loads did not affect germination in either 
study. Additionally, a hypothetical model of pine cone effects on recruitment from the soil seed 
bank following fires is presented. 
KEYWORDS: fuel heterogeneity; fine-scale fire heterogeneity; legumes; germination; seed 
survival. 
 
 
47 
INTRODUCTION 
Recurrent fires maintain the structure and diversity of numerous ecosystems worldwide 
(Bond and Keeley 2005; Bond et al. 2005). Within these systems, variations in fuel loads, types, 
and arrangement are important determinants of fire behavior and intensity (Whelan 1995; Bond 
and van Wilgen 1996), which in turn influence the ecological effects of a given fire (Kennard 
and Outcalt 2006). Even at fine scales (i.e., <10 m2) and in frequently burned ecosystems, 
woodland fuels may exhibit markedly heterogeneous distribution (Hiers et al. 2009), which may 
influence fire?s impacts at similarly small scales. The importance of fine-scale fire heterogeneity 
as a potential driver of plant diversity within fire-maintained systems is becoming increasingly 
recognized (e.g., Rebertus et al. 1989; Mitchell et al. 2006; Thaxton and Platt 2006). 
An example of an exceptionally species rich and fire-dependent ecosystem is the longleaf 
pine (Pinus palustris Mill.) savanna of the southeastern United States (Peet and Allard 1993). 
This system?s diverse groundcover and abundant pine needle litter provide fine fuels necessary 
for carrying fires across the forest floor (O?Brien et al. 2008). Bunch grasses common to this 
system?Aristida spp. and Andropogon spp.?are highly flammable and serve as perches for 
other fine fuels, such as pine needles (Hendricks et al. 2002; Hermann 2007). Together, the 
bunch grasses and perched fine fuels provide a continuous fuel layer and greatly enhance the 
pyrogenicity of longleaf pine stands. Groundcover species alone, however, may not sustain fires 
if pine litter, particularly pine needles, is absent or occurs in insufficient abundance (O?Brien et 
al. 2008). 
In longleaf pine savannas, accumulation of needles and woody litter under pine trees may 
increase fine-scale fire intensities (Mitchell et al. 2009; Jack et al. 2010). Indeed, increased pine 
litter and associated increases in fire temperatures and intensities have been found to reduce the 
 
48 
survival and growth of oaks (Williamson and Black 1981; Rebertus et al. 1989). Elevated pine 
fuel loads have also been found to reduce the survival and resprouting of shrubs in longleaf pine 
stands (Thaxton and Platt 2006) and increase mortality of longleaf pine seedlings (Jack et al. 
2010). 
Fine fuels and coarse fuels from pine trees differ in their burning characteristics and may 
therefore differ in their ecological effects. In contrast to pine needles, which may burn for only a 
few minutes (Fonda 2001), pine cones may burn for well over half an hour (Fonda and Varner 
2005). Although pine cones represent a potentially significant source of fuel (Mitchell et al. 
2009), the ecological effects of pine cones as a fuel source remain poorly understood (Fonda and 
Varner 2005). Because distribution of cones is influenced by the distribution and density of pine 
trees (Platt et al. 1988; Varner 2005; Mitchell et al. 2009), the impact of longleaf pine cones on 
fire effects is likely greater following mast seeding years when pine cones are particularly 
abundant. 
Legume species are an important component of the groundcover flora in some longleaf 
pine forests; in some cases, accounting for nearly 10% of species present (Drew et al. 1998) and 
up to 120,000 stems per ha (Hainds et al. 1999). Within this ecosystem, legumes provide forage 
for wildlife and are important sources of nitrogen replenishment to the soil following loss due to 
volatilization in fires (Hendricks and Boring 1999; Hiers et al. 2003). Many leguminous species 
are hard-seeded, possessing an impermeable seed coat (Ballard 1973; Morrison et al. 1998). Hard 
seed coats impose a physical dormancy on seeds of these species by limiting uptake of water by 
the embryo (Ballard 1973) and thus allow for the formation of persistent seed banks (Baskin and 
Baskin 1998). Exposure to fire enhances germination of some hard-seeded species by breaking 
the moisture barrier that imparts dormancy (Morrison et al. 1998); however, little is known about 
 
49 
how small-scale fire variation may affect species distribution by influencing seed germination or 
mortality of individuals of a species. 
Seeds within the soil seed bank are uniquely positioned to take advantage of new gaps 
created by fires as recruitment sites, provided they can survive the fires. Within ecosystems that 
are regularly burned, the capacity for seeds to survive fires or germinate in response to these fires 
differs among species (Keeley et al. 1985; Clark and Wilson 1994; Danthu et al. 2003). Thus, 
species-specific responses to fine-scale fuel and fire variation may contribute to patterns of 
species recruitment and fine-scale vegetation diversity. 
The objective of this study was to examine effects of variable fine fuel loads and 
presence of longer burning fuels on germination and mortality of a suite of common legume 
species. First, germination and mortality of six legume species following experimental fires 
using a gradient of fuel loads were assessed in the laboratory. Second, post-fire germination 
responses of seven legume species to fuel (i.e., pine needle and pine cones) manipulations were 
determined in the field. 
 
MATERIALS AND METHODS 
Study Site 
Seed germination responses to fuel manipulation treatments were examined in the 
laboratory and in field plots at the Joseph W. Jones Ecological Research Center at Ichauway 
(Ichauway). Ichauway is a 12,000 ha research property located in Baker County, Georgia, USA 
and occurs within the Lower Gulf Coastal Plain ecoregion and within the historic range of 
longleaf pine (Ware et al. 1993). The local climate is characterized by short, mild winters and 
 
50 
long, hot summers (Lynch et al. 1986) with average annual temperatures ranging from lows of 
5?C in winter to highs of 34?C in the summer (Goebel et al. 1997). 
Study Designs and Data Collection 
EXPERIMENT 1: LABORATORY GERMINATION 
Seed mortality and germination were examined in response to ignition of variable fuel 
loads under laboratory conditions. The laboratory setting allowed for ready retrieval of the seeds 
for further testing. Post-treatment retrieval of seeds permitted me to quantify seed mortality as a 
factor in the absence of germination response. 
In this study, seeds of six legume species were used (nomenclature follows Wunderlin 
and Hansen [2003]): Centrosema virginianum (L.) Benth., Crotalaria rotundifolia Walter ex J.F. 
Gmel., Desmodium floridanum Chapm., Lespedeza angustifolia (Pursh) Elliot, Mimosa 
quadrivalvis L., and Rhynchosia reniformis DC. These species were selected based on seed 
availability, common occurrence within Ichauway?s longleaf pine stands, their perennial life 
histories, and characteristics that permitted ready identification of seeds of each species during 
post-treatment retrieval. Additionally, the species represent a range of seed coat hardness as 
determined by their scarification requirements using a Forsberg? tumble scarifier lined with 150 
grit sandpaper (Forsberg, Inc., Thief River Falls, Minnesota). Germination trials following 
scarification indicate that D. floridanum is soft-seeded and does not typically require 
scarification for germination; Ce. virginianum is moderately hard-seeded, requiring some 
scarification (2 s); and Cr. rotundifolia, L. angustifolia, M. quadrivalvis, and R. reniformis are 
extremely hard-seeded, requiring more extensive scarification to elicit germination (9, 9, 15, and 
17 s, respectively; L.K. Kirkman, unpublished data). Rhynchosia reniformis seeds were 
harvested by hand from natural longleaf pine stands at Ichauway in October 2007. Seeds of all 
 
51 
other species used were harvested in 2005 and 2006 from a native seed production garden at 
Ichauway. Seeds were stored in burlap or paper bags in a climate-controlled laboratory at 
approximately 21?C prior to use.  
For this study, four replicate sets of 20 seeds species-1 were used for untreated control and 
four fire treatments (20 seeds ? 5 treatments ? 4 replicates = 400 seeds species-1). Fuel load 
treatment rates were selected to approximate typical accumulation of longleaf pine needle litter 
between 1-2 and 3-4 years since the last fire in natural stands (Ottmar and Vihanek 2000; Ottmar 
et al. 2003). The low fuel treatment was equivalent to 2.8 Mg/ha of longleaf pine needles, while 
the high fuel treatment was equivalent to 5.7 Mg/ha of pine needles. Longleaf pine needles were 
dried for 24 h in an oven at 70?C prior to use. Two longleaf pine cones (65 to 85 g per cone [150 
g total] after oven-drying for 24 h at 70?C) were used for pine cone treatments. Prior to applying 
fire treatments, a 61 ? 61 ? 7 cm wooden box was filled with approximately 5 cm of sand (125-
500 ?m grain size). The commercially prepared (i.e., washed and sieved) sand was oven-dried 
for 4 days at 70?C and allowed to cool for 12 h prior to use. Prior to each treatment, a 6 cm 
diameter hemispherical steel sieve (approx. 1 mm) was buried level with the sand surface in the 
center of the box. Next, seeds of all six species were uniformly sowed across the sand within the 
sieve, lightly pressed into the sand, and then covered with 1-2 mm of sand to minimize direct 
contact with burning material. Following this, pine needles were spread uniformly over a 707 
cm2 circular area (centered over the sieve). For pine cone fuel treatments, pine cones were placed 
side-by-side, lengthwise, in opposing directions on top of the pine needles directly above the 
sieve. 
A portable furnace was used to ignite fuels uniformly for the fire treatments in early 
March 2008. Similar to that described by van Eerden (1997), the circular propane-fueled portable 
 
52 
aluminum furnace (31 cm diameter by 40 cm high) was mounted on a tripod and could be raised 
and lowered as needed using a hand-crank cable mechanism. To ignite the fuel, the furnace was 
lowered to within 2.5 cm of the sand?s surface, was allowed to stand for 10 s, and then raised. 
Four chromel-constantan (E-type) thermocouples were placed on the sand surface in a 15 cm ? 
15 cm square around the sieve and then lightly covered with 1-2 mm of sand (i.e., approximate 
seed burial depth). Thermocouple data were recorded at 1 s intervals using a CR3000 data logger 
(Campbell Scientific, North Logan, Utah, USA). Results from the companion study presented in 
Chapter 2 indicated that brief exposure (i.e., ?10 min) to temperatures ?60?C were not lethal to 
these legume seeds; therefore, when temperatures recorded by all thermocouples fell below 
60?C, seeds were separated from the sand by sifting through a 1 mm wire-mesh sieve. Maximum 
temperature (from any one of the four thermocouples) and duration of temperatures ?60?C (i.e., 
potentially lethal elevated temperatures [Hare 1961] averaged among the four thermocouples) 
were recorded for each treatment. 
Following treatments, seeds were separated by species, counted, and placed on filter 
paper (Whatman No. 1) moistened with deionized water in plastic petri dishes. Seeds were then 
allowed to germinate in the laboratory under ambient light at 21?C for three weeks in March 
2008. Baskin and Baskin (1998) note that 2-4 weeks are typically adequate durations for 
germination experiments. Furthermore, previous experience with these and related species 
indicated that fungal attacks of seeds increase with the time seeds are exposed to damp filter 
paper (pers. obs.). Based on these personal observations and similar to those of Baskin and 
Baskin (1998), three weeks were considered adequate to observe most germination while 
minimizing fungal infestation. Germination (defined as protrusion of the radicle through the seed 
coat) was assessed three times per week. Germinated seeds were removed and discarded after 
 
53 
each biweekly assessment. Additionally, rotten seeds (identified by extensive visible fungal 
infestation and/or by being easily crushed using minimal pressure) were counted and discarded 
following each assessment. Rotten seeds were counted as dead (Baskin and Baskin 1998). 
After the three-week germination period, viability of remaining seeds was assessed using 
a 2,3,5-triphenyl tetrazolium chloride (TZ) test (Peters 2000). Following manual scarification of 
the seed coat with a razor, seeds were placed on filter paper moistened with deionized water in 
petri dishes for 12 to 24 h. Following imbibition of water, excess moisture was blotted from the 
filter paper and then the paper was remoistened with a 1.0% TZ solution. The covered petri 
dishes were then wrapped in aluminum foil and stored in a laboratory at 21?C for 48 h, after 
which time seeds were dissected to assess the embryo for staining. Seeds were considered viable 
if the radicle and cotyledons were both stained or if a radicle had emerged from the seed coat. 
Unstained seeds and seeds with only cotyledons or radicle stained were considered non-viable 
(Peters 2000). 
EXPERIMENT 2: FIELD FIRE INTENSITY MEASURES AND GERMINATION 
To examine germination response of legumes to variable fire intensities caused by pre-
fire fuel variation, seeds of seven common perennial species were used: four that were also used 
in the laboratory experiment (Ce. virginianum, D. floridanum, L. angustifolia, and M. 
quadrivalvis) plus three additional species, Desmodium ciliare (Muhl. ex Willd.) DC., 
Pediomelum canescens (Michx.) Rydb., and Tephrosia virginiana (L.) Pers. Similar to the 
laboratory study described above, these species were selected because they represent a range of 
seed hardness, they are relatively common within longleaf pine stands of Ichauway, and because 
adequate seed quantities were available. Based on scarification requirements for germination as 
described above, D. ciliare and D. floridanum are soft-seeded; Ce. virginianum is moderately 
 
54 
hard-seeded (2 s scarification); and L. angustifolia, M. quadrivalvis, P. canescens, and T. 
virginiana are extremely hard-seeded, requiring greater scarification (9, 15, 7, and 7 s, 
respectively). Seeds were collected from Ichauway?s native seed production garden in 2005 or 
2006. 
This study was performed in three blocks of planted wiregrass (Aristida stricta Michx.) 
beds located within a native plant garden on Ichauway. Soils of the garden plot consisted of 
loamy sand (Orangeburg series, Typic Kandiudult). Planted wiregrass clumps occurred at a 
density of 5 m-2 to approximate natural densities (Clewell 1989). 
A randomized complete block split-plot design of two fine fuel loads and two pine cone 
(i.e., presence or absence) treatments was used for burning treatments. Within the three wiregrass 
blocks, fine fuel loads were used as whole plots and pine cone treatments were the split-plots. 
Within each of the three wiregrass blocks, six plots (2 m ? 3.5 m) separated by 1 m wide mowed 
buffer strips were established in February 2007. Two replicates within each block were randomly 
assigned to one of the main treatments: 1) unburned control treatment; 2) burned with low fine 
fuel load; or 3) burned with elevated fine fuel load. Additionally, two adjacent split-plots (0.5 m 
? 2 m) were established within the center of each plot and then randomly assigned to a pine cone 
treatment: either with or without pine cones. An example of the plot layout (post-fire) is depicted 
in Fig. 3.1. Within each split-plot, four subsample sites were established by inserting steel rings 
(15 cm diameter with 5 cm wall height) until the top of each ring was level with the soil surface. 
Thirty seeds of each legume species were sown into each subsample ring, lightly pressed into the 
soil, and then covered with 1-2 mm of soil to minimize direct contact with burning fuels. To 
prevent seed predation by vertebrates, subsample plots were covered with metal screens (~1.5 
mm mesh size) affixed to the ground with aluminum nails. To avoid potential increased 
 
55 
conduction of heat to the seeds via the metal screens, the screens were removed immediately 
prior to and then replaced following application of the burn treatments. A total of 3,600 seeds 
species-1 were used for this study (5 treatments ? 6 replicates ? 4 subsamples replicate-1 ? 30 
seeds subsample-1). 
Experimental fuel loads used for this study were similar to the treatment loads (per unit 
area) used in the laboratory study described above. In March 2007, air-dried longleaf pine 
needles were uniformly spread throughout each burn plot. Immediately prior to igniting the 
prescribed burns, 20 oven-dried (24 h at 70?C) longleaf pine cones (75.1 ? 29.6 g per cone; mean 
? SD, n = 30) were placed uniformly throughout one split-plot in each fuel treatment plot. In July 
2007, the edges of each plot were ignited using a drip-torch (2:1 diesel:gasoline mixture) and the 
fire permitted to burn into the center of each plot. 
During the treatment applications, infrared thermal imagery (one digital image recorded 
every 5.5 s) was acquired for six plots using a ThermaCAM? S60 (FLIR Systems, Boston, 
Massachusetts) to measure temperatures of the fires. Thermal images were individually analyzed 
using the ThermaCAM? Researcher Pro software package (v2.7, FLIR Systems, Boston, 
Massachusetts). To assess maximum temperatures, two temperature sensitivity ranges of the 
thermal camera were used during the fires: 0-500?C and 300-1500?C. The lower limit of the 
highest temperature range (i.e., 300?C) was used to assess duration of potentially lethal 
temperatures. Maximum temperatures and duration were derived from the thermal images by 
using three uniformly spaced digital transects arranged parallel to the long edge of each split-
plot. Example thermal images obtained during an experimental fire are depicted in Fig. 3.2. 
Beginning approximately two weeks after application of the treatments (July 2007), seed 
germination within the plots was assessed biweekly for one growing season, ending in November 
 
56 
2007 prior to the first frost. Biweekly assessments were made by counting seedlings within each 
sub-plot in the field. Seedlings were readily identified in the field by the distinctive 
characteristics of their cotyledons and first true leaves (if present). Emergent seedlings were 
removed after each germination assessment to avoid repeated counts of the same individuals. 
Statistical Analysis 
Data were analyzed using the SAS software package (v9.1, SAS Institute, Cary, North 
Carolina). Data from each species were analyzed separately for each experiment. Data were 
normalized prior to analyses using arcsine square root transformations for proportions and log 
transformations for counts. 
Addition of pine cones killed all seeds in the laboratory germination study. A one-way 
ANOVA was therefore used to compare germination and mortality responses between untreated 
controls and those burned under low and high pine needle fuel loads (Littell et al. 2002). 
Differences in mean maximum temperature and durations due to treatments (i.e., low and high 
pine needle fuel loads and presence/absence of pine cones) from the laboratory study were 
examined using a fully factorial ANOVA. Significant ANOVA results were then further 
investigated using post-hoc analyses (Tukey?s means comparison tests; ? = 0.05; Littell et al. 
1996 and 2002) to identify differences among treatment responses (i.e., germination and 
mortality). 
Prior to analyzing germination data for the field study, total germination of each species 
was first calculated for each sub-plot. Germination data for this study were first analyzed for 
differences between control and fire treatments (fuel variation pooled at the plot level) and then 
analyzed for differences between burned treatments. To compare controls with fire treatments, 
germination totals for all sub-plots within each whole plot (i.e., controls and fuel loads) were 
 
57 
used to calculate mean germination at the whole plot level. These whole plot mean germination 
values (regardless of fuel load) were then analyzed using a randomized complete block (RCB) 
ANOVA (Littell et al. 2002) with wiregrass beds used as blocks. To compare germination 
between fire treatments, sub-plot totals were used to calculate mean germination for each split-
plot (i.e., pine cone treatments). These split-plot mean germination values were then analyzed 
using an RCB split-plot mixed model ANOVA (Littell et al. 1996) to determine responses to fuel 
treatments. Wiregrass beds were used as blocks, fuel load treatments were used as whole plots, 
and presence of pine cones was assigned as the split-plot for this analysis. Data for temperatures 
and durations of temperatures ?300?C were analyzed using a nested mixed model ANOVA by 
nesting pine cone treatments within fuel loads (Littell et al. 1996). Where ANOVAs indicated 
differences between treatments, post-hoc Tukey?s means comparison tests were performed (? = 
0.05; Littell et al. 1996, 2002). 
 
RESULTS 
EXPERIMENT 1: LABORATORY GERMINATION 
Temperature and duration data are presented in Fig. 3.3. Presence of pine cones increased 
both maximum temperatures (F = 12.401,12, P = 0.0042; Fig. 3.3a) and durations (F = 127.721,12, 
P < 0.0001; Fig. 3.3b). Variation in pine needle fuel loads did not affect maximum temperatures 
(F = 2.451,12, P = 0.143; Fig. 3.3a) or duration of temperatures ?60?C (F = 0.171,12, P = 0.688; 
Fig. 3.3b). No interaction occurred between fuel load and pine cone treatments for either variable 
(maximum temperature: F = 3.051,12, P = 0.106; duration: F = 0.291,12, P = 0.598). 
For all species, germination coupled with TZ tests indicated high initial viability of 
untreated seeds (i.e., >85%); however, germination levels of untreated seeds were highly variable 
 
58 
(Table 3.1). Germination levels generally reflected patterns of relative seed hardness (i.e., hard-
seeded species experienced the least germination). The relatively soft-seeded D. floridanum and 
moderately hard-seeded Ce. virginianum experienced the highest germination levels (about 70% 
and 50%, respectively). 
Overall, burning treatments reduced the germination rate of soft-seeded D. floridanum 
and moderately hard-seeded Ce. virginianum, while germination of hard-seeded species?Cr. 
rotundifolia, L. angustifolia, M. quadrivalvis, and R. reniformis?was unaffected by burning 
(Table 3.1). Regardless of relative seed hardness, mortality of all species increased following 
experimental burns. As noted in the methods, presence of pine cones killed all seeds, regardless 
of species or fuel load. Pine needle load did not affect germination or mortality of any species 
(Table 3.1). 
EXPERIMENT 2: FIELD FIRE INTENSITY MEASURES AND GERMINATION 
Maximum fire temperatures did not differ among fuel treatments (fuel load: F = 0.331,1, P 
= 0.667; pine cones: F = 0.611,3, P = 0.490) (Fig. 3.3c). Maximum temperatures typically ranged 
from about 725 to 875?C and occurred during flaming combustion. These temperatures are 
similar to those reported in longleaf pine stands in southern Georgia (Robertson 2007; Jack et al. 
2010) and are within the range of flaming combustion temperatures reported elsewhere (Neary et 
al. 1999). Contrary to expectations, altering pine needle loads did not affect mean duration of 
temperatures ?300?C (F = 0.141,1, P = 0.771); however, presence of pine cones significantly 
increased duration of temperatures ?300?C (F = 10.201,3, P = 0.049) (Table 3.3d). No interaction 
occurred between fuel load and pine cone treatments for the documented temperature variables 
(maximum temperature: F = 1.211,3, P = 0.352; duration: F = 0.061,3, P = 0.823). 
 
59 
Although the soft-seeded D. floridanum had more than 50% germination on average, 
germination of the remaining species was generally low (<20%) regardless of treatment (Table 
3.2, Fig. 3.4). Relative to unburned controls, burning significantly reduced germination of six of 
the seven legume species, although the magnitude of response to burning treatments varied by 
species and with fuel treatment (Fig. 3.4). Only germination of M. quadrivalvis, which had the 
hardest seed coat of the species used, did not differ from unburned controls. Among burning 
treatments, pine needle fuel load did not influence germination of any species (Table 3.3). The 
presence of pine cones reduced germination of all species except D. ciliare and Ce. virginianum, 
although the former species was marginally reduced (Table 3.3, Fig. 3.5). No interaction between 
fuel load and presence of pine cones occurred (Table 3.3) and no significant block effects were 
found. 
 
DISCUSSION 
By providing a source for regeneration of populations and recruitment of new 
individuals, persistent soil seed banks buffer against localized extirpation of some species 
(Thompson et al. 1998; Stocklin and Fisher 1999) and help maintain genetic diversity within 
these species (Dolan et al. 2008). Fires influence seedling recruitment from the soil seed bank 
both directly and indirectly (Bond and van Wilgen 1996). The present study demonstrated that 
fires have the potential to directly influence post-fire recruitment by killing seeds present in the 
soil seed bank. Although not observed under the conditions studied herein, fires may also 
enhance germination of some species (such as hard-seeded legumes) within longleaf pine stands 
(e.g., Chapter 2; Cushwa et al. 1968; Martin et al. 1975). Fires may also act as indirect filters for 
seedling establishment. Open spaces may be increased after fires via litter consumption and 
 
60 
reduced plant cover (Laughlin and Ful? 2008; Zimmermann et al. 2008; Myers and Harms In 
Press), thus potentially increasing the availability of suitable microsites for seedling recruitment 
(Eriksson and Ehrlen 1992; Setterfield 2002). 
Results of the present studies indicate that fire has the potential to significantly affect 
post-fire regeneration of legumes from the soil seed bank by influencing seed survival and 
germination. Fires did not appear to stimulate germination under the conditions studied, as has 
been reported in other studies of post-fire regeneration from the seed bank (e.g., Shea et al. 1979; 
Moreno and Oechel 1991). On the contrary, fires reduced germination for eight of the nine 
legume species studied. Only germination of M. quadrivalvis was not significantly reduced 
following fires in either the laboratory or field study. Additionally, germination responses 
differed between studies. Of the four species common to both studies (Ce. virginianum, D. 
floridanum, L. angustifolia, and M. quadrivalvis), only Ce. virginianum and D. floridanum 
experienced reduced germination following experimental fires in both cases; Ce. virginianum 
and D. floridanum have the softest seed coats of these four species and, as indicated in Chapter 1, 
may be less tolerant of elevated temperatures than those with harder seed coats. As indicated by 
the laboratory study, post-fire reductions in germination may be the result of increased seed 
mortality. 
Variation in fine fuel loads did not influence germination in either study, while presence 
of pine cones significantly reduced germination in the field study and eliminated it altogether in 
the laboratory study. These observations are likely linked to increased fire intensities under 
burning pine cones compared with pine needles alone. Indeed, a variety of studies have linked 
increased fine fuel loads with increased maximum temperatures during fires in longleaf pine 
systems (e.g., Mejeur 1998; Thaxton and Platt 2006; Jack et al. 2010). Unlike those studies, 
 
61 
however, the fine fuel load treatments used in the present studies did not significantly influence 
temperatures (Figs. 3.3a and c). Thaxton and Platt (2006) found that addition of woody fuels 
further increased temperatures compared to fine fuel addition alone. Similarly, pine cones 
increased temperatures in the current laboratory study. No such trend was found for the field 
study, however. These observations may be due, in part, to the methods used to measure 
temperatures in each study: thermocouples were used in the laboratory to measure temperatures 
occurring just below the substrate?s surface directly under the combusting material, whereas the 
field study used a thermal camera that measured temperatures using emitted radiation from the 
burning material. Residence times of temperatures ?60?C and ?300?C were similarly influenced 
by fuel types and loads, increasing with addition of pine cones, but not with additional fine fuel. 
Soils buffer elevated temperatures during fires and attenuation of temperatures with 
increasing soil depth is a well-documented phenomenon (e.g., Heyward 1938; Beadle 1940; 
Massman and Frank 2004); therefore, location within the soil profile will influence seed fate. 
The temperature range for maximum germination is typically within 70-100?C for hard-seeded, 
heat-stimulated legumes (Cushwa et al. 1968, Martin et al. 1975, Auld and O?Connell 1991, 
Williams et al. 2003). Temperatures reached near the soil surface in the current studies were well 
above this range of typical heat-stimulated germination. Studies of soil seed banks within 
longleaf pine stands that were encountered during the literature search sampled soils to depths of 
5 or 10 cm, but did not describe the vertical distribution of seeds within these samples (e.g., 
Cohen et al. 2004; Ruth et al. 2008). Seed bank studies from other fire-prone ecosystems (e.g., 
South African fynbos [Pieterse and Cairns 1986; Fourie 2008]; Australian jarrah forest [Shea et 
al. 1979]; Argentinian grassland [Laterra et al. 2006]) indicate that most seeds likely occur 
 
62 
within the upper 5 cm. Further studies within longleaf pine stands are needed in order to quantify 
the vertical distribution of seeds within the soil seed bank. 
Conceptual models have been developed to explain and predict post-fire germination 
responses. Martin and Cushwa (1966) note that, below a given threshold, temperatures have no 
effect on germination of heat-stimulated seeds. Above this temperature threshold, germination 
increases linearly with temperature until maximum germination is reached. Auld (1986) 
developed this model further by coupling soil temperature profiles with experimentally derived 
germination curves of the hard-seeded legume Acacia suaveolens (Sm.) Willd. to predict post-
fire germination responses within the soil profile under high- and low-intensity fires. Auld?s 
(1986) models predict that temperatures sufficient to stimulate germination of A. suaveolens 
would occur nearer to the soil?s surface under low-intensity fires than high-intensity fires. Under 
high-intensity fires, lethal temperatures will penetrate deeper into the soil profile, thereby 
increasing the depth at which germination will be stimulated during these fires.  
Burning woody fuels have the potential to elevate soil temperatures both vertically and 
laterally. Monsanto and Agee (2008) found that potentially lethal temperatures (i.e., 60?C) 
readily penetrated to 10 cm beneath and beside burned logs. One can visualize these elevated 
temperatures as a roughly bowl- or cone-shaped area of soil centered on the burning logs. Other 
long-burning fuels like pine cones may have similar effects. For example, a companion study to 
the current studies (Appendix 1) found that burning longleaf pine cones could increase 
temperatures 2.5 cm below the soil surface roughly 20-80?C higher than fires without pine 
cones. 
Based on the above models and observations of the influence of pine cone fuels, a 
conceptual model was developed to illustrate the potential impact of pine cones on seeds within 
 
63 
the soil seed bank (Fig. 3.6). This model assumes that: 1) temperatures decrease with increasing 
distance (both laterally and vertically within the soil) from the pine cone; 2) direct seed mortality 
of all species decreases with increasing distance from the pine cone; and 3) seeds of some 
species will be stimulated to germinate by elevated temperatures from the burning pine cone. 
This model predicts that most or all seeds located under and immediately adjacent to a pine cone 
(or other fuel source that prolongs fire) will be killed by the higher intensity fire (Zone of 
Mortality). As depth and lateral distance from a pine cone increase, temperatures will descend 
into the favorable range for germination (Zone of Germination). Each species, particularly those 
with heat-stimulated germination, will have a particular optimum temperature range and thus an 
optimum distance/depth from a pine cone for germination. Eventually, distance and/or depth 
from a pine cone will be sufficiently great to have no effect on seeds in the soil seed bank 
because soil temperature will not rise into the range that stimulates germination (Zone of No 
Effect). 
By killing mature plants and seeds, pine cones create open spaces that are available for 
colonization. Within the soil seed bank, seeds that are able to survive and germinate closer to 
burned pine cones may be able to take advantage of the open space created, thus providing them 
with an initial competitive advantage over species that are less tolerant of high intensity fires. 
Following germination in the open spaces created by pine cones, other fire-mediated factors, 
such as water or light availability, may be important for seedling survival (Bond and van Wilgen 
1996). In particular, within longleaf pine stands, moisture availability may be the most important 
resource for seedling establishment (Iacona et al. 2010). Because pine cones and other long-
burning fuels are not distributed uniformly throughout longleaf pine stands (Hiers et al. 2009), 
 
64 
these open spaces will exhibit similarly heterogeneous distribution, thus influencing fine-scale 
distribution patterns of species. 
The potential influence of pine cones on fine-scale vegetation dynamics within the 
longleaf pine ecosystem and other frequently burned pine-dominated systems, such as ponderosa 
pine (Pinus ponderosa [Dougl.]) ecosystems of the western United States, needs to be tested in 
the field. To test this hypothetical model, lateral and vertical soil temperature profiles need to be 
characterized for burning pine cones, as well as the influence of soil properties such as moisture 
content, compaction, porosity, and texture. Season of fire may also influence the ecological 
effects of pine cones, as soil and fuel moistures as well as ambient temperatures will likely vary 
with season. Additionally, each species likely has a maximum survivable germination depth. 
Seeds stimulated to germinate below this maximum depth will be indirectly killed by the fire 
because of their inability to reach the surface before exhausting their energy stores. No studies 
were encountered that have investigated maximum survivable germination depth for species 
native to the longleaf pine ecosystem. Finally, the applicability of this model to other long-
burning fuels needs to be assessed by experiments that compare burning characteristics of pine 
cones to other plant materials, such logs and tree limbs.  
 
REFERENCES 
Auld TD (1986) Population dynamics of the shrub Acacia suaveolens (Sm.) Willd: fire and the 
transition to seedlings. Australian J Ecol 11:373-385 
Auld TD, O?Connell MA (1991) Predicting patterns of post-fire germination in 35 eastern 
Australian Fabaceae. Austral Ecol 16:53-70 
Ballard LAT (1973) Physical barriers to germination. Seed Sci Technol 1:285-303 
Baskin CC, Baskin JM (1998) Seeds: ecology, biogeography, and evolution of dormancy and 
germination. Academic Press, San Diego 
 
65 
Beadle NCW (1940) Soil temperatures during forest fires and their effect on the survival of 
vegetation. J Ecol 28:180-192 
Bond WJ, Keeley JE (2005) Fire as a global ?herbivore?: the ecology and evolution of flammable 
ecosystems. Trends Ecol Evol 20:387-394 
Bond WJ, van Wilgen BW (1996) Fire and plants. Chapman Hall, London 
Bond WJ, Woodward FI, Midgley GF (2005) The global distribution of ecosystems in a world 
without fire. New Phytol 165:525-538 
Clark DL, Wilson MV (1994) Heat-treatment effects on seed bank species of an old-growth 
douglas-fir forest. Northwest Sci 68:1-5 
Clewell AF (1989) Natural history of wiregrass (Aristida stricta Michx. Graminae). Nat Areas J 
9:2223-2233 
Cohen S, Braham R, Sanchez F (2004) Seed bank viability in disturbed longleaf pine sites. 
Restor Ecol 12:503-515 
Cushwa CT, Martin RE, Miller RL (1968) The effects of fire on seed germination. J Range 
Manag 21:250-254 
Danthu P, Ndongo M, Diaou M, Thiam O, Sarr A, Dedhiou B, Vall, AOM (2003) Impact of bush 
fire on germination of some West African acacias. For Ecol Manag 173:1-10 
Dolan RW, Quintano-Ascencio PF, Menges ES (2008) Genetic change following fire in 
populations of a seed-banking perennial plant. Oecologia 158:355-360 
Drew MB, Kirkman LK, Gholson AK (1998) The vascular flora of Ichauway, Baker County, 
Georgia: a remnant longleaf pine/wiregrass ecosystem. Castanea 63:1-24 
Eriksson O, Ehrlen J (1992) Seed and microsite limitation of recruitment in plant populations. 
Oecologia 91:360-364 
Fonda RW (2001) Burning characteristics of needles from eight pine species. For Sci 47:390-396 
Fonda RW, Varner JM (2005) Burning characteristics of cones from eight pine species. 
Northwest Sci 78:322-333 
Fourie S (2008) Composition of the soil seed bank in alien-invaded grassy fynbos: potential for 
recovery after clearing. S Afr J Bot 74:445-453 
Goebel PC, Palik BJ, Kirkman LK, West L (1997) Field guide: landscape ecosystem types of 
Ichauway. Report number 97-1. Joseph W. Jones Ecological Research Center at 
Ichauway, Newton 
 
66 
Hainds MJ, Mitchell RJ, Palik BJ, Boring LR, Gjerstad DH (1999) Distribution of native 
legumes (Leguminoseae) in frequently burned longleaf pine (Pinaceae)-wiregrass 
(Poaceae) ecosystems. Am J Bot 86:1606-1614 
Hare RC (1961) Heat effects on living plants. Occasional Paper 183. U.S. Department of 
Agriculture, Forest Service, Southern Forest Experiment Station 
Hendricks JJ, Boring LR (1999) N2-fixation by native herbaceous legumes in burned pine 
ecosystems of the southeastern United States. For Ecol Manag 113:167-177 
Hendricks JJ, Wilson CA, Boring LR (2002) Foliar litter position and decomposition in a fire-
maintained longleaf pine ? wiregrass ecosystem. Can J For Res 32:928-941 
Hermann, SM (2007) How does longleaf pine native groundcover fit with forest management 
goals? In: Estes BL, Kush JS (comps) Longleaf Pine: Seeing the Forest through the 
Trees. Proceedings of the Sixth Longleaf Alliance Regional Conference, November 13-
16, 2006, Tifton, Georgia, USA. Longleaf Alliance Report No. 10, pp 28-31 
Heyward F (1938) Soil temperatures during fire in the longleaf pine region. J For 36:478-491 
Hiers JK, Mitchell RJ, Boring LR, Hendricks JJ, Wyatt R (2003) Legumes native to longleaf 
pine savannas exhibit capacity for high N2-fixation rates and negligible impacts due to 
timing of fire. New Phytol 157:327-338 
Hiers JK, O?Brien JJ, Mitchell RJ, Grego JM, Loudermilk EL (2009) The wildland fuel cell 
concept: an approach to characterize fine-scale variation in fuels and fire in frequently 
burned longleaf pine forests. Int J Wildl Fire 18:315-325 
Iacona GD, Kirkman LK, Bruna EM (2010) Effects of resource availability on seedling 
recruitment in a fire-maintained savanna. Oecologia 163:171-180 
Jack SB, Hiers JK, Mitchell RJ, Gagnon JL (2010) Fuel loading and fire intensity?effects on 
longleaf pine seedling survival. In: Stanturf JA (ed) Proceedings of the 14th biennial 
southern silvicultural research conference. Gen. Tech. Rep. SRS-121. U.S. Department of 
Agriculture, Forest Service, Southern Research Station, Asheville, pp 275-279 
Keeley JE, Morton BA, Pedrosa A, Trotter P (1985) Role of allelopathy, heat, and charred wood 
in the germination of chaparral herbs and suffrutescents. J Ecol 73:445-458 
Kennard DL, Outcalt K (2006) Modeling spatial patterns of fuels and fire behavior in a longleaf 
pine forest in the southeastern USA. Fire Ecol 2:31-52 
Laterra P, Ortega EZ, Ochoa MDC, Vignolio OR, Fern?ndez ON (2006) Interactive influences of 
fire intensity and vertical distribution of seed banks on post-fire recolonization of a tall-
tussock grassland in Argentina. Austral Ecol 31:608-622 
Laughlin DC, Ful? PZ (2008) Wildland fire effects on understory plant communities in two fire-
prone forests. Can J For Res 38:133-142 
 
67 
Littell RC, Milliken GA, Stroup WW, Wolfinger RD (1996). SAS system for mixed models. 
SAS Institute, Cary 
Littell RC, Stroup WW, Freund RJ (2002) SAS for linear models, 4th edn. SAS Institute, Cary 
Lynch JM, Gholson AK, Baker WW (1986) Natural features of Ichauway Plantation, Georgia. 
The Nature Conservancy, Southeast Regional Office, Chapel Hill 
Martin RE, Cushwa CT (1966) Effects of heat and moisture on leguminous seed. Proc Tall 
Timbers Fire Ecol Conf 5:159-175 
Martin RE, Miller RL, Cushwa CT (1975) Germination response of legume seeds subjected to 
moist and dry heat. Ecology 56:1441-1445 
Massman WJ, Frank JM (2004) Effect of thermophysical properties of a dry soil using a new 
model of soil heat flow and a new high temperature flux sensor. Int J Wildl Fire 13:427-
442 
Mejeur RS (1998) Effects of fire on the reproductive ecology of Tephrosia virginiana (L.) Pers. 
in the Carolina Sandhills. MSc Thesis, University of Georgia, Athens 
Mitchell RJ, Hiers JK, O?Brien JJ, Jack SB, Engstrom RT (2006) Silviculture that sustains: the 
nexus between silviculture, frequent prescribed fire, and conservation of biodiversity in 
longleaf pine forests of the southeastern United States. Can J For Res 36:2724-2736 
Mitchell RJ, Hiers JK, O?Brien JJ, Starr G (2009) Ecological forestry in the southeast: 
understanding the ecology of fuels. J For 107:391-397 
Monsanto PG, Agee JK (2008) Long-term post-wildfire dynamics of coarse woody debris after 
salvage logging and implications for soil heating in dry forests of the eastern Cascades, 
Washington. For Ecol Manag 255:3952-3961 
Moreno JM, Oechel WC (1991) Fire intensity effects on germination of shrubs and herbs in 
southern California chaparral. Ecology 72:1993-2004 
Morrison DA, McClay K, Porter C, Rish S (1998) The role of the lens in controlling heat-
induced breakdown of testa-imposed dormancy in native Australian legumes. Ann Bot 
82:35-40 
Myers JA, Harms KE (In Press) Seed arrival and ecological filters interact to assemble high-
diversity plant communities. Ecology 
Neary DG, Klopatek CC, DeBano LF, Ffolliott PF (1999) Fire effects on belowground 
sustainability: a review and synthesis. For Ecol Manag 122:51-71 
O?Brien JJ, Hiers JK, Callaham MA, Mitchell RJ, Jacks SB (2008) Interactions among overstory 
structure, seedling life-history traits, and fire in frequently burned neotropical pine 
forests. AMBIO 37:542-547 
 
68 
Ottmar RD, Vihnanek RE (2000) Stereo photo series for quantifying natural fuels. Volume VI: 
longleaf pine, pocosin, and marshgrass types in the Southeast United States. PMS 835. 
National Wildfire Coordinating Group, National Interagency Fire Center, Boise 
Ottmar RD, Vihnanek RE, Mathey JW (2003) Stereo photo series for quantifying natural fuels. 
Volume VIA: Sand hill, sand pine scrub, and hardwoods with white pine types in the 
south-east United States with supplemental sites for volume VI. PMS 838. National 
Wildfire Coordinating Group, National Interagency Fire Center, Boise 
Peet RK, Allard DJ (1993) Longleaf pine vegetation of the southern Atlantic and eastern Gulf 
Coast regions: a preliminary classification. Proc Tall Timbers Fire Ecol Conf 18:45-81 
Peters J (2000) Tetrazolium testing handbook. Contribution No. 29 to the handbook on seed 
testing. Association of Official Seed Analysts, Lincoln 
Pieterse PJ, Cairns ALP (1986) The effect of fire on an Acacia longifolia seed bank in the south-
western Cape. S Afr J Bot 52:233-236. 
Platt WJ, Evans GW, Rathbun SL (1988) The population dynamics of a long-lived conifer (Pinus 
palustris). Am Nat 131:491-525 
Rebertus AJ, Williamson GB, Moser EB (1989) Longleaf pine pyrogenicity and turkey oak 
mortality in Florida xeric sandhills. Ecology 70:60-70 
Robertson K (2007) Fire ecology: fire behavior notes. Tall Timbers Res Station Res Notes 6(5):2 
Ruth AD, Jose S, Miller DL (2008) Seed bank dynamics of sand pine scrub and longleaf pine 
flatwoods of the Gulf Coastal Plain (Florida). Ecol Restor 26:19-21 
Setterfield SA (2002) Seedling establishment in an Australian tropical savanna: effects of seed 
supply, soil disturbance and fire. J Appl Ecol 39: 949-959 
Shea SR, McCormick J, Portlock CC (1979) The effects of fires on regeneration of leguminous 
species in the northern jarrah (Eucalyptus marginata Sm.) forest of Western Australia. 
Australian J Ecol 4:195-205. 
Stocklin J, Fisher M (1999) Plants with longer-lived seeds have lower local extinction rates in 
grassland remnants 1950-1985. Oecologia 120:539-543 
Thaxton JM, Platt WJ (2006) Small-scale fuel variation alters fire intensity and shrub abundance 
in a pine savanna. Ecology 87:1331-1337 
Thompson K, Bakker JP, Bekker RM, Hodgson JG (1998). Ecological correlates of seed 
persistence in soil in the northwest European flora. J Ecol 86:163-169 
van Eerden BP (1997) Studies on the reproductive biology of wiregrass (Aristida stricta Michx.) 
in the Carolina sandhills. MSc Thesis, University of Georgia, Athens 
 
69 
Varner JM (2005) Smoldering fire in long-unburned longleaf pine forests: linking fuels with fire 
effects. PhD Dissertation, University of Florida, Gainesville 
Ware S, Frost CC, Doerr PD (1993) Southern mixed hardwood forest: the former longleaf pine 
forest. In: Martin WH, Boyce SG, Echternacht AC (eds) Biodiversity of the Southern 
United States: Lowland Terrestrial Communities. John Wiley & Sons, pp 447-493 
Whelan RJ (1995) The ecology of fire. Cambridge University Press, Cambridge 
Williams PR, Congdon RA, Grice AC, Clarke PJ (2003) Fire-related cues break seed dormancy 
of six legumes of tropical eucalypt savannas in north-eastern Australia. Austral Ecol 
28:507-514 
Williamson GB, Black EM (1981) High temperature of forest fires under pines as a selective 
advantage over oaks. Nature 293:643-644 
Wunderlin RP, Hansen BF (2003) Guide to the vascular plants of Florida, 2nd edn. University 
Press of Florida, Gainesville 
Zimmerman J, Higgins SI, Grimm V, Hoffman J, Munkemuller T, Linstadter A (2008) 
Recruitment filters in a perennial grassland: the interactive roles of fire, competitors, 
moisture and seed availability. J Ecol 96:1033-1044 
 
 
70 
 
Species Treatment Germination (%) Mortality (%) 
Ce. virginianum Control 48.8 ? 4.3 a 7.5 ? 2.5 b 
 Low Fuel 10.0 ? 5.0 b 71.3 ? 15.3 a 
 High Fuel 8.8 ? 5.2 b 90.0 ? 6.1 a 
Cr. rotundifolia Control 16.3 ? 7.2 a 15.0 ? 2.0 b 
 Low Fuel 21.3 ? 8.8 a 68.8 ? 12.0 a 
 High Fuel 11.3 ? 8.0 a 77.4 ? 9.6 a 
D. floridanum Control 70.0 ? 4.6 a 13.8 ? 1.3 b 
 Low Fuel 13.3 ? 13.3 b 87.5 ? 12.6 a 
 High Fuel 0.0 ? 0.0 b 97.5 ? 1.4 a 
L. angustifolia Control 1.3 ? 1.3 a 1.3 ? 1.3 b 
 Low Fuel 3.8 ? 3.8 a 87.4 ? 9.2 a 
 High Fuel 2.5 ? 2.5 a 83.6 ? 8.5 a 
M. quadrivalvis Control 7.5 ? 1.4 a 3.8 ? 1.3 b 
 Low Fuel 2.5 ? 2.5 a 56.8 ? 18.3 a 
 High Fuel 2.5 ? 1.4 a 74.2 ? 9.5 a 
R. reniformis Control 2.5 ? 1.4 a 2.5 ? 1.4 b 
 Low Fuel 1.8 ? 1.8 a 64.5 ? 14.3 a 
  High Fuel 0.0 ? 0.0 a 92.5 ? 4.8 a 
 
Table 3.1. Influence of fine fuel load variation on post-burn germination and mortality of legume 
seeds (mean ? SE) for the laboratory germination study. Different letters indicate significant 
differences as indicated by post-hoc Tukey?s means separation tests (P < 0.05).  
 
71 
 
Species 
Model 
Component df F P 
Ce. virginianum Bed 2,14 1.57 0.243 
 Burned 1,14 12.91 0.003 
D. ciliare Bed 2,14 3.34 0.065 
 Burned 1,14 22.39 <0.001 
D. floridanum Bed 2,14 0.06 0.945 
 Burned 1,14 75.01 <0.0001 
L. angustifolia Bed 2,14 0.81 0.466 
 Burned 1,14 38.28 <0.0001 
M. quadrivalvis Bed 2,14 0.57 0.579 
 Burned 1,14 0.59 0.456 
P. canescens Bed 2,14 3.50 0.058 
 Burned 1,14 66.50 <0.0001 
T. virginiana Bed 2,14 0.53 0.602 
  Burned 1,14 13.33 0.003 
 
Table 3.2. Effect of burning on germination of legume seeds (July ? November 2007; n = 30) in 
the field germination study. Bed was used as the blocking component of the RCB split-plot 
ANOVA. Mean germination values from burned treatments and untreated controls are depicted 
in Fig. 3.3. 
 
72 
 
Species 
Model 
Component df F P 
Ce. virginianum Bed 2,2 2.63 0.276 
 Fuel Load 1,2 0.46 0.567 
 Pine Cones 1,16 3.08 0.099 
 Fuel ? Cones 1,16 0.64 0.435 
D. ciliare Bed 2,2 5.00 0.167 
 Fuel Load 1,2 3.71 0.194 
 Pine Cones 1,16 3.71 0.072 
 Fuel ? Cones 1,16 4.30 0.055 
D. floridanum Bed 2,2 3.14 0.242 
 Fuel Load 1,2 2.58 0.249 
 Pine Cones 1,16 10.89 0.005 
 Fuel ? Cones 1,16 2.21 0.157 
L. angustifolia Bed 2,2 1.31 0.433 
 Fuel Load 1,2 0.27 0.656 
 Pine Cones 1,16 7.11 0.017 
 Fuel ? Cones 1,16 0.01 0.910 
M. quadrivalvis Bed 2,2 4.95 0.168 
 Fuel Load 1,2 0.18 0.711 
 Pine Cones 1,16 6.01 0.026 
 Fuel ? Cones 1,16 0.20 0.663 
P. canescens Bed 2,2 5.17 0.162 
 Fuel Load 1,2 0.05 0.838 
 Pine Cones 1,16 10.49 0.005 
 Fuel ? Cones 1,16 0.00 1.000 
T. virginiana Bed 2,2 2.12 0.321 
 Fuel Load 1,2 0.66 0.502 
 Pine Cones 1,16 12.90 0.002 
  Fuel ? Cones 1,16 0.88 0.361 
 
Table 3.3. Effect of fuel variation on post-burn germination of legume seeds for the field 
germination study. Bed was used as the blocking component of the RCB split-plot ANOVA. 
Mean germination values from burned treatments with and without pine cones are depicted in 
Fig. 3.4. 
 
73 
 
 
Figure 3.1. Example of plot layout (post-fire). Solid line is plot boundary (pine needle 
treatments) and dashed line is split-plot boundary (pine cones treatments). White ash within the 
pine cone treatment split-plot is remains of longleaf pine cones and round, black areas are 
wiregrass clumps. 
 
74 
 
 
 
Figure 3.2. Thermal imagery of a typical burn (a) during flaming combustion (at approx. 0.5 min 
after ignition of fire) and (b) during smoldering/glowing combustion with smoldering pine cones 
visible (at approx. 30 min after ignition). Red outlines in the centers of the images indicate 
approximate split-plot boundaries. 
a 
b 
 
75 
 
 
 
 
Figure 3.3. Effect of fuel treatments on mean maximum temperatures and mean durations of 
elevated temperatures for the laboratory study (a & b) and field study (c & d). Error bars 
represent 1 SE. Different uppercase letters indicate significant differences (P < 0.05) between 
fine fuel treatments as indicated by ANOVA comparisons. Different lowercase letters indicate 
significant differences (P < 0.05) between pine cone treatments as indicated by ANOVA 
comparisons. No significant interaction (P > 0.05) between fine fuel loads and pine cones was 
found for either study. LF = low fine fuel, HF = high fine fuel, NPC = no pine cones, PC = pine 
cones. 
c d 
a b 
A 
A 
A 
A 
A 
A 
A A 
a 
a 
a 
a a 
b 
b 
b 
 
76 
 
 
 
Figure 3.4. Germination of seeds from untreated controls and burned treatments (means; n = 30) 
for the field germination study. Error bars represent 1 SE. Asterisks indicate significant 
differences as indicated by ANOVAs (P < 0.05) presented in Table 3.2. Species codes are: CEVI 
= Ce. virginianum, DECI = D. ciliare, DEFL = D. floridanum, LEAN = L. angustifolia, MIQU = 
M. quadrivalvis, PECA = P. canescens, TEVI = T. virginiana. 
* * 
* 
* 
* 
* 
 
77 
 
 
 
Figure 3.5. Effect of presence of pine cones during fires on post-burn germination of legume 
seeds (means; n = 30) for the field germination study. Error bars represent 1 SE. Asterisks 
indicate significant differences between treatments as indicated by ANOVAs (P < 0.05) 
presented in Table 3.3. Species codes are: CEVI = Ce. virginianum, DECI = D. ciliare, DEFL = 
D. floridanum, LEAN = L. angustifolia, MIQU = M. quadrivalvis, PECA = P. canescens, TEVI 
= T. virginiana. 
* 
* 
* 
* 
* 
 
78 
 
  
Figure 3.6. Potential impacts of burned pine cones on germination and mortality of seeds. 
 
79 
 
 
 
 
 
CHAPTER 4 
SUMMARY AND CONCLUSIONS 
 
Results of the studies presented in Chapters 2 and 3 indicate that fires have the potential 
to affect germination and mortality of legume seeds. Such impacts may, in turn, affect 
recruitment of these and similar species, thereby influencing post-fire vegetation patterns. The 
species-specific germination and mortality responses observed indicate that the intensity of each 
fire may be an important determinant of post-fire plant assemblages. Furthermore, fine-scale fuel 
variation, which in turn affects similarly scaled fire behavior, has the capacity to dramatically 
affect recruitment of individual species from seed. 
The results presented in Chapter 2 indicate species-specific germination and responses to 
the various fire surrogates studied: dry and moist heat, smoke, and manual scarification. 
Germination and mortality of most species tended to be relatively unaffected by dry heat, except 
when exposed to the most extreme combination of temperature and duration (i.e., 150?C for 10 
min), which was characterized by reduced germination and increased mortality. Two species, 
however, experienced enhanced germination with increasing heat: the hard-seeded Lespedeza 
angustifolia and Crotalaria rotundifolia; although the effect was most pronounced for Cr. 
rotundifolia and both species exhibited sharply increased mortality at 10 min and 150?C. 
Germination of all species was enhanced by short exposures to moist heat (0.16 to 1 min) 
without experiencing increased mortality. Additionally, seeds of all species survived (15-70%) 
even after the longest exposure to moist heat (10 min), although germination tended to decline 
 
80 
following these longer exposure times. Scarification greatly enhanced germination of all species 
without affecting seed mortality. Generally, germination and seed mortality of the legume 
species examined in this study do not appear to be influenced by plant-derived smoke. The only 
exception was Centrosema virginianum, which exhibited slightly reduced, but statistically 
significant, germination after exposure to smoke. 
Chapter 3 investigated the potential role of fine-scale fuel heterogeneity on various 
parameters of fire behavior (maximum temperature and duration of potentially lethal 
temperatures). Increasing fine fuel loads did not affect intensity of the fire (i.e., maximum 
temperatures or duration of potentially lethal temperatures). While maximum temperatures were 
not affected, adding pine cones greatly increased the time seeds were exposed to potentially 
lethal temperatures. As might be expected, in both studies, germination of all species was 
substantially reduced by exposure to fire, especially if seeds had pine cones burning over them. 
Manipulating pine needle fuel (i.e., fine fuel) loads did not affect germination. 
A conceptual model was presented in Chapter 3 describing the potential effects of pine 
cones, which can smolder long after the fine fuels have stopped burning and cooled. Elevated 
temperatures from fires have the potential to kill seeds and, in locations where smoldering fuels 
promote longer durations of these lethal temperatures, such as under pine cones, more (or all) 
seeds will be killed. Under pine tree canopies, it is logical to assume that all seeds will be killed 
directly under and adjacent to any pine cones consumed in a fire. One can further assume that 
potentially lethal conditions decrease with increasing distance from pine cones until, after some 
distance, lethal conditions are no longer achieved (i.e., seeds are unaffected). By creating small 
pockets of open space, burned pine cones create potential colonization gaps for seeds stored in 
the soil seed bank or from other sources (e.g., unburned plants, animal scat, etc.). Within the soil 
 
81 
seed bank, seeds that are able to survive and germinate closer to the center of burned pine cones 
may be able to take advantage of the open space, survive, and thrive, while others are killed or, if 
they survive, are unable to compete. Differential survival and germination of seeds and 
recruitment of seedlings following fires?mediated by small-scale pre-fire fuel variation?may, 
therefore, influence fine-scale patterns of plant distribution. Furthermore, within frequently 
burned systems, small-scale fire variation may be more important for determining localized 
species distribution?at least during initial stages of (re)colonization?than other factors, such as 
light or soil nutrients. 
The behavior and impact of each fire is unique, affected by fuel loads, type, and 
distribution, as well as by moisture, weather, and local topography. Similarly, individual seeds 
will respond to fires based on a number of intrinsic (e.g., moisture content, presence of a hard 
seed coat, etc.) and extrinsic (e.g., location within soil profile, soil moisture, fire intensity, etc.) 
factors. Of the intrinsic factors, seed hardness and moisture content may be the most important to 
survival during, and germination following, fires. It is clear that cracking (i.e., scarifying) the 
seed coat increases germination of these legume species. Dry heat generated by fires may 
vaporize internal seed moisture and increase internal pressure or cause cells to differentially 
swell and contract, thereby cracking the seed coat. Alternatively, steam generated within moist 
soil or driven off of adjacent biomass during fires may soften or penetrate seed coats, thereby 
initiating germination. In addition, seeds with low moisture contents may have less damage to 
cells following water vaporization and may, therefore, more readily survive fires than seeds with 
high moisture contents. Although high moisture contents may initially buffer seeds against 
elevated temperatures experienced during fires, increased water content may also increase 
cellular damage once the water has vaporized. Because intrinsic and extrinsic seed characteristics 
 
82 
do not exist in isolation from one another, knowledge of these factors and their interactions are 
essential for determining species response to a given fire. 
 
83 
 
 
 
 
 
APPENDIX 
FIELD OBSERVATIONS OF SOIL SURFACE AND SUBSURFACE TEMPERATURES 
AND RESIDENCE TIMES OF SMALL-SCALE EXPERIMENTAL FIRES 
 
Treatment Temperature (?C)a  Residence Time (min)b 
 Surface Subsurface  Surface Subsurface 
March, No Cones 526.3 ? 144.8 30.0 ? 3.8  1.5 ? 0.5 0.0 ? 0.0 
March, Cones 675.5 ? 173.5 52.6 ? 11.7  7.0 ? 2.3 0.8 ? 1.3 
July, No Cones 525.3 ? 157.1 57.0 ? 8.8  14.8 ? 6.6 1.8 ? 2.6 
July, Cones 685.0 ? 83.8 75.7 ? 34.6  35.8 ? 13 8.8 ? 11.9 
 
Temperatures and residence times (mean ? SD) recorded at the soil surface (surface) and 2.5 cm 
belowground (subsurface) during experimental burns of individual Tephrosia virginiana plants 
in a garden plot at the Joseph W. Jones Ecological Research Center in Bryan Co., Georgia in 
March and July 2007. Plants were burned with (cones) or without (no cones) two pine cones and 
a small amount of pine needles (approx. 32 g). Pine cones and pine needles were oven-dried 
prior to use. Fires were ignited using the portable furnace described in Chapter 3. Three 
replicates were used for each treatment. aMean maximum temperatures recorded using chromel-
alumel (K-type) thermocouples. bMean residence time in minutes of temperatures ?60?C (i.e., 
potentially lethal).