2011 SOUTHEASTERN NATURALIST 10(4):659–672 Food Habits of American Alligators (Alligator mississippiensis) in East Texas David T. Saalfeld1,*, Warren C. Conway1, and Gary E. Calkins2 Abstract - American Alligator (Alligator mississippiensis) food habit data are important when establishing management strategies, as diet can directly influence growth rates, body condition, behavior, and reproduction. Diets of American Alligators are hypothesized to vary among habitats as well as geographically; however, few diet studies have been conducted outside of Florida and Louisiana. To address this information gap, 62 diet samples were obtained from alligators ranging in size from 94.7 cm to 386.0 cm (total length) from June 2006–September 2008 in inland freshwater wetlands in East Texas. A total of 33 different prey items (comprising 670 individual prey items) and 1 parasite were identified. Irrespective of size class, sex, and study site, >85% of individual prey items were invertebrates. Nearly all diet samples contained some sort of organic by-catch and/or non-food items (i.e., woody debris, aquatic plants, seeds, rocks, fishing tackle, etc.). Although alligator diets were similar between sexes, non-breeding (<183.0 cm in total length) alligators consumed more invertebrate prey items by biomass and percent occurrence than breeding-size alligators. In general, alligators forage opportunistically; therefore, most habitat-based, local, or geographic variability in food habits among populations are most likely influenced by food availability. As such, regional differences in food availability likely result in geographic variability in life-history characteristics such as growth rates and condition, important factors to consider when establishing management strategies. Introduction Throughout the southeastern United States, Alligator mississippiensis Daudin (American Alligator) exist as top carnivores in aquatic and wetland systems, and perform important functions in structuring coexisting animal populations within their environment (Barr 1994, 1997; Mazzotti et al. 2009; Rosenblatt and Heithaus 2011). Opportunistic predators, alligators exhibit a varied diet, and are adept at exploiting local prey resources that encompass a wide diversity of sizes and taxa, ranging from small insects and crustaceans to large vertebrates (Chabreck 1971, Delany and Abercrombie 1986, Rice 2004, Valentine et al. 1972). Alligator diets vary by specific geographic location, habitat, prey encountered, and prey vulnerability and size, as well as by alligator size (Barr 1994, 1997; Delany and Abercrombie 1986; Dodson 1975; McNease and Joanen 1977; Platt et al. 1990; Wolfe et al. 1987). For example, smaller (juvenile) alligators typically consume invertebrates and small fish (Barr 1994, Chabreck 1971, Delany 1990, Dodson 1975, Fogarty and Albury 1967, Giles and Childs 1949, Platt et al. 1990, Valentine et al. 1972), whereas larger adults mainly consume vertebrates 1Arthur Temple College of Forestry and Agriculture, Stephen F. Austin State University, Nacogdoches, TX 75962. 2District 6 Wildlife Division Field Office, Texas Parks and Wildlife Department, 1342 South Wheeler, Jasper, TX 75951. *Corresponding author - dsaalfeld@gmail.com. 660 Southeastern Naturalist Vol. 10, No. 4 (Delany and Abercrombie 1986, Giles and Childs 1949, McNease and Joanen 1977, Shoop and Ruckdeschel 1990). Alligator diet studies have been concentrated in Louisiana (Platt et al. 1990, Taylor 1986, Valentine et al. 1972, Wolfe et al. 1987), north central and central Florida (Delany 1990, Delany and Abercrombie 1986, Delany et al. 1999, Rice 2004), and southern Florida (Barr 1994, 1997; Fogarty and Albury 1967). Such basic, descriptive studies of alligator diet and food habits allow for speculation about ontogenic, geographic, and temporal variation in resource utilization and food habits for alligators throughout their geographic range (Barr 1994). Diet analyses and food habits are essential data for understanding functional roles of key predators in any ecosystem, but also reveal basic predator-prey relationships and allow comparisons among individuals of different sizes and among habitats (Barr 1994, Rice 2004). Moreover, diet can directly affect growth rates and body condition (Chabreck 1971, McNease and Joanen 1981), which are important lifehistory characteristics to understand when establishing management strategies. As such, food-habit data are important for identifying and isolating possible causes for geographic variation in growth rates and body condition among alligator populations. However, no dietary studies have been performed in the western portion of the American Alligator’s geographic range. Previous work from east Texas inland wetlands has shown that alligators exhibit faster growth rates and poorer body condition than most other populations throughout their range (Saalfeld 2010, Saalfeld et al. 2008, Webb 2005). Therefore, quantifying diets regionally could elucidate potential causes for regional discrepancies in growth rates and body condition. The objectives of this study were to quantify food habits of inland alligators within three wetlands in east Texas and determine any potential variability in dietary habits between sexes and as a function of size or location. Field Site Description This research was conducted on three freshwater wetlands in east Texas: Angelina- Neches/Dam B Wildlife Management Area (Dam B WMA), Kurth Lake, and Little Sandy National Wildlife Refuge (NWR) (Fig. 1). Study sites were arranged along a longitudinal gradient throughout east Texas (Fig. 1) and were selected based upon presence of suitable habitats (i.e., mosaic of open water, floating vegetation, and emergent vegetation) for American Alligators (Webb 2005), annual hunts and spotlight surveys conducted by the Texas Parks and Wildlife Department (TPWD), and landowner permission. Although study sites are representative of most east Texas wetlands, few other wetlands within this region had adequate densities of American Alligators to include as study sites. Dam B WMA is a 5113-ha area located within Jasper and Tyler counties at the confluence of the Angelina River, Neches River, and B.A. Steinhagen Reservoir. Dam B WMA is characterized by riverine, open lake, and shallow marsh habitats, with an average depth of 1.2 m (Webb 2005, Webb et al. 2009). Dominant aquatic plants include Eichhornia crassipes (Mart.) Solms (Common Water Hyacinth), Salvinia minima Baker (Common Salvinia), S. molesta D.S. Mitchell (Giant Salvinia), Alternanthera philoxeroides (Mart.) Griseb. (Alligatorweed), Hydrilla verticellata (L.f.) Royle (Hydrilla), Polygonum spp. (smartweed), and Nelumbo 2011 D.T. Saalfeld, W.C. Conway, and G.E. Calkins 661 lutea Willd. (American Lotus). Dominant woody species include Taxodium distichum (L.) Rich. (Bald Cypress), Cephalanthus occidentalis L. (Buttonbush), Salix nigra Marshall (Black Willow), Triadica sebifera (L.) Small (Chinese Tallow), Quercus nigra L. (Water Oak), Q. lyrata Walter (Overcup Oak), Nyssa aquatica L. (Water Tupelo), and Pinus spp. (pine species) (Godfrey and Wooten 1981). Kurth Lake is a 294-ha reservoir located in Angelina County, comprised of an abundance of deep (i.e., maximum depth of 12.2 m and average depth of 2.4 m) open-water habitat (80.5% is deep open water) and a few shallow bays with isolated pockets of emergent marsh (Saalfeld 2010). Dominant aquatic species include American Lotus, Hydrilla, Ceratophyllum demersum L. (Coontail), and Nuphar lutea (L.) Sm. (Yellow Pond Lily). Dominant woody species along wetland margins include Buttonbush, Black Willow, Chinese Tallow, Water Oak, Overcup Oak, and pine (Godfrey and Wooten 1981). Little Sandy NWR consists of 1539 ha, of which ≈1100 ha are bottomland hardwood forest, located on the northern bank of the Sabine River in southern Wood County. Little Sandy NWR contains four main lentic bodies: Overton Lake, Brumley Lake, Bradford Lake, and Beaver Lake. Of these, only Brumley Lake (an impoundment of Little Sandy Creek) and Overton Lake (an impoundment of Jim Ned Creek) were used as study sites. Overton Lake is approximately 175 ha, and Brumley Lake is approximately 200 ha. Both are connected by several creeks and canals, essentially making these lakes one large wetland, hereafter referred to as Little Sandy NWR. Little Sandy NWR is characterized Figure 1. Location of counties and study sites in east Texas used to study food habits of American Alligators (Alligator mississippiensis), 2006–2008. 662 Southeastern Naturalist Vol. 10, No. 4 primarily by shallow marsh with little open water or creek channels (i.e., average depth of 0.8 m). Dominant aquatic species include Limnobium spongia (Bosc) Rich. ex Steud. (American Frog-bit), American Lotus, Cabomba caroliniana A. Gray (Carolina Fanwort), Coontail, Zizaniopsis miliacea (Michx.) Döll and Asch. (Giant Cutgrass), and Yellow Pond Lily. Woody species include Chinese Tallow, Buttonbush, Black Willow, and Morella cerifera (L.) Small (Southern Wax Myrtle) (Godfrey and Wooten 1981). Methods Capture and handling From May–October 2006, 2007, and 2008, we captured, uniquely marked, and released American Alligators at Dam B WMA, Kurth Lake, and Little Sandy NWR using several capture techniques (i.e., snake tongs, pole snares, by hand, and swim-in live traps; see Saalfeld et al. [2008] and Webb [2005] for complete capture descriptions). At night, spotlights affixed with red filters were used to locate alligators with a 4.9-m Go-Devil® boat outfitted with a 20-hp Go-Devil® mud motor. Alligators <125.0 cm in total length (TL) were captured using snares, tongs, or hands, while swim-in live traps (Ryberg and Cathey 2004) were used to capture larger alligators (>160.0 cm). Upon capture, alligators were restrained with duct tape, and each individual was sexed by cloacal examination (Chabreck 1963, Joanen and McNease 1978). For all captured individuals (regardless of size), we measured the following morphological features: total length (cm; ventral tip of snout to tip of tail), snout– vent length (cm; ventral tip of snout to proximal tip of vent), eye to nare length (cm), total head length (cm; dorsal tip of snout to distal part of head scute), tail girth (cm; circumference of tail directly behind rear legs), right hind leg length (cm), chest girth (cm; circumference of chest directly behind front legs), and mass (kg). All length measurements were obtained using a flexible tape measure, and mass was obtained using a Pesola® hanging scale (Baar, Switzerland). Food habits We obtained diet samples from alligators ranging in size from 94.7–386.0 cm in TL. After capture, we fastened alligators ranging in size from 106.0–244.0 cm (i.e., alligators <106.0 cm TL were too small, and alligators >244.0 cm TL were too large to safely and/or effectively collect diet samples using the hose-Heimlich method) to a plywood board and placed them at an incline with jaws secured open with a piece of PVC pipe and duct tape. We then utilized the hose-Heimlich method (Barr 1994, Fitzgerald 1989, Rice et al. 2005) to remove all stomach contents, by carefully inserting a Teflon hose down the esophagus and into the alligator’s stomach. An external mark for the distance the Teflon hose was inserted into the esophagus/ stomach corresponded to the fourth whirl of scutes anterior to the hind legs (Rice 2004, Rice et al. 2005). A bilge pump connected to a garden hose was then connected to the Teflon hose in the alligator’s stomach, and water was pumped (≈50 L/min) into the stomach until full. With the hose still in place, a mixture of stomach contents and water was then expelled into a collection basin. We continued this procedure until the water flushing the stomach was clear and free of any particulate matter. We 2011 D.T. Saalfeld, W.C. Conway, and G.E. Calkins 663 then poured each stomach content sample through a 0.5-mm mesh sieve, placed the sample into a labeled plastic bag, and stored at -20 °C for subsequent identification. A 0.5-mm mesh sieve was selected as it retained all food items, but allowed for mud and other particulates to flow through the sieve without clogging. We also obtained diet samples (i.e., whole stomachs) of alligators >244 cm in TL from harvested individuals during the TPWD’s annual alligator harvest at Dam B WMA. Individuals were harvested by hook and line baited with chicken. Prior to sieving, baits were removed from stomach contents, as they were easily recognizable (i.e., harvested alligators did not have time to digest the chicken, and usually chicken was still attached to the hook). Similar to those obtained from the hose-Heimlich method, all stomach contents were poured through a 0.5- mm mesh sieve, placed into a labeled plastic bag, and stored at -20 °C. Because the hose-Heimlich method is nearly 100% effective in removing all food contents from an alligator’s stomach (Barr 1994, Fitzgerald 1989, Rice et al. 2005), diet samples from whole stomachs were directly compared with diet samples from the hose-Heimlich in all analyses. We first obtained total wet mass (g) of all stomach contents and then sorted diet samples into identifiable prey items (e.g., fish, reptiles, mammals, birds, amphibians, gastropods, insects, crustaceans, or bivalves) and non-prey items (e.g., rocks, plant material, artificial objects, etc.) and identified the prey items to lowest possible taxa. We determined minimum number of individuals based upon the occurrence of specific items (e.g., fish otolith, water bug thoraxes, etc.). Along with occurrence, we obtained wet masses for each taxon within a given stomach sample. All prey items that were unidentifiable down to species were sorted into categories (e.g., fish, bird, insect mammal, and amphibian/reptile) and included in relevant analyses. Data analysis We used a chi-square analysis (PROC FREQ; SAS Institute 1999) to examine variation in prey presence/absence (i.e., fish, amphibians/reptiles, mammals, birds, and invertebrates) between or among wetlands, sexes, and sizes (breeding size [>183.0 cm TL] and non-breeding size [<183.0 cm TL]; Giles and Childs 1949, Joanen and McNease 1975, Klause 1984, McIlhenny 1934). Because most large alligators were from Dam B WMA, no large-size comparisons were made among wetlands. Additionally, we used an analysis of variance (ANOVA, PROC GLM; SAS Institute 1999) to examine differences in percent composition of prey items by wet mass (i.e., proportion [%] a prey taxon mass comprised of the total mass of a sample) and percent occurrence (i.e., proportion [%] a single prey item comprised of the total number of prey items within a diet sample) between or among sizes, sexes, and wetlands. We could not examine interactions among these variables (i.e., sex, size class, and wetland) due to sample-size limitations. An alpha level of 0.05 was used for these analyses, and least squared means separation was used to examine differences (P < 0.05). Results From 1 June 2006–31 September 2008, we obtained a total of 62 American Alligator diet samples (24 from Dam B WMA, 35 from Little Sandy NWR, and 664 Southeastern Naturalist Vol. 10, No. 4 3 from Kurth Lake; Table 1). We obtained samples from alligators ranging in size from 94.7 cm to 386.0 cm TL. The majority of the diet samples (49) were obtained non-lethally using the Hose-Heimlich technique, while fewer (13) were obtained from harvested alligators collected during hunts conducted at Dam B WMA in 2007 (n = 9) and 2008 (n = 4). Although many prey items were damaged (i.e., presumably due to digestion, jaw pressure, and/or prey capture), we identified a total of 33 different prey taxa (comprising 670 individual prey items) and 1 parasite (Table 2). Irrespective of size class, sex, and wetland, 47.8% of individual prey items identified were giant water bugs (Belostomatidae), with one or more giant water bug documented in 66.0% of all samples. Nearly all (97.3%) samples contained organic by-catch (e.g., woody debris, aquatic plants, seeds, etc.), whereas 52.3% of samples contained rocks/stones and 19.1% contained foreign matter (e.g., plastic bottle caps, fishing tackle, tent spike, shotgun shell). Additionally, 53.0% of all samples had at least 1 parasite. Invertebrates, fish, birds, mammals, and amphibians/reptiles occurred equally between males and females (P > 0.05). Invertebrates, fish, birds, and amphibians/ reptiles occurred at similar frequencies among wetlands (P > 0.05); however, mammals occurred more often in alligator diets at Dam B WMA than at other wetlands (Table 3). Irrespective of wetland, invertebrates occurred more often in non-breeding than breeding-size alligators (Table 4). Conversely, mammals and amphibians/reptiles occurred more often in breeding-size alligator’s diets, whereas birds and fish occurred in both size classes equally (Table 4). Irrespective of size, wetland, and sex, percent composition (by wet mass) was 48.2% food and 51.8% non-food (e.g., stones/rocks, plastic, woody debris, etc.). Overall, percent composition by wet mass of invertebrates, fish, birds, mammals, and amphibians/reptiles (P > 0.05) was similar between male and female alligators. Additionally, percent composition by wet mass of invertebrates, fish, amphibians/reptiles, birds, and mammals was similar among wetlands (Table 5). However, similar to frequency of food items, percent composition by wet mass for breeding-size alligators contained a greater percentage of vertebrates (mean = 42.7%) as compared to non-breeding-size alligators (mean = 25.7%). Additionally, breeding-size alligators consumed more mammals by wet mass (Table 6) than smaller alligators. Both size classes contained similar percent composition Table 1. Sample size (n), average mass (kg), average length (cm), and sex for American Alligators (Alligator mississippiensis) in which diet samples were obtained by month from Angelina-Neches/ Dam B Wildlife Management Area, Kurth Lake, and Little Sandy National Wildlife Refuge, TX, 2006–2008. Average Average Month n mass (kg) length (cm) Males Females June 13 7.5 133.5 3 10 July 16 9.8 132.1 10 6 August 16 6.8 131.5 12 4 SeptemberA 17 51.8 215.6 13 4 AAlligator harvests occurred only in September resulting in larger average mass and length measurements for this month. 2011 D.T. Saalfeld, W.C. Conway, and G.E. Calkins 665 Table 2. Food items documented in American Alligator (Alligator mississippiensis) stomach content samples (identified to lowest possible taxon) from Angelina-Neches/Dam B Wildlife Management Area (DMB), Kurth Lake (Kurth), and Little Sandy National Wildlife Refuge (LSNWR), TX, 2006–2008. Taxon DMB LSNWR Kurth MOLLUSCA Bivalvia Unionida Unionidae (freshwater mussel) X X X PLATYHELMINTHES Cestoda (tapeworm) X X ANNELIDA Clitellata (leech) X X ARTHROPODA Malacostraca Decapoda Palaemonidae (freshwater shrimp) X Cambaridae (crayfish) X X X Insecta Odonata Lestidae (spreadwing damselfly) X X Aeshnidae (dragonfly) X X Orthoptera (grasshopper) X X Hemiptera Nepidae (water scorpion) X X Belostomatidae (giant water bug) X X Coleoptera (beetle) X X X Gyrinidae (whirligig beetle) X Psephenidae (water penny) X Diptera Tipulidae (cranefly larva) X X Arachnida Araneae Pisauridae (fishing spider) X X CHORDATA Actinopterygii Lepisosteiformes Lepisosteidae Lepisosteus oculatus Winchell (Spotted Gar) X X Cypriniformes Cyprinidae Cyprinella lutrensis Baird and Girard (Red Shiner) X Cyprinodontiformes Fundulidae Fundulus spp. Lacèpéde (Top Minnow) X Poeciliidae Gambusia affinis Baird and Girard (Western Mosquitofish) X X Perciformes Centrarchidae Lepomis gulosus Cuvier (Warmouth) X Lepomis macrochirus Rafinesque (Bluegill) X X Lepomis miniatus Jordan (Redspotted Sunfish) X Micropterus salmoides Lacèpéde (Largemouth Bass) X 666 Southeastern Naturalist Vol. 10, No. 4 Table 2, continued. Taxon DMB LSNWR Kurth Amphibia Anura Hylidae Hyla cinerea Schneider (Green Tree fFrog) X Reptilia Testudines Kinosternidae Kinosternon flavescens Agassiz (Yellow Mud Turtle) X Emydidae Trachemys scripta Wied-Neuwied (Red-eared Slider) X X Squamata Colubridae Nerodia spp. Baird and Girard (unidentified water snake) X X Viperidae Agkistrodon piscivorus Lacèpéde (Water Moccasin) X X Crocodilia Alligatoridae Alligator mississippiensis Daudin (Alligator) X Aves Pelecaniformes Anhingidae Anhinga anhinga L. (Anhinga) X Ciconiiformes Ardeidae Bubulcus ibis L. (Cattle Egret) X Gruiformes Rallidae Gallinula chloropus L. (Common Moorhen) X Mammalia Rodentia Myocastoridae Myocastor coypus Molina (Nutria) X Artiodactyla Suidae Sus scrofa L. (Feral Hog) X Table 3. Presence/absence (n), chi-square (χ2), and P-values resulting from chi-square analysis of invertebrate, fish, amphibian/reptile, bird, and mammal prey item frequency among wetlands for American Alligators (Alligator mississippiensis) at Angelina-Neches/Dam B Wildlife Management Area (Dam B WMA), Kurth Lake, and Little Sandy National Wildlife Refuge (LSNWR), TX, 2006–2008. Dam B WMA Kurth Lake LSNWR Variable Present Absent Present Absent Present Absent χ2 P Invertebrates 10 1 3 0 30 3 0.30 0.862 Fish 5 6 2 1 17 16 0.43 0.805 Amphibians/reptiles 3 8 1 2 6 27 0.69 0.710 Birds 1 10 0 3 9 24 2.49 0.287 Mammals 3 8 0 3 1 32 6.53 0.038* *Significant P-values. 2011 D.T. Saalfeld, W.C. Conway, and G.E. Calkins 667 by wet mass of fish, amphibians/reptiles, and birds (Table 6). However, diets of smaller alligators had greater percent composition by wet mass of invertebrates than breeding-size alligators (Table 6). Males and females had similar percent occurrence of invertebrates, fish, amphibians/ reptiles, birds, and mammals (P > 0.05). Among wetlands, alligators also had similar percent occurrence of invertebrates, fish, amphibians/reptiles, birds, and mammals (Table 7). Similar to previous analyses, a greater percentage of invertebrate prey items were detected within non-breeding-size alligator diets (Table 8) when compared to breeding-size alligators, and breeding-size alligators had a greater percentage of amphibians/reptiles and mammals in their diet samples than non-breeding-size alligators (Table 8). Both size classes had similar percentages of birds and fish within their diet samples (Table 8). Discussion American Alligators exhibit an extremely varied diet, as evidenced by their opportunistic strategy of exploiting locally available and/or abundant prey. Many studies have documented alligator prey encompassing a wide diversity of sizes and taxa, from small insects and crustaceans to large vertebrates (Chabreck 1971, Delany and Abercrombie 1986, Rice 2004, Valentine et al. 1972, Wolfe et al. Table 4. Presence/absence (n), chi-square (χ2), and P-values resulting from chi-square analysis of invertebrate, fish, amphibian/reptile, bird, and mammal prey item frequency between size classes (breeding: >183.0 cm in total length, non-breeding: less than 183.0 cm in total length) for American Alligators (Alligator mississippiensis) at Angelina-Neches/Dam B Wildlife Management Area, Kurth Lake, and Little Sandy National Wildlife Refuge, TX in 2006–2008. Non-breeding Breeding Variable Present Absent Present Absent χ2 P Invertebrates 43 4 9 6 8.34 0.004* Fish 24 23 8 7 0.02 0.878 Amphibians/reptiles 10 37 8 7 5.67 0.017* Birds 10 37 1 14 1.66 0.197 Mammals 4 43 7 8 11.34 0.001* *Significant P-values. Table 5. Means, standard errors (SE), and F and P-values resulting from analysis of variance of percent (%) composition by wet mass of invertebrate, fish, amphibian/reptile, bird, and mammal prey items for American Alligators (Alligator mississippiensis) among wetlands (Angelina-Neches/ Dam B Wildlife Management Area [Dam B WMA], Kurth Lake, and Little Sandy National Wildlife Refuge [LSNWR]), TX, 2006–2008. Dam B WMA Kurth Lake LSNWR Variable Mean (%) SE Mean (%) SE Mean (%) SE F P Invertebrates 15.7 5.3 34.6 18.6 16.7 3.4 0.93 0.402 Fish 5.9 4.2 0.8 0.8 10.2 2.8 0.68 0.513 Amphibians/reptiles 11.9 5.1 26.0 26.0 0.3 0.1 1.43 0.248 Birds 0.0 0.0 0.0 0.0 11.3 4.6 2.28 0.111 Mammals 11.4 5.4 0.0 0.0 2.0 2.0 1.94 0.153 668 Southeastern Naturalist Vol. 10, No. 4 1987). Alligator food habits in this study mirrored those observed in previous studies, in that alligators appear to forage opportunistically, and in the feeding process, often ingest relatively large quantities of non-food items. Although most food habit studies documented similar vertebrate prey items (i.e., amphibians/ reptiles, mammals, and fish; Barr 1994,1997; Delany and Abercrombie 1986; Table 6. Means, standard errors (SE), and F and P-values resulting from analysis of variance of percent (%) composition by wet mass of invertebrate, fish, amphibian/reptile, bird, and mammal prey items between size classes (breeding: >183.0 cm in total length, non-breeding: <183.0 cm in total length) for American Alligators (Alligator mississippiensis) at Angelina-Neches/Dam B Wildlife Management Area, Kurth Lake, and Little Sandy National Wildlife Refuge, TX, 2006–2008. Non-breeding Breeding Variable Mean (%) SE Mean (%) SE F P Invertebrates 21.5 3.6 3.5 1.8 7.80 0.007* Fish 8.0 2.2 8.2 6.7 0.00 0.967 Amphibians/reptiles 7.6 3.0 14.6 6.7 1.17 0.283 Birds 8.4 3.5 0.0 0.0 1.85 0.178 Mammals 1.8 1.5 17.2 8.4 8.29 0.006* *Significant P-values. Table 7. Means, standard errors (SE), and F and P-values resulting from analysis of variance of percent (%) occurrence of invertebrate, fish, amphibian/reptile, bird, and mammal prey items for American Alligators (Alligator mississippiensis) among wetlands (Angelina-Neches/Dam B Wildlife Management Area [Dam B WMA], Kurth Lake, and Little Sandy National Wildlife Refuge [LSNWR]), Texas, 2006–2008. Dam B WMA Kurth Lake LSNWR Variable Mean (%) SE Mean (%) SE Mean (%) SE F P Invertebrates 52.0 8.4 70.0 6.9 68.5 5.0 1.75 0.183 Fish 16.6 5.7 18.9 11.6 14.6 3.2 0.08 0.920 Amphibians/reptiles 14.7 5.3 11.1 11.1 6.9 3.2 0.93 0.401 Birds 2.1 1.4 0.0 0.0 7.1 3.3 0.92 0.406 Mammals 14.6 5.7 0.0 0.0 2.9 2.9 2.25 0.115 Table 8. Means, standard errors (SE), and F and P-values resulting from analysis of variance of percent (%) occurrence of invertebrate, fish, amphibian/reptile, bird, and mammal prey items between size classes (breeding >183.0 cm total length, non-breeding <183.0 cm total length) for American Alligators (Alligator mississippiensis) at Angelina-Neches/Dam B Wildlife Management Area, Kurth Lake, and Little Sandy National Wildlife Refuge, TX, 2006–2008. Non-breeding Breeding Variable Mean (%) SE Mean (%) SE F P Invertebrates 71.7 4.3 32.7 1.8 18.72 <0.001* Fish 12.7 2.6 24.4 6.7 3.13 0.082 Amphibians/reptiles 6.6 2.7 21.1 6.7 5.41 0.024* Birds 5.8 2.5 1.7 0.0 0.83 0.365 Mammals 3.2 2.2 20.1 8.4 7.34 0.009* *Significant P-values. 2011 D.T. Saalfeld, W.C. Conway, and G.E. Calkins 669 McNease and Joanen 1977), food availability estimates are difficult to obtain due to the diversity of prey items consumed and the apparently very generalized opportunistic foraging strategy employed by alligators, regardless of region. Such constraints make it difficult to estimate specific food selectivity or preferences, where any geographic differences in food habits are most likely influenced by food availability, rather than species-specific selection processes (Barr 1994, 1997; Delany and Abercrombie 1986; McNease and Joanen 1977). In this study, fish were the most prevalent vertebrates in diet samples; however, most fish were found in samples from Little Sandy NWR, where over 70% of samples in which fish were present were from this study site. Although fish have short residence times in alligator digestive systems (Barr 1997, Rice 2004), they still occurred in diet samples more frequently than other vertebrates (that have longer residence time), indicating that fish may be the most important, or most available, vertebrate prey item for all size classes at Little Sandy NWR. Conversely, fish were less prevalent in the samples from Dam B WMA (25.0% of diet samples contained fish) and Kurth Lake (4.8% of diet samples contained fish). At Dam B WMA, fewer fish were available due to an extended water drawdown event occurring during 2006 and 2007 for repairs to the primary dam. This drawdown lasted over a year and substantially reduced the abundance of both fish and other aquatic vertebrates (e.g., turtles; Bill Adams and Gary Calkins, Texas Parks and Wildlife Department, Jasper, TX, pers. comm.). After repairs were completed and water levels returned to normal pool level, many exotic invasive plant species such as Alligatorweed, Giant Salvinia, and Water Hyacinth became re-established, expanded areal extent, and increased their densities throughout the study site. These exotic invasive plants formed thick mats, and appeared to both reduce alligator movements and limit their access to shallow water habitat where alligators tend to feed more efficiently (Saalfeld 2010). Decreased fish abundances after the drawdown likely forced alligators to focus on alternative prey items, such as mammals and other alligators. As cannibalism was only documented at Dam B WMA after the drawdown, it likely resulted when low water levels and subsequent expansion of exotic invasive plant species led to unusually high concentrations of alligators of all size classes. Despite generally being opportunistic, alligators appear to shift diets from invertebrates to vertebrates as they increase in size, a phenomenon documented in many studies (Barr 1997, Chabreck 1971, Delany and Abercrombie 1986, Delany et al. 1999, Giles and Childs 1949, McIlhenny 1935, McNease and Joanen 1977, Taylor 1986, Valentine et al. 1972, Wolfe et al. 1987). Specifically, within this study, mammals and amphibians/reptiles occurred most often in diets of larger alligators (>183.0 cm), whereas insects, crayfish, and small fish occurred most often in diets of smaller alligators (i.e., <183.0 cm). However, insects may be over-represented in crocodilian diet samples due to longer residence times of exoskeletons and secondary ingestion (Barr 1994, 1997; Garnett 1985; Jackson et al. 1974; Wolfe et al. 1987). Additionally, after the drawdown at Dam B WMA, these species are likely the first to re-colonize, leading to their greater frequencies in diet samples. Because they occurred at similar frequencies among wetlands, these are all important food items for alligators in east Texas, especially for smaller 670 Southeastern Naturalist Vol. 10, No. 4 size classes. As alligators increase in size, they become more capable (i.e., due to changes in ontogenetic skull structure) of exploiting larger food resources (Delany 1990, Delany et al. 1999, Dodson 1975), but also have greater energy requirements and metabolic costs (Dodson 1975, Thorbjarnarson 1993). Therefore, shifting to larger prey items meets increased energy demands (Delany et al. 1999) and maximizes feeding efficiency for larger alligators (Wolfe et al. 1987). Because alligators shift to larger prey items as they increase in size, the prevalence of mammals in diet samples from Dam B WMA could be due to the average size of alligators sampled (Shoop and Ruckdeschel 1990), rather than other compounding factors such as the drawdown and food availability. Overall, alligators within these study sites exhibit faster growth rates, but are in poorer condition (i.e., alligators weigh less than alligators of similar lengths) compared to alligators in other geographic regions (Saalfeld 2010, Saalfeld et al. 2008). Although the role of food availability on alligator growth and condition is difficult to assess, it is likely an important factor influencing geographic variation in growth rates and condition. For example, the water drawdown and rapid expansion of exotic invasive species at Dam B WMA could have reduced prey abundances and/or availability, causing a reduction in body condition of resident alligators. Moreover, the abundance of open water habitat at Kurth Lake could have limited prey diversity and availability resulting in both slower growth rates and poorer condition. Little Sandy NWR contained abundant sources of fish and wading birds (numerous vegetated islands for rookeries); however, many of these prey items were not always available. For example, only young wading birds (which most likely fell out of nests) were found in diet samples. Wading birds synchronize nesting and migration, resulting in only a short time span when these prey items are available to alligators. Additionally, other large vertebrate prey items (e.g., mammals) may not occur in high densities within this wetland due to its small size (375 ha) and control management efforts exerted to reduce their density. Therefore, relying mainly upon fish and seasonally available wading birds may result in the poor condition of alligators observed in this study site. In addition to food availability, the high percentage of non-food by-catch documented during this study could be an indicator of reduced feeding efficiency within these wetlands. No studies have reported such high percentage composition of non-food items (Fogarty and Albury 1967, Platt et al. 1990). Although we did not find a correlation between percentage of non-food by-catch and condition (D.T. Saalfeld et al., unpubl. data), it is possible that habitat conditions such as reduced water clarity or increased vegetation density (both of which were documented within these wetlands) could influence feeding efficiency and, ultimately, condition. To date, few studies have focused on inland American Alligators, especially in Texas. This study supports the notion that faster grow rates and poorer condition of alligators within this region (Saalfeld et al. 2008) could potentially be attributed to food availability. To better manage these populations, future work should assess abundances and nutritive quality of dominant prey items (e.g., mammals, amphibians/reptiles, fish, wading birds, and invertebrates). As diets tend to mirror prey abundances (Barr 1994, 1997; Delany and Abercrombie 1986; Mc- Nease and Joanen 1977), changes in prey populations (e.g., fish kills following 2011 D.T. Saalfeld, W.C. Conway, and G.E. Calkins 671 drawdown), may result in lower quality prey (e.g., invertebrates and small fish) being consumed. Therefore, understanding food availability and quality could provide insights into selection patterns influencing growth rates and condition. Acknowledgments Financial, logistical, and technical support was provided in part by the Texas Parks and Wildlife Department, and the Arthur Temple College of Forestry and Agriculture, Stephen F. Austin State University. Special appreciation to the staff of Martin Dies, Jr. State Park, US Fish and Wildlife Service, Jim Neal, and Abitibi Consolidated Industries, for additional logistical, financial, and technical support. We thank C. Comer, I. Hung, M. Kwiatkowski, and D. Scognamillo for comments and reviews on previous versions of this manuscript. Finally, we thank Ragan White, Allen Rainey, Ben Koerth, Jerry Staton, Austin Wadyko, Matt Buckingham, Jake Rustin, Brandon McNeill, and Sarah Saalfeld for assisting with data collection for this study. Literature Cited Barr, B.R. 1994. Dietary studies on the American Alligator, Alligator mississippiensis, in southern Florida. M.Sc. Thesis. University of Miami, Coral Gables, FL. 73 pp. Barr, B.R. 1997. Food habits of the American Alligator, Alligator mississippiensis, in the southern Everglades. Ph.D. Dissertation. University of Miami, Coral Gables, FL. 243 pp. Chabreck, R.H. 1963. Methods of capturing, marking, and sexing alligators. Proceedings of the Southeastern Association of Game and Fish Commissioners 17:47–50. Chabreck, R.H. 1971. The foods and feeding habits of alligators from fresh and saline environments in Louisiana. Proceedings of the Southeastern Association of Game and Fish Commissioners 25:117–124. Delany, M.F. 1990. Late summer diet of juvenile American Alligators. Journal of Herpetology 24:418–421. Delany, M.F., and C.L. Abercrombie. 1986. American Alligator food habits in northcentral Florida. Journal of Wildlife Management 50:348–353. Delany, M.F., S.B. Linda, and C.T. Moore. 1999. Diet and condition of American Alligators in 4 Florida lakes. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 53:375–389. Dodson, P. 1975. Functional and ecological significance of relative growth in alligators. Journal of Zoology 175:315–355. Fitzgerald, L.A. 1989. An evaluation of stomach flushing techniques for crocodilians. Journal of Herpetology 23:170–172. Fogarty, M.J., and J.D. Albury. 1967. Late summer foods of young alligators in Florida. Proceedings of the Southeastern Association of Game and Fish Commissioners 21:220–222. Garnett, S.T. 1985. The consequences of slow chitin digestion on crocodilian diet analyses. Journal of Herpetology 19:303–304. Giles, L.W., and V.L. Childs. 1949. Alligator management of the Sabine National Wildlife Refuge. Journal of Wildlife Management 13:16–28. Godfrey, R.K., and J.W. Wooten. 1981. Aquatic and Wetland Plants of Southeastern United States: Dicotyledons. University of Georgia Press, Athens, GA. Jackson, J.F., H.W. Campbell, and K.E. Campbell, Jr. 1974. The feeding habits of crocodilians: Validity of the evidence from stomach contents. Journal of Herpetology 8:378–381. Joanen, T., and L. McNease. 1975. Notes on the reproductive biology and captive propagation of the American Alligator. Proceedings of the Southeastern Association of Game and Fish Commissioners 29:407–414. 672 Southeastern Naturalist Vol. 10, No. 4 Joanen, T., and L. McNease. 1978. The cloaca sexing method for immature alligators. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 32:179–181. Klause, S. 1984. Reproductive characteristics of the American Alligator (Alligator mississippiensis) in North Carolina. M.Sc. Thesis, North Carolina State, Raleigh, NC. 85 pp. Mazzotti, F.J., G.R. Best, L.A. Brandt, M.S. Cherkiss, B.M. Jeffery, and K.G. Rice. 2009. Alligators and crocodiles as indicators for restoration of Everglades ecosystems. Ecological Indicators 9S:S137–S149. McIlhenny, E.A. 1934. Notes on incubation and growth of alligators. Copeia 1934:80–88. McIlhenny, E.A. 1935. The Alligator’s Life History. Christopher Publishing House, Boston, MA.. McNease, L., and T. Joanen. 1977. Alligator diets in relation to marsh salinity. Proceedings of the Southeastern Association of Game and Fish Commissioners 31:36–40. McNease, L., and T. Joanen. 1981. Nutrition of alligators. Proceedings of the Annual Alligator Production Conference 1:15–28. Platt, S.G., C.G. Brantley, and R.W. Hastings. 1990. Food habits of juvenile American Alligators in the upper Lake Pontchartrain estuary. Northeast Gulf Science 11:123–130. Rice, A.N. 2004. Diet and condition of American Alligators (Alligator mississippiensis) in three central Florida lakes. M.Sc. Thesis. University of Florida, Gainesville, FL. 89 pp. Rice, A.N., J.P. Ross, A.G. Finger, and R. Owen. 2005. Application and evaluation of a stomach flushing technique for alligators. Herpetological Review 36:400–401. Rosenblatt, A.E., and M.R. Heithaus. 2011. Does variation in movement tactics and trophic interactions among American Alligators create habitat linkages? Journal of Animal Ecology 80:786–798. Ryberg, W.A., and J.C. Cathey. 2004. A box-trap design to capture alligators in forested wetland habitats. Wildlife Society Bulletin 32:183–187. Saalfeld, D.T. 2010. American Alligator (Alligator mississippiensis) ecology in inland wetlands of east Texas. Ph.D. Dissertation. Stephen F. Austin State University, Nacogdoches, TX. 279 pp. Saalfeld, D.T., K.K. Webb, W.C. Conway, G.E. Calkins, and J.P. Duguay. 2008. Growth and condition of American Alligators (Alligator mississippiensis) in an inland wetland of east Texas. Southeastern Naturalist 7:541–550. SAS Institute. 1999. SAS/STAT, Version 9.SAS Institute, Inc, Cary, NC. Shoop, C.R., and C.A. Ruckdeschel. 1990. Alligators as predators on terrestrial mammals. American Midland Naturalist 124:407–412. Taylor, D. 1986. Fall foods of adult alligators from cypress lake habitat, Louisiana. Proceedings of the Annual Conference of Southeast Association of Fish and Wildlife Agencies 40:338–341. Thorbjarnarson, J.B. 1993. Diet of the Spectacled Caiman (Caiman crocodilus) in the central Venezuelan Llanos. Herpetologica 49:108–117. Valentine, J.M., J.R. Walther, K.M. McCartney, and L.M. Ivy. 1972. Alligator diets on the Sabine National Wildlife Refuge, Louisiana. Journal of Wildlife Management 36:809–815. Webb, K.K. 2005. Population ecology and habitat use of American Alligators in inland freshwater wetlands of east Texas. M.Sc. Thesis. Stephen F. Austin State University, Nacogdoches, TX. 123 pp. Webb, K.K., W.C. Conway, G.E. Calkins, and J.P. Duguay. 2009. Habitat use of American Alligators in east Texas. Journal of Wildlife Management 73:566–572. Wolfe, J.L., D.K. Bradshaw, and R.H. Chabreck. 1987. Alligator feeding habits: New data and a review. Northeast Gulf Science 9:1–8.