107 Evaluating Trap Alternatives for Removal of Salvator merianae (Black and White Tegu) Michael L. Avery1,*, John S. Humphrey1, and Richard M. Engeman2 Abstract - Salvator merianae (Argentine Black and White Tegu, hereafter, Tegu) is an omnivorous, burrowing lizard native to South America. Tegus were introduced through the pet trade, and free-ranging populations now threaten many native species in Florida. As Tegu control programs expand and more traps are deployed, the need for a simple, inexpensive trap increases. To date, there has been no experimental effort to compare types of traps or alternative lures. In this study, we evaluated responses of 12 captive Tegus to several alternative trap/bait combinations. We video-recorded each of the trials and scored the outcomes based on the trap the Tegu entered first. Our results suggest that alternative trap/lure combinations, such as traps made of PVC pipe baited with commercial mouse-based trap lure, might be just as effective at capturing Tegus, and thus could be less expensive options for Tegu control programs. Trials with captive animals do not necessarily predict outcomes with free-ranging animals, and we recommend well-designed field trials as a next step. Introduction Salvator merianae (Duméril and Bibron) (Argentine Black and White Tegu), native to South America, is a popular animal in the pet trade. Tegus are established with healthy populations in south (Miami-Dade County) and west-central (Polk and Hillsborough counties) Florida from accidental and/or intentional releases (Krysko et al. 2011). The first Tegu nest in the wild in Florida was recently described from south Florida (Pernas et al. 2012). Tegus are omnivorous burrowing animals and therefore are threats to many native species in Florida, including Gopherus polyphemus Daudin (Gopher Tortoise) and Athene cunicularia floridana Ridgway (Florida Burrowing Owl), each of which is considered a species of special management concern (Engeman et al. 2011). Current population-control measures consist of live-trapping with chicken eggs as bait, and then euthanasia of all Tegus caught. Controlled experiments to compare types of traps or lures to improve trap efficacy are lacking. In this study, we documented the responses of wild-caught captive Tegus to several trap/bait combinations to determine if current standard practice can be improved. Methods Colleagues in south Florida live-trapped Tegus for our study in accordance with the standard operating methods and equipment of ongoing Tegu removal activities. 1USDA/APHIS/Wildlife Services, Florida Field Station, 2820 East University Avenue, Gainesville, FL 32641. 2USDA/APHIS/Wildlife Services, National Wildlife Research Center, 24101 LaPorte Drive, Fort Collins, CO 80521. *Corresponding author - michael.l.avery@ aphis.usda.gov. Manuscript Editor: Michael Cove Everglades Invasive Species 2016 Southeastern Naturalist 15(Special Issue 8):107–113 Southeastern Naturalist M.L. Avery, J.S. Humphrey, and R.M. Engeman 2016 108 Vol. 15, Special Issue 8 We maintained the animals individually at our field station in Gainesville in secure outdoor pens equipped with refugia, water pans, and food bowls (Fig. 1). These home pens (1.5 m x 3.0 m) were made of aluminum frames covered with plasticcoated 2.54 cm x 2.54 cm wire-mesh, and included a floor to prevent escape by Figure 1. (Top) Tegu 89 entering a live trap wrapped in black plastic and baited with commercial mouse paste (My-T Mouse). The standard live-trap on the right side was wrapped in burlap and baited with a chicken egg (16 July 2013). (Bottom) Tegu 4 entering a PVC live-trap baited with a commercial mouse paste (My-T Mouse). The standard live trap on the right side was wrapped in burlap and baited with a chicken egg (19 July 2013). Southeastern Naturalist 109 M.L. Avery, J.S. Humphrey, and R.M. Engeman 2016 Vol. 15, Special Issue 8 burrowing. Each home pen was adjacent to a 3.0 m x 6.1 m test pen, which each animal could access only when the door to its home pen was opened. We conducted trials from 11 July to 12 September 2013 and 10 June to 18 July 2014. At 0830–0900, we placed in the test pen the 2 trap/lure combinations to be evaluated as a 2-choice trial and opened the door from the home cage. We baited 1 live trap (1.5 m long x 20.3 cm wide x 20.3 cm high; Tomahawk Live Trap, Hazelhurst, WI) with a fresh chicken egg, simulating the standard field-trapping situation. In the field, it is standard practice to shade the trap with vegetation to provide a trapped animal relief from the sun. For consistency throughout our study, we provided shade by wrapping the trap with burlap. The alternate trap was either (1) a live trap like the standard trap except covered in black plastic to create a darkened refuge, or (2) a trap made from a 1-m-long piece of 15-cm-diameter PVC pipe with a one-way wire-mesh entry door based on the design used in the successful Boiga irregularis Merrem (Brown Treesnake) trapping program on Guam (e.g., Engeman and Vice 2001, Linnell et al. 1998). Each trap had a one-way exit to allow the animal to escape and reduce possible trap shyness from repeated capture. As funding and budgets are always a concern with species-control programs, we compared costs of live traps from commercial suppliers to the costs of materials obtained at a local home improvement store needed to build an ~1-m-long PVC trap. We baited the alternate traps with an egg, melon-oil scent, commercial trapper lure, or nothing. The melon-oil lure was a commercial cantaloupe fragrance (Peak Candle Supplies, Denver, CO). We presented it by adding 1.0–1.5 ml of the liquid to a cotton ball placed inside a small, perforated plastic vial. My-T Mouse® Blackie’s Blend is a commercial mouse paste used to capture predators such as foxes and cats. We presented the mouse paste by half-filling a 35-mm film canister perforated with 6 small holes. We suspended the melon scent and mouse paste lures from the top of the trap by a thin wire. We randomly determined whether to place each trap on the north or south side of the test pen. When the traps were in place, we opened the access door from the Tegu’s home pen. The trial ended when the Tegu entered one of the traps. We used 12 Tegus in these trials. Each animal experienced each of the standard trap–alternate trap pairings once. We randomized the sequence of the presentations for each animal; consecutive trials were separated by 1–5 d. We video-recorded each trial and reviewed the recordings to document the relevant timing and behavioral outcomes. We noted the trap that each animal contacted first and the trap that each animal eventually entered. We recorded the time each Tegu entered the test pen and the time it entered the trap (latency). We used latency times to trap entry to examine changing behavior towards traps as the experiment progressed and each lizard experienced more trials. We evaluated latency to enter a trap across trials using one-way analysis of variance (ANOVA). Each trial had 2 trap alternatives, but latency could only be measured for the trap the lizard entered; thus, comparisons of latency times between the trap alternatives would have been inappropriate. Each animal was originally captured from the wild using the “standard” field method. Thus, these animals had proven vulnerable to that capture method. Southeastern Naturalist M.L. Avery, J.S. Humphrey, and R.M. Engeman 2016 110 Vol. 15, Special Issue 8 Therefore, our aim was to assess which of the alternative trap/lure combinations had the highest catch rate when in competition with the standard method, and whether any of the alternative trap/lure combinations appeared superior to the standard. We used Cochran’s Q test, a repeated measures chi-square test for categorical data, to detect differences in capture rates among alternative trap/lure combinations competing head-to-head with the standard trapping setup. Results Each animal entered a trap in each of its 6 trials (Fig. 1). We observed no statistically significant differences among the 6 alternate trap-lure combinations (Cochran’s Q χ2 = 8.117, df = 5, P = 0.15). For 5 of the 6 alternate trap-lure combinations, selection of the alternate was ± 1 of 50% (Table 1). For example, when paired with the PVC trap/egg bait, Tegus selected the standard trap 7 times versus 5 times for the PVC trap. In all trials using live traps covered in black plastic, Tegus selected the plastic-covered trap 18 times, a value equal to the rate at which they selected the standard burlap-covered trap. In contrast, Tegus selected the PVC trap only 11 times and the standard trap 25 times. Latency from the time a Tegu left its home pen until it entered a trap varied across trials (F5,66 = 2.35; P = 0.050). Tegus took an average of almost 27 min (range = 1–101 min) to enter a trap during their initial trial (Fig. 2). Mean latency dropped to almost 17 min during trial 2, and then remained less than 10 min for trials 3–6 (Fig. 2). On 62 occasions, we were able to record the trap that the test animal first contacted. The type of trap first contacted and investigated affected the type of trap eventually entered (χ2 = 5.034; P = 0.025). More than half (35; 56.5%) of the first contacts were with the alternative trap. The animal proceeded to enter the alternative trap 19 of those 35 times (54.3%). In 27 trials, the test animal contacted the standard trap first, and 20 (74.1%) of those animals entered the standard trap. Looking at the costs of currently available traps compared to the traps we built shows that the alternate PVC traps were significantly less expensive. Commercial traps cost around $75–$85. The materials for one of our PVC traps cost $24.41, and each required 10–15 minutes to assemble (Table 2). Table 1. Outcomes of trap-choice trials involving 12 Tegus exposed to each of 6 trap/lure pairings. Each trial ended when the test animal entered one of the two available traps. Times Times Standard trap-lure Entered first Alternate trap-lure Entered first Live trap (burlap)-egg 7 vs Live trap (black plastic)-no lure 5 Live trap (burlap)-egg 5 vs Live trap (black plastic)-melon oil 7 Live trap (burlap)-egg 6 vs Live trap (black plastic)-mouse paste 6 Live trap (burlap)-egg 7 vs PVC trap-egg 5 Live trap (burlap)-egg 11 vs PVC trap-melon oil 1 Live trap (burlap)-egg 7 vs PVC trap-mouse paste 5 Southeastern Naturalist 111 M.L. Avery, J.S. Humphrey, and R.M. Engeman 2016 Vol. 15, Special Issue 8 Discussion The latency to enter a trap dropped rapidly after the initial trials. The pattern suggests that after their initial exposure, the Tegus learned from their previous experiences, adapted to the test conditions, and entered one of the paired traps with minimal hesitation. That they were originally caught from the wild using the standard trap method was an initial demonstration of their willingness to enter a trap. Table 2. Cost of commercial live-traps compared to the cost of constructing a PVC-pipe trap of similar length from materials obtained at a local home-improvement store. Smaller commercial live-traps are readily available, and prices can vary considerably. Trap Cost Commercial live-trap (on-line prices as of 8 April 2015) Havahart® x-large 2-door (107 cm x 28 cm x 33 cm; 42 in x 1 1 in x 13 in) $74.99 Tomahawk® large Raccoon 1-door (91 cm x 30.5 cm x 30.5 cm; 36 in x 12 in x 1 2 in) $84.58 PVC pipe trap (in-store prices from Lowe’s, 1 April 15) Pipe 102 cm long, 15.24-cm diameter (40 in long, 6-in diameter) $11.66A Drain cap-end $8.99 Trap door $0.22B Galvanized 1.59 cm (5/8”) steel nut door -weight $0.54 Total materials $24.41 Labor $4.00C Total cost $28.41 A3.05 m; 10 ft PVC green drainage pipe = $34.99; 3 traps per 3.05-m [10-ft] length. B2.44 m x 0/7 m; 96 in x 27 in galvanized steel stucco sheet = $8.47; thirty-seven 25.4 cm x 17.8 cm; 10 in x 7 in doors per sheet, hand-cut to shape. C~10–15 min/trap; 4 traps/hr. @ $16/hr = $4 labor per trap. Figure 2. Mean latency of 12 individually caged Argentine Black and White Tegus to enter a live trap declined cross a series of 6 trials. Capped vertical bars denote 1 SE. Southeastern Naturalist M.L. Avery, J.S. Humphrey, and R.M. Engeman 2016 112 Vol. 15, Special Issue 8 Alternate Tegu trap–lure combinations we tested did not differ in capture effectiveness from the standard system currently in field use. Each of our Tegus had already demonstrated a vulnerability to the standard-trap system; thus, its use as a basis for comparison with each of the alternate systems could imply a severe preference test for the alternates in a 2-choice trial. Our results are encouraging in that there appear to be viable trapping alternatives for Tegu control programs. We recognize that results from tests of captive animals are not necessarily representative of responses by free-ranging animals. Nevertheless, the results we obtained indicate that Tegus are willing to enter traps of various shapes and sizes baited with lures other than eggs. The implication is that encountering the trap is likely the key element in successful trapping, not necessarily the type of trap or bait used, although bait type might be important in attracting the animal so that an encounter occurs. Our findings can be expanded through follow-up pen tests within a larger arena or preferably with field trials. Further evaluation of the PVC traps is especially warranted because they are considerably less expensive than the standard live-trap. Multiple PVC traps can be assembled for the cost of a commercial live trap of similar length, although smaller commercial live-traps and alternative commercial sources might reduce the economic advantage of the PVC approach. Cost is a major constraining factor in wildlife field-projects; thus, deployment of PVC traps would be preferable to standard live traps, provided both have similar degrees of efficacy. If encounter rate is the key factor for capturing Tegus regardless of trap type, then, given a fixed budget and the fact that the total cost of materials and labor for the PVC traps are only about 33–40% as much as the price of the live-traps, PVC traps could be deployed in much greater numbers than commercial live-traps to maximize encounter rates, and therefore capture rates. The importance of trap-encounter rate and the economy of PVC-type traps make them excellent for intensive trapping efforts. Furthermore, damage or losses of traps to vandalism would be less impactful relative to expensive commercial traps, and replacement cost would be minimal. Acknowledgments We thank J. Fobbs and J. Edwards for providing wild-caught Tegus and K. Keacher and E. Bruce for care and maintenance of the Tegus and for contributing to the design and construction of the Tegu burrows. G. Lujano assisted with data collection and animal feeding. Literature Cited Engeman, R.M., and D.S. Vice. 2001. Objectives and integrated approaches for the control of Brown Tree Snakes. Integrated Pest Management Reviews 6:59–76. Engeman, R.M., E. Jacobson, M.L. Avery, and W.E. Meshaka Jr. 2011. The aggressive invasion of exotic reptiles in Florida with a focus on prominent species: A review. Current Zoology 57:599–612. Krysko, K.L., K.M. Enge, and P.E. Moler. 2011. Atlas of amphibians and reptiles in Florida. Final Report, Project Agreement 08013, Florida Fish and Wildlife Conservation Commission, Tallahassee, FL. 524 pp. Southeastern Naturalist 113 M.L. Avery, J.S. Humphrey, and R.M. Engeman 2016 Vol. 15, Special Issue 8 Linell, M.A., R.M. Engeman, M.E. Pitzler, M.O. Watten, G.F. Whitehead, and R.C. Miller. 1998. An evaluation of two designs of stamped metal trap-flaps for use in operational trapping of Brown Tree Snakes (Boiga irregularis). Snake 28:14–18. Pernas, T., D.J. Giardina, A. McKinley, A. Parns, and F.J. Mazzotti. 2012. First observations of nesting by the Argentine Black and White Tegu, Tupinambis merianae, in south Florida. Southeastern Naturalist 11:765–770.