ISSN: 0098-4590 Florida Scientist N N N + + 1 + ^i- + + + + 1 1 + xr m O O + + + ^ + ± ^+ "cd cd N [VOL 68 + + — \ i cd 1 es H + CO S (N _c + C _o e o OO O + co C '3 + CI + + 40 (< 30 subadult 1986) We mm SVL) as established one- each sampling session began two hours after sunset and lasted it mm SVL) or by a number of investigators (Rose and Barbour, 1968; Selcer, excluded any geckos whose size could not be accurately established (< 10 individuals) categorized the perch height of each gecko as 'low' if the gecko was with and was < lm above ground, or 'high' if We >lm —We sampled 50 one-story walls on a weekly basis between September and December Survey 2001 Walls were visited approximately two hours after sunset, and the same categorize adults and subadults In addition to documenting perch height, we criteria were used to also quantified sociality, exposure, and selected surface temperature using the following methods: (1) We defined sociality using these criteria: "alone" when a gecko had no other individual within 50-cm radius, a "pair" when two geckos were within a 50-cm radius, and a "group" when three or more geckos were found together Geckos were labeled 'exposed' when a portion of the trunk was visible, and "not exposed" when the entire trunk of a gecko's body was hidden behind a wall fixture Using a Raytek Raynger ST model temperature gun, we recorded the substrate temperature at a spot a (2) (3) within 5cm of the center of the gecko's body —We Survey sampled 160 one-story walls between the months of March and June 2002 same sampling regime and equipment visited each wall twice using the We used chi-square subadults for all tests to compare perch data from Survey maximum and minimum values for ratios and 95% their exposure for data from Survey in a 2x2 contingency comparison between and Survey and was limited to Survey The odds ratio, defined as the odds of success of an event measure of association between two events (Agresti, 1996) The and pair and group for Survey required the use of an amended pair, estimator (addition of 0.5 to each cell) to calculate the odds ratio as table was zero odds ratio (Agresti, 1996) In cases was inconclusive since Results sightings (26%) at a where the 95% the number one —Survey —We recorded Of the adult observations, 97 low height (Table a high height and 38 confidence intervals were calculated for perch height, sociality, and table, represents a solitary test and exposure between adults and We calculated means, standard deviations, and data We used a paired t-test to compare substrate temperature all temperature of adults with subadults This Odds height, sociality, and Survey We as previously mentioned 1) more than one signifies of the contingency independence (Agresti, 1996) 176 (131 adults, (74%) occurred With respect (84%) were located cell confidence interval contained the number one, the at at to subadults, 45 subadults) gecko a high height, and 34 (16%) were located a low height (Table 1) at Perch height FLORIDA SCIENTIST 208 Table chi-square per [VOL 68 Perch height of adult and subadult H turcicus arranged by survey, a = distribution of table cell; b = p-value of cell using Observations degree of freedom Observed Expected Cell frequency frequency chi-square P-value Perch Height of Adults Survey High 97 Low 34 74.0% 26.0% 77.4 4.96 0.026 53.6 7.16 0.007 247.1 9.66 107.9 22.13 — — Survey High 296 Low 59 83.3% 16.6% High 170 71.7% Low 67 28.3% — — 0.002 < 0.0001 Survey — — Perch Height of Subadults Survey High 15.6% 26.6 14.43 Low 38 84.4% 18.4 20.85 High 105 15.51 116 47.5% 52.5% 153.9 Low 67.1 35.55 — — — — — — < < 0.0001 < < 0.0001 0.0001 Survey 0.0001 Survey High Low was not independent of gecko size groups (x — — p =