We obtained demographic data on more than 60 gopher tortoise (Gopherus polyphemus) populations in Florida before the emergence of upper respiratory tract disease (URTD). We later resurveyed 10 populations to compare demographic profiles at sites where antibodies to Mycoplasma agassizii were detected subsequently and at sites where they were not. We screened for antibodies and checked for potential biases in studying URTD by determining whether individuals caught above ground were representative of a population as a whole and whether the probability of detection of seropositive (plus serologically suspect) individuals is a function of sample size. Some sites with no indication of decline had relatively high percentages of seropositive individuals, and some sites exhibiting substantial population decline had no or low percentages of seropositive individuals; therefore, our results do not unambiguously implicate the presence of URTD in the population declines. Seropositive individuals occurred at 4 sites not known previously to have them; therefore, our results indicate that exposure to Mycoplasma agassizii has been more widespread than heretofore suspected. The percentage of individuals determined to be seropositive (plus suspect) tends to be related positively to number of individuals tested and the sizes of individuals caught above ground were not always expected from the size distribution of a population; therefore, our results indicate that sampling method can influence the estimate of percentage of seropositive individuals in the population. We suggest that the simplistic conservation response that we have taken to the emergence of URTD may need to be reconsidered and that maintaining or creating conditions necessary to minimize the chance of re-emergence of URTD, to prevent URTD from reaching epidemic proportions, and to allow populations to recover from URTD is important.