Rigorous estimates of survival rates are critical for predicting the dynamics and persistence of populations and for devising and implementing management interventions for species conservation.  Despite the conservation importance of the Florida panther (Puma concolor coryi), rigorous estimates of kitten survival are currently unavailable for this critically endangered large felid, and debate persists concerning the demographic effects of the introgression of Texas pumas (P. c. stanleyana) into the population in 1995.  Using long-term radio-telemetry (1982-2008) and passive integrated transponder (PIT)-tagging (1995-2008) data within a live-recapture, dead recovery modeling framework, we estimated kitten survival and assessed the impact of the following variables on survival:  age within the first year of life, sex, season, litter size, and heterozygosity.  We also tested the hypothesis that kitten survival would increase over time following introgression in 1995 due to increased genetic vigor. We used information-theoretic techniques (QAIC_c) for model comparison and statistical inferences. 

Results/Conclusions
A total of 373 panthers was marked, of which 249 were PIT-tagged in the den and 49 were radio-collared as older kittens (approximate ages 5 – 12 months).  The model averaged estimate of kitten survival was 0.406 (95% CI: 0.170 – 0.695).  The considerable uncertainty in this estimate reflects both high standard errors for estimates from individual models and very different estimates of survival from competitively ranked models.  There was strong support for kitten survival being lowest in the first three months of life, but weak support for it varying by sex, season, or litter size.  Heterozygosity positively influenced kitten survival. Results regarding the influence of genetic introgression on kitten survival are presented, and their management implications discussed.  The apparently dramatic growth of this population since 1995 suggests that the models with higher estimates of kitten survival are more plausible, but more work is required to put these estimates into context.