Reestablishment of testicular normal temperature after testicular heat stress is unknown and its effect varies widely. The aim of this study was to investigate the impact of scrotal insulation (IN) on testicular temperature and its relation to semen quality and testosterone blood serum concentration. For this, 33 rams were used: 17 submitted to IN for 72 h (using bags involving the testes) and 16 not submitted to IN (control group). The experiment was performed between August and December, 2013 in Pirassununga, Brazil (21°56”13” South/47°28’24” West). Seminal characteristics, testosterone blood serum concentration, rectal temperature (RT), respiratory frequency (RF), scrotal superficies mean temperature (SSMT) and eye area mean temperature (EAMT) were analyzed 7 days before IN and 21, 35, 49, 63 and 90 days afterwards. SSMT and EAMT were measured by thermography camera FLIR T620®. Testosterone was evaluated by radioimmunoassay. Analysis of variance was used to determine the main effects of treatment, time and treatment-by-time interaction using PROC MIXED of SAS software adding command REPEAT. Pearson correlation test was used to verify correlation between SSMT, EAMT, RT and RF. Significant difference was considered when P≤0.05. At the end of IN, SSMT was higher in insulated group (32.26±0.19oCa) than in control group (30.58±0.18oCb), and the difference between rectal and testicular (deduced from SSMT) temperatures was 1.12oC; in the other times of the evaluation this difference was between 2.91 and 4.25oC in IN group. SSMT was reestablished 24 h after IN. RT and EAMT presented correlation (r=0.59; P<0.0001). There was time-by-treatment interaction for total sperm (P=0.0038) and progressive motility (P=0.01), abnormal spermatozoa (P<0.0001), membranes integrity (P<0.0001), induced TBARS (P=0.05) and DNA integrity (P=0.0004). These semen characteristics were negatively affected 21 days after IN and, excluding induced TBARS and abnormalities, recovered 35 days afterwards; induced TBARS just was affected after 49 days of IN; sperm abnormalities just recovered after 63 days. Testosterone blood serum concentration was lesser in insulated rams (P=0.03). Thus, the difference of 1.12oC between RT and testicular temperature impacts semen quality and testosterone blood serum concentration. Moreover, this study shows that rams can recover testes temperature efficiently towards IN and that infrared thermography is an efficient tool to identify differences on SSMT.