Heat stress (HT) during the dry period affects hepatic gene expression and adipose tissue mobilization during the transition period. In addition, it is postulated that HT may alter insulin action on peripheral tissues. Our objective was to evaluate the effect of cooling heat-stressed cows during the dry period on insulin effects on peripheral tissues during the transition period. Cows were dried off 46d before expected calving and assigned to 1 of 2 treatments: HT (n = 16) or cooling (CL, n = 16). During the dry period, the average temperature-humidity index was 78, but CL cows were cooled with sprinklers and fans, whereas HT cows were not. After calving, all cows were housed and managed under the same conditions. Rectal temperatures were measured twice daily (0730 and 1430h) and respiration rate recorded 3 times weekly during the dry period. Dry matter intake was recorded daily from dry-off to 42d relative to calving (DRC). Body weight and body condition score were measured weekly from dry-off to 42 DRC. Milk yield and composition were recorded daily to 42wk postpartum. Glucose tolerance tests (GTT) and insulin challenges (IC) were performed at dry-off, −14, 7, and 28 DRC in a subset of cows (HT, n = 8; CL, n = 8). Relative to HT, CL cows had lower rectal temperatures (39.3 vs. 39.0°C) in the afternoon and respiration rate (69 vs. 48 breath/min). Cows from the cooling treatment tended to consume more feed than HT cows prepartum and postpartum. Compared with HT, CL cows gained more weight before calving but lost more weight and body condition in early lactation. Cows from the cooling treatment produced more milk than HT cows (34.0 vs. 27.7kg/d), but treatments did not affect milk composition. Treatments did not affect circulating insulin and metabolites prepartum, but CL cows had decreased glucose, increased nonesterified fatty acid, and tended to have lower insulin concentrations in plasma postpartum compared with HT cows. Cooling prepartum HT cows did not affect the insulin responses to GTT and IC during the transition period and glucose responses to GTT and IC at −14 and 28 DRC were not affected by treatments. At 7 DRC, CL cows tended to have slower glucose clearance to GTT and weaker glucose response to IC relative to HT cows. Cows from the cooling treatment had stronger nonesterified fatty acid responses to IC postpartum but not prepartum compared with HT. In conclusion, cooling heat-stressed dairy cows in the dry period reduced insulin effects on peripheral tissues in early lactation but not in the dry period.