The aim of this study was to characterize progressive dehydration, stress responses, and water consumption patterns of horses transported long distances in hot weather and to estimate recovery time after 30 h of transport. Thirty adult mares and geldings were deprived of access to feed and water for 6 h, blocked by age, sex, breed, and body condition score, and assigned to one of the following treatments: penned, offered water (Penned/Watered, n = 5); penned, no water (Penned, n = 5); transported, offered water (Transported/Watered, two groups of n = 5); or transported, no water (Transported, two groups of n = 5). None of the horses had access to feed while on treatment. A commercial, single-deck, open-top, 15.8-m-long trailer was divided into four compartments to accommodate the two Transported/Watered and two Transported groups at 1.77 m2 per horse. At 8, 17, 22, 27, 30, and 33 h after initiation of transport, the truck returned and stopped for 1 h to allow for data collection and to give the Transported/Watered and Penned/Watered horses 10 min of access to water in individual buckets. Treatments for the non-watered horses (Penned and Transported) were terminated after 30 h due to dehydration and fatigue, whereas the watered horses (Penned/Watered and Transported/Watered) could continue for another 2 h. Mean weight loss after 30 h was greater in the Penned (57.1 kg, 12.8%) and Transported (52.2 kg, 10.3%) groups than in the Transported/Watered (20.7 kg, 4.0%) and Penned/Watered (17 kg, 3.5%) groups (P < 0.0001). Respiration, heart rate, sodium, chloride, total protein, and osmolality were significantly elevated in the non-watered horses (P < 0.0001), and sodium, chloride, total protein, and osmolality greatly exceeded normal reference ranges, indicating severe dehydration. Although not statistically significant, the horses penned in full sun, with or without water, had a dehydration response that was slightly greater than that of the transported horses. Plasma cortisol concentrations had a significant time x treatment interaction (P < 0.0001), in which the Penned/Watered and Transported/Watered horses remained relatively consistent, whereas the Transported, and especially the Penned, horses' plasma cortisol concentrations greatly increased. Transporting healthy horses for more than 24 h during hot weather and without water will cause severe dehydration; transport for more than 28 h even with periodic access to water will likely be harmful due to increasing fatigue.