Records were obtained over a 3 year period from six Holstein dairy farms of 300 to 500 cows each in the Phoenix, Ariz. area. Dairies were selected on the basis of similar management practices, herd size, milk production and facilities (with the exception of cooling systems). Microclimatic modifications (two dairies each) were shade only (approximately 3.7 m2/cow), evaporative-cooled shades and low-pressure water foggers under the shades. Data were categorized by season of calving (spring, Feb.-May; summer, June-Sept.; and fall, Oct.-Jan.). Traits evaluated were calving interval, days open and services/conception. Calving interval was shortest for cows calving in the spring (378 days), intermediate in fall (382 days) and longest in summer (396 days). Similar seasonal trends were observed for days open (103, 103 and 119 days, respectively) and services/conception (1.54, 1.81 and 1.93, respectively). All differences between spring and summer were significant (P less than 0.05). Calving interval and days open were less for evaporative-cooled groups (374 and 98 days, respectively), with no difference between shade only and foggers (391 and 392 days, 112 and 116 days, respectively). Services/conception were similar for all groups (1.72 to 1.79). A significant interaction between microclimate and season for services/conception could be interpreted as (i) smaller season differences for evaporative-cooled groups than for shade or foggers, or (ii) a change in the ranking of control and fogger groups during summer versus fall. Evaporative cooling was more effective than fogging for reducing the detrimental effects of seasonal high temperatures on fertility.