The aims of this study were to describe the most suitable views for identification of aortic valve prolapse (AVP) and to report how repeatable this finding was in the horse. Eight healthy Thoroughbred and Irish Draught cross horses (mean±SEM:600±26.7Kg) were recruited. Echocardiography was performed daily for five days on each horse by two operators. Standard images of the aortic valve were obtained and valves were interrogated using continuous, pulsed wave and color-flow Doppler. Images were evaluated by one operator for the five separate examinations and by one operator for a single examination. Inter- and intra-operator repeatability were assessed. Only appropriate long-axis standard views of the aortic valve (AoV) were used for evaluation of the prevalence and repeatability of AVP as non-standard views can artificially create the appearance of valve prolapse. Data were compared using repeated measures ANOVA, Student's T, Friedman, and McNemar's tests as appropriate. Intra-class correlation coefficients (ICC's) were calculated. AVP was observed in 88% and physiological aortic regurgitation (PAR) in all horses on at least two days. Prolapse was observed in more than one cusp in 50% of horses and all three cusps in one horse. Prolapse was observed to affect the right coronary cusp most commonly (100% of horses with AVP), the non-coronary cusp the next most common (50%) and the left coronary cusp least commonly (17%). This was most reliably identified on a right parasternal long axis view of the left ventricular outflow tract and could only be seen on a right parasternal short axis view of the aortic valve if AVP was severe. There were no differences in any of the measurements, time indices, prevalence of AVP or PAR or dimensions of the regurgitant jet identified using color-flow Doppler over five days in the same horse except for free wall diameter in systole (p=0.018) and diastole (p=0.019). There were no differences between measurements performed on images obtained by the two different operators and agreement regarding AVP and PAR between echocardiograms performed by different operators was between 79 and 100%. There were no differences between measurements performed on different occasions or between operators. Intra-class correlation coefficients showed good (>0.85) agreement between echocardiograms obtained by different operators on the same day except for pre-ejection period (0.504) and inter-ventricular septum dimensions (0.304). There was extremely good (> 0.85) correlation between measures taken from the same horses on different days except for free wall diameter in systole and diastole (0.537), AoV thickness (0.666) and PAR jet width (0.390). In conclusion this is a repeatable reliable technique for identification of AVP and PAR in the horse.