Quantitative information concerning pre-ejaculation sexual preparation of bulls was published some years ago by Collins, Bratton &Henderson (1951), who concluded that about 40 more motile sperm could be obtained by imposing one false mount before ejaculation. It had been suggested that one false mount in bulls is essentially as efficient as two such attempts in terms of sperm output (Branton, D'Arensbourg &Johnston, 1952). More recently Hafs, Knisely &Desjardins (1962) have shown that bulls given a planned 5 min sexual preparation yielded 30 more sperm than those given no preparation. Similarly 10 min preparation resulted in a 12 increase in sperm output when compared with a 5 min preparation. These authors also showed that zero, one, two or three false mounts significantly affected motile sperm output. Although it has been suggested by Foote (1969) that sexual preparation procedures are extremely important in males in order to maintain ejaculations of high quality at regular intervals, little data have been published concerning stimulation procedures in rams. Satisfactory conception rates can be obtained following double or single set-time inseminations in ewes (Smith, Boland &Gordon, 1978). Because of the requirement to use freshly collected semen for a.i. it is essential that reliable techniques be available for collection of high quality material. The purpose of the present experiment was to evaluate the effect of the physiological state of the teaser ewe on semen characteristics. In addition rams were subjected to a period of stimulation where they were in sight of other rams ejaculating into the artificial vagina. The basis of their ability to yield ejaculates of acceptable quality into the artificial vagina. Six ovariectomized Galway ewes were used as teasers. Oestrus was induced following a 30 mg Cronolone pessary for 8 days. Each ewe received 250 /ig stilboestrol at the time of pessary withdrawal. Semen was collected from December until July, a period which corresponds to the late breeding season and anoestrus. Alternate rams were used to collect using an oestrous or non-oestrous teaser ewe. Semen was collected using an appropriate artificial vagina. The volume of semen collected together with the sperm density was examined. The libido of the rams, measured in terms of ‘reaction-time’, was determined as the time taken by the ram to ejaculate after reaching the service crate. This was monitored using a stop watch for all collections. A total of 1133 samples were evaluated. In a second experiment 1012 ejaculates were used to study the effect of visual stimulation on semen characteristics. On alternate fortnights rams were allowed to observe the semen collection procedure when they were placed in a pen adjacent to the collecting crate. During the second period rams were penned out of sight of the collecting crate. Each ram was taken from the pen and semen collected from him in isolation, before he was returned to the pen. Semen was evaluated using the following criteria: volume, density and reaction time. Data were subjected to analysis of variance. © 1979, Cambridge University Press. All rights reserved.