nique devised by NAGASE et al. (1964) for freezing bull semen in a pelleted form - which is meanwhile proved by NAGASE et al. (1966) to be suitable for stallion's semen too - we found this method working with horse semen as well. The semen was collected in fractions as outlined by BADER & HUTTENRAUCH (1966) and the results obtained were very encouraging (MERKT & KRAUSE, 1966; KRAUSE & GROVE, 1967). In 1967 we continued in this direction. Thirtysix ejaculates were collected from various stal- lions and the semen preserved by deep-freezing using the pelleting technique. After thawing these pellets in steril-milk at +40 Centi- grade approximately 50 % (15 - 70 %) of the spermatozoa resumed full vigorous motility. There were marked diffe- rences in freezing ability of the semen of the various. males used and also in between ejaculates of the same semen donor. The same applied to thawed semen stored at 5 Centigrade, which showed different degrees of viabili- ty. Twentyone mares were inseminated; ten conceived, nine did not conceive, the result in two mares remains still unknown. Eight mares were settled during first estrous, one during the third and one during the fifth. Pellets were stored at -196 Centigrade in liquid nitrogen over periods ranging from three weeks to fourteen month. Pellets were thawed in 20-ml sterile-milk heated to +40 Centigrade. The number of cells inseminated was ad justed to 100 million cells with a good forward motility. There was one pregnancy achieved with about 60 million motile cells. The problem of inseminating mares at the correct time is a major difficulty. Inseminations were only carried out after rectal follicle control - up to three times daily - in order to deposite the semen as nearly as possible to the moment of ovulation. MERKT, WEITZE & LORRMANN (1967) were able to prove that bull semen pellets stored in liquid nitrogen are liable to loos spermatozoa. It was also evident that these sper- matozoa can attach themselves to other pellets being in the same container. This creates a problem for there seems to be a possibility now that semen of different bulls can be mixed. This in mind it seems to be of some importance to store the semen of different stallions in different containers. Furthermore, to ensure the identity of the foal every in- semination should obligatory be linked with blood group 1582