ABSTRACT. Jugular vein blood samples were collected from 23 young and sexual mature feral stallions to examine the relationship between plasma testosterone concentration and age, breeding season or harem size. Testosterone concentration increased with the age of the stallions until they formed their own harems, at about 4 to 6 years old. Seasonal variations in testosterone concentrations were observed, and found to be significantly higher (P<0.001) throughout the breeding season than non-breeding season, from 3 years of age. Testosterone levels were correlated with harem size for individual stallions. It can be inferred from these results that there is a relationship between plasma testosterone concentration and age, breeding season and harem size. -KEY WORDS: equine (wild), harem, testosterone.J. Vet. Med. Sci. 60(5): 643-645, 1998 Fresh water ponds are plentiful in the range. The grasslands are restricted to the upper area (about 40 ha) half way up the two hills (about 290 m above sea level) which are located in the center of the range and covered with various kinds of native grasses. Other areas of the range are mostly occupied by artificial forests with Japanese cedars and a variety of copses [8]. Generally, the warm climate and abundant rain provide a lot of vegetation in summer. In the first experiment, twenty-three feral male horses, ranging from 1 to 6 years of age, were used to determine the relationship between testosterone levels and age, or breeding season. All animals were in good physical conditions during the experiment. Individual horses were readily identified by the number branded on the hip. Almost Misaki horses were individually recorded the birth date and parents (identified by DNA analysis). Blood samples were collected four times every two weeks from 23 individual horses in May and June of 1994, and mean values of testosterone for each were obtained. These data was used to relate testosterone concentrations to age for the breeding season. Similarly blood samples were collected four times every two weeks in October and November of 1994, to provide data for the non-breeding season.In the second experiment, we randomly selected 7 males at different ages (4-9 years old) with or without a harem, and examined the change of harem size and testosterone levels during two Breeding seasons from 1993 to 1995, in order to investigate their relationship. Blood samples were collected from the 7 individual males in May of both 1994 and 1995 and harem formation was observed. The horses were extremely dangerous to handle and were restrained with Xylazine as recommended by Dodman [5]. A dose of 4 mg/kg body weight was injected using a special syringe, fired from a blowgun at a few meters from the horse (hind quarter), for accuracy. Some of the horses never adjusted to the restraint procedures and so we tried again on other days. Blood samples were taken by jugular vene puncture into heparinised tubes and placed on ice, plasma was obtained by centrifugation and was stored at 20°C until Testosterone is the ...