Territorial behavior of the Uganda kob (Adenota kob thomasi; Reduncini, Hippotraginae) was studied in the Toro Game Reserve, western Uganda. Two types of territories were found: (a) small individual plots, 15-30 m in diameter, aggregated in tight clusters that are called leks, arenas, or territorial breeding grounds (TGs) ; (b) larger territories of 100-200 m diameter, distributed between the arenas, called single territories (STs). The largely permanent TGs, to which most of the breeding is confined, provide the basis for a social organization of the kob population of ca. 15 000. A certain number of kob are, by tradition, attached to a particular TG, so that the total population is subdivided into units, each associated with one TG. STs are spread out between the TGs; their size, number, and distribution vary with season and local conditions. There is an irregular gradient in size and density of territories from the center of a TG through the STs in its vicinity. STs may be aggregated in loose clusters used as temporary or seasonal TGs. Permanent TGs may arise from such clusters. Abandonment of existing and formation of new TGs are relatively rare. The males on the STs are strongly attached to confined areas which they defend against intruding males. Competition for STs is not intense, but males are occasionally replaced. Males defeated from their STs join a male herd and may attempt later to reoccupy the same ST, often successfully. Whistling probably serves for marking the territory or for attracting females to it. Herds of females often pass through or stay on STs, but the males do not possess harems. They court the female and attempt to copulate with them, but most females avoid their approaches. Few copulations occur on the STs; in several cases the females involved proved to be physiologically abnormal, and it is concluded that the males on STs do not contribute significantly to the reproduction of the population. The daily activity of males on the STs is compared with that of males on TGs. The latter spend less time for feeding and have less food available on their territories; this, combined with the higher proportions of fighting and sexual behavior on TGs, is propably the main reason for the much higher rate of interchange of males on TGs compared with STs. Also, the degree of competition for territories is higher on TGs than STS. The males of the kob population studied are, on the whole, divided into two categories: Those frequenting TGs, and those staying on STs. Both types join a male herd when they are not territorial. The age distribution among males on TGs and those on STs is largely equal. Some males occupied both territories on a TG and STs, but such cases are relatively rare. Two young-adult males first occupied a ST for some time, before they appeared on a TG, but this course of behavioral development does not seem to be the general rule. Territorial behavior was found in several other kob populations; the relative number of STs and the development of TGs vary considerably between different areas. Territoriality and lek behavior in other ungulates are briefly reviewed. The Uganda kob is the only antelope known, so far, to exhibit typical lek behavior. In addition, behavioral polymorphism such as the occurrence of different types of territories within the same population has not yet been found in any other species of antelopes. The following conclusions pertaining to the Uganda kob are drawn: STs are the original form of territoriality, still prevalent in small or marginal populations. In large and dense populations the formation of TGs offers certain ecologic advantages, such as providing a social organization and a spacing mechanism to the population and ensuring maximum efficiency of reproduction. Despite these advantages of TGs over STs the latter have not disappeared. Either they provide some social advantage, as yet unknown, or their persistence ensures adaptive plasticity of local populations and the species as a whole, to meet emergencies brought about by changes in the environment.