The present paper presents a first order description of the precopulatory behavior of the golden jackal (Canis aureus syriacus) without using emotionally-toned words. The coarse structure of this behavior which extends over 4 months prior to copulation is described. During this time, jackals roam the fields in pairs, each pair in a particular area. From time to time they perform certain series of actions while circling each other in a specific manner. The term T-sequence is used for this behavior which might be defined as precopulatory. A detailed fine-grain description of selected actual sequences of four captive pairs and three wild pairs is presented. Common features and variation in performance, with special emphasis on the latter, are summarized. Precopulatory behavior of jackals might be viewed as a complex of three groups of events superimposed on one another: the group of events which constitute the T-sequences (such as circling, standing perpendicularly to the mate, putting the head on the mate's back, etc.) is superimposed on a group of patterns which are performed during the first phase of precopulatory behavior - sniffing the ground, urination and scraping. Later, a third group of patterns is superimposed on the first two groups: this group includes the male licking the female's vulva, the female mounting the male and other patterns. At this stage, the frequency of appearance of patterns of the first two groups decreases gradually; the male starts to mount the female, and a few days later copulations follow. Variation in performance was found in the following acts: 1) In certain pairs the male always forms the top of the T during T-sequences, in others the female, and in still other pairs both mates take turns at forming either the top or the base of the T. Relative body position during T-sequences is in correlation with the distribution of events between the mates. It also affects the time of onset of the week of copulations. 2) Some agonistic patterns were observed during T-sequences only in captivity. 3) Certain behavior patterns are performed by some pairs but not by others. 4) Some behavior patterns are performed in different styles by different individuals. The performance of T-sequences is related to the following factors: a) sequential relation of events to copulation, b) time of day, c) change in environment, d) howling, e) duration of pair bond, f) individual differences. Motivational models as well as quantitative analysis were avoided, since the variability of the material calls for the development of special methods for fine-grain analysis. This material deserves phenomenological treatment, in the sense that regularity should be looked for first in the structure of concrete specific behavior sequences, and only then in the behavior in general.