Four aspects of olfaction in host location by tsetse flies, Glossina spp., are discussed as follows: (1) host location and its mechanism, (2) factors affecting host location, (3) kairomones and host location, and (4) kairomones and host selection. Flight behavior in the various phases of host location (i.e., ranging, activation, orientation, and landing) in the absence and presence of olfactory cues is summarized. Movement toward an odor source is effected inter alia through optomotor-steered, upwind anemotaxis. It is still unclear how tsetse employ upwind anemotaxis to realize host location, considering the often highly variable wind direction. Olfactorily induced activation is governed by the olfactory cue perceived and threshold levels set by the internal state of the fly. The former depends on the odor source and distance from it; the latter is determined by species, sex, and physiological state. Wind direction and speed, as well as vegetation and the mobility of the host, interfere with successful completion of odor-induced host location. Close-range olfactory cues (including composition and concentration gradients), visual cues, and nutritional state determine whether a fly, having reached the host animal, will land on it. Carbon dioxide is important in host location because it induces landing and long-range attraction. The role of the other kairomones (acetone, 1-octen-3-ol, 4-methyl-phenol, and 3-n-propyl-phenol) is less clear. Apart from the complacency of various host species under tsetse attack, host choice by tsetse is predominantly opportunistic and primarily the result of the frequency of successful tsetse–host encounters. Nevertheless, host selection based on olfactory cues cannot be ruled out.