Through the comparison of the performance of a group of glandectomized rats with that of a group of sham-operated rats on a double-alternation schedule of reward-nonreward (R-N), the role ofthe harderian gland in the production and/or release of Rand N odor cues was evaluated. The establishment of appropriate patterned responding by both harderianectomized and shamoperated animals argues against this gland's being involved in the odor control of maze learning. Additional potential odor sources are considered.Subsequent to the proposal (Ludvigson & Sytsma, 1967) and establishment (e.g., Davis, Prytula, Noble, & Mollenhour, 1975; Davis, Whiteside, Bramlett, & Petersen, 1981 ; Mellgren, Fouts, & Martin, 1973; Prytula & Davis, 1974, 1976Seago, Ludvigson, & Remley, 1970) of the premise that rats exude characteristic odors on reward (R) and nonreward (N) occasions, several attempts have been made to isolate the anatomical source of such odor production. For example, McNeese and Ludvigson (1975) reported that the discriminable odor cue necessary for the development of appropriate double-alternation (DA) patterning was not produced/released by the preputial gland or the androgen-dependent accessory glands. Moreover, McNeese and Ludvigson (1975) indicated that such odors did not appear to be linked to urine production. More recently, Weaver, Whiteside, Janzen, Moore, and Davis (1982) reported a study in which the potential role of the footpad sweat gland in the production of Rand N odors was evaluated. By spraying the feet oftest subjects with aerosal plastic, this gland was functionally inactivated. The results indicated that precluding odors emanating from the feet significantly accentuated R versus N DA patterning, rather than decreasing it as would be predicted if a source of R or N odors had been eliminated.On the other hand, Thiessen (1984) reported that lipids and pigments that are released from the harderian gland are involved in the determination and/or elicitation of social behavior in the Mongolian gerbil (Meriones unguiculatus). More directly to the question at hand, it has been shown that harderian material may serve as an attractant pheromone (Thiessen, 1984). The relatively large size of the harderian gland and the fact that it secretes a substantial amount of material suggest that it might be a possible location for the production and/or release of odor cues controlling maze performances, as described above. The present experiment was designed to investigate this possibility. Utilizing the technique of running one large Requests for reprints should be directed to S. F. Davis, Department of Psychology, Emporia State University , Emporia, KS 66801. squad of animals composed of separate groups (see PrytoIa, Davis, & Fanning, 1981) leads to several predictions. First, if a group of harderianectomized animals preceded a group of sham-operated controls, minimal, if any, odorbased patterning would be expected on the part of the first several animals in the sham-operated group; this would be due to a disrupt...
