The animals used were both males and females of an outbred strain of rats developed as a cross of three strains: inbred Fischer albinos from Charles River Laboratories; highly inbred DA agoutis maintained in our lab; and outbred S2 black rats obtained from the Netherlands in 1980. Both males and females of this strain are capable of both territorial and competitive fighting when tested under appropriate conditions, but under conditions used in the present experiment, tested in a neutral arena without food deprivation, they normally do not show any offense behavior.
With the test animal under chloropent anesthesia (3.5 cc/ kg), guide cannulas were implanted by stereotaxic procedure in the hypothalamus. The guide cannulas consisted of 15 mm lengths of 23 gauge stainless steel tubing implanted to a depth 7 mm below the skull, approximately 3 mm above the intended target. In some of the animals, two guide cannulas were implanted symmetrically on each side of the skull, at 1.5 mm lateral of the sagittal suture, while in most of the animals only one guide cannula was implanted. To reach more medial placements in some animals, the guides were implanted at a 5° angle; this enabled us to avoid puncturing the sagittal sinus. To secure the guide cannulas in place, three stainless steel screws were implanted in the skull, and both the screws and the guides were embedded in dental cement.
Tests were conducted 2 or more days after surgery in a neutral test arena .consisting of a box 70 cm X 40 cm X 40 cm with a Plexiglas front and wood chips on the floor. A same-sexed opponent of approximately the same weight was introduced at least 5 min before testing and was observed to ensure that there was no spontaneous offense or sexual interactions.
An injection cannula, ranging in length from 17 mm to 18 mm, was inserted into the 15 mm guide cannula, giving a projection of 2 mm to 3 mm ventral to the guide cannula. The injection cannulas had been prepared from 30 gauge dental needles and were connected to a 1 microliter syringe by several feet of coiled PE plastic tubing. Then 0.5 microliter of picrotoxin (1 mg/ml in 0.9% saline) was injected over the course of 1 min. The injection was verified by measuring the movement of a small air bubble trapped in the line. One minute after injecting, the injection cannula was removed. In some cases, 0.9% saline was injected by itself as a control procedure.
The behaviors of the two animals, and especially their social interactions, were monitored for at least 35 min after each injection. The following behaviors were noted and scored for frequency per min: locomotion (transits from one end of cage to other); circling (clockwise or counterclockwise); self-grooming of the face, flank, or ano-genital region; pushing wood chips forward with the forepaws; digging in the wood chips and pushing them backwards with the hindpaws; leaping; standing upright and scanning; sniffing; and eating. The following social behaviors were noted for frequency: bite-and-kick attack; offensive sideways posture; upright posture and boxing; crawl-under; crawl-over; groom other; sniff ano-genital region; mounting with thrusting; on-top posture (formerly called full aggressive posture); and chasing.
On several tests when the injection produced mounting, the opponent was replaced and alternated every 2 min with an opponent of the opposite sex, in order to observe if there was any change in behavior as a function of the opponent's sex. There was no opportunity to test offense against partners of the opposite sex.
Sixty rats (34 males and 26 females) received injections at only one site in the hypothalamus. Thirteen others (all male) received injections on both sides of the hypothalamus through two implanted cannulas. In most of these 13 cases, injections were made at two locations on each side by using first a shorter and then a longer injection needle. These multiple injections were spaced at least 45 to 60 min apart, which gave time for overt behavioral effects of the injection to dissipate. In fact, we found little repetition of the evoked behaviors from such multiple injections (no repetition of offense and only two cases each of mounting and persistent locomotion).
Twenty-six other rats (all males) received injections of picrotoxin at single sites in the ventromedial tegmentum.
Following all testing, the animal was anesthetized again and perfused through the ascending aorta with 0.9% saline and then 10% formalin. The brain was extracted and sliced into 100 micron sections on a Vibratome to determine the location of the cannula track and injection sites. The injection site could be located as the deepest point of the cannula track. In most cases, the site could be confirmed by means of a small area of orange-red stain apparently caused by the picrotoxin. Data were discarded if we could not find the exact location of the track on the brain sections, or if the cannula tip penetrated into the third ventricle or below the brain. Locations of the tracks were plotted for males on drawings from the atlas of Paxinos and Watson (22) and for females on drawings from the atlas of Konig and Klippel (12). For purposes of illustration all female sites were transcribed onto the anatomically equivalent planes in the Paxinos and Watson atlas.