Operant Analysis of Behavior Associated With Oral Self-Administration Of Drug Solutions in Rats

Since many drugs which are abused by man are taken ora11y. it was desirable tc develop a method s11itable to quantitatively and reli ably I measure oral self-administration of drugs and their effects on behavior in experimental animals. Water deprived rats were trained to 1 ick for drug solutions (4.34 ul/lick) and bar press for food on a Fixed Interval (FI) 60 second schedule and press another bar for secondary reinforcements on a Fixed Ratio (FR )-5 schedule as three concurrent operants. Non-discriminated responding was inconsequential. An appropr~ate drug solution was substituted for water or the drugs were injected intraperitoneally before the session. for each drug, a dose-response was determined with usually six replicate sessions per dose for each rat. Three rats were usually used in each study. Substitution of solutions of amphetamine (0.5, 0.99 and 1.99 m Molar) resulted in concentration dependent decreases in discriminated and non-discriminated licking , and discriminated lever pressing for secondary reinforcement. Non-discriminated lever pressing for secondary reinforcement or food pellets increased. Consequential lever ~ress ing for food pellets was unaffected. The effects of amphetamine on lever pressing and licking were similar whether an acute injection was made before the session or amphetamine was self-ingested during the session. Chronic injections of amphetamine (5 mg/kg, I.P., 18 hours before the water session, given daily for 5 days prior to the study_ and during the study for 30 days) resulted in an increased sensitivity to ingested amphetamine . This increased sensitivity was manifested by a shift of

Chlorpromazine (0.5 mg/kg, I.P., 30 minutes before the session) significantly increased the lickin9 rates for solutions of amphetamine (0.5 mM or 1.0 mM).
Substitution of solutions of ethanol (10, 20, 40, 80% v/v) resulted in concentration dependent decreased in discriminated and nonqiscriminated licking, and discriminated lever pressing for secondary reinforcement. Non-discriminated lever pressing for secondary reinforcement or food pellets was increased. Consequential lever pressing for food pellets was unaffected. While the licking rate decreased with increased concentrations of ethanol, the grams of absolute ethanol ingested increased. The effects of oral injections of ethanol (12 ml/ kg, of a 50% v/y solution, 15 minutes before the session on behavior were similar to the effects of ingested ethanol except for a decrease . in number of food pellets obtained. Disulfiram (50 mg/kg, I.P., 60 minutes before the ethanol session) did not affect behaviors for various contingencies during water sessions or initial portions of ethanol (20% v/v) sessions. However, disulfiram pretreatment depressed behaviors completely after the inital ingestion of small quar.tities of ethanol.
Rats were given increasing doses of morphine sulfate, until a total daily dose of 200 mg/kg was attained.
Oeprivation of these rats of their daily morphine for four successive days had little effect on water licking, but licking rates for a solution of amphetamine (0.5 m Molar) decreased dur1ng the abstinence while licking rates for a solution of ethanol (80% v/v) remained con-. -ii-sistently higher than thos.e for amphetamine. Nal orphine (_4 mg/k~ abol-, ished licking for water in three of four rats. indi.cating a difference between nalorphine ind~ced and abstinence induced withdrawal .

ACKNOWLEDGEM ENTS
The author wishes to express his gratitude to Dr. Harbans Lal, his major professor, for his guidance and endless patience tl1roughout the study and in the preparation of this thesis. Special thanks are also conveyed to Dr. Harbans Lal for his part in the shaping of the author's attitude and ability which are essential for a career in research.
The author wishes to express his gratitude to the members of his graduate committee for their guidance in the preparation of this thesis.
The author wishes to express his gratitude to Anne Partenhei mer and Pamela Brink for the typing of this thesis.
The author a 1 so \'fishes to convey his thanks to his fe 11 ow students for their suggestions during these studies and to all the others who have been so helpful in all phases of the preparation of this thesis.
The author' is grateful to the University of Rhode Island for · providing financial assistance during these studies. -iv-

DEDICATION
The author would li ke to dedicate this thesis to his wife, Regina, I without whose constant encouragement, confidence and prayers this thesis could not have been possible. -v~

I. INTRODUCTION
Self-administration of drugs by man is not a unique phenomenon but dates back before recorded history (Le\<1in, 1964j. To understand the basic aspects of drug self-administration, it is essential that it be studied in an experimental situation using experimental animals to e1i minate the complication of psychological and ' personality factors which complicate studies with humans. Since Weeks (1962) described a technique for intravenous selfadministration of morphine in the rat, a considerable amount of research has been devoted to the study of intravenous self-administration of opiates (Thompson and Schuster, 1964), stimulants (Pickens, 1968), and ethano 1 (Deneau et ~, 1969).
Since many drugs which are abused by man are taken orally, it was desirable to develop a method suitable to quantitatively and reliably measure oral self-administration of drugs in experimental animals, in order to extrapolate the results to the human situation. Utilizing the observation of Falk (1961) that rats trained to obtain food pellets on a variable interval 90 second schedule, will consume three to five ·times their normal water intake in a session when water was freely available, Lester (l96J) reported intakes of 5.6% ethanol by rats to the point of intoxication.
The µ·resent investigation utilized a modification of Falks (1961) schedule induced polydipsia with operant licking chosen as the means by which to study oral self-administration of drugs, rather than · using lever pressing for dipper presentation of fluid. Operant licking was chosen because it was a ' novel approach and never utilized in this way, secondly, ingestion by licking is the way in which a rat routinely ingests fluids.
The · present investigation sought evidence to deter~ine: 14 1. Whether fluid acquired by operant licking would provide reliable, quantitative data with which to study oral self-administration of drugs by rats.
2. Whether orally ingested drugs exerted similar effects on behavior as the same drugs injected before the session.
3. Whether chronic treatment with drugs would affect subsequent selfadministration of the same drug.
4. Whether chlorpromazine pretreatment would have any effect on amphetamine self-ingestion. 5. Whether this system could serve as a model with which to study the effects of disulfiram on ethanol consumption.
6. The effect of substitution of solutions of ethanol or amphetamine for water during abstinence induced withdrawal in morphine dependent rats.
7. The behavior of the morphine antagonist, nalorphine, on morphine dependent rats.

II. LITERATURE SURVEY
Self-administration of drugs by man is not 3 unique · pnenomenon but dates back before recorded history (Lewin, 1964). Drugs are used in t\-10 ways -first for their therapeutic effect and secondly in an abusive manner . Although the use of drugs by man precedes recorded history, the recognition of the problem of dependence (Opioid) is a comparatively recen t phenome non (Light and Torrence, 1929). About this same · period, it was discovered that biological factors involved in drug dependence could be studied in animals (Tatum~~. 1929). To place developments in the study of dependence and self-administration of drugs in closer perspec tive yet, two of the first review articles on the subject have been j ust recently published (Schuster and Villarreal, 1967;Schuster and Thompson, 1969).
In one of the earliest experiments which suggested that morphine self-administration might be brought under operant control, Spragg (1940), demons t rated that physically dependent chimpanzees, who had been deprived of morphine, would choose a box containing a morphine-filled syringe over a box containing food. The experimenter then injected the ani mal with morphine. If the animal had recently received an injection of morph i ne, it would choose the box containing food. In 1955, Headlee et~ presented a study designed to show operant conditioning of drug self-administration. They have also been the only investigators to use an intr aperitoneal route of self-administration. In their study, restrained rats were conditioned to turn their heads laterally and interrupti ng a beam of light falling on a photo-electric cell, thus starting a· pump which infused a morphine solution through c needle inserted through the body wall into the peritoneal cavity.
16 Weeks (1962) described a technique for intravenous self-administration of drugs in the rat. He demonstrated clearly that rats, rendered physiologically dependent by programmed intravenous administration of morphine, will then maintain the dependent state by pressing a lever to obtain the drug. The self-administration technique was then extended to physically dependent monkeys with similar results (Thompson and Schuster, 1964). Deneau et ~ (1969) went a step further and were able to demonstrate that monkeys v1hi ch were not phys i o 1 ogi ca lly dependent upon morphine would initiate and maintain self-administration of morphine. These results indicate that the monkeys developed a psychological dependence -a preference to exist under the influence of the drug -before physiological dependence could develop. That non-dependent monkeys will self-administer morphine has also been confirmed by Schuster (1970). Nichols et ~and Coppock (1956) conditioned rats to self-ingest morphine solutions following establishment of physical dependence. Kumar et ~ (1968) have shown it is possible to induce a preference for morphine in rats without making them physically dependent. The method involved makinq the rats accustomed to satisfying their normal thirst during a limited time daily, and then substituting morphine solutions for the water normally given.

Alcohol Self-Administration
The human disease state of 11 alcoholism 11 is so complex that even today it defies adequate description and consequently adequate therapeutic (1) Odor-Taste -The odor and taste of solutions influence the ingestion of alcohol by rats. For exam ple, when ability to taste was diminished by methylpentynol, rats were found to drink a solution of alcohol as high as 20% (Dicker, 1958). Olfactory phenomena also seem to have an affect, since rats made anos mic by extirpation of the olfactory bulbs selected higher concentrations of alcohol than · prior to surgery (Kohn and Stellar, 1960).
(2) Concentration -Rats will preferentially select a solution of alcohol over water if the concentration of alcohol is not excessive. For example, Myers (1968) reported that several strains of rats prefer solutions of alcohol over water only when concentration of the solutions are less than 7%.
(3) Caloric Content -Since solutions of alcohol are a ready source of calories, this factor must also be considered in interpreting 18 results of alcohol ingestion. Rats have been shown to reduce their food intake in proportion to the concentration of alcohol consumed (Richter, 1953). However, the aversiveness of concentrations greater than 7% was sufficient to offset the caloric value of alcohol during severe food deprivation (Myers and Carey, 1961).

(b) Animal Factors
Many factors concerning the experimental animals have also been described in studies of alcohol consumption.
(1) Aoe -Alcohol preference was reported to be greater in young rats (2-3 months of age) than in animals up to 2 years old (Parisella and Pritham, 1964). However, Goodrick ( 1967) _ repo.\ted that a 1coho 1 ingestion increased in rats 1-5 months old, and was greater at every concentration in rats 24 months old as compared to rats 15 months of age.
(2) Sex .-The question of whether male or female animals drink more alcohol has yielded results that are contradictory. Mardones (1960) and McClearn and Rodgers (1959) reported no sex differences in alcohol consumption by rats; Aschkenasy-Lelu (1960) and Eriksson and Malmstrom (1967) reported that female rats consume more alcohol than males; while Schaldewald et~ (1953), Clay (1964) and Powell et~ (1966) reported more alcohol consumed by male than by female rats. A wide individual variation in alcohol consumption within animals of the same strain, sex, age etc. (Mardones, 1960 andEriksson, 1969) has been demonstrated.
(3) Genetic Factors -By outbreeding Wistar rats which differed in their alcohol consumption, Eriksson (1968) has raised two genetically different lines. Marked differences between the sexes and strains were ' evident, with regard to alcohol consumption, by the eighth generation. Mardones and co-workers (1953) had previously shown that a clear correlation existed between alcohol consumption of purents and offspring. Forsander (1967) has also demonstrated a genetic factor in alcohol consumption.

(c) Miscellaneous Factors
In addition, there are significant effects of a variety of miscellaneous factors on alcohol consumption by laboratory animals.
(1) Ambient Temperature -Eriksson (1969) reported that rats maintained at an environmental temperature of s 0 c consumed more alcohol than similar animals in environments of 22°c or 32°c. This work contradicted the earlier report of Myers (1962) that rats consumed more ethanol at 18oc than at 27°c.
(2) Effect of Pretreatment -As mentioned before, rats of several strains will usually prefer solutions of alcohol over water if the concentrations are lower than 7-8% (Myers, 1968). By restricting the animals fluid intake exclusively to solutions of alcohol, ingestion of alcohol in concentrations up to 20% has been reported (Richter, 1953, Mardones, 1960. Other techniques such as injecting the animals with alcohol have also been investigated. Myers (1963) has de~onstrated a preference for alcohol in rats following repeated intracranial infusions of alcohol.
However, Kaz and Mende 1 son ( 1967) \'Jere unab 1 e to produce the same phenomena in monkeys.
(3) Stress -In an attempt to increase voluntary alcohol consumption by imposition of a stressful environment, Myers and Holman (1967) stressed rats by intense shock given randomly around the clock for 14 days. The 20 hypothesis that an increased intake of alcohol would provide relief from the stress was not supported ~s no increase occurred. Preference for alcohol has been reported when a cuP. as a warning light was associated with presentation of shock on a random basis (Cicero et ~. 1968 andPersensky, 1968). However, the preference was only observed during periods of stress and not afterward.
(4) Schedule -Induced Alcohol Consumption -Utilizing the observation of Falk (1961) that rats trained to obtain food pellets on variable intervals of 90 seconds, will consume 3-4 times their normal water intake in a session when water was freely available, efforts have been made to induce alcohol consumption. Using this technique, Lester (1961) reported intakes of 5.6% alcohol solution by rats to a point of intoxication. As with other attempts, this one also failed to create a preference of alcohol over water (Senter and Sinclair, 1967).
(5) Nutritional Factors With regard to alcohol consumption in self-selection experiments, it ~ust also be remembered that the choice is not really a simple one between alcohol and water, but rather a threeway choice in which the caloric value of alcohol can also substitute for food (Hausmann, 1932). This situation has been recently reiterated by Forsander (1967).

ToTerance
In general, forced consumption of alcohol solutions for long periods of time has not led to an increased preference for alcohol in rats (Richter, 1953;Mardones, 1960). In fact, Essig (1968) has reported thatrats show a preference for water after prolonged periods of alcohol consumption.
In a series of self-selection (preference) experiments, Arvola and Forsander (1963)  While tolerance to alcohol has yet to be shown in rats self-ingesting alcohol, it has been observed and measured by behavioral tasks if alcohol is administered intragastrically. Rats trained to escape foot shock have displayed tolerance after receiving alcohol intragastrically for a period of time (Moskowitz and Wapner, 1964;and Chen, 1968). Tolerance, as measured by the ability of rats to run on a motor driven belt suspended over an electrified grid, has been shown to develop maximally within three weeks following intoxicating administration of alcohol and to be significantly reduced after one week during which no alcohol was administered (LeBlanc, et~. 1969).
When investigating the consumption of alcohol by rats for a length of time, the possible development of tolerance, whether due to metabolic changes or to cellular changes in the central nervous system (Mendelson, 1970) is an area for much needed research. More work is needed in the area of self-ingestion as compared to involuntary administration before the mechanism of tolerance can be clearly defined.

22
One related area of particular interest is the metabolic fate of alcohol. While agreement exists that liver alcohol dehydrogenase is the major enzymatic pathway for the metabolic degradation of alcohol (Westerfeld, 1955) the exact relationship of the activity of this enzyme to the plasma haTf-life of alcohol remains unclear. For example, in the studies of Wilson et~ (1961), two strains of mice, the C57 Bl/6J and C3H/Agautie were shown to have widely differing levels of liver alcohol dehydrogenase activity. Despite this difference, the plasma half-life of alcohol in the two strains was virtually indistinguishable (Wilson, 1967). This correlates well with the studies of Asade and Galambos (1963) who were unable to correlate the rate of disappearance of alcohol from blood with liver alcohol dehydrogenase activity in humans.

Behavioral Studies
Research on the effects of alcohol on laboratory animals has the disadvantage that one cannot ask the subjects how they feel, but on the other hand there is much more control and flexibility in animal experimentation than in human experiments. In general, the use of animal behavioral techniques has provided some information on the effects of alcohol.
The earliest report of the effect of alcohol on conflict behavior was made by Masserman et~ (1944by Masserman et~ ( , 1945, and Masserman and Yum (1946).
Cats were first trained to obtain food and were later subjected to an air blast or electric shock at the food, resulting in avoidance behavior.
Alcohol, injected intraperitoneally, restored the food approach behavior.-Similar experiments have also been conducted using rats (Conger, 1951;Barry and Miller, 1962;Grossman and Miller, 1961;Freed, 1967 In an experi ment in which rats were induced to drink alcohol, si milar effects were observed (Freed~ 1968b).
Experiments have given evidence that alcohol reduces frustration as produced by extinction (Barry et~' 1962). Rats under the influence of alcohol (injected intraperitoneally) had faster running speeds during extinction trials (no food available) than saline treated rats.
A generalized depressant effect of alcohol was demonstrated in a lever pressing response for water reward maintained by DRL (Differential Reinforcement for Low Rate) schedule (Sidman, 1955;Laties and W eiss, 1962). The total number of lever presses decreased to less than half the normal amount.
Several studies report that doses of alcohol causing ataxia are required before there is any impairment in avoidance performance in rats trained to avoid shock either in a shuttle box or by jumping up on a pole (Walgren and Savolainen, 1962;Chittal and Sheth, 1963;Broadhurst and Walgren, 1964). Deneau et ~ (1969) demonstrated that monkeys which were not physiologically dependent upon alcohol would initiate and maintain self-injection of alcohol as was also demonstrated for morphine.

Stimulant Self-Administration
Initial studies of drug self-administration by animals were concerned with either alcohol or morphine since it was thought that only drugs which · produce a physiological dependence in man would be self-administered by animals. Yet many stimulants suc h as cocaine and amphetamine do not ,; ·; i . . l 24 induce physiological dependence but are widely abused by man. (Durrant, 1965;Connell, 1968).
The self-injection of sti mulant drugs by mo~keys was first reported by Deneau et~ (1964). Since then, Pickens andThompson (1968), Pickens (1968), and Woods and Schuster (1968) have shown cocaine to be selfadministered intravenously by both rats and monkeys. Methamphetamine has been found to be self-administered intravenously by rats (Pickens, 1968;Pickens, Meisch and McGuire, 1967). Amphetamine was shown to be selfadministered intravenously by monkeys (Deneau et ~' 1969) and rats (Pickens, 1968;Pickens and Harris, 1968). Nicotine has been shown to be self-administered intravenously (Deneau and Inoki, 1967) and inhaled as tobacco smoke by monkeys (Jarvik, 1967). A comparison between stimulant self-administration and opiate self-administration has recently been published by Thompson and Pickens (1970). Stimulant self-administration differs from opiate self-administration in at least four respects: acquisition of self-administration is rapid for stimulants \'Jhile very gradual for opiates. The intervals between infusions is extremely stable for stimulants but variable for opiates. Self-administration of stimulants is cyclic with alternating intake and abstinence patterns while no such pattern is observed with opiates. Finally, a long burst of responses occurs at a very high rate during extinction of stimulant self-administration while fo opiate extinction, responding persists at a low rate for weeks and even months after discontinuing reinforcement.

Minor Tranquilizer Self-Administration
Recently  have applied operant techniques to study 25 oral self-administration of another class of drugs, the minor tranquilizers, in rats. In their study, drugs were used as secondary reinforcers.
Rats were initially tested for preference between water and chlordiazepoxide, meprobamate, LSD, nicotinic acid and quinine. In all instances, the rats · preferred water. The rats were then trained to lick the drug bottle in order to obtain food reinforcement. Subsequent testing revealed that following the association of ingestion of drug with food reinforcement, the animals drank s.ignificant quantities of chlordiazepoxide, meprobamate and nicotinic acid. The effect was not obtained with LSD or quinine.

Materials
Subjects -Male albino-rats of Sprague-Dawley strain, random bred, weighing between 250-300 grams at the start of the experiment, were obtained from Charles River Breeding Farms, W ilmington, M assachusetts.
They were housed individually in an air-conditioned, light-controlled room (lights on 12 hours -off 12 hours). Food was available ad libitum, but water was available only for 30 minutes in the home cage, 2-4 hours after the drug session.
Apparatus -Two animal test cages (LHV Model 1417C) housed in sound attenuated chambers (LHV Model 1417C) were used to train and test the rats. In each cage were two levers, a food magazine and a licking operandum ( Figure 1) for presentation of fluid. Each drop of fluid delivered was 4.35 ul (Table 1) (Table 1) of fluid (primary reinforcement). Further responses on the left lever during this 20 second period were inconsequential. Every five consecutive 20 second cue-light periods were followed by 130 seconds of programmed stimulus not contingent upon lever pressing.
Each lick during this period also provided 4.35 ul of fluid. Left lever · presses during this 130 second period were inconsequential. Responding on the right lever on a FI-60 seconds schedule produced 45 mg Noyes food pellets.
Shapin g -The shaping was accomplished in the following steps: (a) Naive, water deprived rats were placed into the training cage and allowed to lick a drop of water which was hanging from the licking spout. Acquisition to licking for water usually occurred within ten minutes ( Figure 3).
(b) Once rats were licking consistantly for ten minutes or more, licking time was programmed to be contingent on left lever pressing (FR-1). This was termed consequential licking time and was signaled by a cue-light above the left lever. The ratio on the Jeft lever was gradually increased to FR-5 in the same session. Once this behavior was established, the consequential licking time, following five responses on the left lever, was set at 20 seconds and termed self-produced stimulus segment.
(c) A programmed stimulus segment, in which 130 second~ of continuous cue (house light) was programmed to occur following the end of the fifth (20 second) self-produced stimulus segment, was then (d) Rats were shaped to press the right lever on a FI-60 second schedule for 45 mg. Noyes Pellets by the method of successive approximations.
Recording -Responding during each session was recorded on a Harvard Cumulative Recorder (Gerbrands Model C-3). A typical cumulative record of the components of the session is given in Figure 4. During the self-· produced stimulus segment (20 seconds of cue-light following five responses on the left lever), the fifth left lever press was recorded as a downward deflection of the stepper pen which remained deflected for the 20 second duration. A left lever press during the 20 second period was inconsequential and recorded as an upward deflection of the stepper pen. Licking during the self:produced stimulus segment activated the stepper. Licking which occurred between any of the five consecutive self-produced stimulus segments was non-discriminated and non-consequential but also activated the stepper pen. Discriminated licking is distinguished from nondiscriminated licking on the cumulative record in that the stepper pen is not deflected downward before the non-discriminated licks. To differentiate the self-produced stimulus segments from the programmed sti mulus segments on the cumulative record, the stepper was programmed to reset at the end of the fifth consecutive 20 second self-produced stimulus segment. Licking during the 130 second programmed stimulus segments also activated the stepper. Left lever presses during programm~d stimulus were inconsequential and recorded as a downward deflection of the stepper pen. To further differentiate between the two stimulus segments on the cumulative record, the event pen was in the downward position ' for the five consecutive self-produced stimulus segments and in the upward position for the programmed stimulus segment. All right lever presses (as no attempt was made to distinguish between consequential and inconsequential right lever presses on the cumulative record) were recorded as deflections of the event pen, upward during seif-produced, and downward during programmed stimulus segments.
In addition to the cumulative records, all responses (discriminated and nondiscriminated) were recorded on digital counters. The cumulative duration of each stimulus segment, self-produced and programmed, was recorded on elapsed time meters. All recording equ1pm~nt with the exception of the cumulative recorders was programmed to record for only the specified length of the session to insure quantitative recording of the data. The cumulative records were measured (30 minutes/6 inches) from the start of the session to give the exact cumulative record corresponding to the session. The students "t" test was also used to test for differences between control and experimental groups (Dixon and Massey, 1969). The level of significance was determined by comparison of "t" values with values from standard tables: Design Predrug Session -The first of the two daily sessions was designated as predrug session. It served two purposes: first, to determine if the rats were behaviorally fit for the session and, secondly, it served as a control for the drug session which followed 4-6 hours later . . . . . ,_ Drug Session -The second daily session, following the predrug session was designated as the drug session. An appropriate drug solution was substituted for water for the rats to ingest during this session, or 36 the drugs were injected intraperitoneally before the session and water was made available for licking in the training cage. The drug session ' was not run on the day a subject did not perform favorably during the · predrug session.
Drug Studies -For each drug, a dose-response was determined with usually six replicate sessions per dose for each rat. Three rats were usually used in each study. The experiments performed are listed according to the drug involved.
Amphetamine -The effect of amphetamine were studied according to the experiments as follows: Amphetamine Licking -Two Operant Schedules Three rats, Z-29, Z-30, Z-33, were trained on two operant scheduleslicking and left lever pressing for secondary reinforcement. The right lever was present but not programmed to deliver food pellets.
These rats were initially trained to lick for water then log concentrations of amphetamine (0.50, 0.99, 1.98 mM) were randomly substituted for water in the second daily sessions. Each of the concentrations was presented until approximately six replicates at each concentration were obtained.

Effect of Chronic Amphetamine Injection on Amphetamine Licking
The same three rats (Z-29, Z-30, Z-33) as used in the amphetamine licking under two operant schedules, were injected daily . for 40 days with a 5 mg/kg dose of amphetamine intraperitoneally four hours after the second daily session. The following log concentrations of 37 amphetamine (0. 125, 0.25, 0.50 mM) were randomly substituted for water in the second daily sessions during this ti m e. Each of the concentrations was presented until approximately six replicates at each concentration were obtained. At the end of this series, hexobarbital (130 mg/kg I.P.) sleeping time determinations were carried out in these rats as well as in two other groups of nontrained rats which received either daily saline or amphetamine (5 mg/kg I.P.) injections for about 40 days also.

Effect of Chlorpromazine Pretreat men t on Amphetamine Licklr:!_g_
The effect of chlorpromazine pretreatment (0.5 mg/kg injected I.P. 30 minutes before the amphetamine session) on licking rate for amphetamine solution was determined. For rats Z-29, Z-30 and Z-33, the solution of amphetamine was 0.5 mM while for rats Z-50, Z-51 and Z-52, the solution of amphetamine used was 1.0 mM.

Amphetamine Licking -Three Operant Schedules
Three rats Z-34, Z-35, and Z-37 were t rained on two operant scheduleslicking and left lever pressing for secondary reinforcement. In addition, they were also trained to press the right lever to obtain 45 mg Noyes Food Pellets on a FI-60 second schedule. The rats were initially trained to lick for water then amphetamine

Effect of Withholding Morphine Injection on Amphetamine and Ethanol Ingestion
Amphetamine Ingestion -A solution of amphetamine 0.5 m"1 was substituted for water in the second session only on the last day of morphine injection. Responding in the behavioral box was measured for the next four consecutive days also (when no morphine injections were made) and then the rats were given morphine injections for at least one day before the procedure was repeated. This procedure 41 was repeated for a total of five replicates for rats Z-44 and Z-45 and three replicates for rats Z-47 and Z-49.
Ethanol Ingestion -A solution of ethanol (80% v/v) was suostituted for water in the second session on the last day of morphine injection for rats Z-47 and Z-49. Three, five day replicates were made in the same way as for amphetamine ingestion.

Effect of Nalorphine Injection on ·water · Ingestion
The effect of Nalorphine 4 mg/kg, I.P., injected 60 minutes before the second session, was determined i n rats receiving daily morphine injections. Water was available for licking in both sessions. The Nalorphine injection was replicated about six times in both groups of rats Z-44, Z-45 and Z-47, Z-49.

PREDRUG SESSIONS
As previously indicated in the experimental section, rats were run on predrug sessions to determine whether there was any carry-over effect of previous treatment. The decision to run a rat on any given day wa s based on responding throughout the 30 minute session of at least 75 percent of normal.
The data of predrug sessions is presented in appendices in the same sequence data of drug sessions is presented in the results section. This · predrug data has been analyzed by Factorial Analysis of Variance (ANOVA) in the same way as data from drug sessions. Wherever a statistically significant effect occurred between predrug sessions of corresponding drug sessions, further analysis of the data was made by use of Duncans Multiple Range Test (DMRT). The statistical significance is seen not to occur between groups of sessions corresponding to doses or concentrations in any ordered manner (successively increasing or decreasing) but rather in a random manner indicative of variability in day-to-day responding of the rats. -·--·---   The effect of increasing concentrations of amphetamine solutions on consequential licking rate in Rats Z-29, Z-30 and Z-33 is presented graph i cally in Figure 6. These data have been presented and analyzed in Table 2. The substitution of increasing concentrations of amphetamine for water produced a statistically significant effect (P < 0.001). Further analysis by DMRT showed the effect to be concentration dependent with increasing concentrations producing progressive decreases significantly different (P < 0.05) from ' water and with only the 1.98 and 0.99mM concentrations not statistically different (P > 0.05) from each other under both simulus conditions. According to ANOVA, the rats did not differ significantly (P < 0.05) among themselves; there was no significant effect (P > 0.05) due to stimulus condition and none of the possible interactions showed significant relationships (P > 0.05).
As can be seen from Figure 5  Effect of amphetamine self-ingestion on consequential licking rate under self-produced and programmed stimulus in Rats Z-29, Z-30, Z-33. Clear squares represent significant differences (P < 0.05) between amphetamine sessions and corresponding predrug sessions (circles).
"""' lD this effect by DM RT showed, as with consequential licking rates, there was a significant effect (P < 0.05) with increasing concentrations of amphetamine producing progressive decreases significantly different from water as well as between each increasing concentration with the only exception of the 1.98 and 0.99 mM concentrations under programmed stimulus not being statistically different (P > 0.05) from each other. There was a statistically significant difference (P < 0.01) between the two stimulus segments in minutes spent in licking and it can be seen from the data of Table 3  ----There was no significant difference (P > 0.05) between the two stimulus conditions nor among the rats. None of the possible interactions showed any significant (P > 0.05) relationships.

"'
According to ANOVA, the milligrams of amphetamine sulfate ingested by the rats was significantly affected (P < 0.05) as increasing concentrations of amphetamine solutions were substituted for water. The data and analyses are presented in Table 5. Further analysis by DMRT showed the amphetamine ingested at the highest concentration (1.98mM) was significantly less (P < 0.05) than under 53 either of the two lesser concentrations under both stimulus conditions. The amount of amphetamine ingested under the two lesser concentrations (0.99 and 0.50mM) was not significant (P > 0.05) between them.
According to ANOVA, there was no significant difference (P > 0.05) between the two stimulus conditions nor among the rats. None of the possible interactions showed any significant relationships (P> 0.05).

Inconsequential Licking
According to ANOVA, the substitution of solutions of amphetamine for water resulted in a significant effect (P < 0.05) on inconsequential licking rate. The data and analyses are presented in Table 6.
Further analysis by DMRT showed all concentrations produced a significant (P < 0.05) decrease when compared to water. Increasing concentrations of amphetamine solutions produced progressive decreases in inconsequential licking rates, but none were significantly different (P > 0.05) from each other. There was no significant difference (P > 0.05) among the rats nor in the interaction between concentrations and rats.    and analysis by DMRT showed the highest concentration (l.98mM ) produced a significantly lower (P < 0.05) left lever rate than the lowest concentration (0.50mM) but the middle concentration (0.99mM) was not significantly different (P > 0.05) from either the lowest or highest concentrations. According to ANOVA, the rats did not differ significantly (P > 0.05) among themselves nor was the interaction between concentration and rats significant (P > 0.05). The effect of amphetamine on consequential left lever rate in rats Z-29, Z-30 and Z-33 is presented graphically in Figure . 7 along with the effects on inconsequential left lever rate under self-produced stimulus.

Inconsequential Left Lever Pressing
For Secondary Reinforcement As can be seen from Figure 7, the substitution of solutions of amphetamine for water resulted in a marked increase in inconsequential left lever pressing for secondary reinforcement during self-produced stimulus conditions. Apparently due to large variability in the data, this effect was not statistically significant (P > 0.05) according to ANOVA. The data and analysis are presented in Table 8.
There was a significant difference (P < 0.05) between the rates under the two stimulus conditions and the data clearly indicates the rates were higher under self-produced than programmed stimulus conditions. There was no significant difference (P > 0.05) among

AMPHETAMINE (mM)
Effect of amphetamine self-ingestion on discriminated and non-discriminated left lever pressing rate (amphetamine/predrug) for secondary reinforcement during self-produced stimulus segments in Rats Z-29, Z-30, Z-33.
°' __,  The same three rats Z29, Z-30 and Z-33 as used in the previous experiment to determine the effects of amp hetamine self-ingestion were used to determine the effects of a daily intraperitonea1 injection of amphetamine (Smg/kg) given four hours after the daily amphetamine lick i ng sessions. On days prior to the start of the experiment, these rats were also injected daily with 5mg/kg of amphetamine.

Consequential Licking 64
The effect of tncreasing concentrations of amphetamine solutions on consequential li cking rate in these amphetamine treated rats is presented graph ically in Figure 8. Increasing concentrations produced decreased Ticki ng rates, but the major difference between this experiment and the previous one was that concentrations required were considerabl y lower; 0.125, 0.25 and O.SOmM versus 0.50, 0.99 and 1.98mM of t he previous experiment. A comparison of these resu]ts is presented graph ically in Figure 9. According to ANOVA, the licking rates were signifi cantly affected (P < 0.001) when increasing I concentrations of amphetamine solutions were substituted for water.
The data and anal yses are presented in Table 9.  Effect of amphetamine self-ingestion on consequential licking rate under self-produced and programmed stimulus in rats Z-29, Z-30, Z-33 which were chronically i~ected with 5 mg/kg, i.p. of amphetamine, 4 ho~rs after the amphetamine session. Clear squares represent significant difference (P < 0.05) between amphetamine and corres ponding predrug sessions (circles). Ll-J  Figure 9~· Effect of amphetamine self-ingestion on consequential licking rate under selfproduced and programmed stimulus in Rats Z-29, Z-30, Z-33 when they were normal and after chronic injections of amphetamine.

°' °'
r . (R) c x s C X R S X R C X S X R ERROR * s i g n i f 1 c a n t a t P < 0 . 0-5 **Significant at P < 0.01 ***Significant at P < 0.001

°' (X)
• concentrations whiie under programmed stimulus a significa~t difference (P < 0.05) occurred between the 0.125 and 0.25mM concentrations. According to ANOVA, there was no significant difference (P > 0.05) between the stimulus conditions and the rats differed significantly (P < 0.05) among themselves. None of the possible interactions showed any significant relationships (P > 0.05). 69 According to ANOVA, the substitution of increasing concentrations of amphetamine for water resulted in a significant effect (P < 0.001) on the amount of time spent in licking. The data and analyses are presented in Table 10. Further analysis of this effect by DMRT showed that the amount of time spent in licking under either stimulus condition was significantly reduced (P < 0.05) when compared to water.
Under both stimulus conditions, the amount of time spent in licking at the 0.50mM concentration of amphetamine was significantly less (P < 0.05) than at the 0.25mM concentration. The O.l25mM concentration effect was not significantly different (P > 0.05) from the other two. According to ANOVA, there was a significant difference (P < 0.05) between the two stimulus conditions and data of Table 10 indicates more time was spent in licking under the programmed stimulus than the self-produced stimulus. The rats differed significantly (P < 0.01) among themselves and the interaction between concentration and rats showed a significant (P < 0.01) relationship. A graphical comparison of time spent in licking by these rats during the experiment prior to the one in which they received daily injections of amphetamine is presented in Figure 10. Although the concentrations used for licking were considerably lower, the amount of time spent in licking was also generally less when the rats were receiving I  According to ANOVAs the substitution of increasing concentrations of amphetamine for water resulted in a significant effect (P < 0.05) on the milligrams of amphetamine sulfate ingested by the rats. The data and analyses are presented in Table 11. Further analysis of this effect by DMRT showed progressively more amphetamine was ingested with successively increasing concentrations and the amount ingested at the lowest concentration (0.125mM) was statistically lower (P < 0.05) than at the highest concentration (0.50mM) under both stimulus conditions. The amount ingested at 0.25mM concentration was not statistically different (P > 0.05) from either the 0.125 or 0.50mM concentrations. According to ANOVA, there was no significant difference (P > 0.05) between the two stimulus conditions nor among the rats in amounts of amphetamine ingested. None of the possible interactions showed significant (P > 0.05) relationships. A graphical comparison between the amount of amphetamine injection experiments is presented in Figure 11. Here a clear distinction between decreasing amount of amphetamine ingested before, and increasing amount of amphetamine ingested during amphetamine injection can be seen.

Inconsequential Licking
According to ANOVA, the substitution of increasing concentrations of amphetamine for water resulted in a significant effect (P < 0.001) --.an inconsequential licking rates. The data and analyses are presented in Table 12. Further analysis of this effect by DMRT showed all concentrations of amphetamine produced a significant decrease (P < 0.05) in inconsequential licking rate when compared to water. Although each increasing concentration of amphetamine produced a progressive t '    Any two means not underscored by the same line are significantly different at p< 0.05.
"'-J 00 79 decrease in inconsequential licking rate, these effects were not statistically significant (P > 0.05) from each other. According to ANOVA, there was no significant difference (P > 0.05) among the rats nor in the interaction between concentration and rats.

Consequential Left Lever Pressing
For Secondary Reinforcement According to ANOVA, the substitution of increasing concentrations of amphetamine for water resulted in a significant effect (P < 0.001) on consequentia1 left lever pressing for secondary reinforcement.
The data and analyses are presented in Table 13. Further analysis of this effect by DMRT showed all concentrations of amphetamine produced a significant decrease (P < 0.05) in consequential left lever pressing when compared to water. None of the effects produced by the amphetamine concentrations were significantly different (P > 0.05) from each other. This effect is presented graphically for rats Z-29, Z-30 and Z-33 in Figure 12. According to ANOVA, there was no significant difference (P ~ 0.05) among the rats nor in the interaction between concentration and rats.

Inconsequential Left Lever Pressing
For Secondary Reinforcement According to ANOVA, the substitution of increasing concentrations of amphetamine for water had no significant effect (P > 0.05) on ~nconsequential left lever pressing for secondary reinforcement.
The data and analysis are presented in Table 14. The inconsequential left lever rates under self-produced stimulus are presented graphically in Figure 12. The rates were significantly different (P < 0.001) under the two stimulus conditions and the data in Table 14 clearly shows the ,/ ------------*Significant at P < 0.05 . **Significant at P < 0.01 ***Significant at P < 0.001  Figure 12. Effect of amphetamine self-ingestion on discriminated and non-discriminated left lever pressing rate (amphetamine/predrug) for secondary reinforcement during self-produced stimulus in Rats Z-29, Z-30, Z-33 which were chronically injected with 5 mg/kg, i.p. of amphetamine, 4 hours after the amphetamine sessions. Any two means not underscored by the same line are significantly different at P <0.05.
co ~ rates were greater under self-produced stimulus. The rats did not differ significantly (P > 0.05) among themselves and the only interaction which shows a significant relationship (P < 0.01) was between con centration and rats.

Barbiturate Narcosis
In order to test whether any changes in metabolism or central nervous system sensitivity occurred as a result of injecting rats daily over about a 40 day period with amphetamine, two groups of nont rained rats were given a daily intraperitoneal injection of 85 either normal saline or amphetamine (mg/kg). The results of barbiturate narcosis studies on these rats are presented in
stimulus and for rat Z-30 under both stimulus conditions. In another three rats, Z-50, Z-51 and Z-52 offered a 1 .OmM solution of amphetamine, chlorpromazine produced a marked increase in licking rate for amphetamine but the effect was significant (P < 0.05) only for rat Z-50 under self-produced stimulus. In this experiment, only the effect of chlorpromazine on consequential licking rate was considered.

Amphetamine Licking -Three Operant Schedules
Consequential Licking 88 The effect of increasing concentrations of amphetamine solutions on consequential licking rate in rats Z-34, Z-35 and Z-37 is presented graphically in Figure 13. These data have also been presented and analyzed in Table 17. According to ANOVA, consequential licking rate was significantly affected (P < 0.001) due to substitution of amphetamine solutions for water. Further analysis of this effect by OMRT showed the effect to be concentration dependent with all increasing concentrations producing progressive decreases significantly different (P < 0.05) from water. Under self-produced stimulus, the 0.50mM concentration produced a decrease significantly different (P < 0.05) from all the lower concentrations. Under programmed stimulus, there was no significant difference (P > 0.05) between 0.50 and 0.25mM concentrations, but both resulted in significantly ----gr ea t~r (P < 0.05) decreases than 0.125 and 0.0625mM concentrations.
According to ANOVA, there was a significant difference (P < 0.001) between the two stimulus conditions and data of Table 17 shows clearly the consequential licking rates were consistently higher under selfproduced as compared to programmed stimulus. There was also a .I I ,   The milligrams of amphetamine sulfate ingested by -these rats, as can be seen in Figure 13, tended to either increase, remain stable or decrease for rats Z-34, Z-35 and Z-37 respectively, with 92 increasing concentrations of amphetamine solutions. The data and analyses are presented in Table 18. According to ANOVA, this was a significant effect (P < 0.05), but further analysis by DMRT showed the milligrams of amphetamine ingested were not significantly different (P > 0.05) from each other under the various amphetamine concentrations. According to ANOVA, there was no significant difference (P > 0.05) and there was a significant difference (P < 0.001) among the rats. None of the possible interactions showed any significant (P > 0.05) relationships.
According to ANOVA, the time spent in consequential licki.ng was significantly affected (P < 0.001) by substitution of solutions of increasing concentrations of amphetamine for water. The data and analyses are presented in Table 19. Further analysis by DMRT showed there was a progressive decrease in time spent in licking. There was no significant difference (P > 0.05) between the lowest concentration (0.0625mM) and water, but all the other concentrations     ----------*Significant at P "0,05 **Significant at P < 0,01 '° ***Significnat at P < 0.001 two stimulus conditions. There was a significant difference (P < 0.001) among -the rats and in the concentration and rat interaction (P < 0.001).
None of the other possible interactions showed a significant (P > 0.05) relationship.
According to ANOVA, the number of reinforcements (drops of fluid) delivered were significantly affected (P < 0.001) by substitution of solutions of increasing concentrations of amphetamine for water.
The data and analyses are presented in Table 20. Further analysis by DMRT showed there was a progressive decrease in number of reinforcernents delivered with increasing concentrations with all concentrations significantly (P < 0.05) less than water. With the exception of 0.0625mM concentration under self-produced stimulus, there were no significant (P > 0.05) between any given concentration and the concentrations immediately adjacent to it. All other concentrations were significantly different (P < 0.05) from each other. According to ANOVA, there was a significant difference (P < 0.001) between the two stimulus conditions, and it can be seen from the data of Table   20 that more reinforcements were delivered during self-produced than programmed stimulus. There was~ significant difference (P < 0.001) among the rats as a significant (P < 0.01) relationship in the rat and concentration interaction. None of the other possible interactions -~showed a significant (P > 0.05) relationship.

Inconsequential Licking
According to ANOVA, inconsequential licking rates were significantly affected (P < 0.01) by substitution of solutions of   Table 21. Further analysis by DMRT showed there was a progressive decrease in inconsequential 1icking rates with increasing concentrations with a significant decrease (P < 0.05) under all concentrations of amphetamine when compared to water. None of the concentrations were significantly different (P > 0.05) from each other. According to ANOVA, there was no significant difference (P > 0.05) among the rats nor in the concentration and rat interaction.

Consequential Left Lever Pressing
For Secondary Reinforcement Consequential left lever pressing rate for secondary reinforcement decreased with increasing concentrations of amphetamine as can be seen in Figure 14. According to ANOVA, this was a significant effect (P < 0.001). The data and analyses are presented in Table 22.
Further analysis by DMRT, showed there was a progressive decrease in consequential left lever rate with all concentrations except the lowest (0.0625mM) significantly decreased (P < 0.05) as compared to water.
There was no significant difference (P > 0.05) between the remaining amphetamine concentrations. According to ANOVA, there was a significant difference (P < 0.001) among the rats as well as in the concentration and rat interaction (P < 0.001).

Inconsequential Left Lever Pressing
Far Secondary Reinforcement According to ANOVA, the inconsequential left lever rates for secondary reinforcement were significantly affected (P < 0.01) by substitution of solutions of amphetamine for water. The data and  277.80 *Significant at P.:: 0.05 **Significant at P < 0.01 ***Significant at P < 0.001   Effect of amphetamine self-ingestion on consequential and inconsequential left lever pressing for secondary reinforcement during self-produced and programmed stimulus in Rats Z-34, Z-35, Z-37. Food pellets were concurrently available on FI-60" right lever pressing.  analyses are presented in Table.. 23.

106
The rats differed significantly (P < 0.01) among themse..Tves in this respect and this can be seen graphically in Figure T4. Further analysis of this effect by DMRT, showed the effects of amphetamine were not concentration -related.
According to ANOVA, there was a significant difference (P < 0.001) between the stimulus conditions and data of Table 23 clearly shows the rates were higher under self-produced stimulus than under programmed stimulus. There were significant relationships in the concentration and rat interaction (P < 0.05) and in the stimulus segment and rat interaction (P < 0.001). None of the other interactions showed a significant relationship (P < 0.05).

Right Lever Pressing For Food Pellets
The effect of amphetamine self-ingestion on the third operant response in this study, right lever pressing for food pellets is presented graphically in Figure 15. The data is expressed as consequential lever pressing and total lever pressing. This graph indicates that total right lever pressing increased with increasing amphetamine concentrations while consequential lever pressing was relatively unaffected. According to ANOVA, this observation is substantiated in that total right lever rates were significantly affected (P < 0.001) by amphetamine. The data and analyses are presented in Table 24. The consequential right lever rates were ~ot significantly affected (P> o·.05) by amphetamine. The data and analyses are presented in Table 25. Further analysis of the effect of amphetamine on total right lever rates by DMRT showed a progressive increase in lever pressing rate with increasing concentrations of amphetamine. Significant increases (P < 0.05) between sets of     stimulus conditions in either total right lever rates or consequential right -lever rates. There was a significant difference -(P < 0.001) among the rats in total and consequential right lever rates. For total right lever rates, there was a significant relationship (P < 0.001) in the concentration and rat interaction, while for consequential right lever rates there was a significant relationship in the concentration and stimulus condition interaction (P < 0.05) as well as in the concentration and rat interaction (P < 0.001). None of the other possible interactions were significant (P > 0.05).

Effect of Amphetamine Injection on Water Self-Ingestion
Consequential Licking The effect of amphetamine injections can be seen from cumulative records for rat Z-19 in Figure 16. The cumulative records on the left side are of predrug sessions where responding was at a high rate and consistent for all the sessions. On the right are the corresponding drug sessions. When either no injection was made or nonnal saline was injected, there were no apparent differences between predrug ·and drug sessions. The injection of increasing doses of amphetamine produced increasing disruption of responding to where it was totally abolished at the 2.0mg/kg dose.
The effect of amphetamine injections to decrease consequential lickjng rate in rats Z-19, Z-20 and Z-22 is presented graphically in Figure 17. According to ANOVA, consequential licking rates were significantly affected (P < 0.001) by amphetamine injections. The data and analyses are presented in Table 26. Further analysis of  Effect of amphetamine injection on consequential licking rates for water during self-produced and programmed stimulus in Rats Z-19, Z-20, Z-22.
Clear squares indicate significant differences (P < 0.05) between amphetamine sessions and corresponding predrug sessions (circles).
,.    Table 26 indicates the consequential licking rates were greater under self-produced stimulus. There was no significant differ ence (P > 0.05) among the rats. The interaction between dose and stimulus segment was significant (P < 0.05), but none of the Gther interactions showed any significant relationships (P > 0.05).
According to ANOVA, the time, in minutes, spent in consequential licking was significantly affected (P < 0.001) by different doses of amphetamine. The data and analyses are presented in Table 27.
Further analysis of this effect by DMRT showed there was no significant difference (P > 0.05) between saline injections, and the 0.25~ 0.50 and l.Omg/kg doses. The 2.0mg/kg dose resulted in a _significantly lowered (P < 0.05) time spent in licking as compared to -saTfne injection and the other three doses of amphetamine. According to ANO V A, there was a significant difference ( P < 0. 05) bet\'1een the two stimulus conditions and data of Table 27 indicates more time was spent under programmed stimulus than self-produced stimulus.
"' f ' *Significa.nt at P < 0.05 **Significant at P < 0.01 ***Significant at P < 0.001 There was a significant (P < 0.05) interaction between dose and rats, but none of the other interactions showed any significant relationships (P > 0.05).
According to ANOVA, the number of drops of water delivered was significantly affected (P < 0.001) by amphetamine injections. The data and analyses are presented in Table 28. Further analysis of this effect by DMRT, showed there was a general decrease in water consumption with increasing doses of amphetamine, but the only significant decrease (P < 0.05) occurred at the 2.0mg/kg dose with the only other exception of a significant decrease (P < 0.05) between the l.Omg/kg dose and saline under programmed stimulus.
According to ANOVA, there were no significant differences (P > 0.05) between stimulus conditions, among the rats or in any of the interactions.

Inconsequential Licking
According to ANOVA, inconsequential licking rates were significantly affected (P < 0.05) by amphetamine injections. The data and analyses are presented in Table 29. Further analysis of this effect by DMRT, showed there was no dose dependent effect, but the 2.0mg/kg dose produced the greatest decrease which was significantly less (P < 0.05) than saline and 0.25 and 1.0mg/kg -.-Oases. The 0.50mg/kg dose produced a significant decrease (P < 0.05) when compared to saline. According to ANOVA, there wa? no significant difference (P > 0.05) among the rats or in the dose and rat interaction.     According to ANOVA, amphetamine injections had no significant effect _(P > 0.05) on consequential left lever rate for-secondary reinforcement. The data and analyses are presented in Table 30.
Further analysis of this data by DMRT showed there was no dose dependent relationship, but that the 2.0mg/kg dose produced a significant decrease (P < 0.05) when compared to saline and the other doses of amphetamine. According to ANOVA, there was no significant difference (P > 0.05) among the rats or in the dose and rat interaction.

Inconsequential Left Lever Pressing
For Secondary Reinforcement According to ANOVA, inconsequential left lever rates for secondary reinforcement were significantly affected (P < 0.05) by amphetamine injections. The data and analyses are presented in Table 31. Further analysis of this data by DMRT showed there was no dose dependent relationship, but under self-produced stimulus, the 2.0mg/kg dose produced an increase significantly greater (P < 0.05) than that produced by the other doses while under programmed stimulus the l.Omg/kg dose produced an increase significantly greater (P < 0.05) than that produced by either saline or other doses of amphetamine.
According to ANOVA, there was a significant difference (P < 0.01) in   The effect of substitution of increasing concentrations of ethanol for water to decrease consequential licking rate in rats Z-25, Z-26, Z-27 and Z-28 is presented graphically in Figure 18.
According to ANOVA, the substitution of ethanol solutions for water had a significant effect (P < 0.001) on consequential licking rate.
The data and analyses are presented in Table 32. Further analysis of this data by DMRT showed the effect was concentration dependent with increasing concentrations of ethanol producing progressive decreases in consequential licking rate. There was no significant difference (P > 0.05) between the 10% or 20% concentrations and 132 water or themselves, but the 40% and 80% concentrations produced rates significantly lower (P < 0.05) than either 10% or 20% concentrations or water. The rate under 80% was significantly lower (P < 0.05) than under 40%. According to ANOVA, there was a significant difference (P < 0.001) between the stimulus conditions and data of Table 32 shows the rates were greater under self-produced than programmed stimulus. There was a significant difference (P < 0.001) among the rats and a significant relationship (P < 0.001) in the concentration and rat interaction. None of the other interactions were significant.
According to ANOVA, the time, in minutes, spent in consequential _ticking was significantly affected by ethanol solutions. The data and analyses are presented in Table 33.      According to ANOVA, there was a significant difference (P < 0.001) between the stimulus conditions and data of Table 33 shows more time was spent under programmed stimulus than under self-produced stimulus.
There was a significant difference (P < 0.001) among the rats as well as a significant relattonshjp (P < 0.001) in the concentration and rat interaction. None of the other interactions showed a significant interaction (P > 0.05).
According to ANOVA, the drops of fluid delivered were significantly affected (P < 0.001) by ethanol solutions. The data and analyses are presented in Table 34. Further analysis by DMRT showed a progres~ sive decrease in fluid consumption with increasing ethanol concentrations. ' There was no significant difference (P > 0.05) between the 10% concentration and water or 20% concentration, but all other concentrations of ethanol were significantly different (P < 0.05) from water and from each other. According to ANOVA, there was no significant difference (P > 0.05) between the stimulus conditions.
There was a significant difference (P < 0.001) among the rats as well as in the concentration and rat interaction (P < 0.001).
-There was no significant relationship (P > 0.05) in any of the --~ther interactions.
According to ANOVA, there was a significant difference (P < 0.001) in grams of absolute ethanoT ingested under the various ethanol concentrations. The data and analyses are presented in Table 35.             was the most marked decrease seen at 80%, but it was not significantly less (P > 0. 05) than at 40%. According to ANOVA, there was a significant difference (P < 0.001) among the rats and the interaction between concentration and rats was not s ign-i fi ca-nt ( P> 0. 05) , --·--... -

Consegue~tial Left Lever Pressing
For Secondary Reinforcement ---According to ANOVA, consequential left lever rates were significantly affected (P < 0.001) by ethanol solutions. The data and analyses are presented in Table 37. Further analysis of this effect by DMRT showed there was a concentration dependent decrease in consequent ial left lever rates with increasing concentrations of ethanol. All concentrations of ethanol produced a significant decrease (P < 0.05) when compared to water. There was no significant differences (P > 0.05) between 10% and 20% and between 40% and 80% but both 40% and 80% were significantly lower (P < 0.05) than 10% and 20%. According to ANOVA, there was a significant difference (P < 0.001) among the rats and a significant relationship (P < 0.01) in the rat and concentration interaction. The effect of ethanol on consequential left lever rates in rats Z-25, Z-26, Z-27 and Z-28 is presented graphically in the upper portion of Figure 20.

Inconsequential Left Lever Pressing
For Secondary Reinforcement According to ANOVA, inconsequential left lever rates were not s1gnif.icantly affected (P > 0.05) by ethanol. The data and analysis are presented in Table 38. The data is also presented graphically for self-produced stimulus in the lower portion of Figure Table 38, the rates under self-produced stimulus can be seen to be greater than under programmed stimulus. There was a significant difference (P < 0.05) among the rats and a significant relationship (P < 0.05) in the stimulus segment-and rat -interaction.
None of the other interactions showed any significant relationships (P > 0.05).  Table 39. Further analysis by DMRT showed there was a progressive decrease in consequential licking rates under selfproduced stimulus with increasing concentrations. All concentrations, but 10%, were significantly lower (P < 0.05) than water. There was no signifi~ant difference (P > 0.05) between 10%, 20% and 40% and while the 80% concentration was significantly lower (P < 0.05) than all the other concentrations. Under programmed stimulus all concentrations of ethanol produced rates significantly lower (P < 0.05) than water. There was no significant difference (P > 0.05) between the 10%, 20% and 40% concentrations, while the 80% concentration was significantly lower (P < 0.05) than all others. According to ANOVA, there was a significant difference (P < 0.001) between stimulus ~egments and from Table 39        5574. 10 *Significant at P < 0.05 **Significant at P <: 0.01 ***Significant at P < 0.001  Table 40. Further analysis by DMRT showed a progressive decrease in time spent -fn consequential licking with increasing concentrations of ethanol. All concentrations produced a significant reduction (P < 0.05) when compared to water.
------Tnere was ·no significant difference (P > 0.05) between 10%, 20% and 40%. Under self-produced stimulus, the 80% concentration produced a rate significantly lower (P < 0.05) than 10% while under programmed stimulus the 80% concentration produced a rate significantly lower (P < 0.05) than 10% and 20%. According to ANOVA, there was no significant difference (P > 0.05) between stimulus conditions. There was a significant difference (P < 0.001) among the rats and in the concentration and rat interaction (P < 0.001). None of the other interactions showed any significant relationships (P > 0.05).
According to ANOVA, drops of fluid delivered was significantly affected (P < 0.001) by ethanol solutions. The data and analyses are presented in Table 41. Further analysis of the data by DMRT shows there was a progressive decrease in drops of fluid delivered as increasing concentrations of ethanol were substituted for water.
The drops of fluid delivered at all concentrations of ethanol were significantly lower (P < 0.05) than for water. There was no significant difference (P > 0.05) between the 10%, 20% and 40% concentrations.
-~he 80% concentration was significantly lower (P < 0.05) than the -10% and 20% concentration effects. According to ANOVA, there was a significant differen~e (P < 0.01) between the two stimuli and the data of Table 41 shows values under self-produced stimulus are    greater than under programmed stimulus. There was a significant difference (P < 0.001) among the rats. The interaction between concentration and rat was significant (P < 0.01) but none of the other interactions had a significant relationship (p-->-0.05}: · According to ANOVA, the grams of absolute ethanol ingested was significantly affected (P < 0.001) by the various concentrations -~~-r ethanol. The data and analyses are presented in Table 42. Further between the 10% and 20% concentrations nor between the 40% and 80% concentrations but both 40% and 80% were significantly greater (P < 0.05) than the two lower concentrations, 10% and 20%. Under programmed stimulus, there was no significant difference (P > 0.05) between the 10% and 20% concentrations nor between the 20%, 40% and 80% concentrations but both the 40% and 80% concentrations were significantly greater (P < 0.05) than the 10% concentration. According to ANOVA, there was a significant difference (P< 0.01) between the two stimulus conditions and according to the data of Table 42, more ethanol was consumed under the self-produced stimulus than the programmed stimulus. There was a significant difference (P < 0.05) among the rats. There was a significant relationship (P < 0.001) in the concentration and rat interaction. None of the other inter-___ ---actions showed any significant relationship (P > 0.05).

Inconsequential Licking
According to ANOVA, inconsequential licking rates were significantly affected (P < 0.001) by the substitution of ethanol   solutions for water. 169 The data and analyses are presented in Table 43.    Table 45. There was a significant difference (P < 0.001) between the two stimulus conditions and according to the data of Table 45, the rates were higher under self-procfuced sti mulus than programmed stimulus. There was no significant difference (P > 0.05) among the rats. A significant relationship (P < 0.05) exists in the ,concen tra ti on and rat interaction, but all other possible interactions were not significant (P > 0.05).

For Food Pellets
According to ANOVA, the substitution of solutions of ethanol for water had a significant effect (P < 0.001) on the total right lever rate for food pellets. Food pellets were available on a FI-60 second schedule so these rates reflect all responding on the right lever, bQth inconsequential and consequential. The data and analyses are presented in Table 46. Further analysis by DMRT showed there was no significant difference (P > 0.05) in responding between the ethanol concentrations and water under self-produced stimulus, while under program med stimulus the only significant difference (P < 0.05) was in a higher rate under 80% ethanol as compared to water.
According to ANOVA, there was no significant d1fference (P > 0.05) between stimulus conditions. There was a significant difference -,._(p < 0.001) among the rats and in the interaction between concentration and rat (P < 0.05). None of the other possible interqctions showed any significant relationships (P > 0.05).
There was a significant effect (P < Q.05) in the stimulus and rat interaction, but none of the possible interactions showed any ~;fgnifitant (P > 0.05) relationships.

Effects of Oral Ethanol Injections
The three rats, Z-53, Z-54 and Z-55, selected for the study of

Effect of Water or Ethanol Licking Without Ethanol Injection
The results for all variables for the three rats combined are presented and analyzed by Students t Test in Table 48. Total -right lever pressing (consequential and inconsequential) for food pellets and increased significantly (P < 0.01) under self-produced stimulus during ethanol ingestion. There was no significant Ethanol produced a significant reduction (P < 0.001) in consequential licking rates, amount of time spent in licking and amount of fluid ' consumed under both stimulus conditions. The inconsequential licking rate was likewise significantly reduced (P < 0.001) by ethanol.

Effects of Ethanol Injection on Water Licking
The results for all variables for the three rats combined are presented and analyzed by Students t Test in Table 49. With the exception of total right lever pressing, all variables were significantly reduced at a probability value between (P < 0.025 -P <

Effects of Ethanol Injection on Ethanol Lick~
The results for the three rats combined are presented and analyzed by Students t Test in Table 50. With the exception of conseqilential right lever pressing under self-produced stimulus and inconsequential licking rates under self-produced stimulus, there were no significant differences (P > 0.05) in any of the -::Var-tables before or after ethanol injection with ethanol was available for licking. Consequential right lever pressing for food pellets during self-produced stimulus was significantly reduced (P < 0.01) as was inconsequential licking (P < 0.05) by ethanol injections.

Effects of Ethanol Injection on Water and Ethanol Licking
The results for the three rats combined are presented and analyzed by Students t Test in Table 51. With the exception of consequential licking rate and drops of fluid delivered, there were no significant differences (P > 0.05) in any of the other variables between ethanol injection sessions when water or ethanol was available for licking. Consequential licking rates were significantly lower during ethanol licking in the self-produced stimulus (P < 0.01) and the programmed stimulus (P < 0.025). Drops of fluid delivered were also significantly lower during ethanol licking in the self-produced stimulus (P < 0.025) and the programmed _ __ stimulus (P < 0.05).

DisuTfiram Effects on Ethanol Ingestion
The effect of disulfiram on ingestion of ethanol (20% v/v) 187 in rats Z-59, 60, 61 is seen in Figure 21.       consecutive days on consequential licking rates for water, amphetamine· " (0. 5 mM) and alcohol (80% v/v) under self-produced and programmed stimulus · in morphine dependent rats Z-47 and Z-49. Food pellets were concurrently available on FI-60'' right lever pressing. Effect of withholding daily morphine injection (200 mg/kg, i.p.) for four consecutive days on minutes spent in consequential licking for water, amphetamine (0.5 mM) and alcohol (80% v/v) under self-produced and programmed stimulus in morphine dependent rats Z-47 and Z-49. Food pellets were concurrently available on FI-60 11 right lever pressing. for the first two days are presented in the lower two thirds of Figure 27. On the first day, the responses were very low, but greatly increased on the second day. In rat-C-47, the consequential licking rates for ethanol were lower than for either water or amphetamine ( Figure 24). Similar results were observed in amount -.i-::-crf--tfote.-Spe_ nt in consequential licking ( Figure 26). In Z-49 however, the consequential licki ng rate was higher for amphetamine and by 96 hours of morphine deprivation was higher than for water ( Figure   24). Similar results were observed for amount of time spent in consequential licking ( Figure 26).

Effect of Nalorphine Injection
The effect of nalorphine injection was investigated in rats Z-44, Z-45, Z-47, Z-49 which were receiving a daily dose of 200mg/kg of morphine. A dose of 2mg/kg of nalorphine was ineffective in all four rats and this effect is presented for rat Z-44 in the upper half of Figure 28. The effect of 4mg/kg of nalorphine is presented for rat Z-44 in the lower half of Figure 28. At this dose, nalorphine, abolished all responding in rat Z-44. The data for the effect of 4mg/kg of nalorphine in each of the four rats is presented and analyzed by "t" test in Table 52.
In rat Z-44, nalorphine injections resulted in a significant decrease (at least P < 0.05) in consequential left lever rate,   In rat Z-45, nalorphine injections also resulted in decreased responding, but none of the effects were statistically significant (11 > Q.05).

V. DISCUSSION
Oral self-actministration of drugs was studied using rats as exper-tm~ntal subjects. The use of experimental animals as subjects eli minated complications of psychological and personality factors which complicate studies with humans. Oral self-administration of drugs was studied since the drugs are commonly self-ingested and the results obtained could be more reliably extrapolated to human behavior. Oral selfadministration via licking was selected for several reasons. Licking is a response in the rats' natural repertoire, is a rapidly acquired response and is a response which provided optimum high rates of responding. In addition, licking rates provided reliable and quantitative data with which to study oral self-administration of drugs.
Since drug use by humans is under the influence of complex controls as motivation, availability of drugs and availability of funding, the application of simultaneous multiple operants as utilized in these ' experiments afforded a means by which to evaluate the complex effects of self-administered drugs in the same subject. The measurement of licking behavior under two different stimulus conditions provided an indicator of motivation as well as motor ability. During the programmed stimulus phase, fluid was freely available to the rat while during the self-produced stimulus phase, the rat had to first perform five presses on the left lever in order to gain access to fluid from the licking spout. In some instances the right lever was programmed to deliver food pellets on a fixed interval schedule of one minute. Responding or lack of responding on the right lever provided data on the effects of drugs on food motivated behavior, fixed interval responding as well as motor ability. Since consequential as well as inconsequential re-208 spons~s were possible on licking or on either lever, the inconsequential . . . . responses were recorded as well and used as a measure of tne disruptive effects of the drugs on the discriminative ability of the rat~.
In order to assess the applicability of this simultaneous multiple operant design to the study of drug self-administration, several different drugs were investigated.
The first drug to be studied was dextro-amphetamine. Self-administration of solutions of amphetamine via the oral route was readily achieved in rats. A~phetamine ingestion resulted 1n a depression of licking following an initial period of normal licking. That licking persisted for several minutes before it was abruptly depressed is an indication that the taste of the solutions was not aversive. While oral ingestion of amphetamine produced a sudden depression in licking after a period of no effect, intraperitoneal injections of amphetamine caused a depression ?f behavior throughout the entire session. This suggested that a sufficient quantity of amphetamine had to be administered before pharmacological effects were exhibited.
The decrease in consequential licking rates, inconsequential licking rates and consequential left lever pressing for secondary reinforcement suggests a behavioral depression produced by amphetamine. These findings are si~ilar with respect to response rate, to the works of Deneau et ~ (1969) and Pickens and Harris (1968) who studied intravenous selfadministration of amphetamine in rats. When the magnitude of infusions was increased, the rate of lever pressing for intravenous infus. ion decreased in similar fashion to decreased response rates seen with increasing concentrations of amphetamine.
Non-discriminated aspects of responding· as for secondary reinforce-ment ~r food pellet~ incr~ased with amphe~amine. This appears to be the result pos:sibly _of motor hyperactivity and/or disruption of discrimination. A s:.imi 1 ar action of amphetamine was previously reported by La 1 and Brown (1969l, who demonstrated depression by amphetamine, of goal directed behavior and stimulation of irrelevant or purposeless behavior. The complex behavioral effects of amphetamine are best evaluated by simultaneous multiple-operants as utilized in this study since the effect of amphetamine on behavior maintained by different reinforcers and different schedules of reinforcement was studied concurrently in the ~ame subject. While amphetamine shows anorexic properties in a free eating situation (Epstein, 1959), consequential operant responding under the control of fixed-interval schedules for food pellets was unaffected by amphetamine in the present investigation. While other discriminated responses were reduced, the relevant responding on the right lever for food p~llets ~as not so affected. Since responding on right lever was maintained by a fixed-interval schedule this observation is in agreement with that of Kelleher and Morse (1968), who reported an increased rate of lever pressing for food by amphetamine under a fixed-interval schedule, but depressed responding under fixed-ratio schedule. They concluded that the behavioral effects of amphetamine were controlled more by the contingencies: of reinforcement than by the r· einforcement itself. Dews (1958) reported that amphetamine exerts its effect on schedule controlled behavior by increasing low rates ofresponding maintained by schedules such as the fixed interval and decreasing high rates of responding maintained by schedules such as fixed ratios. More recently Weiss (1970) ,. has.. further examined the effect of amphetamine on schedule controlled behavior. Using interresponse time as the primary measure-210 m~nt, b~ attributed tQ~ increased rate of re£ponding during fixed interval schedules ca~£ed by amphetamine to decreased interresponse times early in the interval and decreased rate of responding during fixed ratio schedules to increased interresponse times during the first portion of the fixed ratio.
Drugs are rarely freely available to human addicts. More often than not, drugs are available at specific times and after completion of a specific task in order to obtain sufficient money to purchase drugs.
The hypothesis that rats too will perform some task in order to gain access to drug solutions was tested in this study. The rats were re-9uired to press the left lever (FR-5 schedule) in order to attain an opportunity to obtain drug solutions. These results were then compared to the situation when fluid was freely provided. During predrug sessions when water was available for licking, the licking rates after lever pressing were always higher than when fluid was freely provided.
This effect was abolished when amphetamine was substituted for water in rats exposed to the two operant procedure both when they were normal as well as when they were chronically treated with amphetamine, but was not changed in rats which were exposed to the three operant procedure or in rats which were injected with amphetamine. In no instances were the licking rates higher when fluid was freely provided. Further examination of the data revealed that significantly more time was available for licking under the programmed stimulus than the self-produced stimulus in all groups of rats except the three operant group in which there was no difference. Also since there were no significant differences in the number of reinforcements obtained under either stimulus condition, the observed effects appear due to amphetamines effect on licking rates 211 ra~h~r ~han ~a som~ asp~c~ of th~-~~p~rim~ntal desi~n. The interaction between concentration and stimulus was not significant when amphetamine was available for ingestion, but the interaction between dose and stimulus was significant when amphetamine was ingested, indicating that the ~ifferential dose effect depended upon the stimulus contingencies.
Inconsequential licking was decreased in all the studies with ~rnphetamine whether it was ingested or injected. This appears to be an exception to the general rule that amphetamine increased irrelevant behavior and may be unique to the operant response of licking. Inconsequential licking could only occur during· the self-produced stimulus so there is no stimulus comparison, but in no instance was there a significant interaction between either concentration or dose with rats indieating it was independent of any differences in the rats behavior.
Inconsequential l~ft lever pressing for secondary reinforcement was consistently increased by amphetamine whether it was ingested or injected. Inconsequential left lever rates were also consistently greater under self-produced than programmed stimulus. This appears to be correlated with the fact that left lever pressing on a FR-5 schedule was an integral component of the self-produced stimulus and even during predrug sessions there was a significant increase in inconsequential responses during self-produced stimulus.
Following chronic treatment with amphetamine an apparent increased sensitivity to ~elf-ingested amphetamine occurred in all responses.
This increased sensitivity was manifested by a shift of the concentration response curves to lower concentrations. following chronic treatment, the variability in responding between rats increased significantly in~ dicating different degrees of animal susceptibility to chronic treat-ment with amphetamine.
(19661 for rat~ res_ponding on complex schedulP.s of r~lnforcem~nt. Rela~ tfvely· 1 ittl e is. know-about the effects of long.,..term adminlstration of amphetamine. Behavioral studies have reyealed different changes in the response to stimulant drugs which develop during the course of chronic administration. These include no tolerance (Uehling, 19691 in measures of lever pressing (CRF schedule) for i1lumination change; tolerance (Schuster et~. 1966) as measured by lever pressing (fl or DRL schedules) for food reinforcement, and increase in magnitude of response (Kosman and Unna, 1968} as measured by spontaneous activity of rats injected chronically with d,1-amphetamine.
Inhibited metabolism of amphetamine by hepatic microsomal enzymes was ruled out as a poss.ible mechanism for the increased sensitivity.
The duration of hexobarbital narcosis was not significanlty different in control animals or rats chronically treated with amphetamine. Since mice. This difference may be due to the fact that an acute rather than chronic dose of amphetamine was used. Since barbital is not metabolized (Dorfman and Goldbaum, 1947;Ebert et~. 19641 and duration of barbital narcosis was not significantly affected, there appears to be no change in central nervous system sensitivity in these rats to barbiturates to explain the apparent increased sensitivity to ingested amphetamine. An alternate explanation might be an accumulation of amphetamine in these rats as a result of the combined oral ingestion and chrontc daily injection of amphetamine. In the present investigation, intraperitoneal injections of chlorpromazine markedly increased licking of amphetamine solutions. Since these drugs are presumed to have opposite actions on the adrenergic system, with chlorpromazine blocking (Brodie et ~, 1959) and amphetamine potentiating (Hanson, 1966) the actions of catecholamines these findings of mutual antagonism might be expected. Schuster and Wilson (unpublished data) showed that chlorpromazine increased the intravenous selfadministration of amphetamine by ~hesus monkeys with indwelling intravenous catheters. Chlorpromazine has been reported to antagonize behavioral depression caused by amphetamine in a food reinforced operant in the rat (Brown, 1963) as well as in the pigeon (Davis, 1965). Glick and Jarvik (1969) reported an antagonism between the effects of amphetamine and chlorpromazine on delayed matching performance in monkeys. Maickel (1968) has shown that amphetamine will reverse the decrease in gross behavior and motor activity caused by chlorpromazine in rats.
Self-administration of ethanol via an oral route was also readily achieved in rats. As with amphetamine, the persistent licking for several minutes before depression of behavior suggest that the licking motivation was stronger than the aversion to taste until a high concentration of ethanol was licked. Depression of ethanol licking seemed to be due to an intoxication effect since the behavioral depression was gradual and some licking was maintained throughout the session. 214 In this study, rats have ingested significant amounts of ethanol in concentrations (up to 80% v/v) well above those reported in the literature to date. Myers (1968) has reported that rats will not voluntariiy consume ethanol if the solutions are greater than 7-8%. By restricting the animals fluid intake exclusively to solutions of ethanol, ingestion of ethanol in concentrations up to 20% has been reported (Richter, 1953;Mardones, 1960). Using a technique similar to the one used in the present investigation, Lester (1961)  A significant difference between the effects of ethanol and amphetamine self-ingestion is !eflected in the inconsequential left lever rate which was significantly increased by amphetamine, but not affected by ethanol. As discovered earlier, the ability to increase irrelevant behavior may be unique for amphetamine.
Responding on the right lever which produced food pellets was either unaffected or increased by ethanol indicating that the rats had not lost their motor ability and as a result continued to obtain food pellets in spite of decreased responding associated with ethanol ingestion.
The effects of oral injections of ethanol were similar to effects of ingested ethanol except in right lever pressing for food pellets. As with ingested ethanol, there was an increase in lever pressing for food pellets, but rather than the number of pellets delivered remaining constant, the number decreased. The injected ethanol apparently disrupted the rats temporal discrimination for fixed internal responding so while more responses were made, fewer food pellets were delivered.
The experimental design utilized in the present study provided a simple, inexpensive model by which to test the effects of disulfiram on ethanol consumption in the rat. After pretreating the rats with disulfiram, only a fraction of the normal amount of ethanol was consumed.
These results supported in the rat, the findings of  that ethanol ingestion was terminated in humans, after a critical amount of ethanol was consumed in the presence of c!i.sulfiram. An explanati.on for th.is effect was put forttl by  that disulfiram inhibited acetaldehyde dehydrogenase thereby resulting in an elevated acetaldehyde level in the blood. Apparently, the elevated acetaldehyde levels were toxic to the rat as in man to prevent the further intake of ethanol. A conditioned aversion to ethanol was not established in the rats since they continued to ingest ethanol on succeeding days of the disulfiram pretreatment experiment.
The applicability of the method used in this investigation to the s.tudy of morphine withdrawal was studied. During a four day withdrawal period, the ingestion of water, amphetamine and ethanol was studied.
The greatest depression in behavior was seen on the first day of withdrawal with a trend gradually returning to normal over the next three days. This behavioral effect during withdrawal is well correlated with the duration of electroencephalograph changes observed during withdrawal in morpnine dependent rats (Khazan, 1970). After discontinuation of morphine injections, a marked drop in high voltage, slow waves occurred and persisted for about three days.
In contrast to the ingestion of amphetamine, morphine rats ingested ~igh concentrations of ethanol readily during the withdrawal fro~ morphine. The mechanism for this urge to drink ethanol when morphine is not available is not known. It may be due to ethanol-induced depression of the central nervus system which is hyperexcited during morphine 'tJithdrawal (Goldstein and Goldstein, 1961). It may be also due to the conversi on of ethanol in the central nervous system to a morphine-like compound (Walsh et al, 1970).
--Nalorphine, a morphine antagonist, has been used to demonstrate morphine withdrawal in addicted animals (. Thompson and Schuster, 1964).
In addicted rats receiving morphine by lever pressing for intravenous infusion, nalorphine increased th2 rate of responding proportionally to the severity of withdrawal induced. Nalorphine in a dose of 4 mg/kg reliably abolished responding in three out of four dependent rats, but in the fourth rat a dose as high as 16 mg/kg was ineffective indicating varying degrees of susceptibility to nalorphine-induced withdrawal.
Nalorphine 1n a dose of 4 mg/kg produced no effect or a slight increase in responding in a nonaddicted rat. McMillan and Morse (1967) haye also reported increased operant responding in non-tolerant animals.
Since the mechanism of action of morphine is still unknown, it is not surprising that the mechanism of action of antagonists as nalorphine is also unclear. However, since it is generally assumed that nalorphine acts by displacing morphine from its receptor (Takemori et~' 1969), it is not surprising that this acute onset of withdrawal has a more severe ' effect on behavior than gradual onset of abstinence-induced withdrawal. (1) Oral self-ingestion by operant licking provided a reliable mea~s by which to quantitatively study self-administration of abusive drugs in the rat.
(2) Oral self-ingestion by operant licking utilized a method to study drug self-administration which was in the rats natural repertoire, easily acquired and resulted in high rates of responding.
(3} Selfpingestion of drug solutions was not limited by taste aversion except in extremely high concentrations, but rather by the pharma-GOlogical effects of the drugs. [4) The drugs studied in this investigation affected behavioral responses in a dose-dependent manner.
(5) Rats stopped ingesting amphetamine after ingesting a dose approximately equivalent to the dose of injected amphetamine which abolished responding.
(6} Multiple concurrent schedules provided the means by which to study the complex nature of drug self-administration in rats. (7) Following chronic treatment of rats with amphetamine, an increased sensitivity to ingested amphetamine occurred which was manifested by the same behavioral effects of ingested amphetamine but at considerably lower concentrations.
(9) The study with disulfiram offers an experimental model to test and study pharmacologically -induced aversion to ethanol ingestion.          Any two means not underscored by t~e same line are significantly different at P <0.05.        *Significant at P < 0.05 **Significant at P < 0.01 ***Significant. at P < 0.001        ' ""' **Significant at P < 0.01 ***Significant at P< 0.001
).     1130.97 *Significant at P <0.05 **Siqnificant at P <0.01 * * * S i g n i f i ·c a n t a t P < 0 .                                 Any two means not underscored by the same line are significantly different at P < 0.05. w ..... ""-J