Effect of Selected Drugs on Control of Morphine Withdrawal Hypothermia Induced by Morphine or Conditional Stimulus Associated with Morphine

The present investigation was undertaken to demonstrate how a conditional stimulus (cs) similar to the a ction of morphine, can increase rectal temperature during morphine abstinence. Also, the study implicates certain neurotra11E.rn1iters which are involved in the ef'f'ect of conditional stimulus and of morphine to affect rectal temperature. Rats were given two equally spaced injections of morphine sulfate daily, each injection being paired with a bell. The bell was presented fo r one minute and the injection was given during the last 15 seconds. This procedure was followed for lJ-15 days. Twenty-four hours after the last injection the bell was presented alone. The rats learned to increase their body temperature folloi.ving the presentation of the bell. This increase was specific only to animals that had the bell paired with morphine prior to challenge treatment. This change in temperature was shown to be approximately equivalent to an injection of 12.5 mg/Kg at 24 hr after the last morphine injection. When naive animals were exposed to a bell, no change in temperature was observed. Those rats which had received a random bell or no bell during addiction demonstrated no change in temperature when presented with the CS 24 hr after the last injection. iii Naloxone, a narcotic antagonist, produces hypothermia in normally addicted rats only if given within 12 hr after the last morphine injection. In contrast, when administered to CS-morphine paired animal s which received only the CS 24 hr after the las t morphine injection, naloxone caused a hypothermia. This data suggest that the CS and morphine a re working by either the same or parallel pathways in the brain. The CS induced increase in temperature was b locked dur ing withdrawal ~hen t h e animals we re pret reated wi th phenoxybenzamine (2 mg/Kg ), mecarnylamine (2.5 mg/Kg) , h a loperidol (0. 2 mg/Kg) and benztropine (0 . 625 mg/Kg ) but was not blocked by cyprohe ptidine (2 mg/Kg) . Morphine induced increas e in temperatur e ~~s blo cked by me camyJ.nmine, ph e n oxybenzamine and cyprohept idin e but was not blocked by haloperidol or bentropine. Propranolol (2 mg/Kg ) had little effec t on the increase in t emperature due to t he CS or morphine when given at 24 hr after the last CS-morphine pairing . The CS wa s not able to affect other withdrawal symptoms such as sha kes, ptosis, pi loerection, loss in body weight, or writhing when presented 24 hr after the last morphine dose. These data indicate that the increase in temperature elicited by morphine during withdrawal can be classically conditioned. Such a response required a functional autonomic and centra l nervous system.

tional stimulus and of morphine to affe ct rectal temperature.
Rats were given two equally spa ced injections of morphine sulfate daily, each injection being paired with a bell.
The bell was presented fo r one minute and the injection was given during the last 15 seconds. This procedure was followed for lJ-15 days. Twenty-four hours after the last injection the bell was presented alone.
The rats learned to increase their body temperature folloi.ving the presentation of the bell. This increase was specific only to animals that had the bell paired with morphine prior to challenge treatment. This change in temperature was shown to be approximately equivalent to an injection of 12.5 mg/Kg at 24 hr after the last morphine injection. When naive animals were exposed to a bell, no change in temperature was observed. Those rats which had received a random bell or no bell during addiction demonstrated no change in temperature when presented with the CS 24 hr after the last injection.
iii Naloxone, a narcotic antagonist, produces hypothermia in normally addicted rats only if given within 12 hr after the last morphine injection. In contrast, when administered to CS-morphine paired animal s which received only the CS 24 hr after the la s t morphine injection, naloxone caused a hypothermia.
Thi s data suggest that the CS and morphine a re working by either the same or parallel pathways in the brain.
The CS induced increase in temper at ure was b loc ked during withdrawa l ~hen t h e animals we re pret re ated wi th phenoxybenzamine (2 mg/Kg ), mecarnylamine (2.5 mg/Kg) , h a loperi dol (0. 2 mg/Kg) and benztropine (0 . 625 mg/Kg ) bu t was not blocked by cyproh e ptidine (2 mg/Kg) . Morphine induced increas e in temp eratur e ~~s blo cked by me camyJ .nmine, p h e n oxybenzamine and cyprohept idin e but was not bloc ked by haloperidol or bentropine.
Propranolol (2 mg/Kg ) had little effec t on the incre ase in t emperature due to t he CS or morphine when given at 24 hr after the last CS-morphine pairing .
The CS wa s not able to affect other withdrawal symptoms such as sha kes, ptosis, pi loerection, loss in body weight, or writhing when presented 24 hr after the last morphine dose.
These data indicate that the increase in temperature elicited by morphine during withdrawal can be classically conditioned. Such a response required a functional autonomic and cent ra l nervous system. iv

ACKNOWLEDGEMENTS
The author wishes to express hi.3 g ratitud e to his parents, Mr. and Mrs. John Drawbaugh , and to his wife's parents, Dr. and Mrs. John Locke for th eir ins p iration, thoughtfulness and fi nancial assistance d u r ing the years required for th e th es i.s .
The a u t ho r wishes t o express his gratit ude to his wife for the hardships she incurred during the yea rs requi red for the thesis .
Special thank.'" are conveyed to Dr. Harbans La l, his major p rof esso r, 1\'hose guidance dur ing these experiments and great p atie n ce wi t h the experimen~er were greatly appreciat e d.
The author a lso wi sh es to thank his fellow stud ents ,  (1972) have demonstrated tha t hyp e rthermia can be cond i ti one d during mor p hine ad mini st r a tion by p airing a n e u t r al st imu lus ( be ll) with morphi ne inj e ctions. Th is co n d i t ioning p ro cedure requir e s app r ox imatel y 24 -JO pa i rings of the bell a nd mo r ph ine (Roffman t l . §-1 . , 1972). The resu. l ting d a ta l ed to th e h yp othe s i s tha t the c o ndit i onal stimulu s , act . ing on the brain , may affect the same receptors that morphine ai'fects .
Being able to unders t a n d the conditio n ing associated 1,.;ith mo r phi n e admi n istration may be of great va l ue i n treating human add i cts . It shoul d be r eas oned the n that those b ehav i ors that a r e paired with e a ch mo rphi n e i nj e c tio n mu s t be ext ing u i sh ed along with the a c t ua l ph y s i ca l p r ocess o f drug adm inis t r ation in order to cure addiction.
Conditional r e s pons e s du e to morphine administrat ion were first seen as a salivary reflex by Collins and Tatum (1925). Shortly thereafter Dr. Krylov, of the Tashkent Bacteriological Laboratory in Petro g rad, o b s e rved that a ft e r repeated morphine injections in dogs they would vomit when the investigator entered the room, a response seen initially immediately f ollowing the morphine injection.

Wikler a nd
Pescor (1967) demonstrated furth e r, using the classical conditioning paradi g m, that the environme nt associated with abstinence can act as a conditional stimulus {cs) and can elicit withdrawal symptoms wh en the rats were placed on that environment many months af'ter t he primary abstinence period.
In addition, r ece nt e vidence indi ca t es that the pers is tence of abstinence-associated conditioning in post -morphine dependent monkeys reflects a possible mechanism for the rel apse to drug tak i ng behavior (Goldb erg and Schuster , 1970) .
The present i n vestigation sought evidence to est ablish: l) Wb.ether t:he CS acts on p a thways that a re sensitive to the action of morphine .
2 ) 'if"hether naloxone, a drug which is a pure narcotic antagonist (Blumberg and Dayton,197J), can elicit hyp othermia following the CS in 24 hr abstinence rats, therefore, supporting the hypoth e sis that the CS and morphine affect temperature by similar neuronal pathways .
J) The mechanism of a c tion of the conditional stimulus on the temperature regulatory system of the rat and its relationship to the mechanism of a ction of morp hine on the temperat ure re g ul ato ry system of the rat.

LITERATURE SURVEY
Condit ioning Associated with Narcotic Addiction Conditioning associated with narcotic addiction has been demonstrated in a number of experiments .
Utilizing the salivary conditional reflex a s a conditional respons e , Krylov (1927) Thompson and Ostlund (1965), that animals addicted and withdrawn in one environmen t will selfadrninister a narcotic drug \vh e n placed back in that environment for up to six months afte r the last day of narcotic ingestion . Goldberg and Schuster (1967, 19/0 ) utilized nalorphin.e, a morphine antagonist , to demonstrate conditi'.)ned abs t inence changes induced by nalorphi:ne in post morphine dependent monkeys. They observed that after p ai ring a neutral stimulus (li g ht) with nalorphine injections, the n e utral stimulus could, · when presented a .lone, elicit con.di tional responses (emesis, salivation and decreased heart rate).
These responses are normally only observed following the nalorphine injection in morphine dependent animals. However, they could not condition the hypothermic effect that follo ws nalorph ine administration. Goldberg et al. (1971) demonstrated that monkeys would self-administer saline, to overcome an antagonistic effect, if they previously had been given nalorphine under the same conditions. Thompson and Pickens (1969) reviewed the literature of conditioning and drug dependence through 1969. They concluded that much of drug self-administration can be explained by operant behavior. Antecedent conditions (Kolb, 1962),current stimulus circumstances (Cofer and Appley, 1964), qualitative and quantitative propert ies of the reinforcing drug, as well as stimuli associated with dru g administration (Ausubel, 1964;Weeks a nd Collins, 1964), all have the ability to act as variables t ha t do affec t drugreinforc e d response .
They concluded, finally , that drug dependence can be analyzed using the op e r an t pa radi gm and thus p rovide answers to ~he underlying mechanisms of drug dependence . Beach (1937) reported that env ironme nt al stimuli acted as a secondary reinforcer ir~. morphine dependent r ar. s.
A similar experiment was pub lish ed by Wikler and Pescor (1967). Beach's experiment was intended to change the environment by giving the rats a choice of either the ori g inal environment or a ne w one instead of placing them into their original environment, as was done by Wikl er and Pescor. The animals preferred the environment in wh ich they experienced addiction and withdrawal to the unfamiliar neutral ones.
Thus, it was concluded that rats would, when abstinent, show a preference for distinctive environments which h ad previously been repeatedly associated wi th rel ief of withdrawal symptoms (Kumar, 1972). It was further concluded that environmental stimuli can become secondary reinforcers after repeated pairing with the effects of morphine and that the learning involved may contribute to the maintenance of dependent behavior.
Utilizing a self-administration technique Kumar and Stolerman (1972) showed that animals given morphine in their drinking water would drink large amounts of qui.nine following cessation of morphine in the water source. They coneluded that the bitter taste of qui.nine alone was the reason for the large in ta ke and they further concluded that t aste had become a secondary reinforcer.
Utili z ing both c lassica l and operant p a radigms Crovder et al . (1972) showed that animals given morphine injections p a ired with a buzzer will bar press for the buzze r and a saline i n fu sion . Th ey concluded that t he buzzer a n d the saline injection had acquired secondary reinforcing properties. It was furthe r concluded that a stimulus can become a secondary reinforcer without being a discriminative stimulus for an operant.
Utilizing state-dependent learning Hill et al. (1971) and Rosecrans et al. (1973) showed that rats could discriminate drug (morphine) and non-drug {saline) states.
Hill's group concluded that when an addict takes an injection he is not only attempting to regain the initial unconditioned effects of the drug, but also to reinstate some of the learned or reinforcing experiences which can only occur in the drug condition.
The Rosecrans group did not attempt to explain their results in terms of practical importance, but rather they explained thei.r results in biochemical terms which will be discussed belo w .
Utilizing a cl a ssical paradi g m, Roffman ~ al. (1 972 , 1973) p a ired a bell with morphine in jec tions. The neutra l bell eventua l ly a c q uir e d prop ertie s o f a conditiona lstimulus, s imila r t o mo r p hine, wh ich wa s sho wn to prevent wi t hdrawa l hypo the r mi a dur i n g the 72 h r pe r iod fo llo wi n g t he .las t mo rphine in je ction. The y c o n c l ud e d th at to d emonst r ate that a condit i onal stimulus c a n block one wi t h dr awa l s y mp t o m would be to pa r a llel the r it ual that h u man addi cts fo ll ow to postpo n e the onset of wi th drawal . A human a dd ie t foll o ws a set pattern wh en he a dministers the drug , a nd i f th e drug is not ava i lab l e t he ritual alone (c onditiona l s t imulus) c a n postpone ab s t i ne n ce (Weidman a n d Fellne r, 1 971) .
Conditi oni ng associated wi t h morphine i ngestion ca n thus be demons tra ted by the use of Pavlov's cl as sical co ndition techni q ues. Also, the conditioning c a n b e demonstrated by using an operant conditioning procedure or by combining both classical and operant p rocedures. Yet another way tha t conditioning as sociated with morphine h a s been observed is by using the state-dependent learning paradigm.
Neurochemic al S vs tems Inv olve d in T e mp e r a tur e Re gulation Feldberg and Myers (1963) postulated that body temperature is regulated by the balance of three mo noamines (5-HT serotonin, DA dopamine, and NE norepinephrine) in the anterio r hypothalamus. This hypothes i s was based on experiments in which serotonin or norepinephrine was administered intravent ricul arly and their effect on temperature reco rded .
Serotonin caused hypertherrnia and norepinephrine caused hypothermia in the cat.
In the rat similar evidence has been observed using serotonin and norepinephrine (Feldberg and Lotti, 1967;Breese and Howard , 1971). Besides these amines, dopamine (Kruk , 1972) and acetylcholine (Lomax et al., 1969) might also be involved in temperature regulation.
Utilizing the intraventricular injection t e chni que, Jacob and P eindaris (197J) acl.minist er ed injections of' serotonin to r abbits and observed, like Feldberg and Myers, an increase in body temperature. However, if the a nimals were pretreated with cyproheptidine (antiserotonin drug), the increase in temperature due to serotonin was antagonized.
Jacob and Peindaris also injected NE intraventricularly and observed a n increase in temperature (contrary results to those of Feldberg and Myers). When phenoxybenzamine was given one hr before the norepinephrine, the hyperthermia due to norepinephrine was antagonized. Propra nolol (B adrenergic blocker) did not alter NE hyperthermia in rabbits. Chlorpromazine (a phenothiazine, antipsychotic) drug known to a ntagonize dopamine, caused hypothermia by itself. When norepinephrine and serotonin followed chlorpromazine no change in the norepinephrine hyperthermia was observed and a very slight increase was noted in the serotonin treated animals.
Recent studies have indicated that cholinergic mechanisms in the hypothalamus may be involved in the centra l control of body temperature.
Although the levels of acetylcholine (ACh) in the hypothalamus are relatively low when compared with the monoamines, the enzymes for ACh 's synthesis and degradation are also present, suggesting that acetylcholine could fulfil l a neurotransmitter role in this particular brain region (Hall,197J). However, the role of acetylcholine on temp erature r egulation in the r a t is s t ill questionable (Myers, 1969). ~Jany factors, such as route of administration, amount of s~bstance, and environmental temperature, could account for discrepancies in whether or not acetylcholine directly affects the regulatory system.
Nicotine has been shown to cause a rise in temperature, and if mecamylamine is given before nicotine the rise in temperature is blocked (Lomax and Kirkpatrick, 1969). The main conclusion from this study was that nicotine somehow changes the hypothalamic set point. Thus, nicotinic receptors play some role in the hypothalamic cholinergic thermoregulatory system.

Thermoregulatory Behavior
Homeotherms regulate their body temperature by 1) physiological or autonomic responses mediated by way of the sympathetic nervous system, and 2) behaviorial means involv-ing coordinated and voluntary motor activity. There have been long discussions concerning the terms "autonomic" (physiological) versus "behavioral" thermoregulation since behavior can also be considered physiological . Th ese terms are readily accepted and Ca banac (1972) has sugge s ted that one speak of thermore g ulatory behavior and thermoregul ato ry ph·ysiolo g ic al resp onses in place of t h e more amb i g uous terminology as behavioral thermore g u l ation and phys io logical thermoreg~lation .
Ge n e rall y speaking , a n anima l in his natural environment compensates for fluctuations in temperature simply by movi n g to a warmer or cooler place (Ri chards, 1974) . This movement of the organism to a more desirable therma l env ironment can be called by defin i t ion, thermoregulatory behavior (Hensel, 197J) . The organism contro ls heat ga in and heat loss b y cha nging the physical chara cteris tics of h is environment by behavior such as avoidance, huddling , n estling, or putting on clothing such as is the case with man.
Only recently has there been an increasing appreciation of the fact that, wh en given freedom to choose, homeotherms generally rely more on thermore g ulatory behavior than on thermoregulatory physiological responses to alter their body temperature (Richards, 1974).
How are these responses motivated? It is safe to assume that organisms are motivated by states of "pleasantness" (comfort) a.nd "unpleasantness" (discomfort). There The desired outcome of the system is relativ•~ c onstancy of body te mperature . Th e system also may represent autonomic regulation deriving internal energy rather than external as this diagram represents.
is evidence that consciousness plays a big role in determining what state is desirable to be comfortable. Corbit (1969) and Adair, et a=b_ . (1970) have shown that the rat and monkey will behaviorally control. their environmental te1 rrp era ture whe n their preop ti c-anterior hypothalamic area are thermally stimul..ated .
Th e regulati o n is shm~n schematically in figure 1 .
Tb.is g ener a l diagram repres ent s a modified version of Stolwi.jk and Hardy 's (1966) view of the behavioral system.
This t y·pe ot~ d.iagram can and i s used to explain both Corbit' s a nd Adair's data . Thi s d iagram or one very similar has been used by many phys iolo gists working in the area of thermoregul a t :Lon. The terms such as re fer ence input elements, cont rol.l ing e lements , feedback elemer.t:::; 1 Ed: c . 5 '.nay seem rather general but thi.s f"ield has grown so rapidly in recent years that organi zation of the data c a n bes t be explained us ing these terms in an eng ineering concept of control systems.
Simply, the information has come at one time and no one has been able to synthesize all the ideas and propose a system identif ying specific brain areas as to their exact function within the thermoregulatory system.

EXPERIMENTAL
( 1  (Table 1). The rats were maintained at this dose for 2-4 days and then withdrawn.
The procedure for injection during the morning session was as follows: Each animal was taken out of its home cage (one animal injected at a time), placed in a plastic container and taken to a sound attenuated and temperature controlled room (21°C ± 0.5) 40 feet from the room where the animals were housed. Immediately after entering the chamber the animal was removed from the plastic container and placed i nto a single-pan balance to be weighed and then returned to the plastic cont ai ner. The bell was turned on and after 45 seconds the animal was picked up and securely held, one hind leg a nd the head, so as to prevent the animal from movement and the injection was g iven. The n the animal was again returned to the plastic container, and after a total of 60 seconds had elapsed, the bell was turned off.
The rat was then im.11ediately returne d to his individual c age . Each day the order of animals going through this procedure wa s changed .
The identical proc e dure 1i as follo we d d uring the evening session with the excepti_,n that the body wei ght was not taken at that time. (4) Testing Procedure The test procedure for the experiments using mecamylamine, phenoxybenzamine, propranolol , haloperidol,benztropine and cyproheptidine to eval uate their control of morphine withdrawal hypothermia was the following: 1. The same animals were used throughout each experiment.

2.
Temperatures of each animal were taken 10 min prior to and JO min after their last morphine injection.
J. Temperatures of all animals we re again taken 4. The animals were then divided into two groups, those receiving morphine and those receiving the bell. JO min after the test treatment the temperatures were taken again.
Temperature Measurements All temperature measurements were taken at designated times u sing a digital thermistor thermometer (Di gitee Model 8500-2 by United Systems Corpora tion, Dayton, The rec ta l probe (Model 402 , Yellow Springs Instrument Co ., Maryland) wa s inserted fi v e cm (Myers, 1973)   Piloerection -This symptom was observed when the rat's fur stands out from the body. The occurrence or absence was measured after the animal had time to groom following placement into the cage. This was done so as not to report raised fur that might have resulted from handling.
Changes in body weight and temperature were measured just prior to placing the animals in the observation cages.
All of the measurements were made at O, 24, 48, 72 hours following the last morphine j _njection.
vations were always made in the morning.

(7)
Statistics These obser- The Student's "t" test was used to determine the si g ni ficance of a difference between two correlated means (i.e., pre -challenge and post -challenge temperatures). The two temperatures, pre and post, were recorded for each individual rat and the column desig11ated 11 change" 1vas arriv ed at 1 JY subtracting the post-challenge temperature and the pre-challenge temperature of each animal . The specificity of the bell's effect on rectal ternperature is summarized in Table 4. The bell had no effect on (1) animals that had never received the drug, (2) on morphine-addicted animals which received a random bell   during addiction, and (J) on morphine-addicted animals naive to the bell. The only rats whose rectal temperatures were affected (increase) by the bell alone were the animals who received morphine and bell paired throughout addiction. Table 5 shows that the bell, when presented 24 hr after the last morphine injection, causes an incre ase in rectal temperature but this increase is not attenuated by add i tio:ial presentations at JO minute intervals after the initial presentation .
The three presentations at 24 .5, 25 and 25.5 hours were for only 10 seconds ; only the first p resentation was for one minute.
Different doses o f morph ine were given 2 L~ hr after the last morphine in j ectioa as can be seen in Table 6. As the dose increased, the effect on recta l temperature increased until 25 mg/kg was given .
No difference in chang e o:f rec tal temperature existed bet-ween 25 mg/kg and 1 00 mg/ kg doses. The dose of 12.5 mg/kg was observed to be similar in magnitude to the increase in rectal temperature following the bell wh en presented to conditioned animals.
Data presented in Table 7 show the effect of one dose (100 mg/kg) of morphine over a period of 48 hr. The temperature reached a maximum at JO minutes after the intraperitoneal injection. This temperature was still high two hr after the injection. These data were used to determine the appropriate time at wh ich the temperature should be recorded following the morphine injection or the presentation of the bell.  This information suggests that the bell and morphine were acting on either a sin gle or parallel pathways \~hich meet at some po i nt eliciting the same effect .
To :further substantiate the similarit y of physiolo g ical mechanisms (bell a nd morph ine) the bell or morphine was given following b e ll a nd nalo x one (2 mg/k g) . Neither the bell nor mor phine could reverse the effect of the a n-  3 JJiimals for these groups had received Bel l + Nalo x one bef'ore either the bell agai: n or morphine.
5Re£er to Legend 4 of T a ble J. 9 aefer to Legend 5 of Table 4.
Not e : Saline + Bell and Bell + Naloxone are th e sa me anima ls. The Bell and Morphine gro ups were d e rived i'rorn the 6 animals of Bel l + Naloxone. If both 0( and ~ adrenergic blockers were involved in the hyperth ermic response due to morphine or the CS, then bo th propra nolol and phenoxybenzamine would be nece ssa ry to prevent the incre a se in temperature . Data prese nte d in Table 10 indic ate that mecamylamine ( 2 . 5 mg / kg ) pretreated a nimals (one hour) do not sh o·w any incre ase in temperature due to e i ther treatment by the beJl or a morphine inj ection. Giving rnecamylamine alo n e does not change the t emperat ur e o f 24 hr abstinence r ats .
Morphine alone increa s ed the temperature two degre es (Table   2), and the bell alone at 24 hr o f abstinence increased the temperature by almost one degree (      If more than one transmitter (dopamine, acetylcholine, or serotonin) is involved, then two or more of the compounds might be required to prevent the hyperthenric ef'fect due to the CS or morphine.
Data summarized in Table 13      Data for cyproheptidine (2 mg/kg and I+ mg/kg) pretreated animals (45 min) summarized in Table 15 indicate that following cyproheptidine, the increase in temperature due to the bell was not bloc ked to the same extent as the increase in tempera ture due to morphine wa s blocked.  morphine alone during addiction received either the bell or nothing following the same procedure mentioned above. Table 16 showed that the bell did not significantly affect the withdra1v-al symptoms, at any   Piloerection 25-28 8-10 8-9 10-10 28-28 10-10 9-9 10-10 28-28 10-10 9 -9 10-10 8 8Number of a nimals showing s ymp t om out 0£ t o t a l nu~1or observed in each group . 9Refer to Legend J of Ta bl e 2.
4 Re fers to lo s s from zero tim e .
5Bell present e d at 12 a nd 2J} hr.
6 " II II " J 6 a n d 471 hr .

DISCUSSION
This study is especially significant in that it demonstrates that the conditional stimulus a nd morphine affect the body temperature in morphine addicted rats through a similar neurophysiol ogical pathway initiated by different neurosubstances.
In this study, rats g iven a conditional stiwulus paired with morphine, whe n give n the conditional stimulus alone during 1vithdrawal, exhibited an increase in temperature analogous to the effect of morph ine.
The conditional stimulus was found not to change a ny of these symptoms. Therefore, under the present addiction schedule the change in temperature was the only conditional withdrawal symptom that was measured. However, this does not mean that temperature is the only conditionable withdrawal symptom. Rather, this system was most easily conditioned and by possibly varying the ~onditioning procedures it might be possible to alleviate the severity of other withdrawal symptoms.
The following discussion will cover three areas. The first part will include the evidence establishing the a .bil-.
ity of the CS to cause a rise in temperature during withdrawal, similar to the change seen following a morphine in-44 jection. The second part will deal with the physiological mechanisms involved in mediating temperature changes following the CS and morphine. The last part will deal with the significance of these findings.
The addiction s che d ule was modified from a previous experiment by Roffman t l al . (197J), so t hat instead of four injections per day, only two injections were administered.
The terminal dose (2 00 mg/Kg) was still reached in 1 0 days as in the experiment by Roffman~ al. (197J) . The rationale for the reduction of injections each day was the hope that the CS wil l be more effective \\' "hen g iv e at 12 hr intervals as the a nimal will be more g reatly motivated to relieve withdrawal symptoms, at each injection, unlike the erratic motivational state of the rat in the other exper iment .
Some other withdrawal symptoms, as p re v iously mentioned, were observed at 24 hr after the last morphine or CSmorphine injection. Thirty minutes prior to this measurement the bell was presented to some animals from both groups.
As was previously stated the bell affected only the temperature. This may be due to the inadequate number of pairings, as other experimenters (Wikler and Pescor, 1966;Kumar, 1972) had a minimum of 45 pairings in conditioning experiments, while in the present experiment the maximum pairings was JO.
Another possible reason why the bell did not affect the other withdrawal symptoms may be due to the temporal pairing of the bell and the injection. It might be necess ary to present the bell for a longer period of time after the injection or increase the duration of bell presentation. This would insure that the onset of drug act ion would definitely occur during the presentation of the CS. Also, other stimuli (i.e., visual-strobe light or gustatory-1% saccharine) may be found to be effective in either reduci n g or eliminating wi thdr2,wal symptoms , along with temperature. Another possibility is that more than one stimulu~ may be needed to control specific :::.yrnptoms of the withdrm\-al syndrome.
All of these po~sibilities must be considered in order to realistically evaluate the effect of 8nvirorunental cues on drugt aking behavior in rats . It is ar~ accepted fact that humans go through many rituals (Wikler, 1971) before and during drug administration and that parts of these rituals become conditional stimuli. Theref'ore, it seems prob ab le that animals :receiving morphine c a n be conditioned by different cues either separately or simultaneously. It is just a matter of selecting relevant cues to be paired with the drug administration.
Many investigators measure the rectal temperature one hour following morphine administration (Lotti, et al ., 1969;Gunne, 1960 andMartin et al., 1963). However, each investigator had his own particular addiction schedule and it was thought that since the present schedule was not similar to any of the above, a time to measure temperature following morphine administration should be experimentally determined. Therefore, a dose of 100 mg/Kg was administered to addicted rats and maximum hyperthermia occurred JO minutes following the injection. Thus the time for all temperatures to be taken was JO minutes after each treatment (cs, morphine or CS-morphine).
The effect on tempe rature by morphine was observed not to change at 08JO and 20JO. This ±'actor is important in that diurnal rhythms may have caused the animal t o behaviorally perceive or phys iol ogica lly react differently to the inj ect ion in t he morni n g as compared to the injection at ni g ht. This can be reasoned by the fact that the analges ic effects of morphine are different in the morning as In any case, since the temperature change is the same at the two time periods (08JO and 20JO) it is at least safe to assume that physiologically morphine is a:f:fecting the thermoregulatory center in a similar manner.
Only by experimentation will the behavioral factors be determined as being no different at the 08JO and 20JO t imes.

Control of Morphine-Withdrawal Hypothermia by a Conditional
Stimulus

1+8
The following section c ontains evidence that a conditional stimu l us can, like morphine elicit a rise in temperature d uring i;,.rithdrawal in animals addicted to morphine-CS.

1.
In the presence of the CS, 24 hr after the last morphine-CS pairing, the rats showed a significant increase in rectal temperature. But if no CS was prese nted the conditioned anima_ls exhibi ted no change in temperature at 24 hr of withdrawal.
If morphine was administered the typical increase in tempe ratur e was observed. Also, if the morphine-CS was given at 24 hr withdrawal the usual increase in temperature was observed.

2.
In the presence of the CS, 24 hr a:fter the last morphine injection, the rats showed no change in rectal temperature.
Also, presenting the bell to animals who received the bell randomly throughout addiction, produced no ef:fect on rectal temperature. Thus, the bell acted as a CS only when paired with morphine during the addiction phase. The CS did not, however, cause the same change in temperature as morphine (100 mg/kg) when given 24 hr after the last CS-morphine pairing.
Instead it was approximately equivalent to 12.5 mg/kg of morphine in its effect on a withdrawal animal's temperature. Further, the time of the presentation was found to be only 10 sec in duration to cause the increase in temperature.
And if given at JO min intervals after the initial increase in temperature due to the CS, no cumulative or additional changes were observed.
The neut ral stimulus has acquired cond itional properties .

2.
The magn~tude of the bel l wi th respect to its ability to change the temperature is not as strong as the tenninal dose of mor phi n e it wa s paire d with.
J. The multiple CS presentations did not produce any cumulative ef'fect when given successively at 24 hr withdraKal.
These conclusions should not be interpreted as claiming that the conditional bell equals 12.5 mg/kg because statement three clearly shows that not to be t rue.
If it were equal to 12.5 mg/kg morphine, then continued presentations should cause an increase in rectal temperature that would equal the 100 mg/kg dose of morphine. Also, a more important physiological factor must be considered. Is it desirable for the organism to increase its temperature to a hyperthermic state?
First, a consideration must be made as to the status of the homeostatic mechanisms of the thermoregulatory system. That is to say, does chronic morphine change the "set point" of the thermoregulatory system and therefore change so the temperature at which the organism now calls normal. It has been postulated that such a situation d o es occur (Lotti et al., 1965) where the "set point" chang es. It is difficult to assess which way it might g o, but since much of the withdrawn animal's day is spent in a hyp othermic state ( p res ent experiment) it should be safe to a ssume that his " s et point" Even if the "set point" of the thermoregulatory system does not change , the withd rawn animals may just raise their tempera ture to a comfortable level. Simply, they possess a range (i.e., J7 . 7 -J8.l) at which they find body comfort. It is known that organisms strive t6 maintain a state of comfort (Hardy ~t a l., 1971).
Since the thermoregulatory system is easier to change than other body systems (Richard, 1973), and trying to condition it seems not to be an exception because it is behaviorally regulated, the above reasoning appears logical. This also explains the lack of ability of the CS to cause a cumulative effect by repeated administrations. The rat has reached a comfortable state thus behaviorally he is not motivated to r a i se his temperature any more a nd thus he does not.

Physiological Mechanisms Involved in Mediating Temperature
Changes Following t l:.e CS and Morphine This section contains evidence t:t:.at the e:ffect of the CS and that of' morphine on temperature is mediated by different tran~mitters, but that common paths may exist in the thermor egula tory neural net.
Tr1e use of each compound used to ana lyze the experiment will be discussed separately .

1.
In the presence of mecamylamine neither the CS nor morphine \,'as able to increase the rats' temperature . The dose was determined by the criterion that it by itse lf did not affect the temperature . Those data support the idea that the autonomic nervous system was involved in media ting the temperature changes follo',ving the CS or morphine. This block by mecamylamine was at efferent ganglia, thus preventing any communic ation between the thermoregul atory center and peripheral mechanisms (i.e., adipose tissue, blood vessels) which wo uld create an increase in temperature or a pyrogenic effect.

2.
In the presence of propranolol or phenoxybenzamine the CS and morphine's effect on rectal temperature were unaffected by the former and blocked by the latter. These data suggest that J3 receptors are not involved in the mediating temperature changes that follow the CS or morphine.
Centrally ~-receptors have been shown to play little if any role in thermoregulation (Rudy and Wolf, 1971). Also, peripherally the role of fi receptors within the mechanisms involved in temperature changes are limited to causing a decrease in body temperature and increasing lipolysis to increase heat production (steiner, 197 3) • This latter use of' fa-receptors would not fit because the rise in temperature by t he CS and morphine occurs t oo q uickly and since the p receptors have been blocked, a reduction was observed.
The dose of pro pranolol was determined by i ts inability to change temperature by itself and behaviorally it has been shown not to cause any c hanges in ac~ivity (Weinstock and Speiser, 1974), at the dose used which may indirectly alter the temperature.
The pretreatment time of l hr was used as the peak tissue levels seen in the rat observed at between 45 min and 75 min (Hayes and Cooper, 1971).
The selection of propranolol as a J3-blocker may not have been the best choice. This compound is distributed both centrally and peripherally, therefore i~ effects carmot be localized as with a compound such as practolol which works exclusively centrally (Wong & Schreiber, 1972). Since few )3-receptors, if any, are involved in central thermoregulatory processes this problem is not that critical.
Phenoxybenzamine, the blocker which did block the CS and morphine's effect on temperature was also not the best drug to be used in this kind of study. The use of phenoxybenzamine is widespread, but other more specific 0( blockers (phentolamine) exist and would allow for easier interpretation of data (Goldstein & 53 Munoz, 1961). In this experiment phenoxybenzamine c ompletely blocked the CS a n d allowed morphine to raise temperature only slightly. If the dose was slightly raised the co mplete block probably would have resulted.
The dose of pheno xybenzamine used has been previously sho·wn to block electroencephalogen and blood pressure changes that may result from a stimulation of brain receptors (Goldstein and Munoz, 1961) . The pretreatment time used has been shown to be the optimal time for blocking NA effects on temperatu re (Jacob and Peindaris, 1973). 4. In the presence of benztropine, morphine's hyperthermic effect was not blocked, but the bell's effect on rectal temperature was blocked.
Since the drug is a centrally acting anticholingergic, it was deduced that ACh was involved in mediating the CS hyperthermia in the brain.
The pretreatment time was determined by Puri ~ a l. (1973) as having optimal biochemical effects. Also, because the dose used produced no effect on temperature, it was decided to use this dose.
5. In the presence of cyproheptidine, morphine's effect on temperature was partially blocked but the bell's effect was only slightly reduced. 6. Finally, naloxone affected the CS and morphine in the same way. Naloxone caused a large drop in temperature following the CS or morphine, 24 hr after the last morphine injection. When naloxone is given alone 24 hr after the last morphine injection, only a small drop in temperature is noted. These data suggest that an interaction has occurred between & narcotic antagonist and the learned conditional effects of morphine (Drawbaugh and Lal, 1974).
A working explanation of the above data is presented graphically in Figure  There is little agreement among physiologists on how the thermoregulatory system works; however, they do agree upon the center or controlling system and that feedback loops e x ist.
Morphine affects the reference input elements by means of a transmitter substance. This substance in turn affects a "receptor" which has the ability to be both excited and inhibited. It is this receptor that has many arms to different elements which are labeled controlling elements.
These elements affect vasomotor activity, shivering, sweat and panting, which are located under the heading of controlled system. It is at this point where at least one feedback loop exists which returns to the receptor and inhibits it. This inhibition results in a dr opp ing of temperature in the case of morphine , i.e. as the drug is metabo- lized the e f'~" ect on the "receptor'! by the reference input drop and the :feedback. loop begins to affect the receptor and the temperature begins to fall . (Inhibitio n refers to the abiiity of the system to compensate fo r the inc r ease in temperature due -co morphine and does not necessaTily mean that the receptor is turned off.) Behavioral stimuli work the same way as they are able to affect the reference input elements by specific neurotransmitters. In this study morphine and CS, by different neurotransmitters, affect the reference input elements which in turn cause stimulation of the "receptor." By stimulating the receptor (this does not mean that only one receptor exists), an increase in tempe~ature or hyperthermia exists. The abilj_ty of the CS to eventually raise the temperature by itself may be t8rmed thermal motivation (conscious experience) (Corbit, 1973).
This increase in temperature causes thermal comfort, but to rise to morphine's hyperthermic level would cause discomfort and not be desirable, thus explaining why the CS causes an increase which is considerably less than morphine. Also, because this is a behavioral change it is transient allowing the feedback loop to again affect the "receptor" and the The practical imp ortance of' this finding is related to the use of narco t ic antagonists in the therapy of narcotic addiction. The current rationale behind the use of narcotic antag onists in the treatment of heroin a ddicts is that treatment with these drugs will result in the extinction of heroin consumption because of the blockade of the "high" sought from agonistic effects of illicit heroin.

6J
The present data suggest that narcotic.: antagonists may also be valuable in extingui shing heroin habit associated with the conditional placebo effects of her oin-seeking behavior.
These t::f.fects have been considered to be major factors in the relapse of the addiction (Wikler, 1971), and it is therefore imperative to investigate the site and mechanism of thi.s conditionr:cl ::::ie.havior to perhaps arri-..re at some efficacious method of treatment of' addict ion.
In conjunction with tLis, physiologists are int erested in determining the site a nd mechanisms of' drugs. This interest coincided with the aims of this experiment in the study of the effects of morphine and the CS on temperature in the rat. Temperature changes due to morphine apparently result from a direct action upon thermoregulatory centers within the anterior hypothalamus. Some evidence in support of this view is found in investigations in which rectal temperatures were recorded following microinjections of morphine into various regions of the hypothalamus and surrounding brain areas (Lotti et al., 1965).
The approach taken in this experiment to differentiate the neural pathways used by the CS and morphine is rather unique.
There have been few attempts, up to this time, to determine the neural pathways used by morphine to affect temperature.

CONCLUSIONS
1) The condi tiona l stimulus may evoke activity in the brain pathways which a re specifically sensitive to the actions of morphine.
2) _,The condi_tional stimulus and morphine probably utili ze a peripheral mechanism involving ACh and also receptors in the s~npathetic nervous system. J) Centrally the conditional stimulus acts by means of a dopaminergic pathway .

4) Centrally morphine acts by means of a seratinergic
pa thway in altering body temperature.

5)
The CS and morphine have a common path, however, they converge at this pathway by different routes.