This is a 42 U.S.C. § 1983 action brought by a prisoner housed at the Blanch Correctional Institution, a facility of the North Carolina Department of Correction, against various employees and prison officials of the Department of Correction, to recover damages for alleged cruel and unusual punishment, suffered by him when he was “maced” in his cell by Lieutenant West of the prison staff.1 A jury trial was had. At the conclusion of the testimony, the Magistrate, to whom the action had been committed by agreement, submitted the cause to the jury on, so far as the appellant was concerned, three interrogatories. These interrogatories, as statements of the issues to be resolved by the jury, were apparently agreed to by the parties and were explained to the jury in the Magistrate’s instructions.

The liability of the defendant West, as set forth in the plaintiff’s complaint, was based on an alleged violation of the constitutional prohibition against “cruel and unusual” treatment arising out of his macing by West while confined in his prison cell. Without objection, the cause was submitted to the jury. In his jury instructions on the standard of “cruel and unusual punishment” to be applied by them in resolving the issue of liability in this case, the Magistrate charged the jury that the use of mace “by and of itself does not establish excessive and unjustified force” violative of a prisoner’s constitutional right to be free of “cruel and unusual punishment,” that “the reasonableness of its use (in the prison context) depends on the surrounding circumstances,” that it “may be reasonably used to quell disorders and to compel obedience but not to punish a prisoner,” and that, in determining whether the use was reasonable under the circumstances of the case, the jury should consider “one, the need for the application of the force; two, the relationship between the need for force and the amount of force used; three, the extent of injury inflicted; and four, whether the force was applied in good faith effort to maintain or restore discipline or was used maliciously or sadistically for the very purpose of causing harm.” He, also, instructed the jury that, even if the defendant West had “unjustifiably used ... excessive force against Plaintiff,” he would not be liable if he “acted in good faith and with a reasonable belief in the lawfulness of his ... actions” under all the circumstances but the burden of establishing such “reasonable belief” was an affirmative defense of which the defendant had the burden of proof. This, the Magistrate told the jury, was known as a “good faith defense,” the elements of which were “first, that the particular Defendant reasonably believed that he was acting within his lawful authority, that is, that he reasonably believed that using, authorizing, or acquiescing in the use of mace was justified under the circumstances; and second, that the particular Defendant acted in good faith on the basis of his reasonable belief.”

After these statements of the substantive issues of liability, the Magistrate then advised the jury on the form in which they should express their verdict. For this purpose, he submitted two interrogatories relating to the possible substantive liability of the defendant West. The first interrogatory on liability of the defendant West (interrogatory number 2) called for an answer to the question whether West had *966“violate[d] the plaintiffs constitutional right to be free from cruel and unusual punishment by the use of excessive and unreasonable force against him,” in the use of mace on the occasion in question. The jury was instructed that if their answer to this interrogatory was “Yes,” then they were to go on to interrogatory number 4, which, as it related to the defendant West, involved what the Magistrate denominated as the defendant’s “good-faith” defense. This interrogatory inquired of the jury whether, if the defendant West had “unjustifiably used, ... excessive force against ... plaintiff,” had he (the defendant West) “act[ed] in good faith and with a reasonable belief in the lawfulness of his actions.” The jury was directed that if they found from the evidence that the defendant West had so acted, then they “should answer issue numb r four [which was the issue just quoted], ‘Yes.’ ” He added that, “You [referring to the jury] should not go on to consider issue number five [which was the damages part of the interrogatories] unless you have answered ‘no’ as to” the defendant West in response to interrogatory four. The jury followed these instructions. They answered both the second and fourth interrogatories addressed to them in connection with the defendant West’s possible liability “Yes.” It necessarily followed, therefore, that under the Magistrate’s instructions, the jury was not to award the plaintiff damages and such was the jury’s response. Judgment was accordingly entered in favor of the defendant West and this appeal by the plaintiff followed.

The plaintiff on appeal argues that the Magistrate erred in submitting to the jury the issue of good-faith in the use of mace under the circumstances of this ease, and in admitting evidence in support of such defense. He, also, argues that if the good faith of the defendant West were in issue, the Magistrate should have charged the jury that the good faith defense requires the establishment not of a subjective belief in the lawfulness or reasonableness of an officer’s conduct, but of an objective standard of such reasonable belief.

The difficulty with this argument of the plaintiff is that it was not properly raised at trial. The interrogatories submitted to the jury were, it seems clear from the colloquy between the Magistrate and counsel during discussion of the jury instructions, agreed upon by the parties. Certainly, when the interrogatories were stated and explained by the Magistrate in his instructions to the jury, there was no objection to the submission of the cause in the form of such interrogatories or to the propriety of the interrogatories as proper statements of the issues to be resolved in the case by the jury by their verdict. Under those unobjected to instructions, the jury was, as we have seen, directed to proceed no further (in effect, to make no award of damages in favor of plaintiff) if the answer to the so-called “good faith” interrogatory was “Yes.” Under these circumstances, the plaintiff is in no position to assert error in the instructions or in the ruling of the Magistrate on the interrogatories submitted by him to the jury, the answers to which were to be dispositive of the plaintiff’s claim. Though we consider the affirmance of the judgment below proper for the reasons already stated, the dissent would reverse. It would appear that, since the dissent raises issues not addressed by us in our exposition of our reasons for affirming the judgment below, we should comment on the reasoning of the dissent for its contrary conclusion.

The reasoning on which the dissent would reverse the judgment below is in the form of a syllogism consisting of three interdependent legal premises or conclusions. The first and major premise in this syllogistic exposition is that macing a prison inmate in his cell is “clearly prohibited” by the federal law in this circuit and is accordingly per se unconstitutional. Because it finds that this constitutional principle was “clearly” established in the federal law of this circuit at the time this action arose, the dissent postulates that the defendant “should have known that the use of tear gas under those conditions constituted cruel and unusual punishment” and, since he did or should have known such *967fact, the dissent concludes that “as a matter of law a good faith defense was not available to the defendant.”2 Having reached this point, the dissent would find that the giving of a good faith immunity instruction, though not objected to at trial, the “failure to object [did] not preclude appellate review [and reversal because] the instruction ‘plainly misstate[d] fundamentally controlling substantive’ principles.” It is thus evident that the dissent’s reasoning rests entirely on the correctness of its initial major premise, that macing by the defendant of the plaintiff while the latter was locked in his cell was a constitutional violation under the clearly established law of this circuit. We accordingly address this fundamental and central thesis of the dissent.

In support of its central thesis, the dissent cites two cases, only one of which is a decision of this court, Landman v. Peyton, 370 F.2d 135, 138 (4th Cir. 1966), cert. denied, 385 U.S. 881, 87 S.Ct. 168, 17 L.Ed.2d 108, 392 U.S. 939, 88 S.Ct. 2315, 20 L.Ed.2d 1399 (1968), rehearing denied, 393 U.S. 900, 89 S.Ct. 77, 21 L.Ed.2d 194. Peyton, however, is not cited, as we read the dissent, for the proposition that macing by a prison official of a prison inmate while locked in his cell is per se unconstitutional; the dissent summarizes the decision as merely “admonishing state prison officials] that unnecessary use of tear gas can violate the Constitution.” Actually, though, Peyton is cited in other decisions in this circuit3 and generally in other circuits,4 as authority for the use of tear gas or mace by prison officials for disciplinary purposes, and this construction of the decision seems the one warranted by both the facts and the result in Peyton, which affirmed the use of tear gas in that case. The defendant prison officials in that case were alleged in plaintiff’s complaint to have “used tear gas indiscriminately as punishment against the prison inmates.” The district court had found at trial “that tear gas was used approximately 12 to 15 times in the course of a year ...” against inmates. The prison regulations, as testified to by the defendant Peyton provided the prison guards with “blanket authorization to use tear gas against a ‘recalcitrant’ inmate, in addition to employing it in its customary role to prevent riots and escapes.” 5 On the basis of this regulation and the record, the district court “concluded that this [i.e., the use of tear gas] was done in a legitimate exercise of disciplinary authority.” This court, on appeal of the district court’s decision, noted the prison regulation authorizing use of tear gas on “recalcitrant” inmates and contrasted it with the less permissible federal prison regulations. However, after this comparison, we affirmed the decision of the district court sustaining the use of tear gas in the prison environment against “recalcitrant” inmates, saying (370 F.2d at 138, n. 2):

“Superintendent Peyton testified that the guards have blanket authorization to use tear gas against a ‘recalcitrant’ inmate, in addition to employing it in its customary role to prevent riots and escapes. A written report is required only after a tear gas shell is fired. The superintendent also stated that no record is kept of the number of shells issued and no crosscheck is made to determine if reports are *968filed covering each instance that the gas is employed. By way of contrast, the use of tear gas against a single inmate in a federal prison is permitted only ‘in most extraordinary circumstances and after approval in writing of the warden or superintendent.’ U.S. Department of Justice Bureau of Prisons, Use of Tear Gas and Smoke 3 (1949). While it is our belief that the federal practice is the more enlightened, we will not disturb the finding that the use of tear gas revealed in this record did not amount to ‘cruel and unusual punishment.’ ”

This language which is the dispositive part of the court’s opinion, does not sustain in any way the dissent’s basic premise; in fact, it is clearly authority for a contrary view and, as we have already observed, it has been so generally considered.

The other authority cited by the dissent for its per se rule is Landman v. Royster, 333 F.Supp. 621, 649 (E.D.Va. 1971). The dissent, quoting from this case, declares that the “federal law in this circuit clearly prohibited ‘the use of [tear] gas to disable a man physically who poses no present physical threat.’ ” But the quotation included in the dissent is only a few words taken out of a long sentence which itself in turn was a part of a larger and more extensive paragraph in this decision. The full paragraph of the opinion, with the words extracted therefrom in the dissent italicized, is:

“Tear gas has also been used to silence noisy, misbehaving men while confined to their cells. Thomas Jefferson was gassed three times, and others have been gassed in their cells at the penitentiary. The problem of dealing with convicts who persist in disturbing entire cell blocks and inciting others to join in the disorder is a real one. The Court has not found any instances of gassing men in cells who were not currently disruptive. Yet the use of gas to disable a man physically who poses no present physical threat constitutes a form of corporal punishment, the use of which in such a situation is generally disapproved. Undoubtedly it is effective, but it is painful, and its abuse is difficult to forestall. The problem appears to arise because there appears to be no way to isolate a misbehaving inmate to an area where his rantings will not disturb anyone. This difficulty is, however, one of the system’s own creation. If chaining a man to his bars, punishing him with a strap, and other corporal punishment should be enjoined, Jackson v. Bishop, supra [404 F.2d 571] this Court cannot make a principled distinction which would permit the use of tear gas to punish or control the nonthreatening inmate.” Id. at 649.

As is seen, the full sentence from which the dissent extracted its quotation, did not “clearly prohibit” tear gas “to disable a man physically who poses no present physical threat” but declared that the use of tear gas “in such a situation is generally disapproved.” 6 (Italics added) Moreover, it recognized that control of noisy and disturbing inmates with mace was not unusual or necessarily improper.

That the decision in Royster has not been understood as prohibiting the use of tear gas against “recalcitrant” prisoners while locked in their cells in this circuit is illustrated by the decisions in McCargo v. Mister, 462 F.Supp. 813, 818 (D.Md. 1978) and in Greear v. Loving, 391 F.Supp. 1269 (W.D.Va. 1975). Thus, in McCargo the court said at p. 818:

“The use of tear gas against inmates in their cells is not per se a cruel and unusual punishment. In the landmark case of Landman v. Royster, 333 F.Supp. 621 (E.D.Va. 1971), Judge Merhige noted that ‘[t]ear gas has also been used to silence noisy, misbehaving men while confined to their cells;’ however, in a situation where tear gas was used to punish or control a non-threatening inmate, the court concluded that such use was ‘generally disapproved.’ 333 *969F.Supp. at 649. Similarly, in Greear v. Loving, 391 F.Supp. 1269 (W.D.Va. 1975), vacated, 538 F.2d 578 (4th Cir. 1976), the court remarked that ‘the use of tear gas is not forbidden in all instances but only when there appears to be no. necessity for its use,’ and went on to ‘emphasize the close scrutiny with which it considers the use of tear gas in correctional facilities.’ 391 F.Supp. at 1271.” 7

Perhaps more illuminative on this circuit’s rule with respect to the use of tear gas than McCargo or Royster is the decision on appeal in Greear v. Loving, 538 F.2d 578 (4th Cir. 1976). The prisoner in that case had been gassed while locked in his cell. The district court granted summary judgment in favor of the prison official. In so doing, it cited Landman v. Royster. On appeal, we reversed, finding the case not appropriate for summary judgment because of disputed facts but emphasizing that we were “decid[ing] nothing concerning the merits of Greear’s claims.” Id. at 580. In the opinion, we outlined the differences in the testimony which caused the court to reverse the summary judgment. For the purposes of this case, though, the important feature of the ease is that the court did not hold that it was unconstitutional per se to use tear gas on a prisoner locked in his cell in order to quiet and control him, which it should have done if the “clear” rule of this circuit were as the dissent suggests; the court rather remanded the case for a jury to determine whether the use was illegal under all the circumstances.

Greear illustrates the correct rule in this context. That rule is not one of a per se rule of unconstitutionality but a rule that looks to the totality of the circumstances, the provocation, the amount of gas used, and the purposes for which the gas was used, in determining the validity of the use of tear gas in the prison environment. As Judge Merhige in Royster observed, “[t]he problem of dealing with convicts who persist in disturbing entire cell blocks and inciting others to join in the disorder is a real one.” Most of the cases have held that, provided the quantity of gas used is graduated by the gravity of the occasion, gassing is not violative of constitutional rights. This was the ruling of the Court of Appeals in Spain v. Procunier, 600 F.2d 189 (9th Cir. 1979), modifying the opinion in 408 F.Supp. 534 on which McCargo relied. There the court laid down this rule on the use of tear gas by prison officials:

“If then the tear gas is used in dangerous quantities, we agree with the district court that its use is justified only in those grave circumstances which would justify the use of severe and potentially lethal force.

“We think the record further indicates, however, that use of the substance in small amounts may be a necessary prison technique if a prisoner refuses after adequate warning to move from a cell or upon other provocation presenting a reasonable possibility that slight force will be required. In these circumstances the substance may be a legitimate means for preventing small disturbances from becoming dangerous to other inmates or prison personnel____

“... We agree with the trial court that use of potentially dangerous quantities of the substance is justified only under narrowly defined circumstances, but we further conclude that use of nondangerous quantities of the substance in order to prevent a perceived future danger does not violate ‘evolving standards of decency’ or constitute an ‘unnecessary *970and wanton infliction of pain.’ ” 600 F.2d at 195-96.

The lower court in this case followed the rule stated in Spain in its instructions to the jury. Thus, it said in these instructions:

“The use of a chemical restraint such as mace by and of itself does not establish excessive and unjustified force. As in every instance where force is employed, the reasonableness of its use depends on the surrounding circumstances. Chemical agents should not be indiscriminately employed so you should consider the amount, frequency, purpose of use in determining whether the macing constitutes excessive force. Mace may bé reasonably used to quell disorders and to compel obedience but not to punish a prisoner. In deciding whether the force employed by a particular defendant was unjustified, excessive and unreasonable, you should keep in mind that the Defendants’ actions took place in an environment different from the quiet of the courtroom, that they occurred within the confines of a prison where it was the Defendants’ duty to maintain order and discipline, and [in] such an environment — not every use of force is a constitutional violation but only that use of force which shocks a reasonable person’s conscience because it is unjustified, excessive and unreasonable. Factors which you should consider in deciding whether such force was used against Plaintiff include, but are not limited to, one, the need for the application of the force; two, the relationship between the need for force and the amount of force used; three, the extent of injury inflicted; and four, whether the force was applied in good faith effort to maintain or restore discipline or was used maliciously or sadistically for the very purpose of causing harm.”

In so directing the jury the lower court was not only following what appeared to be the rule of this circuit as we read Peyton and Greear in particular,8 but that of most of the other circuits. See for instance: Lock v. Jenkins, 641 F.2d 488, 496 (7th Cir. 1981); Stringer v. Rowe, 616 F.2d 993, 998 (7th Cir. 1980); Poindexter v. Woodson, 510 F.2d 464, 466 (10th Cir. 1975), cert. denied, 423 U.S. 846, 96 S.Ct. 85, 46 L.Ed.2d 68; Clemmons v. Greggs, 509 F.2d at 1340 (“[t]he use of tear gas when reasonably necessary to prevent riots or escapes or to subdue recalcitrant prisoners does not constitute cruel and inhuman punishment,” citing Landman v. Peyton, 370 F.2d 135; Bethea v. Crouse, 417 F.2d 504, 509 (10th Cir. 1969); Breece v. Swenson, 332 F.Supp. 837, 840 (W.D.Mo. 1971).

We think the above review both of the authorities of this circuit as well as of other circuits makes it abundantly clear that the use of mace on an unruly or “recalcitrant” prison inmate, though confined in his cell, is not plainly per se unconstitutional and thus the rule for denying to the defendant in this case the defense of good faith immunity is not present [i.e., the unconstitutionality of the defendant’s conduct was so clearly established that the defendant should have known his conduct was illegal].9

*971Whether it was constitutional or not depended on whether the jury believed the defendant or the plaintiff. ’ The defendant testified he alone used mace and then only in a single burst for a second or two. The plaintiff, on the other hand, testified he was gassed continuously by three or four guards, of whom the defendant was one, for several minutes.10 Under the Spain rule and that included in the lower court’s jury instruction in this case the jury, if it believed the defendant that the intrusion was as minor as the defendant said, it should have found for the defendant; if on the other hand, it believed the plaintiff’s testimony, the jury should have found for the plaintiff. The jury resolved this point in favor of the defendant and it is not our function to reverse that finding. Even if this were not true, though, we are not convinced that the clear error exception to the mandate of Rule 51, Fed.R.Civ.P. is properly invoked under the circumstances of this ease. Morris v. Travisono, 528 F.2d at 858, we think, supports conclusively a conclusion contrary to that reached in the dissent.

The plaintiffs-prisoners in Morris sued the prison officials, contending that “the use of tear gas to punish nonthreatening prisoners constituted cruel and unusual punishment,” redressable under § 1983. In submitting the cause to the jury, the district judge had given this instruction:

“ ‘[I]f you find that the plaintiffs were gassed while locked in their cells and at the time of the gassing posed no substantial and immediate physical threat to themselves, other prisoners or to the security of the institution or to the correctional officers but rather the plaintiffs were gassed for the mere purpose of punishing them, then you must find for the plaintiffs.’ ” (Emphasis in text)

The verdict of the jury was in favor of the plaintiffs-prisoners. On appeal, the defendants argued that the instruction as given in its definition of cruel and unusual punish*972ment had used “a harsher legal standard than that established by precedent” where the charge of cruel and unusual punishment consisted of the use of tear gas on a prisoner locked in his cell. The court recognized that the challenged instruction was “open to attack as lowering the threshold of cruel and unusual punishment to a level that cannot be said to represent a well-settled principle of constitutional law.” Id. at 858. The court, however, said it was unnecessary to reach that “serious question of substantive law” because of “the failure of defendants to object clearly and specifically to the district court’s instructions before the jury retired to consider its verdict, as required by Fed.R.Civ.P. 51.” Id. at 859. The court proceeded to emphasize that “Rule 51 is of considerable importance for the orderly and just functioning of the judicial system” and that [quoting 9 C. Wright & A. Miller, Federal Practice & Procedure § 2558, at 675 (1971) ], “ ‘If there is to be a plain error exception to Rule 51 at all, it should be confined to the exceptional case where the error has seriously affected the fairness, integrity, or public reputation of judicial proceedings.’ ” It concluded that this was not such a case and affirmed the judgment entered on the verdict. We agree with this decision.

For the reasons stated, we feel that the rationale of the dissent cannot be sustained. To repeat: First, its primary premise that the clearly settled law of this circuit is that a gassing of a prison inmate in his cell is per se unconstitutional is an untenable assumption, not supported by the decisions of this circuit. Since this is so, it cannot be said that the defendant knew or reasonably should have known that his action in gassing the plaintiff was unconstitutional and, therefore, under Harlow, the defendant was entitled, as the magistrate without objection instructed the jury, to a good-faith defense. It follows, therefore, that the instructions of the magistrate were proper but, even if not proper, the plaintiff, who was represented by able and experienced counsel, waived any objections to such instructions by failure to object at trial.

Accordingly, we affirm the judgment below.

dissenting:

I cannot agree that the plaintiff’s failure to object to the lower court’s jury charge precludes him from asserting error on appeal. In my view, an objection was not necessary because the lower court committed plain error when it submitted an instruction on good faith to the jury. For this reason, I respectfully dissent.

I.

In Wood v. Strickland, 420 U.S. 308, 95 S.Ct. 992, 43 L.Ed.2d 214 (1975), the Supreme Court firmly established that a good faith defense is not applicable if a state or federal official “knew or reasonably should have known that the action he took within his sphere of official responsibility would violate ... constitutional rights____” Id. at 322, 95 S.Ct. at 1001. Since Wood was decided, the Supreme Court has twice reaffirmed this aspect of the good faith standard, emphasizing both the subjective and objective elements of the test. See Procunier v. Navarette, 434 U.S. 555, 98 S.Ct. 855, 55 L.Ed.2d 24 (1978); Harlow v. Fitzgerald, 457 U.S. 800, 102 S.Ct. 2727, 73 L.Ed.2d 396 (1982).

From my reading of the record I remain unconvinced that defendant West survives either prong of the test. In 1974, when the macing occurred, the eighth amendment unquestionably protected prison inmates from the unjustified and excessive use of force by prison officials. Wilkerson v. Utah, 99 U.S. 130, 136, 25 L.Ed. 345 (1878); Weems v. United States, 217 U.S. 349, 30 S.Ct. 544, 54 L.Ed. 793 (1909); Louisiana ex rel. Francis v. Resweber, 329 U.S. 459, 67 S.Ct. 374, 91 L.Ed. 422 (1947); United States v. W.T. Grant Co., 345 U.S. 629, 73 S.Ct. 894, 97 L.Ed. 1303 (1953); Trop v. Dulles, 356 U.S. 86, 78 S.Ct. 590, 2 L.Ed.2d 630 (1958); Furman v. Georgia, 408 U.S. 238, 92 S.Ct. 2726, 33 L.Ed.2d 346 (1972); Jackson v. Bishop, 404 F.2d 571 (8th Cir. 1968) (Per Blackmun, C.J.); Inmates of At*973tica Correctional Facility v. Rockefeller, 453 F.2d 12 (2d Cir. 1971); Wright v. McMann, 387 F.2d 519 (2d Cir. 1967); Sostre v. McGinnis, 442 F.2d 178 (2d Cir. 1971) (en banc); Landman v. Royster, 333 F.Supp. 621 (E.D.Va. 1971). The same protection was provided under North Carolina law. See N.C.Gen.Stat. § 148-20 (prohibiting corporal punishment); see also Threatt v. North Carolina, 221 F.Supp. 858 (W.D.N.C. 1963).

In 1974 it was also clear that this prohibition against excessive force applied to the use of tear gas. The leading cases in this circuit stated then, and still state now, that where the actions of a prisoner pose no danger to the safety of guards, or other prisoners, or to the general security of the prison, and where there is no need to move physically a recalcitrant prisoner, the use of tear gas on inmates constitutes a form of corporal punishment that violates the eighth amendment. Landman v. Royster, 333 F.Supp. 621, 649 (E.D.Va. 1971) (“If chaining a man to his bars, punishing him with a strap, and other corporal punishment should be enjoined, Jackson v. Bishop, supra, 404 F.2d 571, this court cannot make a principled distinction which would permit the use of tear gas to punish or control the non-threatening inmate.”) See also Landman v. Payton, 370 F.2d 135 (4th Cir. 1966); McCargo v. Mister, 462 F.Supp. 813, (D.Md. 1978); Greear v. Loving, 391 F.Supp. 1269 (W.D.Va. 1975). Since 1974 we have reaffirmed our opposition to the use of excessive force on several occasions. Slakan v. Porter, 737 F.2d 368, (4th Cir. 1984); (prohibiting excessive use of billy clubs, tear gas, or high pressure hoses); Pritchard v. Perry, 508 F.2d 423, 425-26 (4th Cir. 1975); see also Bruce v. Wade, 537 F.2d 850 (5th Cir. 1976); Johnson v. Glick, 481 F.2d 1028 (2d Cir.) cert. denied, 441 U.S. 913, 94 S.Ct. 462, 38 L.Ed.2d 324 (1979).

This is not to say that the use of tear gas is unconstitutional per se. We have held that under the proper circumstances, the use of tear gas in moderation can be an effective and acceptable means of inmate control.1 The record in this case, however, *974reveals that West’s use of mace was unnecessary and abusive. Bailey was incarcerated in a one-person cell in a building with only four cells. His sole offense consisted of making profane remarks to a prison guard. I do not believe that Bailey could have posed a danger to the personal safety of West, other inmates, or the general security of the prison. Notwithstanding West’s self-serving testimony about the need to guard against a riot, this was not a situation where the density of the prison population made it either possible or likely that an abusive remark would spark a general disturbance or an uprising. Cf. Clemmons v. Greggs, 509 F.2d 1338 (1975) (use of gas sanctioned where near-riot conditions prevailed in heavily populated cell block). West himself could not have been in physical danger because he had no reason to enter Bailey’s cell and the incident involved only an exchange of words. There was no evidence in the record to indicate that Bailey had damaged the cell or was inflicting injury on himself, and there was no danger to other prisoners because Bailey was in a single-person cell. Immediate incapacitation of the prisoner was simply not required. Cf. Greear v. Loving, 538 F.2d 578 (1976) (gas needed to stop prisoner from setting fire to his mattress and destroying the plumbing in the cell).

Regardless of whose testimony one believes, the record paints a vivid picture of gross over-reaction to a relatively minor offense. Prisoners are often unruly and insubordinate. In some instances punishments are required. In this case, however, what little disruption Bailey’s comments might have posed to the peace of the prison could have been handled by removing Bailey to another cell, ignoring the comments entirely, or threatening the prisoner with a milder form of non-corporal punishment. In short, the macing amounted to a gratuitous infliction of bodily injury. The majority appears to forget that the price Bailey paid for his cursing went beyond temporary discomfort; Bailey has permanently lost part of his sight.

In sanctioning the use of tear gas under certain aggravated circumstances, this court never intended for gas to be used routinely to discipline prisoners who stray from the letter of the prison code. Were gas to be used in this manner, it would cease to be a means of inmate control and would become a form of corporal punishment to be dispensed at the discretion of prison guards. As this court recognized in Landman v. Peyton, 370 F.2d 135 (1966), absolute power is corrupting, whether it lies in the hands of prison guards, or other men. Id. at 140. It is the responsibility of the courts to ensure that unnecessarily violent prison practices do not pass unchecked. Although prison officials must have some discretion over the application of punitive measures to protect themselves and other inmates, carte blanche use of tear gas has never been, and should never be, included in those protective or disciplinary measures. Were we to permit such a practice the door would be opened to routine infliction of excessive and unnecessary corporal punishment.

However difficult it may be in some instances to draw the line between punishment and legitimate use, there is little question in my mind but that the line was crossed here. This case involves none of the exigent circumstances that we have relied on in the past to justify the use of tear gas.2 In my view, therefore, West’s actions violated clearly established case law prohibiting the use of excessive force. The constitutionality of West’s response was not a matter for the jury to decide. West reasonably should have known that under these circumstances the use of mace violated Bailey’s constitutional rights, and thus I believe West was not entitled under Harlow to a good faith immunity defense.

In addition to my concern over West’s ability to satisfy the objective prong of the *975Harlow standard, I do not believe West can meet the subjective element of the test. Although West claimed statutory authority under North Carolina law3 to use mace on Bailey, he testified that he knew prison policy required the least degree of force necessary to enforce discipline. West conceded that he knew he “didn’t have to use mace” on Bailey, and even went so far as to enumerate other, less harsh, measures he could have taken.4

I conclude, therefore, that as a matter of law a good faith defense was not available to the defendant. The lower court erred in failing to instruct the jury accordingly.5

II.

The majority correctly points out that plaintiff’s counsel did not object to the court’s instructions on good faith after they were read to the jury. This court has repeatedly held, however, that failure to object need not preclude appellate review if the instruction “plainly misstates fundamentally controlling substantive principles”, Miller v. Premier Corp., 608 F.2d 973, 983 (4th Cir. 1979), and thereby prejudices the verdict.6

In my view the court’s decision to tender instructions on good faith amounted to a misstatement of fundamental principles. It matters little whether a court charges incorrectly on legal principles that should be tendered, or charges correctly on principles that should not be presented to the jury;7 in either event the law applicable to the case at hand has been mistated. Thus, *976although the court here did not actually místate a required element, Miller still applies.

There is little question but that the lower court’s error was prejudicial. Because the jury rendered a special verdict, we know as a matter of record that the jury found defendant West guilty of violating the plaintiff’s eighth amendment rights, yet immune from damages under the good faith defense. The court’s decision to instruct on “good faith”, therefore, effectively denied the plaintiff damages he should have received. Clearly, then, both parts of the plain error standard — “plain misstatement” and prejudice — have been satisfied.8

III.

For the foregoing reasons, I would reverse the ruling of the lower court and remand for a determination of damages.