OPINION
On petition of Mary Ann Mireles, we issued a writ of certiorari to the Court of Appeals to decide (1) whether the doctrine of res ipsa loquitur is restricted to events from which the jury, without assistance of expert testimony, could infer negligence from common knowledge that such events do not otherwise ordinarily occur; and (2) what responsibility the trial court has.in response to a request for a jury instruction on a theory to *447which a party is entitled but which has been requested in language that is unsatisfactory to the court. This is a medical malpractice action in which Mireles has sued Dr. Thomas Broderick, her anesthesiologist. Based on expert testimony that the anesthesiologist positions and cushions the patient’s arm to avoid nerve compression injury during surgery, Mireles requested the trial court to describe the injury-causing occurrence that was within the exclusive control or management of the anesthesiologist as “inadequate protection of plaintiffs extremities during anesthesia.” Because of alleged error in the statement of this element of the doctrine, the Court of Appeals affirmed the trial court’s refusal to instruct the jury on res ipsa loquitur as requested by Mireles. Mireles v. Broderick, 113 N.M. 459, 827 P.2d 847 (Ct.App.1992). We reverse and remand for a new trial. In so doing, we address additional issues raised by Dr. Broderick — the propriety of instructing the jury on res ipsa loquitur in medical malpractice actions generally and in this case in particular, and the necessity of establishing exclusive control in order to submit the instruction to the jury.
Facts and proceedings. Shortly after undergoing a bilateral mastectomy, Mireles experienced numbness in her right arm. The numbness subsequently was diagnosed as ulnar neuropathy, a condition marked in her case by degenerative nerve damage to the fourth and fifth fingers of her right hand. Mireles brought this action against Dr. Broderick, alleging separate counts of medical negligence, battery, and res ipsa loquitur. The case went to trial before a jury on the negligence and res ipsa loquitur theories. Mireles’s expert witness, Dr. Randall Waring, testified that the ulnar nerve can be injured if it is compressed. He testified he believed that Mireles’s ulnar injury, “in all probability, occurred while she was under anesthesia for [the] surgery” and that such injury was totally preventable by proper care. He testified that the ultimate responsibility for protection against injury lies with the anesthesiologist, who should properly position and cushion the arm to avoid compression and should monitor the arm during surgery to be sure that proper positioning and cushioning are maintained while the patient is unconscious. At the close of Mireles’s case, the presiding judge stated that he was not going to allow Mireles to go forward with the case on the theory of res ipsa loquitur “because it doesn’t come under the exclusivity rule.” The court later refused Mireles’s requested instruction on res ipsa loquitur.
Focusing on the content of the requested instruction, the Court of Appeals affirmed the trial court. Citing SCRA 1986, 1-051(1) (Repl.Pamp.1992) (stating that “a correct instruction must be tendered” to preserve error in instructions), the Court based its affirmance on the conclusion that the tendered instruction was “not a proper res ipsa instruction.” Mireles, 113 N.M. at 461, 827 P.2d at 849. The Court also advanced a broader rationale, namely that there could be no error in refusing the requested res ipsa loquitur instruction because, as framed by Mireles, the instruction was “at best, an ‘unnecessary crutch’ that set forth an obvious proposition for which no additional instruction was necessary.” Mireles, 113 N.M. at 465, 827 P.2d at 853.
Propriety of res ipsa loquitur instruction in medical malpractice actions. In contending that res ipsa loquitur is inapplicable to this medical malpractice action as a matter of law, Dr. Broderick advances two arguments. First, he contends that res ipsa loquitur is available only when an inference of negligence is articulable from the common knowledge and experience of the lay person. The major thrust of Dr. Broderick’s argument is that the common-knowledge requirement is the “historical premise” of res ipsa loquitur, and to permit expert testimony to establish the inference of negligence would constitute an “end run” around this premise. According to Dr. Broderick, medical malpractice plaintiffs should be required to base their cases either on expert testimony or “common-knowledge” res ipsa loquitur, but not both. Second, according to Dr. Broderick, when a plaintiff has attempted to explain the exact medical cause of the injury, she should not have the benefit of the res ipsa loquitur instruction.
—Expert testimony may support an inference of negligence. Dr. Broderick argues *448that, because of the rule that negligence of medical providers generally must be proved by expert testimony, res ipsa loquitur is limited in malpractice cases to the common-knowledge exception alluded to by this Court in Cervantes v. Forbis, 73 N.M. 445, 448, 389 P.2d 210, 213 (1964). He argues that only when the inference of negligence is within the common reservoir of knowledge of the jurors may the jury be charged on the res ipsa loquitur doctrine. In Cervantes, we stated that -without expert witness testimony demonstrating departure from medical standards “there can be no issue of fact as to the negligence or proximate cause unless the case is one where exceptional circumstances within common experience or knowledge of the layman are present, or one where the res ipsa loquitur rule is applicable.” Id. at 448-49, 389 P.2d at 213 (emphasis added).
By focusing on the rule of Cervantes, Dr. Broderick’s argument loses sight of the dispositive principle at issue in the application of res ipsa loquitur. Res ipsa loquitur describes a set of conditions to be met before an inference of negligence may be drawn. See Hepp v. Quickel Auto & Supply Co., 37 N.M. 525, 528, 25 P.2d 197, 199 (1933) (quoting from Plumb v. Richmond Light & R. Co., 233 N.Y. 285, 135 N.E. 504 (1922), that res ipsa loquitur is a “rule that the fact of the occurrence of an injury and the surrounding circumstances [of the defendant’s control and management] may permit an inference of culpability on the part of the defendant, make out plaintiffs prima facie case, and present a question of fact for the defendant to meet with an explanation”). As such, the central issue is not whether common knowledge alone is sufficient to establish an inference of negligence. Rather, the issue is whether there is a factual predicate sufficient to support an inference that the injury was caused by the failure of the party in control to exercise due care. The requisite probability of negligence may exist independently of the common knowledge of the jurors. The common-knowledge exception to the expert testimony rule may inform but does not delimit the application of res ipsa loquitur.
We join the growing consensus of courts from other jurisdictions and adopt scholarly commentary to hold that the foundation for an inference of negligence may be formed by expert testimony that a certain occurrence indicates the probability of negligence. See, e.g., Seneris v. Haas, 45 Cal.2d 811, 291 P.2d 915, 923 (1955) (in bank); Walker v. Rumer, 72 Ill.2d 495, 21 Ill.Dec. 362, 364, 381 N.E.2d 689, 691 (1978); Perin v. Hayne, 210 N.W.2d 609, 614-15 (Iowa 1973); Buckelew v. Grossbard, 87 N.J. 512, 435 A.2d 1150, 1157-58 (1981); Mayor v. Dowsett, 240 Or. 196, 400 P.2d 234, 243-44 (1965) (in banc); Jones v. Harrisburg Polyclinic Hosp., 496 Pa. 465, 437 A.2d 1134, 1138 (1981); Van Zee v. Sioux Valley Hosp., 315 N.W.2d 489, 492-93 (S.D.1982); Horner v. Northern Pac. Beneficial Ass’n Hosps., Inc., 62 Wash.2d 351, 382 P.2d 518, 524 (1963); Fehrman v. Smirl, 20 Wis.2d 1, 121 N.W.2d 255, 268 (1963); Restatement (Second) Torts § 328D cmt. d (1965) (“[E]xpert testimony that such an event usually does not occur without negligence may afford a sufficient basis for the inference [of negligence].”); W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 39, at 247 (5th ed. 1984) (stating that when a basis of common knowledge is lacking, expert testimony may provide a sufficient foundation for an inference of negligence); Thomas A. Eaton, Res Ipsa Loquitur and Medical Malpractice in Georgia: A Reassessment, 17 Ga.L.Rev 33, 52-56 (1982) (recommending that Georgia permit inference of negligence in medical malpractice actions to be based on expert testimony).
Contrary to Dr. Broderick’s contentions, we are aware of no reported New Mexico opinions either expressly or impliedly prohibiting the founding of a res ipsa loquitur inference on expert testimony in medical malpractice actions or, for that matter, in any tort action. Dr. Broderick suggests that this Court in Buchanan v. Downing, 74 N.M. 423, 394 P.2d 269 (1964), expressed reluctance to permit an expert-based res ipsa loquitur inference. To the contrary, Buchanan appears to encourage the practice. There, the plaintiff sued his doctor for injuries alleged to have resulted from the negligent administration of a hypodermic injection. On appeal from an adverse summary judgment, the plaintiff contended that application of the *449doctrine of res ipsa loquitur presented a factual issue of negligence sufficient to avoid summary judgment. Id. at 424, 394 P.2d at 270. Affirming, this Court noted that the plaintiffs failure to present evidence “by deposition or affidavit, of any witness, expert or otherwise, supporting his allegation of negligence or the proximate cause of the injury,” id. at 424, 394 P.2d at 271, was fatal to his claim that summary judgment was inappropriate. Id. at 427, 394 P.2d at 273. The Court reviewed the deposition testimony relied upon by the plaintiff and concluded that the testimony dispelled rather than raised an inference of negligence. Id. at 425, 394 P.2d at 271. We find nothing in Buchanan that would suggest reluctance to sanction, in the proper case, an expert-based res ipsa loquitur inference.
Dr. Broderick asserts, with little analysis, that by allowing expert testimony to form the basis for the res ipsa loquitur inference of negligence, “res ipsa loquitur [would lose] its roots as a form of circumstantial evidence.” We fail to perceive any such historical limitation on the application of res ipsa loquitur.1 As we have discussed, courts in other jurisdictions long have permitted expert testimony to form the foundation for an inference of negligence; commentators also endorse the use of expert testimony to raise the res ipsa loquitur inference. Stripped of technical argument, Dr. Broderick’s contentions are a plea for a policy-driven decision to except medical malpractice defendants from the application of res ipsa loquitur. This we are unwilling to do.
—Evidence of specific cause of injury did not preclude res ipsa loquitur inference. Dr. Broderick contends that because Mireles adduced testimonial evidence concerning what could have been a cause of her injury, she is precluded from relying on the doctrine of res ipsa loquitur. He identifies the testimony of Dr. Waring that Mireles’s arm was probably moved during surgery so that the ulnar nerve was compressed for between thirty and forty minutes on the edge of the padded board that held her arm. Citing Marrero v. Goldsmith, 486 So.2d 530, 532 (Fla.1986), Dr. Broderick’ contends that when a plaintiff undertakes explanation of the accident as it actually occurred, “there is a point where the plaintiff destroys any inference of other causes and dispels the need for the inference of negligence.” Recognizing that the Court of Appeals has held that some evidence of the cause of the injury does not obviate the need for res ipsa loquitur, see Harless v. Ewing, 81 N.M. 541, 545-46, 469 P.2d 520, 524-25 (Ct.App.1970), Dr. Broderick contends nonetheless that the evidence in this case rose above the “some evidence” standard of Harless. We find Prosser, supra, to be illuminating on this point.
Plaintiff is of course bound by his own evidence; but proof of some specific facts does not necessarily exclude inferences of others. When the plaintiff shows that the railway car in which he was a passenger was derailed, there is an inference that the defendant railroad has somehow been negligent. When the plaintiff goes further and shows that the derailment was caused by an open switch, the plaintiff destroys any inference of other causes; but the inference that the defendant has not used proper care in looking after its switches is not destroyed, but considerably strengthened. If the plaintiff goes further still and shows that the switch was left open by a drunken switchman on duty, there is nothing left to infer; and if the plaintiff shows that the switch was thrown by an escaped convict with a grudge against the railroad, the plaintiff has proven himself out of court. It is only in this sense that when the facts are known there is no inference, and res ipsa loquitur simply vanishes from the case. On the basis of reasoning such as this, it is quite generally agreed that the introduction of some evidence which tends to show specific acts of negligence on the part of the defendant, but which does not *450purport to furnish a full and complete explanation of the occurrence, does not destroy the inferences which are consistent with the evidence, and so does not deprive the plaintiff of the benefit of res ipsa loquitur.
Prosser § 40, at 260 (footnotes omitted).
Significantly, Mireles was actually unable to provide by way of an expert opinion direct evidence that Dr. Broderick’s failure to use proper care resulted in her injury. No expert “went so far” as to testify that the injury to Mireles was proximately caused by the failure of Dr. Broderick to possess and apply the knowledge and to use the skill and care ordinarily used by reasonably well-qualified anesthesiologists. See SCRA 1986, 13-1101 (Repl.Pamp.1991) (Uniform Jury Instruction, duty of doctor). Because Dr. Waring’s testimony did not purport to furnish an opinion on the ultimate issues of medical malpractice, it did not deprive Mireles of the benefit of res ipsa loquitur.
Erroneous refusal of res ipsa loquitur instruction based on disputed craftsmanship. The parties and the Court of Appeals have focused on alleged error committed by the trial court in refusing the New Mexico Uniform Jury Instruction for res ipsa loquitur, which states:
In support of the claim that _ was negligent, ___ relies [in part] upon the doctrine of “res ipsa loquitur” which is a Latin phrase and means “the thing speaks for itself’. In order for the jury to find _negligent on this doctrine,_has the burden of proving.each of the following propositions:
1. that the injury or damage to _ was proximately caused by
(name of instrumentality or occurrence) which was under the exclusive control and management of_; and
2. that the event causing the injury or damage to __ was of a kind which does not ordinarily occur in the absence of negligence on the part of the person in control of the instrumentality.
If you find that_proved each of these propositions, then you may, but are not required to, infer that_was negligent and that the injury or damage proximately resulted from such negligence.
If, on the other hand, you find that either one of these propositions has not been proved or, if you find, notwithstanding the proof of these propositions, that _used ordinary care for the safety of others in [his] [her] [its] control and management of the_then the doctrine of res ipsa loquitur would not support a finding of negligence.
SCRA 1986, 13-1623 (Repl.Pamp.1991).
Pertinent to our discussion, Mireles requested this instruction with the following insertion which we have italicized:
1. That the injury to Plaintiff was proximately caused by inadequate protection of Plaintiffs extremities during anesthesia while her condition was under the exclusive control and management of Dr. Broderick.
2. That injury to Plaintiff was of the kind which does not ordinarily occur in the absence of negligence on the part of the person in control.
In essence, the Court of Appeals appears to have reasoned that the instruction as worded was unnecessary, that it did not provide for the finding of a factual premise upon which to base a res ipsa loquitur inference of negligent causation. The Court believed the jury was instructed to find the fact issue of negligence rather than the premise of exclusive control. Mireles, 113 N.M. at 464, 827 P.2d at 852 (stating the instruction “begins after the jury has crossed the res ipsa bridge”). Had the jury found to be true the proposition that Mireles’s injury was caused by “inadequate protection” of her extremities, the jury would not need a res ipsa loquitur inference to determine liability. Id. Judge Bivins similarly opined that the term “inadequate protection” describes the “specific acts of negligence Plaintiff relied on to prove Defendant negligent.” Id. at 466, 827 P.2d at 854 (Bivins, J., specially concurring).
This Court has long held that a requested instruction that is erroneously or inaccurately drafted need not be given by the trial court. See, e.g., Childers v. Southern Pac. Co., 20 N.M. 366, 375, 149 P. 307, 309 *451(1915); Hanks v. Walker, 60 N.M. 166, 174, 288 P.2d 699, 704 (1955) (“A request for an erroneous instruction is properly refused.”). The formulations of the rule suffer little tolerance. See Childers, 20 N.M. at 375, 149 P. at 309 (stating that the requested instruction must be in a form so that the trial court can submit it to the jury without qualification or modification; if the requested instruction is erroneous “either wholly or in part,” it is properly refused). It should be clear, however, that these statements of the rule beg rather than answer the question we face today: When is an instruction so legally incorrect or deficient to justify wholesale rejection by the trial court?2 An instruction is correct, and thus proper to submit to a jury, when the instruction is consistent with the law, cf. Ambassador Ins. Co. v. St. Paul Fire & Marine Ins. Co., 102 N.M. 28, 32, 690 P.2d 1022, 1026 (1984) (approving an instruction that “accords with New Mexico law”), and articulates fairly, completely, and succinctly the relevant law applicable to the facts, cf. Gerrard v. Harvey & Newman Drilling Co., 59 N.M. 262, 267, 282 P.2d 1105, 1108-09 (1955) (stating, “only statements of law to be applied in the examination and determination of the issue” should be included), so that the jury when so instructed will be able to comprehend and apply the law, see Haynes v. Hockenhull, 74 N.M. 329, 334, 393 P.2d 444, 447 (1964) (stating that the purpose of instructing the jury “is to make the issues that they are to determine plain and clear”). In this case, we agree with Judge Pickard’s dissent that the majority of the Court of Appeals read the “requested instruction in an unnecessarily technical fashion,” see Mireles, 113 N.M. at 468, 827 P.2d at 856 (Pickard, J. dissenting), and conclude that the instruction in this case was legally correct and should have been submitted to the jury as requested or as edited by the trial court.
The issue is whether insertion of the phrase “inadequate protection of Plaintiffs extremities during anesthesia” in the space provided for “instrumentality or occurrence” fairly apprises the jury of the premise required by law. It is true that the inserted phrase articulates a specific act, and does so with the term “inadequate”, which is often associated with breach of duty. The fairest interpretation of the phrase, however, is that it identifies to the fullest extent possible the occurrence that allegedly caused Mireles’s injury. We agree with Judge Pickard that the use of “inadequate protection” in that part of the instruction does not necessarily imply a specific act of negligence. “Adequate protection” simply means protection that is sufficient to prevent injury. Adequate protection may not be possible with the best of care. Some medical procedures are high risk because there is no way to protect against complications. Here, however, an expert testified that there are means of protecting against ulnar nerve damage and that complications do not ordinarily occur in the absence of negligent failure to employ such means.
This issue turns on the distinction between views on legal drafting and the measurement of legal sufficiency by objective criteria. As the Court of Appeals suggested, we also could propose different ways to draft the res ipsa loquitur instruction in this case and suggest that perhaps “positioning of the plaintiffs body” may have been a better description of the injury-causing occurrence. Indeed, we would not be surprised to find many different approaches to the same problem. The task for the trial court in circumstances calling for drafting by counsel, however, is to determine whether the instruction as tendered informs the jury of the relevant law to be applied to the facts in the case. When draft language gives rise to concern, the trial court is nonetheless under a duty to instruct the jury succinctly and accurately on the issue of law presented. See SCRA 1-051(B) (“The court shall instruct the jury regarding the law applicable to the facts in the cause unless such instructions be waived by the parties.”); Gerrard, 59 N.M. at 273, 282 P.2d at 1112 (“[I]t is the duty of the court at every trial to give to the jury the fundamental law applicable to the facts in the case and ... unless waived by the parties, *452instructions to that extent at least, must be given whether requested or not----”). The trial court may submit the instruction as tendered or change the instruction, with or without consultation with counsel, to suit his or her particular proclivity and style. Only legal or factual insufficiency will justify rejection. The trial court should not have refused to instruct the jury on Mireles’s theory of the case.
Mireles provided sufficient evidence that Dr. Broderick had exclusive control within the meaning of res ipsa loquitur. Dr. Broderick contended, and the trial court agreed, that Mireles could not satisfy the element of exclusive control because more than one doctor had control of Mireles’s body while she was unconscious. Dr. Broderick argues that the requisite control must be “sole” control, citing to Waterman v. Ciesielski 87 N.M. 25, 26-28, 528 P.2d 884, 885-87 (1974), and Begay v. Livingston, 99 N.M. 359, 363, 658 P.2d 434, 438 (Ct.App.1981), rev’d on other grounds, 98 N.M. 712, 652 P.2d 734 (1982). In Tipton v. Texaco, Inc., 103 N.M. 689, 697, 712 P.2d 1351, 1359 (1985), and recently in Trujeque v. Service Merchandise, 117 N.M. 388, 872 P.2d 361 (1994), we expressed that the meaning of “exclusive control” in res ipsa loquitur cases is fact specific within any given case. “The essential question becomes one of whether the probable cause is one which the defendant was under a duty to the plaintiff to anticipate or guard against.” Restatement (Second) Torts § 328D cmt. g (quoted in Trujeque, 117 N.M. at 401, 872 P.2d at 364). Dr. Waring testified that it was the ultimate responsibility of an anesthesiologist (Dr. Broderick, in this case) to ensure that a patient’s arm was properly padded and positioned and to maintain the arm in the proper position during surgery. This testimony satisfied the exclusive control evidentiary requirement such that the question of exclusive control should have gone to the jury.
Conclusion. Because Mireles provided evidence that entitled her to proceed to the jury on the theory of res ipsa loquitur and tendered a legally correct instruction, the court erred in refusing to give the tendered instruction or an edited version thereof. We reverse the Court of Appeals and the trial court and remand for a new trial.
IT IS SO ORDERED.
BACA and FROST, JJ., concur.