FRIEDMAN, J.
Plaintiff, a passenger on a scheduled plane flight, sues the airline as the result of a landing accident. He appeals after a defense verdict and denial of his motion for new trial.
At the trial plaintiff relied upon the inference of negligence drawn from the res ipsa loquitur doctrine, particularly as expressed in passenger actions against carriers. The airline produced evidence designed to show exercise of care on its part and to establish a
‘ ‘
non-negligent ’ ’ explanation for the accident. The jury were fully instructed on res ipsa loquitur. With one minor exception, no procedural errors or mishaps are urged. There is no claim that the jury’s verdict might
have been based on a negative finding of damage. The issue on appeal is presence or absence of substantial evidence to support the jury’s implied finding that the res ipsa loquitur inference had been overcome.
Plaintiff was a passenger on a Trans World Airlines jet plane leaving Los Angeles at 1:30 p.m. April 29, 1960, and scheduled to arrive in Kansas City at 6 p.m. The jet, a Boeing 707, had a passenger capacity of 117 and carried a crew of eight, consisting, of four officers and four stewardesses. On this particular flight the plane carried a full crew plus 101 passengers and a cargo of air freight. Captain Eugene Gerow, a transport pilot of extensive experience, was in charge of the plane. At the time of the Los Angeles departure, the Kansas City weather forecast called for rain and for a visibility ceiling between 1,500 and 2,000 feet. By the time the flight arrived in the vicinity of Kansas City, the weather had deteriorated and the ceiling was only 500 feet. This condition required a “blind’’ approach by instrument landing system (abbreviated ILS). The flight plan designated St. Louis as an alternate destination if unsafe conditions prevented a landing at Kansas City.
The Kansas City airport is bounded on three sides by the Missouri River. Runway 18 was 7,000 feet in length and was the only runway equipped for an ILS approach. The runway lay in a north-south direction, but an ILS approach could be made only from the north. A levee was situated between the south end of the runway and the Missouri River. There were markers 5.5 and 1.5 nautical miles from the north edge of the runway. The wind at the time of the landing was shifting between north and east, its velocity varying between 6 and 12 knots. Normally a plane would land against the wind. Since the ILS approach necessitated a landing from north to south, this particular flight would have to land with the wind. TWA landing regulations for this type of plane, as approved by the Federal Aviation Agency, provided at that time that the landing on this runway should not be attempted with a tailwind in excess of 10 knots per hour. Minimum landing speed for this particular plane was 135 knots. The plane was equipped with four stopping mechanisms for ordinary use in landings:
(1) Flaps, or downward extrusions from the trailing edge of the wing.
(2) Spoilers or speed brakes, which are extensions upward from the top of the trailing edge of the wing. These are four
in number, one inboard and one outboard on each wing. Their primary function is to change the contour of the wing, thereby creating a drag against air resistance.
(3) Hydraulic brakes on the eight wheels of the main landing gear. These brakes are equipped with an antiskid mechanism based upon the factor that maximum braking power is achieved just at the point before the wheel commences to skid. When the brakes are applied to the point where the wheels commence to skid, this electrical mechanism causes the brake to release automatically. The hydraulic pressure exerted through the brake pedal then reapplies the brakes so that the wheel is constantly at or near the skid point. Thus the brakes are kept in a cyclical action which maintains the maximum efficient friction load on the wheels. Annunciators or flashing lights on the instrument panel demonstrate the cyclical braking and releasing action. These annunciators are located above the heads of the pilot and copilot, being monitored by the flight engineer. When the annunciators indicate that the wheels are skidding slightly or only occasionally, the maximum braking effect is being achieved.
(4) Reversal of the direction of the engine jet flow. Simultaneously with the application of the brakes with his foot, the captain places the jet engines in “reverse thrust” by pulling individual thrust levers, one for each engine. These levers are mounted on a single standard and are physically manipulated by the left hand alone. The levers may be pulled to a position of maximum reverse thrust, where the full power of the jet engines is applied in reverse, or to a position of normal reverse thrust, which achieves about 90 per cent of the power of the maximum position. Maximum reverse thrust overheats the engines and is used only in emergencies. The four jet engines are numbered consecutively from One to Four and from left to right. Thus engines One and Four are the outboard engines, while engines Two and Three are the inboard engines.
Additionally, there was an emergency braking system referred to as the “air bottle.” By releasing a compressed air container, the pilot could cause all the wheels to lock and put the plane in a skid. This system is used only in emergencies, because the pilot loses directional control of the plane when it skids.
When the April 29 TWA flight approached Kansas City, it was placed under orders from the Kansas City approach control until it was cleared for landing. Approach control kept
the plane circling for 15 or 20 minutes at approximately 2,500 feet of elevation until it was cleared for landing. At that point it came under control of the Kansas City airport tower. The plane crew requested and received frequent wind reports from the airport control tower. When the plane was near the outer marker (5.5 nautical miles from the airport) the tower reported a 12-knot wind from the north, which was in excess of the maximum permissible tailwind. The plane, however, was not “committed to land” at that time. As Captain Gerow continued the approach, he received additional wind reports. At the 1.5-mile marker, the wind had dropped to 10 knots. Since all operating conditions were at or above those specified for safety, Captain Gerow decided to proceed with the landing. The plane broke through the ceiling about 3,000 feet from the approach end of the runway. From this point onward, good visibility existed. The runway was wet, but there was no standing water on it. The jet touched down 400 feet from the beginning of the runway. At that point the actual tailwind velocity was 8 knots, 2 knots below maximum. Speed at point of touchdown was 138 knots, a normal speed and slightly above the minimum permissible speed. Flaps having been extended during the descent, the speed brakes or spoilers were raised immediately upon touchdown. Captain Gerow applied the wheel brakes. He noticed that the pedal pressure was “soft and rubbery” as though the pedal were “dancing.” This action indicated that, because of the wet and comparatively slick runway, the anti-skid mechanism was releasing and resetting the brakes at very fast intervals. The annunciators likewise were blinking rapidly. Captain Gerow “backed off” on the pedal to reduce the skidding and secure more braking power. According to his testimony he was achieving the maximum possible braking effect on the wet runway. There was no brake failure; rather, the slick condition of the runway prevented the brakes from exerting the stopping power available on a dry runway.
Simultaneously with application of the wheel brakes, the captain commenced the process of placing the engines in normal reverse thrust position. Engine Nos. One, Two and Three went into normal reverse thrust immediately, but engine No. Four, the outboard engine on the right wing, did not go immediately into reverse because the thrust lever stuck. To balance the thrust of the plane, Captain Gerow placed engine No. One in an idling position, then tugged on No. Four lever until it went into reverse thrust. The time lag caused by the
sticky lever was about five seconds. Within 8 to 10 seconds after landing, all four engines were in normal reverse thrust.
The plane continued down the runway for another 8 to 10 seconds. At that point, with approximately 3,500 feet of runway remaining, the plane speed was between 90 and 100 knots, and Captain Gerow realized that he was not getting adequate deceleration. He placed the engines in maximum reverse thrust, but this step did not accomplish enough speed reduction. With 1,400 feet of runway remaining, Captain Gerow decided that it was not possible to bring the plane to a stop within the length of the runway and that he would have to turn to one side in order to avoid crashing into the levee at the runway’s south end. He commenced to search for an appropriate turnoff point. Rain had turned the earth soft and muddy. With 400 feet of runway remaining, Captain Gerow turned the plane off the paved runway and onto the soft earth. Just before its turn, the plane was traveling at about 40 knots (approximately 46 miles per hour). The plane traveled 170 feet across the soft earth, slowing constantly all the while. The suction point of the jets threw mud up against the windows. At its stopping point the wheels sank a foot into the mud. Various crew members and passengers testified that the stop was quite smooth. Nothing had been thrown around inside the passenger area or the cockpit. None of the passengers, including plaintiff, reported any injuries. According to his testimony, plaintiff had been thrown forward and had struck his head on the seat in front of him, discovering the next day that he had suffered a whiplash injury.
Asked to explain the plane’s failure to achieve adequate speed reduction within the length of the runway, Captain Gerow testified: “Undoubtedly something associated with the weather. ... I am practically convinced that it was some kind of a sheer [sic] wind coming in off the river, quartering tail wind angle blowing across from the river. Now, the tower is on the other side, so to speak, from the river and they would be the last to get a wind shift like that. ... It is more than a gust, it would be called a sheer [sic] condition—a jet—a small jet stream of air would be traveling along on its own, so to speak, down the river. ”
Defendant does not suggest that res ipsa loquitur is inapplicable to this kind of accident involving a scheduled airliner. (See
Gafford
v.
Trans-Texas Airways,
299 F.2d 60;
Lobel
v.
American Airlines,
192 F.2d 217, cert. den. 342 U.S. 945 [72 S.Ct. 558, 96 L.Ed. 703]; Harper and James, The Law of
Torts, § 19.12, p. 1106; Res Ipsa Loquitur in Aviation Accidents, 6 A.L.R.2d 528-547; Note,
The Liability of Airlines for Injuries to Passengers,
31 So.Cal.L.Rev. 319, 324.) The doctrine creates an inference that the accident was caused by negligent conduct of the defendant, imposing upon the defendant the obligation of going forward to rebut the inference.
(Burr
v.
Sherwin Williams Co.,
42 Cal.2d 682, 688 [268 P.2d 1041];
Kohl
v.
Disneyland, Inc.,
201 Cal.App.2d 780, 783 [20 Cal.Rptr. 367].) The defendant’s evidence need only offset or balance the inference of negligence, and he need not prove himself free from negligence by a preponderance of proof.
(Williams
v.
City of Long Beach,
42 Cal.2d 716, 718 [268 P.2d 1061].) A defendant has alternative methods by which to rebut the inference. First, he may offer a satisfactory explanation of the accident, that is, a definite cause for the accident in which no element of negligence on his part inheres. Alternatively, he may show such care in all possible respects as necessarily to lead to the conclusion that the accident could not have happened from want of care, but must have been due to some unpreventable cause, although the exact cause is unknown.
(Dierman
v.
Providence Hospital,
31 Cal.2d 290, 295 [188 P.2d 12];
Talbert
v.
Ostergaard,
129 Cal.App.2d 222, 228 [276 P.2d 880].) In a suit by a passenger against a carrier, the latter must show exercise of the utmost care and diligence, and adequacy of the defendant’s rebuttal proof must be measured accordingly.
(Hardin
v.
San Jose City Lines, Inc.,
41 Cal.2d 432, 437 [260 P.2d 63];
Kohl
v.
Disneyland, Inc., supra,
201 Cal.App.2d at p. 784.)
Res ipsa loquitur in the hands of a jury is one matter; in the hands of an appellate court, another. Whether the inference of negligence is rebutted by the defendant’s evidence is a question of fact for the trier of fact.
(Scott
v.
Burke,
39 Cal.2d 388, 398-399 [247 P.2d
313]; Kohl
v.
Disneyland, Inc., supra,
201 Cal.App.2d at p. 784.) Appellate judges cannot substitute their own “verdict” for that of the fact trier. If the record discloses substantial evidence in rebuttal of the res ipsa loquitur inference, the verdict of the jury must be upheld.
(Kohl
v.
Disneyland, Inc., supra,
201 Cal.App.2d at p. 784; see also,
Ybarra
v.
Spangard,
93 Cal.App.2d 43, 46 [208 P.2d 445].) The jury, however, cannot arbitrarily disregard the inference of negligence; their instructions are to find for the plaintiff if the defendant fails to meet or balance the inference.
(Burr
v.
Sherwin Williams Co., supra,
42 Cal.2d at p. 691;
Druzanich
v.
Criley,
19 Cal.2d 439, 444 [122 P.2d 53];
Gerhardt
v.
Fresno Medical Group,
217 Cal.App.2d 353, 361 [31 Cal.Rptr. 633].) Where the res ipsa loquitur inference is properly drawn, an appellate court will set aside a defense verdict if there is no substantial evidence to rebut the inference.
(Dierman
v.
Providence Hospital, supra,
31 Cal.2d 290;
Gerhardt
v.
Fresno Medical Group, supra,
217 Cal.App.2d at p. 361.)
The appellate court’s approach to its task of limited review is described in
Gerhardt
v.
Fresno Medical Group, supra,
217 Cal.App.2d at page 361: “The first question that naturally comes to mind is what constitutes substantial evidence? The term is difficult to define as an isolated concept because the evidence to be weighed as well as the issue which the evidence is professed to prove is always peculiar to the individual case. The United States Supreme Court has said that ‘Substantial evidence is more than a mere scintilla. It means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion. ’
(Consolidated Edison Co.
v.
National Labor Relations Board,
305 U.S. 197, 229 [59 S.Ct. 206, 217, 83 L.Ed. 126, 140].) California courts have approved the reasonable-mind standard in testing the sufficiency of evidence. For example, in the early case of
Houghton
v.
Loma Prieta Lumber Co.,
152 Cal. 574 [93 P. 377], the Supreme Court, in discussing the term ‘substantial evidence ’ uses the test of whether the evidence would appeal to a fair and reasonable mind. ’ ’
Pointing out that Captain Gerow was an expert transport pilot whose opinion was entitled to weight, defendant puts forth his opinion testimony of a “shear wind” or “jet stream” as substantial evidence which the jury, as trier of fact, could have accepted as an explanation for the accident. In order to meet or overcome the inference of negligence, an explanation must offer a
definite
cause for the accident.
(Dierman
v.
Providence Hospital, supra,
31 Cal.2d at p. 295; see Prosser,
Res Ipsa Loquitur in California,
37 Cal.L.Rev. 183, 227-229.) Defendant cites
Williams
v.
City of Long Beach, supra,
42 Cal.2d 716,
Danner
v.
Atkins,
47 Cal.2d 327 [303 P.2d 724], and
Kohl
v.
Disneyland, Inc., supra,
201 Cal.App.2d 780, in support of its argument that Captain Gerow’s opinion constitutes substantial evidence. Without pausing for extended discussion of these decisions, we simply point out that in each of them the court found substantial evidence to support the finding that the defend
ant had exercised care in all respects, notwithstanding a speculative coloration in some of the defense evidence. In none of these cases was the court concerned, as we are at this point, with adequacy of the evidence to satisfy the “definite cause” concept, available as an alternative method of meeting the res ipsa loquitur inference.
The demand for definite causation must be observed with some literalness. An explanation drawn out of thin air, so to speak, will not do. The defendant must offer tangible evidence that the accident was caused by an agency beyond his control. Opinion testimony, particularly in scientific or technical fields, may run the gamut from the concrete to the abstract, from the actual to the remotely possible. It may be expressed as a scientific certitude drawn from known facts, or only in terms of refined guesswork based upon partial data or assumptions. The
problem is
not one of admissibility (see 7 Wigmore on Evidence (3d ed.) § 1976, p. 121), but of the opinion’s weight in counterbalance to the gravitational force of the res ipsa loquitur inference. Opinion testimony resting upon speculation, or even upon a balance of probabilities, is entitled to no weight at all against direct evidence.
Direct evidence may itself be outweighed by the circumstantial force of the res ipsa loquitur inference.
(Ybarra
v.
Spangard, supra,
93 Cal.App.2d at p. 46.)
Since the res ipsa loquitur inference is permitted to outweigh direct evidence, while a conjectural opinion may not, then in all logic the conjecture cannot offset the inference. The victim of an accident which does not ordinarily happen in the absence of negligence should not lose the benefit of inferred negligence simply because the defendant hypothesizes some agency over which he has no control as a possible cause of the accident. (See
Doke
v.
Pacific Crane & Rigging, Inc.,
80 Cal.App.2d 601, 607 [182 P.2d 284] ; Witkin, Cal. Evidence, § 79, p. 100.) If the law’s demand for a definite explanation on the part of the defense is to have any meaning at all, then the court—either on motion
for new trial or on appeal—may not leave adequacy of an explanation entirely in the jury’s hands, but must set the balance right by rejecting a speculation which the jury may have mistaken for a definite explanation.
Captain Gerow’s “shear wind” opinion was, at worst, nothing but theory, conjecture; at best, an appraisal of possibilities without any basis in the known facts of the Kansas City accident. His statement that he was “practically convinced” of the wind explanation was only a verbalism cloaking an educated guess. A reasonable mind could accept it as a possible explanation but not as a definite explanation. Thus it possessed no weight to balance the res ipsa loquitur inference. The defense verdict must find evidentiary support elsewhere in order to stand.
Much of defendant’s evidence was aimed at demonstrating the exercise of care in all respects on the part of the airplane crew members. This evidence tended to show that the landing conditions conformed to permissible standards approved by the Federal Aviation Agency; that tailwind velocity was at or under the 10-knot maximum; that touchdown occurred within the first 400 feet of runway, although 1,500 feet was allowed; that the temporary stickiness of one jet reversal control was a normal phenomenon on this aircraft; that the five-second delay in the application of reverse thrust to all four engines did not contribute to the accident in any event; that the pilot was justified in not resorting to maximum reverse thrust earlier in the landing process; that there was no failure of the wheel brakes; that the antiskid mechanism achieved maximum stopping power possible on the wet runway; that the runway, while wet, was not unusually so, and had no standing water on it; that Captain Gerow’s action in turning the plane off the runway was a skillful improvisation which saved plane and passengers from more serious consequences; that the plane actually came to a smooth stop on the muddy field. All this was substantial evidence which supported the jury in rejecting in
ferred negligence in the actual process of the landing and in the mechanical condition of the aircraft.
Defendant, however, had the burden of going forward with evidence that it exercised care as to all preventable causes of the accident.
(Dierman
v.
Providence Hospital, supra,
31 Cal.2d at p. 295.) Its burden was augmented by the passenger-carrier relationship, which imposed on it the duty of exercising the utmost care.
(Hardin
v.
San Jose City Lines, Inc., supra,
41 Cal.2d at p. 437.) Evidence that the airline employees conformed with standard operating procedures did not cover all possible causes of the accident. The procedures themselves were a possible cause. The burden of care rested not only on the pilot and crew in conforming to established standards, but also on the carrier itself in the formulation of these standards. The carrier’s own rules did not as a matter of law fix the standards by which its negligence was to be determined.
(Smellie
v.
Southern Pac. Co.,
128 Cal.App. 567, 579 [18 P.2d 97, 19 P.2d 982].)
TWA’s operating manual for the Boeing 707 permitted a landing on Kansas City runway 18 with a maximum tailwind of 10 knots. The evidence demonstrates that the relative wetness of the runway lowered the speed at which the antiskid mechanism was activated, thus materially extending the space requirements for the plane’s landing roll. The tailwind allowance was the same for a dry or wet runway. Conceivably, care in formulation of the standard would demand a lower tailwind maximum for wet or slippery runway conditions. Under these circumstances, the soundness of the landing regulations was an issue in the case. Absence of evidence of care in formulating this operating standard would be a gap in the line of defense against inferred negligence.
This landing regulation had been submitted to and approved by the Federal Aviation Agency. Federal law empowers the Administrator of the Federal Aviation Agency to adopt regulations and minimum standards in the interest of safety in air commerce. (49 U.S.C. § 1421.) The agency’s regulations require that each air carrier maintain an approved flight manual for each type of transport airplane it operates. (14 Code of Fed. Regs., §§ 40.50, 40.53.) “Approved” means approved by the Administrator of the Federal Aviation Agency.
(Ibid.,
§ 1.1.) The manual must contain “landing and take-off mínimums” for each airport.
(Ibid.,
§ 40.51(a) (7).)
Compliance with a safety standard fixed by law or administrative regulation is a fact which a jury may weigh in determining the issue of care. (See
Beeks
v.
Joseph Magnin Co.,
195 Cal.App.2d 73, 79 [14 Cal.Rptr. 877]; 2 Harper & James, The Law of Torts, § 17.6, p. 1014; Prosser on Torts (2d ed.) pp. 163-164.) Safety regulations of the Federal Aviation Agency have been regarded by the courts as impartial and authoritative criteria for judging the conduct of persons covered by these regulations. (See
Prashker
v.
Beech Aircraft Corp.,
258 F.2d 602 [76 A.L.R.2d 78];
Moody
v.
McDaniel,
190 F.Supp. 24.) While the TWA operating manual for the Boeing 707 did not have the status of a governmental safety regulation, it had been submitted for expert examination and review by the agency delegated by law to achieve safety in air transport. Federal Aviation Agency approval of the TWA flight manual, and specifically the maximum tailwind regulation, provided the jury with an authoritative standard by which to measure not only the affected conduct but also adequacy of the regulation itself. Federal Aviation Agency approval constituted substantial evidence which the jury could accept as evidence of care in the formulation of the company’s operating standard.
Carried to an extreme, the demand for evidence of care as to all preventable causes of the accident would call for an examination of almost unlimited causal possibilities, including those which hindsight alone might eliminate. Our examination of the trial record leads us to conclude that the defense fairly and candidly produced an array of evidence reasonably designed to demonstrate that it had not been negligent in any respect. There was substantial evidence to justify the jury in concluding that the inference of probable negligence had been balanced or overcome.
Judgment affirmed.
Pierce, P. J., and Sehottky, J., concurred.