Opinion

WIENER, J.

This is an appeal by the estate and heirs of Dr. O. Dale Buckwalter (hereinafter the Buckwalters) from a judgment entered against the Buckwalter estate in favor of Airline Training Center 1 and the estates of Ulrich Euler and Holger Behrens arising out of a 1977 midair collision of two private planes near Needles, California. Buck-waiter, Euler and Behrens all died in the accident. Both sets of parties *550 filed complaints against each other for wrongful death alleging negligence of the respective pilots. The liability issue was tried separately from the question of damages. A jury returned a verdict finding that only Dr. Buckwalter was negligent in causing the crash. We have concluded that two errors in the trial of the liability issue mandate reversal and remand for a new trial.

Factual and Procedural Background

The midair collision which forms the basis of this action occurred at approximately 7:23 a.m. on March 15, 1977, near the uncontrolled airport at Needles, California. The aircraft involved were a Mooney M20F piloted by Dr. Buckwalter and a Beech 35-F33A flown by Ulrich Euler with observer pilot Holger Behrens also on board. Euler and Behrens were both German nationals undergoing training to become commercial pilots for Lufthansa Airlines.

There were no eyewitnesses to the accident. Both aircraft were on approach to the Needles airport at the time the collision occurred. Euler and Behrens were enroute to Needles from the Airline Training Center (ATC) in Lichtfield, Arizona. Euler received his U.S. private pilot’s license on February 1, 1977, approximately one and one-half months before the accident. Buckwalter was a dentist with offices in both El Cajon and Needles. A private pilot for some 18 years, Buckwalter was enroute to Needles from El Cajon when the accident occurred. During this 13 years of dental practice in Needles, Buckwalter made the flight from El Cajon at least once a week.

Because there is no control tower and/or radar at the Needles airport, the only definitive information regarding the position of the planes, in the minutes prior to the collision was obtained from the records of the communications between the pilots and the field service station (FSS) at the airport, which provides incoming aircraft with weather and landing condition advisories. According to those records, the ATC Beech aircraft reported at 7:08 a.m. that it was 20 miles east of the airport. The FSS operator recommended a landing northbound on Needles runway number 01. 2 At 7:13 a.m., the Beech aircraft radioed the FSS *551 that it would not use the recommended runway number 01, but rather would use runway 19. At 7:16 a.m., Buckwalter’s Mooney aircraft reported its position to the FSS as 15 miles southwest of the airport. The FSS advised Buckwalter that the recommended runway was runway 01. At 7:17 a.m., the Beech pilot or co-pilot advised he had changed his mind and that the ATC aircraft would now use the recommended runway 01. The FSS issued a traffic advisory to the Beech regarding the presence of the Mooney aircraft in the area. Immediately following, the FSS advised the Mooney as to the last reported position of the Beech. The Beech reported downwind at 7:19 a.m., 3 which was the last radio transmission received from or broadcast to either aircraft. At 7:23 a.m., the locator beacon signal sounded indicating that the aircraft had crashed.

Since there were no eyewitnesses, both parties employed accident reconstruction experts to determine the circumstances surrounding the crash. Both experts relied on a wreckage plot diagram prepared by a National Transportation and Safety Board official on the day after the accident. They also examined pieces of the wreckage and reviewed the FSS communication records. Predictably, these two experts came to diametrically opposed conclusions as to how the collision occurred. The Buckwalter expert, Ivan Stracener, concluded that the Mooney was northbound on final approach for landing when it was struck by the Beech aircraft which was inexplicably flying in a westerly direction. The ATC expert, Norman Horton, suggested that the Beech aircraft was northbound on final approach when it collided with the Mooney aircraft flying inexplicably due east.

In addition to the above mentioned evidence, Stracener based his opinion on the testimony of Beverly Buckwalter, wife of the deceased dentist, who had flown with her husband to Needles on over 200 occasions. Mrs. Buckwalter testified that Dr. Buckwalter always used the recommended runway. Stracener also reviewed the pilot training records for Euler and Behrens, concluding they demonstrated a marked *552 propensity for becoming confused during the landing procedure. Although he was allowed to testify as to his opinion that the ATC pilots had become confused when they switched from a planned landing, on runway 19 to a planned landing on the recommended runway 01, the trial judge refused to allow Stracener to testify as to the basis in the training records for that conclusion, holding that it was inadmissible character evidence. (See Evid. Code, §§ 1101 and 1104.)

In reaching his conclusions, Horton also relied on the hearsay statements of Eddie Loera, one of the managers of the Needles airport. Loera told Horton that Dr. Buckwalter had a habit of using runway 28 (the westbound runway) whenever there was no wind. Horton accordingly concluded that Buckwalter was on the downwind eastbound leg of his approach when the crash occurred. The court permitted Horton to testify as to Loera’s hearsay statements as forming the basis for his expert opinion. Based on this extremely limited factual record, the jury apparently accepted Horton’s reconstruction of the accident and returned a verdict finding that only Dr. Buckwalter was negligent in causing the collision.

Discussion

Among other things, the Buckwalters argue that the trial court erred in refusing to allow their reconstruction expert, Ivan Stracener, to rely on the training records of Ulrich Euler and Holger Behrens in explaining his conclusion that they became confused following their decision to switch runways for the upcoming landing. Additionally, the Buck-waiters claim that the trial court improperly excluded the highly relevant deposition testimony of an ATC employee 4 which corroborated Stracener’s version of the accident.

Relying on the ATC pilot training records, the Buckwalters argue that Ulrich Euler and Holger Behrens had previously and

*553 consistently demonstrated severe shortcomings and deficiencies in the skills necessary to make a proper and safe approach and landing at an airport such as Needles. 5 They assert two independent bases for placing this information before the jury. While we agree with the trial court that the past behavior of Euler and Behrens clearly does not suggest the uniformity and consistency of response and similarity of condition necessary to qualify as “habit” evidence (see Evid. Code, § 1105; Witkin, Cal. Evidence (2d ed. 1966) § 352, p. 312; see also Wilson v. Volkswagen of America, Inc. (4th Cir. 1977) 561 F.2d 494, 511), we think the paucity of evidence available in this case would allow the Buck-waiters’ accident reconstruction expert, Ivan Stracener, to reasonably rely on the ATC training records in forming his opinion and to explain that reliance to the jury. (See Evid. Code, § 802.)

Prior to Stracener’s testimony, the Buckwalters’ counsel informed the court that Stracener had relied on Euler’s and Behrens’ training records in arriving at a reconstruction of the accident. Counsel explained that Stracener’s reconstruction of the route flown by the ATC Beech aircraft was based in large part on a review of landing procedures taught at ATC. Stracener then studied the students’ training records to determine how closely their previous performances matched the ATC guidelines. Based on this review, Stracener concluded it was not unlikely that one or both of the Beech pilots became confused following their decision to switch from runway 19 to runway 01 for the planned landing.

Evidence Code section 801, subdivision (b) provides that an expert’s testimony may be “[b]ased on matter ... whether or not admissible, that is of a type that reasonably may be relied upon-by an expert in forming an opinion .. . .” (Italics supplied; see Kelley v. Bailey (1961) 189 Cal.App.2d 728, 737-738 [11 Cal.Rptr. 448].) The reasonableness of an expert’s reliance is a question of degree, and may well vary with *554 the circumstances. Where, as here, there is little or no direct evidence upon which the expert can base an opinion, the expert may have to turn to forms of circumstantial evidence on which he might not otherwise rely. In such circumstances, the necessity for the information dictates that courts accord to experts somewhat greater latitude in sources of information than might otherwise be the case. (See generally Jefferson, Cal. Evidence Benchbook (1972) § 29.4, p. 509.)

The circumstances of the instant case demonstrate the need for such latitude. With no eyewitnesses and very little radio communication data, the accident reconstruction experts were forced to attempt to recreate the routes flown by the respective airplanes even though the wreckage plot strongly suggested that at least one of the aircraft was not where it should have been. Since the ATC training records indicated instances of past confusion on the part of both Euler and Behrens during landing approaches, they provided the experts with at least one possible reason why one of the planes might have been out of position. Although the trial court certainly retains discretion to limit the amount of detail to which the expert may testify in explaining the basis for his opinion (see Grimshaw v. Ford Motor Co. (1981) 119 Cal.App.3d 757, 788-789 [174 Cal.Rptr. 348]), Ivan Stracener was entitled to rely on the ATC training records of Euler and Behrens 6 and should be allowed on retrial to explain the general basis for that reliance to the jury. (See generally People v. Johnson (1970) 5 Cal.App.3d 844, 847 [85 Cal.Rptr. 238].)

Another issue which will arise on retrial involves the admissibility of the deposition testimony of the chief pilot and managing employee for the ATC, Willis J. Ennis. (See ante, fn. 4.) Shortly after the crash, Ennis was notified and he immediately flew to Needles from Lichtfield, Arizona to investigate. He personally surveyed the accident site and flew what he surmised were the flight paths of both aircraft.

In a deposition taken on June 18, 1979, Ennis “speculated” as to the probable flight paths of the two planes. Significantly, he concluded that *555 at the time the Beech pilots changed their minds as to runways, they had neither sufficient time nor space to properly realign themselves for an approach and landing on runway 01. He therefore opined that the ATC pilots had failed to follow recommended procedures which placed them in an unexpected and improper approach to the Needles airport at the time of the collision.

Although it is true that Ennis speculated as to probable flight paths, his qualifications as a pilot training and accident investigation expert were never challenged. 7 While Ennis’ testimony is perhaps more colorful, 8 we see little to distinguish his “speculation” from the “opinions” of the other experts in this case, except perhaps that Ennis was the only one to personally visit the scene and view the wreckage. And while Ennis’ testimony was certainly prejudicial to ATC’s case, neither ATC nor the trial judge suggested why it was improperly so. Given Ennis’ qualifications and his intimate familiarity with ATC procedures, the testimony must be deemed to be significantly probative. (See generally Kessler v. Gray (1978) 77 Cal.App.3d 284, 292 [143 Cal.Rptr. 496].) We are also not unimpressed by the fact that if Ennis’ deposition testimony is so prejudicially speculative, ATC is always free to call him as a witness, thus depriving the Buckwalters of the direct use of the deposition.

Disposition

The judgment is reversed.

Staniforth, Acting P. J., and Reed, J., * concurred.

The petition of respondent Airline Training Center for a hearing by the Supreme Court was denied October 28, 1982. Mosk, J., was of the opinion that the petition should be granted.

Also joined as parties were Pacific Southwest Airlines and Lufthansa German Airlines, joint operators of Airline Training Center.

Physically, the Needles airport has two runways which come together in the shape of a “T.” But because landings may be made from either direction, there are in fact four runway patterns which are utilized.

The vertical portion of the “T” runs approximately north-south. When a plane lands northbound on the runway, its compass heading is 10 degrees. For that purpose, the *551 runway is designated runway 01. Landing on the same physical runway, but in the opposite direction, the plane’s heading would be 190 degrees. Thus, for southbound landings, the same runway is desigated runway 19.

The “downwind” leg is part of the standard approach pattern at uncontrolled airports and involves flying parallel to the runway in the opposite direction from the planned landing at a distance of approximately one mile from the airport. When the aircraft is one to one and one-half miles past the airport, a 90 degree bank is executed placing the plane on a mile-long “base” leg of the approach. The aircraft then executes another 90 degree turn to align with the desired runway and make the landing.

Willis J. Ennis, the chief pilot and managing employee of the ATC, is a resident of Arizona and unreachable by the subpoena power of a California superior court. ATC refused a request to produce Ennis voluntarily so that he might testify at trial. The trial court ruled Ennis unavailable so that his deposition testimony might be used by the Buckwalters, but then excluded his comments relating to the position of the two aircraft prior to the collision, concluding that they constituted speculation and that the probative value of Ennis’ opinion was “substantially outweighed by the bent to mislead.” (See Evid. Code, § 352.)

For example, an ATC instructor’s comments for Euler on January 13, 1977, stated, “Weak area was the traffic pattern and landings. Pattern position and air speed control was very poor and did not seem to use any of the previous landing techniques learned.” On January 14, the instructor wrote, “Problem with understanding all radio transmissions from approach control. [¶] Did not ask for repeat or indicate to the controller that he did not understand the instructions ...”

Behrens’ training records indicate even more serious deficiencies. A January 18 entry stated, “Student either does not plan thoroughly or completely forgets what he had planned when approaches a strange airport ... [H]e was totally incapable of complying with instructions and was completely confused as to what to do, to the extent he was neither flying the aircraft nor responding to tower instructions ...”

ATC claims that the training records of Holger Behrens are irrelevant since the testimony indicated that Ulrich Euler was the pilot-in-command and was in fact flying the aircraft. On retrial, therefore, any reference by Stracener to Behrens’ training records must be preceded by a foundational showing of their relevance. We leave the determination of the sufficiency of that showing to the sound discretion of the trial judge.

Although Ennis’ deposition testimony strongly suggests adequate qualifications, we do not mean to preclude an appropriate challenge on retrial. (See Evid. Code, § 720.)

One portion of Ennis’ testimony took the form of role-playing: “And I can just see the doctor. He’s out so wide [thinking], ‘I’m not going to run into anybody out this far.’”

Assigned by the Chairperson of the Judicial Council.