DAVID T. LEWIS, Circuit Judge.

This appeal is taken from judgments entered by the District Court for the Western District of Oklahoma against the United States under the Federal Tort Claims Act, 28 U.S.C. § 1346(b), and in favor of the several appellees for property damage allegedly caused by sonic booms intentionally generated by the United States in a test program conducted in and over Oklahoma City. The appellate questions presented are very limited and the principal issue probes the quality and quantity of evidence necessary to sustain an affirmative finding of causation of damages occasioned by the government's wilful tort. The United States frames the issues thus:

1. Whether, as a matter of law, plaintiffs failed to sustain their burden of proving that sonic booms generated by Government aircraft caused damage to their properties.

2. Whether the district court's determination that the sonic booms caused the damage is clearly erroneous.

3. Assuming that the sonic booms did cause the damage, whether the district court erred in not finding that the damage caused was de minimis.

Nine homeowners are involved in this appeal; however it was stipulated below that if the judgment of the district court is affirmed that the government will pay sixty additional homeowners a certain stipulated amount based on the cost of repairs.

Beginning February 3, 1964 and continuing through July 30, 1964, controlled test flights of from one to eight were flown daily over Oklahoma City by aircraft flying at supersonic speeds and at altitudes of from 21,000 to 50,000 feet. The general purpose of the tests was to gain information probing the feasibility of developing supersonic commercial aircraft and the particular purpose was to measure structural response to sonic booms as well as "determine the normal reaction of ground population over a significant period of time to sonic boom pressures. * * *" Eleven test houses in the Oklahoma City area, varying in age from new to 50 years old, were used in the program. Five of the test houses had instrumentation designed to measure the response of their various structural components to the sonic booms. Instruments were also set up in special stakes driven into the ground to measure and record movement in the earth ("seismic response") caused by sonic booms. In addition, three measuring stations were established to measure the overpressure from the sonic booms. One of these measuring stations was located directly under the flight path the test aircraft were to take; the other two were located at five and ten miles distance from the flight path.

Twelve hundred and fifty-three test flights were flown over Oklahoma City during the test program. No accidental deviation from the planned format occurred. A similar program was subsequently carried out at White Sands, New Mexico, and the record reflects references to other instances of known structural response to sonic booms occurring through accidental exposure to the phenomenon. All in all, the record contains over 3,000 pages of testimony and hundreds of exhibits nearly all of which are directed to the technical expository statements and conclusionary opinions of experts in the field of seismology and the effect of vibration on various types of structural material. It is obviously impractical, and we deem it unnecessary, to detail the evidence beyond that required to focus on the legal issues.

A sonic boom is a shock wave striking the earth's surface and caused by an object, here an airplane, passing through the atmosphere at a speed in excess of that of sound. The intensity of the shock wave and the resultant vibration on the ground are affected by the speed, height and size of the aircraft and other variables of perhaps less importance such as atmospheric conditions. As a result of these recurrent booms, many types of structural damage were claimed by homeowners; however a majority of the claims were for the formation of new cracks; glass breakage; the reopening of old cracks which had been repaired prior to the booms; and the popping of nails in plasterboard. Appellees testified in some instances that particular items of damage occurred in front of their eyes at the instant of a boom; more often this testimony was to the effect that defects in the homes were not present in the structures before the booms and were found for the first time after the booms or worsened during the progress of the continuing program.

The government emphatically attacks both the legal competency and the value of this lay testimony to in any way probe the ultimate issue of causation. Starting from the premise that the threshold of human irritation from auditory and motion stimuli is much lower than the irritability threshold for structures the government argues that "in light of their lower threshold, it could seem to them that a house was about to fall apart, when in fact no structural damage was occurring" and that this would be especially true when the homeowners were forewarned, as here, of the occurrence of sonic booms and the expressed willingness of the government to pay for resultant damages, if any. We acknowledge this argument as one properly urged on the issue of credibility but that issue is one for the fact-finder and not this court. Southwestern Investment Co. v. Cactus Motor Co., 10 Cir., 355 F.2d 674, 676; Glens Falls Insurance Co. v. Newton Lumber Mfg. Co., 10 Cir., 388 F.2d 66. We reject the argument as legally negativing the competency of the witnesses or their testimony. Such testimony, if deemed credible, may be considered by the fact-finder as establishing a circumstance which in turn is pertinent to the issue of causation. The lay witnesses could not, of course, competently testify as to the forces that sonic booms create for such a subject requires specialized knowledge and understanding. However the desirability or need for expertise in testimony probing an ultimate fact does not preclude, as a matter of law, all other evidence for "expert evidence does not forcelose lay testimony concerning the same matter which is within the knowledge and comprehension of the lay witness." Stafos v. Missouri Pacific R.R. Co., 10 Cir., 367 F.2d 314, 317; Cundick v. Broadbent, 10 Cir., 383 F.2d 157, 161, cert. den. 390 U.S. 948, 88 S.Ct. 1037, 19 L.Ed.2d 1139. Although the judgments in the case at bar are dependent on acceptance of the testimony of the lay witnesses by both the fact-finder and the expert witnesses called by appellees, we hold that such acceptance is legally proper in both instances.

Appellees supplemented the lay witness factual testimony by that of experts in the fields of seismic study, geology, engineering and construction. These witnesses had each been present in Oklahoma City during the sonic boom program, had examined each home in question, and had obtained from the individual homeowners a history of the structural damage to each building. Applying their varying expertise to their visual knowledge obtained by inspection of the homes and the homeowners' account of damage occurrence these witnesses gave opinion testimony from which it could be concluded that the claimed damage was caused by vibration from sonic booms. Other possible causes of the damage, such as settling and faulty construction, were given consideration by these witnesses. Particular emphasis was given to the sub-soil geology of the Oklahoma City area which was established as highly fractured, a condition which would amplify the effect of vibration. Importance was placed on the occurrence of structural fatigue, the weakening of material from continuing exposure to vibration until finally a shock, insufficient in itself to cause original damage, will be the "straw that broke the camel's back."

The United States asserts the appellees' experts gave no testimony having probative value because they did not give consideration to the factual data obtained by the government from the Oklahoma City and White Sands programs; that the combined testimony of appellees' witnesses rises no higher than speculation for at best the evidence can only establish a possibility of sonic boom causation. And, in contrast, the United States points to its own evidence. This data, largely obtained from its tests in Oklahoma City and White Sands, summarily stated consists of the following: 1) The highest overpressure actually recorded in Oklahoma City was 4.42 p.s.f. with the mean overpressure between 1.4 and 1.5 p.s.f. and that such is comparable to non-boom forces such as thunderstorms and door slamming; 2) that in the White Sands test no damage was caused by booms of strength lower than 4.7 p.s.f. average and then ony minor extensions of the old damage and that "virtually no permanent distortion of the buildings was caused by booms of strength lower than about 24 psf"; 3) that with the exception of acceleration of hairline cracking of paint finish at corners and over nail heads in walls and ceilings the government's test houses in Oklahoma City revealed no discernable damages attributable to the sonic booms; and 4) that according to "generally accepted engineering standards" a particle velocity of at least 2.0 inches per second is required in order to cause structural damage and the highest recorded particle velocity measured in a structural element (ceiling) in Oklahoma City was 1.0 inches per second and the highest in the ground was a little over 1.0 inches per second.

Citing Galloway v. United States, 319 U.S. 372, 63 S.Ct. 1077, 87 L.Ed. 1458, and Jackson v. Nelson, 10 Cir., 382 F.2d 1016, the government asserts that expert opinion that fails to take into account basic and undisputed facts is sheer speculation. We agree to this rule as a generality but do not agree that its application is determinative in the case at bar. The experts of appellees were not bound to accept the scientific data advanced by the government as undisputed fact or as leading to scientific certainty. And appellees' experts affirmatively rejected much of the data as fact or, if fact, the scientific relevance of such fact. The White Sands data was rejected as being inappropriate for comparison with Oklahoma City because of a great difference in the geology of the two areas. The sub-soil at White Sands is stable; at Oklahoma City it is not. The potential of shock intensity magnification through the fractured sub-surface at Oklahoma City was not considered in the government data. The measured overpressures in the Oklahoma City report were not taken at the sites of claimed damage, a significant factor which the government's own exhibit highlights by the statement that "measured overpressures at the three fixed stations indicate that current prediction methods tend to * * * underestimate overpressure values occurring at 5 and 10 miles away from the flight path." We conclude that the testimony of appellees' experts was not faulty through failure to consider undisputed and determinative facts.

The remaining two contentions of the United States directed to the issue of causation — that the total of appellees' evidence showed no natural nor scientific probability of damage attributable to sonic boom but merely projected one possibility and that the court clearly erred in any event because the government evidence overwhelmed appellees' case — present problems that are unique to this case and must be pragmatically projected against the total circumstances of the case.

The government has here exposed appellees' property and persons to a deliberate tort for the very purpose of determining what, if any, damage would result to property and person. It now argues that no property damage occurred that can be proved under traditional rules probing causation and attempts to support that argument by asserting that the persons of appellees were so damaged that they could temporarily not trust their own senses of sight and hearing in determining whether their homes were damaged. Having thus eliminated lay testimony as unreliable the government then insists that causation can only be proved by experts who must rely solely on scientific data to lead them to a conclusion of causation by sonic boom and the elimination of all other potentialities such as vibrations from door slamming and thunderstorms. Appellees' experts could not testify with this certainty for such scientific data does not exist unless the White Sands and Oklahoma City tests are accepted as such without alternative. Some of the damage claimed by appellees could have been caused by disturbances other than sonic booms and this possibility could not be scientifically negated by the expert witnesses. Ordinarily, proof of causation must rise higher than the presentation of mere alternative possibilities and must reach the level of a reasonable probability. Commercial Standard Insurance Co. v. Feaster, 10 Cir., 259 F.2d 210. But appellees cannot be held to a standard of proof impossible for them to sustain under the circumstances here imposed on them by the deliberate acts of the government. The quality of proof under such circumstances can only be, and need only be, minimal. This standard was met by the experts' opinion that the subject damage, viewed in the light of the homeowners' account of their properties' history, was "likely" caused by the sonic booms.

Next, the United States asserts that the district court's determination of fact of causation was clearly erroneous in view of the high quality of its own evidence and the certainty of its experts' opinions. We agree that the government's evidence is impressive. But, as we have indicated, the results of its testing need not be accepted by the court as an absolute. Indeed, a committee appointed by the FAA Administrator at the conclusion of the Oklahoma City test program to review standards and procedures for processing of damage claims concluded that in their "opinion that there is at this time too little scientific data available to discover what type of damage the booms can cause." The trial court characterized the government's evidence as "theoretical and does not in fact negate the actual factual evidence offered by plaintiffs which the Court finds to be fully credible and true." We do not agree that the finding is clearly erroneous.

Finally, the government asserts that damage should have been found to be de minimis. This contention is again premised on the program tests which showed only hairline damage at the test sites. The claim must thus fail.

Affirmed.