This case arises out of a highway accident that occurred on U. S. Highway No. 82 in Bowie County, Texas, on December 27, 1956. The accident involved three vehicles: a five-ton dump truck, a grain truck, and a Ford passenger automobile. The dump truck was owned by the United States and was being driven by one Rufus Springfield, an employee of United Transport, Inc., under a contract between the United States and United Transport whereby United Transport had agreed to transport the truck from Red River Arsenal, Texarkana, Texas, to State Arsenal, Topeka, Kansas. The grain truck was owned by R. L. Willard and was being driven by Hugh Stone. The Ford automobile was owned by Dwight Wood and was being driven by his wife, Marie Wood.

Immediately prior to the accident, the dump truck was proceeding in an easterly direction on Highway No. 82, the grain truck was proceeding in a westerly direction, on the same highway, and the Ford was also proceeding in a westerly direction, following along immediately behind the grain truck. The speed of the dump truck was estimated at 40 to 50 miles per hour, while the speed of the grain truck and that of the Ford were somewhat less.

As the two trucks approached each other the dump truck moved over into the westbound lane and struck the trailer of the grain truck a glancing blow. A collision then followed between the dump truck and the Ford. The driver of the dump truck and the driver and two other occupants of the Ford were killed instantly. A fourth occupant of the Ford sustained permanently incapacitating injuries.

Actions were commenced under the Federal Tort Claims Act1 to recover damages from the United States on the theory that the accident was caused by its negligence in either creating or failing to discover and correct a defect in the dump truck. The United States filed an answer denying negligence on its part and a third-party complaint against United Transport, Inc., alleging that the accident was caused by the negligence of the driver rather than by a defect in the truck. The third-party complaint sought recovery from United Transport for damages to the dump truck as well as indemnity with respect to any judgment rendered in favor of the original plaintiffs. United Transport answered, denying negligence on the part of its driver, and counterclaimed for $5,434.83, the amount owed by the United States for services rendered under the contract between the parties but which had been withheld pending the outcome of this litigation. The cases were tried to the court and resulted in determinations adverse to the contentions of the United States in every instance except as to United Transport's counterclaim, which was dismissed on the ground that the United States had not consented to suit. The United States appealed from the adverse portions of the judgment; United Transport appealed from the dismissal of its counterclaim.

The principal controversy at the trial related to the right front wheel of the dump truck. It is undisputed that that wheel came off the truck either before the collision or at the moment thereof. The Government's theory was that the wheel came off at the time of the collision as a result of the impact and that, consequently, the losing of the wheel could not have caused the accident. The appellees' theory, on the other hand, was that the wheel came off prior to the collision; that, after the wheel became disengaged, it passed back under the rear of the right fender and the running board of the truck, thus lifting up the right side of the truck; and that this caused the truck to lurch over into the westbound lane and collide with the oncoming vehicles. Both sides to the dispute placed their reliance primarily upon the testimony of expert witnesses, the basic facts being largely undisputed. The trial judge chose to accept the version of the occurrence presented by the appellees' expert. The principal question presented on this appeal is whether that version is so inherently incredible that the findings of fact predicated thereupon must be rejected as clearly erroneous. Rule 52(a), Federal Rules of Civil Procedure, 28 U.S.C.A.

The Government's attack upon the appellees' theory of the accident is two-pronged: first, it contends that the accident could not possibly have occurred as opined by appellees' expert because it would be physically impossible for a truck constructed as was the truck involved in this case to run over its own loose front wheel and be thrown in the direction away from that loose wheel; and, secondly, it contends that, even were it to be assumed that such an occurrence is theoretically possible, the conclusion that it did in fact occur in this particular case is untenable because such a conclusion is entirely inconsistent with the undisputed facts. These two contentions will be treated separately.

The contention that the appellees' theory of the accident is an impossible one rests upon three sub-contentions. The first of these is that, since the front fender is located further from the truck's right wheel than the truck's axle is from the highway, if the right wheel did become disengaged while the truck was in motion, the axle would fall to the highway and pull the truck to the right before the wheel reached the fender. We do not think that such a conclusion is necessarily correct. It is undisputed that when the right wheel came off it did so suddenly and with considerable force. That is evidenced by the fact that a part of the velocity housing to which the wheel was attached was broken off and remained with the wheel. This force doubtless reacted on the wheel in some manner and it is possible that it threw the right wheel backward against the rear of the right front fender. That this could happen before the axle fell to the highway, even though the distance between the axle and the highway was less than that between the wheel and the fender, seems clear. The force of inertia acting upon the axle would prevent its falling directly to the highway and increase considerably the distance it would have to travel.

The Government's second argument on impossibility is that the right, front fender and running board would have collapsed rather than have supported the truck as appellees theorize. An examination of the record, however, does not support this argument. The appellees' expert witness testified that the fender was made of metal sufficiently heavy to support the truck momentarily and that the running board, enforced as it was with brackets, could do likewise. The only contrary testimony was that neither of these parts was designed to support the truck. On that state of the record, this Court certainly cannot hold that the implied finding that the fender and running board could momentarily support the truck is clearly erroneous.

Thirdly, the Government contends that appellees' theory of the accident is impossible because, even if the right wheel did get under the right fender and lift the truck, it would certainly pull the truck to the right rather than throw it to the left. Clearly, if all other factors remain equal, the lifting of the right side of the truck would cause it to veer to the left. The Government's argument is that the right wheel would have slowed down to such an extent that it would tend to brake the right side of the truck enough to more than compensate for this tendency to the left. The extent of this braking action would be roughly proportionate to the reduction in the speed of the right wheel. There is nothing in the record to indicate how much the speed of the right wheel would have been reduced. The rear of the right front fender, where under the appellees' theory contact with the wheel was made, sloped gently down and back. While it may be assumed that some friction would have accompanied this contact which would tend to decrease the speed of the wheel, the amount of such decrease is incalculable. When the fact is kept in mind that the fender and the wheel were traveling at almost exactly the same speed, it certainly does not seem impossible that the increased friction would not be adequate to compensate for the tendency to the left caused by the lifting of the right front of the truck.

We conclude, therefore, that there is nothing in the construction of the truck which would render appellees' theory of the accident inherently impossible. Before coming to the Government's second contention, i. e., that appellees' theory is completely inconsistent with the undisputed facts, a fuller description of the two competing theories here would seem helpful.

The Government's theory is that the driver of the dump truck fell asleep at the wheel. In order to put this theory in perspective, it seems necessary to point out that the accident occurred in mid-afternoon, on a straight stretch of highway, approximately eight miles from the Red River Arsenal where the trip began. A great deal of testimony, impliedly credited by the trial judge, was to the effect that the driver was rested and alert when he departed on his journey. Moreover, the driver of the grain truck testified that just before the collision it appeared to him that the driver of the dump truck was wrestling with the wheel of his vehicle. Finally, the Government's own expert testified that his examination of the wreck convinced him that the driver of the dump truck was properly positioned at its wheel at the moment of impact.

The appellees' theory is that the right front wheel came off because the studs which attach the wheel to the velocity housing of the truck stripped out. There are ten studs which serve this purpose, the attachment being effected by inserting the studs into coarsely threaded holes bored in the velocity housing. The studs are secured to the wheel by means of nuts which fit fine threads on the other end of the studs.

It is undisputed that at least six of these studs came out of the velocity housing by stripping the coarse threads out of the holes. Two of the studs did not strip out but, rather, were freed when the portion of the velocity housing into which they were set broke off. The remaining two studs together with the corresponding piece of the velocity housing were never recovered.

The appellees' theory is that these studs stripped out either because they were tightened too tight or were left too loose, and it is admitted by the Government's witnesses that either could cause stripping. At the same time, it is admitted by appellees' witness that stripping can be caused by impact. The question then is: What in fact did cause this particular stripping?

The Government's first argument on this point is that, because the coarse threads were stripped out of the velocity housing rather than the fine threads out of the nuts, the stripping must have occurred as a result of impact. In support of this argument the Government points that the coarse threads could not have been stripped out as a result of overtightening of the nuts because the fine threads of the nuts would have stripped first. Next, the Government contends that the coarse threads could not have been stripped out as a result of overtightening the studs themselves, because to do so would require such a tremendous amount of pressure that it could only be accomplished by a deliberate effort to do so and with the aid of an unusually heavy tool. Finally, the Government contends that the stripping of the coarse threads could not have resulted from looseness, because such looseness would necessarily have been accompanied by evidence of wear on surrounding parts and it was clear that no such wear occurred, and, also, because any looseness would have been discovered by the inspectors at Red River Arsenal before the truck was released. Since the three contentions considered collectively cover all possibilities under the appellees' theories, if these contentions are correct the Government must prevail. On the other hand, since the appellees' theories are alternative, if any one of the three contentions fails, the Government's entire argument on this point must fail.

We consider it unnecessary to pass upon the first two of these contentions for we are satisfied that the Government cannot prevail on the third. Clearly, the Government's contention that the looseness theory must be rejected because looseness would have been discovered is no argument at all; for it assumes the whole ground of the argument to which it is directed by assuming that there was no negligence on the part of the Government employees at Red River Arsenal who were charged with the responsibility of seeing that the truck was roadworthy. The contention that looseness would have produced wear, though stronger by reason of the fact that it is not inherently fallacious, is equally unavailing. Even the Government's own witness did not testify unequivocally that looseness would necessarily be accompanied by evidence of wear not here present. His testimony is merely that such things as alleged by the appellees "generally" occur over a period of time and, when they do, are accompanied by evidence of wear. We cannot hold that the mere absence of such wear precludes a reasonable finding that the studs were loose.2

A second argument offered by the Government is rested primarily upon the fact that two studs did not strip. The Government showed that the tensile strength of each of these studs was more than sufficient to bear the weight of the entire truck and argues that, since that is so, the stripping of the six studs could not possibly have been the cause of the accident because the remaining two were easily strong enough to carry the truck. The fallacy in this argument is obvious. The issue is not whether the two studs were strong enough to carry the truck but whether the velocity housing was strong enough when all of the weight was shifted to two studs rather than distributed evenly among ten. For it is clear that the studs did not give way. The velocity housing fractured. Appellees contend that this fracture was caused by the concentration of all of the weight normally carried by the ten studs on only two of them. There is no evidence in the record to indicate that such a theory is impossible.

Moreover, an examination of the record reveals that there is considerable support for the appellees' theory. The rear of the right front fender and the running board showed evidence of damage of the precise nature to be expected under the theory that the truck had overrun a loose wheel. The rear of the fender was raised and straightened; the running board was pushed back and raised several inches. This damage becomes more significant when considered in connection with the fact that the front of that fender showed no evidence of damage. In addition, a battery which was mounted on the running board was dislodged and found lying on the right-hand side of the road and behind the point of impact, a position consistent with the appellees' theory and inconsistent with that of the Government. Finally, the wheel itself was found beside the truck rather than in front of it, as would normally be expected if it came off as a result of the impact.

The Government has, of course, attempted to explain away each of these facts. To some extent, it has succeeded. If we were sitting as the trier of facts in this case, a difficult problem would be presented. But our function is more limited. The trial court found, as a fact, that the accident was caused by the truck's overrunning of its own right front wheel. We have only to decide whether that finding is clearly erroneous. Rule 52(a), Federal Rules of Civil Procedure. We conclude that it is not.

The Government also contends that, even if the accident was due to a defect in the truck, there is no evidence to sustain a finding that the United States either created such defect or failed to exercise reasonable care to discover it. This contention too must fail. The Government's own witnesses testified that, if the claimed defect had been present, it would have manifested itself to proper inspection. This testimony, designed to negative the existence of a defect, is relevant to the question of due care once the defect is established. Again, we are unable to say that the finding of the trial court is clearly erroneous. The judgment must, therefore, be affirmed insofar as it holds the United States liable for damages suffered as a result of the accident.

There remains only the question of the appeal by United Transport. The trial court dismissed United Transport's counterclaim against the United States on the sole ground that the United States had not consented to suit. Since it is clear that the claim sought to be asserted is within the Tucker Act3 and could therefore be asserted in an original action, this Court is presented with the question of whether the jurisdiction conferred by that Act may be invoked by way of a counterclaim.

Though this is a question of first impression in this Court, other Courts of Appeals have found occasion to consider it. In United States v. Silverton,4 the First Circuit, speaking through Chief Judge Magruder, said:

"It is conceded by the government that the defendant in this case could have brought an original action in the court below against the United States for breach of contract, under the Tucker Act. If he had done so, of course the court below, under Rule 42, F.R.C.P., 28 U.S.C. could have consolidated such action with the pending action brought by the United States. It would be the emptiest technicality to hold that the same jurisdiction could not be invoked by way of counterclaim in the action already brought by the United States."

Subsequently, in Thompson v. United States,5 the Fourth Circuit said:

"Although the appeal must be dismissed for lack of jurisdiction, we think that, in the interest of a prompt disposition of the litigation involved, we should say, in line with our action in Carolina Mills, Inc., v. Corry, 4 Cir., 206 F.2d 76, 77, and Ford Motor Co. v. Milby, 4 Cir., 210 F.2d 137 that all of the judges of this court are of opinion that the counterclaims here involved may be asserted in the action instituted by the United States. They arise out of contract and involve less than the sum of $10,000. Congress has given its consent in the Tucker Act, 28 U.S.C. § 1346(a) (2), that the United States be sued on such claims in the District Courts; and we can think of no sound reason why the suit should not take the form of counterclaim filed to a suit instituted by the United States in a District Court, where the amount of the counterclaim does not exceed $10,000."

On the other hand, in United States v. Nipissing Mines Co.,6 the Second Circuit took the opposite position:

"Moreover, in our opinion, the Tucker Act of 1887 which gives the District Courts jurisdiction over certain suits against the United States, is not broad enough to permit the recovery of demands upon counter claims. We think that that statute refers to original suits and prescribes procedure inconsistent with its use as the basis of a counter claim."

Upon close consideration of the question, we think the view expressed by the First and Fourth Circuits is the correct one. We think the statute should be interpreted from a practical rather than a technical standpoint. Nothing in the statute itself precludes such an interpretation. And in a situation where no substantive rights are at stake, the only question being whether litigation is going to be disposed of in an expeditious or inexpeditious manner, we think that an interpretation which lends itself to sound judicial administration is justified.

It follows that the judgment of the trial court must be reversed insofar as it dismisses United Transport's counterclaim. On the appeal of the United States the judgments are affirmed. On the appeal of United Transport, Inc., the judgment is reversed and the cause remanded with directions to enter judgment against the United States in the amount of $5,434.83. See 28 U.S.C.A. § 2106.

Judgment against United States affirmed; judgment against United Transport, Inc. reversed with directions.

TUTTLE, Circuit Judge (dissenting).

With the greatest deference to the views of the majority, I must dissent.

The recovery allowed by the trial court was, in my opinion based on nothing more than the possibility (on an extremely remote possibility) that the government truck ran across the road to its left and into the oncoming truck because its right-hand front wheel came off due to the government's negligence.

There were several eye witnesses to everything that happened. Several of them testified at the trial. Unfortunately several of them, including the driver of the government truck, died in the crash. All of the eye witnesses testified that the Army truck angled to the left of the center line and that they did not see any wheel come off of the truck or any untoward event before the crash that brought a pile-up and fire involving two other vehicles.

In this state of facts there was no burden on the government to prove any theory as to what caused the accident. Unexplained, this state of facts would require a judgment that the injury was caused by the negligence of the driver. The opinion, it seems to me, is too much preoccupied with knocking down the government's theory of the case. It needs no theory. In McNamara v. American Motors Corp., 5 Cir., 247 F.2d 445, 447, we said:

"* * * we think the plaintiff's claim is untenable in law and in fact. It is untenable in law because it unwarrantedly assumes, contrary to settled law, that theory and speculation, as to how decedent's death occurred, can serve as evidence, satisfying plaintiff's burden to make out her case, and shifting to defendant the burden of showing by evidence that plaintiff's theory was wrong, or of coming forward with a theory of its own, as to decedent's death, and evidence showing it to be a better, that is a more plausible, theory than the one plaintiff advances."

In all reason, it is the plaintiffs' theory that must be examined, because it is only by establishing by competent proof that this theory of the occurrence is what happened that the plaintiffs could recover.

Examining the plaintiffs' theory, we find it to be that, due to some negligence at the government repair installation, some of the studs which held the right-hand front wheel were improperly attached to the velocity housing; that this improper attachment caused or permitted six of the ten studs to pull loose by stripping their threads; that this caused or permitted the wheel to come off; it then dropped back and wedged under the right fender; this raised the right side and threw the truck to the left and into the other vehicle. This theory is expounded in spite of the fact that no one saw any of these things happen, but because an automobile repair expert gave the opinion that the wheel became dismounted in such a manner. The only observable physical facts upon which this opinion was based were that after this crash, which occurred between three automobiles whose combined speed at time of impact was between 60 and 90 miles per hour, the right-hand front wheel was in fact off, and the threads of six of the studs had been stripped and two remaining studs that had not been stripped were still firmly screwed into a part of the velocity housing which had itself been torn apart, with a part of it still attached to the two studs and wheel.

I think it is self evident that the chain of causation which the plaintiffs' expert built up, embodying, as it does, inferences from the one circumstance of the loss of the wheel as a result of or before the crash, that the improper insertion of the studs rather than the tremendous impact of the crash tore this wheel from the truck with such force as to break the casting of the velocity housing and strip the heavy threads of six studs; that this wheel, when lost, rolled back and under a lower running board before the axle assembly could drop to the ground; that such action caused the front end to turn to the left instead of to the right; that such loss of the wheel was the result of some negligent act of the government at its repair shop, although it was not shown that the studs were removed or needed to be removed in the overhaul job, would be too tenuous to be the basis of a fact finding even though the known physical facts made it appear equally reasonable with any other theory. See Smith v. General Motors, 5 Cir., 227 F.2d 210; McNamara v. American Motors Corporation, supra.

Where, as here, such chain of causation is not only too tenuous to be considered, but when the only theory on which the plaintiff could support a claim of negligence is itself, as I firmly believe, inherently incredible, I am convinced that the judgment of the trial court based on such theory is clearly erroneous.

It simply seems to me so contrary to common knowledge of physical facts that the accident could have occurred as contended for by the appellees that I think a more detailed statement of the facts should be made than is contained in the opinion of the court.

Stone, the driver of the grain truck which was first struck by the government vehicle, testified that when he was about 50 to 100 feet from the Army truck he saw it angling towards him, across the center line of the road; he himself was doing 20 to 30 miles per hour; the Army truck was driving 40 to 45 miles per hour; he immediately veered to the right shoulder and in the time it would take to snap his fingers twice the truck had struck the left rear of his vehicle; at the time of the impact he was completely off the road on the right shoulder; he did not see the wheel come off and he saw no sudden jerk or any sudden elevation of the front end of the Army truck.

Mrs. Phelps, who was in a car following the truck, saw it suddenly swerve to the left across the road; she saw no wheel come off and saw nothing "mechanically defective about the Army truck as it crossed the road."

Mrs. Story, who was in an automobile following the Wood car, which was demolished, testified she just "noticed the Army truck coming over in our side of the lane," and "it just left its lane off the road it was going east and we were going west. And it just came angling over and the grain truck (Stone) swerved his cab to * * *." She testified she did not see a wheel come off of the Army truck, and, when asked whether the Army truck looked level or tilted, she said: "Well, the best I remember it looked level."

These are the only eye witnesses to the accident who testified at the trial.

An examination of the wheel showed that six of the ten steel studs, which were threaded with heavy or coarse threads into the velocity housing, had been pulled out of their threaded seating and two others were still firmly embedded in a part of the housing that had been literally torn away from the main part of the housing; two other studs were missing and not accounted for; the end of the axle assembly was not disfigured and no marks on the road indicated that it had dropped to the road while the truck was still in forward motion.

There was a slightly bent portion of the right-hand fender and running board which, according to appellees' theory, was caused by the wheel passing back and raising these elements sufficiently to cause the truck to lurch to the left. The diameter of the wheel and tire was much greater than the height of the running board from the ground, so that some unexplained force would have had to raise it sufficiently for the wheel to wedge under it enough to raise the right side of the truck. The running board was 22 inches from the pavement, and the wheel was 42½ inches tall, so that this 42½-inch wheel would have to pass under a 22-inch high running board to achieve this effect. This would, of course, have canted the cab of the truck sharply to the left, but no eye witness saw any such tilt.

Then, if we assume that all of this happened, we must next believe that instead of this jammed wheel pulling the truck around to the right because of the tremendous drag thus created, the wedging of the 42½-inch tire under the 22-inch high running board in some way heaved the truck over towards the left, and all in such a way that it would not be noticed by the three eye witnesses who saw only that the truck angled over to the left in an apparently normal fashion.

I think that no careful analysis of the physical facts, most of which were without dispute, could possibly permit the conclusion that the accident occurred as opined by the plaintiffs' single expert witness.

It is, of course, unusual for an appellate court to reverse the judgment of a trial court, even when sitting without a jury, on the weakness of the factual proof. We are, however, enjoined to do so when a finding of the trial court is clearly erroneous. I am firmly convinced that for the reasons stated this judgment was clearly erroneous, both because there was no adequate proof that would permit the making of an inference that the wheel came off before the accident and because even if there were such affirmative proof, it would be so contrary to known physical facts and physical laws as to be inherently unbelievable.

I would reverse the judgment on the main appeal and affirm on the cross-appeal.