delivered the opinion of the court.
This appeal is from a judgment for $5,000 damages for personal injuries received by appellee (plaintiff), a brakeman in the employ of appellant (defendant). The count of the declaration relied on is predicated on the Federal Employers’ Liability Act and the Federal Safety Appliance Act. It charges, in substance, that defendant was a common carrier engaged in interstate commerce; that at the time of the accident both it and plaintiff, its brakeman, were engaged in interstate commerce; that defendant permitted one of the couplers on the car in question to be in a defective condition so that it would not couple automatically and could not be coupled without a brakeman going* between said car and the one next to it; that plaintiff was thus obliged to go between the cars and while in the act of opening a coupler stepped and slipped on a piece of coke or coal and was caught under a wheel of one of said cars and thereby injured.
There was conflicting evidence as to the identity of the car, plaintiff’s tending to show it belonged to the “Big Four” road, and defendant’s, that it belonged to the “B. & 0.” The weight of the evidence supports defendant’s contention. But as we view it, it is im*157material to the legal questions presented, to which of the two railroads so designated the car belonged, as both were engaged in interstate commerce and the car was received by defendant from one of them for delivery over its tracks to the consignee.
After the car was unloaded it was switched by defendant to the scale track in its yards to be weighed. At the north end of the scale track was a scale used for weighing cars. Standing on this track were three empty cars, the southernmost of which was a “coke” car. In a switching movement just before the accident by a crew to which plaintiff belonged, two empty flat cars were being pushed north on this track by an engine for the purpose of coupling to said coke car the north flat car on which plaintiff was riding. At the first impact the cars failed to couple. The engine came to a stop and plaintiff went between the flat ear and the coke car to adjust the knuckle on the latter, and after adjusting it gave a signal to the engineer to go ahead, and while walking out stepped on a piece of coke or coal causing his left foot to slide under the wheel of the car just about the time of the second impact when again there was a failure to couple.
Appellant contends that the evidence fails to show that defendant was engaged in interstate commerce or that the car with the alleged defective coupler was being hauled or used on defendant’s line in moving interstate traffic. There can be no question on this record as to the applicability of both Federal acts. As the car was used on a road engaged in interstate commerce, it was required by the Safety Appliance Act to be furnished with an automatic coupler (Southern Ry. Co. v. United States, 222 U. S. 20); and as defendant received loaded freight ears from roads engaged in interstate commerce and participated in the transportation to their place of destination, it was a common carrier engaged in interstate commerce. (Devine v. *158 Chicago & C. River R. Co., 174 Ill. App. 324, and 259 Ill. 449; United States v. Colorado & N. W. R. Co., 157 Fed. 321.) Being so engaged, the additional test of the application of the Federal Employers’ Liability Act was whether the work engaged in at the time of the injury was a part of the interstate commerce in which the carrier was engaged. (Pedersen v. Delaware, L. & W. Ry. Co., 229 U. S. 146, 3 N. C. C. A. 779; Illinois Cent. R. Co. v. Behrens, 233 U. S. 473, 10 N. C. C. A. 153.) The work being done was the moving of an empty car to be weighed in order to determine the net weight of an interstate load. Upon a very similar state of facts in Wheeling Terminal Ry. Co. v. Russell, 209 Fed. 795, 799, it was held that it not having been shown that the cars being weighed had been withdrawn from service in interstate commerce, the' presumption was that they remained in it. There was no proof in the instant case tending to show any such withdrawal. Hence the same presumption must obtain.
Appellant also urges that defendant’s negligence, if any, to maintain a proper appliance was not the proximate cause of injury, and that the court should have so held- as a matter of law and have directed a verdict in its favor. The contention is that plaintiffs injury was due primarily to his negligently signaling the engineer to go ahead before he stepped out from between the cars; and secondarily, to his slipping on the piece of coal or coke. Both plaintiff’s signaling at the time and place and his slipping on the piece of coal might be regarded as mere links in the chain of causation connecting defendant’s alleged negligence to maintain properly its coupling appliance with plaintiff’s injury, and thus present a question of fact as to which different conclusions might be reached, and therefore a question for the jury. (Donegan v. Baltimore & N. Y. Ry. Co., 165 Fed. 869; Erie R. Co. v. Russell, 183 Fed. 722; Grand Trunk Western Ry. Co. v. Lindsay, 201 Fed. 836.)
*159But we agree with appellant that the preponderance of the evidence is manifestly against plaintiff’s claim that there was a defective appliance. There was no question but that the car was furnished with automatic couplers and other appliances that obviated the necessity of going between the cars to couple or uncouple them provided they were in working order. The question of fact was, was there any defect therein so that they would not work if properly handled. Plaintiff alleged the existence of a defective condition. The burden was on him to prove the allegation. He made no attempt to prove any specific defect, but relied on making out a prima facie case. .The unquestioned fact that the cars did not couple on impact was supplemented by plaintiff’s testimony that as his car approached the coke car he observed that the couplers on both were in position to couple; that after they failed to couple on the first impact he found the device on the flat ear operated properly but that the knuckle of the coke car was open, and the pin down; that he raised it, closed the knuckle and dropped the pin again, and that the cars failed to couple on the second impact, but that he “did not observe anything about the coupler as to why it opened.” The proof of a defect, therefore, consisted entirely of the" fact that the cars failed to couple automatically on impact, and plaintiff’s testimony that the couplers were in a proper position for automatic coupling. If, however, as the uncontradicted evidence shows, the couplers operated automatically immediately before and immediately after the occasion in question, the palpable inference would be that the couplers were not placed in a proper position for coupling, for no matter how perfect the mechanism unless the couplers are in a position for coupling it cannot be effected with even automatic couplers. The duty of seeing that they were in such position manifestly rested upon the brakeman. If they *160were not, then the failure to couple on impact cannot be said to be due to a defect in the appliances.
It was said in Chicago, R. I. & P. Ry. Co. v. Brown, 229 U. S. 317, 3 N. C. C. A. 826, where a violation of the Safety Appliance Act was charged,, that the Taylor case, 210 U. S. 281, and Chicago, B. & Q. R. Co. v. United States, 220 U. S. 559, settled “that the failure of a coupler to work at any time, sustains a charge of negligence in this respect.” Citing these cases as authority, counsel for appellee seems to regard the mere fáct of the failure of cars to couple on impact as conclusive of 'the question of negligence. It would be strange indeed, though it be the carrier’s absolute duty to maintain the appliances required by said act, as held in said cases, if the carrier were not permitted to show in defense that it had performed such duty and that such failure was in fact due not to a defective appliance but to something else. Such proof is quite different from reliance on want of knowledge or exercise of due diligence, neither of which is a defense where the duty alleged to be violated is absolute. But proof that cars properly equipped to couple automatically did not couple on impact, while prima facie evidence of a violation of the act, is not conclusive of the fact that the appliances were out of order when the contrary is susceptible of proof.
Appellant contends that the alleged position of the coupler and pin on the coke car after the first impact was physically improbable if not impossible. We need not discuss the contention, for whether so or not, defendant was not precluded from rebutting the inference of negligence that might be drawn from plaintiff’s proof, and we think it was completely refuted by affirmative and uncontradicted evidence that the appliances were in good working order.
Plaintiff himself testified that there was no difficulty in coupling or uncoupling this particular car on two occasions on the same day before the accident, one *161before and one after it was unloaded, and a fellow switchman testifying to the same thing said also that there was no difficulty in coupling or uncoupling said car immediately after the accident—once when it was placed on the scales to be weighed and once when it was placed and left on another track. The yard inspector and the yard manager each testified that immediately after the accident, before the coke car in question was moved or anything done thereon, he examined its coupling appliances and found them in good order. The manager also testified that he then identified the car as a B. & O. car and took its number. There was proof that a B. & O. freight car of that number was inspected on its arrival at S'outh Chicago by the ,B. & O. railroad inspector and again by another of its inspectors on January 3rd following, when it was delivered by defendant to the B. & O. railroad, and that the same car was inspected by defendant’s inspector on December 26th when it arrived in defendant’s yards as well as immediately after the accident, and that on each occasion the safety devices were found to be in good order.
Against the force of such evidence counsel for appellee argues that it was not inconsistent with an existing defect and that its credibility was for the jury to determine. In support of the first contention he calls attention to plaintiff’s testimony that the coupler in question was a “ Janney” coupler, and proceeds to argue that its construction is such that when certain parts are slightly worn the coupler will at times fail to operate on impact and on other occasions will operate without difficulty, but we find no evidence in the ■ record on which to base such argument. The difference in construction or mechanism between a “Janney” and any other type of coupler is a subject on which the record is silent and of which we have no knowledge and cannot take judicial notice.
The credibility of defendant’s witnesses was not di*162rectly assailed, there was nothing inherently improbable in their testimony and they were not impeached nor directly contradicted. The jury’s unquestioned prerogative of determining the credibility of witnesses does not justify an arbitrary rejection of testimony of witnesses who have in ‘ no way been discredited and the logical inferences therefrom, especially when it is reasonable, corroborated, uncontradicted, and opposed at the points of difference only by the unsupported testimony of a solitary, interested witness.
The failure of the cars to couple was attributable either to a specific defect not proven and which did not exist if defendant’s evidence was true, or to a mistake or misrepresentation as to the position of one or both of the couplers before or after the impact. In his routine plaintiff may not have been particularly observant and could have been mistaken, and such a conclusion is most probable in view of the unquestioned evidence that the appliances not only worked perfectly just before and after the accident, but were examined and found in perfect condition immediately after it. Hence, we can reach no other conclusion than that the verdict was manifestly against the weight of the evidence. There is nothing in the record to indicate that other evidence on this subject is obtainable. In accordance, therefore, with the duty of this court to weigh the evidence and its power to render final judgment in such a case (Borg v. Chicago, R. I. & P. Ry. Co., 162 Ill. 348; Donelson v. East St. Louis Ry. Co., 235 Ill. 625), we reverse the judgment and find as a fact that defendant did not permit the couplers to be in a defective condition as alleged in the declaration.
Reversed with finding of fact.
Finding of Fact. We find the issues for the appellant, the Chicago Short Line Railway Company, that it was not guilty of negligence as alleged in the declaration herein of Donald A. Curran, appellee; that it did *163not permit one of the couplers on its car, as therein alleged, to be in a defective condition so that it would not couple automatically without a brakeman going between said car and the one next to it.