Opinion of the Court by

Judge Clay

Affirming.

Enos Caudill, an employee of the Louisville & Nashville Railroad Company, brought this action under the Federal Employers’ Liability Act (45 TJSCA secs. 51-59) to recover damages for personal injuries. At the conclusion of Caudill’s evidence the trial court directed a verdict in favor of the company, and he appeals.

The evidence briefly stated is as follows: The railroad “company was engaged in picking up loaded cars at the different mines on its Rockhouse branch in Letcher county, and taking them to Hazard, where they were assembled for further shipment. The crew consisted of the engineer, the conductor, Caudill, the head brakeman, and the swing brakeman, Elkins, who carried the switch list and gave directions as to what cars were to be taken from the various mines. On completing their work on the Eden mine the swung brakeman directed Caudill to go down and head in on the load track at Marion. The track at that point was on an ascending grade, and to prevent the cars from moving it was necessary for the miners to set, and they did set, the hand brakes on each ear as it loaded and dropped down on the track. When the swung brakeman gave the order to Caudill he stated that after cutting the switch he would start in the middle of the cut of ears to be picked up, and would couple up the air hose and release the hand brakes on the cars. It was Caudill’s duty to couple the air and release the brakes on the remainder of the cars. He began the performance of this duty by first coupling the air hose and turning the angle cock, so that the air brakes on the engine might be applied. He then started toward the rear of the train *688 and released the hand brakes and coupled the air hose on some five or six cars. Before he completed this work the engineer in response to a signal from one of the men started the ca'rs moving. Caudill and the conductor immediately discovered that the hand brakes on one of the cars which Caudill had not reached had not been released, and gave the engineer the signal to stop. When the train stopped Caudill climbed the ladder at the end of the car next below the one on which the brakes were holding, and placed his left foot on the end sill of that car and his right foot on the couplers. In order for him to release the brakes it was necessary for him to use a pick handle or brake club that had been issued by the company. Holding to the lower car with one hand he took the brake club and reaching across the opening between the cars prized the latch loose. This released the brakes and the car with the slack out started to run in. When he stepped back his left foot slipped off and went down between the end sill and the drawhead. His foot was mashed and had to be amputated near the ankle. It was customary in a case of that kind for the engineer to apply the air when he stopped the train. Had he applied it on the occasion in question there would* have been no movement of the car and his foot would not have been injured. He did not couple the air hose between the two cars. He did not look to see if it was coupled, nor could he say that it was coupled. He had an idea that it was, as the brakeman is supposed to get the air or see that it was all back there on the train before he moved it out.

The only negligence relied on was the failure of the engineer to apply the air. Whether, in view of the fact that there was no evidence tending to show that the engineer knew, or by the exercise of ordinary care could have known, that appellant was between the cars, such failure, if it had been shown that it caused or contributed to appellant’s injury, would have constituted negligence authorizing a recovery, we need not determine. The only evidence that the engineer did not apply the air is the assumption that, if he had done so, the cars would have remained set. It is at once apparent that this inference is far-fetched and unreasonable in the light of the facts. The application of the air would have had no effect whatever unless the air hose was coupled at the place of accident. Appellant admits that he did not couple the air, *689 or see that it was coupled, and was unable to say that the air was coupled. He merely inferred that it was because the brakeman was supposed to see that it was done before the train was moved out. There is no basis whatever for this inference. The brakeman who released the brakes also coupled the air. Therefore, even if it had been shown that the car on which the brake had not been released was in that portion of the train that the swing brakeman was to look after, the fact that the brake was unreleased was a circumstance tending to show that he had not been there, and that the air had not been coupled. But the evidence goes further. Appellant admits that the train was moved before he reached the car in question, and that the car was one that he was to look after. In the circumstances he had no right to assume that the duty of coupling the air had been performed by some one else, and his mere “idea” that it was is not sufficient to show that such was the case at the time of his injury. In view of this showing it cannot be said that there was any evidence tending to show either that the engineer failed to apply the air, or that such failure, if any, caused or contributed in any way to appellant’s injury. The case is simply one where liability is predicated on a strained inference not based on an established fact, but based entirely on a prior inference unsupported by any fact, and therefore equally far-fetched—a situation not sufficient to carry the case to the jury. Sutton’s Adm’r v. L. & N. R. Co., 168 Ky. 81, 181 S. W. 938; National Surety Co. v. Redmon, 173 Ky. 294, 190 S. W. 1081. It follows that the motion for a peremptory instruction was properly sustained.

Judgment affirmed.