BUFFINGTON, Circuit Judge.

In the court below, the United States, by 18 counts, charged the Erie Railroad Company with a violation of Act Cong. March 2, 1893, c. 196, § 1, 27 Stat. 531 (U. S. Comp. St. 1901, p. 3174), which provides:

“It shall be unlawful for any common carrier engaged in interstate commerce by railroad to use on its line any locomotive engine in moving interstate traffic not equipped with a power driving wheel brake and appliances for operating the train-brake system, or to run any train in such traffic after said date that has not a sufficient number of cars in it so equipped with power or train brakes that the engineer on the locomotive drawing such train can control its speed without requiring brakemen to use the common hand brake for that purpose.”

It will be observed that these counts do not charge the railroad with violation of that absolute duty referred to in Delk v. St. Louis, 220 U. S. 586, 31 Sup. Ct. 617, 55 L. Ed. 590, of properly equipping its cars initially with statutory safety devices. What they do charge is that the railroad subsequently unlawfully used such initially properly equippéd cars in that it operated trains in interstate traffic wherein 75 per cent, of the cars had not their air brakes connected with the engine.- The case has been tried twice in the court below, and twice brought to this court for review. It involves the movement of freight cars in the great tidal yard terminal system of the Erie Railroad Company contiguous to New York Harbor. The present writ is an effort on the part of the government to have this court review and, reverse its former decision, or in the event of our adhering to such view, to put the case in such shape that an application to review it may be made to the Supreme Court. On the first trial the lower court declined to admit testimony tending to show that the operations complained of were switching yard movements. It further instructed the jury that the movements complained of violated the statute. On error to the entry of judgment thereon, this court, in an opinion reported at 197 Fed. 287, 116 C. C. A. 649, which opinion we by reference make part hereof, remanded the case “with permission to the court below to grant a new trial, if' so moved by the government.” On the case as then presented in this court two questions arose: First, did the act in question require, in switching operations, the coupling of air brakes ? If not, a second question arose, namely, Were the car operations therein involved switching movements? The first question, which was one of statutory construction, we decided in the negative, saying:

“Giving then to the act the construction that it was not meant to cover bona fide switching operations, we next inquire whether the car movements here in question were really switching operations of the railroad.”

After discussing the proofs we then said:

“To us it seems clear that there was evidence tending to show that-the whole triangle formed by the Bergen, Jersey City, and Weehawken yards constitutes unitedly a single terminal classification yard. While the distance between two of these points is over three miles, it is evident from the narrow strip of land that is left along the river from the Palisades that this wide spread of space is topographically required to permit the complicated car transfer and classification of a great railroad’s terminal traffic. It is in such yards that the great contest against inextricable confusion and freight congestion must be waged.”

*856Indeed, the case had been tried and the evidence adduced on the theory that it turned on the definition of the word “train,” and it was contended that if the cars coupled together in this intersubyard transfer could be called a train, the statute was violated, whether such transfer was really switching or not. Setting aside this contention, and pointing out what was the real question involved, we said:

“It is urged that considerable numbers of cars, initially classified at the three respective points, are moved on the main freight tracks to and through the tunnel, and that they constitute trains, and such movement therefore falls within the wording of the statute. Of course, 35 cars coupled together and drawn by a locomotive make a train, for such connected ears are drawn and follow in the engine’s train; hut this mere word definition does not settle the question before us. It is not a wrangle over mere names, but rather whether the railroad is here doing a bona fide switching work which the law confessedly was not meant to cover. * * * We have here a question of a great terminal located in a most congested freight center. . * * »' Indeed, the testimony of the government’s inspector » * * shows that practically the same terminal situation as that at Jersey City exists at Buffalo and other terminal points. We are pointed to no conditions incident to-this short run which makes the use of the air brakes essential to the safety of the shifting crew. The run is a brief one. There are neither grades to-encounter, stops to be made, nor protracted exposure.”

In the absence of testimony pertinent to the real issue involved, this-court might have also said, as was done by the Circuit Court of' Appeals of the Eighth Circuit in C., B. & Q. R. R. v. United States, 211 Fed. 12:

“This case was tried mainly by the dictionary. We have much reasoning of counsel upon general principles. What we would have preferred would be: An accurate description of the development of the terminal yards at Kansas City; the present structure -of those yards; the methods of handling trains therein; the speed at which transfer trains are moved between the yards; the control over such trains afforded by the coupling up of the air upon a part of the cars only; whether in actual practice, with the air coupled up on 6 to 10 cars, the engineer can control the speed of these transfer trains from the locomotive, ‘without requiring brakemen to use the common hand brakes for that purpose’; what, if any, accidents have resulted from the failure to couple up 75 per cent, of the air; the time that would be consumed in coupling up 75 per cent, of the air on such trains; the number of trains that are moved in the yard; the effect upon the movement of cars in such terminal yards if 75 per cent, of the air had to be coupled up on all these strings of cars. In other words, the evidence should do all that could be done to place the court in the same position as an experienced railroad man in judging of these transportation questions.”

[1] Feeling, however, that the government might have such testimony to offer, we, as seen by our order, and to afford an opportunity to produce the same, gave the government leave, if it saw fit, to move for a new trial. On the second trial, however, the government has given no testimony whatever on the question which we -pointed out as the vital one in the Case, namely, “whether the railroad is here doing a bona fide switching, work.” The only testimony bearing on the foregoing question is that of one witness, an inspector for the Interstate-Commerce Commission as follows:

“Q. Now are you familiar with the local situation of Jersey City? A. Tes,, sir. Q. What would you call the collection of 'tracks at Jersey City on which the cars are shifted? A. Classification tracks for the making up of trains.. *857Q. What would you call the whole group of classification tracts there, limiting your answer to those classification tracts at Jersey City? A. A yard. Q. Is there a similar group of classification tracts at Bergen? A. There are. Q. What, among railroad men, is such a classification group of classification tracts at Bergen called? A. A yard. Q. Are you familiar with a similar group of classification tracts at Weehawten? A. Xes, sir. Q. Will the same answer apply to the designation of the group of classification tracts at Weehawten that is applied to the others? A. It does. Q. In each case it is a yard? A. Xes, sir.”

[2] It scarcely need be said that this testimony, as to the mere existence of three individual yards, threw no light whatever on the real issue whether “the whole triangle formed by the Bergen, Jersey City, and Weehawken yards constitutes unitedly a single terminal classification yard.” Indeed the express limitation of the question asked, “limiting your answer to those classification tracks at Jersey City,” precluded the witness from giving any answer on the real issue. It would therefore seem that when the government omitted to furnish any evidence on that issue, and expressly precluded its only witness from giving any testimony upon it, it cannot now justly complain that such issue was not submitted to the jury; for, as it seems to us, the situation, in view of the trial procedure, narrows itself' to this: If the case had been submitted to the jury and it had found for the government, would the court below have been bound to set the verdict aside ? The recognized test in that regard is that:

‘‘When the evidence given at the trial with all the inferences the jury could justifiably draw from it is insufficient to support a verdict for the plaintiff, so that such a verdict, if returned, must be set aside, the court is not bound to submit the case to the jury, hut may direct a verdict for the defendant.”

In that regard the trial judge, in directing a verdict for the defendant, said:

“It has been held by the courts that these requirements of the safety appliance act are not obligatory in shifting movements in a yard. If they were, the mere movement of a car without the required appliances in a yard would be a violation of the act; and the mere movement of the train, in shifting, without the coupling of the cars, together with the air-brake appliance, so that the required number of cars were so coupled, would be a violation of the act. But, as I have said, the act does not require these appliances in a terminal yard where shifting is done.

“The evidence in this case appears to me to be undisputed that the railroad company receives cars at Jersey City, receives cars at Weehawken, and receives cars at Bergen. The evidence discloses the fact that cars — many of them — are received at Jersey City from lighters, and also at Weehawken, and that those cars must be removed quickly to make room for others that may be coming. There is an abundance of evidence that is uncontradicted, as to the great number of cars that are received and handled at these two points. The evidence is uncontradicted that at Jersey City the defendant railroad company is hemmed in by the Pennsylvania Railroad and by other conditions, so that it has there but a limited number of tracks — 60, I think, 50 or 60 — and the evidence shows that at Weehawken, also, the defendant is hemmed in, able to have but some 80 tracks there. * * * The evidence shows that the movement of these trains extends from what they call the terminals at Jersey City and Weehawken, through the tunnel out to Bergen, where there are something like 115 tracks, where cars can be classified, where they can be inspected and repaired, and where the air hose may be coupled up and the air brakes thoroughly tested, taking all the time that is necessary, before they start for their various destinations. As I say, the facts are not *858disputed. This is one yard, under the evidence in the case, although separated by a mountain and connected by a tunnel, which, by the way, the evidence shows, does not contain the main traveled tracks of the railroad. Being one yard, therefore, under the evidence in the case, it is my duty to instruct you, gentlemen, that the railroad company is not guilty of a violation of the act in its train movements, as charged, and that your verdict must be for the defendant.”

[3] In view of the absence of evidence on the part of the government, and of the undisputed facts proved by the defendant, the question whether these three yards are or are not a switching unity is not debatable. Unquestionably they are. That the cars in question were being subjected to switching classification in successive yards of this system is undeniable. Courts no more than individuals can close their eyes to the fact that they were being moved in switching, not road, service, and that this service was continuously and uninterruptedly a switching service none the less because barriers of nature — the Hudson river and Palisades at front and rear, and the laws of man in a prior use by common carriers, north and south — necessitated part of such switching' being done on one side and part on the other of the tunneled Palisades. Of the physical fact that this was a switching operation there can be no doubt. Of the legal results that flow from that physical fact there is of course a question, but as to the facts themselves there is no question.

Our decision has commended itself to the Circuit Court of Appeals of the Eighth Circuit in C., B. & Q. R. R. Co. v. United States, supra, wherein it was said:

“The identical question which is here presented was before the Circuit Court for the Third Circuit in Erie Railroad Company v. United States, 197 Fed. 287 [116 C. C. A. 649], and, we think, was there properly decided, notwith-' standing its criticism in United States v. Pere Marquette R. R. Co. [211 Fed. 220] in the District Court of the United States for the Western District of Michigan, decided September 5, 1913.”

We have carefully considered the case of United States v. Pere Marquette R. R. Co., referred to, but as the facts, of that case are stated, the legal question involved in our case, namely, whether the act of 1903 covers switching operations, was really not involved in that case. There the Wyoming yard was the switching outpost on the main line of the railroad terminating at Grand Rapids. It was two miles from the city, and while it was within the yard limits of that city, all the classification and switching incident to the distribution of incoming, and the making up of outgoing, trains was done in the Wyoming yard. As stated in the opinion:

“Wyoming yard and freighthouse yard are both within the general yard limits of the city of Grand Rapids, but they are about two miles apart and each has its own system of switching trades. Trains passing from one yard to the other must run for the entire distance upon defendant’s main line, over which its regular passenger and freight trains, as well as switching and transfer trains, are operated. This part, of the main line has some minor grades and curves, and crosses at grade five city streets and a street car line. All of defendant’s freight trains entering Grand Rapids are taken directly to Wyoming yard, and are there broken up and the cars switched and classified; some being put into outgoing trains and forwarded to their destination, while others, containing local merchandise, are switched or transferred to the *859freighthcrase yard, or to some city side track, to be unloaded and to have their cargoes rearranged. All outgoing freight trains are made up at and started from Wyoming yard.”

It is therefore clear that the entire switching of incoming cars intended for other lines took place in the Wyoming yard, and such cars never went to the city freightyard. As to the incoming local freight, it also was classified and switched in the Wyoming yard and from there sent in over the main road to the city freightyard for unloading. The Wyoming yard was in fact the sole terminal switching yard. There and there alone classifications and interchange were made. Under these terminal conditions it could no more be contended that the subsequent transfer to the city freightyard of these already 'switched trains for two miles, over the main passenger and freight line of cross streets at grade and intersecting trolley tracks, was a switching operation than would be the case if the Wyoming yard were 22 instead of 2 miles distant from the city freightyard. In other words, the switching was initially and finally finished at the Wyoming yard and the switching there, so far as the make-up of trains was concerned, was in no way affected by, dependent upon, or correlated to anything thereafter or theretofore done in the city freightyard. The track between Wyoming yard and the city freightyard was not a nexus that served to correlate two dependent yards, but was a hiatus or break between two independent yards. To us it is clear that the Wyoming terminal is wholly different from the Brie, because in the Erie we have three switching yards interdependent of each other, each supplementing and each necessary to complete the partial switching done in the other, and all forming, a unitary combination switching system topographically indispensable to the handling of incoming and outgoing trains.

[4, 5] It is, however, contended that our former decision overlooked and disregarded the effect of the act of 1903 (32 Stat. 943), and that it is in conflict with the decisions of' the Supreme Court referred to below. It will be observed, however, that federal air-brake legislation had in view two distinct objects, namely: First, to compel railroads to have their cars equipped with such brakes, inter alia, under all circumstances ; and, second, to compel the use of air-brake equipped cars under certain circumstances. These duties of air-brake equipment and air-brake use are separate and distinct. As noted at the outset of this opinion, it is this duty to equip that is referred to in Delk v. St. Louis, 220 U. S. 586, 31 Sup. Ct. 620, 55 L. Ed. 590 as “an absolute duty to provide and keep proper couplers at all times and under all circumstances.” The duty to equip was the subject-matter involved in St. Louis v. Taylor, 210 U. S. 282, 28 Sup. Ct. 616, 52 L. Ed. 1061; in the reaffirming case of Chicago v. United States, 220 U. S. 559, 31 Sup. Ct. 612, 55 L. Ed. 582; in Southern Ry. Co. v. United States, 222 U. S. 24, 32 Sup. Ct. 2, 56 L. Ed. 72; and in Wabash v. United States, 168 Fed. 1, 93 C. C. A. 393. It will be noted, also, that the act of 1893, as construed by the Supreme Court in Johnson v. Southern Pacific Co., 196 U. S. 21, 25 Sup. Ct. 158, 49 L. Ed. 363, was neither changed nor enlarged by Act March 2, 1903, c. 976, 32 Stat. 943 (U. S. Comp. St. Supp. 1911, p. 1314). That act was pass.ed to meet the decision in Southern Ry. Co. v. Johnson, 117 Fed. 462, 54 C. C. A. *860508, but needlessly so, for, as said by the Supreme Court in reversing the latter case:

“The latter act is affirmative, * * * and, in effect, only construed and applied tlie former act.”

It would seem, therefore, that none of these cases involved the question here involved, namely, the compulsory use of air-brake equipment in switching operations. On that question, which is one of statute construction, we hold the act does ,not compel the air coupling of cars in switching movements. We so hold, amongst others, for these reasons : First, because had Congress meant to compel air-coupled switching, it would have said so; second, by providing automatic coupling Congress-had already provided as far as it could against the avoidable dangers incident to switching; third, if the law includes switching, it covers all switching without limitation or exception, an impracticable and impossible thing; fourth, if the act covered switching, and Congress meant to except any switching therefrom, it neither did nor by language made it possible to now decide what switching was excepted. Indeed, a careful study of this act shows the use in the statute of' terms- and words which in common use are applied to road, as contrasted with switching, operations. The act deals first with the locomotive alone as distinguished from the train. It makes it unlawful for the railroad “to-use on its line” — and line, main line, is a word which, in the common speech of railroad work, distinguishes the line of the road from switches and terminal yards. But the act proceeds, “to use on its line any locomotive engine in moving interstate traffic.” Surely the words,, “in moving interstate traffic,” in connection with the use of a locomotive on its line, is aptly applied to draft of' trains in their transit between states. But the act proceeds, the locomotive “on its line” which is “moving interstate traffic” must be equipped with “appliances for operating the train-brake system.” Surely these words, “operating the train-brake system,” mean that the system, the train-brake system, operated by the locomotive “on its line” and in “moving interstate traffic,” refers to a running, father than a switching, movement. And the further words of the statute, which make it unlawful for the road “to run any such train-in such traffic * * * that has not a sufficient number of cars in it so equipped with power or train brakes that the engineer, on the locomotive drawing such train can control its speed without requiring brakemen to use the common hand brake for thqt purpose,” are words that aptly describe train movement. The operation of1 “the locomotive drawing such train” is in marked contrast with the push and pull of a switching engine, and “control its speed” refers to a train that is speeding, for the appliances must be such that “the engineer on the locomotive drawing such train can control its speed without requiring brakemen to use the common hand brake for that purpose.” All these terms and words-of railroad parlance are applicable to line travel and fitly descriptive thereof. In railroading, “line” is contracted with “switch,” “yard,” and “terminal”; main line, branch line, with switches. A “locomotive drawing such 'train” is in contrast with the push and pull of a yard switching engine. And not only is this the common and practical meaning of the' *861words, but when the practical operative consequence of giving these words the meaning here contended for are considered, the unwisdom of such construction is apparent. As well said by Judge Amidon in the Eighth Circuit case quoted above:

“The attempt is made to deduce the decision of the case from the definition of the word ‘train’ by a process of abstract reasoning. One fundamental trouble with such reasoning is that it proves too much. The word ‘train’ of course covers any string of cars hauled by an engine. But if the statute is to be applied to all trains falling within this definition, then it would cover all movements of cars by means of a locomotive in switching operations, and it would mate no difference whether that movement was on a main track or a siding. Such a result reduces the reasoning to an absurdity, because its application to railroads would operate as an embargo upon commerce.”

That it would be an unworkable embargo on terminal operations is a conclusion justified by the proofs in this record. These proofs show that the air coupling 6i trains, such as covered by the present indictments, to make the switch movement through the tunnel and between the Bergen, Jersey City, and Weehawken subyard's, would require a delay of 1 hour and 25 minutes in the movement of each train. In this connection we pass by as untenable the contention of1 the government that part of the delay might be avoided by dispensing with testing; that the government officers would be satisfied with coupling. We cannot accede to this, and we think on mature reflection it would not be pressed. It suffices to say that if Congress meant that cars were to be air coupled while being switched, it did not intend to relieve the railroad from a thorough inspection of such couplings before they were used. It would not only be negligence for the company to fail to so inspect, but the manifest injustice of leading men to place reliance on the efficiency of untested appliances borders so near on positive and misleading wrong that we dismiss such contention without further discussion. We hold that air coupling in switching was not included in the statute. If we are wrong in that, if air-coupled switching is the statutory duty, then there must go with it such preliminary tests of such appliance^ as will insure their fitness to fulfill the statutory purposes in view, namely, that the engineer on the locomotive drawing such train be able to control its speed without requiring brakemen to use the common hand brake for that purpose. To forego inspection and the time needed therefor is not to be countenanced. The mere fact that the cars in these terminals come from other systems, whose standards of maintenance differ, would of itself require a more careful and thorough test and inspection before the cars were subjected to the exactions of an air-coupled service.

As to the remaining eight counts, we need only say, as we did in our former opinion, that as therein indicated our disposition of the main case disposes of these matters.

Adhering, therefore, to our former decision, we affirm the judgment below. ,