[853] BARRETT, C.

Ruth Crabtree was awarded a judgment against the trustees of the St. Louis-San Francisco Railway Company for the death of her husband, Russell Crabtree, and the definitive question is whether the facts proved constitute a violation of Section 2 of the Safety Appliance Act. (45 U.S.C.A., Sec. 2.)

Crabtree was a brakeman on a local freight train between Fort Scott, Kansas and Kansas City, Missouri. On his thirteenth trip he was killed during a switching movement at Lenexa, Kansas. The substantial facts relied upon and which the jury could and did reasonably find are as follows: After several switching movements the engine was standing on a switch track about 400 feet north of a crossover switch. At that time there were three cars attached to the front of the engine and one to the tender. The purpose of the next movement, which was to be a "flying" or "drop" switch, was to place the three cars on the track west of the crossover track. The conductor, who was in charge of the movement and the train, was stationed at the crossover switch on the east side of the northbound track — the track the engine and cars were on. Dellinger, the fireman, who was acting as engineer, was at the throttle on the east side of the engine. While the conductor, Edwards, was "lining" the switch Crabtree walked from the north end of the three cars and stopped about even with the front end of the engine and on the east side of the cars. With these three members of the crew in their respective positions, which they would "assume in making the cut," the conductor simultaneously signalled the engineer and Crabtree to make a "drop" or "flying" switch of the three cars. With the signal, Crabtree stepped upon the pilot sill or step on the engine, on the east side, and after the engine and cars had travelled about one-half the distance to the crossover switch the conductor gave an "easy" or "slack" signal and the engineer applied the brakes and "slowed up" so as to create "slack" and permit Crabtree to "pull the pin," thus uncoupling the three cars from the engine. After the engine had been "slowed up" the engineer looked back at Crabtree for a "go ahead" signal and could then see enough of him that he could see his hand and knew that he was standing on the pilot step. He got the signal from Crabtree, released the brakes on the engine, opened the throttle and looked back in time to see Crabtree rolling out from under the second car from the engine.

There was an automatic coupler on the front of the engine and an uncoupling lever ("pin lifter") on both the right and left sides (a double lever) of the engine. Of the three cars to be switched the one next to the engine was a coal car with automatic couplers on each end but the uncoupling levers were single and diagonally placed, consequently there was one uncoupling lever on the "left side of end of car" and none on the right side. It is fair to say that the only inference to be drawn from all the evidence is that the couplers and the uncoupling levers were such devices as are permitted by the rules of the Interstate Commerce Commission and that they were free from mechanical defects during the switching movement in which Crabtree was killed.

Crabtree's duty in connection with the "flying" switch was "to pull the pin" so as to effect an uncoupling or disconnection of the engine and the coal car. The pin can only be lifted during the moment the "slack" appears after the engineer acts in response to the "easy" or "slack" signal. Since the cars are to be dropped and the switching movement is to be executed by the joint and cooperative efforts of the conductor, the engineer and the brakeman, the only reasonable inference of fact to be drawn is that it is most practicable and desirable that these three members of the crew be on the same side of the engine and cars in order that all of them may receive and know of the signals given and act accordingly. In this instance they were all properly in such positions (on the "working side of the [854] movement") that they could see and directly signal the engineer or follow the instructions of the conductor or the signals of Crabtree.

With cars and an engine coupled as these were the jury found that the safest, proper, customary place for the switchman to stand and perform his duty of lifting the pin during a "flying" switch was on the pilot sill or step on the front of the engine and that he could not stand on that step or elsewhere on the east side of the engine and cars and lift the pin without and except that his body be between the engine and car to be disconnected.

The plaintiff, Mrs. Crabtree, does not contend that the railroad was negligent in any manner with respect to her husband's death. Illinois State Trust Co. v. Missouri Pac. R. Co., 319 Mo. 608, 5 S.W.2d 368; Harlan v. Wabash Ry. Co., 335 Mo. 414, 73 S.W.2d 749. Neither does she rely upon a mechanical defect in the uncoupling devices as a violation of the Safety Appliance Act. Chicago, St. P., M. O. Ry. Co. v. Muldowney, 130 F.2d 971; Truesdale v. Wheelock, 335 Mo. 924, 74 S.W.2d 585. Nor does she claim that the uncoupling devices had failed to function or operate, for any cause, and that her husband was between the engine and coal car for that reason. Peters v. Wabash Ry. Co., 328 Mo. 924, 42 S.W.2d 588; Alcorn v. Missouri Pac. R. Co., 333 Mo. 828, 63 S.W.2d 55; Johnson v. Southern Pac. Co., 196 U.S. 1, 25 S.Ct. 158, 49 L.Ed. 363. Nor does she base her cause of action upon the railroad's failure to equip its cars with appliances not complying with the standards fixed by the rules and orders of the Interstate Commerce Commission. Atchison, T. S.F. Ry. Co. v. Scarlett, 300 U.S. 471, 57 S.Ct. 541, 81 L.Ed. 748.

Her position is that the movement to be executed, the "flying" switch of the three cars (which involved as a component part of such a movement an uncoupling of the cars from the pilot step of the engine), made it necessary for her husband to go and be between the end of the car and the engine and that such facts and conduct constitute a violation of the absolute standards fixed by the act itself. "It shall be unlawful for any common carrier engaged in interstate commerce by railroad to haul or permit to be hauled or used on its line any car used in moving interstate traffic not equipped with couplers coupling automatically by impact, and which can be uncoupled without the necessity of men going between the ends of the cars."

The railroad's contention is that these facts and circumstances do not establish that it violated the Safety Appliance Act but on the contrary affirmatively demonstrate that all the appliances met the requirements of the act and the specifications of the Interstate Commerce Commission and, therefore, it had fully discharged its obligation to the plaintiff's husband.

Admittedly, there were no defects in the appliances and the couplers, as well as the uncoupling devices, were standard equipment and yet the railroad violated its statutory duty to the plaintiff's husband. The evil sought to be eliminated by the legislation and the hazard to employees which the law was designed to obviate was "the necessity of men going between the cars" and especially for the purpose of coupling or uncoupling them. House Report No. 1678, 1st Session, 52nd Congress; Senate Report No. 1049, 1st Session, 52nd Congress; Vol. 24, Pt. 2 Congressional Record, 1893, (52nd Congress, 2nd Session) pp. 1246-1252; 1273-1288. In this respect the language of the act is perfectly clear and unambiguous. Paraphrasing, it says that is unlawful for a railroad to use "any car" which does not couple automatically on impact "and which can be uncoupled without the necessity of men going between the ends of the cars." The railroad's duty with respect to couplers and uncoupling devices is absolute "at all times and under all circumstances" to provide devices which may be uncoupled without the necessity of men going between the cars for that purpose. Delk v. St. Louis-San Francisco R. Co., 220 U.S. 580, 31 S.Ct. 617, 55 L.Ed. 590. In the Delk case there was a defective coupler and it may be that as to some appliances there must be a mechanical defect before there is a violation of the act proximately causing the injury complained of or the object causing the injury must be one of the things specifically designated by the statute (Compare Riley v. Wabash Ry. Co., 328 Mo. 910, 44 S.W.2d 136; Zachritz v. St. Louis-San Francisco Ry. Co., 336 Mo. 801, 81 S.W.2d 608; Truesdale v. Wheelock, supra; Fryer v. St. Louis-San Francisco Ry. Co., 333 Mo. 740, 63 S.W.2d 47 and Lilly v. Grand Trunk Western R.R. Co., 317 U.S. 481, 63 S.Ct. 347, 87 L.Ed. 323), but, under this section of the act. "If the railroad does, in point of fact, use cars which do not comply with the standard, it [855] violates the plain prohibitions of the law, and there arises from that violation the liability to make compensation to one who is injured by it." Chicago, B. Q. Ry. Co. v. United States, 220 U.S. 559, 575, 31 S.Ct. 612, 615, 55 L.Ed. 582, 588. It is not broken or mechanically defective devices which the law prohibits, neither does the law compel the use of any particular type of device. "The test of compliance with section two" is an uncoupling device which may be used "without the necessity of men going between the ends of the cars." Johnson v. Southern Pac. Co., 196 U.S. 1, 18, 25 S.Ct. 158, 162, 49 L.Ed. 363, 370. "The test of compliance is the operating efficiency of the couplers with which the car is equipped." Chicago, St. P., M. O. Ry. Co. v. Muldowney, 130 F.2d 971, 975. "The broad aim of the second section is to eliminate the risk of injury by an absolute prohibition of the use of any car unequipped with such a coupling device as that cars may be coupled and uncoupled `without the necessity of men going between the ends of the cars.'" United States v. Philadelphia R. Ry. Co., 223 F. 215, 216.

In this case, usually and ordinarily, the coal car and the engine could have been uncoupled with the appliances present and without the necessity of the brakeman being between the car and the engine. But, as the jury found, during a "flying" switch it was the function and duty of the brakeman to uncouple the cars and he could not reasonably do so without the necessity of his being between the car and the engine, with the appliances at hand, and that plainly is a violation of the statute. Christy v. Wabash Ry. Co., 195 Mo. App. 232, 191 S.W. 241. (Writ of error dismissed for lack of jurisdiction 246 U.S. 656, 38 S.Ct. 424, 62 L.Ed. 924.) The movement to be executed, of necessity, placed him in that hazardous position condemned by the statute. Consequently, no imperfection in the appliances was needed to create the hazard imposing liability (Peters v. Wabash, supra; Truesdale v. Wheelock, supra); the operation itself presupposes and implies a violation of the statute. As Judge Ellison said in the Christy case: ". . . it was shown by the evidence that defendant's engine and the car being switched were properly provided with automatic couplers which would couple by impact and which could be uncoupled without going between the cars. But as we have said, they could not be utilized without going between the cars. There was abundant evidence to show that a `flying switch' made by the employee standing on the footboard in front of the moving engine could not be accomplished without standing between the engine and car, or without hanging to the end of the car by placing one foot in a stirrup and reaching some part of the body around between the two. We think in these circumstances a case was made for plaintiff. For notwithstanding an interstate carrier complies with the Safety Appliance Act, yet if it operates the cars so that the appliances cannot be used without doing the thing the Act seeks to avoid, i.e. going between the cars, it violates the statute as fully as if it had failed to install the appliances." In other words, is compliance with the statute and liability to be evaded merely by furnishing the proper appliances and then not using them?

It is not the device which is condemned — one which does not operate is unlawful, not because the device is defective, but because such a device necessarily places an employee in that dangerous position condemned by the statute, between the cars. For example, in Hohenleitner v. Southern Pac. Co., 177 F. 796, there was nothing wrong with the coupling or uncoupling devices but when the cars were on a track with a twenty degree curve they would not operate without the necessity of men going between the cars and that was held to be a violation of the act. In United States v. Illinois Cent. R. Co., 177 F. 801, the couplers met all the standards but the cars were so loaded with lumber that the uncoupling devices could not be utilized and that was held to be a violation of the act. There is no difference in principle in these cases and this one. Here the railroad operated its train in disregard of its appliances and executed a "flying" switch which of necessity placed its employee between the engine and the cars for the purpose of uncoupling them whether he could and did use the appropriately provided devices or not and so violated the express provisions of the act. Compare to the contrary Devine v. Calumet River R.R. Co., 259 Ill. 449, 102 N.E. 803, though the case is otherwise distinguishable because a defective engine was, in fact, the proximate cause of the fatality.

In this connection the railroad contends that the violation of the Safety Appliance Act was not the proximate cause [856] of Crabtree's fall and death because the evidence affirmatively disclosed that the uncoupling had been effected and, therefore, he did not fall from the pilot step, between the engine and car, while in the act of or by reason of his effecting an uncoupling. Peters v. Wabash Ry. Co., supra; Illinois State Trust Co. v. Missouri Pac. R. Co., supra. The evidential basis of its argument is this: The conductor gave the "slack" signal, Crabtree "pulled the pin," and the cars were then uncoupled from the engine. Crabtree gave a "go ahead" signal, indicating that the uncoupling had been effected. He was then standing on the pilot sill. The engineer acted in response to the signal, looked back, and Crabtree was then rolling out from under the second car. From these facts it is apparent that Crabtree did not fall while and during the instant the pin was in his hand nor even during any of the time he was manually and immediately engaged in lifting the pin from the couplers and so, in that sense, he was not killed while engaged in an uncoupling operation. But, as a "flying" switch includes and assumes an uncoupling, so an uncoupling under such circumstances presupposes the man is to prepare for the operation by getting himself into the proper position between the cars, lift the pin, signal that he has done so and then he is entitled to a reasonable length of time in which to return to a position of safety and it cannot be said that he had remained there such a period of time thereafter that the uncoupling operation was not the proximate cause of his fall. Here, as when a defective coupler necessitates an employee's going between the cars, the brakeman is there between the cars in response to the movement to be executed and for the purpose of uncoupling them. Since that is the fact, "It (being between the cars) did not cease to operate as the proximate cause of his being between the cars until he got out or had a reasonable time to get out, and would be the proximate cause of any injuries he received during such time. Therefore, if there is substantial evidence in this case that he did not get out, but was injured on his way out, then there was a case for the jury." Truesdale v. Wheelock, supra. The same argument was made in the Christy case, that is, that the fall must have occurred as the deceased was actually engaged in pulling the pin or exerting himself in an effort to effect the uncoupling and not before. The court pointed out that Christy had to go in between the cars to uncouple them and that he was killed because he went in and said: "It certainly can make no difference whether he was killed an instant before he begun the act of uncoupling." And here, that he was killed an instant after the uncoupling had been effected should not and does not make any difference; the proximate cause of his death was the necessity of his having to be between the cars. Chicago, M., St. P. P.R.R. Co. v. Linehan, 66 F.2d 373.

It is urged that instruction one hypothesizing the plaintiff's theory of liability is erroneous as applied to the facts and as a matter of law as well. The objections are that it does not require a finding that the railroad violated the provisions of the act nor that there were any defects in the equipment or that it failed to operate. It is said that the instruction ignores the defendant's contention that the uncoupling could have been effected from a position on the cars other than the pilot step; that it permits the jury to speculate and guess that Crabtree fell from the step while in the process of effecting the uncoupling and permits the judgment of the jury to be substituted for that of the Interstate Commerce Commission.

What we have previously said with reference to the railroad's liability answers most of these objections. Liability in this instance, as we have indicated, does not suppose or require a defect or a failure of the appliances to properly operate in a mechanical sense — the violation of the law was an operation or movement in disregard of the devices, one which could not be practicably executed or effected except by violating the admonition of the statute. Also, we pointed out that from the evidence most favorable to the plaintiff the uncoupling could not have been effected in such a movement except that the brakeman, whatever position he might take, be between the end of the car and the engine. So it was with reference to the exact moment he fell — it need not have been at the instant he lifted the pin and the cars separated. If he was between the cars for the purpose of uncoupling them and as a result of that fact or because of it he was caused to fall, during, immediately before or so shortly after uncoupling them but during a period of time necessarily and reasonably connected with the uncoupling movement the act was violated and liability would follow. Truesdale v. Wheelock, supra.

[857] In the phraseology of the statute the instruction states that it was the duty of the railroad to equip and use cars which could be uncoupled without the necessity of men going between them. It then said that if "under all the facts and circumstances" such as the lay of the tracks, the necessity of signals being communicated between certain members of the crew and the location of the appliances "it was necessary in the execution of a drop or `flying' switch . . ., contemplated or being executed," that Crabtree should take a position on the step between the engine and car for the purpose of uncoupling them and "while in the performance of said uncoupling, or immediately thereafter as a consequence thereof" he fell and that the necessity of his going between the cars was the direct and proximate cause of his death the plaintiff was entitled to recover. Obviously, if the plaintiff's theory of liability is sound, the instruction is likewise sound. The jury was not to find what the applicable legal principles were — the court told the jury what the railroad's duty was, in the language of the statute, and they were to find the facts and, of course, apply them, and if the facts were as the jury found the law was violated and liability followed as a matter of course. The defendant's given instruction D specifically required the jury to find that the cars could not be uncoupled without the necessity of Crabtree going between the engine and cars. Its instruction F specifically required the jury to find that Crabtree fell from the pilot step "while in the act of effecting an uncoupling" and the court properly added "or as a consequence thereof" before the plaintiff was entitled to recover. And its instruction N specifically told the jury the plaintiff could not recover if they found that Crabtree "could have executed the uncoupling . . . from a position upon either side of the coal car and without the necessity of getting upon the pilot step of the engine and going between the engine and coal car." Aside from these facts the instruction correctly and properly follows the suggestions and instructions in Terminal R. Ass'n. of St. Louis v. Kimbrel, 105 F.2d 262 and Alcorn v. Missouri Pac. R. Co., supra.

The only possible available witnesses possessing personal knowledge of the occurrence resulting in Crabtree's death were the members of the train crew and especially the conductor and acting engineer who were engaged with Crabtree in executing the "flying" switch and they were called as witnesses by the plaintiff. About eleven months prior to the trial of the case the plaintiff took the conductor's deposition but upon the trial she called the conductor, Edwards, as a witness. During his examination in chief he was reminded of the fact that he had given the deposition and then asked whether it was necessary for himself and Crabtree to be visible to one another, that is on the same side of the train, for the purpose of executing the maneuver of the "flying" switch and he answered: "No, sir," it was not. The plaintiff then proceeded to call his attention to the deposition, which he had signed only the day before he was testifying and that procedure was objected to as violating the rule against cross-examining one's own witness. Plaintiff's counsel then stated that he wished to refresh his memory and the court permitted counsel to continue. The witness then admitted that he had answered "Yes" in his deposition but explained his previous contradictory answer by saying that what he meant was that it was only necessary that he and Crabtree have a previous, common understanding as to the movement to be made and after that "he can have a big red barn between me and the signal giving," that it was not necessary for them to be able to see one another and the signals given. Subsequently, the witness was asked whether he gave any signals other than the "slack" signal after the movement began and he said he did not. Counsel then referred to the deposition again and after some discussion counsel said: "I am going to show that he testified differently and I am surprised at his statement today." He then read from the deposition that the witness had given a further signal for the engine to pull away more rapidly so as to give the witness more time in which to continue with his part of the switching.

On cross-examination of this witness the railroad proved that Crabtree had a choice of four positions from which to "pull the pin," three of them different from the one he assumed and two of them from the west side of the cars. The conductor then said that if Crabtree had been on the west side of the cars he, the witness, would have crossed the track from the switch stand and signalled that they were to proceed with the movement and he would need no further signals. (Subsequently, the engineer, Reppert, testifying for the defendant, explained that, customarily, [858] when the switchman was on the west side of the cars or opposite the side the signalling conductor, the fireman, here Reppert who would act as fireman, received the signals and relayed them to the engineer.) Plaintiff's counsel then asked the witness whether he recalled that the morning after the accident he had met Emmett Crabtree, Russell's brother, and Mr. Atkins, the father-in-law, in Fort Scott and he did. Counsel then said: "I will ask you if you said to them on that occasion that Mr. Reppert, the engineer, (then acting as fireman) was on the ground at the time of this last movement?" It was objected that the plaintiff could not, in this manner, impeach her own witness. The court overruled the objection and the witness said: "I never made such a statement."

Subsequently, the plaintiff proved by the two witnesses, Atkins and Emmett Crabtree, that on the day following the accident Edwards told them that Reppert was on the ground at the time of Russell Crabtree's death.

The railroad does not complain of the plaintiff's cross-examination of witness Edwards with reference to the contradictions in his evidence and his deposition, nor of her refreshing his memory from the deposition; its only complaint is of her impeaching his testimony with reference to the conversation in Fort Scott. The particular problem is more readily understood and appears in its proper aspect, however, in view of the substance of all that occurred.

The railroad's position is that whether Reppert was on the ground or on the engine at the time of Crabtree's fall and death was a collateral issue and, therefore, it was prejudicial error for the plaintiff to impeach her own witness as to whether he had previously said the engineer was on the ground or on the engine. Its argument is that the plaintiff was not surprised or entrapped by Edwards' testimony in that regard and, therefore, she was concluded by his denial of having made the statement.

In the first place, whether the engineer was on the ground or in the cab of the engine was not a "collateral" issue by any conceivable definition of the term and as applied to the witness Edwards and his testimony. 3 Wigmore, Evidence, Secs. 1003, 1020-1023. As we have pointed out, the railroad claimed and Edwards testified that Crabtree could have performed his task of pulling the pin from the west side of the cars. In short, the railroad's defense was that the movement to be executed did not contemplate the necessity of the brakeman's going between the cars to uncouple them and that was the gist of Edwards' testimony. One of the instructions specifically compelled a verdict for the defendant if the jury found that Crabtree could have executed the uncoupling "from a position upon either side of the coal car and without the necessity of getting upon the pilot step of the engine and going between the engine and coal car." When it was made to appear that it was necessary for the brakeman to receive and give signals (and that was an important factual issue) and it developed that had Crabtree been on the west side of the cars the signals would have been relayed to the engineer through the fireman, the whereabouts of the fireman during the movement became vitally important. Even if the jury had thought Crabtree could have uncoupled the cars from the west side and without the necessity of being between the cars and the engine, yet if they had also thought and found, as they undoubtedly did, that it was necessary for Crabtree to receive and give signals they could not have found for the railroad if the fireman had been on the ground and, therefore, in no position to relay the necessary signals to the engineer. What the witness had previously said about the engineer's whereabouts during the operation, standing alone, may have been "collateral" in the sense that merely what the witness had previously said about it was a subsidiary or secondary fact and question, but when it is understood that the presence of the engineer at one place would destroy a defense, while his presence at another place might defeat liability, the matter cannot be said to be "collateral" and the only problem is whether the subject was handled with propriety.

We have always followed the view that a party may not call a witness and then impeach his testimony, either by cross-examining him or by proving previous contradictory statements, merely because his evidence proved to be unfavorable or unsatisfactory. Gallagher v. S.Z. Schutte Lumber Co. (Mo. App.), 273 S.W. 213. Although the tendency is to relax the rule, if not to abolish it entirely, and leave the whole matter to the discretion of the trial judge, as we now do when a witness proves to be hostile and adverse. Burnam v. Chicago, G.W.R. Co., 340 Mo. 25, 100 S.W.2d 858. See the American Law Institute's Model Code of Evidence, Rules 105 and 303. It is probable that the true [859] rationale of the rule, as it is generally applied, is that "the party (calling the witness) ought not to have the means to coerce his witness" into testifying as he desires because he has the power to injure him as a witness if he testifies unfavorably or to make him a good witness in the event his evidence does not prove to be disappointing. 3 Wigmore, Evidence, Sec. 899. In short, the true purpose of the rule is to protect the witness against coercion and abuse. However, we have always advanced as the basis of the rule the traditional reason that "a party . . . vouches for his credibility and should not be allowed to impeach him." Detjen v. Moerschel Brewing Co., 157 Mo. App. 614, 617, 128 S.W. 696. Woelfle v. Connecticut Mut. Life Ins. Co., 234 Mo. App. 135, 147-148, 112 S.W.2d 865, 872, summarizes our existing applicable rules: ". . . a party calling a witness will not be allowed to discredit or impeach him by showing that he has theretofore made other statements contradictory of his testimony on the stand, unless it is first made to appear that either the witness himself, or else the adverse party, has by some statement or artifice misled or entrapped the party into calling the witness, so that the adverse party has thereby gained an advantage in the case which he would not have enjoyed if he himself had called the witness. (Citing cases.) In other words, to warrant the impeachment of one's own witness, there must always exist the element of actual, and not mere feigned, surprise at the testimony which the witness gives, and even then only in the event that the testimony is of such an affirmative character as to be favorable to the adverse party, and therefore prejudicial to the party who has been misled or entrapped into calling the witness."

Under this rule one may not impeach the witness he has called when he well knows in advance that the witness' evidence is going to be unfavorable and adverse, as, for example, when the witness' deposition has been previously taken and his evidence was then adverse to the party calling him. In that event there is no surprise to the party calling him and no unexpected benefit to the adverse party. And, of course, in that event the party is not misled or entrapped by the witness. Woelfle v. Connecticut Mut. Life Ins. Co., supra; In re Largue's Estate, 198 Mo. App. 261, 200 S.W. 83; Glenn v. Metropolitan St. Ry. Co., 167 Mo. App. 109, 150 S.W. 1092. When, however, a party is unexpectedly surprised by unfavorable, adverse evidence given by his own witness he may interrogate such witness as to previous inconsistent statements made by him and introduce testimony of other witnesses showing the previous inconsistent and contradictory statements. 6 Jones, Commentaries on Evidence, Secs. 2430-2431; 3 Wigmore, Evidence, Secs. 904-905 with the Missouri cases collected on page 410; Detjen v. Moerschel Brewing Co., supra.

In the instant case Edwards' deposition was taken about eleven months prior to the trial, though he did not sign it until the day before the trial began. In giving the deposition Edwards testified that it was necessary that he and Crabtree be visible to one another in carrying out their respective duties, thus implying that they should both be on the east side of the cars. When his attention was called to the fact he explained away the force of his previous testimony. Upon the trial of the case he repudiated his previous testimony that he had given a signal other than the "slack" signal. At this point in the case plaintiff's counsel told the court that he was surprised at the witness' change in his testimony and the court's ruling indicates that the court thought the plaintiff was surprised, if not entrapped, by the witness. While the plaintiff and her counsel could not reasonably expect the conductor to have been a partisan witness in their behalf, yet they could reasonably suppose that he would stand by his deposition and testify, at least as to material matters, on the trial as he had in his deposition. The witness not only changed his testimony as to certain vital matters but turned out to be a most willing and partisan witness for his employer, so much so that it was obvious that he was an adverse and hostile witness as far as the plaintiff was concerned.

As we have indicated, the railroad does not complain of these matters but complains only of the plaintiff's impeaching the witness' testimony as to where the engineer was at the time of the fatality. It does not appear whether he had been asked about this matter in his deposition. The plaintiff's foundation question as to what he had previously told the two specified witnesses about the engineer's presence came near the close of his evidence, after he had given fifty-three pages of testimony and on redirect examination. By that time it was plain that he was an adverse, hostile [860] witness and the plaintiff was justified in putting the question to him in the manner she did and after receiving an adverse answer, laying the proper foundation, and subsequently proving by the two witnesses that he had previously made a statement as to the engineer's whereabouts, which was contrary to his testimony on the trial and it was not error for her to do so. If the railroad desired to limit the effect and purpose of the impeaching and contradictory evidence it should have made an appropriate request or offered a delimitating instruction. Sotebier v. St. Louis Transit Co., 203 Mo. 702, 102 S.W. 651; Courter v. Geo. W. Chase Son Merc. Co. (Mo. App.), 299 S.W. 622.

The appellant railroad assigns as error the giving of instruction two on the measure and amount of damages the jury might award in the event they found for the plaintiff. It is objected that the instruction does not confine the pecuniary loss of the widow to the evidence and is, therefore, a roving commission to the jury to find such loss in any manner they might fancy. It is objected that the instruction charged the jury to consider the age and life expectancy of the deceased husband when there was no evidence as to his age and it wholly omitted to direct the jury that they should also consider the age and life expectancy of the plaintiff, the widow. It is also objected that the instruction is confusing and misleading.

The instruction stated that the jury "should assess such damages as shall fully compensate the widow for all pecuniary losses, as hereinafter explained, suffered by her" as the result of her husband's death. In arriving at the pecuniary loss the jury was to consider the earning capacity of the deceased prior to and at the time of his death, probable future earnings, what his widow might have received from his earnings and as to the latter they were to consider his age, health, habits, etc., and his expectancy of life as shown by the evidence.

The measure of the railroad's responsibility to those entitled to recover under the act is the pecuniary loss to them. Chesapeake O. Ry. Co. v. Kelly, 241 U.S. 485, 36 S.Ct. 630, 60 L.Ed. 1117. And, "No hard and fast rule by which pecuniary damages may in all cases be measured is possible." Michigan Cent. R.R. v. Vreeland, 227 U.S. 59, 72, 33 S.Ct. 192, 57 L.Ed. 417. But the items mentioned in instruction two and to which the jury's attention was directed are the evidence of and proper matters for the jury to consider in arriving at that loss. Siberell v. St. Louis-San Francisco Ry. Co., 320 Mo. 916, 9 S.W.2d 913, 916. At the request of the railroad the court specifically told the jury that "the measure of damages to which the plaintiff is entitled, is the pecuniary loss suffered by the plaintiff by reason of the loss of the deceased." They were then directed that there were certain specific items they could not consider "but you should take into consideration the position the deceased occupied, and the support that he might have afforded to his wife had he lived, and in arriving at this amount, you will take into consideration his earning capacity, physical condition, his habits and surrounding circumstances as you may find them to be from the evidence." If the railroad thought the jury might improperly consider other items or that instruction two was too broad, a roving commission and confusing it should have requested further instructions on such matters or incorporated them in the one given at its request. Simpson v. St. Louis-San Francisco Ry. Co., 334 Mo. 1126, 70 S.W.2d 904.

The plaintiff neglected to prove Russell Crabtree's age. However, his brother testified and he said he was thirty-one years of age and that he was Russell's older brother. There was a picture of Russell in evidence. Without objection on that ground the plaintiff proved that a man twenty-four years of age had a life expectancy of about forty years. The plaintiff, his wife, testified and, although she did not state her own age, the jury saw her and could come to a reasonable conclusion as to her age and whether she was older or younger than Russell and even her life expectancy. Cervillo v. Manhattan Oil Co., 226 Mo. App. 1090, 49 S.W.2d 183. In this respect instruction two is not a model but under the circumstances of this case we do not think it prejudicially erroneous. As we have indicated, the jury could find the ages of both the husband and wife and their life expectancies as well. Hunt v. Chicago, B. Q.R. Co., 303 Mo. 107, 259 S.W. 481; Jenkins v. Wabash Ry. Co., 232 Mo. App. 438, 107 S.W.2d 204. Aside from this, however, we do not think it necessary for us to decide whether the plaintiff's pecuniary [861] loss must of necessity be measured by the husband's life expectancy, her own, the younger of them, both of them or by which the jury did in fact measure the loss. See and compare: Hunt v. Chicago, B. Q.R. Co., supra; Jenkins v. Wabash, supra; Chapman v. Terminal R.R. Ass'n. of St. Louis (Mo. App.), 137 S.W.2d 612; Stottle v. Chicago, R.I. P. Ry. Co., 321 Mo. 1190, 18 S.W.2d 433; Morrow v. Missouri G. E.S. Co., 315 Mo. 367, 286 S.W. 106; McIntyre v. St. Louis-San Francisco Ry. Co., 286 Mo. 234, 227 S.W. 1047 and McCord v. Schaff, 279 Mo. 558, 216 S.W. 320; 25 C.J.S., Sec. 90(4), p. 1233. The railroad's instruction, which we have quoted, obviously adopted the plaintiff's theory as to the measure of damages, and specifically eliminated certain things from the jury's consideration. If it desired more or wished to limit the plaintiff's loss further it should have requested further instructions on the subject. Simpson v. St. Louis-San Francisco Ry. Co., supra; annotation 87 A.L.R. 910, 934-937; 16 Am. Jur., Secs. 209, 228, 371; 25 C.J.S., Sec. 90(4).

Finally, it is contended that the verdict of $15,000.00 is excessive. The railroad does not brief this point (Homan v. Missouri Pac. R. Co., 334 Mo. 61, 64 S.W.2d 617), but in its reply brief argues that even if it be assumed that Crabtree was twenty-four years of age and that his wife's life expectancy was longer than his the annuity table (Mo. R.S.A., Sec. 3522) shows that the present value of an annuity of $1.00 a year at the age of twenty-four is $13.541 which multiplied by the deceased's earnings of $540.00 for the previous year was $7,312.14 or double an amount authorized by the evidence. As we have indicated, that is only one of the elements to be considered. If Crabtree was only twenty-four years of age he had an expectancy of about forty years. He and the plaintiff were married December 11, 1938. He was killed October 5, 1939 and though his average earnings for the previous five years were about $50.00 a month we cannot say this verdict is excessive as a matter of law and for the reasons pointed out in the reply brief. Crecelius v. Chicago, M. St. P. Ry. Co., 284 Mo. 26, 223 S.W. 413; Southern Ry. v. Bennett, 233 U.S. 80, 34 S.Ct. 566, 58 L.Ed. 860.

The judgment is affirmed. Westhues and Bohling, CC., concur.

PER CURIAM:

The foregoing opinion by BARRETT, C., is adopted as the opinion of the court. All the judges concur.