On March 6, 1897, defendant company sent out from Barnesville a passenger train bound for Grand Forks. It was a “double header”; the head engine (No. 600) being in charge of engineer Henry Gowenlock, and the second one (No. 143) in charge of engineer Lewis Mason. The head engine was of large capacity, fully equipped with air brakes. No. 143 was smaller, and had no air equipment, and no brake except the usual hand brake on the tender. Air was piped from the head engine, by means of a rubber tube, through the second engine, to the train behind. This placed the train under the control of Gowenlock, so far as the air brakes were concerned. The means of communication between the two engineers was by whistle. A short, sharp whistle by No. 600 was the signal for No. 143 to shut off steam. When this train was approaching Mallory station, between Crookston and Grand Forks, the front trucks of No. 600 left the rails at a point about three hundred feet east of the switch. The engine kept to the rails in this condition until it reached the switch, when it left the main track and followed the switch or southerly, track. The second engine and the train kept on the main track. At a point about one hun*452dred twenty feet west from the frog, No. 600 left the track, and fell upon her side, on the south side of the track. No. 143 passed on west about forty feet, and was then derailed to the north of the main track. The head coach piled in between the engines, and there seems to have been a general wreck, — at least, of the front end of the train. Gowenlock and his fireman’ were killed, and this action was brought to recover damages for the death of the engineer, Gowenlock.

The specific charge against defendant is that, as soon as the deceased discovered that something was wrong with his engine, he turned on the air brake, shut off steam from his engine, and signalled Mason to shut off steam, but that Mason paid no attention to the signal, but kept his engine going, with the result that, when No. 600 struck the switch and partly separated from the train, No. 143 pushed her along and off the track, causing the engineer’s death. Plaintiff had a verdict for $5,000, and defendant appeals from an order denying a motion for a new trial.

The case was in this court upon a former appeal (76 Minn. 146, 78 N. W. 972), and was reversed for error occurring at the trial; and, while the facts now before us are substantially the same as were presented at that time, we will examine the evidence with reference to its sufficiency, without regard to the intimation of the court upon the former appeal that the verdict at that trial was supported by the evidence. A proper analysis of the evidence may be made under the following heads: (1) Were the front trucks of No. 600 off the rails? (2) Did Gowenlock discover it and give the signal? (3) Did Mason hear the signal, or, if he did not, should he have heard it? (4) Was the failure by him to shut off steam and reverse his engine the proximate cause of the accident?

1. The train was approaching the station at twenty or twenty-five miles an hour, and ran through a snowdrift about two feet thick, three hundred twenty-five feet east of the switch. At that time the front trucks of No. 600 left the rails and traveled close to them on the south sides of the rails. This is shown by an examination of the xfiarks upon the rails, spike heads, and ties immediately after-wards; and one witness (Geddes), who stood about two hundred feet west of the switch, says he heard the thumping of the engine. The *453front trucks must have been off, because they followed the rails of the side track, which came in at the switch from the south side, and which were then open.

2. That Gowenlock discovered that something was wrong with his engine appears from the fact that about the time he passed the drift, between two hundred and three hundred feet east of the switch, he gave the sharp-whistle signal for Mason to shut off steam. The sounding of this whistle is positively sworn to by three different witnesses, — McCabe, who was in the fourth coach of the train; Geddes, who stood about two hundred feet west of the switch; and Davidson, who was up by the station, about twelve hundred feet west of the switch. Although the position of the train at that time is not definitely located, the signal was heard shortly after the station whistle, and while the train was at least two hundred feet east of the switch. The positive testimony of these witnesses was sufficient to justify the finding that the signal was given immediately after striking the snow, although there was considerable testimony of a negative character to the contrary.

3. Mason testified that he did not hear the signal; that he felt no shock until his engine was derailed, but heard the station whistle just within the mile limit, and then opened his window on the right-hand side and looked out, saw a man standing there, but did not see him make a gesture. His cab was fixed up with canvas and curtains to prevent snow from entering. If Mason heard the station whistle, why did he not hear the second one? If his window was open, his opportunity to hear it was better than before. If McCabe, on the train, felt a jar east of the switch, why did Mason not feel it? From the evidence bearing on this point, the jury were justified in finding that Mason heard the signal and did not act upon it. But in this connection must be noticed appellant’s eighth assignment of error. The court charged the jury as follows:

“But if it was given, and Mason actually heard it, or if he did not hear it, but could have heard it by the exercise of that care and caution which an ordinary, skilful engineer ought to have exercised under like circumstances, then it would be his duty to answer it and to shut off steam, and in that case he would be deemed guilty of negligence in failing to shut off stea-m. And if the jury believe that such negligence was the natural and proximate cause of the injury, *454and that the deceased was not guilty of contributory negligence, then the plaintiff would be entitled to recover.”

Mason was occupying a very responsible position. He knew that the object of putting two engines on that train was to clear the track where snow had drifted. He knew that his only means of communication with the head engineer was by means of a whistle. He knew that his engine was not provided with air brakes and had no means of stopping independently, and that he was under the control of Gowenlock. Mason says he opened the window to look out and see where they were, and that snow was flying. As the snow bank was three hundred twenty-five feet east of the switch, he must have continued to look out for a distance of over two hundred feet; for Nesbit, who stood opposite the switch, saw him still looking out as he passed, and motioned to draw his attention to the head engine. Was Mason’s attention so diverted by what he saw outside and by the flying snow that for a moment he was oblivious to the signal whistle? Under the conditions, he was charged with grave responsibility, and the question submitted to the jury was, could he have heard the signal by the exercise of that care and caution which an ordinary, skilful engineer should exercise under such circumstances? The instruction presented the proper test, and contained no error.

4. We come now to the critical question in the case: Was the failure of Mason to shut off the steam from and reverse the engine, after hearing the signal, the proximate cause of the injury and death of plaintiff’s intestate? Is there anything to- show that the accident resulted from the head engineer’s own carelessness, or in a failure to act promptly?

The witness Geddes testified that he saw no steam from the head engine, and that she did not seem to be working, but that No. 143 was working steam as they approached the switch. Nesbit stated that No. 143 was puffing smoke and making a noise, and that he saw no steam coming from No. 600, as they passed him at the switch. WThen No. 600 was examined immediately after the wreck, the brake lever was found in the position which indicated that it had been pulled to apply the air brakes. The brakes were also found to be *455set ón the engine. The evidence in reference to the four facts mentioned (that is, the giving of the signal, the shutting off steam, the position of the brake lever, and the position of the brakes) was sufficient to justify the jury in finding that G-owenlock shut off the steam and set the air brakes as soon as he reasonably could after discovering that all was not right with his engine. And thus is eliminated the question of the negligence of deceased.

There is a discrepancy as to the speed of the train from the point where the trucks left the rails until it reached the switch. The highest point marked by any witness was twenty-five miles per hour, and the lowest ten miles. One claimed that the train slowed up considerably while passing through the snowdrift, and then increased in speed as it emerged. Others claimed that the speed was about the same until it passed the switch. It is fairly inferable that the speed was not less than fifteen nor more than twenty miles per hour while the train was covering the three hundred twenty-five feet immediately east of the switch. No. 600 was a “Mogul,” and was capable of one hundred eighty pounds pressure to the square inch, while No. 143 had a capacity of one hundred forty pounds. The train consisted of seven passenger cars, which, with the engines, made a total of over five hundred feet in length. It is seventy-five feet from the switch to the frog, and the side track which connects at the switch gradually diverges to the south for a distance of about two hundred fifty feet, and then runs parallel to the main track; the space between them being about ten feet. - The head engine was found turned end to end on the south side of the side track at a point about two hundred feet west of the switch. No. 143 passed on about forty feet further, and stopped, on the north side of the main track, retaining an upright position, completely off from and close to the north rail. The front coach stopped on the main track, partly derailed, the front end close to the rear end of No. 143, and the rear half of the car being just north of and by the side of No. 600. The total distance traveled by No. 600 after her front trucks went off was about five hundred twenty-five feet.

There was some attempt to show by experts within what distance such a train, going fifteen to twenty miles per hour, could have *456been stopped by applying the air. One witness called by plaintiff stated two hundred, and another two hundred twenty-five, feet. Other witnesses called by defendant stated from three hundred to four hundred fifty feet. Geddes, whose position was nearly opposite the place where the engines went off, stated that No. 143 seemed to be pushing No. 600. The location of the large engine indicated that she was thrown off and turned around by means of a force from the rear. The question is, what was that force? Was it the momentum of the train, without regard to the energy of No. 143, or was it the strength put forth by No. 143, regardless of the train’s momentum? When was the air applied by Gowenlock? If at the time of giving the signal, then would the brakes stop the train before reaching the point where the engine left the track, and, if they would, could the energy put forth by No. 143 overcome the resistance of the brakes and pull the train along, and at the same time push the engine ahead?

All of these possible contingencies and inferences arise, and appellant argues with much force that it is not possible to draw the conclusion that the failure to shut off the steam from No. 143 by Mason was the proximate cause of the overthrow of No. 600. But there were many things open to the consideration of the jury. It may reasonably be supposed that Gowenlock gave the signal to Mason when at a distance not less than two hundred fifty to three hundred feet east from the switch, and that Mason, if he had been attentive, would have at once shut off steam and reversed his engine. It would require a certain amount of time for Mason to act, but it is not unreasonable to assume that he might have acted while the train was passing over one hundred or one hundred fifty feet, so that his engine could have been made dead weight, if not a positive force in resistance by reversing, at a point at least one hundred feet east of the switch. The side track diverged from the switch very gradually, and the further the two engines proceeded, the further apart they became. If they were fastened together, the •one following the side track and the other the main, there would come a time when they must separate on account of the increasing space between the tracks. While running over the first one hundred or one hundred fifty feet west from the switch they would not *457be far apart, and the tendency for No. 143 to push No. 600 ahead without upsetting her would seem to be less than for the next one hundred feet. The tendency to push her over to one side rather than ahead would increase rapidly in proportion to the distance. There is no.way to tell at what particular place they became separated, or what held them together so long. But it is a fair conclusion to draw that, if the force exerted by No. 143 had been removed when it might have been, the engines would not have come to that particular relation with each other which resulted in the force which overturned No. 600. From all the circumstances and conditions surrounding and connected with the incident, the jury were fairly warranted in coming to its conclusion without entering the realm of speculation and conjecture.

As to the other assignments of error, two only require notice. McCabe, a witness for plaintiff, was asked this question:

“Do you know whether or not the whistle was about the time the train started off?” to which he answered: “I should judge it was about the time. Q. By Mr. Wellington: That is a matter of judgment? A. Yes, sir.”

Defendant moved to strike out the answer as irrelevant, incompetent, and immaterial. The answer to the question was not responsive, but the objection was not directed to that point. But merely admitting, in answer to defendant’s counsel, that his statement was based on his judgment, would not make the answer subject to the objection made. The judgment of the witness was necessarily based upon his recollection. His answer was equivalent to saying, “To the best of my recollection.” The inference that the witness was merely guessing is not warranted.

Again, counsel for plaintiff asked the following question of an expert witness:

“Assuming that all the testimony is true that you have heard in this case, what could the second engineer have done to have aided in stopping that train?”

It was answered by the statement that he could have shut off steam and reversed the engine. Objection was made upon the ground that there were no facts given upon which to predicate an *458opinion. It is argued that it did not appear that the witness had heard all the evidence. The question assumes that the witness had heard the testimony, and the opinion called for was based upon such evidence as was in the case. The sufficiency of that evidence goes to the weight of the opinion based upon it, rather than to its competency.

Order affirmed.