Weaver, J.
Johnson was in the defendant’s service as a railway brakeman. At the time of the accident he was employed on a freight train moving from Marion to Ottumwa. The train was a heavy one and being pulled by' two engines. On approaching the station at Sigourney, Johnson, acting as he claims in the line of his duty as brakeman, went out upon the top of a box car at or near the reat end'o.f the train and passed from car to car in the direction of'the engines. In the course of this trip he was suddenly precipitated forward from the top of a box car down upon a flat car or coal *344rack immediately in front of the car from which he was thrown, and sustained the alleged injuries on which this action is. based. The negligence charged against defendant is: (1) In the unnecessary and unusual application of the air brake by the engineer, and the sudden and violent stopping or retarding of the speed of the train, whereby Johnson was thrown forward from the box car; (2) in failing to have a sufficient number of the cars equipped with efficient automatic brakes; and (3) in employing and continuing in its employment- an incompetent and reckless engineer. Issue was taken by the defendant upon all the plaintiff’s allegations of negligence, and the cause was tried to a jury. The exceptions argued by appellant are too numerous for discussion in detail without unduly extending this opinion, and we will confine our attention to those questions which seem to be decisive of the appeal.'
1. Evidence: hearsay. I. Among the witnesses testifying in behalf of appellee was its car accountant at the general offices of the company m the city of Chicago, 111. It is conceded that this man had n0 personal knowledge as to what cars corn-posed the train and had never examined them, yet he was permitted, to testify in great detail as to their equipment with air brakes. ITe was also allowed to give it as his opinion that the “ foreign cars,” or cars of other roads making up a part of the train, “ must have been ” thus equipped. Each interrogatory calling for this evidence was objected to at the time, and the objection was overruled with a suggestion on the part of the court that, if defendant did not make it competent, it would “ go out.” At the close of this witness’ testimony the plaintiff renewed the objection and moved to strike the answers which had been given. The objection and motion were overruled with the same reservation by the court.
That the testimony was incompetent and its admission erroneous is too clear to require argument. If a witness'may be heard to testify to the physical condition or equipment of *345cars which he has never seen or examined, then all the rules against hearsay evidence and mere conclusions and opinions may as well be blotted from the books. The witness was not called, nor did he testify as an expert as to the manner in which freight cars in general are constructed and equipped, but he undertook, from the reports and records to be found in his office and from his knowledge of freight cars in general, to tell the jury just how these particular cars which he had never seen were or must have been equipped. It is hard to conceive on what theory this matter could have been offered or admitted in evidence, or how it could have been made competent for any purpose. When the witness clearly disclosed that he had no personal knowledge of the facts to which he assumed to testify, then there was no possible way of making his answers competent over the plaintiff’s objection, and they should have been ruled out.
2' ?v¿enTce? T0 waiver. It is said, however, that conceding the incompetency of the evidence plaintiff should have renewed his motion to exclude it at the close of the trial, and failing so to do the objection is waived. We think this case does not call for an application of the rule relied up0ri the appellee. The authorities cited go to the effect that the mere order of the introduction of evidence is within the discretion of the court, and, where the testimony offered is such as may become competent by making a proper preliminary showing or laying the proper foundation, the trial court may within reasonable bounds permit a change in the usual and natural order of proof. Cramer v. Burlington, 42 Iowa, 319 ; Rutledge v. Evans, 11 Iowa, 287" court="Iowa" date_filed="1860-12-10" href="https://app.midpage.ai/document/rutledge-v-evans-7092189?utm_source=webapp" opinion_id="7092189">11 Iowa, 287; Dorr v. R. R., 128 Iowa, 359" court="Iowa" date_filed="1905-06-14" href="https://app.midpage.ai/document/dorr-cattle-co-v-chicago-great-western-railway-co-7111595?utm_source=webapp" opinion_id="7111595">128 Iowa, 359; Lundvick v. Ins. Co., 128 Iowa, 376" court="Iowa" date_filed="1905-07-12" href="https://app.midpage.ai/document/lundvick-v-westchester-fire-insurance-7111598?utm_source=webapp" opinion_id="7111598">128 Iowa, 376.
But where the testimony offered is in its very nature inadmissible upon any conceivable showing, we think there is no rule or precedent for holding that the party who has made timely objection to its introduction, and thereafter has renewed his objection by motion to strike, shall be deemed *346to have waived the error, because he fails to repeat the motion. In our judgment such a rule would impose an unnecessary, not to say unfair, burden upon the party prejudiced by the error complained of.
8' admission'of hearsay. It is further urged by the appellee that the error was without prejudice, because the same matter testified to by the car accountant from hearsay was afterward shown by a witness who spoke of his own knowledge. This / contention is but another way of saying that, so ]0I1g as there is any evidence given to the jury of an alleged material fact, the admission of wholly incompetent evidence of the same fact is error without prejudice. No authority to which our attention has been called goes to this extent, and we think the error complained of was not cured. In view of the record as a whole, we should, perhaps be disinclined to reverse the judgment below on this ground, but, in view of the necessity of a new trial for reasons hereinafter indicated, we think it proper to rule upon the exception.
i. negligence: tíal evidence” ' II. It was the theory of plaintiff that the engineer controlling the air brakes unnecessarily and without warning made what is called in the parlance of trainmen an “.emergency ” application of the air and checked the speed of the train so suddenly as to throw JohnS011 from the h>OX car- I*1 Support of this claim, Johnson, who was an experienced trainman, testified that the check in speed which threw him forward was “ the setting of the air, stopped suddenly; sudden setting of the air, very suddenly and violently. In the emergency application the brakes take hold very severe at once. I would call it an emergency application.” And on cross-examination-: . “ It was an unusual stop. It was .what we. call an emergency stop — rough and quick.” And again on redi-: rect: “ You can tell when the air is set on a car, feel it take hold. I believe the air was applied in the emergency because of the manner in which the train stopped.” On part of *347the defendant testimony was offered tending to show that the emergency stop had not been used, and that the speed of the train had been checked in the manner usual in approaching a station. Upon this feature of the case the trial court, among other things, instructed the jury as follows:
(9) Even if you find it to be a fact that the box car upon which Johnson was riding bunted against the car ahead of it, and then immediately the car in its rear bunted against it, and this jolting of thé car caused Johnson to lose his balance and fall, yet this cannot be regarded by you as proving that engineer Kissinger applied the air in the emergency, or otherwise improperly applied the air brakes too violently, unless you also find affirmatively that such movement of the car is inconsistent with, and cannot be as reasonably accounted for on any other theory, for, if such movement of the car as described by Mr. Johnson may reasonably have occurred from other causes than the setting of the brakes, then its value as circumstantial evidence is destroyed.
(10) The testimony of Mr. Johnson to the effect that the box car that he was on bunted against the car ahead of it, and then the car to the rear bunted against it, is merely circumstantial evidence, which it is claimed by the plaintiff indicates that engineer Kissinger had applied the emergency; but if the proof shows that such movement or action of the car, viz., its bunting against the car ahead of it and then the car to the rear bunting against it, may also be caused by the mere shutting off of the steam on the engine, while they and the' forward cars were on a sharp curve, and the car where Johnson was and those to the réar were on a comparatively straight track, and this movement of this car is as reasonable with the one theory as the other, then you are instructed that you would not be warranted in saying from this testimony of Mr. Johnson alone on this point that this movement 6f the car that he was on shows that the engineer had applied the air in the emergency, because it may have occurred from another cause.
In still other instructions attention is called to other isolated facts and circumstances in the plaintiff’s case, and the jury are in each instance told in substance that, if such *348particular fact or circumstance could be reasonably accounted for upon any other theory than the application of the emergency stop, then it value as circumstantial evidence in plaintiff’s favor would be destroyed.
In our judgment these instructions cannot be approved. In giving them, the trial court doubtless relied, as do appellee’s counsel in argument, upon the rule laid down by this court in Asbach v. Railroad Co., 74 Iowa, 248" court="Iowa" date_filed="1888-03-12" href="https://app.midpage.ai/document/asbach-v-chicago-burlington--quincy-railway-co-7103303?utm_source=webapp" opinion_id="7103303">74 Iowa, 248, and followed in Rhines v. Railroad Co., 75 Iowa, 598, and Wheelan v. Railroad Co., 85 Iowa, 167" court="Iowa" date_filed="1892-05-16" href="https://app.midpage.ai/document/wheelan-v-chicago-milwaukee--st-paul-railway-co-7105321?utm_source=webapp" opinion_id="7105321">85 Iowa, 167. We do not intend here to enter upon any discussion of the soundness of the. rule of the cited cases, but, accepting it as the law of this state, consider whether it was properly applied in the case before us. An examination of the precedents referred to will readily demonstrate that in no instance have we gone farther than to hold that, where proof of the alleged negligence and the resulting injury rests solely upon circumstantial evidence, and the proved circumstances when taken as a whole and fairly considered are consistent with the exercise of due care on the part of the defendant, the plaintiff cannot recover. To say, however, that each particular circumstance constituting the array of evidence on which the plaintiff relies may be taken separately and subjected to that test, and that if it be found consistent with any other theory than that of negligence its value as evidence is destroyed, is an altogether different proposition. And this, it appears to us, is the vice of the charge given by the trial court. It takes up one by one I he various alleged circumstances shown upon part of plaintiff and tells the jury that, if such happening is one which “ might ” have occurred from some other cause than the negligence charged, it must be disregarded as without probative weight and value in reaching their verdict. It is scarcely too1 much to say that, if this rule is to be approved, no case, civil or criminal, can ever be established by circumstantial evidence. It may easily happen that each of a series of facts and circumstances, separately considered without reference *349to other proved facts, may be entirely consistent with due care, and yét when viewed as a whole may not be inconsistent with the conclusion of negligence.
5. Sami. We have thus far discussed these instructions upon the áppellee’s theory that the case of the appellant is wholly dependent upon circumstantial evidence, and that instructions upon that theory were called for; but we do not ^ understand the record. The plaintiff’s evidence upon the essential matters of controversy is direct and not circumstantial, and the giving of instructions applicable only to the latter was of necessity misleading.
It is true that plaintiff does not claim that he saw the engineer move the lever and apply the air to the brakes, but he does show that he was a trainman of many years’ experience and was able to tell by the manner in which the brakes took hold of the wheels and the manner of the stop that the air had. been so applied. The fact that this testimony is to some extent in the nature of a conclusion, and that such conclusion is based on circumstances other than actual observa: tion of the acts of the engineer, does not serve to classify it as circumstantial evidence. Speaking generally, circumstantial evidence, in law, is evidence tending to prove collateral facts from which the jury may infer the existence or nonexistence of the fact in controversy. When employed in any case, civil or criminal, the witness is confined to testimony concerning the collateral fact or facts relied upon, and the inference to be drawn therefrom is for the jury alone. But when, as in the present ease, he testifies to a material fact in controversy as of his own knowledge, its character as direct evidence is not always lost simply because it appears that his statement is in the nature of a deduction from other relevant facts. This is especially true as to the testimony of a skilled or experienced observer, whether he be testifying simply as an expert or to conclusions based upon his personal observation — conclusions which accord with the common experience of those engaged in the same line of employment. Not sight alone, but *350all of the senses contribute to a man’s knowledge of external circumstances, and it is easily credible that a railway brakeman, by long experience in riding upon and handling freight trains, noting the effect of controlling speed by shutting off steam, by applying hand brakes and by air applied with varying degrees of force, may be able to distinguish between the methods employed, although not in position to see the act by which the check or stop was accomplished. Such testimony, whatever its value, and however much its force may be weakened by cross-examination or by the testimony of the opposing witnesses, is nevertheless direct evidence. The instructions given by the trial court and above quoted are not in harmony with the views here expressed, and the exception taken thereto must be sustained.
Other exceptions have been argued, but, as the questions argued are not likely to arise upon a new trial, we shall not stop to note them more particularly. We think it proper to suggest that the argumentative tone which pervades the instructions as a whole is one which should be avoided. It is right and proper that the jury should be warned against allowing their verdict to be swayed or influenced by anything save the evidence and the instructions of the court, and that their attention should be clearly and forcibly directed to the rules laid down for their guidance; but this supervision should not be pressed so far that either party may justly complain that its right to have .the jury pass upon every fact question in the case has been unduly narrowed.
For the reasons stated a new trial is ordered.— Reversed.