of

when we assume that all train orders were repeated back at least three times. If we thus multiply the average daily number, namely sixty-six, by the average daily repetitions, it gives us 198. This is the number of times that the method employed by the appellant was applied each day on only one of the divisions of its system. If we multiply the daily repetitions by thirty, it gives us a monthly average 5,940, while the yearly average amounts to 71,380 times. The testimony further shows that there were nine divisions on appellant's system west of Omaha, Neb., and that the trains of the Oregon Short Line Railway passed over the rails of the appellant east of Granger, Wyo. From this it is only fair to assume that the average number of repetitions of train orders as we have given them is below rather than above the average. As we have stated, the rule became effective on February 24th, 1902, and was thus in force at least thirty-two months at the time of the accident. The number of repetitions on the Evanston division within that time, according to the average we have adopted, would amount to 190,080. So far as the evidence discloses, the method adopted by appellant of transmitting and transcrib ing train orders thus failed, if it failed at all, once in 190,000 times upon one of the nine divisions of its railroad system. It may be that it failed more often, but if it did it would have been an easy matter for respondent to have shown the fact. In view that train orders are sent out as often as we have shown, and that they necessarily must be received by a great number of trainmen and station operators, it would seem absolutely impossible to conceal the defect in or insufficiency of the rule prescribing the method, if such be the fact. But notwithstanding the fact that not a single instance was shown where the method used by appellant failed, if followed by the employee, the jury condemned the method, and did so upon the sole ground that a considerable number of other railroads used a different method. 10, 11 There is no evidence whatever that by following the method used by such other roads mistakes are obviated, nor is

there any that under the method used by appellant mistakes are not avoided if the rules are followed and obeyed by the employees. The jury simply assumed that such was the fact, and, under the instructions as given by the court, they were authorized to assume it. Not only were the jury permitted to enter the realm of speculation with regard to the foregoing facts, but they were also permitted to assume that appellants' rules were insufficient when followed.

The test of whether a rule is insufficient or not lies in the fact that it fails to afford adequate protection when followed, and not that it fails because it is disregarded. Quite true, a rule may be so framed that it is impracticable to follow it, but, if it is, it may be insufficient for that reason. Moreover in the instruction it is in effect assumed that a rule or method which provides that numbers in train orders shall be stated in both words and figures is a proper rule, and a standard to which all other methods shall conform. With regard to this the court told the jury that, if they found that ordinary care required that the defendant "should have adopted a rule providing that when time is stated in the body of a train order it should be expressed both in words and figures and not simply in figures, then the court charges you that the failure of the defendant company to exercise such ordinary care would constitute negligence." The duty cast upon the appellant by this instruction is not that it was required to exercise ordinary care and diligence to provide rules which under the circumstances would prove adequate when followed to protect the employees against avoidable and unnecessary dangers, and which would make the conduct of its business reasonably safe, but the appellant was in effect declared negligent unless it adopted a particular method, namely, that of stating the time in the body of its train orders in both words and figures. We know of no law by which an employer is required to adopt one method rather than another in conducting his business. The law, as we understand it, upon this point is well stated in

12

Norfolk & W. R. Co. v. Cromer, 101 Va. 671, 44 S. E. 899, in the following language:

"Courts and juries cannot dictate to railway companies a choice between methods, all of which are shown to be reasonably adequate for the purposes intended to be subserved. Thus to subject them to the varying and uncertain opinions of juries in questions of policy, and to substitute the discretion of the latter for their discretion would be wholly impractical, and would prove alike disastrous to the companies and the public."

13

If this is the law, it is obvious that whether appellant was negligent or not with regard to the conduct of its business cannot be determined by a mere comparison of its methods with those employed by others, although it were conceded that the method used by others was reasonably safe. It does not necessarily follow that because one method is safe a different one is not.

As we have pointed out, there is no evidence in this case which shows, or tends to show, that the rules adopted by ap pellant which governed the transinission and transcribing of train orders were less effective if followed than were the rules of other companies upon that subject. As we have seen, the rules of appellant required each order and each repetition thereof to a different operator to be repeated back to the dispatcher who issued and transmitted the orders. If it was repeated back correctly, it received the "O. K." of the dispatcher; otherwise not. In repeating it back, the operator thus had the correct order before him, and by another rule he was required to trace all copies from the order which was proved to be correct. If the operator thus had the time correctly stated in figures, and if he traced the additional copies from the correct order, it is not easy to see why all copies would not have been the same. No doubt, if the operator disobeyed the tracing rule and attempted to copy from memory, a mistake might occur in the copy although the original was correct. But can it be said that a rule is insufficient simply because a mistake may occur if it is not followed? Would it not be quite as easy and just as probable

that the operator would make a mistake in copying a train order in which the time was stated in both words and figures as in copying one in which the time was stated in figures only? If this is so, how can it be said, in the absence of any evidence upon the subject, that a rule which required time to be stated in both words and figures is sufficient, while one which required the time to be stated in figures only is insufficient? The question here is not that the appellant was negligent in issuing and transmitting an incorrect train order. There is no claim nor evidence that the train dispatcher did not issue and transmit the order giving the running time of No. 3 correctly, nor is there any claim or evidence that the order was so imperfectly written that it either confused or misled the trainmen. The order in question was plain and easily understood. The only defect in the order was that "50" was, by some one and in some way not disclosed by the evidence, substituted for "30," and in that way gave a difference of twenty minutes in the actual running time of No. 3. If the dispatcher had issued such an order and had thus misled the trainmen, and this order had been the proximate cause of the collision, a different question would be presented. No such a claim is made in the complaint, and there is no evidence to substantiate such a claim. Neither was the case submitted to the jury upon such a theory, but, as we have seen, it was submitted to them upon the theory that the rules of the appellant with regard to the transmission and transcribing of train orders were insufficient upon the sole ground that they did not require the time in train orders to be stated in both words and figures.

A striking illustration that a rule which requires time to be stated in both words and figures may not, under all circumstances, be the best, is afforded in the case of McLeod v. Grinther, 80 Ky. 399. In that case a train order was issued which gave a certain train until 10 o'clock to make a certain point. The order was written thus: "Fish extra east; can have until ten, 10 o'clock a. m., to make Beards for number 2 and number 4." This order was intended to mean 10

o'clock, while the conductor construed it to mean ten minutes after 10 o'clock. The company was held liable in that case for not exercising ordinary care in writing the message, in that the figures "10" should have been placed in brackets to show that they were intended as a repetition of the word "ten," and not as additional time. We cite this case merely to show that neither a court nor jury is justified in assuming that one rule or method is to be upheld and another should be condemned upon the sole ground that they differ with regard to the methods that are to be followed in doing a particular thing which requires careful and constant regulation. In this connection it should also be kept in mind that, where rules have been deliberately formulated and promulgated by the master for the government of a complex and dangerous business, they should be condemned only when it is clearly made to appear that they are in fact insufficient. "The presumption is that rules which the master has adopted for the government of his servants, to prevent injuries to them, are reasonable and sufficient." (26 Cyc. 1159; Rex v. Pullman Palace Car Co., 2 Marvel (Del.) 337, 43 Atl. 246; Little Rock, etc., Ry. v. Barry, 84 Fed. 944, 28 C. C. A. 644, 43 L. R. A. 349; Vedder v. Fellows, 20 N. Y. 126.)

14

While in the last two cases cited the doctrine that the question of the sufficiency of a rule is one of law and not of fact is enforced- a doctrine which we think is against the weight of authority-yet, so far as we are aware, the courts, whether they hold to the doctrine that the sufficiency of rules is a question of fact for the jury or one of law for the court, all agree that the insufficiency of a rule should. clearly be established before it is condemned. This court is committed to the doctrine that the sufficiency of a rule is a question of fact. (Johnson v. U. P. Coal Co., 28 Utah 46, 76 Pac. 1089, 67 L. R. A. 506.) The controlling reason, however, which has impelled a majority of the courts to hold that the question is one of fact rather than of law is because it requires evidence from a competent

15