VANN, J.

The person selected by the parties to make the estimate was in the employ of neither, yet, as chief engineer of the railroad company, he sustained such a relation to both as to make it the interest of each that his estimate as to the materials furnished and work done by the plaintiffs should be as large as possible, for it determined the amount of the plaintiffs' compensation as sub-contractors and of the defendant's profits thereon as contractors. This case, therefore, is unlike those, so frequently arising, in which the certificate or estimate is required from an architect or engineer in the employment of one of the parties. In that class of cases the danger that the person acting as an arbitrator might favor his employers is obvious. While neither natural nor legal disabilities hinder a person from being an arbitrator, provided the fact is known to the parties at the time of the submission, still, as he is the agent of both parties alike, and impartiality is the fundamental requisite, the courts closely scrutinize the action of an arbitrator whose relation to one of the parties was such as to naturally influence the judgment even of an honest man. (Morse on Arb. and Award, 99; Russell on Arb, 105.)

In this case, however, there was no reason why the person selected should not be wholly disinterested and impartial. The parties stood upon an equal footing, their contract was without legal objection, and the arbitration clause is as binding and should be enforced the same as any other provision. In one sense, as was said in a case somewhat analogous, the submission to the determination of the engineer is more obligatory than any ordinary submission to arbitration, inasmuch as, being upon consideration, it is not revocable, and the obligation upon the defendants to pay did not, by the terms of the contract, arise until the estimate was made by the engineer. ( Herrick v. Vermont C.R. Co., 27 Vt. 673, 679.) A valid award or estimate operates as a final and conclusive judgment, and however disappointing it may be, the parties must abide by it. (Id.; Perkins v. Giles, 50 N.Y. 228; Fudickar v. Mut. Life Ins. Co., 62 id. 392; Kidwell v. B. O.R.R. Co., 11 Gratt. 676; O'Reilly v. Kerns, 52 Penn. St. 214; Vanderwesker v. Vermont Central R.R. Co., 27 Vt. 130; Ranger v. Great Western Railway Co., 5 H. of L. Cas. 71; 2 Wood's Railway Law. 995; 1 Redfield's Law of Railways, 438.)

The estimate made by the chief engineer should not, therefore, be set aside or disregarded unless some good reason is shown for such action. The trial court, without deciding that there was any error in the estimate, adjudged, by its interlocutory decree, that, if upon the reference ordered, any error should appear in the estimate, it should be corrected, and that the party in whose favor a balance then appeared should recover the same from the other. The only reason appearing in the findings or suggested by the evidence for thus disturbing that which the parties had expressly stipulated should be final, is, that the chief engineer did not personally measure the work, and that when the final estimate was about to be signed he refused to allow the plaintiffs to call a witness to contradict the statements already made to him by the subordinate engineers. This involves an inquiry into the nature of the power intrusted to the chief engineer. Was he an arbitrator, as that term is understood at common law, or was it his duty in estimating quantities to simply make a summary computation, as held by the learned General Term? The answer to this question must be found in the contract, which is both the source and limit of the power under consideration. The contract, however, is to be construed in the light of the surrounding circumstances, and in connection with the agreement with the railroad company, and the actual intention of the parties thus ascertained. The power in question was confided to a man, who, as engineer in chief, was building a railroad extending from the Missouri river to the Pacific ocean. The single division of the road to which the contracts related was more than two hundred miles long, and it was to be completed between the 29th of September, 1871, and the 1st of July, 1872. Could it have been within the contemplation of the parties that the head of so great an enterprise should make the measurements himself, or even personally superintend them when made by others? The plaintiff's contract provided that the "square timber and plank in structures and flattened timber in structure, as well as for all pile and trestle, and all other timber structures required on the road-bed of the Dakota division, "should be paid for at so much per thousand feet, "board measure;" piling at a certain price "per lineal foot of piles driven;" and "all iron used in above work, at ten cents per pound." The bridges were to be paid for at so much per lineal foot, with an increase of price as the spans increased in length.

Considering the extent of the railroad, the time provided for its completion, the details and complications in the measurements and the nature of the duties of the chief engineer as implied from his position, even if it would be possible for him to give the requisite personal attention to the subject, it would be unreasonable to expect it. As said by the court in the Herrick Case ( supra), "When we come to know that practically the chief engineer never does and never can make these estimates, or even verify those made by others, that the thing is altogether impracticable, we must conclude that the parties had reference to something which was usual, or at least possible in such cases." It was accordingly held that an estimate by the assistant engineer was sufficient in that case. When the chief engineer was constituted the sole judge between the parties of the quantity of work done and materials furnished by the plaintiffs, they did not provide that he was to measure, but that he was to estimate. In McMahon v. New York Erie Railroad Company ( 20 N.Y. 463), the contract provided that the measurements were to be made by the engineer, but the parties did not require that in this case, where compliance would have been virtually impossible. They evidently meant that he was to act in some way that was possible and practicable, as otherwise they could not expect him to act at all, for neither of them had any control over him. But how was he to make the estimate if not from personal measurement or observation? Upon what was he to base it? How was he to get at the facts? The contract, interpreted in the light of the surrounding circumstances, suggests the answer that he was necessarily to rely upon the reports of his subordinates. No other way was practicable for estimating so great a work, extending over so many miles of territory. Even if one man could do it, the head of the engineering department that was building a railroad across a continent would not be selected for the purpose. The position of the chief engineer made him conversant with the general facts, and gave him a thorough knowledge of the engineers under his control. He was in a situation to exercise his judgment upon the reliability of their reports, and could direct others to revise their measurements if he deemed it necessary. The same means that he employed to protect the railroad in its payments to the defendants were apparently regarded as sufficient to protect the parties as between each other. The evidence shows that the estimates were made from actual surveys and measurements by engineers of the railroad company in the presence of the plaintiffs' foreman. After the work was finished one of the plaintiffs, with one of the defendants, met the chief engineer and the division engineer, and together they made up an estimate which, according to the deposition of a witness read in evidence by the plaintiffs, and not contradicted, so far as appears, was agreed upon by all, after certain concessions had been made. This estimate included everything, except certain iron and lumber then on hand, which it was agreed should be estimated and added to make it final.

It is to be observed that the plaintiffs, when they learned that the estimate was about to be signed, made no objection to the action of the chief engineer because he had not personally measured the work, nor did they request him to measure it or to cause new measurements to be made. They did ask him, however, to take the evidence of their foreman as to the true quantity of work. He replied that he must rely upon the reports of his subordinates as to the quantity, and that he could not take the testimony of the contractors or their employes.

Was it the duty of the chief engineer to hear evidence? We have already held that the nature of his trust was such as to permit him to rely upon the reports of his subordinates in making his estimate. The same reasoning which led us to that conclusion applies with equal force as an answer to this question. Did the parties expect him to try a law suit, for if the door is opened to admit one witness, why should not all who know anything about the matter be allowed to come in? And what would this involve? The statement introduced in evidence by the plaintiffs upon the trial consists of sixty-five printed pages of items, considerably exceeding one thousand in number. Nearly every item is an aggregation of other items not appearing in the statement. Each states the number of pieces, their designation, size in inches, length in feet and inches, quantity by board measure, kind of timber or material, and in some instances other particulars. This includes but three-fourths of plaintiffs' work, as no question was raised in relation to one residency or subdivision. Each item and each sub-item might give rise to an issue, and become the subject of controversy. Did the parties expect that a man charged with the responsibility of building a great railroad could stop long enough to enter upon an investigation, through witnesses called and sworn, with such possibilities? To ask this question is to answer it. We think that it was the intention of the parties to clothe the chief engineer with the power of summary computation, based upon his experience in building railroads, his general knowledge of this road, the original surveys, measurements, plans, specifications and such other data as would be presumed to be in his possession, but chiefly upon the reports of the skilled engineers working under him. It thus became his duty to exercise his judgment upon all the facts thus ascertained, and to fairly make the estimate. There was no delegation of authority further than was impliedly authorized by the contract. ( Wiberly v. Matthews, 91 N.Y. 648; Billings Law of Award, 76, 77).

What was the effect of the estimate made by the chief engineer? The intention of the parties to make it final is evident from the language used in their own agreement, and it is emphasized by the reference to the agreement between the railroad company and the defendants, which contains a provision making the estimate of the same person conclusive upon the parties to that instrument also. The defendants evidently intended that the work, which they had agreed to do for the railroad company and which they sub-let at a profit to the plaintiffs, should be estimated under both contracts by the same person, so that the estimate which was conclusive between the railroad company and themselves should also be conclusive between the plaintiffs and themselves. The plaintiffs, by signing their contract, united with the defendants in the effort to carry this intention into effect. The construction of the contract, its performance and all matters of difference that might arise in relation thereto, were submitted in advance to the chief engineer, whose decision was made final and conclusive upon both parties, each of whom waived any right of action, suit or other remedy "in law or otherwise," under or arising out of the contract. Furthermore, it was made a condition precedent to the payment of the plaintiffs that the chief engineer should furnish to the defendants his certificate that the work had been completely finished in every respect, according to the stipulations and specifications, together with his estimate of the quantity of work done, which estimate it was agreed should not only be final and conclusive upon both parties, but it was also fixed upon as the basis to determine the amount due to the plaintiffs. Compliance with this provision thus became a condition precedent to any recovery by the plaintiffs. ( Butler v. Tucker, 24 Wend. 447; Smith v. Briggs, 3 Den. 73; D. H. Canal Co. v. Penn. Coal Co., 50 N.Y. 250; United States v. Robeson, 9 Peters, 319; Story's Eq. Juris. § 1457a.)

The chief engineer had the general powers and duties of an arbitrator, except as they were expressly or impliedly restricted or increased by the contract of the parties. We have held that they could not have intended that which they knew to be impossible, but, as they agreed that he should be an umpire to decide all matters arising or growing out of the contract, the same effect should be given to his estimate, when made as to the decision of an umpire or arbitrator. No court has any general power of supervision over the awards of arbitrators. There is no claim of fraud, corruption or bad faith on the part of the chief engineer, and the only misconduct charged against him has already been noticed. It is, however, claimed that he made a mistake in his estimate. The court did not so decide when it made its interlocutory decree, which is based wholly upon the failure to personally measure, and the refusal to hear evidence. It is well settled that while an award may be set aside for a palpable mistake of fact in the nature of a clerical error, such as a miscalculation of figures, still, in general, such mistake to be available must appear on the face of the award or in some paper delivered with it. ( Fudicker v. Guardian Mut. Life Ins. Co., 62 N.Y. 392; Morris Run Coal Co. v. Salt Co. of Onondaga, 58 id. 667; Woods v. Monell, 1 Johns. 101; Todd v. Barlow, 2 id. 551.)

The party who seeks to set aside an award upon the ground of mistake must show from the award itself that but for the mistake the award would have been different. (Id.) No mistake appears upon the face of the estimate made by the engineer. The alleged error did not appear until after the entry of the interlocutory decree, and at the close of a protracted trial before the referee. The merits of an award, however unreasonable or unjust it may be, cannot be reinvestigated, for otherwise the award, instead of being the end of the litigation, would simply be a useless step in its progress. The parties provided by full and explicit stipulations for a final and conclusive estimate by the chief engineer as an essential prerequisite to any payment by the defendants. Of what value are these carefully drawn provisions if a mistake is to open the whole matter and throw it into the courts for re-examination? Is it within human power to precisely compute or exactly estimate a work of such magnitude, and with so many details, without any mistake? Is an error of a few dollars in a hundred thousand to throw open the gates of litigation? If not, why should any error, unless it be so great as to be evidence of fraud, corruption or bad faith? By what other rule can a mistake be measured so as to establish the standard for interference by the courts?

We regard the estimate of the chief engineer as conclusive, and that, in the absence of proof of corruption, bad faith or misconduct on his part, or palpable mistake appearing on the face of the estimate, neither party can be allowed to prove that he decided wrong as to the law or facts. ( Perkins v. Giles, 50 N.Y. 228.)

We do not desire to intimate, however, that he decided wrong in any respect, for the evidence strongly tends to sustain his conclusions, but whether his estimate was right or not, the parties by their contract conclusively committed their rights to him and they must abide the result.

The order of the General Term; reversing the judgment of the Special Term and ordering a new trial, should be affirmed and judgment absolute ordered against the plaintiffs in accordance with their stipulation, with costs.

All concur.

Order affirmed and judgment accordingly.