delivered the opinion of the court.

We agree wiili the court below that, if we assumed the truth of the direct testimony of all the witnesses, we should be unable to determine which party was at fault. But a careful analysis of the evidence satisfies us that the-court below arrived at a correct conclusion. The testimony on behalf of the schooner is confusing and unreliable. It is manifest that her captain and crew had no accurate notion of distance or of navigation.- The Mosher was hut two miles off shore, and her master and crew locate the .Rochester when she was first observed so that the latter would be navigating half a mile inlaud from the shore. The witnesses, either ignorantly or willfully, misstate the facts. The testimony on the part of the .Rochester is much more reliable and consistent. She observed the Mosher when a mile and a half or two miles distant; the latter exhibiting her green light. The Rochester immediately starboarded, going to port. Upon this course the vessels exhibited to each other their green lights, and would, confessedly, had no change of course occurred, have passed each other in safety- What then induced the change which brought about this collision? The Mosher suddenly exhibited to the Rochester both her lights, and then her red light. According to the story of the captain of the Mosher, her course was not changed at any time. The learned and asiute proctor for the Mosher would therefore have us believe that this change of lights was produced by the yawing of the vessel, and claims fault on the part of the Rochester, in that she did not allow therefor, and starboard her wheel long enough or strong enough to give the Mosher a sufficiently wide berth. This notion would seem to have had being in the imagination of the proctor, for there is no suggestion of it by the schooner’s witnesses; nor do we understand there was, nor can we assume there could have been, sufficient yawing of the schooner to shut out the one light exhibited, and to exhibit the other light to the Rochester, within the distance the two vessels were apart, and within the time intervening before the collision. It is true, it was the duty of the steamer to keep out of the way of the sail vessel; the latter keeping her course. This the steamer attempted to do. and undoubtedly would have accomplished but for the remarkable change of course of the schooner. That the latter should change her course seems unaccountable, and is only explicable upon the fact that the master on two occasions left the wheel loose, which brought about a change in her course. If the result would be to cause the vessel to luff up into the wind, as is insisted by the Mosher, we think the clear inference is that the change in these lights was brought about by au attempt of the master, when lie resumed the wheel, to put her hack on her course. If, on the contrary, the effect would be to send the schooner to starboard, that of itself would sufficiently account for the change. Whatever would result, it is clear there was this sudden change in the course of the schooner, threatening collision, and which impelled the steamer to change her course to the westward. The master of the Mosher had *368charge of the wheel. The crew were engaged in setting canvas. The vessel was at a point where the commerce of the lakes converges to the port of Chicago. The master had no right to assume the duty of wheelsman, under such circumstances. His duty at that time was to keep a vigilant outlook,- — to be on hand on the deck where he could observe the movements of approaching vessels, and give orders accordingly. The City of Augusta, 50 U. S. App. 39, 44, 25 C. C. A. 430, and 80 Fed. 297. The court was of the opinion that no one on the schooner, except the captain, was on the lookout, and that his disadvantageous situation obliged him to leave the wheel, whereby the schooner went to starboard, indicating to the steamer the change in her course. We are inclined to agree with the court below upon this proposition, ■notwithstanding the testimony on the part of the schooner asserts the presence of a lookout. It would seem remarkable, if a proper lookout was stationed, that the captain should have heard from him but once during the approach of the steamer, and should have appealed to the crew engaged in setting canvas, and not to the lookout, to ascertain what light the schooner was exhibiting. It is said the steamer was at fault in having no vigilant lookout. We find no foundation in fact for this objection. The captain and the mate were on watch, in addition to the usual lookout, and they all appear to have been vigilantly employed in the performance of their duty.

It is also objected that the steamer should have stopped and backed, instead of porting her helm. Rule 21 provides that every steam vessel which is directed by the rules to- keep out of the way of another shall, on approaching her, if necessary, slacken her speed, or stop or reverse. The difficulty with the application of this rule herb is that. the steamer ported her helm and went to the starboard to avoid the schooner, and the vessels would have passed each other safely, but for the faulty action of the schooner. We think it very doubtful whether, when the schooner exhibited her red light, the Rochester could have avoided a collision by stopping and reversing. If, however, that could have been done, the course adopted was taken in extremis, to avoid an impending collision induced by the fault of the schooner, and for which the steamer should not be held blameworthy. The decree will be affirmed.