(after stating the facts as above). [1] We think that this was a crossing, not a passing, case. We must choose between the story of the Newburgh, backed up by Eldridge, and the supposition that each vessel was passing the other port to port on directly opposite courses 300 feet apart. If so, it seems to us highly improbable that the Newburgh should ever have sought a starboard to starboard passing, even though we should agree that, had she done so, the' McGuirl would have assented. The suggestion that the Newburgh meant to avoid the New York shore further up because of possible shipping appears fantastic. On the other hand, it seems hard to accept Eldridge’s story that the McGuirl was heading only two points downstream, because not only does it contradict the story of the McGuirl, but there was no conceivable reason for her still so doing half a mile below Guttenburg. Munson says that the Mc-Guirl was headed “about the same” as he, which is consistent with a slight heading on the New York shore. Furthermore, there is no improbability that the McGuirl had already begun to head in towards her slip a mile below; the tide had not yet become a serious factor. Finally, there is every indication that the Newburgh, taken unawares, was presented with a sudden situation which in her judgment demanded some action on her part. The only relative position of the vessels, consistent with that fact, is a crossing case, though at a very small angle, probably about one point.
The learned trial judge believed that the “McGuirl starboarded after accepting the Newburgh’s proposal to cross her (the Newburgh’s) bows,” and in this he is supported by the apparently disinterested tes*439timony of Eldridge. Yet Eldridge heard no signals from the McGuirl, and swore that she starboarded five points at the last moment, which is certainly impossible. His observation must therefore have been faulty. Moreover, we regard it as most unlikely that a burdened vessel, having agreed to cross another’s bows, and having started to do so, should in the middle of that movement change and do the exact opposite for no ascertainable reason. We think it much more likely that she relied upon the superior speed and turning power of the Newburgh, and either ported slowly or did nothing at all. It is, of course, possible that her master, after starting to cross, supposed the Newburgh would not follow her signal, and in haste attempted to starboard, yet that presupposes an emergency in his judgment, and such an emergency would ordinarily have called for an alarm on his part. As we are forced to choose between the supposition that she substantially kept her course, or first ported and then starboarded, we accept the former.
[2] These facts being established, it seems to us clear that the Newbtirgh made too close a shave, or, having a faulty lookout, was taken by surprise, as is made plausible by the short interval between her signals and the little distance left her after the McGuirl’s assent. If we were to assume that she did not shave too close, but merely found herself at too close quarters, it would still follow that she had allowed the McGuirl to get too close for safe navigation after she got her assent. Either she ought to have stopped and backed at once, when the answer was too long delayed or without making any such proposal, or she had distance enough to execute her proposal after getting the Mc-Guirl’s assent, and failed in her own navigation. In any event, she is in a dilemma which puts her in fault.
[3-5] We have decided that the McGuirl did little or nothing after assenting to the Newburgh’s two whistles, but substantially kept her course and speed. Her navigation in this view raises a question upon which we have found very little in the books. It is good law that, when the burdened vessel decides to “keep out of the way” by crossing the bows of the privileged vessel, though she gets an assent to such a proposal, she assumes the risks involved in choosing that method. The Nereus (D. C.) 23 Fed. 448, 455, 456; The Greenpoint (D. C.) 31 Fed. 231. The duty of the privileged vessel in such cases is to co-operate, and she need not keep her course. The Sammie (C. C.) 37 Fed. 907; The M. Vandercook (D. C.) 88 Fed. 559, affirmed 105 Fed. 1004, 45 C. C. A. 183. The situation, at least in this circuit, after the agreement, is one of special circumstances. The George S. Shultz, 84 Fed. 508, 510, 28 C. C. A. 476 (semble). But such an agreement, initiated by the privileged and assented to by the burdened vessel, might be regarded as creating other duties. It could be considered as a proposal that the duties of the vessels should be reversed, and that the burdened (now the privileged) vessel hold her course and speed, so that the privileged (now the burdened) vessel might be able to forecast her positions at future moments, precisely as the rule requires when no agreement has been made.
We have been unable to find much in the books which touches on the precise point. In The Susquehanna (D. C.) 35 Fed. 320, 322, the bur*440dened vessel was exonerated because she did not “thwart” the proposal, having apparently kept her course. On the other hand, in The Columbia (D. C.) 29 Fed. 716, 720, Judge Brown thought it.a matter of indifference which vessel proposed the change; the burden always remaining upon the vessel originally burdened. In Stetson v. The Gladiator (D. C.) 41 Fed. 927, Judge Nelson said that the exchange justified the burdened vessel in keeping her course.
In none of these cases was the originally burdened vessel held to any duty to keep her course. On the whole, we are disposed to think that any agreement to change the usual rules should be treated as creating thereafter a position of special circumstances. If so, we think that, although the proposal emanates from the privileged vessel, and should, be taken as meaning that she will undertake actively to keep but of the way, it need not absolve the burdened vessel from her similar and original duty also to keep out of the way, nor will it impose on her a rigid duty to hold her course and speed. It is true that that duty is imposed by the.rules generally as a correlative to the duty to keep out of the way, but only in cases where no agreement has been reached. Some convention is essential when neither knows the other’s purposes, but when both have agreed upon a maneuver by an exchange of signals, their accord should be left for execution by movements adapted to the circumstances. For example, if the angle of crossing is wide, it will usually be best for the originally burdened vessel to hold her course and speed; but, if it be narrow, it is safest for both to starboard and pass at a greater distance. No doubt the proposal involves the proposer in a duty to give a wide enough margin for safety, even though the-assenting vessel do not starboard.
[6] In the case at bar the McGuirl, having assented, did nothing, or at most very little, as the position at collision proves. The crossing was at a narrow angle, and the assent involved for safety some active co-operation on her part under those special circumstances. Each was under a duty to keep out of the way of the other, and to this extent the original duty of the McGuirl remained. Her co-operation included more than the mere absence of anything which might “thwart” the New-burgh. Had the angle been wide, her best co-operation might well have been to hold her course and speed. In this respect her fault may have been less than the Newburgh’s, who had'suggested the change; but upon that we do not feel disposed to speculate. It is enough that the agreement in fact created a situation not much removed from a starboard to starboard passing, in which both vessels have an active duty.
While, therefore, we cannot agree in imputing to the McGuirl the final porting, which the learned trial judge accepted, the result is the same, and both vessels were at fault.
The decree is affirmed, with costs.