These are cross-appeals from a decree holding both vessels at fault in two consolidated cases arising out of a collision in the North River, between the New York and New Jersey shores, shortly after half past six in the morning of August 22, 1956. Visibility was from one to three miles, despite a slight haze, the wind was light from the Northeast and there was almost no current as it was at the end of an ebb tide. The running lights on both vessels had not yet been switched off.

The steam tug Esso New Hampshire had two barges in tow: the Esso No. 21 made fast to her starboard and the Esso No. 17 made fast to port. Both the tug and her tow were owned by Esso Standard Oil Company. Each barge carried from 6000 to 7000 barrels of fuel oil and the destination of this flotilla was the north side of Pier No. 3 where a United Fruit Company steamer was to be bunkered with fuel oil from Esso No. 21 at 8 A.M. The tug was 98 feet long, 27 feet abeam and had a 13 foot draft. The two barges, each 176 feet in length, with a beam of 38 feet were lashed to the tug in such fashion that each barge extended about 100 feet forward of the stem of the tug. As the flotilla headed up the North River past the Battery, it reached Pier No. 3 about 6:15 A.M. and found the United Fruit Company steamer had not yet berthed. So the tug stopped about 50 feet from the river end of the pier, heading upriver parallel to the line of piers, and the master decided to cross the river and return after leaving one of the barges at a Pennsylvania Railroad berth on the New Jersey side of the river. Somewhere in the neighborhood of Pier 7 the tug turned hard left at half speed and started across; but very shortly the master observed two steamei’s, one inbound and one outbound obstructing his course and he stopped again and drifted. It is impossible with any accuracy to fix the then position of the tug and tow with relation to the Manhattan piers but they were out in midstream at least several hundred feet. After the two steamers passed there was other traffic, including two Central R.R. of New Jersey ferryboats passing in opposite directions, one bound for New York and the other returning to New Jersey. In addition the President Garfield, a Victory ship owned by American President Lines, Ltd., with which Esso No. 21 later collided, was slowly coming upstream favoring the New York side of the river and she was followed at a distance of a half mile or so by another steamer on the same course.

The effect of the wind and the ebb current soon caused the tug and tow to swing to port and there came a time when the master of the tug ordered half speed ahead and steered hard left in an endeavor to return to the Manhattan side of the river. This speed was maintained until just before the collision, when the master ordered full speed ahead in an attempt to get across the bow of the President Garfield. This maneuver almost but not quite succeeded, as the point of contact was only 25 or 30 feet from the stern of the barge Esso No. 21, lashed to the starboard side of the tug.

What took place in the few minutes immediately prior to the collision seems to us to be overwhelmingly established. We shall presently describe the facts in some detail. We think the collision was caused solely by the faulty navigation of both vessels in the brief interval of time during which the facts we are about to describe occurred.

By way of preliminary, and when the vessels were separated by over half a mile, the Sandy Hook pilot and the master of the President Garfield saw the “broad red light” of the tug and “surmised” she was on her way across the river. Acting on this assumption the President Garfield, without giving any signal of her intention to do so, changed her course slightly to starboard to give the tug and tow clearance. As the findings below are not as clear as they might be, we cannot tell whether the tug and tow were at this time on a steady course or whether they were drifting along and running off their way. In the view we *542take of the case, however, subsequent events alone caused the collision and we shall assume, arguendo, that the master •of the tug was contemplating an attempt to return to the Manhattan shore, and that the tug and tow were drifting and running off their way, with the engines •stopped. Moreover, if, as we hold, the •collision was caused solely by the faulty navigation of both vessels at a later time in the sequence of events, the •change of course by the President Garfield was not a contributing cause of the •collision, and we have no occasion to decide whether or not this change of course •constituted a fault.

It was evidently at about the time the President Garfield steadied back on her ■course that the master of the tug decided to return to the Manhattan side of the river. In any event, as we interpret the findings, Judge Knox concluded as matter of fact that the tug and tow had crossed the bow of the President Garfield from starboard to port. We arrive at the same conclusion, and perceive no basis for rejecting this finding as clearly •erroneous.

It was under these circumstances and with a distance of no more than half a mile between the converging vessels that the master of the tug ordered half speed ahead with his wheel hard left and gave a signal of two blasts. This was the first of two two-blast signals given by the tug, and we shall discuss the other one in connection with what occurred immediately before the collision. There is not the slightest doubt that this first two-blast signal was addressed to the President Garfield and it was given several minutes before the collision. Receiving no reply to his signal the master of the tug continued at half speed ahead and his wheel hard left on a semi-circular course back across the bow of the President Garfield from port to starboard, having already crossed her bow in the opposite direction. Even though the President Garfield was proceeding upstream slowly, at perhaps three or four knots over the ground,1 due to the fact that she was not scheduled to dock until 7 A.M., it was gross negligence and faulty navigation for the slow-moving tug and tow to attempt to re-cross the bow of the President Garfield in order to get back out of the traffic to the Manhattan shore. Moreover, the giving of the first two-blast signal under the circumstances was an additional act of negligence. It could not possibly have been interpreted as an invitation for a starboard to starboard passing, as we shall see; and it was completely worthless as an indication that the tug and tow desired to return to the Manhattan shore. Viewed from any angle it was confusing. We shall later refer to other shortcomings of the tug and tow.

While the fault of the President Garfield is less glaring, we think the trial judge properly held both vessels to blame. The pilot of the President Garfield “observed that the Esso tug was swinging to the left more, and getting dangerously close to us, continuing her left hand swing.” And yet he did not sound the danger signal. The master of the President Garfield testified that when he heard the first two-blast signal from the tug he could not imagine that it was meant for the President Garfield. He looked around and saw no other vessels to which the signal could have been addressed. The ferryboats had already passed. As a matter of fact, the New York bound ferryboat passed between the President Garfield and the tug and tow, so a starboard to starboard passing as between that ferryboat and the tug and tow was just as impossible as was a starboard to starboard passing between the President Garfield and the tug and tow. The suggestion that it might have been meant for the steamer following the President Garfield on the same course will not hold water either. The net result was that the master of the President Garfield had no alternative *543other than to admit that he did not understand the course the tug and tow intended to follow, that he should have sounded the alarm but that he did not. His testimony on this point is as follows:

“Q. Captain, referring back to your Coast Guard testimony from the answer which we have in the record, you stated, ‘Since the tugboat was past our port bow, it would have been impossible for us to answer the situation.’ What did you mean by that statement, ‘it would have been impossible for us to answer the situation’? A. At that time, Captain McNally [the pilot] and I believed it was impossible to give a hard left and have a starboard-to-starboard passing. We would have to go on the Jersey side of the river.

“Q. If that was the case, Captain, and you recognized the Esso tow was indicating an impossible situation, why didn’t you blow the danger signal? A. I don’t know why we didn’t. We failed to blow the danger signal.

“Q. You would agree, wouldn’t you, Captain, that if there is another vessel in close proximity to yours where the risk of collision is obvious from their proximity to each other, if there could have been given a signal as to the impossible situation, as you see it, the danger signal was certainly required ? A. It should have been blown. I agree to that.”

As held by Judge Knox the failure to give the danger signal was a statutory fault and a violation of Inland Rules, Art. 18, Rule III, 33 U.S.C.A. § 203. See National Motorship Corp. v. United States, 2 Cir., 1948, 171 K.2d 413; James McWilliams Blue Line v. Card Towing Line, 2 Cir., 1948, 168 F.2d 720. This being the case the President Garfield had the heavy burden of establishing that this fault could not have contributed to the collision. The Pennsylvania, 1874, 19 Wall. 125, 136, 86 U.S. 125, 136, 22 L.Ed. 148; Afran Transport Co. v. The Bergechief, 2 Cir., 1960, 274 F.2d 469, 473; Sinclair Refining Co. v. The Morania Dolphin, 2 Cir., 1959, 272 F.2d 192.

That the President Garfield fell far short of sustaining this burden is clear from what occurred after the first two-blast signal was given by the tug. Receiving no reply the tug and tow went right ahead hard left at half speed. She then sounded another two-blast signal, as if to say “I am going right ahead across your bow and you had better look out.” This was another fault on the part of the tug and tow. Immediately thereafter the President Garfield gave three blasts, reversed her engines and then put them in emergency reverse, with the result that at the moment of impact she was making sternway. Upon hearing the three blasts the master of the tug ordered full speed ahead, steadied his wheel and made a desperate effort to turn right and get his flotilla past the stem of the President Garfield. We do not find it necessary to decide whether or not this full speed ahead order was an additional fault on the part of the tug. The real significance of the maneuver is. that, had the President Garfield sounded the danger signal when it should have been sounded, that is to say, immediately after hearing the first two blasts from the tug, the master of the tug might have reversed his engines or at least might not have given the full speed ahead order and the collision might have been avoided. In any event, we think it abundantly clear that the President Garfield did not sustain the burden of showing that the failure to sound the danger signal could not on any hypothesis consistent with the testimony and the findings have contributed to the collision.

Affirmed.