This is an appeal from a decree in the admiralty, holding the tug “Cynthia” solely liable for a collision in the Passaic River on the night of December 23d, 1934. The injury was to the barge, “Seaboard No. 333,” owned by the libelant, and in tow of the tug, “Agnes A. Moran,” which collided with a barge in tow of the tug, “Cynthia” ; the owner of the “Seaboard No. 333” libelled both tugs in rem, charging each with fault, but the judge exonerated the “Moran.” The facts are as follows. The “Moran” was bringing the “Seaboard No. 333” downstream and, as she thought the channel too narrow to take her alongside and believed she would not tow well upon a hawser, she was pushing her stern forward, made fast by four cables from the bow bits of the barge to the stern bits of the tug, tightened by turn buckles. At the stern of the barge (which was the forward end of the flotilla), was a cabin fourteen feet wide, the distance between whose after end and the tug’s lights was 175 feet; the barge was light and the top of the cabin was at least seventeen feet above the water; the tug’s lights were only fifteen feet, ten inches. The horizontal distance between the outboard sides of the vertical side pieces of the screens of the tug’s running lights was thirteen feet, one inch, and the lights themselves must have been between nine and ten inches in diameter. The result of this make-up of the tow was that an oncoming vessel, while in front of the cabin of the barge, would not see the tug’s running lights, or at best would only see their diffused glow. All the lights were, however, in order, and the barge had a white light on each of her after corners.
The “Cynthia,” with a barge 100 feet long on her starboard hand, was coming upstream, and passed through the northern draw of the Pennsylvania Railroad bridge, making out the white lights of the “Moran” and her tow as she did; the vessels were_ then about half a mile apart. The master of the “Cynthia” and her engineer, who was on deck, swore that they did not see the “Moran’s” red light at any time, and that for this reason they assumed that she was bound upstream in the same direction as they. The “Cynthia” followed behind on the same side of the channel as the lights; but when the vessels were within about 1,000 feet of each other, she discovered that they were upon an approaching tow, blew an alarm, backed, and by backing headed around to port, thus throwing the barge across the bow of the “Seaboard No. 333.” The “Moran” had herself made out the “Cynthia” and her tow, when they came out from the abutment of the bridge, but as they were then necessarily well over on the north side of the channel, she kept on, expecting to pass without difficulty, until, as she said, the sudden backing of the “Cynthia” threw the barge into her water. The collision happened 600 or 700 feet above *465the bridge and while the “Moran” was on the extreme south line of the channel. The judge held the “Cynthia” for not earlier, making out that the “Moran” was bound downstream, and not seeing her running lights, especially as the channel was winding; he appears to have assumed with the “Moran” that, as she backed, she threw herself and her barge across the channel and into the “Moran’s” water, where she had not been before.
It seems to us that on the contrary the “Cynthia” must have got into the “Moran’s” water at some time before she stopped and backed. We start with the fact that in collision the “Seaboard No. 333,” which was straight in the channel was over as far as she could get to the south. The “Cynthia’s” barge was only 100 feet long and its starboard corner struck the stern of the “Seaboard No. 333” about midships. Since the “Cynthia” was not under a left rudder after she began to manoeuvre for the “Moran” but merely backed, she and her barge would turn upon a point somewhere near the course she had been making, and, as the barge’s bow was only thirty feet in front of her, it could have moved to the left very little — not more than fifteen feet. It follows that in collision it could not have been to the left of the course which the “Cynthia” herself had been following. This is corroborated by the fact that the master and engineer of the “Cynthia” never saw the “Moran’s” red light; which presupposes that she was in the blind sector, though her barge may have been to the north of the “Moran’s” projected course.
The “Cynthia’s” fault follows from this, for she had got over in the “Moran’s” water. She said that it was the “Moran” which got over into hers, but that is impossible unless the collision happened on the north side of the channel, which the “Cynthia” does not now assert, though her engineer so swore. It was very natural that she should have been on the wrong side, if she supposed that the white lights were those of an upbound tow, for the channel is not marked, and its thread is much to the north of the thread of the stream measured from shore to shore. The “Cynthia’s” master, who had no license, may have judged by the shores', in which event he would not have been very far out of position. Finally, the “Cynthia” would hardly have backed when she discovered her mistake, unless the “Moran” was ahead of her. In View of this it is unnecessary to consider those faults with which the judge charged her.
He did not consider the faulty make,up of the “Moran,” or find whether it could have contributed to the collision. There can be no doubt that it was faulty. True, subdivisions (b and c) of article 2 of the statute, section 172, 33 U.S.Code (33 U.S.C.A. § 172), only require that the running lights shall throw their light “from right ahead to two points abaft the beam”; but the clause, “from right ahead,” must be read with subdivision (d), which requires the running lights to be fitted “with inboard screens projecting at least three feet forward from the light, so as to prevent these lights being seen across the bow.” If a ship’s light were merely a mathematical point, it would not be visible at all across the inboard screen, but the line from the center of the light to the end of the screen will necessarily be at something of an angle to the fore and aft line of the ship. In the case at bar for instance the lamps were at least nine inches in diameter, and the line from their center to the end of the screen was at an angle of seven degrees to the fore and aft line of the tug. From this it follows that at a distance of a quarter of a mile there was a sector over 300 feet wide across which both lights were visible. Hence the expression, “to prevent these lights being seen across the bow,” must not be taken literally; and this indeed is obvious from the sailing rules, for rule 1, art. 18 (33 U.S.C.A. § 203) presupposes that there is such a sector where both lights can be seen by each of two opposing vessels. The cabin of the “Seaboard No. 333” almost altogether eliminated this sector for an approaching vessel, which would not see the running lights at all, or at best only a faint glow from part of their outer half. There is every probability that the “Cynthia” was in just that sector. While at that time the inspector’s rules did not forbid such a make-up, they were changed on January 23, 1935, by a rule promulgated June 18, 1935, so as to provide that pushed barges shall themselves carry running lights. This recognizes that it had been dangerous to rely upon the tug’s own lights, as it certainly had; the practice vwas always indefensible. In the case at bar the “Moran” altogether failed to prove that this fault had nothing to do with the collision; on the contrary there is every probability that it did have, that it caused the “Cynthia” *466to leave hér own water, where she was when she came out from the bridge. Whether the “Moran” was also at fault for failing to make her out after she got in front of her, we need not inquire.
Decree modified to hold both vessels at fault