On November 26, 1951, shortly after twelve o'clock noon, the "Rincon Hills," a Canadian tanker, grounded on Dimond Reef, off Governors Island in New York Harbor, and the damage to the ship and cargo is said to exceed one million dollars. Deep Sea Tankers, Limited, as owner of the "Rincon Hills" and bailee of her cargo, and Shell Oil Company, Incorporated, as charterer, within a few days filed a libel against certain tugs and tows of Central Railroad Company of New Jersey and Metropolitan Sand & Gravel Corporation, and against the Railroad Company and Metropolitan, asserting liability for the grounding because of alleged crowding and interference with the navigation of the "Rincon Hills" as she passed along the East River Range for deep draft vessels in her course up the East River. Three proceedings for exoneration from or limitation of liability were then brought by the Railroad Company on behalf of her tug and two carfloats; Metropolitan answered the libel of Deep Sea Tankers, and in the answer included a prayer for relief by way of limitation of liability to the value of Metropolitan's interest in her tug and three sand scows, in the event that they should not be exonerated. There was no separate limitation proceeding initiated by Metropolitan. The above described causes, together with another not involved on this appeal, were consolidated and tried together before Judge Goddard who died on August 26, 1955, after the conclusion of the trial but before the submission of briefs or other argument by counsel. It was stipulated that the case would be submitted to Judge Clancy on the trial record made before Judge Goddard and, upon this record and the briefs of counsel, but without oral argument, Judge Clancy found the "Rincon Hills" solely at fault, exonerated the two tugs and tows and dismissed the libel. This appeal by libelants followed and, on motion of appellees, we have already held that Judge Goddard's trial notes formed no part of the trial record stipulated to be considered by Judge Clancy and hence are not part of the record now before us. 2 Cir., 242 F.2d 433.

Since the opinion and findings below are based upon the cold record alone, without the benefit of seeing and hearing the numerous witnesses who testified before Judge Goddard, and, as this Court has noted, "we have an eventual responsibility for the facts as well as for the law," The Bellhaven, 2 Cir., 72 F.2d 206, 209; Thorne, Neale & Co. v. Reading Co., 2 Cir., 87 F.2d 694, 696; see also The Seeandbee, 6 Cir., 102 F.2d 577, 581, we have carefully read and studied the entire transcript of 6793 pages and the 196 exhibits and have made new findings of fact in accordance with the clear weight of the evidence. Because of the unusual circumstances to which reference has just been made, we shall set forth our findings in considerable detail.

A bird's eye view of the area within which the tanker and the tugs and tows were proceeding to their respective destinations is given by Chart 369 of New York Harbor. The "Rincon Hills" had left her anchorage at Quarantine in the Narrows, bound for 138th Street, East River; her intended course was through the Upper Bay, past Governors Island, along the East River Range for deep draft vessels, past Dimond Reef and thence up the East River. The Railroad steam tug "Long Branch" and the two carfloats were following and gradually overtaking the Metropolitan diesel tug with three fully loaded sand scows as they passed under the Brooklyn Bridge down the East River; the destination of the Railroad flotilla was the yards in Jersey City, New Jersey, in a general westerly direction from the Battery, and that of the sand tow was a meeting place with another Metropolitan tug off the Battery, where one of the three sand scows was to be transferred to the other Metropolitan tug. A seemingly wide expanse of open water between the Battery and Governors Island, where small craft may go at will, is disclosed by Chart 369 to be in fact an exceedingly narrow passage for vessels of deep draft. Dimond Reef, in 24 to 27 feet of water, juts out like a sore thumb and the East River Range, with a steady 40 feet of water, is obviously designed as an indispensable aid to navigation for seagoing vessels as they proceed up the East River. In terms of fairway for such vessels, as they steam up and down stream, there is little more than 325 yards of deep water between the Range and the relatively shoal area in the neighborhood of the piers and ferry slips in the vicinity of the Battery. The Governors Island Ferry, the Ellis Island Ferry and the Staten Island Ferry all operate across the westerly end of the Range or thereabouts and they as well as most small craft are readily maneuverable and can quickly get out of the way of large steamers or tankers. The two tows involved in this case, however, as will shortly appear, were huge, slow-moving, cumbersome flotillas and it requires little argument to demonstrate that they would constitute a substantial menace to the navigation of large vessels in the vicinity of Dimond Reef, especially on an ebb tide, unless kept clear of the Range.

The "Rincon Hills" was a fully loaded turbo-electric tanker of 10,653 gross tons, 523.5 feet in length and 68.2 feet abeam, drawing 29.8 feet of water forward and 29.9 aft.

The diesel tug "Metropolitan No. 2," of 650 horsepower and 144 gross tons, was 80.6 feet long, beam 23 feet, depth 9.6 feet; the hawser connecting her with the sand scows was 12 to 15 fathoms in length; the scows C-G 202, C-G 150 and C-G 104, respectively 112, 113.4 and 113.3 feet long, fully loaded, were made up in tandem and "fast up close to each other" with hawsers and cross lines — a flotilla of about 500 feet overall. The other dimensions of the sand scows were: C-G 202, 320 gross tons, beam 33.1, depth 8.8; C-G 150, 308 gross tons, beam 32 feet, depth 8.9 feet; C-G 104, 327 gross tons, beam 32 feet, depth 9.4.

The steam tug "Long Branch," of 950 horsepower and 229 gross tons, was 104.2 feet long, beam 25.7, depth 14.2. The carfloats Central 56 and Central 61 were of identical dimensions and of steel construction, 291 feet long, beam 40 feet, depth 10.6 feet. Each was a 3-track float, with a capacity of 17 freight cars. There were two empty coal cars and one loaded freight car on one of the floats, the other was light. The floats were on either side of the "Long Branch," all three tightly lashed together, with the bows of the floats brought nearly together ahead of the stem of the "Long Branch," which forced out the sterns of the floats fanwise. There are certain minor discrepancies and contradictions in the record respecting some of the figures, but they are substantially accurate. The sand scows and the carfloats were passive instruments without propulsion equipment of any kind.

At this point we think it will be helpful to separate the facts we deem beyond all reasonable doubt established by the record, so that the more controversial features of the case may be discussed against what we consider to be an unassailable background.

(1) The "Rincon Hills" sounded a single whistle blast, followed by the danger signal of four or five short blasts; and there was no response to either of these signals by either of the two tugs.

(2) As the two tows proceeded downstream the Railroad carfloats gradually came closer and closer to the sand tow, intending to pass to port of the sand tow; and the carfloats did pass to port of the sand tow, the actual passing taking place after the carfloats went under the bow of the "Rincon Hills."

(3) The stem of the "Rincon Hills" barely missed collision with the carfloats as they crossed her bow.

(4) It was clear weather, with excellent visibility and a fresh westerly wind; the current was running on the ebb a full knot and a half in the direction of Dimond Reef, a fact which the masters of the two tugs did not take into consideration, as it was just about dead low water at the Battery and the two masters of the tugs, who knew it was about low tide, were not aware of the fact that the current continued to run on the ebb well over an hour after low water at the Battery.

(5) Just as or immediately after the carfloats crossed the tanker's bow the pilot gave the order "hard left, full ahead," which was repeated by the second officer and by the helmsman; but the helmsman, because of a "reaction" or intuitive fear of collision, disobeyed the hard left order and put the wheel hard right.

(6) The pilot in charge of the navigation of the "Rincon Hills" was a man of ripe experience, thoroughly familiar with New York Harbor and the set of the current on and near the Range. The masters of the Railroad and Metropolitan tugs were old hands at their trade and should have known what they were about.

There were many witnesses, several on board the tanker. The sounding of the danger signal alerted some of these witnesses to the fact that something unusual was going on Some of them were on the tows or on ferryboats in the vicinity, one was in his office on the top floor of the Whitehall Ferry Building on shore. Obviously, some of these witnesses were interested; they varied as to intelligence, accuracy of observation and ability to understand the questions, some of which were highly technical. There was an almost universal and quite bewildering disposition by practically all eye-witnesses to guess at distances and time intervals in terms of seconds or minutes. The documentary evidence is, as always, important and we have been particularly impressed by the tanker's Bridge Bell Book. We reject the charge by appellees that the claim of crowding was manufactured out of whole cloth as an afterthought, and that the entries in the Bridge Bell Book were deceptively written up after the event to support the allegedly fictitious claim of crowding and to cover up the "wrong wheel movement of the helmsman." On the other hand, we take such colorful expressions in the testimony as "this carfloat came shooting across the bow of the ship," and "the ship took a nosedive," to be no more than rhetorical flourishes of little consequence or significance.

Except for those charged with the responsibility of navigating the tanker and the two tows, who were under a duty to observe other traffic in the East River, there was nothing to attract the attention of anyone to the scene off Dimond Reef until the danger signal was given. Thus the danger signal alerted some, including the master of the tanker, who was at that time in the chart room abaft the bridge, and the purser, who was in his stateroom counting money. The close proximity of the tanker and the carfloats when a collision seemed unavoidable drew the attention of others to what occurred immediately prior to the grounding. Doubtless the attention of many others was attracted after the vessel grounded and was motionless. We have necessarily viewed with some skepticism elaborate descriptions, by witnesses not then aboard the tanker or the tows, of the relative positions of the vessels at various times prior to the grounding as honest but at least to some extent unreliable. Viewed from another angle, it seems to us perfectly natural for a person whose attention is thus suddenly focused on an event or occurrence to see the part of the scene to which his attention is attracted, without taking in the whole picture. For example the master of the tanker ran out from the chart room upon hearing the four or five short blasts. He saw the carfloats which the tanker almost ran down; but did not notice the sand tow. We do not infer from this that the sand tow "was never in the picture," as found by the court below.

In the discussion which follows we shall endeavor to find the correct answers to the following controversial questions: (1) what were the relative positions of the tanker on the one hand and the two tows on the other when the one-blast signal was given; (2) how close was the sand tow to the carfloats when the carfloats crossed the bow of the tanker; (3) what were the relative speeds of the tanker and the tows just before the "hard left, full ahead" order was given by the pilot; (4) what was the function of the two Meseck tugs, one to port and the other to starboard of the tanker as she steadied on the Range; and did any act or omission of the pilot or those aboard these tugs contribute to the grounding; (5) had the helmsman promptly obeyed the hard left order of the pilot would the tanker nevertheless have grounded; (6) did the pilot change his course to the right after giving the danger signal and before he ordered "hard left, full ahead."

It is clear to us that the tugs and tows seriously obstructed the navigation of the tanker and that the negligent operation of both the tug "Metropolitan No. 2" and of the Railroad tug "Long Branch" contributed to the grounding of the tanker on Dimond Reef. We so find. But no fault is found against any of the sand scows or the carfloats nor against the tug "Metropolitan No. 3."

What made our task unduly difficult at the outset was not only the fact that the scene as portrayed in the opinion below is painted with a large brush, but also the confusion caused by a parade of witnesses, who had little or nothing to contribute to the case, and by a welter of inconsistencies and contradictions caused by the marking of charts, and the measuring of distances with dividers, and so on, by other witnesses who at times did not have the remotest idea of what they were doing. But a careful digesting of the entire record lends perspective; and the seeming fog and confusion are readily dissipated. We shall work our way back from the grounding.

While it is probable that the tanker bumped or scraped along the rocky surface of the reef before she fetched up, there is no doubt that she held fast at some part of the northerly tip of the reef, parallel with the Range and hence well south of the Range. We also find that the tanker was south of the Range when the "hard left full ahead" order was given. We shall discuss this point in some detail later on. As we have already indicated, the record in its entirety literally compels a finding that the stem of the tanker missed the stern of the port carfloat by the narrowest of margins. As this finding is of such critical importance, we shall briefly review the testimony and some of the exhibits marked by some of those who testified.

One of the most impressive witnesses was Robert J. Wilcox, the helmsman on the tanker. This young man frankly admitted that he gave the wheel two or three turns to the right although the order "hard left" had been repeated both by second officer Richard and by himself. He said he did this instinctively and without thinking, because he saw the carfloats practically dead ahead and he thought he would ram them. The fact is that the stem of the tanker had already cleared the stern of the port carfloat, and there is at least a possibility that the stranding would have been avoided had Wilcox done as he was told. But we credit his testimony that the carfloats were so close that he thought he would ram them if he obeyed the "hard left" order. We shall return to Wilcox later when discussing the question of whether or not we should find that the tanker was also to blame.

Ernest Kuinlivan, the boatswain, was up in the forecastlehead to stand by the anchors in case of emergency, and perhaps to make fast a line from one or both of the assisting tugs. He could not have been in a better position to see the distance between the stem of the tanker and the stern of the carfloats. He testified that the clearance was about 30 feet. Philip, the purser of the tanker, ran out of his cabin when he heard the danger signal, saw the carfloats off the tanker's starboard bow, thought there was going to be a collision and went back to his cabin when he saw the tanker was going to clear the carfloats.

There were two assisting tugs. The "William Meseck" had put Rowohlt the pilot aboard the tanker at Quarantine. McConnell, the mate of this tug, in charge in the absence of Rowohlt, had first gone to port of the tanker but was directed by the pilot to move over to Starboard, which he did. He saw the sand tow cross the bow of the tanker as the danger signal was blown and testified that the carfloats were "right around the stern" of the sand tow. Edward R. Hoffman, at the wheel of the tug "Hayward Meseck," on the port side of the tanker, testified that the carfloats cleared by 10 feet. Mullock, the master of the tanker, fixed the distance as 25 or 30 feet. Richard, the second officer, testified it was 20 or 25 feet. Rowohlt, the pilot, said he just cleared by about 20 feet. It is claimed that this is contradicted by his testimony at the Coast Guard hearing, a few days after the grounding, but we find no contradiction in view of the diagram (Exhibit 9) prepared by Rowohlt at the Coast Guard hearing. He may have misplaced the tows by not properly portraying the crossing but the proximity of the vessel and the tows is unmistakable.

Henry G. D. Hoffman operating the Governors Island Ferry "Lt. Col. Robert E. Shannon," which left the Battery at 12:15 P.M. and arrived at Governors Island at 12:21 P.M., testified that the two tows and the tanker "seemed to all blend together" and that the clearance at the crossing of the tanker's bow was 10 feet — "the floats just missed getting hit." Roland H. Whelan, quartermaster of the ferry "Ellis Island," which left Ellis Island at 12:15 P.M. and tied up in her slip at the Battery at 12:29 or 12:30 P.M., thought there would surely be a collision. He testified, it "looked very close to me, maybe 10, 20 feet, and I couldn't see how an accident could have been avoided." William G. Tuttle, at the wheel of the "Gold Star Mother," the Staten Island Ferry arriving at her slip at the Battery 12:30 or 12:31 P.M., gives the clearance as 20 to 30 feet, "they were all in very close quarters," with the carfloats about 100 feet behind the sand tow. We find no trace of animus in the testimony of these witnesses.

It is perhaps understandable that Hutton and Molloy operating the Metropolitan No. 2 tug, with the sand scows, and the "Long Branch," with her carfloats, respectively, should place their tugs and tows at so great a distance from the tanker as completely to negative any interference with the navigation of the tanker. Hutton was completely discredited; and Molloy's testimony at the Coast Guard hearing that his tow cleared the tanker by only 70 feet would seem an insurmountable obstacle to the acceptance of his trial estimate of 200 or 175 or 150 feet in view of the overwhelming testimony to the contrary.

Albert H. Reichert, at the wheel of the tug "Metropolitan No. 3," following some little distance behind the two tows with orders to take off one of "Metropolitan No. 2's" sand scows somewhere in the neighborhood of the Battery, adds little to appellees' version of the stranding. His testimony, when considered in the light of his statement (Exhibit YY), leaves us with the impression that he saw little if anything of the events preceding the actual grounding of the tanker. In any event, he says that the sand tow and the carfloats "were up and out in the North River when the ship was coming up on the range," that the tanker grounded when he was abreast of her and that the two tows "were in the North River at the time." It requires little argument to demonstrate the unreliability of this testimony. Rodden of the Staten Island Ferry "Cornelius Kolf," in the offshore pilothouse in No. 5 Slip South Ferry, is another broken reed. His testimony is a mass of confusion, as is indicated by his insistence that "I saw the Meseck tug blowing the alarm," and that he actually saw the steam from her smokestack as she blew the alarm. But it is clear beyond cavil that the danger signal was blown by the tanker and not by the Meseck tug. Bielinski, the first deckhand or mate of the "Long Branch" gives the nearest point of the tanker to the carfloats at the time of passing as 400 to 450 feet. There is no trace of partisanship in his testimony, and no outbursts or profanity as in the case of Hutton, but he is clearly mistaken on many points, and we think it may have been unwise to have him present during the entire trial of four months, except for a week or so when he was ill, listening to the testimony of the other witnesses. He was called near the very end of the trial. Bernard Sheridan was sitting on a settee behind Tuttle on the "Gold Star Mother." If it was thought that he would contradict Tuttle's testimony in some vital particulars, his testimony must have been disappointing. He added precisely nothing to the case.

The owners of the tugs and tows lean far too heavily upon the witness William M. McGuire, who described what he said he saw after hearing the alarm signal as he sat at his desk in his office on the top floor of the Whitehall Ferry Building. If his attention was first attracted by the four or five short blasts, as he said, he could not possibly have seen the tanker as she steadied on the Range and he locates the two tows so far away from the tanker as to make his testimony of little value. (See dots placed by McGuire on Exhibit 78.) He characterizes the grounding as "the strangest thing I ever saw," and all the probabilities of the situation militate strongly against a version that describes the tanker, in charge of an experienced pilot, slowing down, then starting up, taking a sudden "sheer" to starboard and fetching up on Dimond Reef, all on a clear day, without any other vessel anywhere near her, and after blowing a meaningless danger signal. No explanation worthy of the name is given for his failure to mention to the representative of appellants' counsel, on the various occasions of their talks together, that he claimed to have seen what he later testified to; nor is it clear to us why he read parts of the testimony given at the Coast Guard hearing and over 1000 pages of the trial transcript in preparation for his appearance on the witness stand.

Thus we have two long, unwieldy tows, one immediately behind the other and in a position to pass to port of the first tow, the intervening distance between the two being 100 feet or so, crossing the bow of the tanker on a westerly or northwesterly course, at a point where the tanker is already south of the Range. How did the tows get there?

Without discussing the testimony in detail, it will suffice to say that there is abundant evidence to support a finding that, as the tanker came up the bay and past Governors Island she swung on to the Range at about which time the pilot gave the one-blast signal. There was no traffic in sight up the East River except the two tugs and tows. We find the location of the tanker and the two tows was such as to make this signal for a port-to-port passage quite in order, and that he was, as he testified, signalling to the two tows. It was, at the time the single blast signal was given, a meeting head and head. We cannot agree that, as found below, this signal blast was not addressed "to any particular vessel but was only to warn all traffic in the East River of the Rincon Hills' approach." No such signal is provided for in the Navigation Rules for Harbors, Rivers, and Inland Waters, 33 U.S.C.A. § 203, or in the Pilot Rules for Inland Waters, 33 C.F.R. §§ 80.01-80.36; nor was the one-blast signal here given with any such intent. Unnecessary signals are forbidden, Section 80.35; the proper response if the tugs contemplated a port-to-port passing was a single blast, Section 80.03; if there was some misunderstanding on the part of the tugs, or if they had appreciated the difficulties which necessarily would arise if they turned and exposed the entire string of sand tows and carfloats broadside to the ebb tide, they should have sounded four or more short blasts, the danger signal, Section 80.1. See also La Boyteaux, The Rules of the Road at Sea (1920) 181, 185.

But I must never fail to remember that in giving forth these "blasts," I must only convey to my opponents the course I am about to follow in accordance with the Steering and Sailing Rules. It must always be present to my mind that if I depart from the Rules, and make signals by "blasts," which may indicate a course or manoeuvre not authorized by the Rules, I and my owners and my ship will probably be condemned in costs * * *. Gray, Rule of the Road (4th ed. 1884) 32.

Moreover, the tugs and tows and the tanker were clearly in such relative positions that passing signals were in order; but the tugs made no response to the signal from the tanker, and, to make matters worse, they pursued, even after the danger signal, a course which made disaster almost inevitable.

The pilot waited about 30 seconds for a response from the tows, and then, as neither tug made any response whatever, he blew four or five short blasts — the danger signal. What he meant by this is too obvious for comment. He thought a dangerous situation was developing and he was quite right. But neither of the tugs made any response whatever; and there is undisputed proof in the very testimony of those operating the tugs that, with a reserve power always available for emergencies, no direction was given to the engineer of either craft to increase the speed of his engines. Moreover, there was a strong westerly wind of about 23 miles an hour that could not have failed to slow up the progress of the tows to an appreciable extent.

We hold that passing and danger signals are to be taken to mean precisely what is stated in the Pilot Rules, supra. They have a clear and settled meaning as above stated, and the master who fails to make the appropriate and required response is guilty of a statutory fault and must take the consequences. It may well be that passing signals are frequently unnecessary, but such appears not to be the case here. The disregard of the danger signal is particularly blameworthy.

We are asked by appellants to rule that the approximately 325 yards of deep water from the 23-26-25-21 foot line off the lower Manhattan ferry slips and piers at the Battery to the westerly side of the Range off Governors Island is a "Narrow Channel," within the meaning of the Rule of Navigation, 33 U.S. C.A. § 210, which requires vessels and craft when proceeding in such channel to "keep to that side of the fairway or mid-channel which lies on the starboard side of such vessel." Appellees insist that it is well settled law in this Circuit that this body of water is not a Narrow Channel, citing The Wrestler, 2 Cir., 232 F. 448; The Bern, 2 Cir., 74 F.2d 235; City of New York v. American Export Lines, Inc., 2 Cir., 131 F.2d 902. See also Griffin, The American Law of Collision (1949), § 36 n. 14 at p. 94. We are inclined to agree with this, especially in view of the extent and conformation of the harbor area involved and the cross or east and westbound traffic, although these cases do not deal with the precise body of water here involved. But we need not give that point the consideration here that it perhaps deserves, as we hold the tows at fault for being where they had no business to be at a time when a deep draft vessel was steaming up the Range. Even if not a statutory fault this is plain, ordinary negligence. See The Wrestler, supra; The Concho, S.D.N.Y., 58 F. 811, affirmed 2 Cir., 63 F. 1023. We hold that when large steamers or tankers or other deep draft vessels are on the Range or steadying on a course past Dimond Reef up the East River, tugs and tows and other unwieldy craft must keep away from the waters near the Range. On an ebb tide the slightest interference with the navigation of deep draft vessels is apt to end in disaster, as there is practically no room in which to maneuver vessels several hundred feet in length and the current is bearing in a southwesterly direction right over the face of the reef. The danger to the entire waterfront as well as to other ships in the vicinity is also manifest, as oil or gasoline bubbling up to the surface from fractures of the cargo tanks caused by grounding on Dimond Reef or by collision, may at once or later be ignited.

Thus, even in the absence of any lack of response to signals, and irrespective of how they got there, the tows were at fault for not keeping away from the immediate vicinity of the Range, where they interfered with the navigation of the tanker.

The record demonstrates in the clearest possible manner, however, just how the tows found themselves on the Range. It was actually dead low water at the Battery at 12:18 P. M., eight minutes later than the predicted time. The current continued to run out over Dimond Reef on the ebb for about 1½ hours after low water at the Battery. We accept the data prepared by the Currents and Oceanography Branch of the Division of Tides and Currents of the U. S. Coast & Geodetic Survey (Exhibit 100) and reject conflicting data prepared by others. According to the testimony of Don C. Piper, Assistant Chief, the "current comes out of lower East River and is carried in a southwesterly direction, crosses Dimond Reef and on to Governors Island where it is turned to the westward." This witness gives the velocity of the current over Dimond Reef at the stage of the tide already indicated as one and a half knots, and the velocity was greater further up the river as indicated in Exhibit 100.

It so happened that both Hutton and Molloy, in charge of the two tugs, miscalculated the current cycle. Each was under the misapprehension that at low water at the Battery the current was slack or practically so. It is hard to believe that two such old hands at towing in New York Harbor would make such an error but their testimony to that effect is unequivocal and convincing. Each insists he made no allowance for the current, which is another way of saying that he operated his tug without taking into account that he had underneath him a current carrying him down in the direction of the Range and Diamond Reef at the rate of 1½ knots an hour.

As the incidents already described all took place in the few minutes preceding 12:30 P. M. on the day in question, it is highly probable that both Hutton and Molloy thought they barely had time to make the Battery with the very end of the ebb current, a matter of vital consequence to them. This would explain their remaining in midstream longer than usual, as there was less current near the piers. In any event, in a statement made on December 13, 1951, very shortly after the grounding, by Hutton to the attorney retained to represent Metropolitan, and reduced to writing and signed, Hutton says that after passing below the Brooklyn Bridge "the tow was favoring the middle of the river." He says that he saw the tanker "when the tow was about off Pier 9," and that "when approaching or in the vicinity of Pier 5 we began to pull in toward South Ferry so as to be in a position to round the Battery." Every effort to get away from this at the trial was unavailing. Finally, Hutton confirmed the statement, saying "the statement is true," and "it must be right, I signed it." Under these circumstances, and in view of the position of the vessels when they barely missed collision, we have no other alternative than to hold that the finding below to the effect that Hutton began to pull in when off Pier 8 rather than when off Pier 5 is clearly erroneous.

But, if Hutton's sand tow was favoring the middle of the river until he reached Pier 5, this fixes the position of Molloy and the carfloats following close behind but slightly to port of the sand tow. This is something Bielinski, standing in the pilothouse of the "Long Branch" alongside of Molloy as a sort of observer or lookout, would be likely to notice. And Bielinski's testimony on the subject is explicit. He testified that at Pier 8 the sand tow was ahead "of the starboard carfloat," "a little to our right," with the stern of the last sand scow just ahead "of the center of the starboard carfloat." Accordingly, the carfloats were further out in the river than the sand tow, and, as Hutton waited until he reached Pier 5 to turn in, the carfloats were still to his port. And this relative position of the two tows continued until after they passed the tanker, although they came closer together with each passing moment. Bielinski said the two tows were 200 or 250 feet apart when off Piers 8 and 9, that the distance was slowly closing; and we have already found that they were about 100 feet apart as they crossed the tanker's bow.

On the basis of these facts what happened to the two tows in the few minutes following Hutton's decision to pull in toward South Ferry at Pier 5 is merely an operation of the laws of physics. As the two tugs turned to the West the tows caught the force of the ebb current broadside and they swung out into the river, each now headed West or substantially so, but they were making about as much progress down toward the Range as they were making ahead. Thus strung out the two tows covered a distance of almost 900 feet, including the 100 feet between the two tows, through which 100 feet the tanker could not have passed, and they completely blocked off the tanker. The probability is that Hutton and Molloy were looking toward their respective destinations, completely oblivious of the danger to the tanker, until they suddenly saw her close aboard.

In view of the foregoing we do not find it necessary to pass upon the question presented by the medical proof relative to Molloy's competence to manage the "Long Branch." We shall return to this subject and discuss it briefly, together with the matter of Charts and Tide and Current Tables, when we come to the subject of limitation.

We find it more difficult to determine whether or not there was fault on the part of the tanker. This involves a consideration of her speed, the various directions given by the pilot after the one-blast signal, whether the assisting tugs should have been ordered to make fast before they started to come alongside and found it impossible to do so, and the circumstances attending the disobedience by the helmsman of the "hard left" order. We have concluded that the tugs "Metropolitan No. 2" and "Long Branch" were solely to blame.

We accept the testimony of Rowohlt, the pilot. There was considerable credible corroborative proof, but we find it not necessary to describe it in detail. Most of the alleged discrepancies between his extensive testimony at the trial and his brief interrogation at the Coast Guard hearing disappear on close analysis. We do not read his testimony at the Coast Guard hearing as stating that there was no interference with his navigation by the two tows. On the contrary, he answered "yes" to the following question, "You said that if the tugs and tows had not been where they were, you would have taken the `Rincon Hills' on a turn into the East River at about a point where the ship would have been opposite South Ferry, is that correct?" There is more to the same effect. When he testified that his "hard left" order was not to avoid a grounding or to avoid the tows, we interpret this as meaning that he thought he was a safe distance from the reef and that the tows had already crossed his bow. When he gave it, he believed the "hard left, full ahead" order would take him safely up the East River. But, by way of hindsight, he testified at the trial that, even if the "hard left" order had been executed the tanker would have grounded nevertheless. While at the Coast Guard hearing he does not specifically say that he gave the right wheel order which brought the tanker south of the Range, this is because no questions were asked about his orders to the helmsman after the tanker had steadied on the Range; and he did testify, "We were running parallel with the range and the ship was about half the distance south of the range." He explicitly stated at the Coast Guard hearing that when he gave the one-blast signal the position of the vessel and the tows was one of meeting head and head. Moreover, he testified that both the single blast and the danger signal were addressed to and intended for the tows, and this entire testimony at the Coast Guard hearing must be read in the light of the sketch he then drew, Exhibit 9 on this trial, to which reference has already been made.

The substance of his recital at the trial of the tanker's maneuvers prior to the "hard left, full ahead" order was: after the single blast he steadied on the Range in a matter of seconds at half speed; that 30 seconds after the single blast he sounded the alarm, ordered the engines stopped and gave the right wheel order which was executed, bringing the tanker slightly south of the Range; that the two tows were a good ship's length and a half away when he sounded the alarm; that two minutes elapsed between the sounding of the alarm and the grounding. He said he could not bear to port because of the presence of the sand tow and that he could not reverse his engines as this would turn the tanker to starboard and on the reef. The Bridge Bell Book, which we consider the most important piece of documentary evidence in the case, gives the "full ahead" order at 12:26½ P. M. and at 12:27 P. M. contains the entry "stop struck bottom (Diamond Shoal)." We say this despite the discrepancies between these entries and those made by the oiler in the Engine Bell Book. Moreover, the little curved arrow leading to the words "(Diamond Shoal)" on the preceding line seems to us to be a natural way of making the record in this small note-book. In view of the other entries in the book there seems to us to be no reason to doubt the explanation that the use of the arrow and the previous line was thought preferable to placing "(Diamond Shoal)" on the line underneath. We find nothing suspicious about this entry or the entries in the rough log of the vessel. Accordingly, we find, on the basis of this and other proof, not only that the refusal to find that the right wheel order was given and executed was clearly erroneous, but in addition that the tanker was continuing to swing slightly to starboard when the "hard left" order was given, that she was still on a right swing when she hit the reef, and that the interval of time between the turning of the wheel to the right rather than to the left as ordered, and the correction of this mistake by the helmsman, was no more than 15 or 20 seconds at most. In response to the order to correct his mistake he was turning the wheel to the left as she struck. As we are favorably impressed by the competency of the experts who testified on the tanker's behalf and the accuracy of their opinions on the subject of the lateral movement of the vessel, we are convinced that in all probability the tanker could not have cleared the reef if the "hard left" order had been properly and promptly executed. See Knight, Modern Seamanship (11th ed. 1948) 294. But we cannot be certain of this and hesitate to make a finding to that effect because there is no way of knowing with any degree of accuracy what the precise position of the tanker was with relation to the reef when the order was given.

The speed of the tanker was approximately 7½ knots, even after the engines were stopped. Under normal conditions this would not be excessive. We have found it impossible to estimate the speed of the tugs and tows. Perhaps the sand tow was making 3 knots or so and the carfloats more. We estimate all these speeds through the water rather than over the ground. The question then is: from the moment it was apparent that a dangerous situation was developing ahead, did the pilot take every reasonable and available means of meeting it. When the single blast signal was given the intended course of the two tows could not have been known by Rowohlt. It is still difficult to understand Molloy's persistence in following close behind the sand scows and to port of the sand scows. With the power at his disposal, and the fairly light load as compared with a full load of 17 cars, he could easily have passed the sand tow to starboard and got out of the tanker's way. It is not unlikely that his mental operations were affected to some extent by his chronic hypertension and by the sedatives he was taking on his doctor's advice. For all Rowohlt knew the carfloats were headed up Buttermilk Channel. In any event, it was not until he got no response from either tug to his single blast signal that Rowohlt knew or could have known that a dangerous situation was developing. We can only conclude that the failure to respond to the single blast signal and the manner in which the two tows proceeded thereafter was due to negligence and inattention on the part of Hutton and Molloy, and their testimony gives ample support to this conclusion.

Accordingly, we must test the question of fault on the part of the tanker by ascertaining what, if anything, Rowohlt could have done to avoid running down the carfloats or coming perilously close to the reef after he sounded the danger signal. He did stop his engines. Could he have safely reversed his engines? We have concluded, on the basis of the expert opinion testified to, and other supporting proof, that this maneuver would have brought the tanker on the reef. In this connection it must be borne in mind that a current of one and one-half knots was flowing in the direction of and over Dimond Reef. There was little or no possibility that reversing the engines could have had any effect other than to increase the danger. How about the assisting tugs? Rowohlt had already ordered one of them from the port to the starboard side of the tanker. Surely there was no occasion to have more than two assisting tugs and it seems quite reasonable to place one of them to port and the other to starboard. The dangerous situation, due in no small part to the failure of the two tows to respond to the single blast signal, developed only when it became apparent that the two tows were oblivious to or unmindful of the way they were obstructing the progress of the tanker on the Range; and this, as it later turned out, was due to the miscalculation by both Hutton and Molloy of the force, velocity and direction of the current. At the time the single blast signal was given McConnell was in the very act of bringing the "William Meseck" in close to make fast to the tanker on or near her starboard bow. But this became impossible as Rowohlt gave his right wheel order, and the swing to starboard which brought the tanker south of the Range forced McConnell to drop back. We may add, parenthetically, that the only fair inference to be drawn from McConnell's testimony, taken in its entirety, is that the right swing he was describing was the result of the right wheel order given after the sounding of the danger signal, and not to any action of the tanker due to the mistake of the helmsman later on. Edward R. Hoffman testified that he was in charge of the tug "Hayward Meseck" to port of the tanker, and that he was prevented from making fast to the tanker's port bow by the proximity of the sand tow. The sand scows came so near his tug that he dropped in close alongside the port side of the tanker, and he was forced to stay there as the carfloats immediately followed the sand scows. This brief review of the testimony we think disposes of the claim that proper use was not made of the assisting tugs.

Accordingly, this phase of the case comes down to the act of the helmsman in disobeying the "hard left" order of the pilot. Ordinarily, such disobedience, if it contributed to a collision or grounding, would constitute a fault chargeable to the vessel. It is the most serious charge made against the tanker in this case, and the one that has caused us the most difficulty. Taking all the attendant circumstances into consideration, however, we have concluded that an in extremis situation existed and the natural and instinctive reaction of the helmsman, faced as he thought with the certainty of ramming the carfloats, is excusable. It will be recalled that the "hard left" order was repeated both by the second officer Richard and by the helmsman himself. We find the tanker free from fault.

The Railroad, within six months after notice of claim, filed petitions for exoneration from or limitation of liability as owner of the "Long Branch" and of each of the two carfloats. All procedural requirements were concededly complied with; and we are presented with several interesting questions of law, none of which was passed upon by the court below, as the tanker was found to be solely at fault and the tug and the carfloats were exonerated.

Appellants, without citing any authority, claim that the "Long Branch" and the carfloats were not engaged in any maritime undertaking but were acting merely as a ferry in an exclusively railroad operation, and they would have us hold that the legislation embodied in 46 U.S.C.A. §§ 183 and 185 was not intended by the Congress to be applicable to railroad lighterage in New York Harbor, for the reason, among others, that the inclusion of vessels engaged in railroad car lighterage would not contribute in any way to putting American shipping upon an equality with that of other maritime nations. And we are referred to clauses used in Uniform Bills of Lading as in some way supporting this view. But we think it too clear to require extended discussion that the tug and tow were vessels within the meaning of the words "any vessel" in the Limitation of Liability Act. The operation here involved was maritime; those aboard each of these vessels doubtless came under the protection of the Longshoremen's and Harbor Workers' Compensation Act, 33 U.S.C.A. § 901 et seq.; and the fact that the carfloats were designed and built for the specific purpose of transferring railroad freight cars from one railroad terminal to another in and about New York Harbor does not make the operation any the less maritime in character.

Appellants also contend that neither the Railroad nor Metropolitan is entitled to limit its liability because both the tug "Long Branch" and the tug "Metropolitan No. 2" were unseaworthy with the knowledge and privity of their owners. With respect to both tugs it is claimed by appellants that no copies of the Charts, Tide Tables and Current Tables issued by the United States Coast and Geodetic Survey were provided to the tugs to be kept on board always available for use. But both Hutton and Molloy were obviously familiar with the waters covered by the Charts of New York Harbor. As to the Railroad the evidence is that Charts as well as Current Tables were available and could have been obtained on demand at the Dispatcher's Office at any time by Molloy or other tug captains in the employ of the Railroad, if they were to operate in any part of the Harbor concerning which they felt further information should be had or if they had any other special need of them. The evidence concerning Metropolitan is less clear but it is probable that Charts were available or actually on "Metropolitan No. 2" but the Current Tables were not. The "Long Branch" and the "Metropolitan No. 2" each had on board a Commercial Tide Calendar giving high and low water at Sandy Hook, the Battery and Hell Gate. Taking the evidence as a whole we think both the Railroad and Metropolitan acted reasonably in committing the navigation of their tugs to these two men of many years experience towing in the very waters involved in this case without making further provision for either Charts or Current Tables. Many of the witnesses testified that it was not customary to have them on board harbor tugs. We are still at a loss to understand how Hutton and Molloy miscalculated the current cycle; but the mere fact that they did so does not convince us that the failure to make further provision for Charts and Current Tables aboard these tugs was a wilful fault by the owners such as to deprive them of the benefits of the Limitation of Liability Act.

As to Molloy appellants most earnestly press the claim that there was wilful fault on the part of the Railroad in permitting him to continue as Master of the "Long Branch," despite knowledge of his chronic high blood pressure. We think, however, that it is a sufficient answer to say that a competent medical examiner, acting on behalf of the Railroad, passed him subject to reexamination as fit for duty. It may well be that there was inexcusable delay in producing some of the records and others were never produced; but what we consider to be the whole story came out in the end and we find nothing to impugn the good faith of the medical examiner.

Appellants also raise an interesting and important question not previously passed upon by us, so far as we are aware, and on which there are conflicting decisions in the District Courts for the Eastern and Southern Districts of New York and elsewhere.1 It will be recalled that Metropolitan filed no separate or other petition for limitation of liability but, long after the six months period after notice of claim as provided in 46 U.S.C.A. § 185, Metropolitan alleged the Limitation Act, 46 U.S.C.A. § 183(a), as a partial defense to the libel. Appellants assert that there can be no limitation of liability unless a proper claim therefor is made, either by the filing of a petition for limitation or by way of defense to a libel or other court proceeding within six months after notice of claim. In other words, appellants would have us hold that every claim of limitation of liability asserted under 46 U.S.C.A. § 183(a) is subject to the six months requirement of 46 U.S.C.A. § 185.2

The law is well settled that a petition to limit liability pursuant to 46 U.S.C.A. § 185 is valid and effective only if it is filed within six months after the shipowner receives written notice of claim. The Fred Smartley, Jr., 4 Cir., 108 F.2d 603; Standard Wholesale Phosphate & Acid Works, Inc. v. Travelers Ins. Co., 4 Cir., 107 F.2d 373; The Irving, D.C.S.D.N.Y., 28 F. Supp. 585, affirmed Conners Marine Co. v. U. S., 2 Cir., 107 F.2d 1011; The Grasselli Chemical Co. No. 4, D.C.S.D.N.Y., 20 F. Supp. 394.

There is, however, a separate and distinct procedure by which a shipowner may limit his liability in accordance with 46 U.S.C.A. § 183(a). A shipowner may institute a separate proceeding by the filing of a petition for limitation or he may plead the limitation statute, 46 U.S.C.A. § 183(a), in his answer as a defense to suit against him. See 3 Benedict on Admiralty (6th ed. 1940) Section 508.

The Third Circuit, in a well-reasoned opinion by Judge Goodrich, has held that the six-month period does not govern a plea of limitation by way of answer. The Chickie, 3 Cir., 141 F.2d 80, 84-85. In reaching its decision in The Chickie the Court reviewed the historical development of the two different procedures available to shipowners to limit their liability. We are in accord with the analysis developed in Judge Goodrich's opinion and we hold that a plea of limitation by way of answer is not subject to the six-month time limit in Section 185. It is to be noted, however, that all those interested, including the "Rincon Hills" as bailee of the cargo, are parties to the consolidated proceeding now before us in which the respective claims of limitation are asserted.

The first statute authorizing the limitation of liability, the forerunner of Section 183, was enacted in 1851, 9 Stat. 635. In 1872 the Supreme Court promulgated admiralty rules of procedure which permitted a shipowner to petition for limitation of liability. Shortly thereafter the Congress enacted the substance of these rules into law. In commenting on the purpose of the rules, the Supreme Court said they "were adopted for the purpose of formulating a proceeding that would give full protection to the shipowners in * * * a case" where there were "various parties claiming damages against them for injuries sustained by mishaps to the ship or cargo." "They were not intended to prevent them from availing themselves of any other remedy of process which the law itself might entitle them to adopt. They were not intended to prevent a defence by way of answer to a libel, or plea to an action, if the ship-owners should deem such a mode of pleading adequate to their protection." The Scotland, 105 U.S. 24, 33-34, 26 L. Ed. 1001. It is unnecessary to comment on the variety of situations that may arise, in some of which concursus is highly desirable if not indispensable and in others not.

In practice it developed that the proceeding by way of petition for limitation of liability was often a source of delay and circuity of action, because shipowners did not institute this independent proceeding until after the question of liability had been litigated and determined against them. To alleviate the abuses inherent in this situation, the Congress, on June 5, 1936, amended Section 185 to require that a petition for limitation must be filed within six months of notice of claim.

Section 185, as amended, does not refer to Section 183(a), and no reason has been advanced by appellants for making the right to limitation as provided in Section 183(a) subject to the six months time bar. Indeed, the Congress amended both Section 183(a) and Section 185 simultaneously in 1936 without specifying any time limit in Section 183(a). Moreover, the abuses which resulted under Section 185 prior to its amendment are not encountered when limitation is pleaded by way of answer, because the libelant, by filing its libel, controls the time within which the remedy must be invoked. Therefore, in accord with The Chickie, supra, we hold that the Metropolitan plea of limitation in its answer was not barred by the six months time limit of 46 U.S.C.A. § 185.

Appellants, pursuing what seems to be another phase of the point just above discussed, assert that Metropolitan "posted no bond or stipulation." If all that is meant by this is that the requirements of Section 185, as to time and the posting of a bond or stipulation, are to be read into Section 183, it will suffice to say that we hold otherwise. On the other hand, when limitation of liability is asserted by way of defense in the answer there are interlocutory remedies open to a libelant to prevent loss or depreciation of the vessel by requiring surrender or the posting of security. These are procedural matters which do not appear to affect the substantive rights of the parties and they are normally litigated before or at least during trial. As the record discloses that certain proposed stipulations were served by Metropolitan on the office of libelants' proctors but were apparently mislaid, and there are references to attempts to agree upon a form of stipulation similar to the one entered into with the Railroad, we have concluded that, in this unclear state of the record, we shall and do leave this point open, to be decided by the District Court on the remand.

We next consider whether the Railroad and Metropolitan limitation funds must include the value of all craft in each of the two flotillas, as is urged by appellants, or whether the surrender of the tugs "Long Branch" and "Metropolitan No. 2," and not the scows or carfloats, is sufficient for the purpose of limiting liability, as contended by the respective appellees. If the question were an open one we would be disposed to include the whole flotilla if we could, especially in the case of the carfloats, but we must take the law as we find it, and the question is no longer open. In Liverpool, Brazil & River Plate Steam Navigation Co. v. Brooklyn Eastern District Terminal, 251 U.S. 48, 40 S. Ct. 66, 64 L. Ed. 130, libelant's steamship was damaged while it was moored at a pier in Brooklyn when a carfloat lashed to the port side of respondent's tug collided with it. The tug was proceeding up the East River with the carfloat on its port and a disabled tug on its starboard side. All three vessels were owned by respondent. The sole question before the Supreme Court was "whether the value of the whole flotilla should not have been included in the (limitation) decree." The Court held that only the value of the tug was required by law to be included, because the carfloat, although it was the vessel which collided with the steamship, was a "passive instrument in the hands of" the tug, and "for the purposes of liability the passive instrument of the harm does not become one with the actively responsible vessel by being attached to it." Previously there had been disagreement among the Circuits regarding this question. See The Begona II, D.C.Md., 259 F. 919. But the Supreme Court, in an opinion by Justice Holmes, in the Liverpool case agreed with the Second Circuit view that the tug and tow did not become one vessel and placed the question at rest on the theory that the words "any vessel" in the Limited Liability Act meant the offending vessel and not others attached to it, even though all were in common ownership and engaged in a common venture. We are not at liberty to disregard this binding precedent simply because a contrary view may seem to reach a conclusion more in keeping with the realities of this particular case.

As we have previously noted, it is clear that the carfloats and sand scows in tow of the tugs "Long Branch" and "Metropolitan No. 2" were merely passive instruments of navigation. They had no motive power of their own. Accordingly, only the value of the two tugs need be included in the limitation fund.

Those cases cited by appellants, in support of their position that the passive instrument as well must be surrendered, involved a breach of contract where the libelant had entered into a contractual agreement with the respondent, either for the transportation of himself and his goods, The Big Chief, D.C.E.D. Mo., 75 F. Supp. 496; see also, The Columbia, 9 Cir., 73 F. 226, or of employment, Standard Dredging Co. v. Kristiansen, 2 Cir., 67 F.2d 548. As no such contractual relationship existed between any of libelants and either the Railroad or Metropolitan, the cases cited by appellants are not in point. This distinction is noted in Sacramento Nav. Co. v. Salz, 273 U.S. 326, 332, 47 S. Ct. 368, 370, 71 L. Ed. 663, a case involving exoneration under the Harter Act, 46 U.S.C.A. § 192, as follows:

"This court held (in Liverpool) that it was necessary to surrender only the active tug, saying `that for the purposes of liability the passive instrument of the harm does not become one with the actively responsible vessel by being attached to it.' But this is far from saying that the entire flotilla might not be regarded as one vessel for the purposes of the undertaking in which the common owner was engaged at the time of the collision. The distinction seems plain. There the libel was for an injury to a ship in no way related to the flotilla. It was a pure tort; no contractual obligations were involved; and the simple inquiry was, What constituted the `offending vessel'? Here we must ask, What constituted the vessel by which the contract of transportation was to be effected? a very different question."

See also, In re O'Donnell, 2 Cir., 26 F.2d 334.

Moreover, if fault is attributable to more than one of the vessels engaged in a common venture or enterprise, each with motive power of its own, then each such vessel must be surrendered. United States v. The Australia Star, 2 Cir., 172 F.2d 472. Likewise, if a vessel in tow of a negligent tug is found at fault, for absence of lights for example, the value of both vessels must be included in the limitation fund. Petition of Lake Tankers Corp., S.D.N.Y., 1955 A. M. C. 55; The Bowling Green, D.C.E.D.N.Y., 11 F. Supp. 109, affirmed Czarnikow Rionda Co. v. Ellerman & Bucknall S.S. Co., 2 Cir., 81 F.2d 1017. But, in the case before us, as in Liverpool, no fault whatever is attributable to the sand scows or to the carfloats. It is true that they made the flotillas more cumbersome, but it does not follow that they were at fault, and we hold they were not.

The decree below is reversed and the consolidated cases are remanded for further proceedings not inconsistent with this opinion.