The barge Devon, belonging to the Susquehanna Coal Company, brought a cargo of coal from South Amboy, N. J., to Boston. It was to be delivered at the wharf of the Bay State Fuel Company, on Broad Canal, in East Cambridge. The Boston Towboat Company’s harbor tug Vim undertook to tow the barge to her discharging berth. While in tow of the Vim the barge grounded and stuck fast in the canal, before reaching her berth at the wharf. This happened during the afternoon of Monday, September 10, 1906. The work of getting her off, which required the discharge of her cargo by means of lighters, was done by the Towboat Company. Before she .could be taken off, the barge lay aground during several tides and sustained considerable damage. The Eastern Dredging Company had been engaged, shortly before she grounded, in dredging that part of the canal, for the purpose of making a wider and deeper channel in it. In the first of the above libels the owner of the barge seeks to recover for the barge’s injuries, either from the Dredging Company, upon the alleged ground that it negligently obstructed the channel in *819;be course of its operations and failed to warn against such obstruction, or else from the Towboat Company, upon the alleged ground that it negligently performed its undertaking to tow the barge safely to her berth. In the second libel the Towboat Company claims compensation for salvage services rendered to the barge after she had grounded.
The barge was 176 feet long. Her breadth of beam, including her guard rails, was 35 feet 6 inches. She was drawing, with this cargo on board, 14 feet 5 inches aft, and 13 feet 8 inches forward.
The canal, at that part of its length with which this case is concerned, was about 100 feet wide; but by no means all of its width was available, even at high water, for vessels of this barge’s size. A short distance above its mouth First street crosses the canal by a drawbridge. In order to get the barge to her destined discharging berth, which was not far above the draw, on the north side of the canal, she had to be taken through the draw in the First Street bridge, and thence through a deep water channel in the canal extending from the draw to the berth. There was another discharging berth for barges, nearer the-draw than hers, also on the north side of the canal. If, as was the case on September 10, 1906, there happened to be another barge already occupying this lower berth, she had to pass that barge in order to reach her own berth. At the point where the other barge had to be passed, the deep water channel was supposed to be just wide enough to let one barge pass another. Outside it, and between it and the southern bank of the canal, the water in the canal was entirely too shallow to float loaded barges at any time of tide. The southerly side of the deep water channel sloped somewhat abruptly to the bottom, from this shallow water portion of the canal. The northerly side of the deep water channel was formed by the northern bank of the canal. The above conformation of the bottom in that part of the canal referred to was in part due to dredging done there in July, 1903, for the .purpose of providing a deep water channel sufficient in width to permit one barge 35 feet in width and drawing 14*4 feet to pass another similar barge lying against the north bank in the lower discharging berth. A plan made at the time shows a width for the deep water channel to be secured by the dredging then contracted for of about 68 feet, measuring from the north bank of the canal. This channel was deep enough at high water to float a barge of the Devon’s draft, but it was scarcely more than deep enough. At low water it would not have nearly enough water in it to float such a barge at all. In order to get such a barge up the canal, through the channel from the draw to the upper discharging berth, it was necessary to take her up at high water and to be careful that she was kept in the deep water channel. The risk of grounding her upon the sides of the channel, where the water was not quite deep enough, by getting her slightly out of mid-channel, where the deepest -water was to be found, was, as is obvious, a risk which had to be always anticipated; and the less the general depth in the channel at the time, the greater, of course, would be the risk of such an occurrence. Barges were, for this rea*820son, always taken up as nearly as possible at the exact time of high water in the canal.
The Devon had, on many previous occasions, been taken up the canal to the same discharging berth to which she was destined on September 10, 1906. She had been so taken up several times in each year since 1902. On many of these occasions she had been taken past another barge lying in the lower discharging berth. She had never before got fast aground while being towed up. The last time she had been so towed up before September 10th was on the 2Sth of the preceding August, and the last tíme before that on the 8th of the same August. On each of these occasions there was a greater depth of water at high water than there was on September 10th.
. The barge in the lower berth on September 10, 1906, was c.alled the Beechwood. I do not find her exact dimensions given in the evidence, but she did not differ much in size from the Devon. If, as seems probable, she was the “scow schooner” or “schooner barge” Beechwood, of Philadelphia, the official list of 1905 gives her length as 186.5 feet and her breadth as 35 feet. She lay heading up 'the canal. In .trying to pass her, the Devon, towed by the Vim, went far enough up the canal to get about half her own length past the Beechwood’s stern and there became fast aground. Efforts to move her from this position were almost wholly unavailing; her bow being swung a few feet only. At low water she was left aground on the bottom nearly where she first stranded, with only 4 or 5 feet of water underneath her.
Shortly before September 10th, and since the Devon was last in the canal before that date, dredging had been done in this part of the canal by the Eastern Dredging Company. It was done as part of certain excavation and improvement work contracted for by the Charles River Basin Commission, a state commission appointed under chapter 465 of the Acts passed by the Massachusetts Legislature in 1903. The dredging in this canal, which was included in the total work proposed, had. been sublet to the Eastern Dredging Company, and it was being done there with the approval of the commission and under the supervision 'of its engineer, according to plans and specifications forming part of the contract for it. The contract called for a considerable widening and deepening of the existing deep water channel which has been described. Not less than 17 feet of water was to be provided to and at all the wharves on the canal as far up as the Third Street draw, situated some distance above the berth to which the Devon was bound.
The Dredging Company began its work in that part of the canal with which this case is concerned on September 5th. The Beechwood was not then in tfie canal. Using its dredge Mystic, it made a straight cut 22 feet wide, beginning at the First Street draw, and extending up the canal therefrom about 200 feet, to a point nearly opposite that at which the Beechwood’s stern rested, when that barge was in her berth, and somewhat more than 80 feet, therefore, if the Devon was overlapping the Beechwood by half her own length, below the point where the Devon’s bow rested when she grounded on September 10th. *821On its southern side, this cut was nearly coincident with the southern side of the old deep water channel. The bottom of the cut was from 3 to 5 feet deeper than the old bottom of that channel, in which the cut was made. The operation of the scoop' of the dredge was such as to leave the upper end of this cut semicircular in outline from side to side of the cut. Having been continued on September 6th, the work of dredging was discontinued on September 7th, because of the Beechwood’s arrival and occupation of the lower berth. The dredge was then moved further up the channel to a point above the berth to which the Devon was bound. She was at work there on September 10th when the Devon came.
The operation of the scoop of the dredge in making such a cut as was made on September 5th, 6th, and 7th is such as tended to leave around the edges of-the cut made, both along its parallel sides and around its semicircular end, what is called a “heading,” consisting of more or less material from the bottom cut into, loosened, disturbed, or dragged upward by the scoop, or fallen from the scoop in the process of lifting its contents in order to dump them above water.
That the Devon, after she had once become fast aground, was strained by having to lie at low tide partly over the cut which the dredge had made, is sufficiently probable upon the evidence. That she was stopped and caused to become fast aground at that place by a “heading” left by the dredge, as the libelant contends,' seems to me by no means so clearly shown.
[ 1 ] 'The libelant, for the purpose of proving that the Devon was caused to ground by such a “heading,” relies largely on the undisputed facts that she had passed through the same part of the same channel, loaded in substantially the same way, around another barge in the lower berth, on the previous occasions referred to, without grounding, but that she failed to pass through, for the first time, upon the first attempt she made after the dredge had been at work there. Much of the apparent force in the argument from these facts, however, disappears in view of the other facts disclosed by the evidence, which are next referred to.
Kotwithstanding the success of the Devon in passing this point on previous occasions, it is impossible to say that the channel had ever been safe and sufficient for her passage. It seems to me clear, on the contrary, that she had always and of necessity run a very considerable risk of grounding at this same place, through failure to employ the few minutes during which the water was highest, or through failure to keep exactly in the right course. A very slight want of care in these respects must always have been all that was necessary to get her aground. Even with the tide at its height, she never had more than á few inches to spare under her bottom, around her bilges, or alongside, as the soundings later referred to show. So narrow were the limits within which water enough for her could be found between another barge like the Beechwood and the southern limits of the channel that a very slight deflection from the proper course would be enough under the most favorable circumstances to make her touch or drag on the southern side. As has been stated, the channel *822at this point was intended for two barges of 35 feet each in width, and it had been made barely sufficient for two such barges. The Devon’s width exceeded 35 feet by 6 inches; the Beechwood’s was 35 feet, at least. It had not been uncommon for the Devon, in going up the channel fully loaded, to touch upon and drag over or through portions of the bottom, said to have consisted of mud. Grounding in this way a little harder than usual might so easily involve delay enough for a fall of tide such as would make starting her again before the next high water an impossibility that I must regard the risk of’ such a result as a serious risk necessarily incurred whenever she was taken up.
Proof, therefore, that the Devon’s attempt to pass the Beechwood on September 10th was made while the tide stood at its height, and that in making it she was not allowed to deflect at all from that course which would have involved the least risk of grounding, is essential before it can be safely concluded that because the grounded there must have been in the channel a new obstruction, put there since her last passage through it, and that this was what caused her to ground.
I think the libelant fails to show that the barge grounded at high water. This proof is the more essential to its case because a low course of tides was prevailing, and the best water ever to be expected was not to be had at all on that day. Witnesses who were present and say that the barge grounded at about high water are not wanting ; but I am not satisfied that any of them gave much exact attention at the time to the question when the tide was or was to be at its maximum height in the canal, or noted with any accuracy the precise time when the barge grounded. It appears that when the Devon got to the First Street draw on her way up, the drawtender would not let her through, because the tide was not then high enough. It appears that when the draw was opened she had to be dragged through it, scraping the bottom as she went, and that a good deal of time was thus occupied before she was clear of the draw and ready to proceed up the channel above it. The drawtender was not a witness, and it did not appear just when she arrived at the draw, nor just how long it took to pull her through it. There was some evidence that the draw was opened about half'past 4. The calculated time of high water in Boston Harbor on that afternoon was 4:40 p. m. High water in the canal cannot have been more than a few minutes later. The readings of automatic tide gauges at Moon Island and at Craigie Bridge were in evidence. The Moon Island reading agrees well with the calculated time. The Craigie Bridge reading, however, would put the time of high water there at 4 p. m., which is, of course, impossible, if the calculated time or the Moon Island gauge were anywhere nearly right, and so impossible as to indicate some inaccuracy in the working of the gauge too serious to permit any reliance on it. There was evidence that it was nothing unusual to have to drag the Devon through the draw, as was done on this occasion, and that whenever she could be dragged through at all, even if it took from half an hour to two hours, there would be water enough to get her to her berth. The *823fact that every tide must fall 9 feet at least in 6 hours seems to make it impossible to accept this statement; but the fact that it is made by those who were concerned in towing the Devon up may perhaps indicate that the}' felt no pressing need of haste in their operations. 1 have found nothing in the evidence to render the supposition at all improbable that one reason, at least, why the barge grounded was a fall in the tide before she reached the place of grounding. Her master testified, it is true, that it was 4:45 p. m. by his cabin clock when she grounded; but this testimony he did not give until he was recalled, near' the end of the trial, and after the testimony as to the time of high water had been heard. I am unable, on the whole, to accept it as fixing the time of grounding with reliable accuracy, or as sustaining the burden which is on the libelant to show that the grounding was not caused by attempting to get through after the tide had begun to ebb.
Nor do I think the libelant has made it sufficiently clear that the barge did not deflect at all from the proper course while trying to get by the Beechwood. Three plans of soundings, one of them made under the libelant’s auspices, indicate her position while finally aground. All agree in locating her slightly crosswise of the safest course. Each shows her bow slightly inclined toward the southern side of the channel. The evidence is that a hawser made fast on that side and taken to the barge’s steam winch was used to help pull her up channel from the draw. The tendency of this was and could only have been to carry her bow toward that side of the canal. And it appears, also, that in the efforts to pull her off, which followed her grounding, her bow was swung several feet to the northward, before she settled into the position indicated on the plans. If so, she was at first more crosswise of the channel than,the plans show.
On the evidence thus far considered it seems to me quite as probable that the barge would have grounded where she did, whether the dredging of September 5-7, 1906, had been done or not, as that it was a heading thrown upon the bottom by the dredge which grounded her. The libelant’s case might nevertheless be established by proof that the dredge did actually have a “heading” sufficient to ground the barge at the point where it must have been left in order to stop her. Of direct evidence, however, that the dredge actually left a “heading” at the point referred to, there is very little. No one claims to have seen any such “heading.” and it may well he that, if a “heading” had been there, no one could have seen it, on account of the depth of the water over it, even at low tide. No one claims to have found anything of the kind by any method of examination between the cessation of work by the dredge on September 7th and the barge’s attempt to go up on September 10th. A few soundings, not accurate enough and not sufficient in number to be significant, were made around the barge as she lay aground. After she was removed, more complete and accurate soundings of the whole locality, including the place where the barge had been, were made on behalf of both parties. I have been unable to find in them sufficient indications that the dredge had in fact left any “heading” around the edges of the cut it made, such as *824can reasonably be supposed to have grounded the barge. The soundings seem to me to show the existence of a shoal place in the old deep water channel, further up than the place where the dredge’s cut ended, over which the Devon could not have expected to pass without at least touching, except upon a depth of water greater than the average depth at ordinary high tides. The only circumstance in favor of the conclusion that the dredge had anything to do with the existence of this shoal place is that the dredge’s cut may be said to have ended, generally speaking, about where the shoal spot referred to begins, in going up the channel. But the shoal place, as indicated by the soundings, occupies far too much space in the channel, both lengthwise and crosswise, and indicates an inequality in the bottom far too extensive in its total contents, to permit the conclusion that it was a heading made by the dredge which caused its existence there.
This view of the matter is strongly supported in another way. Soundings of this same part of the channel, made in December, 1903, November, 1904, and December, 1905, were also shown on plans produced. They seem to me to agree in indicating that the same shoal place was found to exist in the same part of the channel at the times referred to. It appeared that the dredging done in July, 1903, as already above stated, was in part occasioned by the existence of a bank projecting into the deep water channel at this point from the southern side of the canal. The widening then done consisted in great part of a cut made into this bank. The evidence as to the width of .the cut then made, and the contour of the bottom as disclosed by the soundings since made, down to and including those of December, 1905, seem to me to show that enough of this bank was left to constitute the shoal place in the channel whose existence all the soundings agree in indicating. I do not think the soundings made since the Devon’s accident indicate that it was then found to be of more importance as an obstruction in the channel than it had previously been. They do not indicate any substantial addition to what was there before. That those soundings, or any of them, show a “heading” also found in the channel, or the remains o.f one, it does not seem to me possible to say. The shoalest places in the channel which appear upon the plans are not found sufficiently near to the point where a “heading” around the cut would have been to make this a probable conclusion.
The testimony of persons experienced in dredging was offered, on both sides, as to the size and character of the “heading” which such a dredge would be likely to leave around such a cut as was here made, in a bottom composed of such material as was here found. This evidence is, of course, to be carefully considered in connection with the indications to be gathered from the plans of soundings. On behalf of the libelants there was evidence that the heading left would be in height from 6 inches to 2% .feet, and of a width on the bottom about equal to its height, but depending upon the skill with which the dredge had been operated, and that it might cause such a barge to ground. On the defendant’s behalf there was evidence that the “heading” would not exceed 6 'inches or a foot in height, and 1% or 2 feet in *825width, that it would be on the extreme edge of the cut, that it would be loose and soft in character, and that it would not obstruct the barge more than so much mud. I am not satisfied, on all the evidence, that any “heading,” such as the dredge can reasonably be thought to have left here, would, after the tide had flowed over it between the 7th and 10th of September, constitute an inequality in the bottom sufficient in size to be indicated, or found at all, by soundings made, as these were, not nearer together than 5 feet, which is the least distance between soundings on any of the plans. I am therefore unable to regard it as satisfactorily shown that the dredge did in fact cause any material shoaling of the channel, or create any material obstruction to the barge’s progress.
[2] If not responsible for getting the barge aground, I do not think the dredging company can be made responsible for her injuries, so far as caused by straining at low tide, even though these were be-1 cause she happened to ground while partly over the dredge’s cut. There is no evidence that the place would have been safe for her to lie aground in before the cut was made; but, even if this be assumed, it is not shown that vessels were accustomed to lie aground there, or anywhere in the canal, except at the discharging berths along its banks. Those who invited vessels to use and occupy such berths were bound to keep them reasonably safe for such vessels as might lie in them upon such invitation, and would be liable for damage due to defects in them; but it is difficult to see how anything done to a part of the bottom where lying aground was not to be anticipated can constitute a violation of any rights, especially in a case where, as here, all that was done was with the sanction and approval of the proper public authorities and there was no obstruction to navigation. Failure to warn against a danger created by such work might no doubt under some circumstances constitute negligence on the part of a contractor for such work, as in Casement v. Brown, 148 U. S. 615, 13 Sup. Ct. 672, 37 L. Ed. 582. This contractor, however, was not bound by its contract to give such notices, as was the contractor in that case, and its liability to warn against dangers cannot, it would seem, be extended beyond such dangers as it ought to have anticipated that some vessel might incur. The danger to this barge in lying where she did at low water does not seem to me to fall under that description. I am obliged. therefore, to dismiss the libel in the first case as against the Dredging Company, for want of proof sufficient to sustain it.
f 3 | The libelant’s case against the Towboat Company stands upon a different basis. Having undertaken to tow the Devon through the difficult and dangerous channel already described, the Towboat Company was bound to know its difficulties and dangers, and to use that degree of care and skill which was necessary for the purpose of avoiding them. The Devon was without motive power of her own, her navigation was wholly under the tug’s control, all her movements were subject to the tug’s direction, and there is no suggestion of any independent action on the Devon’s part having a tendency to cause her to run aground. Under such circumstances the fact that the Devon grounded raises a presumption that the tug was negligent, as in Burr *826v. Knickerbocker, etc., Towage Co., 132 Fed. 248, 65 C. C. A. 554; The W. G. Mason, 142 Fed. 913, 915, 74 C. C. A. 83. The burden is on the tug to explain the cause of the disaster.
The Towboat Company relies, for the purpose of showing that it was not negligent, upon the same evidence which is claimed to charge the Dredging Company with the responsibility for the Devon’s grounding, and which has been above held insufficient for that purpose. If the evidence is not sufficient to charge the Dredging Company, it "follows that the responsibility must rest upon the Towboat Company.
Even if it had appeared that anything done by the Dredging Company had increased the dangers and difficulties of- that part of the channel in'which the Devon grounded, there was evidence in the case which seems to me to prevent the exoneration of the Towboat Company from responsibility. It appears that the master of the Vim 'knew that the dredge had been at work in the part of the channel in question between September 5th and 7th. If, knowing this, and knowing, as he must be held to have known, the slender margin of safety upon- which every attempt to tow such a .barge through such a place must at best be made, he attempted to tow the Devon through without inquiry as to changes in the bottom made by the dredge, I do not see how the company which employed him can escape responsibility for the consequences.
As against the Towboat Company, therefore, the libel brought by the owner of the Devon is sustained. The Towboat Company seeks in its answer to limit its liability, if any, to the value of the Vim, and its right to such limitation is not disputed. It is understood that in any case the damages sustained by the Devon must considerably exceed the value of the Vim. Some evidence regarding the value of the Vim on September 10, 1906, has already been heard. If the inquiry is 'completed upon that point, I find that her value was $800, for which sum, with costs, there may be a decree in the first case against the Towboat Company. If the inquiry has not been fully completed, there may, instead, be an interlocutory decree against the Towboat Company, and such further evidence as either party desires may be taken before the assessor, who will also fix the amount of the damages to the Devon, unless agreed by the parties.
The libel in the second case, brought by the Towboat Company against the Devon for salvage services rendered after she had grounded, must stand dismissed, with costs.