In the view we have taken of the case the defenses based on the claims of exemption from liability by reason of the Harter act and of certain provisions in the bill of lading need not he discussed. The cargo was shipped at Baltimore on respondent’s steamship Bulgaria to be forwarded to Hamburg. When the steamship reached a point in the River Elbe opposite Brunshausen, and some 14 miles below Hamburg, the water in the river was low, and under a provision of the bill of lading, and in accordance with the usual procedure in such conditions, the Bulgaria, on the afternoon of May 27, 1902, anchored at the customary place, and some 300 or 400 yards from a point where the steamship Patricia, also belonging to the respondent, was anchored, and proceeded to discharge a portion of her cargo into lighters belonging to the respondent. One of the lighters was made fast on the port side of the Bulgaria near her stem, and the portion of the cargo for which damage is claimed herein had been discharged into this lighter. During the 27th and *2728th, with lighters alongside, the steamers had swung clear of each other with the turning of the various tides. At about 4 o’clock on the morning of May 29th, just as the tide was commencing to flood, the Bulgaria began to swing and drift broadside up the river toward the Patricia, and, while so swinging, it is claimed that she stuck in the mud forward. In any event, her stern continued to swing, and her officers saw that she was likely to collide with the Patricia, and started her engines ahead, and ordered the lighter cast off, in order to prevent it from being crushed between the two steamships. The lighter went adrift, caught on a safety hook of the rudder of the Patricia, sprung a leak, broke the hook, and sank before it could be towed ashore.

The evidence indicates that one of the steamships had dragged her anchor during the preceding night. The court below held that the preponderance of evidence showed it was the Patricia which dragged her anchor, and that this w^as the cause of the collision.

The testimony of the officers of the Patricia that she did not drag her anchor is indefinite and inconclusive, tier chief officer was asleep at the time of the accident and, therefore, could not testify as to the position of his vessel at that time. His second officer testified that he took the bearings of the Patricia the night before the accident, and they were the same as when they first anchored, and that when he came on deck just before the collision the ships were the same distance from each other as on the previous night, “as far as I can say,” or “as far as I could see,” but he admitted, “I didn’t notice; I didn’t take hearings.”

The log of the Bulgaria and the testimony of the Bulgaria’s officers are to the effect that it is “very likely” that one of the ships dragged her anchor the night before; that it was much more likely it was the Patricia than the Bulgaria; that if the Bulgaria had dragged during the night, she would have drifted down and away from the Patricia, and her officers would have noticed it; that they did not notice any such dragging, and did not believe the Bulgaria dragged.

In the face of this testimony, counsel for respondent argues that either the Bulgaria dragged her anchor on the flood tide on the morning of the accident, or that the Bulgaria swung with the flood tide, while the Patricia remained stationary. The unsoundness of the first contention appears from the fact that while the Bulgaria was swinging around with the flood tide there would be no strain on her anchor; the second contention is a mere conjecture, and is not supported by any testimony, except “a guess” on the part of one witness, who ascribes the cause of the accident to the dragging of one of the anchors, and the statement of the second officer of the Patricia, who said that when the Bulgaria was swinging “she took away the current from the Patricia.” The fact that all the witnesses on the part of the respondent concur in ascribing the collision to other causes, and that no such blanketing or swinging had occurred on any of the previous tides, is a sufficient answer to this second contention.

The foregoing considerations establish the correctness of the find*28ings and conclusion of the court below, for the reasons stated in its opinion, that the dragging of the Patricia’s anchor was the cause of the collision, and that the respondent was liable for negligence, in failing to discover the fact that the two vessels had drawn so near together that there was risk of collision, and in not taking measures to avert the danger.

The decree is affirmed, with interest and costs.