The plaintiffs excepted for the refusal of the following prayer for instruction: “That in no aspect of the case can you consider the run or thread of the stream as it existed in 1889 (at the date of the Iiallyburton deed), or as it now exists, at its intersection with the eastern margin of Depot Street, as defining the beginning corner.” It not having been found or admitted that the stream had changed its location, this prayer could not have been given, though it is true that the doctrine of accretion has no application here. That applies where a stream is a boundary. In such cases the slow, gradual, imperceptible accretion to the bank from the shifting of the stream belongs to the proprietor on the gaining side, and the stream remains the boundary. But here the stream is not the boundary. It is mentioned as one. of the means to locate the beginning point of the survey of a town lot whose boundaries were actually run out and marked on the ground, said beginning being further measured at the time as being 170 feet from ascertain angle in the road. The change in said road or location of said angle would not shift the location of the beginning point in the survey of said lot. No more would the shifting of the bed of the stream change the location and boundaries of the lot. But as there was evidence tending to show that the bed of the stream had not changed, in that event (should the jury so find) its location would be- a factor, and the prayer that “in no aspect of the case” should the run of the stream, as it existed at date of the deed or as it exists now, be considered, was properly refused. Even where the doctrine of accretion applies, if the beginning corner is on the bank of a stream, in running the other lines measurement must be made from “X,” where the beginning corner stood at date of deed, else the whole tract' would move with the bank of the stream. But the court erred in excluding the map made by the city surveyor in 1893, whereon at the time he entered the finding of the stake resembling a surveyor’s stake at the point where the evidence of the *314plaintiffs tended to locate the beginning corner or stake. This evidence was offered to corroborate the testimony of the city engineer and was competent for that purpose.
“After the jury had retired and had been out several hours considering the issues in this case, they came back, and at about 11 o’clock at night asked for further instructions. In reply to their interrogatory as to whether or not they could locate the beginning corner by commencing on the alley and by running back the first two lines according to their courses and distances, the courses reversed, the court replied that they could not.” The plaintiffs excepted.
This instruction, taken in connection with the circumstances under which it was given, must have misled the jury into the belief that they could not, under any circumstances nor for any purpose, reverse the two first courses of the ITallyburton deed, whereas, in certain aspects of the case, it was entirely proper for them to have done so. The rule laid down in,our authorities is, that the survey, where practicable, must be run out in the same order in which the surveyor, at the time of the deed, made it. But when there is, as here, the search for a lost corner or line, if corners or lines further on are located, the courses and distances may be reversed and run backward to find the lost corner or beginning. In Harry v. Graham, 18 N. C., 77, Ruffin, C. J., says: “Eor example, if this deed had said that the line from the corner chestnut and red oak ran to a black oak near the patentee’s other line, and gave neither course nor distance, or only one, and thence N. 45 E. 220 poles to a post oak, his own and Beard’s corner, the line might be reversed from the post oak to ascertain the corner of that and the next preceding line, because that affords the only evidence (the black oak not being found or its locality otherwise identified) of the point at which the one line terminated and the other began.” In Dobson v. Finley, 53 N. C., 498, Pearson, G. J., says: “Supposing the pine to be established as the second corner, could the first, a beginning corner, be located. *315by reversing the course and measuring the distance called for from the pine bach — that is, on the reversed course? His Honor ruled that the beginning corner could be fixed in this way. "We agree with him. If the second corner is fixed, it is clear to a mathematical certainty that by reversing the course and measuring the distance you reach the first corner.” Both these cases are cited, approved and followed in Lindsay v. Austin, 139 N. C., 467, 468, in which case the authorities are reviewed and the principles applicable are clearly stated by Mr. Justice Brown, who also points out that “there is nothing in. Norwood v. Crawford, 114 N. C., 518, which militates against the position that this is a proper method of determining the beginning corner.” The same principle that reversing the line may be resorted to when a lost corner or line cannot be located by proceeding in the regular course is laid down in Duncan v. Hall, 117 N. C., 446. Here, one at least of the subsequent calls, the alley way, was identified, or the jury could so find, and, if they did so, the courses and distances could be reversed to locate the beginning.
It is not necessary to consider the other exceptions, as there, must be a new trial, since in the above particulars there is found reversible error.
Error.