delivered the-opinion of the court.
On the trial of this action of ejectment in the court below, the Longdale Iron company, plaintiff, relied on the commonwealth’s grant to John Barclay, of March 7th, 1796, for 20,000 acres, as the same is laid down by the report and plat of survey made and returned by A. M. Lusk, the surveyor appointed by the circuit court of Rockbridge ' county to do the required surveying for the parties in this suit.
To the northwest of the Barclay patent lies the Joseph Grubb patent of date 15th October, 1795, for 20,000 acres, which is also laid down on the said plat and report of survey made by the said Lusk. The plat shows that a small piece of land, triangular in shape, is covered by both of these grants, constituting an interlock, which being comparatively insignificant in quantity, and wild land never in the actual occupancy of any one, it was conceded at the trial, that it belonged to the Grubb tract, by virtue of the seniority of the Grubb patent over the Barclay patent; and accordingly, the judgment of the court in favor of the plaintiff, the Longdale Iron company, based on its ownership of the Barclay tract, excluded from this Barclay tract, as laid down by Lusk, the said small, triangular interlock. On the 4th day of June, 1835, Jasper Cope, then the owner of the Grubb patent, obtained from the commonwealth a grant of 26,250 acres of land “in conformity with an inclusive survey made on the 19th of October, 1832, by *459virtue of an order of the worshipful county court of Bath county,” which grant embraced the 20,000 acres of the Grubb patent, as part of the 26,250 acres covered by it. At the time of the institution of this«action, the defendant, George Randolph claimed to be the owner of both the Grubb patent and the Cope patent, and he relied on both of them to defeat the plaintiff’s claim under the Barclay patent.
The plat and report of survey made by Lusk, in the cause, on which the Grubb and Cope patents are laid down in connection with each other, show that all the land lying east of the eastern line, “ R S,” of the Grubb patent, and embraced within the lines of the Cope patent, is new territory, and that the land in controversy between the parties, is a part of this nao territory, and is covered as well by the Barclay patent, as by the Cope patent.
The suit was brought in the circuit court of Bath county, and, by consent of parties, was removed to the circuit court of Rockbridge, where it was tried at a special term in July, 1884. Heither party requiring a jury, the whole matter of law and fact was submitted to the court in pursuance of section 36, chapter 158, Code 1873. There was a judgment in favor of the plaintiff for all the land included within the lines of the Barclay patent, as determined and laid down by Lusk, except the small portion (the narrow triangular slip, constituting the aforesaid interlock,) which the plaintiff did not claim; and, thereupon, the defendant filed a bill of exceptions to the opinion and rulings of the court, setting forth only the evidence which had been adduced on either side.
Upon this bill of exceptions, this court, to which the case is brought by writ of error, will consider the case under the rule applicable to a demurrer to evidence. .The rule by which an appellate court must be governed in a case where, like this, the court below has passed upon the questions both of law and fact, is laid down in Backhouse’s Executrix v. Selden, 29 Gratt., *460585-6; Dodson v. Culpeper, 23 Gratt, 356-7; Hodge’s Executor v. First National Bank of Richmond, 22 Gratt., 56.
Upon a demurrer to evidence, the court, says Judge Lacy, in Clark’s Administrator v. Richmond and Danville R. R. Company, 78 Va. (3d Hansbrough) on page 712, is “to consider the demurrer as if the demurrant had admitted all that could reasonably be inferred by the jury, from the evidence given by the other party, and waived all the evidence on his part which contradicts that offered by the other party, or the credit of which is impeached, and all inferences from his own evidence which do not necessarily result from it.”
The only important question of fact controverted upon the trial of the case, is whether Lusk, the duly appointed surveyor in the case, had correctly run and laid down the lines of the Barclay patent; and, indeed, the general correctness of the location of the Barclay patent of 20,000 acres, as laid down by Lusk, is not controverted by the plaintiff in error; only the location of the lines “ IT I” and “I J” are called in question.
It is seldom that a grant for so large a quantity of land as 20.000 acres, in a mountainous region, issued so long ago as 1796, can be as accurately located as the one to John Barclay involved in this case. There is usually a difference in compasses—scarcely any two being exactly alike—and old surveyors in running long lines through wild and rough mountain lands were seldom exactly accurate as to courses and distances. From want of skill, and from want of proper care, both on their own part and by their chain carriers, the surveyors of the last century who were engaged in running off wild lands in the western part of the State, are known to have been more or less inaccurate. This is well understood by all—courts and counsel—who have had occasion to look into old surveys and patents of wild lands. In this case, the perfectly ascertained and admitted lines on the east side of the Barclay patent of 20.000 acres, from the beginning corner “A” around to “H,” *461although connecting original corners, do not, in a single instance, correspond exactly in length with the patent calls; and the courses also differ from the patent calls from one-fourth of a degree to three-fourth of a degree—about one-half of them varying west, and the other half of them varying to the east. But the location of this Barclay patent was made by Lusk according to original corners, line trees and other natural objects and land-marks, which, according to the wise and well settled rule of law, controlled the (in this case, slightly varying) courses and distances.
The original survey of the Barclay patent was made by Alexander Campbell, March 19th, 1795. Alexander Campbell was the surveyor of Rockbridge county. The land was sold by Barclay’s heirs, to Jordans and Irvine between 1828 and 1827, and Edward McLaughlin, a surveyor of great reputation, re-surveyed it at the instance of Jordans and Irvine, with reference to their purchase. When McLaughlin made this survey, Alexander Campbell and his chain carriers, and others who accompanied him on the survey, were, some of them, if not all, still alive; and his intimate acquaintance with the country and its inhabitants gave him exceptionable advantages in identifying the boundaries, which he found and fixed to his entire satisfaction; and where he found the corner trees of the original survey gone or in a perishing condition—from forest fires, the depredations of timber plunderers, or of huntsmen, or from other causes—he marked the surrounding trees as “pointers,” thus establishing the corners and enabling surveyors to follow him. So McGuflin, who in 1870 found some of these marked trees, says: “I passed line trees from H to I that appeared old—one particularly that I supposed to be an original Barclay line tree; it was a 'chestnut oak near Ricely’s house, marked with a tomahawk—not now standing—and apparently an- old line tree of the Barclay patent.” John A. McGuflin was one of the finest surveyors and woodsmen in the State, and when he ran the Barclay tract in 1870, when this *462litigation was not thought of, he found an original line tree on the line “ H E,” a “ chestnut oak, near Hicely’s house, marked with a tomahawk, not now standing.” The evidence is that the Barclay survey was all marked with a tomahawk. Mr. A. D. Trotter, a surveyor of large experience, made the same location of the Barclay patent in 1852, under an order of court in the suit of Jordans v. Campbell. The Barclay survey, the Grubb tract, and the Cope tract are proved to be correctly laid down on Lusk’s map or plat by the testimony of the following experienced surveyors, most, if not all, of whom ran the lines and surveyed the tract at different times between the years 1795 and 188-, viz:A. M. Lusk, John II. Jones, John A. McGuffin, Major Jed Hotchkiss, and James II. Callison, and by the survey of the tract, made as aforesaid by A. D. Trotter in 1852. Mr. James H. Callison, a surveyor of great experience, was employed by the appellant, Randolph, to locate the lines “H I” and “I J,” so as to exclude the land in controversy from the limits of the Barclay patent, but Mr. Callison found that he could not conscientiously do it, and thereupon the appellant dropped him. And no experienced or competent surveyor has ever been found who could locate the lines “III” and “I J” differently from the location fixed by the judgment of the court below. How, in opposition to all this concurrent testimony to the correctness of the survey made by Lusk of the Barclay patent, is only the testimony of John Carmichael, who had been a civil engineer for two and one-half years, and who ran some lines “II I” and “I J” by the direction of the appellant, George Randolph, and not pretending to make a complete survey or to locate the Barclay tract. Mr. Carmichael says: “I never saw Campbell’s official survey or a copy thereof;” and he attempted to fix the lines “II I” and “I J” by some field notes of Campbell, which were incomplete, and different in every( respect, fronrtlie actual and official survey made and returned by Campbell of the Barclay . tract, and which do not even appear to relate to the Barclay *463survey as made by Campbell. Carmichael only ran two hypothetical lines, to suit the appellant, Randolph, which it seems the experienced surveyor and woodman, Mr. James II. Callison, could not conscientiously do; but everr if his testimony were worth anything, scientifically and practically, as against the overwhelming testimony of the appellee, it could not be considered by this court in conflict with the appellee’s witnesses.
The petition states that Lusk ran the lines “II I” and “I J” according to the reading of his compass, without making any allowance for the magnetic variation; and that the officials of the United States’ coast survey, in 1884, by observations on the spot, determined that at the date of Lusk’s survey the variation was one degree and five minutes west. But the record shows that these same officials, at the same time and place and by the same observations, ascertained the variation at the time and place in 1795, when Alexander Campbell made the original survey to have been two degrees and thirty-five minutes east, and not four degrees east, as Campbell in his survey says it was. So the real difference between the running of Campbell and of Lusk, when both surveys are set to the true meridian, is twenty-five minutes, which trivial difference is readily accounted for by the difference in instruments alone. And the difference between the two as to the line “I J” is only'ten minutes.
It is plain, therefore, that the location of the Barclay tract of 20,000 acres, as made by Lusk, and confirmed by the judgment of the court below, is correct; whether in locating the lines “H I” and “I J” (which only are in dispute,) we are guided by the coast survey observations and calculations, or by the line and corner-trees found upon the ground. The law wisely places marks and monuments found upon the land before and superior to courses and distances into which errors may creep even in the transcribing of figures, as well as otherwise.
The question of law is as to the legal operation and effect of *464the Jasper Cope patent of June 4th, 1835, so far as it affects the land in controversy, covered both by it and by the Barclay patent. The defendant, George Randolph, (appellant) contends that the survey upon which this Cope patent is founded, although designated in the patent itself as an inclusive survey, was intended to be only a re-survey of the Grubb patent for the correction of mistakes and uncertainties in the bounds thereof; and that) in legal contemplation, it was nothing more than the original Grubb survey with the courses and descriptions of its metes and bounds rectified and properly defined; and that the Cope patent, therefore, relates back to the date of the Grubb patent in 1795, for 20,000 acres, and gives the appellant, Randolph, a title by operation of law to not only the undisputed, clearly defined, and distinct tract of the Grubb patent for 20,000 acres, but to 6,250 acres more outside of the limits of the Grubb tract, and trenching upon and within the limits of the Barclay patent.
The statute under which Cope proceeded, will be found in the Code of 1819, volume I., section 61, page 335, and this statute uses the term re-survey to designate all new surveys, whether new rights are included or not. The notice to adjoining laud-owners and the petition required by the statute, are lost, and the records of Bath county show nothing relating to the matter but the two orders of the county court of-Bath county designating it as a re-survey, while the patent issued to Jasper Cope designates it asan lí inclusive survey.” But the fact is that only one of these orders of the county court of Bath county was sent to the land office with the Cope survey, and it calls for 26,250 acres, and does not call itself a re-survey. These orders do not show to whom notice was given, but the Cope survey itself and the patent issued uqion it as an inclusive survey, both show the adjoining land owners who were recognized by Jasper Cope and notified as such and it thus aqapears both from the survey and patent, that Messrs. Jordans and Irvine, the then owners and occupiers of the Barclay tract, *465under whom the appellee claims, were not so recognized and not notified. Ho man can be estopped by any record, unless it is shown that he is a party to it, and this should be shown-by the record itself. (Chrisman v. Harman, 29 Gratt., p. 499, and cases there cited. 1 Greenleaf on Evidence, § 522, et seq. Russell v. Place, 91 U. S., 606.)
The appellant contends that the efiect of the re-survey made by Cope was merely to accurately define and make certain, lines and courses of the original Grubb patent for 20,000 acres, which were uncertain and indefinite, and that the legal efiect of the new grant to Jasper Cope, based on this re-survey, was not to give him new territory, but to define his rights under the Grubb patent of 1-795. An inspection of the Grubb patent and the Cope patent, as laid down on Lusk’s map together, shows that while the Grubb patent is bounded by nineteen lines, the Cope patent is bounded by eighty-nine lines; and the Cope tract contains 6,250 acres not within the lines or limits of the Grubb tract.
The statute says (§61, p. 336): “the party upon producing new rights for whatever surplus land appears to be within the bounds, more than the before-mentioned allowance of five acres for every hundred, may sue out and obtain a new grant for such lands thereupon, in which shall be recited the dates and other material circumstances of the former title; and the title papers shall be delivered by the register to the new owner.” The “ new rights ” required by the statute mean land warrant, entry, and survey. The Cope survey was worthless unless based upon a land warrant duly entered according to law. If the “ new rights ” were produced,' the title to the new territory—6,250 acres—cannot antedate the Cope grant of June 4th, 1835. If, however, the contention of the appellant is right, that no “new rights” were produced, it is evident that the Cope grant, so far as it covers the new territory, 6,250 acres, more than, and outside of, the lines of the Grubb patent, was issued not only without authority of law, but in direct *466violation of law, and is absolutely void as to the 6,250 acres of new territory. “If a patent is absolutely void upon its face, or"tlie issuing thereof was without authority, or was prohibited by statute, or the State had no title, it may be impeached collaterally in a court of law, in an action of ejectment.” (Patterson v. Winn, 11 Wheaton, 384; Lomax’s Digest, vol. 2, side page 388; Polk v. Wendal, 9 Cranch, 99.) The Cope grant, so far as it embraced new territory, was either void because issued without authority of law, or, as to the new territory of 6,250 acres, embracing the land in controversy, it was founded on “new rights” acquired long after the date of the Barclay patent. In either event the Cope 'patent of 1835 is not in the way of the Barclay patent issued in 1796.
We are of opinion that the judgment complained of is right and that it must be affirmed.
Richardson, J., dissented.
Judgment aeeirmed.