delivered the opinion of the Court.
*669This is an action of ejectment/ in which the only question presented is one of boundary. The land in dispute is covered by the defendant’s title papers, which originate with Morgan county entry No. 2636, on which grant 27127 was issued August 15, 1859. The question is whether the complainants’ title papers which originate with Morgan county entry No. 1782, upon which grant No. 20670 was issued August 24, 1836, cover and include the land in dispute. If so, complainants are entitled to recover because their title is older; otherwise they are not.
The grant from which complainants derive title describes the land as follows:
“There is granted by the State of Tennessee unto the said Stephen Haight and his heirs a certain tract or parcel of land containing five thousand acres lying in the county aforesaid on the south of Daddy’s creek, in the fork of said creek, and Yellow creek, beginning at a white oak and large hickory five poles above the junction of said creek on the south side; then up Daddy’s creek, south five degrees west one thousand two hundred fifty poles to a stake and pointers above the mouth of Crab Apple branch; thence south eighty-five degrees east six hundred forty poles to a stake at the foot of Crab Orchard Mountain; then along the same north five degrees east one thousand two hundred fifty poles to a stake on the bank of Yellow creek; then down said creek a direct line to the beginning, including the flat land between the mountain and Daddy’s creek.”
*670If the streams and the foot of the mountain were in fact located as indicated by the courses and distances called for in the grant, there would bé' no difficulty in immediately locating the land in the form of a parallelogram one thousand two hundred fifty poles in length and six hundred forty poles in width. All agree that the beginning corner is a certain tree definitely located near the junction of the two creeks, but the undisputed proof shows that Daddy’s creek, instead of running south five degrees west, as indicated by the calls of the grant, runs in a very irregular course, and generally very much more westward of south. Yellow creek, instead of running north eighty-five degrees west as indicated by the calls of the grant, runs in an irregular direction, but generally almost due south. The foot of the mountain, instead of running north five degrees east, as indicated by the 'grant, runs something like forty-five degrees east of north. The question therefore arises as to whether or not the boundaries of the land shall' conform to the location Of the streams and the mountain, or to the courses and distances called for. The complainants contend that the land is to be located by running the lines according to the course and distances, regardless of the natural objects referred to. Whereas the defendant contends that the land is to be located by running from the beginning corner up Daddy’s creek with its meanders, the distance called for, so as to reach a point beyond and above the mouth of Crab Apple branch, from thence the course and distance called for, although such a line would not reach the foot of Crab Orchard Mountain; that the third line should run the course *671called for until it reaches Yellow creek, although the line would fall short one thousand poles of the distance called for, and from thence follow the meanders of Yellow creek to the point of beginning, whatever the distance or direction.
The attached diagram will illustrate the contentions of the parties with respect to the proper location of the complainants’ land.
It is quite manifest that there is a conflict in the calls of the complainants’ grant, when compared -with the true location of the natural boundaries referred to therein. How, then, shall we determine the correct location? There are no marked lines on the grant to indicate just where the surveyor upon whose certificate the grant issued located it. If the surveyor actually ran the lines, he did not *672mark them on the grant, and of course was mistaken as to the direction, or of the course of the streams, and the foot of the mountain.
The mere fact that there is a conflict in the calls of the survey, or that the surveyor was mistaken with respect to the location of the objects called for, would not void the grant, if by any reasonable interpretation of the grant its boundaries can be ascertained and determined with reasonable-certainty.
The primary and fundamental rule to which all others relate and must yield is that the intention of the parties gathered from the whole instrument, taken in connection with surrounding circumstances, must control. Pritchard v. Rebori, 135 Tenn., 330, 186 S. W., 121; 9 Corpus Juris, 152.
We must therefore, in arriving at the true intention of the parties, conform to those rules which have been long recognized as best calculated to accomplish that end, bearing in mind, too, that there are exceptions to these rules, and -that sometimes it is apparent that the true intention of the parties does not conform to the conclusion which a strict adherence to the rules of interpretation would require.
In this case it is evident that the surveyor made a mistake,- and that his mistake was in concluding that Daddy’s creek ran practically north and south, and that Yellow creek ran practically east and west, and that the foot of Crab Orchard Mountain ran practically north and south.
*673A just and reasonable rule which the courts have long followed in harmonizing conflicting calls in a deed or survey is to ascertain which calls are locative, and which are merely directory, and conform the lines to the locative calls. Directory calls are those which merely direct the neighborhood wherein the different calls may be found; whereas locative calls are those which serve to fix the boundaries. Therefore locative calls control in case of conflict. It is less likely that The surveyor or the parties to the instrument under interpretation would make a mistake in describing the exact boundaries than when merely calling attention to the neighborhood where they should be found. Roberts v. Cunningham, Mart. & Y., 68; Disney v. Coal Co., 11 Lea, 607.
If the conflict be between locative calls, then those referring to natural or fixed objects prevail over those for course and distance, unless this results in an absurd conclusion, and one which from the whole instrument and other competent evidence was one that was manifestly not intended. This is upon the principle that in determining the boundaries of a tract of land it is permissible to disregard calls when they cannot be applied and harmonized in any reasonable manner, and thus reject a call which is manifestly false or mistaken. Such a situation might well be developed by reference to the official plat of the surveyor.
The complainants’ grant refers to the land, the boundaries of which are subsequently specified, as lying on the south of Daddy’s creek, in the fork of said creek and Yel-*674Ioav creek. This, of course, is a directory call, one which • directs the inquirer to the general locality of the land. It does not undertake to define and fix the exact boundary. The call beginning at a white oak and large hickory five poles above the junction of the creek on the south side is a specific and located call. The surveyor could not have made a mistake with respect to that, and that point must be accepted, as it is conceded, as the point from which to start in fixing the boundary of this tract. The next call is up Daddy’s creek south five degrees west one thousand two hundred fifty poles. If the surveyor intended that the boundary line should run and be governed by the creek, then he was very much mistaken as to the course of the creek. If he had said, up D'addy’s creek with its meanders, without designating the course, that would have constituted a locative call, and fixed the creek as the boundary line. The call to run' south five degrees west is specific, and can be harmonized with the call for Daddy’s creek, by considering that the line called for was in the direction of the course of Daddy’s creek. Daddy’s creek could not be followed without departing absolutely from the specific course called for. We have in the grant an apt ¿llustration for the rule requiring calls for course and distance to yield to natural or other fixed objects, in the call which fixes the end of the first line at one thousand two hundred fifty poles from the beginning, and at a point above the mouth of Crab Apple branch. If the distance of one thousand two hundred fifty poles fails to reach a point above the mouth of Crab Apple branch, then the distance., must yield and the line go to such a point; but, if the dis*675tance of one thousand two hundred fifty poles would carry the line to a point above the mouth of Crab Apple branch, as the evidence shows'in-this case it does, then both calls' may be harmonized. The same illustration may be found in the next call, which is south forty-five degrees east six hundred forty jtoles to a stake at the foot of Crab Orchard Mountain. If the foot of Crab Orchard Mountain is susceptible of being definitely ascertained, then of course that would control the distance of six hundred forty poles. It would either stop short when it reached that point or be extended until it did reach the point. A call similar to the first one is found in that which runs along with the foot of Crab Orchard Mountain north five degrees east one thousand two hundred fifty poles to a stake. If this call should conform to the foot of the mountain, it would run more than forty-five degrees east of north. This call can only be reconciled by concluding that it was the intention to run the line south eighty-five degrees east, since that conforms to the general course rather than to the specific course of the foot of the mountain. In the third call we find the line to run north five degrees east one thousand two hundred fifty poles to a stake on the east bank of Yellow creek. Here we come upon the mistake of the surveyor in assuming that Yellow Creek ran practically east and west. The second line, if run to the foot of Crab Orchard Mountain or anywhere near there, would have already crossed Yellow creek, and there would be no certaipty of the direction of the line that would be run in order to reach the bank of Yellow creek. But the call is *676to a stake on tbe bank, and not tbe creek itself. Tbe next call is to run down tbe creek a direct line to tbe beginning. This line clearly does not make tbe creek tbe- northern boundary line, but a direct line from tbe stake to tbe beginning.
If the call for Daddy’s creek and for Yellow creek and tbe foot of Crab Orchard Mountain be treated as directory calls, and tbe calls for course and distance as locative, there is no difficulty in arriving at tbe conclusion that tbe complainants’ contention is tbe correct one.
We might well conclude that such was tbe intention of the surveyor, taking into consideration tbe description .in the grant as a whole, and applying it to the situation actually found upon tbe ground.
The conception of tbe locality and form of the boundaries of this tract of land, gained from reading tbe grant or survey, is materially disturbed if tbe courses are departed from and made to follow the lines of tbe streams and tbe foot of tbe mountain. An entirely new picture than that gathered from tbe paper itself is made upon the mind. Tbe form and position of tbe grant which one gathers from reading its description is materially changed when you undertake to move it around to conform to these streams. It is inconceivable that tbe surveyor, judging by what be said in bis report as it is . written in tbe grant, intended to place this tract of land in tbe shape and position which it would necessarily occupy if the- streams and the foot of tbe mountain be considered as tbe locative calls of the grant. It is so radically different that the mind *677naturally resents such a conclusion. It is impossible to harmonize the calls of the grant upon that theory.
In .the case of Roberts v. Cunningham, Mart. & Y., 67, 68, the grant called to lie on the south side of Cumberland river. The course and distance called for in one of the lines, carried a portion of the land across the river, and it was held that the land must be surveyed according to course and distance. This conclusion was reached by treating the reference to Cumberland river as a directory call, and the course and distance as locative. The court said:
“Let it be borne in mind . . . that we are construing a grant, the evidence of legal title; that our sole object is to ascertain, in technical language, identity; and that to give the grant this identity or locality, we .must have boundary — we must have lines; and to draw those lines we must have a point at which those lines are to begin and end. Those calls then in the grant, which serve to fix this boundary, or these lines' and points of beginning and ending, are certainly of more importance to the existence and effect of the present than any call which professes only to throw the whole tract or body of the land into one part of the country or another, on one stream or another, or one or the other side of a creek or river. . . . If a grant has a call for a beginning which is proved, and course and distance from it, it is wholly independent of the assistance of all. other calls.”
Furthermore, we find that the surveyor was mistaken as to the location of the natural objects referred to. He *678supposed that 'Daddy’s creek ran practically north and south,, and that the foot of Crab Orchard Mountain ran practically north and south, and that Yellow creek ran practically east and west. He was not undertaking to locate the direction of these streams. He was fixing the boundary of the land. He was surveying the lines thereof. It is not reasonable that he would make a mistake in the direction of the lines which he was establishing as the boundary of the land, since that is what he was surveying. It would be perfectly natural to make a mistake as to the course of the streams, since he was not surveying the streams.
This view is fortified by the fact that in the line from the beginning he does not call to run with the creek, as he most probably would have done if he intended to make the creek the boundary. Instead of doing that, he calls for a certain course and distance up. the creek, evidently meaning thereby that the line was running in the general direction of the course of the stream. We find, too, that by adopting the cóurse and distante for the first line the call for the foot of Crab Orchard Mountain is substantially complied with. There is some controversy as to just where the foot of the mountain is, as the rise is gradual; but, if the line pursues Daddy’s creek, then the next line certainly can never reach anywhere near to the foot of Crab Orchard Mountain. It cannpt even reach to Yellow creek, which is between the southwest corner and the foot of Crab Orchard Mountain. The foot of Crab Orchard Mountain is just as much a locative call as is Daddy’s *679creek, or yellow creek, and if we adopt the theory that these calls are all locative, and that the line must run with the creek, we would be forced to abandon the call for the foot of Crab Orchard Mountain. If all these natural objects be treated as directory, as undoubtedly they are, then the calls for course and distance are locative, and therefore of greater weight in determining boundary than the directory calls for natural objects. It will be noticed that the east boundary line is not called to end at Yellow creek, but at a stake on the bank of Yellow creek, thus showing, that the surveyor did not actually measure that line, but made a mistake as to the location of Yellow creek, therefore, in assuming that the line would reach to the bank of the creek. If the south boundary line be run to the foot of Crab Orchard Mountain, then the call to a stake on the bank of Yellow creek is ridiculous, for the line had already crossed Yellow creek.
It results, therefore, that the defendant’s, theory that these natural objects constitute locative calls cannot be applied to all of them. Some of them must be disregarded, or the description held void for uncertainty.
. Not only does construing these calls for the streams and the foot of the mountain as directory enable one to harmonize all' of the calls and give to the grant a location with reasonable certainty, but that such was the intention of the surveyor when he certified this description to th,e officers of the state for the issuance of the grant is manifest from his plat of the survey which accompanied the certificate. If a doubt with regard to boundary arise from the calls of the grant, the plat and certificate of survey *680may be referred to, to remove it. In such case the plat shall have not a directory, but a locative, operation in determining its precise limits. This is true although it may result in disregarding a fixed natural or artificial object called for in the grant. Dallum v. Breckenridge, Cooke, 152, Fed. Cas No. 3547; Patton v. Carothers, Cooke, 148, Fed. Cas; No. 10833; Childress v. Holland, 3 Hayw., 274; Roberts v. Cunningham, Mart. & Y., 73; Disney v. Coal Creek M. & M. Co., 11 Lea, 607; Bell v. Hickman, 6 Humph., 398; Tate v. Gray, 1 Swan., 74.
The plat of the surveyor in this case shows this tract of land to be in the form of a parallelogram. It is true it shows this parallelogram to lie entirely between Daddy’s creek and Yellow creek, but it also shows lines running according to the course as constituting the boundary line of the land, with the creeks separated therefrom. It is impossible to reconcile the diagram accompanying the official survey, with a form and appearance, made to conform to the contention of the defendant, The plat demonstrates that the surveyor did not intend to bound the lands by the streams and the foot of the mountain, and that he did hot actually survey the streams as constituting the boundary lines, but merely platted in what he conceived to be the general course and direction thereof.
Applying the principles herein indicated to this grant, we conclude that the land lies in the form of a parallelogram, running the courses and distances called for, starting from the fixed object in the forks of the two creeks.
The chancellor reached this conclusion, and his decree is affirmed.