The defendants, Dana and Alice Hinkley, own a summer camp on Sheepscot Lake. Their lot has common boundary lines on the west and south with land of the plaintiff, Ralph Proctor. The location of those boundary lines is in dispute in this case. Proctor appeals from a judgment of the Superior Court, Waldo County, adopting a referee’s report that found for the Hinkleys and awarded them five hundred dollars in damages for trespass. On appeal, plaintiff contends that the referee made certain errors of law and certain cleаrly erroneous findings of fact. We affirm the judgment in part and vacate it in part.
Both parties derive title to their adjoining lots from Tressie Malcolm. The Hink-leys purchased their lot from her by deed, dated October 15,1942, describing the lot as follows:
A certain lot or parcel of land situated in Palermo, Waldo County and State of Maine bounded and described as follows to wit: Beginning at the low water mark on the shore of Sheepscot Lake in Palermo and extending in a westerly direction to the high water mark, then continueing westerly One Hundred (100) feet by land of Earlе M. Hinkley to a stake and stone, then southwest ninety (90) feet by land of grantor to another stake and stone; then east one hundred (100) feet by land of grantor to the high water mark of said lake; then continueing east to the low water mark, then Ninety (90) feet North by the lake shore to the first mentioned bounds.
Earle M. Hinkley was Dana’s father, who had owned since 1941 the lot adjoining Tressie Malcolm’s parcel on the north.
In 1946, Proctor purchased from Tressie Malcolm her parcel of land out of which the Hinkleys’ lot had been conveyed, “excepting and reserving а lot of land 100 feet long and ninety feet wide, at the northeasterly corner ... being the land conveyed by me [Malcolm] to [the Hinkleys] by deed ... reference to which deed is hereby had for a more particular description of said excepted lot, and right of way appurtenant thereto.”
The boundary dispute between Proctor and the Hinkleys began in June of 1977. Dana Hinkley testified that the Hinkleys built a tool shed on their right of way across Proctor’s land after discussing the matter with Proctor and getting his oral consent; that after the shed was completed, Proctor changed his mind and ordered the Hinkleys to move it off the right of way; and that the Hinkleys complied by moving the shed six or eight feet to what they considered to be their property. Plaintiff claimed title to the property on which it stood and again demanded that the Hink-leys move the shed. They refused, claiming title to the property under their deed.
Contrary to Dana Hinkley’s testimony, Proctor denied that he ever gave the Hink-leys permission to build on the right of way. Referring to the structure as a garage and denying that it had been initially placed on the right of wаy, he testified that he only gave the Hinkleys permission to place it one foot away from his property line. On the morning construction began, he observed that the structure was being built on his land and ordered the contractor to stop. The building was completed there but was later moved by the Hinkleys eight feet east, toward their property. On the premise that it still remained on his land, Proctor demanded that the defendant move it again.
Each party had the Hinkleys’ lot surveyed during the fall of 1977. The surveys established different western and southern boundaries. In September, 1978, Proctor filed a complaint in Superior Court, Waldo County, alleging, among other acts of trespass, that defendants had constructed a ga *468 rage on his property. He sought damages and injunctive relief and a judgment establishing title to the common boundaries. Defendants answered and counterclaimed, seeking similar relief. The case was tried before a referee in June, 1981.
At trial, Proctor contended that the crucial issue was the location of the high-water mark of Sheepscot Lake in what he referred to as “the northeast corner” of the Hinkleys’ lot. He assumed that the remaining three comers of the Hinkley lot must be located by making measurements from that starting point in accordance with the distance and direction calls in the deed; i.e., that the northwest corner should be established by measuring one hundred feet west from high-water mark along the land of Earle M. Hinkley; that the southeast corner should be located at the high-water mark ninety feet southerly of the starting point; and the southwest corner should be located one hundred feet west of the southeast cоrner and ninety feet southwesterly of the northwest comer.
The key witness supporting Proctor’s theory was surveyor Thibodeau. In making his survey, he first established a high-water mark in what he referred to as “corners” of the Hinkleys’ lot on the northeast and southeast. 1 Both those high-water marks were located several feet below the top of the bank. Based on conversations with natives in the area, Thibodeau gave as his opinion that the 1977 high-water mark was one to three feet higher than the high-water mark in 1942. 2 Thibodeau made his survey with respect to the high-water marks ascertained on that basis, thereafter locating the northwest and southwest corners of the Hinkleys’ lot by measuring one hundred feet west from the high-water marks in accordance with the distance and direction calls in the deed. Thibodeau did not find the monuments called for by the deed (“stake and stone”) at the northwest and southwest corners but did notice an iron pin some number of feet westerly of the northwest corner as fixed by his survey.
The Hinkleys relied on a survey conducted in November of 1977 by J.R. Curtis. He conducted his survey with respect to three iron pins which were pointed out to him by Dana Hinkley. He started his survey at an iron pin located 2.1 feet northwest of a hemlock tree on the top of the northeast bank. Although the Hinkley deed unambiguously states that the northerly line of the Hinkley lot goes westerly from high-water mark, Curtis seems to have proceeded on the assumption that the deed was wrong; that the pin near the hemlock tree on the top of the bank was intended to be the monument marking the starting point for the second call of the deed. Curtis admitted that he did not conduct his survey from high-water mark as cаlled for by the deed. Like Proctor’s surveyor, he located the high-water mark several feet below the bank on which the hemlock tree was located.
Curtis placed the northwest corner at an existing iron pin 99.49 feet west of his starting point — apparently the same iron pin Thibodeau had noticed as being west of the northwest corner he had established. Dana Hinkley testified that this pin was an original boundary marker.
From the pin in the northwest corner, Curtis measured 82.64 feet southwest to a third “pin”, which he used to establish the southwest corner of the Hinkleys’ lot. According to testimony of the Hinkleys, that pin was not the original monument called *469 for by the deed. It was, in fact, an iron stake that Dana Hinkley testified had been placed there by Proctor pursuant to an agreement by the parties sometime before 1966 to replace the original monument, which was then missing. The Hinkleys testified that both parties had then agreed that the new stake would mark the location of the southwest corner. According to Curtis’s survey, the Hinkleys’ western boundary line was about ten to twenty feet westerly of the line established by Proctor’s surveyоr.
Following a line in an easterly direction from the new stake to the shore, Curtis was unable to find any monuments. He set a nail 90 feet southwest of the iron pin found near the hemlock tree. He used the nail to establish the southeast corner; it was 100.-22 feet east of the stake in the southwest corner.
In August, 1981, the referee found in favor of the defendants, establishing the boundaries in accordance with the survey conducted by Curtis 3 and awarding defendants $500 in damages for a fence destroyed by plaintiff and other minor damage and inconvenience. The refereе ruled that Proctor had failed to prove the location of the high-water mark, saying “If anything was proven with respect to high-water mark the proof was in favor of defendants; namely that the high-water mark was at the top of a bank that runs along the shore.” To support his ruling, the referee stated, “The witness Curtis was able to actually and with reliable, admissible evidence, place monuments on the four corners of defendants’ property. The most important monument, which I find to be a true monument on all the evidence, is an iron pin located on [Curtis’s survey map] with the notation ‘hemlock tree (15 inch DIA and xh inch Rebar found’).” The referee also ruled that he believed the Hinkleys’ testimony regarding the placement of a stake by Proctor at the southwest comer of the Hinkley lot.
Plaintiff moved in Superior Court to set aside the referee’s report on the grounds that several findings by the referee were clearly erroneous and that the referee erroneously admitted extrinsic evidence to contradict the plain language of the deed. The Superior Court adopted the refеree’s report in its entirety and entered judgment accordingly. With regard to Proctor’s objections to the report, the presiding justice ruled that the referee had properly admit/ ted extrinsic evidence to locate on the face of the earth the monuments mentioned in the deed and that the findings of fact were not clearly erroneous.
In finding that “the most important monument” was the pin located near the hemlock tree on top of the bank in the northern line of the Hinkley lot, the referee was clearly in error. Since neithеr the hemlock tree nor the pin is referred to, even indirectly, in the description of the Hinkleys’ lot, neither could serve as a “monument.” A physical object is not a monument for the purpose of locating a boundary unless the description makes reference to it for that purpose. Furthermore, the referee could not correctly find that the hemlock tree or pin near it was at the high-water mark of the lake, for there was no competent evidence to support such a finding.
What the boundaries are, as ascertained from the language of a deed, is a question of law; where they are on the face of the earth is a question of fact.
Bailey v. Look,
What the referee did, in effect, was to reform the Hinkleys’ deed, informally and
sub silentio,
to substitute the pin near the hemlock tree for “the high-water mark” of the lake as the monument marking the westerly terminus of the first line call. Whether or not such informal reformation might have originally been proper as against Tressie Malcolm, the Hinkleys’ grantor, it was improper as a basis for determining rights against Proctor, a subsequent purchaser for value of the entirе Malcolm parcel excepting the Hinkleys’ lot already conveyed. There being no evidence that Proctor had notice at the time of purchasing from Malcolm that- the Hinkleys might have a right to reformation of their deed, Proctor took from Malcolm free of that equitable right if it existed. In the absence of reformation, the referee had to apply the terms of the description in the deed.
Perreault v. Toussaint,
Although neither the pin nor the hemlock tree may serve as a monument in determining the boundaries of the Hinkleys’ lot, competent evidence supports an implicit finding by the referee that the hemlock tree is the very tree referred to in the earlier deed by which Earle M. Hinkley acquired his adjoining lot on the north. In that instrument, Earle Hinkley’s southern line (the northern line of Dana and Alice Hink-ley’s lot) is described as follows:
[TJhence Easterly one hundred (100) feet, more or less along the line of said barbed wire fence to a hemlock tree and thence in the same line to low water mark on the shore of said Sheepscot Lake.
Since Malcolm’s deed to Dana and Alice Hinklеy calls for their northerly line to go “by land of Earle M. Hinkley,” Earle’s southerly line, if ascertainable, is a monument for locating their northerly line. Thus, although the tree is not a monument for determining Dana and Alice’s northerly boundary line, it does mark a point that must be located on that line.
Although the referee erred in finding that the pin near the hemlock tree marked the westerly terminus of the first line call
(i.e.,
from low- to high-water mark) and the easterly terminus of the second call, Proctor’s method of determining the Hink-leys’ northwest corner is not correct either. He erroneously assumes that once the high-water mark of the lake is located, several feet east of the top of the bank, the remaining three corners of the Hinkley lot must be established by measurements from that point strictly in accordance with the distance and direction calls of the Hinkleys’ deed. But the critical determination for
*471
settling the Hinkleys’ northwest corner is the location, not of the high-water mark, but of the western terminus. Because the deed states that that terminus is marked by a stake and stone, the referee was obliged to determine, if рossible, the location of that monument on the face of the earth. He correctly admitted extrinsic evidence for that purpose.
Perreault,
On the basis of competent evidence at trial, the referee found that the Curtis survey correctly located the northwest comer of the lot, placing it at an iron stake called to his attention by Dana Hinkley. In making the finding, the referee referred to the stake as a monument.
The location of that monument was a question of fact for the referee, whose finding must stand if supported by competent evidence.
Bailey,
The fact that the stake was more than one hundred feet from the high-water mark according to plaintiff’s еvidence does not invalidate the referee’s determination that the stake marks the western terminus of the Hinkleys’ northern boundary. If there is a discrepancy between calls based on ascertained monuments and distance calls, the former prevail.
Edmonds v. Becker,
The location of the southwest corner presents a different issue. The iron pin that the referee found established the southwest corner is not the original monument. Thе Hinkleys testified that Proctor placed the existing stake at this location to replace the missing stake; Proctor denies that he did so. Sometime before 1966, Dana Hinkley discovered that the stake marking the southwest corner of his lot was missing and spoke with Proctor about the matter. The Hinkleys testified that Proctor then placed an iron stake at a place one hundred feet from the top of the bank to replace the missing stake. From the testimony, the referee was justified in inferring an agreement of the parties that the new stake was located at or near the site of the original stake and that the agreed-on stake marked the southwest corner of the Hink-ley lot.
Proctor denies agreeing that the new stake was located at the position of the missing stake. Citing
Bemis
v.
Bradley,
It was within the referee’s province to sort out the conflicting testimony and assess the credibility of the witnesses. There was competent evidence to support a conclusion that the parties did agree sometime before 1966 that the new stake was being placed at the site of the original monument to serve as the southern terminus of the western boundary. Although such an agreement may not be conclusive if occupancy referable to the agreed-on boundaries has lasted less than twenty years, nevertheless the agreed-on location is strong evidence that the boundary thus located is correct; it should be set aside only if it is shown to have been settled by mistake.
*472
Bemis,
The referee determined, as a matter of fact, that the agreed-on stake marked the location of the southwest corner, implicitly rejecting Proctor’s claim that the parties mistakenly agreed on the wrong location of that corner. The possibility that the agreed-on monument may have been more than one hundred feet from the high-water mark established by Proctor does not vitiate the referee’s conclusion that the stake correctly marked the southern end of the western boundary. If there is such a discrepancy, the distance call must yield to the location of the ascertained monument. The referee did not err by establishing the westerly boundary of the Hinkleys’ lot as the line between the two stakes, in accordance with the Curtis survey.
Although we uphold the referee’s determination of the westerly boundary line, we must vacate that part of the judgment declaring the southerly boundary of the Hinkley lot. The fourth call of the deed is for a line running “east one hundred ... fеet by land of grantor to the high water mark of said lake.” The referee correctly placed the westerly terminus of that line at the agreed-on stake, discussed above, marking the southerly terminus of the westerly boundary. However, he found the nail, fixed by Curtis near a hardwood tree on the bank, to be a “monument” marking the southeasterly corner of the Hinkleys’ lot. Because the deed calls for the terminus of the line to be at the high-water mark of the lake, this finding suffers the same infirmity as his earlier finding of the location of “high-water mark” at the easterly end of thе northern boundary. Neither the nail nor the hardwood tree is referred to in the deed as a monument, and there is no competent evidence that they are at high-water mark.
Contrary to Proctor’s contention, no difficulty arises from the possibility that the distance from the agreed-on stake to the high-water mark along the southerly boundary may exceed one hundred feet: the monuments, if ascertained, must prevail over the distance call. The real difficulty in settling the southerly boundary comes from the requirement of the deed that the line run “east ... by land of the grantor.” Since the “land of the grantor” is merely what was left of Tressie Malcolm’s original parcel after she conveyed to the Hinkleys, “line of grantor” is not a monument for determining the course of this boundary: the “line of grantor” is the very line in dispute.
The call to go “east” from the stake does not help. It is obvious from even the most casual inspection of the parties’ survey maps that to establish the boundary by a line running due east from the stake would result in the Hinkleys’ having little or no frontage on the lake at all — a result both absurd and inconsistent with оther provisions of the description. In context, “east” in the Hinkleys’ deed means only “in an easterly direction.”
See Milliken v. Buswell,
Thus, the
direction
of the southerly boundary cannot be fixed solely by reference to the fourth call of the deed and monuments specified therein. It is necessary to resort to the fifth and sixth calls of the deed in order to fix, if possible, the point on the high-water mark of the lake to which the line from the agreed-on stake should be drawn to form the southerly
*473
boundary. The fifth call requires the southerly boundary to run from high-water mark to low-water mark; the sixth describes a boundary running from low-water mark “Ninety (90) feet North by the lake shore to the first mentioned bounds”, that is, to the low-water mark where the northerly boundary begins. The sixth call requires measurement along the contour of the low-water mark, for the deed is plainly intended to convey the grantor’s rights to the lake bed between high- and low-water marks; and the low-water mark must be measured along the contour in the absence of any contrary indication in the deed.
6
See Hodgdon v. Campbell,
Working backward from the last two calls of the deed, we determine that the extreme southeasterly corner of the Hink-leys’ lot is a point on the low-water mark ninety feet, measured along the contour, southerly from the extreme northeast corner of the lot where the northerly boundary line intersects the low-water mark. The fifth call of the deed requires a “continuation” of whatever line is established pursuant to the fourth call from high- to low-water mark. Hence the segment of the southerly boundary between high- and low-water marks is simply an extension of the line required by the fourth call. 7 It fоllows that the southerly boundary must be determined by running a straight line from the agreed-on stake at the southerly end of the westerly boundary to a point ninety feet along the contour of the low-water mark southerly of the point of beginning.
Neither party presented evidence bearing on the location of the low-water mark of the lake. Without such evidence it is impossible to locate the Hinkleys’ southerly boundary on the face of the earth, even though it is possible to determine what that boundary is as a matter of law. Both parties sought a judgment declaring their common boundaries. The referee correctly located their common boundary along the westerly side of the Hinkleys’ lot. However, neither party has established by competent evidence the location of the common boundary along the southerly line of the lot. The judgment of the trial court must therefore be vacated insofar as it purports to determine that boundary.
It is obvious that the referee, faced with a set of poorly drawn conveyances, did his best to bring this litigation to a sensible and fair conclusion. It is rеgrettable that his definition of the parties’ common boundaries must be partially undone because of technical error when the only practical result is to leave unsettled the ownership of a few square feet of terrain having little economic value. This Court urges counsel for both parties to try to persuade their clients to enter into some practical but formal and recordable agreement locating the boundary line in question.
Plaintiff Proctor also contests the award of $500 damages by the trial court. An award of damages by a referee
*474
acting within the scope of the reference will be held excessive only if it is not rationally supportable.
Cf. Souza v. Bangor HydroElectric Co.,
The entry is:
Judgment affirmed in part and vacated in part.
The judgment establishing the boundary line of the defendants’ property in accordance with the Curtis survey is affirmed only with respect to the westerly boundary line described in that survey.
In all other respects, the judgment establishing the boundary of dеfendants’ property is vacated.
The award of damages is affirmed.
Each party to bear its own costs on appeal.
Remanded for entry of judgment accordingly.
All concurring.
Notes
. Thibodeau’s testimony and his survey referred to the lake as the southern boundary of the Hinkleys’ lot and the disputed boundary lines as the northern and western lines of the lot. That conflicts with the directions given in both the Proctor and Hinkley deeds and with the directions used by defendants’ surveyor. In fact, the lake front runs roughly in a northeast-southwest direction. For the sake of clarity, the directions as described by Thibodeau are conformed in this opinion to the directions used in the deeds.
. Other witnesses testified that the water level is about the same today as it was in the 1940’s. The referee made no finding of the location of either high- or low-water mark, as of either 1942 or 1977.
. According to the Curtis survey, the northeast comer is marked by an iron pin near the hemlock tree on the bank; the northwest comer by an iron pin located flush with defendants’ driveway; the southwest comer by an iron stake which the parties placed by agreement; and the southeast comer by the nail set in the ground by Curtis.
. This Court has construed “high-water mark”, when used with reference to a non-tidal river, as “the line on the river bank reached by the water when the river is ordinarily full and the water ordinarily high. Not the highest point touched by the water in a freshet ... but the highest limit reached when the river is unaffected by freshets and contains its natural and usual flow; the highest limit at the ordinary state of the river.”
Morrison,
Other courts have defined the term “high-water mark” in substantially similar terms when applied to an artificial pond or non-tidal stream.
Neponset Reservoir Corp. v. Bashaw,
. See generally Browder, The Practical Location of Boundaries, 56 Mich.L.Rev. 487, 498 (1958).
. A reference to the “shore” is inappropriate for most nontidal bodies of water.
Morrison,
. If the fifth call had not specified that the line from high- to low-water mark is a continuation of the line described by the fourth call, it might be theoretically necessary to invoke the complicated rules for division among coterminous owners of the area between low- and high-water marks. See generally M. Frankel, supra note 4, at 47 — 49. At least that problem is not raised by the terms of this conveyance.