Part one in a series

What makes surveying memorable? From here in the Pacific Northwest I would list distressing encounters with tiny barbless fishhooks on blackberry thickets growing in the wasted territory along the boundaries of vacant lots. From my years in the South I remember the humid heat. Of course, these are merely unpleasantries of the field; surveying is also fun, and, to quote a prospective employer of years ago, "at least you get to eat your lunch in the woods." In the office, both the most frustrating and the most gratifying of jobs must lie in the interpretation of old, meager legal descriptions.

For this and my next column I plan to explore a certain type of confusing description, one that depends on a reference to an unrecorded plat map. The theoretical brambles can be thick along the boundaries of lots in a plat that has never been deposited in the public records but has nevertheless been referenced in the sale of land parcels. Often grantors describe such parcels both by lot numbers and by some other form of description, creating internal conflicts. Often they show appurtenances necessary for the enjoyment of such parcels only on the face of such unrecorded plats, leading to disputes over the location and extent of easements. In this installment I will address the confusion arising from boundary descriptions possibly (or certainly) created by unrecorded plats. Next time I will address easements created by such documents.

From the lawyer's standpoint, property descriptions referencing unrecorded plats cause difficulties in two related areas. One is whether the document establishes a valid property interest (meaning, for these cases, whether the interest survives the original transaction to bind subsequent parties dealing with the land) while the other concerns the extent of notice given by an unrecorded plat. To create a valid property interest you generally must use a conveyance, which requires an adequate legal description. This requirement is usually couched in terms of the statute of frauds (a concept that, strictly speaking, covers contracts to convey land-as well as certain other contracts-but which has been adopted by U.S. courts to cover the act of the conveying property itself):

"By insisting on a written deed, the statute of frauds mitigates the possibility of perjury or fraud in agreements for the conveyance of real property¿.The statute of frauds reflects a recognition that fraud or perjury is more apt to occur when an agreement is based only on the strength of a handshake, rather than on the specifics of a written instrument. In contracts for the purchase or sale of real property, the statute of frauds precludes reliance on oral testimony by requiring that a written agreement meticulously describe the location of the property, or refer to another instrument that describes the location.¿By requiring the clear delineation of property boundaries, the statute of frauds ensures a systematic and organized transfer of property." Berg v. Ting, 125 Wn. 2d 544, 886 P. 2d 564 (1995).

Surveyors frequently face deed descriptions that reference some other document, only to find that the other document does little to provide a "clear delineation of property boundaries." When that other document is an unrecorded plat, the surveyor can legitimately question whether the lines shown are truly in existence, whether they have ever been surveyed, or whether they have been changed since the original lots were created. The famous old case of Seng-felder v. Hill, 21 Wn. 371, 58 P. 250 (1899) (cited in R. Skelton, The Legal Elements of Boundaries and Adjacent Properties, §2 (2) (1930)) concerned title to a lot in a plat that a surveyor had been directed to record. The surveyor failed to record the plat before the grantor sold the lot. Heirs of the grantor contested the sale, arguing it was invalid because the conveyance contained an inadequate legal description, in that the lot was described by reference to a plat not recorded at the time of sale:

"To render a deed void for uncertain description, there must be either a fatal infirmity in the description, appearing on the face of the instrument, or an inability to apply the given description to any particular property. When the objection to the description is made upon the first ground, its sufficiency is determined by an inspection of the deed itself, without ulterior inquiry. But, before a deed will be declared void for the latter reason, all sources of inquiry which the description itself and the circumstances surrounding the parties and the conditions existing at the time of its execution naturally suggest must be exhausted in a vain effort to locate the property. Parties making a conveyance are presumed to make it with reference to the state or condition of the premises at the time, and, if the description be sufficient when made, no subsequent changes in conditions can make it invalid.... It is not the office of a description to identify the premises, but to furnish the means by which they can be identified." 21 Wn. at 378-79.

The court held that one source of inquiry was suggested by the fact that the lots shown on the plat as it was finally recorded substantially matched the lots on the unrecorded version in existence at the time of the sale. Also, the grantor was the same person whose name was on the finally recorded plat. Thus, by making what the court felt were a few simple inquiries, any person dealing with the property could discover the unrecorded plat and find a sufficiently definite location for the property; consequently the sale was valid. In contrast, the same court held in Napier v. Runkel, 9 Wn. 2d 246, 114 P. 2d 534 (1941) that a description reciting the numbers of lots in an unrecorded plat, with no accompanying metes and bounds description, was insufficient to support a sale by tax foreclosure. In the Sengfelder case a description by lot number in an unrecorded plat was satisfactory because the parties were dealing directly with the subdivider and could easily obtain a copy of the plat. In Napier a similar description, when it appeared in a court proceeding and a legal notice of foreclosure in a newspaper, did not give sufficient notice to the property owners that their property was about to be taken for nonpayment of taxes. The Napier court made these statements concerning such plats:

"The disadvantages which follow from the use of a so-called unrecorded plat are obvious. The surveys of the land shown thereon have never been checked by official authority; streets and alleys have not been regularly dedicated as such to the use of the public; and in so far as the plat itself is concerned, the verification of descriptions of land, with reference to the plat, cannot be made by statutory record. Such a plat may be lost or destroyed, there can be no official copy of such a plat, it may be changed from time to time, and there are other objections to its use....

"Of course, a person who prepares a plat, whether recorded or not, should be held to recognize a description of land referring to that plat, but an unrecorded plat would give no notice at all to one unfamiliar with such a document, as to the location of land described therein, and it might afford an insufficient basis for the execution of a deed conveying land foreclosed for delinquent taxes. An unrecorded plat, being no part of the official county records, is not constructive notice to anyone under the recording statutes, and unless the name of the plat might happen to be well known and suggest to persons seeing it a recognized locality, it would probably convey to very few people actual notice concerning the land described." 9 Wn. 2d at 251, 253-54.

From the surveyor's standpoint, property descriptions referencing unrecorded plats create uncertainty. Surveyors often apply proportionate measurement techniques to lots within blocks in subdivisions created by recorded plats, and several well-recognized principles support the practice. Do those principles apply to unrecorded plats as well? Was the unrecorded plat ever surveyed, and, if not, does the plat have any significance to the property's location? Is the creation of the several lots in an unrecorded plat a series of simultaneous or sequential events?

Finally, in some cases the property description does not refer to an unrecorded plat, but investigation reveals that one exists-should the information on such a plat affect the boundary?

The Cases
Willis v. Campbell, 500 So. 2d 300 (Fl. App. 1 Dist. 1986).

This case concerns property described as a portion of the NW 1/4 of the NW 1/4 of a section 6. A private grantor created the description while carving several parcels out of a larger holding. A north-south running fence existed along what the property owners believed to be the east line of the quarter-quarter section, beginning at a point 1,320 feet east of the northwest section corner.

The dispute arose after a surveyor subdivided the section using standard government rules for surveying such sections. Treating the NW 1/4 of the NW 1/4 of this section as a government lot, the surveyor placed the east property line along a north-south running line beginning 1,333 feet east of the northwest section corner. A second surveyor, after considerable research, later determined that the fence matched a boundary set by a survey for an unrecorded plat of part of the overall land holding:

"At trial Willis established that the original government grant of the lands at issue was the North 1/2 of the Northwest 1/4 of the section. The original federal grant did not establish the boundary between the two parcels involved in this controversy.¿The uncontradicted testimony of two daughters indicated that the fence had been built as a boundary and not as a livestock barrier¿[The second surveyor] testified that based on information he received from other surveyors there had been an unrecorded subdivision done by surveyor Cook within at least the Northeast 1/4 of the Northwest 1/4 of the section. [The second surveyor] indicated that he found certain monuments left by Cook and that the old fence line would have been consistent with the boundary if established by Cook's survey. [The second surveyor] further testified that the error in locating boundaries in this particular area appeared to be the result of the fact that surveying protocol requires that all error within an entire township be corrected in the northwest section of the township which is Section 6, the section in which the property in controversy is located. According to [him], Cook surveyed the lands as if it were a normal section having a 5,280 foot boundary when in fact the north boundary of this Section 6 is actually 5335.8 feet.¿

"[The second surveyor] was able to locate the original corners of the Cook survey and a bolt Cook used. At the corner where the description of the lots of the Cook survey began he also found some of Cook's original corners in the lots and was able to locate the west boundary line of Willis' lot 1 of Cook's unrecorded survey which is also the east boundary line of the parcel owned by the Campbells. He further stated that about one foot outside of that line there was an existing fence line but that it 'was right along on the west line' of the Willis property." 500 So. 2d at 302-03.

Other evidence linking the Cook survey to the line in question included the neighbors' treatment of the fence as a boundary and their use of a road running along side the fence to access lots carved out of one of the parcels. The court held that the second surveyor's retracement of the Cook unrecorded plat correctly determined the disputed boundary, basing its decision on principles of control by an original survey:

"The boundary line established by that unrecorded plat will control over a subsequent resurvey purporting to accurately locate the boundary line, if the original survey lines can be located and determined.¿In making a resurvey, the question is not where an entirely accurate survey would locate the lines, but where did the original survey locate such lines.¿The object of a resurvey is to furnish proof of the location of the lost lines or monuments, not to dispute the correctness of or to control the original survey. The original survey in all cases must, whenever possible, be retraced, since it cannot be disregarded or needlessly altered after property rights have been acquired in reliance on it." 500 So. 2d at 303

From a legal standpoint this case is interesting because the deeds did not reference the Cook unrecorded plat, yet the survey of that plat controlled the boundaries. Many attorneys would argue that Cook's monuments were irrelevant unless called out in the deeds. This court's contrary conclusion, however, is not surprising given the amount of evidence that the Cook survey did in actuality display the intent of the original common grantor. The key is not so much whether the deeds are technically correct as whether the evidence exists to correct technical deficiencies in the deeds as you find them. Obviously, in order to preserve the status quo in a situation such as this, the landowners would be well advised to put such evidence in the record. From a surveyor's standpoint this case illustrates very well the lengthy research necessary to perform a justifiable resurvey. It would be all too easy to dismiss the Cook survey as simply "wrong"; however, further investigation might reveal that it nevertheless controls the boundaries.

Clausing v. Kassner, 60 Wn. 2d 12, 371 P. 2d 633 (1962).

In 1939 a landowner hired a civil engineer to subdivide his property along Puget Sound. The plat of the subdivision delineated tracts 14 and 15, of which tract 15 was sold first. All of the conveyances (except for the first, which was later corrected) described the parcels both by tract numbers and by metes and bounds descriptions prepared by a title insurance company, as the plat was never recorded.

"The disputed property is a triangular tract. The southern leg extends westward 374.94 feet from the apex. The western leg, bordering on Puget Sound, is 12.7 feet.¿

"The principal question is whether the triangular strip belongs to Tract 14 on the north, owned by defendant, or to Tract 15 on the south, owned by plaintiff. Tracts 14 and 15 are designated on a subdivision plat, unrecorded at the time the tracts were conveyed by the common grantor. Defendant claims the southern leg of the triangle as the southern boundary of Tract 14 because the metes and bounds description in the 1957 conveyance to him describes the southern leg of the triangle. The plaintiff claims the northern leg of the triangle as the northern line of Tract 15 on two theories: (1) the line was established by survey, marked by stakes by the common grantor, and subsequent conveyances were made with reference thereto and with knowledge thereof; and [(2) adverse possession]." 60 Wn. 2d at 13.

The defendant relied mostly on the metes and bounds description and a surveyor's affidavit locating the lines of that description. The plaintiff relied on affidavits from former owners, who swore that the properties were purchased and sold in reliance on stakes set by the surveyor of the unrecorded subdivision plat. Furthermore, prior owners had improved the properties with landscaping, sprinklers, shrubs and a retaining wall built along the staked boundary. One of the prior owners had built a sea wall terminating at the west end of the staked line. The court made short work of this dispute, determining with little discussion that the line staked along the unrecorded tract boundaries controlled over the metes and bounds description:

"The defendant does not dispute the rule of law that the location of a boundary line by a common grantor is binding upon the grantees and their successors in interest, who take with reference thereto; nor that the line established upon the ground by the parties is presumably the line mentioned in the deed." 60 Wn. 2d at 14-15.

Again, this case is of interest to surveyors in its illustration of the evidence the court used to support its conclusion. The deeds described the disputed line in two inconsistent manners, as the metes and bounds line apparently did not match the tract line staked by the surveyor. From that scenario the court easily determined that the staked line controlled the ambiguity; however, you should take note that the court used evidence of occupation and the landowner's actions to prove that the grantees and their successors in interest had improved their land "with reference" to the staked line.

Windsor v. Bourcier, 21 Wn. 2d 313, 150 P. 2d 717 (1944).

In this case the legal descriptions at issue also purported to describe the lands by tract numbers from an unrecorded plat as well as by metes and bounds. Unlike the situation in Clausing, however, there was a dispute over the location of the stakes set by the surveyor of the unrecorded plat. Predecessors to the litigants bought the respective tracts when the survey stakes were in place. They noted a cedar tree growing on the staked boundary, and that a stake near the public road was later removed and replaced with a power pole. By the time the litigants obtained the property the other survey stake was missing and the cedar tree had been cut down, but its stump remained. The predecessors pointed out the power pole and the cedar stump to the litigants and stated that those items marked the boundary.

In an unusual twist for these cases, one of the litigants hired the surveyor who originally staked the unrecorded plat to survey the boundary. Of course, the metes and bounds descriptions for these tracts merely repeated the dimensions from the surveyor's unrecorded plat. Thus he set the stakes for his resurvey based on these same dimensions and reported that the resulting line did not run through the power pole or the cedar stump, leading to the dispute. The court held that the power pole and cedar stump controlled the line. "Here, however, the court has held the line contended for by respondents to be the true line and the agreement between the parties was merely to accept markers, other than the original stakes, as fixing the true line. The appellants had notice of the substitute markers and the absence of the original stakes.¿The situation in this case is that the common grantor surveyed and staked the land and the predecessors in interest of the parties hereto bought with reference thereto. These predecessors agreed that the line fixed by the stakes was also fixed by the pole and cedar tree. The stakes are now gone but the pole and cedar stump remain. The court found that they marked the true line as staked. This must govern." 21 Wn. 2d at 315-16.

Booten v. Peterson, 34 Wn. 2d 563, 209 P. 2d 349 (1949).

The lots involved in this dispute are 100-foot wide tracts carved out of a "government lot 3" bounded on the north by Hood Canal. The tracts were shown on an unrecorded plat named "Happy Hollow Tracts." One of the tracts in the dispute was described as "The west 100 feet of the east 520 feet of government lot 3¿also known as tract no. 8 of the unrecorded plat of Happy Hollow Tracts.¿" while the other was described as "The west 100 feet of the east 420 feet of government lot 3¿also known as tract no. 9 of the unrecorded plat of Happy Hollow Tracts."

The common grantor who first used the descriptions at issue was named Hoshors. The litigants strongly disputed whether the unrecorded plat had ever been surveyed, or whether any true copy existed. The litigants asked the trial court to set the line based on a theory that the true property line had been lost or had become uncertain, which called for a statutory procedure whereby the trial judge commissioned a panel of surveyors to inspect the property and issue a recommendation. The panel recommended the acceptance of two iron pins and a barbed wire fence as the boundary.

There was no evidence as to the origin of the pins, however, some occupation had been erected following to the line between them (this case also includes an adverse possession claim, not important for the issues I am covering in this article).

The trial court decreed the panel's line to be the boundary, and one party appealed. The Supreme Court reversed:

"The respondents vigorously contend that the evidence shows that the line fixed by the court is the exact line which was established by the Hoshors' plat of the property. This amounts to a contention that the language in the Hoshors' own deeds, following a definite and adequate description of the actual land conveyed thereby, is overridden by the added statement: 'Said land being known also as Tract No. 8 [in the one instance]' and 'Tract No. 9 [in the other] of the Unrecorded Plat of Happy Hollow Tracts, Mason County, Washington.' "In our opinion, what might be called the 'metes and bounds' descriptions in the deeds overrule the superfluous language last above quoted. It seems to be by far the most reliable information as to the location of the alleged lost boundary line. (Earlier in the case the court stated that the tract numbers were of no value in locating the boundaries 'since the plat was never recorded, and for aught we know is not even in existence.' 34 Wn. 2d at 567.)

"The respondents, however, contend that the line established by the court in its decree as the common boundary between the Bootens' and the Petersons' lands is the exact line that was actually staked by the Hoshors in platting and selling the lands in government lot 3. The evidence that it is is not very convincing. Furthermore, the respondents admit, in their brief, that the Hoshors' staking on the unrecorded plat of government lot 3 are not infallible. We quote briefly from the respondents' brief:

"'In approaching this case on the merits it helps to have in mind the chain of circumstances which led to this trouble. The trouble first started some years ago when it was found that Tract 1, Government Lot 3, was encroaching about 14 feet over into Government Lot 4 which adjoins Government Lot 3 on the west. Thereupon, several property owners at the west end of Government Lot 3 moved 14 feet east by agreement and reformed their deeds to read from the west line of Government Lot 3.'

"In other words, the Hoshors' stakes as to tract No. 1 of government lot 3 enclosed a portion of government lot 2 (sic.) which the Hoshors did not own. In our opinion, the descriptions made with reference to the east boundary line of government lot 3 by the Hoshors in their deeds are much more likely to reveal the true boundary between the lands of the plaintiffs and Mrs. Peterson than the Hoshors' staking of tracts Nos. 8 and 9 on the unrecorded plat of government lot 3." 34 Wn. 2d at 577-78.

Booten seems to run counter to the previous cases described above. Most likely this result stems from the vigorous disputes in Booten over whether Happy Hollow Tracts had ever been surveyed, or that any authentic copy of the plat map existed. As a result, the court reverted to the portions of the descriptions calling out portions, measured from the east line, of the government lot as being more "certain." This conclusion should give pause to any surveyor reading the facts, with their recitation that other portions of the government lot are measured from the west line, a classic condition for uncertainty. In any case, a major factual distinction between Booten and the preceding two cases is that in Booten the unrecorded plat is little more than a sketch map, while in the other two the unrecorded plats more obviously represent on-the-ground surveys.


In general, these cases demonstrate that the courts, when faced with descriptions apparently created by unrecorded plats, will search for the most certain way to define the property boundaries. If the evidence clearly shows that the unrecorded plat was surveyed, and that the landowners have treated their property holdings as being bounded by the surveyed lines, the court will adopt those lines as being more certain than the metes and bounds lines. If either feature is missing or particularly uncertain, the courts will use the metes and bounds description instead. The portion of this summation keying on the property owners' treatment of their properties according to the surveyed lines parallels the courts' general view of evidence from unrecorded documents affecting real property: Federal Rule of Evidence 803 (15) allows for the admissibility (over a hearsay objection) of statements contained in unrecorded documents purporting to establish or affect an interest in real property "unless dealings with the property since the document was made have been inconsistent with the truth of the statement or the purport of the document."

This search for the most certain of the competing ways to define the boundary can be used by the surveyor to predict the outcome of applying rules such as proportionate measurements to unrecorded plats. To the extent such proportioning helps the resurveyor actually retrace an original survey, and to the extent that such an original survey actually controls the metes and bounds descriptions, proportioning would obviously be appropriate. To the extent that proportioning is used only to reach some perceived "equitable" conclusion, it would most likely not be appropriate.

For example, R. Skelton, in The Legal Elements of Boundaries and Adjacent Properties, (1930) explains proportioning of excess or deficiency in a block in mathematical terms related to the probability of relocating the lot lines in their originally surveyed locations:

"Where the excess or deficiency is due to the original chain or tape being too long or too short, rigid mathematics dictate that the error be distributed proportionally among the subdivisions of the line, and where a small discrepancy is due to careless surveying, and there are no circumstances suggesting that there is a gross blunder in any part of the survey, the law of probability supports the rule." R. Skelton at § 216.

Thus where the original unrecorded plat appears to have been carefully prepared, and actually surveyed without large errors, one can support the opinion that applying proportionate measurement between found monuments of the unrecorded plat gives the most certain relocation of the platted lot lines. If the platted lot lines control (because the parties' actions and the completeness of the unrecorded plat lends the necessary certainty to overcome the metes and bounds lines) then proportioning would be appropriate.

Consequently, in a situation such as that illustrated by the Booten case, proportionate measurement would not be appropriate (even if the unrecorded plat had been proven to exist) because the Happy Hollow Tracts map contained large errors. In a similar vein, if it can be shown that the original intent of the grantor (demonstrating so-called "original intent" almost always requires some direct evidence of the original grantor's actions) was to follow the unrecorded plat proportioning would help preserve that intent. (See Skelton §216).

In Evidence and Procedures for Boundary Location, 3rd ed., (1994) the authors state that proportioning in a plat is appropriate where all the deeds reference a common instrument (the plat) and thus the rights of the various grantees are created simultaneously. C. Brown, W. Robillard, D. Wilson, at §13-9. This principle would apply to unrecorded plats in those instances that support the use of proportioning as described above, i.e., when the unrecorded plat was actually surveyed, reasonably accurate and had been apparently adopted by the property owners. In such cases the actual conveyances, which would generally be sequential, would nevertheless be bound by the simultaneous survey of the unrecorded lot lines just as in a recorded plat. On the other hand, an unrecorded plat that is essentially a sketch map of the original grantor's layout of the subdivision would have no such control, and the conveyances would be treated as sequential.