Deeds, plans and boundary features


On this page we are considering the following question: "Who owns the disputed land according to 'the deeds'?". That person is commonly known as the “paper owner”. A number of factors will be looked at but as will be seen the key point is to find the original document that divided the two pieces of land. It will then be necessary to go out on site with that document, which will hopefully have a plan attached - albeit more likely than not an inadequate one, and look at the land and try to work out what the position was at the time that the land was divided into two. This is a detailed article covering the following:

  • The starting point - finding the original document that divided the two pieces of land.
  • The general boundaries rule.
  • Use of OS plans in boundary disputes.
  • Measurements
  • Plans and parcels clauses
  • "For identification purposes only"
  • Conflicts on plans and documents
  • Extrinsic evidence.
  • Features on the ground

All sorts of documents will be produced in boundary disputes but it cannot be stressed too highly that the key document is that which divided the land into two portions now held by each party. The reason for this is really fairly obvious: A person cannot convey land that he has already conveyed to some other person. It is sometimes necessary to go back many years and through many transactions to find the original owners of both pieces of land:

“The first resort in the event of a boundary dispute is to look at the deeds. Under the old system of unregistered conveyancing, this means the chain of conveyances and other instruments, going back beyond the period of limitation, which demonstrates that the owner’s title is in practical terms secure as against adverse claims. These conveyances will each identify the subject matter in a clause known as the parcels clause, which contains the description of the land. Sometimes it is no more than a reference to the land conveyed by an earlier conveyance, which will then have to be consulted.” (Alan Wibberley Building Ltd v Insley [1999] 2 All ER 897 (HL), Lord Hoffman at 901, 904j).


The conveyancing documents are only conclusive as to paper title where it is possible to find the first conveyance. If both plots of land were always in separate ownership those documents will be of no great assistance. Intrinsically, they will be no more important than any other piece of evidence that is before the court.


General boundaries rule

The rule

The original document that divided the land into two may be an old fashioned "conveyance" or "deed" that divided the land before it was registered. Or, it may be a Land Registry transfer. The parties to a boundary dispute often refer to their Land Registry documents as their "deeds" and rely heavily on the Title Plans to support their respective contentions as to the true position of the boundary. However, it is extremely important to understand that Land Registry plans are not intended to show precise boundaries. They are only intended to show "general boundaries". This flows from s60 of the Land Registration Act 2002:

(1) The boundary of a registered estate as shown for the purposes of the register is a general boundary, unless shown as determined under this section. (2) A general boundary does not determine the exact line of the boundary.

For a case where the rule was discussed - although on pretty complicated facts - see Drake v Fripp [2011] EWCA Civ 1279. The case re-emphasises that the boundary line on a title is only a "general boundary" and is not intended to show the precise boundary between the two properties.

Getting the exact line determined

Although the normal position is that a Land Registry title plan only shows a general boundary it is possible to have the "exact line of a boundary" determined. This is provided for in s60(3)

    (3) Rules may make provision enabling or requiring the exact line of the boundary of a registered estate to be determined and may, in particular, make provision about-
      (a) the circumstances in which the exact line of a boundary may or must be determined, and (b) how the exact line of a boundary may be determined (c) procedure in relation to applications for determination, and the recording of the fact of determination in the register or the index maintained under section 68.

Rules have been made under this section: Land Registration Rules 2003, rules 117 to 122 (as amended). Some key points arising out of those rules are as follows:

  • The application is made using Form DB and must be accompanied by a plan, or a plan and verbal description “identifying the exact line of the boundary claimed” and must be supported by evidence (r118).
  • If the registrar is satisfied with the application in various respects he must give notice of the application to the adjoining land owners; but need not do so if the evidence consists of an agreement in writing with the adjoining owners or there is a court order determining the line of the boundary (r119). (Once a boundary dispute is settled, making an application under this rule is probably the best way of ensuring that it is clearly binding on both parties and their successors in title. Such clarity will also make it much easier to sell the property).
  • Where the exact line of part of the boundary of a registered estate is determined “the ends of that part of the boundary are not to be treated as determined for the purposes of adjoining parts of the boundary the exact line of which has not been determined. (r121).


Ordnance survey plans

The plan attached to the document that divided the two pieces of land may be an ordnance survey plan and this may well seem to be very accurate. However, there are some important to things to remember about such plans. Firstly, they are intended to show features, not boundaries. The surveyors who draw up the plans are not interest in where the boundary lies between the two bits of land. They are interested in the features on the land. Secondly, the thickness of a line can often represent several metres on the ground. The width of a line on an OS map at a scale of 1/1250 represents about 0.3metre and at a scale of 1/2500 the same line represents about 0.6 metre on the ground. This difficulty is recognised by the courts in all boundary disputes:

“… if a plan is intended to control the description, an OS map on a scale of 1:2500 is worse than useless. The plan or other drawing bound up with the deed must be on a much larger scale so that it clearly shows with precision where the boundary runs” (Scarfe v Adams [1981] 1 All ER 843, per Cumming Bruce at 845h).

“… there are considerable difficulties in the way of Mr Willsher's case …. First, it is to be noted that he relies heavily, though not entirely, on what may be gleaned from the plans: in particular the Ordnance Survey maps … However .. the Ordnance Survey maps do not purport to fix or record the legal boundary between adjoining plots of land. They will mark features, such as hedges or fences, which may well in fact reflect a legal boundary, but the map will simply indicate a line down the middle of such a feature whether the boundary is in the middle or not. They are drawn to a scale of 1:2500 and the line so drawn may be out of true by +/- 2.3 metres …. Mr Powell's evidence [the single joint expert], accepted by the judge ….was that neither the parcels clauses in the conveyances, the deed plans or the Land Registry plans could "accurately transpose the true position of boundaries onto the ground". I take the judge to have meant that such materials (at least in the circumstances of this case) are not definitive or conclusive as to the precise boundary line. …In all these circumstances the Ordnance Survey maps offer, in my judgment, an uncertain guide as to the precise boundary line.” (Willsher v Scott [2007] EWCA Civ 195 Laws J at para 18 and 19).

Parties to a conveyance can of course choose to use such a map to mark the boundary and if so they will be held to what the map shows: Fisher v Winch [1939] 1 KB 666. The following passage was cited from that decision:

    “Of course, the fact that the boundary is shown in a particular place on an ordnance map is in itself no evidence of what the true boundary is as between the parties, but where the party's title is derived from a document which refers to the ordnance map, it is necessary to look at the ordnance map and ascertain where the boundary shown on that map is truly positioned." (see para 30)

However, even in these circumstances there can be “difficulty in ascertaining on the ground what is the feature the map purports to show, for the purpose of understanding what it tells about the contractual boundary. As to that, the Ordnance Survey maps are subject to the uncertainties to which I have already referred.”



When looking at measurements on a plan it is necessary to be very careful. Does it seem that the draughtsman was seeking to specify the length of the boundary? Or, was he specifying the length of a particular feature on the ground? In the latter case he may have been wrong or the features may have moved. It is necessary to be very cautious. The most certain measurements are those that are expressed in relation to a building – if the building was present at the time of the conveyance and is the same now as it was then.


Plans and "parcel clauses"

The amount of importance to be attached to a conveyance plan depends on the words used in the document that refer to it. The first place to look is the "parcels clause". If the conveyance / deed / transfer states that the land to be conveyed is that shown “on the plan annexed hereto”, the plan is highly important. In some cases the court will attach substantial importance to the plan even where it is not specifically referred to (e.g. Dunning & Sons v Syke & Sons Ltd [1987] 1 All ER 700). In every case it is a matter of construction of the deed.


"For identification purposes only"

The phrase "for identification purposes only" can mean that the plan is not as important as other parts of the conveyance and, as will be seen from many comments in a number of cases referred to below, it is necessary to be very cautious about plans marked as such. However, it does not necessarily mean that the plan cannot be referred to for the purposes of determining the boundary: Wigginton & Milner Ltd v Winster Engineering Ltd [1978] 1 WLR 1462 at 1473G):

“When a court is required to decide what property passed under a particular conveyance, it must have regard to the conveyance as a whole, including any plan which forms part of it. It is from the conveyance as a whole that the intention must be ascertained. To the extent that the conveyance stipulates that one part of it shall prevail over another part of it in the event of there being any contradiction between them in the ascertainment of the parties intentions, the court must of course give effect to that stipulation. So if the conveyance stipulates that the plan shall not control the description of the parcels, the court must have due regard to that stipulation; but insofar as the plan does not conflict with the parcels I can see no reason why, because it is described as being “for identification only”, it should not be looked at to assist in understanding the description of the parcels. The process of identification is in fact the process of discovering what land was intended to pass under the conveyance, and that is the precise purpose which the plan is said to serve. Accordingly so long as the plan does not come into conflict with anything which is explicit in the description of the parcels, the fact that it is said to be for identification purposes only” does not appear to me to exclude it from consideration in solving problems which are left undecided by what is explicit in the description of any parcel”.

Some other phrases however really don't help much. The phrase “for identification purposes only more particularly delineated” is particularly confusing in that the first half of the phrase might be said to imply that the plan does not define boundaries but the second half of the phrase suggests that it does. In such a case “the parcels clause gives absolutely no help in defining boundaries of the land conveyed” (Affleck v Shorefield Holidays Ltd [1997] EGCS 159). However, that does not mean that the plan should not be referred to. The plan in Affleck was carefully drawn and on a large scale. The CA held that the judge made “a serious error in not deriving assistance from the plan”.


Conflict between plans and body of document

Smith v Royce Properties Ltd [2001] EWCA Civ 949 provides an example in the field of property law where there was a conflict between a plan and the relevant clause of an option agreement. The case concerned two parcels of land: OS 0062 (which was very large) and OS 0052 (which was very thin). The large plot was stuck behind the thin plot. On the ground it was impossible to distinguish between the two. Unless the option included the thin plot there would be no access to the large plot and the option agreement would be useless. However, the option agreement itself only referred to OS 0062. The CA nonetheless had regard to the land registry plan which referred to both plots to contradict the clear words of the option clause. (The original plan was missing). To do otherwise would not make commercial sense. Indeed it was “inconceivable that [the parties] intended that the option should only apply to 0062”. Tuckey LJ:

“I do not think the judge’s finding about the plan can stand. He should have given greater weight to the filed plan. As the only plan in existence, he should have started with the presumption that it did accurately reflect what was show on the plan attached to the conveyance, not only because it was prepared by the Land Registry, but also because that is what one would have expected it to show. Having therefore concluded that there was an ambiguity between the wording of the option and the plan, how should the judge have resolved it? He does not appear to have taken account of the commercial context to which I have referred. Common sense compelled the conclusion that the parties intended the option to cover the whole field. If the judge did feel compelled to decide that the wording prevailed over the plan because of the nineteenth century cases which were cited to him, I think he was mistaken. Where there is a conflict of this kind I think the modern approach is well summarised in

Lewison: Interpretation of Contracts (2nd ed., 1997) at paragraph 10.07 which says:

      ‘Whether a plan controls a verbal description or a verbal description controls a plan is a question of construction of the particular conveyance. There is no presumption either way.’"

In Cook v JD Wetherspoon plc [2006] EWCA Civ 330 the relevant conveyance contained a plan that stated that the property to be transferred was "defined on the attached plan and shown edged red". The problem arose because the relevant strip of land was stated on the plan as having a width of 40 feet, whereas the scaled width of the strip off the same plan was 30 feet. The claimant argued for the scaled off measurement; the defendant for the stated measurement. It was held that in these circumstances neither measurement prevailed on its own and that in accordance with normal principles it is necessary to have regard to the physical features on the ground at the time of the conveyance (see further below). Sir Martin Nourse:

    “The applicable principle established by these authorities is that, where there is a conflict between (1) dimensions in figures on a plan by which the property conveyed or transferred is described and (2) dimensions arrived at by scaling off the plan, the conflict is to be resolved by reference to such inferences as may be drawn from topographical features which existed when the conveyance or transfer was executed.”

In this case, a boundary line giving the defendant 40 feet would have taken the boundary through a building that existed at the date of the conveyance. This factor (amongst some others) led the court to the conclusion that the scaled off measurement of 30 feet was the correct one.

In Taylor v Lambert [2012] EWCA Civ 3 the precise measurement in a conveyance was held not to be conclusive to determine the position of a boundary. The words used in the conveyance were:

    “All that messuage … which said premises contain an area of 553.4 square yards or thereabouts and are for the purposes of identification only delineated on the plan annexed hereto.”

However, no-one could have worked out what the correct boundary was from the measurement alone and a further survey was needed. Lloyd LJ at para 48:

    “I agree with the judge that the element in the parcels clause which must give way is the description by measurement. That part of the description is inadequate both because it was inconsistent with the other words in the description, with the plan and with the reality on the ground, and because it could not by itself identify the southern boundary as contended for by Mr Taylor. It is not saved by the introduction of extrinsic evidence which explains how it was arrived at. … It would be absurd to allow this irrelevant and out of date measurement to override the conclusion to be drawn from the other parts of the verbal description in the parcels clause, supported as it is both by the plan and by the reality on the ground at the time of the transaction.”


Extrinsic evidence

Quite apart from the factors mentioned in the above paragraph the courts have traditionally resorted to "extrinsic evidence" to construe documents that are unclear.

What is it?

Extrinsic evidence is evidence that is outside the document that conveyed the land. E.g. sales particulars, the contract that led to the conveyance, correspondence, builders plans, old maps, oral evidence, aerial photography.

When is it permitted

The general rule is that it is not permitted if the first conveyance or transfer (dividing the land) is clear as to the position of the boundary. Extrinsic evidence may only be used to determine the paper owner when the conveyance is ambiguous (Scarfe v Adams [1981] 1 All ER 843):

    "The principle may be stated thus: if the terms of the transfer clearly define the land or interest transferred extrinsic evidence is not admissible to contradict the transfer... But, if the terms of the transfer do not clearly define the land or interest transferred, then the extrinsic evidence is admissible so that the court may (to use the words of Lord Parker in Eastwood v Ashton..) 'do the best it can to arrive at the true meaning of the parties upon a fair consideration of the language used'". ( per Griffiths LJ at 851c).

In Ali v Lane [2006] EWCA Civ 1532 the court was concerned with a conveyance made in 1947. One of the items of extrinsic evidence upon which one of the parties sought to rely was a building running parallel to the boundary which was built in the 1970’s. There was evidence that the then land owner who built the building (a Mr Attridge, senior) did so “as close as he could to the boundary and not more than six inches away from it”. However, the CA held that this evidence could not be relied upon to help establish the boundary at the time of the 1947 conveyance. Firstly, the unilateral actions of the owner of one side (ie Mr Attridge) could not be relied on as binding the owner of the other (para 38); and secondly the events that took place in the 1970’s were not relevant to the intentions of the parties in 1947. (Carnwarth LJ at paras 36-38).

"The conclusion I would be inclined to draw from this review is that Watcham [ie Watcham v Attorney General of East Africa Protectorate [1919] AC 533)] remains good law within the narrow limits of what it decided. In the context of a conveyance of land, where the information contained in the conveyance is unclear or ambiguous, it is permissible to have regard to extraneous evidence, including evidence of subsequent conduct, subject always to that evidence being of probative value in determining what the parties intended.

The qualification is crucial. When one speaks of 'probative value' it is important to be clear what needs to be proved. In this case the issue concerns the line of a boundary which was fixed not later than 1947. Evidence of physical features which were in existence in the 1970s is of no relevance to that unless there is some reason to think that they were in existence in 1947, or they are replacements of, or otherwise related, to physical features which were in existence in 1947. Similarly, evidence of Mr Attridge Senior's understanding of the position of the boundary, or actions by him apparently relating to that boundary, is of limited probative value, even if admissible. Such evidence begs the questions whether his understanding of the boundary was well-founded, and if so how strict he was in observing it, particularly having regard to the disused state of the disputed land during that period.

I would add that in principle reference to the intentions of the parties means the parties to the original conveyance. Thus in Watcham the user relied on by the Privy Council was that of the Watcham family, who were the beneficiaries of the original certificate. In none of the cases reviewed above was account taken of the conduct of subsequent owners. Megarry J [in the later case of Neilson v Poole [1969] 20 P&CR 909] might possibly have been willing to go further. Where the evidence of the intentions of the original parties is unclear, long and unchallenged usage may, as he said, be 'good reason for tending to construe the (original) conveyance as having done what the parties appear to have treated it as having done'

I do not read that as necessarily confined to long usage by the original parties. We need not decide whether that is a permissible extension of the Watcham principle. It would only apply if there were evidence of a long period of acceptance of a specific boundary by a succession of parties on both sides of the boundary. That is not this case. The unilateral actions of the owner of one side (in this case Mr Attridge) could not be relied on as binding on the owner of the other."

Note again that it is necessary to go back to the original conveyance; ie the original conveyance that established the boundary between the two parcels of land. In this case the 1947 conveyance that was referred to was not the original conveyance (see para 8 where Carnwarth LJ said "we do not know when or by whom the disputed boundary was first established. The earliest conveyance we have relating to the disputed land was dated 25 March 1947). It is therefore even possible to suggest that any reference to the 1947 itself was largely irrelevant although the judge at trial did say that he thought that the surveyor who drew the plan to the 1947 conveyance measured the land with great care and having had the advantage of considering the original conveyances and the plans".

In Owen Ernest Wood v Hudson Industrial Services Limited [2012] EWCA Civ 599 the plan attached to a deed of gift of land was inconsistent with the wording of the gift. There was however sufficient extrinsic evidence for the court to determine the extent of the land conveyed. The annexed plan to the deed of gift, which was stated to be for identification purposes only, showed an area of 3¼ acres. However, the description in the parcels clause described an area of “one acre or thereabouts”. This was only part of a series of transfers of land of around 22 acres. The question for the court related to what was the area of land transferred to D. It was held that there was a substantial amount of evidence that could be drawn upon to determine the scope of the land conveyed; including the physical layout of the land, the use of the land at the time of the conveyance and a contemporaneous application for planning permission. All this evidence indicated that the parties intended to deal only with the yard in the deed.



Where the documents do not show the boundary presumptions can sometimes help but they are just that, presumptions, and may be rebutted by evidence.


Features on the ground

Unfortunately, the search for deed plans can often turn out to be a bit of a waste of time because when found the original conveyance will not clearly show the boundary! Nor will there be any presumptions to assist. Almost invariably one has to resort to features on the ground, which may possibly assist in showing how previous owners have sought to define the boundary; either topographical features or perhaps something like the existence of an old fence. That is not to say that one ignores the plan altogether. It is necessary to take the plan out onto site and to look at it in the context of the features on the ground – remembering always that it is necessary to consider the position at the time that the two pieces of land were divided into two. To quote Lord Hoffman in Wibberley:

    “The parcels may refer to a plan attached to the conveyance, but this is usually said to be for the purposes of identification only. It cannot therefore be relied upon as delineating the precise boundaries and in any case the scale is often so small and the lines marking the boundaries so thick as to be useless for any purpose except general identification. It follows that if it becomes necessary to establish the exact boundary, the deeds will almost invariably have to be supplemented by such inferences as may be drawn from topographical features which existed, or may be supposed to have existed, when the conveyances were executed.” (p899b).

There have been a number of recent cases in which the relationship between the title deeds and the topography of the site has been considered (Pennock v Hodgson [2010] EWCA Civ 873; Dixon v Hodgson [2011] EWCA Civ 1612; Brown v Pretot [2011] EWCA Civ 1421; Taylor v Lambert [2012] EWCA Civ 3; Cameron v Boggiano [2012] EWCA Civ 157; Shaw v Grouby [2017] EWCA Civ 233 ). Here are some quotations from those cases. In para 12 Mummery LJ in Pennock v Hodgson said:

“Looking at evidence of the actual and known physical condition of the relevant land at the date of the conveyance and having the attached plan in your hand on the spot when you do this are permitted as an exercise in construing the conveyance against the background of its surrounding circumstances. They include knowledge of the objective facts reasonably available to the parties at the relevant date. Although, in a sense, that approach takes the court outside the terms of the conveyance, it is part and parcel of the process of contextual construction.”

And Mummary LJ in Brown v Pretot

    "1. The construction process starts with the conveyance which contains the parcels clause describing the relevant land, in this case the conveyance to the Defendant being first in time. 2. An attached plan stated to be "for the purposes of identification" does not define precise or exact boundaries. An attached plan based upon the Ordnance Survey, though usually very accurate, will not fix precise private boundaries nor will it always show every physical feature of the land. 3. Precise boundaries must be established by other evidence. That includes inferences from evidence of relevant physical features of the land existing and known at the time of the conveyance. 4. In principle there is no reason for preferring a line drawn on a plan based on the Ordnance Survey as evidence of the boundary to other relevant evidence that may lead the court to reject the plan as evidence of the boundary."


When trying to determine where the boundary lies between two pieces of land, the starting point is to find (if you can) the original document (conveyance, deed, transfer etc) that divided them into two. Hopefully a plan will be attached but it is no good looking at the plan in isolation, not least because it is unlikely to tell you precisely where the boundary lies. This may be because it is not clearly drawn, because measurements may be inaccurate or it is not clear where they are taken from. Furthermore, the thickness of a line on a plan represents a large area of land on the ground. If one is looking at a Land Registry document the general boundaries rule makes it clear that the plan is not intended to show the precise boundary. There may some 'extrinsic evidence' such as earlier plans or previous sales particulars that may help, or it may be possible to have resort to a legal presumption. However, whatever other factors there might be that might assist in determining the boundary it is always crucial to go out onto the land, with document and "plan in your hand", and to look at the features on the ground - always bearing in mind that one is trying to ascertain what those features were at the time of the original division of the piece of land.



Assuming that one can determine the precise boundary according to the above (which is rarely the case) that is not necessarily the end of the story. At some point the "paper owner" may have lost title by virtue of adverse possession or there may be something in the way of a boundary agreement which may affect the position.


Back to top

Copyright © Property Law UK