Roy Stephen Farrow v Duncan Boag

This appeal arises out of an application to HM Land Registry by the respondent, Mr Boag, for determination of the exact boundary between land in his registered title and land in the registered title of his neighbour, the appellant, Mr Farrow. The application was made on 21 June 2019 under section 60, Land Registration Act 2002 (the 2002 Act) and was opposed by Mr Farrow. It was referred by the Registrar to the First-tier Tribunal, Property Chamber (the FTT) for determination.

Mr Boag’s property comprises a house and gardens at 2 St Austin’s Grove, Sheringham, Norfolk, and is registered with title no. NK177701. Mr Farrow’s land is a proposed development site lying immediately to the west of Mr Boag’s, and is registered with title no. NK418177 under the description Land on the North side of St Austin’s Grove, Sheringham.

On 14 June 2022, after a contested hearing, the FTT (Judge Michell) directed the Chief Land Registrar to give effect to Mr Boag’s application for the exact line of the boundary between the two titles to be determined in accordance with a plan prepared by Survey Solutions as if Mr Farrow’s objection had not been made. Mr Farrow now appeals against that order with the permission of this Tribunal. It appears that the Land Registry has already acted on the order and Mr Farrow therefore seeks an order for the deletion of the relevant entries in the two titles.

At the hearing of the appeal Mr Farrow was represented by Mr Andrew Gore, who had also appeared before the FTT. Mr Boag represented himself, as he too had previously done. I am grateful to them both for their submissions.

Boundaries shown on the register of title are general boundaries only and do not define the exact line of the boundary between adjoining parcels of land, unless they have been determined under the relevant statutory procedure (section 60(1)-(2), 2002 Act). That procedure is found in rules 118-122, Land Registration Rules 2003, which enable the exact line of a boundary to be determined and recorded in the register. The Land Registry provides a commentary on the rules and an explanation of its practice in its Practice Guide 40, supplement 4.

Rule 118(1) permits the proprietor of a registered estate to apply to the registrar for the exact line of the boundary of that registered estate to be determined. Such an application must be in a prescribed form and, by rule 118(2), must be accompanied by–

The line of the boundary may already been agreed or determined by a court, in which case the registrar will give effect to the application. If not, and if the registrar is satisfied that the plan supplied with the application, or the plan and verbal description, identifies the exact line of the boundary claimed, and that the applicant has shown an arguable case that the exact line of the boundary is in the position shown, the registrar will give notice of the application to the owners of the land adjoining the boundary and will inform them of their right to object (rule 119(1)). If no objection is received the registrar will complete the application by making an entry in the register of the applicant’s title and in any registered title affecting the other land adjoining the boundary, stating that the exact line of the boundary is determined under section 60, 2002 Act. If an adjoining owner objects to the application and it is not possible to dispose of that objection by agreement, or on the basis that it is groundless, the registrar will refer the matter to the FTT under section 73, 2002 Act.

A determined boundary shows “the exact line of the boundary” between two titles, and for that reason rule 118 requires both a plan showing the exact line claimed and evidence that the exact line of the boundary is as shown on that plan. An application must be rejected if the plan submitted is inaccurate or insufficiently detailed. The Land Registry’s Guidance explains at paragraph 4.4 that the plan must identify the start, end and any turning points of the determined boundary, must describe the relationship with physical features where the boundary coincides with them (for example, on which side of the physical feature the boundary runs, or through which point it passes), and must describe points of reference. By paragraph 4.4.1 any measurement shown on the plan must be accurate to +/- 10mm. These are the Land Registry’s own requirements and do not form part of the Rules, but they indicate the degree of precision required to enable particulars of a determined boundary to be recorded on the title plan. It follows that the evidence necessary to establish the exact line of a boundary must be evidence from which the boundary can be identified with the precision required to enable such a plan to be prepared.

Between any two titles to land there will always exist an exact boundary, where the land in one title ends and the land in the adjoining title begins. When a court or tribunal is asked to resolve a boundary dispute it will always do its best with the material which exists to ascertain where, on a balance of probability, the boundary lies. In Neilson v Poole [1969] 20 P & CR 909, Megarry J explained why it was appropriate to refer to evidence of the subsequent behaviour of a common vendor as an aid to the interpretation of a conveyance:

It is important to note that Megarry J was not being asked to determine the exact location of the boundary for the purpose of section 60, 2002 Act (or its predecessor, rule 278, Land Registration Rules 1925) and was not required to consider whether the evidence satisfied the requirements of rule 118 that it establish the exact line of the boundary. The boundary in Neilson v Poole was eventually ascertained from the verbal description in the conveyance which was found to be consistent with evidence of a boundary agreement between the vendor and purchaser.

It will not always be possible to identify a boundary line with the precision required to enable it to be determined, because the evidence required to demonstrate exactly where the boundary lies may not be available. This was acknowledged by the Upper Tribunal (HHJ Dight) in Murdoch v Amesbury [2016] UKUT 3 (TCC), at [96] (referring to Megarry J’s exhortation to “produce a decisive result”):

St Austin’s Grove is a private road running east from Curtis Lane, which itself runs north from Beeston Common. The disputed boundary was created by an Indenture of 26 July 1913 (the 1913 Indenture) by which Sheringham Estates Ltd (the Company) conveyed part of its St Austin’s Estate at Beeston Regis in Norfolk (the Estate) to Juliette Helene Comte. The boundary runs north from St Austin’s Grove terminating at a passageway running east from Curtis Lane. The passageway proves access to a garage at the rear of Mr Boag’s property.

In 1911 the Cremer family conveyed land on the east side of Curtis Lane to the Company for the development of the Estate. The land was conveyed by reference to Ordnance Survey field numbers shown on a plan which gives no indication of there having been any building on the land at that time The first house built by the Company was on the north side of St Austin’s Grove at its junction with Curtis Lane. The house was called St Austin’s, and is now called 5 Curtis Lane. Although it appears to have been completed by 1912, it was retained by the Company until 1921.

Mr Farrow’s land was originally part of the grounds of 5 Curtis Lane and is located to the east of the house and its formal gardens.

Long before the sale of 5 Curtis Lane the Company conveyed a plot of land immediately to the east to Mlle Comte by the 1913 Indenture. That plot was described in the 1913 Indenture as land “with the dimensions (be the same little more or less) boundaries and abuttals thereof more particularly delineated on the plan drawn herein and thereon coloured pink.” The plan is shown below.

The plan drawn on the front of the 1913 Indenture is small, diagrammatic and not to scale. It shows the plot as having straight boundaries on all sides. The dimensions shown include measurements of 77’ along its southern boundary to St Austin’s Grove and 77’ across its northern boundary where it abuts the passageway from Curtis Lane. The western boundary is marked as 165’6”. The plan also marks the southern boundary of 5 Curtis Lane as “about 127’0””, that being the distance from Curtis Lane to the southern end of the boundary which is now in dispute.

The land comprised in the 1913 Conveyance is now the western part of Mr Boag’s property at 2 St Austin’s Grove. The house which now stands there had been built by 1919 and named St Hugh’s Cottage. In that year the Company conveyed a further plot of land to the east of St Hugh’s Cottage to Mlle Comte. That parcel now comprises the eastern part of Mr Boag’s property.

On 16 June 1921 the Company conveyed 5 Curtis Lane, including Mr Farrow’s land, to Francis George Joseph. The 1921 Conveyance includes a scale plan on which the southern boundary of 5 Curtis Lane is marked as being “132’ or thereabouts”. This is the same boundary as had been marked as “about 127’” on the 1913 Indenture plan of its immediate neighbour at 2 St Austin’s Grove (there with the additional qualification “be the same little more or less”).

The FTT found that there was no evidence that any physical feature corresponding with the line of the western boundary shown on the 1913 Conveyance plan existed on the land at the time the 1913 Indenture was executed. Nor was there any evidence that the plot had been staked out or that trees which were present when Mr Farrow acquired his land in 1988 had been there in 1913.

The Judge inspected the land and his decision includes a description of the boundary as it existed at the time of his visit. The land slopes upwards from St Austin’s Grove and on Mr Farrow’s side of the boundary it has been terraced in recent years in preparation for development. A low concrete block wall runs north from St Austin’s Grove for a short distance in the vicinity of the disputed boundary and for about one fifth of its length. The concrete block wall is perpendicular to a section of flint wall along the southern boundary of Mr Farrow’s land. It is not known when the wall was constructed, and although it is assumed that it was built by owners of 5 Curtis Lane who may have intended to build it on their own land, the FTT placed no weight on it as indicating the true position of the disputed boundary in that location as the builder of the wall may have been incorrect about where that line was. There is also a mature beech tree halfway along the disputed boundary. Vegetation which previously existed at the northern end of the boundary was removed by Mr Farrow in 1997 along with the remnants of old wire fencing.

The boundary claimed by Mr Boag is shown on the plan prepared by his surveyors, Survey Solutions, which accompanied his application. The boundary is shown as a straight line running from points marked A and B respectively at its southern and northern ends. Both points are described by Ordnance Survey coordinates and by metric distances to fixed points on adjacent buildings, certified by the surveyor as accurate to the required degree.

The line A-B does not coincide with the concrete block wall at any point. Point A is shown as lying 6cm to the west of the southwestern corner of the wall which then runs north at a slight angle to the claimed boundary line.

The surveyor who prepared the application plan was not called to give evidence, nor did he provide a statement explaining why he had plotted points A and B in the locations shown. I assume he did so on Mr Boag’s instructions. The plan does not show a distance along the combined southern boundary of Mr Farrow’s land and 5 Curtis Lane but the FTT recorded that point A was “said to be 127 feet from Curtis Lane”. I take that to be a reference to Boag’s case and to a second plan which he prepared and submitted in support of his application.

In a letter of 28 October 2019 objecting to the application Mr Farrow’s solicitors asserted that the boundary between the two properties was straight and ran perpendicular to the southern boundary of Mr Farrow’s land along the eastern face of the concrete block wall for its full length before continuing in the same direction until it reached the northern boundary. That position was maintained in Mr Farrow’s statement of case to the FTT dated 25 February 2021. At the hearing before the FTT Mr Farrow argued for a slightly different line which had been plotted by his surveyor, Mr Plumb. That line initially followed the line of the concrete block wall (which Mr Plumb said was “helpful to inform the southern boundary division”), but then turned at a slight angle towards the northeast. Mr Plumb had placed no weight on the 1913 Indenture plan but had relied on a 1959 conveyance of St Austin’s and the 1988 conveyance of Mr Farrow’s land. The FTT rejected Mr Plumb’s suggested boundary because it was not a straight line and so was inconsistent with the line shown on the 1913 Indenture plan.

The FTT considered a number of issues in its decision but the only challenge in this appeal is to its conclusion that the application plan shows the exact line of the boundary. As to that issue, since the disputed boundary was created by the 1913 Indenture, and since neither party had submitted that the position of the boundary had changed since its creation, the FTT directed itself that the issue depended on the construction of the 1913 Indenture.

The FTT addressed the issue at paragraphs 30 to 34 of its decision, as follows:

The statement in paragraph 32 that there seemed to be no real dispute about the position of point “A”, referred to a later discussion of Mr Plumb’s evidence. That discussion came in paragraph 45 of the decision when the Judge referred to a line marked “A-B” on the plan relied on by Mr Plumb and said: “He does not state in his report what point “A” is but it appears to be the point Mr Boag took as “A” on his determined boundary plan.” The Judge then observed that Mr Plumb had not relied on the 1913 Indenture or the 1921 Conveyance of 5 Curtis Grove but had taken a dimension of 132 feet for the St Austins Grove frontage of 5 Curtis Lane and other dimensions from a 1959 conveyance “those being dimensions copied from the 1921 conveyance”.

The appellant was granted permission to appeal on two grounds:

The first ground of appeal is expressed in unhelpfully tendentious terms which obscure the point which the appellant relies on. It suggests, wrongly, that the Judge proceeded on the basis that it was not necessary that he be satisfied that the boundary line proposed in the application accurately represented the line created by the 1913 Indenture. It would be very surprising indeed if the Judge had proceeded on that basis and I can find nothing in the decision to suggest that he did.

Mr Gore submitted that the Judge had failed to ask himself whether the evidence supplied in support of the application established with the required degree of precision where the boundary between the two titles lay. He had referred to Megarry J’s warning in Neilson v Poole that the court could not leave the plot conveyed fuzzy at the edges but had paid insufficient attention to the fact that the application was to determine the boundary and not simply to identify a general boundary, and that rules 118 and 119 require that the boundary shown in the register should remain “fuzzy” unless there is evidence capable of establishing exactly where it lies.

Mr Gore cited several examples of tribunals refusing to determine a boundary because they were not satisfied that the evidence established its location with the required degree of accuracy (including the decision of Morgan J in Lowe v William Davis Ltd [2018] UKUT 206 (TCC)). He also referred to Witt v Woodhead [2020] UKUT 319 LC as an example of the precision required in a determined boundary application; in that case the Tribunal (Judge Cooke) rejected an application for a determined boundary because there was a discrepancy of 4cm between the boundary shown on the application plan and the boundary established by the evidence.

As it was presented orally the first ground of appeal was simply an introduction to the appellant’s main complaint which was that the available evidence was incapable of establishing the exact line of the boundary. That complaint forms the second grounds of appeal.

Although both parties relied extensively on the conveyancing history after the creation of the disputed boundary in 1913, the Judge cannot be criticised for approaching the key question as one of interpretation of the 1913 Indenture. In doing so he directed himself by reference to what was said by Carnwath LJ in Ali v Lane [2006] EWCA Civ 1532, at [36] and [37]:

In this case, as Mr Gore acknowledged, there was no extrinsic evidence with any real probative value. The Judge was right to reject Mr Plumb’s suggested boundary both because it did not follow a straight line for its full length and because it was based on relatively recent conveyancing material and physical features which it was not suggested had been present in 1913. The 1921 Conveyance could not assist since it was not evidence of any consensus between the parties to the 1913 Indenture and may not have been consistent with the case being put forward on either side. The only material the Judge had on which to determine the exact line of the boundary was therefore the 1913 Indenture plan.

The Judge based his analysis on the document and on what was known about the condition of the boundary in 1913. In paragraph 30 of the decision he identified the key information which would have been available at that time as the three features which could be deduced from the Indenture plan, namely, that the distance of the southern end of the boundary from Curtis Lane was “about 127 feet”, that the boundary was perpendicular to the southern boundary of 5 Curtis Lane, and that the boundary was a straight line. He acknowledged that the first of those pieces of information was “not expressed to be an exact measurement” but he nevertheless felt driven to use it to fix the southern end of the boundary because, he said, it was the only information that a person standing on site with the Indenture in their hand would have to fix the southwestern corner point of the land conveyed.

The Judge’s assumption that the purchaser in 1913 would have had no guidance as to the location of the boundary of the land they had acquired other than what could be deduced from the plan is questionable. It is certainly the case that “there is no evidence of any physical feature in this location at the date of the Indenture” but I do not think that the absence of evidence justifies an assumption that nothing had been done to identify the boundary (as occurred in Neilson v Poole) such as inserting a peg or marker at the point which was taken to leave 5 Curtis Lane with a southern boundary of “about 127 feet”. Nor can it be assumed that no feature marked the other end of that southern boundary where the Company’s land met Curtis Lane and from which a measurement could have been taken. The most that can be said is that the only material about the position of the boundary which is now known to have been available to the person standing on site with the Indenture in their hand is contained in the plan, and that any temporary features which might have been present to assist in that exercise no longer exist. As a result, the task of interpreting the 1913 Indenture and identifying exactly where the boundary it created lies might now be more difficult than it was in 1913.

Mr Gore submitted that it is simply not possible to tell exactly where the boundary created in 1913 runs on the basis of the information now available.

Mr Boag supported the FTT’s determination although he acknowledged that the Judge had not arrived at his conclusion in the way that he had done. Mr Boag’s case was based on what he said was the consistency of conveyances of the parcels comprised in 2 St Austin’s Grove by the Company and its successors after 1913 with his case that the southern end of the boundary was at point A on the Survey Solutions plan and not at a point 5 feet further to the east, as it would have been if the plan on the 1921 Conveyance of 5 Curtis Lane was accurate. He emphasised that point A on the application plan had been plotted by Survey Solutions who (contrary to Mr Gore’s suggestion) had not been directed by him where it should be located.

The difficulty I have with Mr Boag’s submissions is that there was no evidence from Survey Solutions about what they had done to plot point A, or why they had done it. It did not coincide with the most obvious physical feature on the ground, the end of the concrete block wall, so it must have been arrived at by measurement from some point. But what that point was, why it had been chosen, and what measurement from it arrived at point A have never been explained.

The Judge appears to have been untroubled by exactly what point A was intended to represent, because he understood there to be no disagreement between the parties about its location or that it is the point where the disputed boundary meets the southern boundary of each property. But that suggested consensus certainly did not exist on the parties’ written cases. Mr Farrow maintained his original position that the eastern face of the concrete block wall formed the boundary between the neighbouring properties. That was also the position adopted by Mr Plumb in his report in which he described the wall as being “helpful to inform the southern boundary division”. The Judge correctly understood Mr Plumb to be depicting the end of the wall as the location of the boundary at the southern end, and he said so in paragraph 5 of the decision (“the line that Mr Plumb, the expert witness called by Mr Farrow, identifies as the boundary … runs along the line of the concrete block wall to the northern end of that wall but then turns at an angle …”). It is true that Mr Plumb did not say precisely where on the wall point A was located, nor did he provide exact coordinates, but there is nothing in his report to suggest he intended point A to lie 6cm to the west of the southwest corner of the wall as it was shown by Survey Solutions on the application plan and as the Judge described it in paragraph 4 of the decision. Nor did the Judge suggest that Mr Plumb had changed his position during his oral evidence or give any reason for rejecting the proposition that the southern stretch of the boundary was marked by the wall.

It was only because the Judge understood that there “seems to be no real dispute about the position of point A” that he was able to determine the line of the boundary from the other two pieces of information provided by the 1913 Indenture plan, that the boundary was perpendicular to the southern boundary of 5 Curtis Lane, and that it was straight. Mr Gore submitted that in fact the southern boundary of 5 Curtis Lane was not itself a straight line and that the 1913 Indenture plan was schematic and insufficiently precise to enable the angle to be measured. He also submitted that the application plan did not show the boundaries to be perpendicular but at a slight angle. It is not possible to verify any of these propositions from copy documents using the naked eye, and I disregard them. But it is not possible to disregard the uncertainty over the position of point A.

I agree with Mr Gore that the exact line of the disputed boundary cannot be identified from the 1913 Indenture. The text of the document does not purport to define the boundaries of the land with precision, describing it as having “the dimensions (be the same more or less) … more particularly delineated on the plan”. The plan is not to scale, so only the dimensions marked on it can be of any assistance. While the dimensions of the land itself are recorded without any further qualification, the same is not true of the only reference point provided from which those relatively precise measurements must be taken; that point, where the disputed boundary meets the southern boundary of 5 Curtis Lane, is identified only as being “about 127’0”” from the corner of St Austins Grove and Curtis Lane. Not only is that measurement imprecise, but the critical starting point from which the measurement is taken is not defined by reference to any feature the position of which can now be located with confidence. Just as there is now no evidence of any physical feature having been present at the disputed boundary in 1913, so there is no evidence of any feature at the south-western corner of 5 Curtis Lane, where Curtis Lane and St Austins Grove met. I was told that it was common ground before the FTT that Curtis Lane was an unmade road in 1913, but even if it was clear at that time where the Company’s Estate ended and the road began, there is no way of identifying that point now. Whatever may have been apparent to a purchaser in 1913 all that can now be said is that the southern end of the disputed boundary is about 127 feet, more or less, from a point the exact location of which is unknown.

In my judgment there was insufficient evidence from which the FTT could find the exact location of the disputed boundary. In those circumstances it was impossible to ascertain whether the boundary line shown on the application plan was the exact line of the boundary and the application to determine the boundary should have been refused.

Before the FTT there was no application from Mr Farrow to determine the boundary in the location he advocated. Nor is this a case in which the FTT, or this Tribunal, could determine that the exact position of the boundary is in a different location from that shown on the determined boundary application plan (as it has power to do where the evidence permits: Bean v Katz [2016] UKUT 168 (TCC), at [24]-[27]; Lowe v William Davis Ltd [2018] UKUT 206 (TCC), at [51]-[55]). For the time being at least the boundary shown in the parties’ respective titles must therefore remain a general boundary.

For these reasons I allow the appeal and direct the Chief Land Registrar to delete the entries which have already been made to give effect to the FTT’s decision (entry 6 in the Property Register for Title no. NK177701 and entry 5 in the Property Register for Title no. NK418177).