*41This opinion concerns the second part of a bifurcated trial regarding land called "Lega'oa” in the Village of Leone. The facts and procedural history of the case are stated in the opinion rendered by the Court after the first part of the bifurcated trial, Willis v. Fa'ivae, 10 A.S.R.2d 121 (1989) [hereinafter Willis I], aff’d, 12 A.S.R.2d 37 (1989).
The second hearing was held July 23-27, 1990. At this hearing the parties presented evidence with respect to the following issues:
1) The precise boundaries of "the flat land" in Lega'oa, within the meaning of To'omata v. People of Leone, 1 A.S.R. 142 (1906), and as further defined in our opinion in Willis I.
2) The dividing line between the inward half of the flat land (awarded in the 1906 decision to To'omata, Tali, and Va) and the seaward half (awarded to the chiefs of Leone).
3) The boundaries of land belonging to each of the various Leone families within the seaward half of the flat land.
4) The boundaries of such other tracts claimed by any party which, although not within the flat land of Lega'oa, are within the much larger survey offered for registration in 1982 by Tony Willis and Dora Asuega on behalf of the heirs of Va. (A number of the present parties objected to this offer of registration, and the dispute gave rise to LT No. 45-82, one of the consolidated cases now before us.)
5) Claims by any party to have acquired, by adverse possession or otherwise, an interest in land awarded to another party in the 1906 case.
7. Boundaries of "the Flat Land"
The Willis/Va plaintiffs submitted a-1990 survey of what they believe to be the flat land within Lega'oa. This survey is Drawing No. P-36-90, and is depicted by a solid black line in Plaintiffs’ Exhibit No. 1, a large composite map. To'omata, a cotenant of the plaintiffs under the 1906 decision, also submitted a map depicting the extent of his claim.
We find that the Willis survey more accurately depicts the boundaries of the flat land than does the To'omata survey. In most places in this valley the land slopes almost imperceptibly up to the foot *42of a fairly steep slope. Along the east and west sides of the valley, this steep slope generally begins at about the 25-foot contour line. Toward the narrow northern tip of the valley, the abrupt rise does not begin until a point near the 75-foot contour line. The plaintiffs’ survey follows closely, although perhaps not perfectly, the bottom edge of this abrupt rise. The To'omata survey, on the other hand, appears to reflect the present To’omata’s understanding of the boundaries he has worked out with his neighbors rather than an attempt to track the boundary of the flat land. Although evidence of such negotiated boundaries may be most helpful in determining claims based on adverse possession and related contentions, it is of little use in determining the starting point of our analysis: what is to be considered "flat."
Accordingly, we find the 1990 Willis survey to be the best evidence of the boundaries of the flat land. (Two slight exceptions are noted in Part IV of this opinion.)
II. The Boundary Between the Inland and Seaward Portions
Plaintiffs’ 1990 resurvey also draws a line attempting to demarcate the boundary between the inward half of the flat land, held to belong to Va and her cotenants, and the seaward half, held to belong to the various families of Leone who had been occupying Lega’oa before 1906. This line is drawn so as to create two precisely equal parcels of 27.875 acres each. The line runs about fifty feet to the south of a common boundary recognized by plaintiffs’ cotenant To’omata and his southern neighbors, the Fai'ivae and Le‘alaialoa/Avegalio/Aigamaua families. This relatively close correspondence, together with the recognition of even a slight margin of error in calculating the land designated "flat," suggests that the traditional boundary is the best evidence of the dividing line. (The Fai’ivae survey, one of the three that recognizes this boundary, appears to be an accurate retracing of a survey prepared in 1915, a few years after the 1906 decision whose mandate we are interpreting.) Accordingly, we accept the To‘omata/Fai‘ivae/ Le'alaialoa/Avegalio/Aigamaua boundary as the dividing line between the two halves of Lega’oa. (A slight discrepancy between the boundary defined by To’omata and that defined by Fai’ivae is discussed in connection with the Taeleifi/To'omata/Willis conflict, Part IV(A) infra.)
*43 III. Boundaries Within the Seaward Portion of Lega ‘oa
The 1906 High Court decision ordered the families of Leone who were held to be the owners of the seaward or southern portion of Leone to divide it among themselves. It appears that this was done; a number of interlocking survey maps made in 1915 present a reasonably coherent picture of how the land seems to have been divided at that time. As is perhaps inevitable with the passage of eighty years or so, however, several minor and one or two major boundary disputes exist. We discuss first the pattern of land occupation that appears to have existed in 1915. We then discuss conflicts in the present surveys, arising in most cases from subsequent changes in that pattern.
A. The Pattern of Settlement in 1915
A number of parties presented claims within the seaward half of the flat land comprising Lega‘oa. We list them in order, more or less, from south to north: Puletu, on behalf of the Puletu family and also on behalf of the heirs of Mrs. Thomas Meredith; Suafo‘a; Tuiteleleapaga; Uo; the (Roman Catholic) Diocese of Samoa-Pago Pago; Su‘a; the Heirs of Sekio Avegalio; Le’alaialoa/Avegalio/Aigamaua; Fai‘ivae; the Estate of Edward Ripley; Le‘oso; Iuli; Taeleifi.
The Diocese and the Heirs of Sekio rely on freehold land grants approved by the Land Commission of Samoa, which operated in Apia under the supervision of the then-Supreme Court of Samoa prior to the coming of the present government. (The Diocese relies on a grant to the "French Roman Catholic Mission"; the land in question was generally known as the "Sisters’ Land.") This court is bound by statute and treaty to recognize Land Commission freehold grants, and lands subject to such grants were specifically excluded from the 1906 decree dividing Lega’oa. The Ripley estate, the Meredith heirs, and the Uo family rely on surveys made in August 1915 and registered shortly thereafter in the Register of Native Land Titles. All of the remaining parties are chiefs of families of the Village of Leone, as are Uo and the current successors of Meredith and Ripley.
The five parties who are relying on long-registered land titles have submitted old survey maps depicting the extent of their holdings. Of these, all but the Heirs of Sekio have submitted recent retraces of the old surveys. Fai’ivae has also submitted an August 1915 map, made by the same surveyor who prepared the other maps but apparently never registered. The remaining parties have not submitted old maps, but all *44of these parties without exception are listed on at least one of the other parties’ old maps as having then occupied the same general area they now occupy. All things considered, the six old surveys fit together quite well, the parties have generally been able to trace them with what appears to be reasonable accuracy, and there are remarkably few conflicts.
Three composite maps submitted to the Court depict reasonably well the relationship among these ancient surveys. Plaintiffs’ Exhibit 1 unfortunately omits the Fai’ivae resurvey. It also places the Uo survey as depicted in theUo family’s recent resurvey (i.e., where the Uo family has decided it belongs) rather than in the apparent relation of the 1915 Uo survey to the other old surveys. Two other composite maps, Fai'ivae Exhibits 21 and 23, include the Fai’ivae survey but omit the Uo survey. None of the three composite maps depicts the Sekio survey (based on a grant to a person named Krause, later purchased by "Captain Allen" and then by Sekio Avegalio), because the heirs of Sekio unfortunately did not resurvey it as part of their preparation for this case.
We have found it convenient to use Exhibit 21 as a basis for our analysis of this area. It appears to be in agreement with Exhibits 1 and 23 in all important particulars. The Court has made notations on Exhibit 21 to indicate where we think the Uo and Krause/Allen/Sekio tracts were located in relation to the others already depicted in the exhibit.
We find that the Meredith, Ripley, Fai’ivae, and Diocese resurveys retrace as nearly as possible the original boundaries of the surveys on which they were based.
In particular, we note (1) that the northwestern comer of the Diocese survey was a sharp comer at the intersection of two streams, and that this feature is still present; (2) that the western boundary of the Diocese survey was the meandering course of a stream, and that the courses in that survey substantially follow the course of the present stream at the location of the western boundary of the resurvey; (3) that the northern boundary of the Ripley survey is marked by a large hole in the ground ("to") which is still present; (4) that the lower 442 feet of the eastern Fai’ivae boundary is identical to the western Ripley boundary; and (5) that the southernmost boundary of the Meredith survey appears to coincide closely with the traditional southern boundary of Lega’oa. The Diocese, Fai'ivae, and Ripley surveys each take note of common boundaries with the others, and the recent retraces of these surveys fit together (with a space in the middle for the Heirs of Sekio tract) almost *45as well as the parcels themselves evidently did in 1915. The Meredith survey begins about fifty feet to the north of the spring or pool called "Punaloa," which has long been a prominent landmark in this area. Punaloa appears to have been part of a tract of land of the same i:, ame, registered by the Land Commission of Samoa as the land of Matthew Hunkin and defined by Talamaivao in 1906 as the southern boundary of Lega'oa.
We further find that each of these ancient surveys depicted a tract of which the present claimant was then (ca. 1906-1918) generally reputed to be the owner; and that each of the claimants has occupied its tract, albeit in some cases intermittently, down to the present day. With such minor exceptions as are noted below, we hold these parties to be the owners of their resurveyed tracts.
We find, however, that the Uo resurvey does not accurately reflect the location of the Uo tract ("Leifi”) originally surveyed in 1915. The original survey denotes "Fai‘ivae" and "Leoso" as the neighbors to the north, and the northern boundary of the 1915 Uo tract appears to have corresponded, with some variations, to the southern boundary of the Fai'ivae and Ripley surveys. (Edward Ripley was a foreigner married to a member of the Le‘oso family.) The Uo retrace, instead of following the southern boundary of these tracts, instead overlaps substantially with them, taking up roughly the southern half of the Ripley tract and the southern third of the Fai‘ivae tract. As we have noted, the location of the to at the north of the Ripley tract and of the streams bounding the Diocese tract strongly suggest that the Ripley, Diocese, and Fai‘ivae retraces are where they ought to be. If this is correct, then the Uo tract should be about 200 feet further to the south.
Moreover, the western boundary of the 1915 Uo survey almost entirely bordered land occupied by Suafo'a. At the extreme northwest of the survey, about twenty feet south of the northwest comer, the map depicts what appears to be a fence jutting off to the west. To the north of this fence, and to the northwest of the Uo survey, were the "Marist Sisters." The recent Uo resurvey, however, would so situate the tract that its whole western boundary would be parallel to the Sisters’ land, and there would be no boundary at all with Su‘afoa. This is shown on Plaintiffs’ Exhibit 1, and can also be seen from examining the locations of Suafo'a, the Sisters’ land, Fai‘ivae, and Ripley on Exhibits 21 and 23.
An exhibit prepared by surveyor S.T. Taua‘i and submitted by Uo, purporting to show relative locations of the Uo, Suafo’a, and Sisters’ *46tracts and certain present-day monuments, is flatly wrong. This map (Uo Exhibit 37) would have the present paved road going across a comer of the Suafo‘a survey and through the middle of the Sisters’ land, whereas in fact the road does not touch either of these tracts and runs parallel to, not across, their eastern boundaries. Nor does either the Sisters’ land or the Suafo’a survey come anywhere near the Fai’ivae houses depicted on this map. A larger composite map also submitted by Uo, Exhibit 35, while inconsistent in certain minor respects with the other composite maps submitted to the Court, would also be inconsistent with Uo Exhibit 37 if the Sisters’ land and the Suafo'a surveys had been placed upon it in anything like their actual locations. Indeed, it appears that the Uo surveyor did trace the outline of the Suafo’a survey on Exhibit 35, just west of the Puletu survey where it belongs, but then thought better of it; possibly this was because the tracing on Uo Exhibit 35 depicts the Suafo’a land in a dramatically different place (relative to the Uo survey) than it appears on Uo Exhibit 37. Either Uo is wrong, or Fai’ivae, Ripley, Suafo’a, and the Diocese are all wrong.
Nor is the location of the Uo retrace itself based on any enduring monuments; as we shall discuss, it appears to have been chosen because the Uo family member who ordered the retrace preferred to contend with Fai'ivae and Ripley than with her neighbors to the south.
The dotted and solid red lines inserted by the Court on Exhibit 21 depict what we believe to have been the most probable location of the original Uo survey of "Leifi." Its boundaries with Suafo’a, the Sisters, Fai’ivae, Le'oso (including the Ripley tract), the "steep hillside," Tuiteleleapaga, and "Pule" (Puletu/Meredith) are all approximately the same distances, and in most cases approximately the same angles, as those depicted in the original. This is also quite close to the location estimated by plaintiffs’ surveyor at the 1988 hearing. (See Plaintiffs’ Exhibit 29, 1988 Hearing.)
We have also inserted on Exhibit 21 what appears to have been the most probable location of the Krause/Allen/Sekio tract. A 1901 conveyance of this property (F. Avegalio Exhibit 60) denotes a common boundary with Fai’ivae on the east and with the Sisters’ land on the west. Save for a slight variation on the western boundary, the tract is exactly the same size and shape as that depicted in the original registration and in subsequent conveyances.
The other parties claiming land in this area — Suafo’a, Tuiteleleapaga, Le'alaialoa/Avegalio/Aigamaua, Su‘a, Iuli, Le'oso, and *47Taeleifi — did not submit old surveys. Each of these families, however, was denoted on the maps of other parties as occupying land circa 1906-1918 in approximately the same location it now claims, and each appears over the years to have occupied at least some of the land it now claims.
B. Suafo'a/Diocese Conflict
Suafo‘a, now as in 1915, occupies a tract to the west of the Puletu/Meredith and Uo tracts and to the south of the Sisters’ land. The survey submitted by Suafo'a contains a small overlap with the Diocese survey. Because the Diocese tract is already legally registered as freehold land, the only way Suafo’a could prevail with respect to this overlapping section would be by adverse possession. Suafo’a testified that he has long occupied the whole area within his survey, but counsel for the Diocese introduced photographs of the overlap area which show no signs of recent occupation in the northwest comer of the Suafo'a survey. This area is on the far (northwest) side of a stream that roughly divides the survey of Suafo'a from that of the Diocese. Accordingly, we hold that the Diocese prevails on the portion of the disputed area on the northwest side of the stream.
With respect to a comer of the Diocese survey that crosses over onto the southeast bank of the stream, there is no direct evidence of whether this area was ever occupied by the Sisters. In general, the evidence is to the effect that the Sisters’ land was used for plantations until the mid-1960s by students at the Sisters’ school in the village proper. We find it unlikely that the Sisters or their students would have taken the trouble to cross the stream in order to cultivate this small strip of land. Accordingly, Suafo’a prevails with respect to this portion.
The far western portion of the Suafo’a survey appears to be outside the traditional boundaries of Lega’oa, and within the aforementioned Hunkin survey of land called Punaloa. This can be seen on Exhibits 1 and 23. We express no opinion on this conflict.
With the aforementioned exceptions, we find that Suafo’a owns the land within the Suafo’a survey.
C. Su ‘a/Diocese and Su ‘a/Willis Conflicts
Su‘a has no conflicts with anyone except a small conflict along the stream that forms his eastern boundary with the Diocese. At trial it appeared that Su‘a and the Diocese agreed that the center line of the *48stream should be the boundary. This is consistent with the best documentary evidence, the Land Commission grant to the Roman Catholic Mission.
Part of the Su‘a survey is outside the flat land of Lega'oa, on the western slopes of the valley. At the recent hearing plaintiff Tony Willis testified that his relatives (heirs of Amelia Va) had planted and cultivated these slopes even after the 1906 decision had excluded them from the land awarded to Va and her cotenants as part of Lega'oa. We find the testimony of Su'a, to the effect that members of his family have long cultivated this area, to be credible. We note that Su'a is listed as an occupant of this area in the original 1906 To'omata survey. Any cultivation by the heirs of Va was therefore neither "exclusive" nor "continuous" and did not give rise to acquisition of title by adverse possession. (For the reasons stated here and in footnote 1 infra, we also reject the Willis/Va claim to acquisition of the mountain slopes by adverse possession against other families in the eastern and western slopes of the valley. A dispute concerning the mountainous area near the northern tip of the valley is discussed in Part IV(D) infra.)
Accordingly, we hold that Su'a is the owner of the land within Su'a survey map No. 582-8-90, except insofar as it may cross the eastern stream into land belonging to the Diocese. (It appears from the two maps that the Su'a survey does follow the stream and that only the Diocese survey will need to be adjusted.)
D. Le ‘alaialoa/Diocese/Fai ‘ivae/Galea ‘i/Sekio Conflicts
The Le'alaialoa/Avegalio/Aigamaua survey (the three families are closely related and have joined as one party in the present litigation) also has a small overlap with the registered freehold survey of the Diocese. The only direct testimony with respect to occupation of this area (other than acknowledgment by various parties that it has long been occupied by "Lealailoa Felise") was the testimony of Avegalio. He said he and his family and the girls from the Sisters’ school cultivated this area side by side during the 1950s. Relations were friendly, and the boundary reflected in the Avegalio survey was the one observed by both sides. (The boundary drawn in the Avegalio survey was a straight line. The boundary reflected by the Diocese survey is a meandering one.) Avegalio went off island in 1957 and returned in 1968; by then the Sisters’ school had closed down and there were no more girls working the adjacent tract; and the Avegalio/Aigamaua/Le'alaialoa family had *49built houses on their tract, although not in the small area that overlaps the Diocese survey. The 1971 aerial photograph and the Court’s viewing of the land confirm that the line drawn by Avegalio reflects the actual boundary of Le‘alaialoa/Avegalio/Aigamaua occupation, at least d ring the last twenty years. (A 1963 aerial photograph cited by the Diocese as evidence of a different boundary is very difficult to read and provides no support for either side.) Although it is possible that the Le’alaialoa/Avegalio/Aigamaua people did not occupy the area now in dispute until after the Sisters left during the 1960s — and that their occupation would be just short of that necessary to have acquired title by adverse possession1 — the contrary testimony of Avegalio is credible, as well as uncontradicted. We therefore find that the Le’alaialoa/Avegalio/ Aigamaua families own the land within their survey, with the exception of a small area to the north discussed in connection with the To’omata and Willis/Va surveys, Part IV(E) infra.
There is also a small overlap between the Fai’ivae survey and the Diocese survey within the southeast panhandle of the latter. This small overlap (a strip about ten feet wide and fifty feet long) is also *50within a survey registered in 1979 as the individual land of Pat Galea‘i, a member of the Fai‘ivae family who was represented by Fai‘ivae during the trial of these cases. As a matter of law, an earlier survey registered in accordance with law prevails over a later one. As there is no evidence of twenty or thirty years’ adverse possession of this area by Fai‘ivae or Pat Galea‘i, the Diocese prevails with respect to this strip of land.
For the same reasons, the Diocese prevails with respect to the larger portion of the Pat Galea‘i survey which overlaps the Sisters’ land. A house constructed by Pat Galea‘i during the 1980s is only partly within the Pat Galea‘i survey, entirely outside the Fai'ivae survey, and entirely within the registered Diocese survey. The land on which this house was built belongs to the Diocese. Another portion of the Pat Galea‘i survey, to the north of the house and mostly outside the rock wall that surrounds it, overlaps the previously registered Krause/Allen/ Sekio survey. This land belongs to the heirs of Sekio.
Just to the north of the narrow strip that overlaps the Diocese, there is a long triangular piece of land included within both the Fai‘ivae and the Pat Galea‘i surveys. Because the Pat Galea'i survey has been registered according to law and the Fai‘ivae survey has not, we are bound by the land registration statute, A.S.C.A. § 37.0101 et seq., to recognize this land as the individual property of Pat Galea* i. (The law is in this respect unaccommodating to Fai‘ivae’s position that the individual registration by Pat was just a convenience and the land was supposed to remain family land. If Pat wishes to deed the land back to the family, he is, of course, free to do so.)
In the area of the Diocese, Fai‘ivae, Le‘alaialoa, Pat Galea‘i, and Sekio surveys there are also three small areas claimed by no one. These areas, marked by the Court on Exhibit 21 as Parcels 1 through 3, were included within another Fai'ivae survey which Fai‘ivae withdrew during the trial. That survey also included substantial portions of the Diocese, Sekio, Le‘alaialoa, To'omata, and Willis/Va surveys; its withdrawal was an act not only of admirable statesmanship but also of plain good sense. However, Parcels 1 and 2 within this area are closely appurtenant to lands occupied by Fai‘ivae, and are held to be property of the Fai‘ivae family in the absence of better evidence of title by any other party. Parcel 3 is more nearly appurtenant to the Le‘alaialoa/Avegalio/Aigamaua tract, and is held to belong to those families.
*51 E. Puletu/Uo/Tuiteleleapaga Conflicts
Perhaps the greatest disparity between the pattern of occupation documented in the 1915 maps and the pattern of claims among the present parties is in the southernmost portion of Lega‘oa. The parties with claims in this area are Puletu/Meredith, Uo, and Tuiteleleapaga. The dispute in question also spills over into the southern portions of the Fai'ivae, Ripley, and Le‘oso surveys.
We have already discussed the 1915 Uo and Meredith surveys. As noted, a recent retrace of the Meredith survey places this tract where it appears to belong: at the southernmost extreme of the central portion of Lega‘oa, just above Punaloa to the south, along an agreed boundary with Suafo‘a to the west, and identical in metes and bounds to the 1915 map. This resurvey (Ripley Exhibit 24) was prepared in 1987 at the request of Lafine Meredith and with the apparent co-operation of Puletu Meredith, son of the late Mrs. Thomas Meredith and also the matai of the Puletu family of Leone. (Mrs. Meredith was herself the daughter of a Puletu title holder and appears to have acquired her land from Puletu.) In the present litigation, however, Puletu Meredith has not limited his claim to the 1987 retrace of the 1915 Meredith map; rather, he claims a larger tract described by a 1990 survey, Uo Exhibit 36. This survey includes almost the entire Meredith retrace, as well as additional areas to the south, to the southeast and, most importantly, to the north.
The 1990 Puletu/Meredith survey extends about 200 feet north of the 1915 Meredith survey. This is almost exactly the north-south dimension of the 1915 Uo survey. Indeed, the northern portion of the 1990 Puletu/Meredith survey appears to occupy roughly the western third of the area which we have determined to have been the most probable location of the 1915 Uo survey. (This can be seen on Exhibit 21.)
The recent history of the Uo tract made it a likely prospect for anyone in the neighborhood who felt he needed more land. In 1928 it was leased by Uo to the Diocese for a term of forty years, to be used by the girls at the Sisters’ school for the purpose of "gardening." (The lease, registered and approved by the Governor, is Diocese Exhibit 48.) Several witnesses did remember seeing the Sisters’ students on this land during the 1940s and 1950s. After the school closed down in the early 1960s, the land appears to have been ignored by the Uo family —with the crucial exceptions of some planting by one elderly member of the Uo family during the mid-1960s (according to the testimony of Fai‘ivae) and perhaps some clearing by another such person on the other side of the *52road at about the same time (according to witness Pio Sagote). With these exceptions, the Uo tract appears to have been vacant between the early 1960s (when the Sisters’ school closed) and the late 1970s (when Tuiteleleapaga people began moving into it from the south).
Puletu himself did not present any evidence that Puletu or Meredith people ever occupied the area north of the original Meredith survey; the present house in this area, built in the late 1970s or 1980s, belongs to a member of the Tuiteleleapaga family. The situation was similar on the other side of the road: witness Tuise'e for the Tuiteleleapaga family remembered that in the early days, "we all worked together on the seaward portion of the land" now claimed by Tuiteleleapaga (emphasis added). Tuise’e said that Tuiteleleapaga people did not begin occupying the inland portion of their current claim (i.e., the area corresponding to the eastern part of the old Uo tract as depicted by the Court on Exhibit 21) until the mid-1970s when Faletoi and Uila began working there. The Tuiteleleapaga people did not move to the west side of the road until even later, when the land was "given to us," apparently by Puletu.
The situation was complicated in 1982 by the return to the Territory of Lucy Uo Ah Ching. A granddaughter of the Uo who had registered this land in 1915 and leased it to the Diocese in 1928, Mrs. Ah Ching returned after an absence of about thirty years and determined to relocate and repossess her family land. She and other members of her family had some difficulty in deciding upon the exact location of the land, but finally settled upon a spot agreed upon by their neighbors to the South. Their neighbor to the North, Fai'ivae, was a strong dissenter from this consensus. Their other northern neighbors, Ripley and Leoso, were not consulted, although a former Leoso titleholder was apparently present at the Uo survey in his capacity as paramount chief of the County. The presence of several southern neighbors and only one northern one may account for the decision to locate this survey about 200 feet to the north of the apparent location of the 1915 which it purported to retrace.
For the reasons we have discussed, we believe the Fai’ivae and Ripley surveys accurately portray the locations of these families’ land and that the correct location of the Uo property registered in 1915 is to the south of these surveys and just to the north of the original Meredith plot, as depicted in Exhibit 21. Although the Uo family has ignored this land for so long that they have forgotten exactly where it is, they have not quite lost it to adverse possession. This is because the Sisters’ students *53used the land at least until the late 1950s, and there was some Uo use of it in the mid-1960s. Moreover, there is no persuasive evidence that anyone else occupied the land until the 1970s. (The red dotted line on Exhibit 21 designates the Court’s judgment with respect to the ori ginal northern and western boundaries of the Uo tract. In order to rationalize the neighboring boundaries, however, and in the absence of persuasive claims by other parties, the small strips of land between the dotted line and the recent Suafo'a and Fai’ivae survey boundaries are also awarded to Uo. A minor adjustment in the Uo/Ripley boundary is also necessary to compensate for a slight discrepancy between the two 1915 registered surveys.)
We find that the 1987 retrace of the Meredith survey accurately retraces the 1915 registered survey of Mrs. Thomas Meredith and is the property of her heirs. We further find that a strip of land about 111 feet long and 20 feet wide, north of the original Meredith survey and south of the Uo tract, designated P-2 on Exhibit 21, is the property of Puletu. Puletu Meredith has at least recently begun authorizing people to occupy this area (as he has within the larger area to the north designated P-1 and held to be the property of Uo); it is within the 1990 Puletu survey, and it is claimed by no other party. Another area within the Puletu survey, immediately to the east of the southern part of the original Meredith survey and designated P-3 on Exhibit 21, is also Puletu property. Puletu presented persuasive evidence that this area has in fact been traditionally occupied by Puletu family members. A further portion of the Puletu survey, to the south of the original Meredith tract, appears to be mostly or entirely outside the traditional boundaries of Lega’oa and therefore outside the scope of this litigation.
The area to the east of the Puletu and Meredith surveys and to the south of the area marked "To Uo" on Exhibit 21, and contained within the plaintiffs’ resurvey of the flat land of Lega’oa (the solid line on Exhibit 1, approximately retraced by the Court in blue on Exhibit 21) is entirely within the Tuiteleleapaga survey of land designated "Punaloa" and is held to be the property of Tuiteleleapaga. The Tuiteleleapaga survey also encompasses a large area on the steep slopes to the east. Although these slopes may well be the property of the Tuiteleleapaga family, they are neither within the flat land of Lega’oa nor within the survey offered for registration in LT No. 45-82 and are therefore outside the scope of this litigation.
*54 F. Ripley/Le ‘oso/Iuli/Taeleifi
The Ripley resurvey, as we have stated, is an accurate retrace of the 1915 registered Edward Ripley survey. The only disputes in the present litigation are the dispute with Uo already discussed, and a claim by Tuiteleleapaga to a small section of the Ripley survey in the south. Because there was no persuasive evidence of Tuiteleleapaga occupation of the northern portion of the present Tuiteleleapaga claim prior to the 1970s, we hold that this area has not been acquired by adverse possession from Ripley. The Estate of Edward Ripley is the owner of the tract described by the Ripley resurvey. (Similarly, there is no evidence of historic, or even recent, Tuiteleleapaga occupation of a small area in the southwest comer of the Fai'ivae survey also claimed by Tuiteleleapaga. Fai'ivae is the owner of this area.)
The Le'oso family finds itself in an awkward posture with respect to any litigation concerning Lega'oa. On the one hand, the 1906 decision specifically excluded Le'oso from the list of Leone chiefs entitled to have a share in the seaward half of Lega'oa. (In Le'oso v. Ripley, LT No. 18-1920, the Court rebuffed an attempt by Le'oso to relitigate this issue with the observation, "Case thrown out of court." The opposing litigant was Edward Ripley, a Le'oso in-law who, it would appear from the old maps, almost certainly had acquired his property from Le'oso in the first place.) On the other hand, Le'oso was clearly present in this area in 1906 and appears still to have been present in 1915. He seems to have lost the 1906 case because he withdrew his objection to the massive To'omata survey in a misconceived effort to make a separate peace with To'omata. Counsel in the present case, moreover, has presented persuasive evidence that the area immediately to the east of the Ripley tract (designated as belonging to "Le'oso" in two of the 1915 maps) has been occupied by Le'oso people (principally Sagote and his immediate family) at least since the early 1950s. As this land is outside the "flat land" and therefore outside the 1906 court’s definition of Lega'oa (see Exhibit 1), a holding that this area belongs to Le'oso is not inconsistent with the 1906 or 1918 decisions. In any event, the Le'oso occupation of this area appears to have been open, notorious, exclusive, continuous since the 1950s, and hostile to any other claim on the land. We therefore find that Le'oso is the owner of the land due east of the Ripley tract and within the Willis survey offered in LT No. 45-82 that defines the scope of this litigation. (We express no opinion on the more steeply mountainous area further to the east, which, although within the recent Le'oso survey, is outside the scope of these consolidated cases.)
*55The Le‘oso claim also extends to the north of the Ripley survey and to the east of the Fai'ivae survey. On the 1915 Fai'ivae map the land in this area is designated as belonging to Iuli and Taeleifi. The 1915 Ripley map also identifies Iuli as the neighbor to the north.
We are convinced that the Sagote family, acting as members of the Le‘oso extended family, has long had plantations in the area designated on the old maps as belonging to Iuli and Taeleifi. Indeed, the Sagote people have even built a house on the hillside in an area that may once have belonged to Iuli and may formerly have had a small Samoan working house squarely within the 1915 Iuli tract. We have also been presented, however, with persuasive testimony that the two areas identified in the 1915 maps as belonging to Iuli and Taeleifi were in fact traditionally regarded as lands of those families and that both families have made at least such use of these lands as would have prevented Sagote or anyone else from having "exclusive" or "continuous" possession for twenty years. Specifically, Leota Toloa testified credibly that he and other members of the Iuli family, including a matai named Atina, worked this area for many years and that he himself only stopped going there in 1979. This testimony was confirmed by other witnesses and was partly confirmed even by Pio Sagote, who explained Atina’s presence on the land by reference to a connection with the Le‘oso family.
There was abundant testimony as to the presence of Taeleifi people in the overlap between Taeleifi and Le'oso. Indeed, at the present time Taeleifi people and Le‘oso people seem to be more or less working around each other in the overlap between the two surveys. This tends to confirm what would in any case have been an obvious hypothesis: that the Le‘oso/Sagote plantations began within the area belonging to Le‘oso, gradually extended northward into the Iuli area, and began more recently to encroach on the Taeleifi tract.
The one thing that is impossible to tell with precision from the 1915 maps is the depth of the Iuli and Taeleifi tracts: how far they extended toward the mountain. From the maps we can tell only that these tracts were reputed to extend at least 200 feet or so east of the Fai'ivae boundary. The Taeleifi survey, however, extends 369 feet east from this boundary (up to about the 100-foot contour) and the Iuli survey extends 500 feet east (to the 200-foot contour).
We find that the Iuli family has presented persuasive evidence of traditional ownership of the area immediately adjacent to the Fai‘ivae and Ripley surveys, and of at least such occupation of this area as would *56prevent its acquisition by adverse possession, but has presented no evidence of such ownership or occupation further east than 200 feet from the Fai‘ivae boundary. (We further find that the northern boundary of the Iuli tract should be modified very slightly to correspond with the southern boundary of the Taeleifi survey of "Lalolelata," which more accurately reflects the 1915 Fai‘ivae map. The eastern boundary should be calculated by reference to a line perpendicular to the southern boundary of the Taeleifi survey, extending to the southern boundary of the Iuli survey.)
The best evidence of ownership of the area within the Iuli survey which is more than 200 feet east of the boundary is the occupation of this area by Pio Sagote and his family. Although we have no direct evidence of when this occupation began, in the absence of direct evidence of any prior occupation, this fact works for rather than against the Le‘oso claim. Accordingly, we find this area (insofar as it is also within the survey offered in LT No. 45-82 and therefore within the scope of this litigation) to be the property of Le‘oso.
In the case of the Taeleifi survey there is convincing evidence not only of historic ownership but also of more recent and more intensive occupation than in the case of the Iuli tract. We also have direct evidence of the eastern extent of the Taeleifi land: Taeleifi testified that until the recent hurricane there was a Taeleifi house on the slope near the southeastemmost boundary of the "Lalolelata" survey. Although Pio Sagote also has plantings within this survey, they are insufficient to acquire ownership by adverse possession. Accordingly, we find that Taeleifi is the owner of the land within the Taeleifi survey of "Lalolelata," insofar as it is also no further north than the northern boundary of the Fai‘ivae survey. (Parts of the Taeleifi claim that are north of this line are disputed with the Willis/Va plaintiffs and with To'omata. We discuss these disputes in Part IV, infra.)
IV. Boundaries Within the Inland Half of Lega ‘oa and the Surrounding Slopes
Property rights within the inland portion of Lega‘oa are somewhat less difficult to analyze than those in the seaward portion. The starting point is the holding of the 1906 and 1918 decisions: that the inland half of the flat land belonged to Tuitele To'omata, Tali, and Amelia Va as tenants in common. Other parties claiming land in this area would have to prove either that their lands were subject to freehold grants from the former Land Commission or that they had acquired title *57by adverse possession since 1906. The slopes surrounding the flat land belong to whoever owned them before 1906 or to whoever has acquired them since by original occupation or adverse possession.
A. Willis /To ‘omata/Taeleifl Conflict
The major conflict with respect to the inland portion of the flat land has to do with the boundaries in the southern and eastern section. The contestants in this area are the Willis/Va plaintiffs, To'omata, and Taeleifi. (As previously noted, Le‘oso also claims some of the land within the Taeleifi survey of "Lalolelata." For the reasons already stated, we reject this claim.) The dispute among these parties partakes partly of a boundary dispute — with Taeleifi placing the boundary of the seaward portion of Lega'oa somewhat further to the North than other parties have placed it — and partly of a dispute over original occupancy or adverse possession. Willis claims his relatives have long occupied not only the flat land awarded to them in 1906 but also the surrounding slopes that were excluded from the decision. Taeleifi claims his people have long occupied the section of the flat land that is within the Taeleifi survey of "Lalolelata," even if it is above the inland/seaward boundary mandated by the 1906 decision.
The situation is complicated by the fact that the To'omata, Willis/Va, and Taeleifi families are all related to each other. Each party can therefore plausibly explain the presence of the others on "his" land by reference to family relationship. For instance, during the 1970s there seems to have been an agreement among the Fai'ivae, Taeleifi, To'omata, and Willis/Va families that a certain hibiscus hedge would be the northern boundary of land occupied by Pepa Taeleifi. Taeleifi seems to regard this either as a recognition that this was always the boundary or a gift of title to the land. To'omata and Willis, however, regard the agreement as a license for Pepa to live on To‘omata/Va land; they point to the fact that Pepa was a first cousin of So'oto, the Va heir who agreed that Pepa should be allowed to live in the area south of the hibiscus hedge.
Similarly, So‘oto appeared in this litigation as a Va heir, and appears to have been the principal Va representative in this area before Tony Willis returned to the Territory in the late 1970s. So‘oto lives in a house on the slopes overlooking the inland portion of the flat land, and the Willis/Va plaintiffs rely on this as important evidence of occupation of these slopes by Va people. Taeleifi, however, points out that So‘oto is a member of the Taeleifi family and that "So'oto" is a matai title *58closely connected to that family. Taeleifi contends that the father and mother of the present So'oto began living in this area, long reputed to be a part of Taeleifi land called "Tiafau," because of the connection of his father So‘oto Saka to the Taeleifi family. But So‘oto says the important connection was through his mother, a Va heir.
We find as follows: first, the boundary between the seaward and inland portions of this land is an eastward extension of the line that marks the northern border of the Fai‘ivae tract. The boundary between Taeleifi and "Tuitele" (i.e., To‘omata) is so marked on the 1915 Faifivae map; this boundary is still recognized by To‘omata and Fai‘ivae as their boundary and by To'omata and the Le‘alaialoa\Avegalio\Aigamaua family further to the west; and it comes quite close to achieving a mathematically precise division of Lega'oa into two equal portions, as ordered by the 1906 court. Taeleifi‘s argument that the boundary is further north, and To‘omata’s contention that it jogs to the south in this area, seem principally designed to bolster their arguments about who gave permission to whom.
We further find that no party has acquired land by adverse possession in this area. Even if So‘oto was acting at all times as a Va heir and not as a member of the Taeleifi family, he could not have possessed land adversely to a family of which he was a member. Similarly, the parties to the hibiscus-hedge incident may or may not have reflected on whether they were giving a license, giving away land, or recognizing a pre-existing property right; had they so reflected, the various parties might well have had different understandings about what was going on. If, however, the land was To‘omata/Va land to begin with, then the Taeleifi family cannot hold the To‘omata/Va heirs to any agreement by which a Taeleifi family member gave away To‘omata/Va property to his own first cousin. (The cousin might well prove, however, that in building his house in this area he reasonably relied on a license given by the Va heir who seemed to be in charge at the time.)
Applying these findings to the claims-of the parties, we find that Taeleifi is the owner of the land on the slopes within his survey of "Lalolelata" (i.e., east of the Willis/Va retrace of the "flat land”) and to the north of the inland/seaward boundary (as defined by an eastward extension of the Fa‘ivae/ To‘omata/Le‘alaialoa boundary to the west). Taeleifi is also the owner of his survey of "Tiafau," which is on the northeastern slopes and which is regarded by all non-Va neighbors as Taeleifi property. The flat land within the Taeleifi survey of "Lalolelata" *59(i.e., inside plaintiffs’ retrace), and north of the inland/seaward boundary as defined above, belongs to the heirs of To‘omata, Tali, and Va.
B. Willis/To ‘omata/Iuli Conflict
Iuli has submitted a survey of land in the northeastern portion of the inland area, just south of the Táeleifi land called Tiafau. This Iuli survey is partly on the slopes and partly on the flat land. Iuli witnesses testified that they have long occupied the whole tract. There is strong evidence to the effect that a Iuli family member, Willie Ah Kuoi (Kuoy), had one of the first houses in this area, on the sloping portion of the present Iuli survey; that members of his immediate family have lived in the area for thirty years or so; and that other Iuli people may have occupied at least some of this land even earlier.
The Iuli people believe that they were "given" their land by Tuitele long ago. It appears that Iuli may have been one of the persons to whom Tuitele To'omata sold land prior to 1918, giving rise to the Court’s holding in Falesau v. Tuitele, 1 A.S.R. 298 (1918), that he was only a co-tenant and was not entitled to sell land within Lega'oa. This holding would not, however, have applied to land outside the "flat land" which was the Court’s definition of Lega'oa. As it happens, the evidence of historic Iuli occupation is far stronger on the slopes than within the flat land. The only Iuli houses built before 1980 were on the slopes; the two small houses on the flat land appear to have been built during the pendency of the present litigation. A member of the Ah Kuoi family testified that his family has had crops in the flat land, but there is also strong evidence that So‘oto and perhaps other Va people have had crops in this area. The burden would be on Iuli to prove at least twenty years (beginning in or before 1962) of uninterrupted and exclusive possession of the flat land in order to acquire any of it. This he has not done. Accordingly, we conclude that Iuli owns the portion of his survey that is on the slopes but that the portion within the flat land (again, defined by plaintiffs’ retrace) belongs to the heirs of To‘omata, Tali, and Va.
C. Willis/To ‘omcita/Sekio Conflict
The Heirs of Sekio Avegalio also claim land within the eastern portion of this area. They rely on a freehold grant from the Land Commission for lands called "Lefuafua and Legoa." The Willis/Va plaintiffs acknowledge the validity of this grant but contend that the land is not within Lega'oa and must be somewhere else. We conclude, *60however, that the grant is within Lega‘oa and that the resurvey submitted by the Heirs of Sekio is the best evidence of its exact location.
The physical description of the land in the freehold grant, although varying in some respects from the current appearance of this land, is strikingly similar in other respects. There is a steep hill just where the survey says "hill." A small stream crosses the road exactly where it ought to. Disinterested witnesses testified that there used to be a swamp about where the survey indicates a swamp should be. The absence of another stream listed on the survey may be explicable by reference to some bulldozing done by So'oto during the mid-1970s, which is said to have changed the course of the main stream in this area. (This bulldozing, which appears to have involved a major portion of Lega'oa, figured heavily in several parties’ evidence. To So‘oto it was evidence of his suzerainty over the area. A Iuli witness was angry because he said he had paid for the bulldozer to level some land for his house and that So’oto — who, it develops, is or was employed as a bulldozer operator for the Government — was doing his own bulldozing on Iuli time. A witness for the Taeleifi family, however, testified that she was the one who paid for the bulldozer. The bulldozing led to a lawsuit with the Heirs of Sekio and perhaps to another with To'omata.)
Failautusi Avegalio, the principal witness for the Heirs of Sekio, appeared to be somewhat confused about this land. He insisted, for instance, that the Marine rifle range had passed directly over the land, whereas the great weight of the evidence shows that it did not. Despite the doubts raised by this testimony, however, we find that the Sekio people have made at least such use of their land over the years as would prevent its acquisition by adverse possession. Almost all the neighboring land occupants testified that they had seen Sekio Avegalio people on the land at various times. Most recently, the 1975 lawsuit (Avegalio v. So‘oto, LT No. 1526-75) in which a permanent injunction was issued against further encroachment on this land by So‘oto, tends to locate the land as part of that bulldozed by So’oto and also to indicate that the heirs of Sekio were still taking an active interest in the land. The result in the bulldozing injunction case may not preclude relitigation of title in the present case; it was a judgment by default after So‘oto failed to appear, and in any case might not bind So‘oto’s cotenants. At the very least, however, the injunction is sufficient to interrupt any acquisitive prescription by So‘oto on behalf of himself and his cotenants. That the suit was brought at all, moreover, bolsters the other evidence that this is the location of the Sekio grant and that the Heirs of Sekio were making *61some use of it. We also note that Lefuafua was identified several times in the pleadings and testimony of the 1906 case as land within Lega‘oa.
We conclude that the location of the freehold grant to "Lefuafua and Legoa” belonging to the Heirs of Sekio Avegalio is where it appears on Exhibit 1 (marked "Se") and Exhibit 23 (marked "Failautusi.")
D. Willis/Tuiteleleapaga/Avegalio Conflict
The remaining substantial conflict in this area has to do with the northern slopes. Tuiteleleapaga has submitted a survey of land the family calls "Tiafau." It is entirely within the mountainous area and just to the north of the Taeleifi survey of the same name. Avegalio has submitted a survey of land just to the west of the Tuiteleleapaga survey. This land is also primarily in the mountainous area, although the southern portion of it is within plaintiffs’ retrace of the "flat land" and appears to the Court to be reasonably flat. The Willis/Va plaintiffs claim this entire area (as, indeed, they also claim the eastern and western slopes) by reason of original occupation and/or adverse possession. Tony Willis testified that his relatives have long occupied this area.
We believe this land to be the traditional communal land of the Tuiteleleapaga and Avegalio families. We note the 1906 testimony of Fai’ivae Alfred Hunkin, apparently a disinterested witness with respect to this section of the land, that the northern border of Lega‘oa is "with land named Tiafau over which Taeleifi Avegalio and Tuiteleleapaga had the pule." (1906 transcript, as translated by the Court interpreter for the 1988 trial, at p. 9.) These are the same three families who claim the land today. Moreover, we heard extensive testimony at the recent hearing with respect to Tuiteleleapaga and Avegalio activities on their respective claims in this area during the middle part of this century. (Such activities, which consisted mainly or exclusively of plantings, appear to have diminished after the 1966 hurricane.)
In the mid-1970s Tony Willis returned after many years in the United States and began to assert his authority in this area. This quickly led to two lawsuits; one against the Tuiteleleapaga people, in about 1976, of which we have no specific details and which Mr. Willis says "just vanished" during the 1970s, and another against the individual heirs of Ese'ese Avegalio, one of the Avegalio people who had been working in this area.
*62The latter case went to trial. It concerned the land in the immediate vicinity of the area where Tony Willis was then building his house. Willis contended that this land was part of Lega‘oa and as such had been deeded to his grandmother, Amelia Va, and held to be her property in the 1906 case. The Court rejected this contention, finding that: (1) the land was not "flat" and therefore was not part of Lega‘oa within the meaning of the 1906 case; (2) at least some parts of the land had been cleared and cultivated by Ese'ese Avegalio during the 1940s, 1950s, and 1960s; (3) after the 1966 hurricane destroyed his crops, Ese'ese had abandoned the land, although some other people continued to plant a few crops there with his permission; (4) some of the crops that had been planted with the permission of Ese‘ese were destroyed by Willis when he began building his house in the late 1970s. The Court concluded that the land was not part of the Va land called Lega‘oa, but was not the property of the heirs of Ese‘ese because he had "abandoned" it and therefore could not lay claim to individual ownership. The latter holding had to do with certain distinctions made by the Court between the nature of individual and communal ownership of land in Samoa. Leuma v. Willis, 1 A.S.R.2d 48 (1980).
We conclude that the land within the Tuiteleleapaga survey of Tiafau is communal land of the Tuiteleleapaga family. We also conclude that the land within the Avegalio survey, with the exception of that below the 75-foot contour line, is the communal property of the Avegalio family. Mr. Willis is estopped to deny that the land is not flat, that it is not part of Lega‘oa, that it was cultivated by Ese‘ese Avegalio, and that he destroyed crops planted with the permission of Ese‘ese when he laid the foundation for his present house. The testimony at the recent hearing convinces us that the activities by Ese‘ese and other members of the Avegalio family on this land were pursuant to the traditional status of this land as Avegalio communal land.
We conclude that the portion of the Avegalio survey below the 75-foot contour land, which includes another Willis house, is "flat land" belonging to the heirs of To‘omata, Tali, and Va. (The 75-foot contour line is slightly below plaintiffs’ retrace of the fiat land. This adjustment is necessary to conform to the Court’s holding in Leuma — which appeared to involve the area around the inland Willis house, but to make no finding with respect to the land around the seaward house — and to our observation of the land in this area.)
*63 E. Willis ÍLe ‘alaialoa/Olo/Avegalio/Su ‘(Conflicts
There is also a slight conflict between the Willis/Va plaintiffs and several other parties with respect to the western boundary of Lega'oa. The Willis retrace draws the line at approximately the fifty-foot contour line; the other parties draw a slightly different line at a stream that meanders along the lower part of the western slope, between the 25-foot and 75-foot contour lines.
This is one of those instances in which mathematical boundaries appear to have given way to natural ones. The difference between the boundary depicted in plaintiffs’ retrace and that afforded by the stream is a narrow strip which in most places is no more than 20 to 30 feet wide. Despite some contrary testimony by Va witnesses Willis and So‘oto, we are convinced that the Va/To‘omata neighbors to the west have long occupied the land up to the stream. To‘omata, the co-tenant of the Willis/Va plaintiffs, recognizes the stream as his western boundary. Moreover, the boundary appears to have been a traditional one: Fai’ivae testified in 1906 that the western boundary of Lega‘oa is "the contributory of the large stream." (1906 transcript at p. 9.) This appears to be the same stream recognized today by all parties but the Willis/Va plaintiffs. Moreover, the only convincing evidence of any occupation of this strip by Va people concerns a brief incursion by So‘oto during the rogue-bulldozer episode in 1975. Olo testified that he confronted So‘oto, who then asked his permission to plant crops on the land. Olo said So‘oto could plant taros (an annual crop) but not bananas or coconuts (which might be later construed as evidence of ownership). The crops and structures presently on this land all belong to the western landowners. We conclude that the western boundary of the land belonging to the heirs of To'omata, Tali, and Va is the stream near the bottom of the western slopes.
It appears from the To'omata survey (Drawing No. 54-15-90, not marked as an exhibit but admitted without objection and incorporated into Exhibits 1 and 23) that the Le‘alaialoa/Avegalio/Aigamaua survey of "Leui" to the south may include a small area on the east side of the stream. Because this land is part of the inland half of Lega‘oa, and because little or no evidence was presented to support a finding of adverse possession by Le‘alaialoa et al., we find that this land is the property of the heirs of To‘omata, Tali, and Va. We hold that the remaining land within the northwestern surveys of Le'alaialoa et al., Olo, Avegalio, and Su‘a, insofar as they are also within the survey offered for *64registration in LT No. 45-82, to be the property of the respective claimants.
F. Willis/Va Claims on Northeast Slope
We find that two areas outside plaintiffs’ retrace of the flat land do belong to plaintiffs and their cotenants. The northernmost of these areas is just to the east of the northernmost area of the flat land. It is bounded by the Avegalio survey on the north, the Tuiteleleapaga survey of "Tiafau" on the east, and the plaintiffs’ retrace of the flat land on the west. This area is in the vicinity of the southernmost Willis house. In the absence of strong evidence of prior occupation by any party, this proximity and Willis’s general testimony of his family’s occupation of all the slopes is the best evidence of ownership.
The second area is just to the south of the first. It is a gently sloping area bounded on the east by the Taeleifi survey of "Tiafau,” on the south by the Iuli survey, and on the west by plaintiffs’ retrace of the flat land. So‘oto testified that he and his family have long had crops in this area, and that the crops were under the authority of his mother, a Va heir. In the absence of a strong claim by any other party — in particular, we note that the area is outside the Taeleifi survey that includes the present house of So‘oto — we accept this testimony as the best evidence of ownership.
These two areas are held to be the property of the heirs of To’omata, Tali, and Va.
V. Conclusion
The inward half of the "flat land" of Lega'oa is the property of the heirs of To’omata, Tali, and Va as tenants in common. This land is defined as that land which is within the following boundaries: (1) on the south, a line defined by the northern boundary of the Le’alaialoa and Fai’ivae surveys and by an eastward extension of this boundary through the Taeleifi survey of "Lalolelata"; (2) on the east, the easternmost boundary of the Willis/Va plaintiffs’ 1990 retrace of the flat land within Lega‘oa; (3) on the north, the 75-foot contour line from its eastern intersection with plaintiffs’ retrace of the flat land to its western intersection with a stream near the western boundary of the Avegalio survey; (4) on the west, the stream that forms the eastern boundary of the Su‘a, Avegalio, and Olo surveys.
*65The heirs of To‘omata, Tali, and Va are also the owners of two parcels just to the east of the above-described flat lands, on the northeastern slopes above Lega’oa, as more fully described in Part IV(F) of this opinion.
The families of Leone who have made claims within the seaward portion of Lega’oa are the respective owners of the parcels described in Part III of this opinion.
Certain families of Leone are also the owners of land along the slopes surrounding the inland portion of Lega‘oa, as further described in Part IV of this opinion.
The Diocese of Samoa-Pago Pago is the owner of the tract described in Part III of this opinion and illustrated by the Court on Exhibit 21.
The Estate of Edward Ripley and the heirs of Mrs. Thomas Meredith are the owners of the land within their respective surveys, as further described in Parts 111(E) and 111(F), respectively.
The heirs of Sekio Avegalio are the owners of a freehold parcel within the seaward part of Lega‘oa, as further described in Parts III(A) and III(D) of this opinion, and of another freehold tract within the inland part of Lega'oa, as further described in Part IV(C) of this opinion.
We express no opinion on the ownership of any land other than that described above. We also express no opinion on the ownership of any land which is both outside the survey that was offered for registration in LT 45-82 (one of the cases consolidated herein) and outside the plaintiffs’ 1990 retrace of the flat land of Lega’oa.
We express no opinion on the validity of a mortgage alleged by plaintiff Levi, who seems to have dropped out of the case after the first hearing.
Finally, we note that none of the lawsuits giving rise to these consolidated cases was an action for eviction. Several people have houses on land which has been held to belong to other people. Some of these people appear to be licensees of the true owners; others may be entitled, if eviction is sought, to compensation for improvements made in good faith. We have every expectation that such issues can be resolved by negotiation rather than by further litigation.
*66Judgment will issue in accordance with these findings and conclusions. It is so ordered.
A.S.C.A. § 37.0120, the adverse possession statute, was amended in 1962 to change the prescriptive period from twenty years to thirty years. Land could be acquired by twenty years of actual, open, notorious, hostile, exclusive, and continuous occupancy provided that such occupancy began in 1962 or earlier. Occupancy beginning later than the effective date of the 1962 amendment — or which, at any time since 1962, has been interrupted or has not been exclusive — cannot yet have resulted in acquisition of title by adverse possession, because thirty years have not yet passed since 1962. Thus the only parties to the present litigation who have acquired land by adverse possession are those whose possession began in 1962 or earlier and has been exclusive, hostile, etc., since that time.
Prior to the 1966 hurricane, Lega'oa was used almost exclusively for plantations and contained few residential houses. Nor do the plantations appear to have been the sort that would have necessitated clear-cutting of forest trees; although a number of old coconut trees are still in existence, most of the plantations seem to have been bananas and taros interspersed among the pre-existing features of the landscape, as is still fairly common. It should also be noted that for several years during World War II the Marines maintained a rifle range in Lega'oa and the whole area seems to have been vacated.
Subject to these constraints on the type of "occupancy" that prevailed until recently in Lega'oa, all parties to the present litigation were shown to have made some use of their own land in the years between 1906 and 1962. Although such use was in some cases intermittent, no party abandoned its land for the applicable prescriptive period (twenty years beginning in or before 1962, or thirty years thereafter). Any occupation by other parties was therefore not "exclusive." The only exceptions, in which adverse possession claims have been proved, have to do with minor boundary disputes in which one party presented the best evidence of what the boundary was circa 1906, but another party persuaded the Court that a slightly different and more "natural" or "obvious" boundary — e.g., a straight line or a stream — had been recognized by all parties over the years.