delivered the opinion of the court:
The Spokane Tribe challenges a ruling of the Indian Claims Commission that the Tribe held Indian title in the 19th century to a lesser area than it claims in this proceeding. 9 Ind. Cl. Comm. 236 (Docket No. 331) (1961). The appeal is interlocutory, before any determination of value has been made. The United States did not cross-appeal. Under 25 U.S.C. § 70s(b), it would be free at a later time to contest the Commission’s determination on any ground, but its brief in this court declares unequivocally that the findings of the Commission determining the extent of the area exclusively used and occupied by the appellant “are well supported by the evidence.” We take this to be appellee’s deliberate acceptance of the evidentiary basis of the Commission’s specification of appellant’s area. Those limits mark, therefore, the minimum boundaries of any area for which appellant may be entitled to compensation.1
The territory claimed by the Spokane Tribe lies in northeastern Washington, near the Idaho border. It came under acknowledged United States dominion by the Treaty With Great Britain, June 15, 1846, 9 Stat. 869, and was first included within the Oregon Territory established on August 14,1848, 9 Stat. 323, and then within Washington Territory when it was constituted on March 2, 1853, 10 Stat. 172. In 1855 Congress authorized the extinguishment of Indian title in Washington Territory east of the Cascade Mountains. Pacts were soon made with many of appellant’s neighbors but, despite a series of councils, discussions, and abortive *61agreements, no final understanding with appellant was consummated until the Agreement of March 18,1887, by which the Spokane Indians ceded all the “right, title and claim which they now have, or ever had, to any and all lands lying outside of the Indian reservations in Washington and Idaho Territories” — and agreed to take out individual land patents, or to remove to and settle upon the Coeur d’Alene Reservation in Idaho, the Colville Reservation in Washington, or the Jocko Reservation in Montana.2 This Agreement became binding upon its ratification by Congress on July 13, 1892, 27 Stat. 120, 139. The Indian Claims Commission set that as the date of the taking of the lands on which the present action is based. The appellant’s claim is that the terms of the Agreement should be revised, under Section 2(3) of the Indian Claims Commission Act, 25 U.S.C. § 70a(3), on the ground that an unconscionably low consideration was paid and the Tribe was entitled to more.
I
The Spokane Indians were a land-using and fishing group, numbering between 500 and 900 individuals during the last half of the 19th century.3 They were divided into three connected bands or groups — the Upper, the Middle, and the Lower Spokanes — on whose behalf appellant sues (as well as on its own behalf). They were surrounded by other Indian groups and tribes with most of whom they were friendly. Particularly after 1877, white settlers crowded the Spo-kanes away from their fisheries and settled upon their hunting and food-gathering grounds outside the Spokane Reservation (see footnote 1, sufra). Once the missionaries had introduced farming, during the earlier parts of the 19th century, many Spokanes developed small cultivated areas, but these were often lost because the Indian owners failed to *62make proper filings under tlie land laws and incoming settlers claimed title through a railroad grant or homestead filing. At the time of the 1887 Agreement with the Federal Government, the general condition of the Tribe had seriously deteriorated.
The lands covered by the Agreement, and therefore involved in the present suit, were all those aboriginally owned by the Tribe. Appellant claims approximately 3,140,000 acres, less the 154,898 set aside and reserved for the Spokane Tribe in 1881 — a net of some 2,955,102 acres. The Indian Claims Commission determined the aboriginally-owned area to be an irregular oblong region (some 70 miles long and 45 miles wide) of about 1,854,858 acres, including the Spokane Indian Reservation (154,898 acres), a net total of about 1,700,000 acres.4 (The present city of Spokane is located within the perimeter of this area.) The Commission’s decision was based on a record containing the oral testimony of the parties’ anthropological witnesses, as well as reports, studies, maps, and findings of prior students, travellers, and officials. Appellant attacks most of the Commission’s boundaries as too limited; only the northern and northwestern sides are left unchallenged. We shall consider each of the disputed borders separately.
A. Northeast houmdary: The Commission places the northeastern line of the Spokanes’ area as running from a point a short distance northwest of Milan, Washington, southeasterly *63to Milan and then on to Peone, and from there in a southerly direction to Opportunity. The line cuts off Mt. Spokane and the territory to the east of that spot. Appellant calls this boundary the Commission’s most glaring error and argues that it is wholly unsupported by any substantial evidence. We are compelled to agree. Neither of the two expert witnesses and none of the writings of the other main students supports so restricted a line; on the contrary, all the major materials bearing directly on the northeast border of the Spokane land indicate that the Indians’ territory extended, on that side, considerably beyond the line fixed below. All or substantial part of the area eliminated by the Commission’s line was found to be occupied exclusively by the Spokanes by Dr. Verne F. Eay, appellant’s expert witness; by Mr. Stuart Chalfant, appellee’s expert; and by independent observers or anthropologists such as James Teit, Edward S. Curtis, James Mooney, John E. Swanton, Leslie Spier, George Gibbs, and Father DeSmet (probably) — whose writings and maps are in evidence.
The only materials cited by appellee in support of the Commission’s northeastern line are three maps which do not bear on the point. One is a map made in 1841 by Charles Wilkes, commander of a naval exploring expedition, after a very short visit to the area; this map does not help in placing the northeast boundary since in that region (at least) Wilkes did not distinguish between the Spokanes and other Salish-speaking Indians (such as the Kalispel, Coeur d’Alene, Flathead, etc.), lumping all of them together in a large territory. Appellee also refers to one of several maps prepared from the observations of Father DeSmet, a Jesuit missionary who lived in the area in the 1840’s and early 1850’s. This particular map covers very large segments of western and midwestem United States and, with respect to the relatively tiny area with which we are now concerned, is far too general and imprecise to be of any aid. Appellant points out that another, more detailed, map by Father DeSmet of the area of our present interest can be read as intimating that the Spokanes extended beyond the northeastern line drawn by the Commission. The third map on which appellee relies is one by Dr. Eay (appellant’s *64expert) included in an ethnological article in 1936. This is a small-scale, diagrammatic sketch depicting the areas of over twenty Indian groups; it must necessarily be taken together with a more-detailed map in the same article showing that, in Dr. Day’s view, the Spokanes’ territory went beyond the Commission’s northeastern line.
Thus, appellee cannot properly rest on the maps it cites as providing any substantial foundation for the Commission’s determination of this sector of Spokane land. The other pertinent evidence, as we have said, supports appellant. From the Commission’s opinion (see 9 Ind. Cl. Comm. at 264) one can infer that it denied the Tribe’s claim to the area west of (and adjacent to) Mt. Spokane because it attributed significance to the absence of references in the early accounts of the region to this outstanding topographical feature; the Commission apparently felt that the mountain would have been mentioned if it had been an identification point in the boundary between the Spokanes and other friendly tribes (as appellant contends that it was). In our view this thin speculation could not be substantial evidence outweighing the strong, cumulative, and uncontradicted testimony and materials on which appellant can count.5 The Commission’s northeastern boundary must be set aside as unsupported by substantial evidence in the record as a whole.
B. Southeast boundary: The Commission’s southeast line runs from Opportunity (near the junction of the Spokane Diver) in a southwesterly straight line to Spangle, to Malden, and then to Pine City. Appellant accepts Opportunity as the starting point but attacks the boundary itself as too retracted. For this sector the state of the record is that appellant’s protest against the Commission’s line is supported by its expert, Dr. Day, and by the views of a number of earlier students and officials (including Spier and Mooney, whom *65appellee incorrectly cites as sustaining the ruling below),6 while the Commission finds almost all its corroboration in the evidence of appellee’s expert, Mr. Chalfant. The difference in the two positions is that Chalfant believes the line adopted by the Commission to represent the farthest extent of the Spokane’s exclusive occupancy, the lands to the east being used in common with other tribes; the rest of the observers think that these more easterly lands were occupied solely by the Spokane Indians (in the sense in which “exclusive occupancy” is employed in cases under the Indian Claims Commission Act) although other groups may have visited there.
The mere counting of witnesses’ heads is not the office of a reviewing court, but where a single witness is arrayed against a number of others it is a prime part of the judicial function to examine his testimony with care to see whether in itself it affords substantial support to the determination made by the fact-finders. That is one of the meanings of the Congressional directive that we consider the record as a whole, and not only a particular piece of evidence in isolation. Universal Camera Corp. v. National Labor Relations Board, 340 U.S. 474, 487-488 (1951); Osage Nation of Indians v. United States, 119 Ct. Cl. 592, 606, 608-611, 612-613, 615, 672, 97 F. Supp. 381, 388, 389-391, 391-392, 393, 425 (1951), cert. denied, 342 U.S. 896 (1951).
In this light, we have reviewed Mr. Chalfant’s evidence on the southeastern boundary.7 His whole presentation on this part of the case is tied to his conception that, though Spokanes were often found east of the restricted line he advocated, their only areas of “exclusive occupancy” were west of that line. In the excluded territory, he says, they commingled with other groups in a common area owned by none, oT they were guests on lands belonging to others, or they had only seasonal habitation. He explains the maps or writings of earlier observers — most of whom had affirma*66tively placed the Spokanes east of his line, and none of whom had denied that the Spokanes lived there — as expansively representing all the areas used or roamed by the Spokanes and not merely their areas of true exclusive occupancy.
We find three crucial defects in this general position. The first is that there is no adequate basis for assuming that the prior students included in their delineations of Spokane territory (in this section) any land which was not aboriginally owned by this Tribe according to the standards necessary for recognition under the Indian Claims Commission Act (see The Sac and, Fox Tribe v. United States, 161 Ct. Cl. 189, 201-02, 315 F. 2d 896, 903 (1963), cert. denied, 375 U.S. 921. The main indications in their maps and writings are contrary to Mr. Chalfant’s hypothesis and he has only unpersuasive bits and pieces out of which to build his supposition that the other observers and ethnographers included commonly-used lands. Second, Chalfant’s view of Indian occupancy is too narrow; he includes only areas in which there were permanent villages or habitations and excludes seasonal or hunting areas, even though intermittent or seasonal use has been accepted as showing Indian title. Delaware Tribe of Indians v. United States, 130 Ct. Cl. 782, 789, 128 F. Supp. 391, 395 (1955); Iowa Tribe v. United States, 6 Ind. Cl. Comm. (Docket No. 135) 464, 478 (1958); Pawnee Indian Tribes. United States, 5 Ind. Cl. Comm. (Docket No. 10) 224, 291-292 (1957); Omaha Tribe v. United States, 4 Ind. Cl. Comm. (Docket No. 225-A) 627, 637, 664, 667 (1957); cf. Mitchel v. United States, 9 Pet. 711, 746 (1835). His depiction of the areas of exclusive Spokane occupancy must therefore be suspect as founded on an erroneous postulate. Third, when we measure Mr. Chalfant’s unelaborated and unproved statements that the Spokanes used all of the land east of his southeastern line in common with the other Indians of this region (or as mere visitors) against the mass of other evidence on this point, we cannot accept his solitary opinion as a substantial enough foundation upon which the Commission could rest its southeastern line. Both the nature of his statements and the significant contradictory materials detract seriously from the weight which can be given his views. For these *67reasons, we must overturn the Commission’s determination of the southeast boundary as supported by insufficient evidence in the record as a whole.
C. Southern boundary: The Commission placed the southern boundary of the Spokane country along a straight line running slightly southwest from Pine City to Ritzville. We think that this sector is adequately supported by substantial evidence. It accords, in general, with the southern boundary estimated by several observers,8 although it is more restricted than that espoused by others.9 We do not have in this instance the overwhelming weight of the evidence massed strongly against the Commission’s line, as we have found to be the case for the northeastern and southeastern boundaries. We have, instead, a good body of materials both for and against the decision below. With the scales so balanced, we cannot upset the ruling.
In an effort to undermine that finding, appellant attacks, on the one hand, the Commission’s reliance on Wilkes’ map of 1841 as well as its reference to the Spokanes’ limited claim at the Grande Ronde meeting of the area Indians in 1854; on the other, appellant challenges the Commission’s refusal to accept Governor Stevens’ more southerly line of 1857. Whatever their merit, these and like arguments fall into the class of contentions properly addressed to a fact-finder; they do not rise to the level of demonstration that the Commission’s boundary is bereft of any substantial support in the record. In the presence of such substantial support, we cannot evaluate for ourselves the competing arguments.
D. Western boundary: For the western edge of the Spo-kanes’ land, the Commission drew straight lines slightly northeast from Ritzville (the southwestern point of its boundary) to Davenport and then somewhat sharply north*68west from Davenport to Peach. This cuts off a substantial wedge of territory which would be included if the line ran straight north from Ritzville to Peach (as well as any territory to the west). The Commission’s reason for excluding this wedge was that it believed the area not to be occupied exclusively by the Spokanes but to be jointly used together with other natives of the region (the Nespelem and Sanpoil Indians). Not inconsistently, the Commission had found, in an earlier case, that the eastern end of the Sanpoil area of exclusive occupancy extended from a point about 15 miles west of Davenport (the central point of the wedge) to just west of Peach (the northern point of the wedge). Confederated Tribes of the Colville Reservation v. United States, 4 Ind. Cl. Comm. (Docket No. 181) 151, 162 (Fdg. 15) (1956). Appellant disputes this finding that the land between the Sanpoil line and the Commission’s western Spokane line was used in common and was therefore not held by the Spokanes under Indian title.
The Government’s brief supports this finding on two pieces of evidence — (a) a statement of Dr. Edward Curtis in a 1911 study of the Spokanes, and (b) the testimony and report of appellee’s expert, Stuart Chalfant. The Curtis statement, as appellant points out, is not at all relevant to our question since it deals only with the common use of certain territory by the different groups of Spokanes, not with a common use by the Spokanes and alien tribes. Chal-fant, however, does say that the Spokanes used the omitted area (or part of it) in common with other tribes, but his evidence demands rigorous analysis since there is very little other material in the record supporting this contention. See supra. The Commission’s western line, we find on examination, cannot be rested on Chalfant’s view. He relied almost wholly on information supplied by members of neighboring tribes (Nespelem, Sanpoil, Colville) that in the 19th century they had seasonally used land in the direction of the disputed area to get food. He was quite unsure, however, of the limits of this use. Nor does his evidence adequately consider whether this use by other Indians was by permission and at the sufferance of the Spokanes, or as a matter of right; if the former, the alien visits would not diminish *69the appellant’s Indian title. Though Chalfant’s report declares that there was no such host-visitor relationship, that summary statement is obviously infected by his erroneous assumption (mentioned previously) that true Indian occupancy could not exist without more-or-less permanent sites or villages. The essence of his position seems to be simply that there were other Indians in the area at some times of the year, but he did not show what lands they used or under what conditions they happened to be there. This is too slim a base for the affirmative finding that the other natives joined the Spokanes in common, equal, use of the entire wedge excluded by the Commission. To this defect must be added the significant fact that, to a substantial extent, the ruling below cuts off parts of the wedge which Chalfant, even with his too-narrow position on Indian occupancy, was still willing to assign to the appellant as its own. The Commission, in sum, had no substantial evidence leading it to find that all the excluded area — the omitted wedge of territory, as well as the land further west — was jointly occupied together with other Indians. Chalfant’s material is gravely insufficient, and there is nothing else of telling significance.
This does not dispose of the point. The particular western boundary chosen by the Commission may fall for lack of a substantial grounding, but since appellant has the burden it camiot prevail unless there is adequate evidence that it occupied some lands to the west of that boundary. This record does contain solid material to that effect. Several students or observers have given substantially all the excluded wedge or still greater territory to the Spokanes.10 These separate items link to form a strong chain of evidence. In this connection, we note that we do not value the testimony of appellant’s expert, Dr. Eay, as highly as appellant would have us, since his general theory of Indian title seems inadmissibly broad as Mr. Chalfant’s appears too narrow.11 *70But the combined impact of the evidence as a whole forces the conclusion that some area beyond the Commission’s western line should have been included.
II
We have found that the Commission erred in setting the northeastern and southeastern line, as well as the southern and central portions of the western boundary, of the Spokane country. To this extent the Commission’s border is not supported by substantial evidence and must therefore be set aside. The next problem which immediately emerges is whether this court should establish the correct lines or should remand the case to the Commission to perform that function.
In the federal system it has been the usual rule that, where the reviewing court exposes the legal error in the decision of an administrative agency, it should not decide for itself those remaining issues within the agency’s special competence, unless there could be only one answer; if the agency would have room to choose, the court should remit the case to allow that discretion to be exercised. Cf., e.g., Federal Power Commission v. Idaho Power Co., 344 U.S. 17, 20-21 (1952) ; Securities & Exchange Commission v. Chenery Corp., 332 U.S. 194, 199-201 (1947); Jacob Siegel Co. v. Federal Trade Commission, 327 U.S. 608, 613-614 (1946); Ford Motor Co. v. National Labor Relations Board, 305 U.S. 364, 372-374 (1939); Crolley v. Tatton, 249 F. 2d 908, 911-912 (C.A. 5, 1957); Public Utility Dist. No. 1 v. Federal Power Commission, 242 F. 2d 672, 683, (C.A. 9, 1957); Local 1229, Int'l Bhd. Elec. Workers v. National Labor Relations Board, 202 F. 2d 186, 189 (C.A.D.C., 1952), rev'd on other grounds, 346 U.S. 464 (1953).
This court has followed that standard on factual or discretionary questions brought here from the Indian Claims Commission. See Osage Nation of Indians v. United States, 119 Ct. Cl. 592, 667, 672, 97 F. Supp. 381, 422, 425 (1951), cert. denied, 342 U.S. 896 (1951); Pawnee Indian Tribe of Oklahoma s. United States, 124 Ct. Cl. 324, 373-374, 386, 401, 404, 412-413, 109 F. Supp. 860, 889, 896, 904, 906, 911 (1953); *71Snake or Piute Indians v. United States, 125 Ct. Cl. 241, 255, 269-270, 286-287, 112 F. Supp. 543, 552, 560, 569 (1953); Miami Tribe of Oklahoma v. United States, 146 Ct. Cl. 421, 469, 175 F. Supp. 926, 953-954 (1959); Yakima Tribe v. United States, 158 Ct. Cl. 672, 682-83 (1962). Congress has designated the Commission to “hear and determine” the claims (25 U.S.C. § 70a); this court’s function is that of review (25 U.S.C. § 70s) .12
In this case, a remand is plainly in order for the northeast and southeast boundaries. Although the Commission has been too restricted in drawing these lines, the record does not constrain any particular degree of expansion or require us to decide that the eastern borders must be placed at this or that precise point or region. All we hold is that the Commission’s present lines are unsupported. Further proceedings will be needed to determine how far eastward those lines should be pushed.
In the west, the situation is somewhat but not altogether different. As we have indicated, all the substantial evidence points to coverage of at least the entire wedge bounded by lines running from Peach to Davenport to Eitzville and back to Peach. The unresolved problem is whether the boundary should be placed west of the straight line running south from Peach to Eitzville. Some of the materials indicate that the Spokanes held Indian title for various distances beyond that line, while other evidence suggests that the line is itself the proper western edge. On that issue the record before us does not coerce a particular answer. We hold, therefore, that the appellant’s area must be defined to reach westward at least as far as the line from Peach to Eitzville, but at the same time we leave to the Commission the question of whether it went beyond that line and, if so, to what extent.
*72III
The Commission’s findings, as amended, declare that appellant “is entitled to institute this action in a representative capacity on behalf of all such survivors or descendants of the membership of the Spokane Tribe as it existed July 13, 1892.” The Commission’s order (also as amended) states that appellant “is entitled to prosecute this action before the Commission on behalf of all the members of the Spokane Tribe as it existed July 13, 1892, or the descendants of said members.” Both parties attack these references to survivors and descendants of the tribal members of 1892. Recently, in Minnesota Chippewa Tribe v. United States, 161 Ct. Cl. 258, 270-72, 315 F. 2d 906, 913-914 (1963), we held a similar declaration by the Commission erroneous as a matter of law. The appellant does not sue on behalf of the individual survivors and descendants of the Spokane Tribe as it was constituted in 1892, but on behalf of an entity or entities. To reflect this principle, the parties are now agreed that the pertinent parts of the Commission’s finding and order should be modified to read: “The petitioner, the Spokane Tribe, has the right and is entitled to institute this action.” We adopt this proposal and so order. The references to individual survivors and descendants will be excised.
The findings and order of the Indian Claims Commission are affirmed in part and reversed in part, as indicated in this opinion, and the case is remanded to the Commission for further proceedings consistent with this opinion.
Affirmed in part; reversed in part.
In the Commission below, appellee contended (in addition to its argument on tbe extent of the area) that the appellant lacked standing to bring this action under the Indian Claims Commission Act and also that the date of taking by the United States should be earlier than that fixed by the Commission. These issues are not now before us.
In. 1872, an Executive Order reservation (the present Colville Indian Reservation) was created for the non-treaty Indians of northeastern Washington, including the Spokanes. Most of the Spokanes refused to leave their own lands to reside at Colville. In 1881, an Executive Order set aside another tract for the Spokane Tribe; most of the Lower Spokane Band resided within this area, but few, if any, members of the Upper or Middle Bands of Spokanes moved there prior to 1888. There are Spokane Indians now residing upon the Coeur d’Alene, Spokane, Colville, and Jocko Reservations, as well as elsewhere.
There had been about 1400 souls in 1780.
The area found by the Commission is bounded as follows (finding 31) : Commencing on the Columbia River at the mouth of Hunters Creek and running thence up Hunters Creek to the fork thereof, thence by a direct line to Bald Mountain immediately north of Deer Lake, thence eastward around Deer Lake and by direct line southeast to the mouth of Beaver Creek on the Little Spokane River a short distance north of the present town of Milan, Washington; thence southeast by a direct line to the fork of Deadman Creek near the present town of Peone, Washington, and thence southeastwardly to the northeast corner of the present townsite of Opportunity, Washington, on the Spokane River; thence southwest to the mouth of Rock Creek on Latah Creek, and continuing in a southwesterly direction to the southwest corner of the present townsite of Spangle, Washington, the southwest corner of the present townsite of Malden, Washington, the southwest corner of the present town-site of Pine City, Washington, and the southwest corner of the present townsite of Ritzville, Washington; thence northeasterly along a straight line to the southwest corner of the present townsite of Davenport, Washington; thence northwest along a straight line to the southwest corner of the present town-site of Peach, Washington, thence northward to the Columbia River; thence northerly up said Columbia River to the place of beginning at the mouth of Hunters Creek.
In the face of these materials (including systematic anthropological studies), scattered observations that other Indians were at one or another time found at one or another spot east of the Commission’s northeastern line cannot be given any substantial weight. These casual observations do not indicate whether the alien Indians were there as visitors, on. a temporary basis, etc.
The following would give the Spokanes a more extended southeast boundary than the Commission drew: Teit; Spier; Curtis ; Mooney ; Swanton; Oregon Hist. Soc.; Gibbs; Wilkes (semble) ; Kay.
Chalfant submitted a written report and also testified. One factor to be considered in evaluating his evidence, though it is not decisive, is his relative inexperience both In ethnological studies generally and in studying the Indians of this area.
Leslie Spier (whose line is not straight but in its curvature includes slightly less territory than the Commission’s) ; Edward Curtis (whose southern line seems somewhat less favorable to appellant) ; James Mooney; John R. Swanton ; Charles Wilkes ; Stuart Chalfant.
We are indebted to appellant for including in the appendix of its reply brief a series of 13 maps showing the various borders given for the Spokanes’ territory by different observers (including the two live expert witnesses, Dr. Ray and Mr. Chalfant). These maps have been very helpful.
James A. Teit; Governor Isaac I. Stevens; Oregon Historical Society; George Gibbs (semble) ; Verne E. Ray; Winans.
Teit; Spier ; Curtis ; Mooney; Swanton ; Gibbs ; Oregon Hist. Society ; Bay.
Dr. Bay apparently believes that Indian title could be obtained by mutual recognition between neighboring tribes that a certain area “belonged” to one or another of them — even though that “owner” did not use or occupy the land assigned to it. This does not accord with the type of Indian title which the Indian Claims Commission Act takes into account. See supra.
It is only wliere the court believes that a particular finding or disposition is compelled by the record that it decides the case on. the basis of that finding or disposition, or orders the Commission to so find. M.g., Chitto v. United States, 133 Ct. Cl. 643, 659-661, 138 F. Supp. 253, 263-265 (1956), cert. denied, 352 U.S. 841 (1956) ; Quapaw Tribe of Indians v. United States, 128 Ct. Cl. 45, 55-79, 120 F. Supp. 283 (1954) ; Miami Tribe of Oklahoma v. United States, 150 Ct. Cl. 725, 281 F. 2d 202 (1960), cert. denied, 366 U.S. 924 (1961).