(after stating the facts as above). [1] The question to be determined in this case is the westerly boundary of the Yakima Indian reservation as described in the treaty of June 5, 1855 ( 12 Stat. 952 ). -

It will be convenient to repeat the description contained in the treaty, identifying its calls by numbers as follows:

(1) Commencing on the Yakima river, at the mouth of the Attahnam river.

(2) Thence westerly along said Attah-nam river to the forks.

(3) Thence along the southern tributary to the Cascade Mountains.

(4) Thence southerly along the main ridge of said mountains, passing south and cast of Mt. Adams, to the spur whence flows the waters of the Klickitat and Pisco rivers.

(5) Thence down said spur to the divide between the waters of said rivers.

(61 Thence along said divide to the divide separating the waters of the Satas river from those flowing into the Columbia river.

(7) Thence along said divide to the main Yakima, eight miles below the mouth of the Satas river.

(8) And thence up the Yakima river to the place of beginning.

There is no controversy as to the location of the point “commencing on the Yakima river, at the mouth of the Attah-nam river,” nor is there any controversy as to the second call, “westerly along said Attah-nam river to the forks,” nor as to the third call, “along the southern tributary”; but the controversy begins with the terminal point of this call, “the Cascade Mountains.”

The call is limited by the Schwartz survey to the southern boundary of the Attah-nam river and terminates where that tributary takes its rise in a range of the Cascade Mountains. But it is now known that this is not the main range of the Cascade Mountains, and that the southern tributary of the Attah-nam river does not take its rise in that range. In the general description of his survey Schwartz reported that this range along which he ran the western boundary of the reservation “is not the main range of the Cascade Mountains.” His instructions were that the western boundary line coming up from the south should he “the main ridge of the Cascade Mountains, thence northerly along said ridge, passing south and east of Mt. Adams, to the southern tributary of the Attah-nam river.” He was further instructed that the question as to which was the main ridge of the Cascade Mountains referred to in the treaty was a subject of disagree *954 ment; the Indians claiming that the main ridge extended to the base of Mt. Adams on the south and east, and white men with diverse interests claiming the said ridge to be further east. He was further instructed :

Schwartz appears to have consulted with the Indian agent in charge of the Yakima agency, and with other white persons and Indians, and found and identified the main ridge of the Cascade Mountains. But he did not carry his survey.along that ridge, as instructed, btit along a ridge 15 to 20 miles further east. His reason for running his line along the eastern ridge instead of the main ridge to the west was that the former was a ridge dividing the waters of the Satas and Klickitat rivers, and along which he could carr)*- his line to the source of the southern tributary of the Attah-nam river “without crossing the Klickitat river, and the treaty did not call for that.” But his instructions did not require him to follow that ridge or to avoid crossing the Klickitat river. On the contrary, he was instructed to ascertain and follow the main ridge of the Cascade Mountains. Nor did any call of the treaty require him to follow the ridge dividing the waters of the Satas and Kickitat rivers.

This brings us to the consideration of one of the controlling questions in this case. Between the main ridge of the Cascade Mountains, along which the fourth call of the treaty locates the western boundary and the apparent terminal point of the third call of the treaty at the source,of the Attah-nam river in the Cascade Mountains, there is a gap of 15 to 20 miles; that is to say, the third call of the treaty following the southern tributary of the Attah-nam river does not reach the main ridge of the Cascade Mountains, but appears to terminate in an inferior eastern ridge of the Cascade Mountains. Between that ridge and the summit of the main ridge there is a distance of from 15 to 20 miles, and a straight line between these two points leaves the eastern ridge and crosses the headwaters of the Klickitat river to the main or western ridge. Is this gap in the description of the treaty boundary of such a character as to defeat the western boundary along the main ridge of the Cascade Mountains? We think not. Gov. Stevens, who negotiated the treaty with the Indians, had a map made in 1857, two years after the making of the treaty and two years before its ratification, showing the location of the Indian nations and tribes of the territory of Washington and the territory of Nebraska west of the mouth of the Yellowstone. On this map is a tracing of the Yakima Indian reservation which clearly carries the western boundary of the reservation to and southerly along the main ridge of the Cascade Mountains, passing south and east of Mt. Adams as required in the fourth call of the treaty. The map is, however, incorrect, in this, that it appears to locate the rise of the southern tributary of the Attahnam river in the main ridge of the Cascade Mountáins. There is in *955 the record a copy of another map, the original of which is said to have been deposited at the Yakima agency, having, it is alleged, been given by Gov. Stevens to an Indian named White Swan; hence the map is named the 'White Swan Map.” This map shows the boundaries of the Yakima reservation and appears to be a reproduction, in part at least, of the tracing of that reservation as it is shown on the larger map and transmitted to Washington by Gov. Stevens in 1857. On ¡his map the northern boundary of the reservation follows the southern tributary of the Attah-nam river to a point manifestly on the main ridge of the Cascade Mountains, thence southerly along the main ridge for a short distance, when the boundary turns eastwardly passing south and east of Mt. Adams.

The error in these maps in locating the head or source of the .¿southern tributary of the Attah-nam river in the main ridge of the Cascade Mountains is not difficult of explanation. The Cascade Mountains is a range of considerable length and elevation, extending north and south through Oregon, Washington, and British Columbia, and run-, niug. nearly parallel to the Pacific Coast line. In the state of Washington this range forms an elevated plateau with a general elevation of from 5,000 to 7,000 feet, rising into a rugged and complex order of ridges and peaks from 50 to 100 miles in width. Along the main axis of the range are a number of high peaks, among othei s M t. Adams, with an elevation of -12,325 feet. This mountain with its connecting ridge and spurs is a prominent feature in the landscape, and would be a natural monument, and the connecting main ridge of the Cascade Mountains a natural boundary for such a large tract of land as the Yakima Indian reservation described in the treaty of 1855. But at the time this treaty was made the country had not been fully explored, the approaches to this main ridge by water courses had not been definitely located, and the topography of the country was not accurately known. Mr. Barnard in his report dated January 12, 1900, concerning the disputed western boundary line of the reservation, says:

With the information available to Gov. Stevens he supposed the source of the Attah-nam river to be in the main ridge of the Cascade Mountains, and he so placed it upon his map. Now, while this has since been discovered to be a mistake, the fact remains that he located the western boundary of the reservation on the main ridge of the Cascade Mountains, and as this is the most certain and material call of the description it should prevail. This is in accordance with the general rule that has been adopted by the courts for solving such a question.

In Newsom v. Pryor, 7 Wheat. 7 , 5 L. Ed. 382 , the Supreme Court of the United States had before it a similar question, and the opinion of the court, delivered by Mr. Chief Justice Marshall, is peculiarly applicable to the facts in the present case. He says:

Under this rule distance must be disregarded and the more material and certain call accepted.

But it is .contended by the appellant that the Barnard survey recognized by Congress in the Act of December 21, 1904, as locating the boundaries of the reservation in accordance with the treaty of June 9, 18S5, and followed by, the court below in its decree, is purely arbitrary; that the lines so recognized and adopted do not follow the treaty calls from the head of the south fork of the Attah-nam river, but, on the contrary, it is contended they are in direct collision with such calls. This objection is based upon the claim that by the fourth call the line from the point where the Cascade Mountains were reached was to run southerly along the mountains and divides separating the waters of the Klickitat and Pisco rivers. The language of the fourth call is as follows: .

If we are correct in our opinion that the topography of the country was not definitely known or correctly understood by the framers of the treaty, this objection is without any serious force, if, indeed, it is not completely answered by the fact that on the Stevens map the Klickitat river is shown as taking its rise on the west side of the main Cascade Range and flows south on the west side of Mt. Adams, while the Pisco is there shown on the east side, and according to Barnard’s report it was believed that it reached the summit of the Cascade Mountains. If this was the information possessed by the framers of the treaty and in accordance with that information they were running this call southerly passing south and east of Mt. Adams to a spur on that mountain, then the Barnard survey locates the fourth call of the *957 western boundary in accordance with the understanding and intention of the parties to the treaty.

The United States might object to the Barnard survey on the ground that it does not in fact reach the ridge of the Cascade 'Mountains for a beginning of the fourth call, but starts on a ridge a little to the east of the main ridge. As this feature of the boundary does not, however, affect any of the land in suit and the objection is not made, we do not stop to discuss that” question.

The fourth call of the treaty terminates, as we have said, on a spur of Mt. Adams which has been reached by a line passing south and east of the mountain. Mt. Adams is a prominent peak in that region and a monument easily identified, and a spur on the southeast side would also be easily identified, and the evidence appears to support that conclusion.

In Barnard’s report he says:

Barnard in commenting upon this call of the boundary says:

The Schwartz survey does not pretend to reach Mt. Adams or any spur of that mountain, but turns aside at the fifty-first milepost and proceeds northerly to avoid crossing the Klickitat river. We think the superior monument of Mt. Adams should prevail here as the more certain call of the main ridge of the Cascade Mountains in the third call, and for the same reasons.

The fifth call of the treaty is as follows:

And the sixth call continues:

*958 The last call carries the boundary to, and perhaps beyond, the fifty-first milepost mentioned in the Schwartz survey, where the controversy ends.

If we are correct in following the language of the treaty and locating the western boundary of the reservation along the main ridge of the Cascade Mountains to a spur of Mt. Adams, reached by passing south and east of that mountain, then the Barnard survey must be accepted as properly locating the fifth and sixth calls, and the objection that the divide mentioned in these two calls is not continuous and is broken by the crossing of the Klickitat river and its tributaries yields, as in the fourth call, to the superior calls of Mt. Adams and the main ridge of the Cascade Mountains.

[2] It is next contended that, if there is a doubt as to the location of the boundaries called for by the treaty, that doubt should be resolved in favor of the patents issued to the Northern Pacific Railway Company and its predecessor in interest. We do not admit that there is any dtoubt as to the proper location of the western boundary of this reservation; but, assuming that there is, the rule which resolves such a doubt in favor of the patent issued by the United States does not obtain in this case.

[3] The United States brings this action for itself and on behalf of the Indians as their trustee and guardian. The Yakima Indian reservation is a tract of land reserved by the Indians out of a much larger tract claimed and occupied by them prior to 1855, and which larger tract they ceded andl conveyed to the United States with an agreement with respect to the reserved tract that it should be “set ■apart and, so far as necessary, surveyed and marked out,for the exclusive use and benefit of said confedei'ated tribes and bands .of Indians as an Indian reservation.”

The United States owed the duty of trustee andl guardian to preserve the rights of the Indians in the tract reserved, and by the treaty it agreed to survey the boundaries of the reservation, and set it apart for the exclusive use and benefit of the Indians. It could not therefore by an incorrect survey deprive the Indians of their right of occupation of the land within the legal boundaries of the reservation as established by the treaty. This is the general and well-established law of trust andl guardianship in support of which authorities need not be cited. The law has been applied to controversies relating to lands occupied by the Indians in the broadest terms.

In Leavenworth, etc., R. R. Co. v. United States, 92 U. S. 733 , 23 L. Ed. 634 , the-railroad'company claimed title to certain lands within the Osage country in Kansas. -These lands had been certified by the Commissioner of the General Land Office upon the approval of the Secretary of the Interior, and by such certificate conveyed to the ■ Governor of Kansas as forming part of a grant made by Congress in 1863 to the state to aid in the construction of certain railroadis. Act March 3, 1863, c. 98, 12 Stat. 772 . The lands were conveyed by the Governor by patent to the railroad company. At the time of the grant by Congress to the state of Kansas the lands were occupied by the Osage Indians, but subsequently the Indians conveyed these *959 lands by treaty stipulation to the United States. It was contended by the United States, in support of its- suit to establish title to' these lands,.that Congress had not disposed of the Osage lands by the act of 1863, and had not intended to do so. The railroad company contended that, although the grant did not operate upon any specific tracts until the railroad was located, it then took effect upon those in controversy, as, by reason of the extinction of the Osage title, the lands had become, in the proper sense of the term, public lands. The Supreme Court denied the contention of the railroad company and held that:

It was also held in that case that:

It was further said that:

In the late case of Stewart v. United States, 206 U. S. 185 , 27 Sup. Ct. 631, 51 L. Ed. 1017 , referring to the sale of these lands under the treaty with the Osage Indians, the Supreme Court said:

In Minnesota v. Hitchcock, 185 U. S. 373, 389 , 22 Sup. Ct. 650, 656 ( 46 L. Ed. 954 ), the Supreme Court held that:

[4] The Indian’s right of occupancy to the entire tract within the boundaries of the reservation as described by the treaty of 1855 had not been extinguished by the United States, at the time of the grant of lands to the Northern Pacific Railroad Company by the Act of July 2, 1864 ( 13 Stat. 365 ). That act granted certain sections of land on the line of the road to which the United States had “full title” and which had not been “reserved, sold, granted, or otherwise appropriated and free from pre-emption or other claims or rights at the time the line of said road was definitely fixed and a plat thereof filed in the office of the Commissioner of the General Land Office.” To the lands between the Schwartz and Barnard surveys the Indians had a “right of occupancy” which, by the express terms of the grant to the railroad company, excluded it from that grant, and the*' officers *960 of the Land Department were without authority to issue patents to the railroad company for any of the lands within the reservation. Leavenworth, etc., R. R. Co. v. United States, 92 U. S. 733, 739 , 23 L. Ed. 634 . The patents so issued were therefore issued by the Land Department through inadvertence and mistake and without the fault, concurrence, or acquiescence of the Indians, or the United States, their guardian and trustee, and should be canceled.

[5] The appellants claim that they were bona fide purchasers, and as such entitled to protection under the Act of March 2, 1896, c. 39, 29 Stat. 42 (U. S. Comp. St. 1901, p. 1603); and that under section 8 of the Act of March 3, 1891, c. 561, 26 Stat. 1095 , 1099 (U. S. Comp. St. 1901, p. 1521), the statute as to all patents issued more than six years prior to the commencement of the action was barred. In our opinion these statutes relate to patents issued for lands within the public domain, and the contention of the appellants is answered by the conclusion we have reached — that the lands within the Yakima Indian reservation had never been a part of the public domain, and never subject to grant or sale under any statute providing for the disposal of the public lands of the United States.

In the court below the deposition of the author of the Barnard survey was taken on behalf of the United States. He identified the lands mentioned in the bill of complaint as being within the reservation, with certain exceptions, which he said he could only determine upon field examination. In the decree these excepted lands were excluded from its operation, and we think correctly, upon the evidence before the court.

The decree of the Circuit Court is affirmed.