dissenting.—I dissent. Conceding that the state is not a “third person” within the meaning of that phrase as employed in section 15 of the act of 1851, there is but one question left for determination in this case. It is a question simply of construction. It is not a question of collateral attack. The decree and patent must be read together. In fact, we know that it is the practice of the land department to make the decree a part of the patent.
1 The case in effect is one simply where the plaintiff claims that the defendant’s title depends upon a deed describing the property by natural boundaries, and also by metes and bounds; that the description by metes and bounds includes lands not included within the natural boundaries; that defendant claims an interest in the lands lying between the natural and the artificial lines; that plaintiff is the owner of all the lands between the two lines, and his title thereto should be quieted, because the natural and permanent boundary lines given in the deed should prevail over the description by metes and bounds. The city could in no way be prejudiced by a determination of this question. No survey would be required to fix new lines. The work *404that has been done would not be undone. The location of the south line would not be disturbed. To determine the issue here, it would be necessary only to draw upon the earth a line corresponding with the metes and bounds given in the patent; and unless the plaintiff could show that some portion of the land in controversy lies between that line and existing, well-defined, natural, and permanent boundary lines indicating the line of high-water mark referred to in the decree, they would, of course, fail to make out their case.
It cannot be presumed in aid of a demurrer that the shore of the sea which bounds the land in question on three sides has been obliterated. The complaint alleges as a fact, which must be taken as true, that the lands in controversy lie below ordinary high-water mark. If a patent should describe a tract of land as bounded on three sides by well-known, natural, and perpendicular walls, like those of Yosemite, or should describe it as lying in a triangle at the junction of navigable streams, like that adjacent to the confluence of the Sacramento and San Joaquin rivers, and should also give a description by metes and bounds, the former, for the purposes at least of stating a cause of action,—a prima facie case, —would be preferred to the latter. The lands are described in the decree as being bounded on three sides by the bay of San Francisco and the Pacific Ocean. The court cannot say as a matter of law that no portion of the land in controversy, although admitted to be below high-water mark, lies in a portion of the peninsula where the ocean or the bay is not a natural, fixed, and certain boundary line, and that, therefore, the metes and bounds given in the survey are in no way controlled by the natural boundaries. The admission kills the conclusion.
The questions here are not whether the state courts can review the action of the officers of the land department, or whether the land department of the United States could divest the state of her title to the land in *405controversy, or whether the patent protects the grant of the confirmee from collateral attack. The questions are: What is the effect of the action of the land department ? Has the state been divested of her title through this patent, in view of its descriptions and recitals? Which description shall prevail, that by metes and bounds, or that by natural monuments? In determining these questions, there is no violation of the principles which forbid a collateral attack upon such instruments. There is no such attack. Surely, if the surveyor had run his lines through the bay of San Francisco and included lands of the state in the county of Alameda, which she became owner of by virtue of her. relations to the paramount source of title, the state, as such owner of land across the bay, would not be bound by the survey so long as that natural monument and limit of boundary —the bay of San Francisco—remained where it is, a notice to all the world; and the decree of confirmation, which is a link in the chain of title, remained as a public muniment of title.
No case has been cited inconsistent with the rule that in cases of this kind the natural and permanent boundaries will prevail. In many of the cases cited, there was no variance between the decree of confirmation and the patent. In Teschemacher v. Thompson the grant was assumed to be one of quantity only. The grant was confirmed by the court, and its boundaries, as defined in the decree, were followed in the patent, which was issued in November, 1857, at which time the federal courts controlled the surveys, and were empowered and expected to make them conform to the decree.
In none of the cases cited is the question of the power of the officer to issue a patent for land not embraced in the decree considered. More v. Massini, 87 Cal. 432, is directly in point. In that case the patent did not show upon its face that any land below high-water mark was included in the tract granted, but the fact was shown at
*406the trial by witnesses,—surveyors; and it was there held that the clause of the decree of confirmation which confirmed the tract bounded by the sea-shore should prevail over a description by metes and bounds. In the case of Chipley v. Farris, 45 Cal. 528, upon which respondents place their chief reliance, there were two inconsistent descriptions, but both were descriptions given by a surveyor, and it is expressly admitted by respondent in that case, in his written points, that if the side lines of the tract in controversy were the sea-shore, a different rule would apply, and that More v. Massini, 37 Cal. 432, would be in point, the line of high-water mark in such cases being considered by the government the more certain boundary line. Of course, where both descriptions are lines established by surveyors for the decree and for the patent, the last, that one contained in the survey and incorporated into the patent, will control. Where watercourses, mountains, or other natural objects are called for in patents, it has been said that distances must be lengthened or shortened, and courses varied, so as to conform to these objects, because mistakes in distances and courses are more probable and more frequent than mistakes as to trees, rivers, mountains, and other objects capable of being clearly and accurately fixed. (McIver v. Walker, 9 Cranch, 177.)
In the case at bar, three sides of the triangular tract described are bounded by the shores of the ocean and bay. Unless we must shut our eyes to what every resident of the peninsula can see, we know judicially that a great portion of those lines have remained for ages as they now appear; but if they have changed, it is a matter of defense which should be alleged, and which may be easily proved.