To show that Luis Peralta's title was a perfect one the plaintiff produced in evidence the documents on which it was founded. They are set out in the bill of exceptions, and are the same that were before this court in the case of United States v. Peralta,2 when the claim was confirmed. In that case the court intimated an opinion that the title was perfect for at least a part of the rancho (embracing a part of the premises now in question), but the point was not material in the case, because the claimants were equally entitled to a confirmation, whether their father's title was perfect or imperfect, legal or equitable; so that the intimation was nothing but an obiter dictum of the judge who delivered the opinion. The title, in some of its aspects, again came before the Supreme Court of California, in 1864, in the case of Minturn v. Brower,3 but, as both parties in that case deemed it their interest to concede the title to be a perfect one, the observations of the court on the subject cannot be regarded as precluding further examination. Such examination, exhaustive in its character, was given in 1870 by the same court on this identical title, and on the very point in question, in the case of Banks v. Moreno;4 and the court, with all the documents before it which have been proven in this case, decided that the title was imperfect. If this were a case depending merely on the local land laws of California, we should be bound by that decision. But as the appellant, in case the title is adjudged a perfect one, invokes the guaranty stipulations of the treaty of Guadaloupe Hidalgo in his favor, independent of any action of the commissioners, the question ceases to be a mere local one, and devolves upon this court the duty of deciding it on its merits. An examination, however, of the reasoning of the Supreme Court of California, in the case last cited, satisfies us of its soundness. The point of the decision is, that the rancho of San Antonio never had any clearly defined boundary on the east. In this we concur with that court. The new claim now made to extend that boundary beyond the crest of the mountain, and to take in the eastern slope on the pretence that the Leandro Creek is the boundary to its ultimate source, is itself conclusive to show the uncertainty with which it has always been invested.

Luis Peralta's occupation of the rancho goes back to 1820. In that year he presented to Governor De Sola his petition for a grant, describing the tract as follows: 'At the distance of eight leagues from the mission of San Jos e, in a northerly or northwesterly course, along the coast, there is a creek named by the reverend fathers of the aforesaid mission, San Leandro, and from this to a little hill adjoining the seabeach, in the same direction and along the coast—there may be four or five leagues more or less, or about—which place and land he asks and solicits may be granted to him that he may establish a rancho.' Here, certainly, is nothing definite. Supposing the creek, San Leandro, as the point of beginning, and the little hill four or five leagues beyond, as fixed and ascertained points; and suppose the shore of the bay on the west to be meant for the boundary on that side; there is no hint of a boundary on the east. Nor is the quantity specified. Had that been done, perhaps it might have enabled a surveyor to fix a boundary by relation. This is the first and original document on which the title is based—the foundation of all the rest.

Upon this petition, the governor, by an order of August 3d, 1820, directs Captain Arguello to appoint an officer to put Sergeant Luis Peralta in possession of the lands petitioned for, and to 'place landmarks on the four points of the compass, that it may be known at all times the extent of said lands which have been granted to him.' Lieutenant Martinez being detailed for this service, on the 16th of August, 1820, reports his action as follows: 'The boundaries which separate his land were marked to him, to wit: The deep creek called San Leandro, and at a distance from this (say about five leagues), there are two small mountains (cerritos). The first is close to the beach; next to it follows that of San Antonio, serving as boundaries, the rivulet which issues from the mountain ranges, and runs along the foot of said small mountain of San Antonio, dividing or separating the land; and at the entrance of the little gulch there is a rock elevating itself in the form of a monument, and looking towards the north. On both boundaries were fixed firm landmarks . . . I put in possession of the said land the above named Luis Peralta.' Here we have, again, the two extremities of the tract along the bay, the creek San Leandro, at one end, and the rivulet that runs by the cerritos, at the other, and nothing more.

Next we have a complaint of the fathers of the San Francisco mission, that Peralta has been put in possession of a portion of their land at the north end of the tract; the result of which is that Peralta is limited, on the north, to the Temescal Creek, or Willow Grove Creek, about a league and a half south of the cerritos. This occurred in September, 1820.

On the 18th of October an entry was made in the public records to the effect, that 'this day was issued in favor of Sergeant Luis Peralta, by the governor of this province, the certifying document for the land which has been granted to him, as appears in this folio by the writ of possession, which the lieutenant of his company, Don Ignacio Martinez, gave him agreeably to an order issued by the government.' We also have the certifying document itself of the same date, which adds nothing to the definiteness of the description.

Now the grant on which the appellant's counsel relies as conferring perfect title is not the certifying document above referred to, but the previous act of directing possession to be given to Peralta, and the actual delivery of possession to him. It is perfectly manifest that Peralta could not have been put into manual possession of several leagues of land. He could only have been put into possession of a certain part or parts in the name of all; and the exterior boundaries of the tract must have been indicated by language or monuments. But we have no evidence of any description of boundaries, or monuments to designate them, except the bay on one side, and the extreme limits of the tract along the bay. The interior line between those limits is entirely wanting in all the documents thus far presented. The title relied on, therefore, is necessarily imperfect, and requires some authoritative survey to distinguish what was intended to be granted from what remained in the public domain.

If we examine the remaining documents we shall not derive any material aid to help us out of the difficulty.

In October, 1820, Peralta addressed a remonstrance to the governor against the curtailing of his tract on the north. The only expression which this paper contains going to show what the tract was which Peralta supposed was granted to him, are the following: 'The reverend father says to the honorable governor that I do not need the land, and that I occupy a great extent; but I would represent that five leagues does not seem to me much in a narrow tract, as you know it is, from the beach to the mountain range, and that not all of it is good, as my lieutenant is aware, for great portions contain hills, creeks, and ravines, not fit for the purpose.' This would seem to indicate that the rancho extended from the bay to the foot of the mountain.

In 1823, whilst the revolution was in progress, Peralta's captain, Arguello, had become Governor of California, and Peralta renewed his application to have the curtailment of his rancho annulled. He speaks of the tract which he originally applied for, as follows: 'Which tract of land, though it appears to be large is not so, for two reasons: 1st. Because it is situate on the coast, and the space between the beach and the top of the mountain is too narrow.' This would indicate the top of the mountain as his supposed boundary. The governor promptly made an order that the part which had been taken from him should be restored, and Lieutenant Martinez put him in possession accordingly; but nothing yet appears in the lieutenant's return or elsewhere to identify or fix the eastern boundary of the rancho, much less to fix it beyond the eastern slope of the mountain, as since claimed by the parties.

In 1827 some new regulations made it necessary for every proprietor to make a return of all lands occupied by him, with the titles annexed; and, in December of that year, Peralta made a return accordingly, describing his rancho as follows: 'Along the coast of the mission of San Jos e, in a northwesterly course, there is a deep creek called San Leandro, forming the dividing boundary of said mission of San Jos e; thence to a small round mountain called San Antonio, the dividing boundary with my neighbor Francisco Castro; which space is a little over four leagues long, and, as it is the narrowest portion of the coast, it at most contains half a league in breadth from the mountain to the sea.'

In 1844 Ignacio Peralta, on behalf of his father, whose title-papers he says were mislaid, petitioned the then governor, Micheltorena, to order the issue of a new title, extending to the top of the range, and accompanies his petition with a dise no, or rough map of the property. The governor ordered a grant to issue, as requested, extending to the top of the hill range, but not to prohibit the inahbitants of the Contra Costa from cutting wood for their own use. This order was not signed by the governor, and seems never to have been carried into execution. And this is the last of the documents on which the plaintiff, the now appellant, relied for a perfect title. Leaving out the proceedings of 1844, which are admitted to be imperfect, no human being can tell, from the language of the various documents, what was the eastern boundary of the rancho. It certainly would seem not to embrace the eastern slope of the hills, as is now claimed; but what it did embrace, or where it did run, is not ascertainable from any of the documents which have been adduced; and no parol testimony can aid this defect as regards the question now under consideration. Parol testimony was very properly adduced before the commissioners for the purpose of showing where equity required that the line should be run, in order to separate the rancho from the public domain. But it cannot make that title perfect which was not perfect before.

The Supreme Court of California, in Banks v. Moreno,5 well observed: 'The precise point under discussion is, whether or not the title of Peralta, as exhibited by the plaintiff, was a perfect title conveying the fee, and which invested him with absolute dominion over a specific parcel of land without any further action on the part of the United States; or whether, at the time of the cession of California, something remained to be done by the government which was necessary to invest Peralta with a complete legal title to the specific tract.

'In every complete grant conveying a perfect title it is essential that the thing granted be sufficiently described to enable it to be identified. In grants of real estate it is not always necessary to describe it by metes and bounds, or by a reference to actual or artificial monuments, nor by courses and distances. If the tract granted have a well-known name, and the boundaries of the tract known by that name are notorious and well-defined, a grant of the tract by its name would, doubtless, convey the title to the whole. In like manner, a grant describing the tract by reference to the known occupation of the grantor or another—or to another instrument containing a sufficient description of the premises—would be sufficient. In short, any description will suffice which identifies the land granted with such certainty that the specific parcel intended to be granted can be ascertained either by the calls of the instrument, as applied to the land, or by the aid of the descriptive portions of the grant. But it is equally certain that to constitute a complete and perfect grant to a specific parcel of land, it must, in some method, appear on the face of the instrument, or by the aid of its descriptive portions—not only that a specific parcel was intended to be granted, but it must also be so described that the particular tract intended to be granted can be identified with reasonable certainty. It would be a contradiction in terms to say that a specific tract was granted if there was nothing in the grant by which it could be ascertained with reasonable certainty what particular parcel was intended to be conveyed.'

We entirely concur in these views; and, therefore hold that the title of Peralta was an imperfect title, and necessarily required confirmation in order to vest a full legal estate in private parties.

But it is contended that the confirmation of the title enured to the benefit of the parties really interested, both at law and in equity, and not merely to the benefit of the confirmees. This is undoubtedly true so far as the segregation of the lands from the public domain and the extinguishment of the government title or claim of title is concerned; but as it respects the legal estate, the confirmation enures to the confirmees alone. The eighth and ninth sections of the act require the claimant to show not only the original title, but his own title by deraignment therefrom. Having established these the object of the inquest is attained. It satisfactorily appears that the land does not belong to the government, and the claimant appears to be the person prim a facie entitled to the legal title. Hence the thirteenth section goes on to declare that for all claims finally confirmed a patent shall issue to the claimant upon his presenting to the general land office an authentic certificate of such confirmation and a plat or survey of the said land duly certified and approved by the surveyor-general of California, whose duty it shall be to cause all private claims which shall be finally confirmed to be accurately surveyed and to furnish plats of the same.

This language is utterably irreconcilable with the hypothesis that the legal estate devolves, upon the confirmation, to any other parties than the confirmees. The patent is to be given to them, and the legal title cannot be separated from the patent.

It is true that the fifteenth section of the act declares that the decree of confirmation shall be conclusive between the United States and the claimants only, and shall not affect the interests of third persons. But this was intended to save the rights of third persons not parties to the proceeding, who might have Spanish or Mexican claims independent of or superior to that presented by the claimant; or the equitable rights of other parties having rightful claims under the title confirmed. The former class could still present their claims without prejudice within the time limited by the statute. The latter class, those equitably entitled to rights in the land under the title confirmed, were not to be cut off. Their equities were reserved. But they must seek them by a proceeding appropriate to their nature and condition. The legal title is vested in the confirmees, or will be when the requisite conditions are performed. It is not in these equitable claimants. They cannot maintain an action of ejectment against the confirmees, or those claiming under them; but must go into equity, where their rights can be properly investigated with a due regard to the rights of others. Had the daughters as well as the sons of Luis Peralta gone before the commissioners, it is possible that they would have participated in the legal advantages of the confirmation. It may now be inequitable on the part of the sons to withhold from them a due share of their father's estate. But other rights may have grown up in the meantime, rights of bon a fide purchasers and others ignorant of the equities existing between the original parties, which it would be unjust to disturb. These questions can be much better examined in an equitable proceeding than they can be in this action, in which, indeed, they are entirely inadmissible.

This view of the relative position of the parties is supported by the weight of authority. The case of Wilson v. Castro6 is directly in point to show the form of proceeding proper for those who claim against the confirmee. In that case the claim was confirmed to the widow, who really had no interest. The brother and sister of the owner, as his heirs at law, brought a suit in equity against the widow, and obtained a decree declaring her to be seized, as trustee, for their use. In Estrada v. Murphy,7 the court says: 'A court of equity will control the legal title in his [the confirmee's] hands, so as to protect the just rights of others. But in ejectment the legal title must prevail;' and it decided the case accordingly against the plaintiff in ejectment. In Banks v. Moreno,8 the same conclusion was reached. In that case, as in this, the plaintiff claimed under the daughters of Luis Peralta; the defendant under the sons; and it was held that the action did not lie. The same view was taken by this court in Beard v. Federy,9 and Townsent v. Greeley.10 In the last case the court uses this language: 'The confirmation only enures to the benefit of the confirmee so far as the legal title is concerned. It establishes the legal title in him, but it does not determine the equitable relations between him and third parties.'

The case is somewhat analogous to that of patents granted upon a pre emption right for public land. Whilst the patent in that case confers the legal title, and admits of no averment to the contrary, the patentee may be subjected in equity to any just claim of a third party, even to the extent of holding the title for his sole use. The grounds of equitable jurisdiction in such cases are stated in the opinion of this court in the recent case of Johnson v. Towsley.11

The action of ejectment in this case cannot be maintained. The judgment of the Circuit Court is

AFFIRMED.