delivered the opinion of the court.
The controversy we are called upon to determine in this case is one of long standing, and involves facts and legal principles concerning which there has been for many years much variety of opinion, both in-the department of the federal government and the judicial tribunals, whose action upon it has been invoked.
The conflicting titles in the case are that of the inhabitants of Carondelet to commons, and a claim of Gabriel Cerré lying within its supposed limits. The title of Carondelet to her *502commons originated in 1812; that of Cerré took its origin, for all purposes now material to be regarded, from an act of Congress passed in 1836. A survey, purporting to be a designation of the Carondelet common, was made anterior to 1817, and subsequently retraced in 1834. That of Cerré’s confirmation was made in 1887.
It is manifest that the title of Carondelet is the better title, unless its locality, as fixed by the two surveys alluded to, can be disturbed. The whole controversy, therefore, turns upon the surveys.
On the part of the claimants under Cerré, it is insisted that Rector’s survey, which was made previous to 1817, was unauthorized by law, or by the orders of his principal; that it was a private and unofficial survey and was never approved ; and that Brown’s survey in 1834, which purported to be a resurvey of Rector’s, was disapproved and set aside by the executive department of the government which had the supervision of this branch of the public service; and that its ultimate approval in 1855 was conditional and qualified, and gave the title based on it no efficacy in opposition to that of Cerré. If these propositions can be maintained, it must follow that the claimant under Carondelet is not entitled to recover.
We do not think it necessary to determine, in this case, whether the commissioner of the General Land Office, prior to the act of July 4, 1836, could legally exercise any supervision or control over the surveys executed and approved by the surveyor general of Illinois and Missouri. Such a power may be very proper, perhaps necessary, to the performance of the duties confided to that bureau by the act of Congress which established it in 1812. Although no express provision has been found giving the land office or its commissioner such control, it may be a necessary implication from powers which were expressly given. As the conclusion we have reached is not inconsistent with the concession of such a controlling and revising power in the commissioner, we have not deemed it necessary to examine the subject more *503particularly. No such power was attempted to be exercised in reference to Rector’s survey. There is reason to believe that it never was contemplated by any of the acts of Congress concerning the public surveys, nor by any practical construction of them either here or in the department at Washington, that a survey, duly and legally made and approved by the surveyor general here, of a private claim had, as a matter of course, to be transmitted to the general land office, and receive the formal approval of the commissioner, and of the department to which he was subordinate, before it could be regarded as an authoritative survey. On the contrary, under the practical construction of the laws previous to 1836, both here and at Washington, the survey, when made and approved by the surveyor general, stood as the authorized governmental survey until it was set aside by some authority having a supervision over the surveyor’s office. As Rector’s survey was never attempted to be disturbed by any one, or at any time, so far as this record shows, until it became merged in Brown’s in 1834, and was recorded in the office as the government survey, it can be nowise important to inquire whether it could have been set aside, if proper steps had been taken for that purpose. We speak of it now as a survey de facto, as practicably approved and acted on. Whether it was legally null for want of authority in the officer to make it is another question, which we proceed to consider.
The act of April 29, 1816, (1 Land Laws, 278) expressly made it the duty of the surveyor “ to cause to be surveyed the lands in the said territories, [Illinois and Missouri,] the claims to which have been or hereafter may be confirmed by any act of Congress, which have not already been surveyed according to law.” Rector’s survey, it appears, was in the office of the surveyor general at St. Louis in 1817. Whether made prior or subsequent to the passage of this act does not appear. In our view, it is immaterial; for as the act of 1816 required the survey to be made, even if no law previous to that did, it was not necessary for the surveyor to cause a new *504survey where one already existed which met his approval. In the absence of proof to the contrary, the presumption is that the officer has done his duty ; and if no survey was authorized before the act of 1816, we would presume that a survey found in the office in 1817, and recognized there as an official act, was made at a time when the law permitted it. Rector’s survey can not then be regarded as a mere private and unauthorized survey. It appears that the surveyor general, under whose administration it was made, approved it in the only mode then customary in his office, and that it was treated as an official survey by those who at that time and subsequently had control of this office. The presumption is that it was made in conformity to directions and instructions from the officer who sanctioned it when it was reported to him.
No particular importance is attached to Rector’s survey, so far as the decision of this case is concerned, except as it constitutes the basis of Brown’s survey in 1834. This last survey was unquestionably authorized by law, and made under instructions issued by the proper authority. Brown was directed to retrace Rector’s lines, and to connect them with the public surveys. It can be of no consequence to inquire whether Brown’s survey, made under these instructions, did in all respects conform to them, for, besides that the surveyor general and all the departments having control over the subject did not complain of the correctness of Brown’s survey in this respect, it is evident that so far as the Cerré claim is concerned, it is immaterial which survey is regarded as the true one, as either will undoubtedly embrace the confirmation.
For five years this survey remained in the office at St. Louis as the approved suxwey of the Carondelet commons, and regarding Brown’s survey as a mere resurvey of Rector’s, it had stood as the approved governmental survey for twenty-two years before any effort whatever, on the part of individual claimants or government officers, was made to disturb it.
*505In 1839, after this survey bad thus been recognized for nearly a quarter of a century by the surveying department, and had been apparently acquiesced in by those supervising tribunals who possessed the power, if they had thought proper to exercise it, to set it aside, it seems that the War Department became alarmed about their title to a military post which had been established at Jefferson Barracks within the limits of this survey, or at all events thought it expedient to appropriate a considerable tract of land, including that post, for the convenience of the government, and entertained apprehensions that the title of the inhabitants of Carondelet might prove an obstacle to this appropriation. At what time, or by what authority, this post was first established there is not shown by any document in the record; nor is it material, for it will be seen in the sequel that the government succeeded not only in securing their title to the enclosures around the post, but to 1702 acres of land which were laid out and surveyed around it. The title to this tract cuts no figure in this case; it is merely alluded to as showing the origin of the inquiry, which, in 1839 or 1840, was instituted into the legality and propriety of the two surveys of Rector and Brown. The attention of the Land Office was in this way directed to the subject, and the district attorney for Missouri, and the Solicitor of the Treasury, and the commissioner of the General Land Office, all concurred, on the first suggestion of the matter, that “ Brown’s survey embraced several thousand acres of land to which Carondelet had no legal or equitable title.” This sentence was a summary one; but it was never executed. From this period, for a succession of years, a discussion was carried on between the department at Washington and the surveyor general here, which it is unnecessary particularly to notice. For a short period the inclination of the Land Department seemed to be adverse to the Carondelet title and to the survey which the government had furnished as evidence of its locality. The government secured her tract of 1702 acres for the Barracks, and a portion of the land south of the Des Peres *506was actually entered or preempted at the land offices of the district although against the express orders of the department at Wasliington.
An examination of this correspondence will show that all the surveyors general, from the date of Rector’s survey until the conclusion of the dispute, including Vm. Rector, Lang-ham, McOree, Dunklin, Spalding, and Milburn, uniformly maintained the integrity of Brown’s survey, and pronounced it official and binding. The recorder of land titles concurred in this view when officially called upon for his action; nor did the heads of departments and bureaus at Washington concur in a different view. On the contrary, with the exception of the officers who first took up the subject in 1839 or 1840, and the Secretary of the Interior in 1853, they also regarded these surveys as authoritative, and declined to disturb them.
It is not deemed important to refer particularly to this official correspondence, except so far as it may be necessary to ascertain what has been the action of the government in relation to those surveys.
In 1841, the solicitor of the treasury (McRoberts) makes a communication on this subject to the Land Office, reviewing and examining the title of Carondelet, and after pronouncing an unfavorable opinion of its merits and recommending 1702 acres to be surveyed and reserved for military purposes, and after designating a specific extent of'the survey which he supposed sufficient to satisfy the claim of Carondelet, which he regarded as limited by the quantity originally claimed before the old board of commissioners, he advises a subdivision of the remaining part of the tract with a view to its sale at auction as public land. In this report and recommendation the commissioner (Whitcomb) concurred. The Barracks tract was accordingly surveyed; portions of the remainder were ordered to be subdivided and sold; and, subsequently, but after this order was countermanded, some of the land was permitted to be entered at the Land Office; but in September, 1845, the then commissioner *507(Shields) came to the conclusion that it was inexpedient to direct the survey of these lands included within the survey of the Carondelet common. In November, 1846, the same officer, in an official communication, says: “ I deem it inexpedient to interfere in any manner with the survey of that common (Carondelet) as originally made by Elias Rector prior to 1817, and retraced by Joseph C. Brown, deputy surveyor, in March, 1834, more especially as the surveyor general, in a certificate to a plat of this common dated April 15 th, 1840, has reported to this office that under an act of the legislature of the state of Missouri, passed in pursuance of the act of Congress approved January 27, 1831, the corporate authorities of the town of Carondelet laid off into small tracts the land within Brown’s survey of the Caronde-let commons, except a portion around Jefferson Barracks, and disposed of the same, or the greater part thereof, to private individuals.” The commissioner concludes his communication by saying: “ I would recommend, in view of all the facts, that the claim of the town of Carondelet to the whole common as surveyed by Joseph C. Brown in 1834 be confirmed, reserving as touch as may be necessary and proper for the United States post at Jefferson Barracks and all valid interferences.”
In 1848 another commissioner (Young) declared, substantially as Gen. Shields had done, that “ it would be inexpedient to interfere in any manner with that survey.”
In 1852, the commissioner (Butterfield) makes a careful examination of the subject, and his conclusion is as follows: “ On a full examination of this matter, I find that the proceedings of solicitor Burchard and commissioner Whitcomb were received in September, 1845, by commissioner Shields, who, by letter of the first of that month to the surveyor general, stated that he had refused an application for the survey of ■ a part of the Carondelet commons directed in 1841; and, in a letter of February 14,1846, to the Hon. J. W. Tibbatts, stated that he deemed it inexpedient to interfere in any njafto; ner with the survey of that common as originally mille;-by>. *508Elias Eector prior to 1817, and retraced by Joseph 0. Brown in 1834; that the lands within Eector and Brown’s surveys are not subject to preemption or ordinary entries, &c., &c. This course was adhered to by acting commissioner Piper and commissioner Young, and has not since been departed from. As the decision of the solicitors and the instructions of commissioner Whitcomb were not executed, I regard them as overruled by the subsequent action of commissioners Shields, Piper and Young, and by the action of the three last named of my predecessors. It has been settled that this office must not disturb or in any manner interfere with the aforesaid survey by Brown, unless Congress shall otherwise order by further legislation; but that said survey is subject to such adverse valid rights as may exist under confirmed private claims, and also to the right of the United States to the military reservation,” &c. “I therefore regard this office, under existing laws, as having no further power or control over the subject.”
In the same year (1852) the commissioner of the General Land Office (Wilson) examines the subject somewhat at large; considers both surveys official and valid, and declares “ that the United States, by these proceedings, have parted with the fee in the land, and any question of conflicting interest in them is, in my opinion, purely a judicial one, and in which the executive can now properly exercise no control.”
In 1853, the Secretary of the Interior disregarding these opinions, directed a new survey; but, before any definite action was had, his successor finally in 1855 settled the question in favor of noninterference. After reciting the history of the official proceedings relative to these surveys, and the acts of Congress and decisions of the supremo court supposed to bear upon the subject, the Secretary says: “ In view, therefore, of all the facts and circumstances in this case, and of the opinions expressed by the highest tribunal known to our laws, I am compelled to the conclusion that, as the surveys of 1816 and 1834 were executed by competent *509authority, were duly approved, and were for a series of years acquiesced in by tlie inhabitants of Carondelet, both the government of the United States and the inhabitants of Caron-delet were estopped and concluded thereby. My decision therefore is, that the survey of 1816 or 1817, as retraced by Joseph C. Brown in 18S4, should be sustained, excluding, of couse, the 1702 acres heretofore set apart and reserved for the use of -the military post at Jefferson Barracks; and that the parties claiming adversely to Carondelet should not be hindered thereby from establishing and settling their rights before the judicial tribunals of the country.”
Thus, it will be perceived that the department at 'Washington in 1855 returned to the point from which they had started in 1839, except so far as the military reservation was concerned, and to secure which had been from the beginning the principal cause of anxiety. Eector and Brown’s surveys were declared to have been executed by competent authority, to have been duly approved, and to have been acquiesced in by both the government and Carondelet for a series of years sufficient to conclude them both; conflicting claims were left to the decision of the courts.
In the discussion of this case it was urged, on behalf of those who claim under the Cerré confirmation, that this decision of secretary McClelland is not an approval of Brown’s survey in its original entirety; that it distinctly and expressly excludes the tract of 1702 acres retained as a military reserve, and that it also, in the same manner and to the samo effect, excludes the private claims ; that this survey is therefore no evidence against those who claim under the confirmations of 1836.
We do not concur in this interpretation of the secretary’s, language and meaning. It is true that great care- is manifested in particularly excluding the military reserve of 170£ acres from the survey. The validity of such an exclusion is. not a matter which is a subject of inquiry here; and it appears from the official correspondence in the- record that, Carondelet had conveyed all her title to, this reservation, if *510she bad any, to the government. But the language employed by the secretary in relation to private claimants is totally different from that used in relation to the military reservation. He sustains the survey, “ excluding, of course, the 1702 acres,” &c.; but he does not exclude any other tract from the survey. He expresses the opinion that “ the parties claiming adversely to Carondelet should not be hindered thereby from establishing and settling their rights before the judicial tribunals.” . Had the secretary entertained a design of excluding any private claim from the survey he would surely have specified it as he did the tract of 1702 acres, or so described it, or the class to which it belonged, as to have made its ascertainment easy. No private claim is referred to by the number of the survey, or the name of the claimant, or the character of the claim. The terms employed in speaking of adverse claims are broad enough to include, not only those of Spanish origin already located by survey, but such as never had been surveyed. They would also embrace pretensions originating under the present government, and might be construed to include mere trespassers. A possession, with a claim of title, makes a party as much an adverse claimant, and sometimes quite as formidable a one, as an inchoate title under the Spanish government. We can find no line of discrimination at which this asserted exclusion is to cease; and according to this construction of the secretary’s decision, the survey could be of little, if any, use to Carondelet, seeing that whenever a controversy exists it can be of no avail, and when there is no controversy she can do very well without it.
It was not the intention of the executive, as we think, to give any such approval, or rather disapproval, of Brown’s survey as has been contended for. The exception or reservation of the rights of private claimants was merely an expression of opinion, similar to those previously expressed by the heads of the land department, that the survey was “ subject to such adverse valid rights as may exist under confirmed private claims,” and that “ any question of conflicting inter*511est in them was purely a judicial one, in-which the executive could properly exercise no control.”
Regarding this decision in 1855 as settling the action of the government in relation to the validity of Brown’s survey, it remains to be considered at what date the survey is to be held as a designation of the land by the government. It will be understood, from what has been already said, that we do not consider the determination or decision of the department in 1855 in the character of an original approval of a survey. It is rather viewed simply as a final conclusion on the part of the executive of the United States, through the proper department, that the survey should not be disturbed; that its previous approval by the surveyor general, under whose directions it was made, and by his successors, and by former commissioners of the Land Office, should stand. Conceding the right of the commissioner of the Land Office, even prior to 1836, to revise the surveys approved by the surveyor general, we do not understand, as before observed, that the acts of Congress on this subject, either in their letter or spirit, or in the practical construction they have received, require a survey to be regarded as incomplete until it receives the final and formal approval of the commissioner and the department of the Interior to which he is attached. Many of the private surveys, particularly those made under the’ act of 1815 concerning Neiv Madrid certificates, reached the Land Office only through the medium of the recorder of' land titles; and the laws have not made it a prerequisite to their validity that they receive the formal sanction of all the bureaus and departments through which they may, if contested, have to pass. Such a construction ,of the _ law would lead to great embarrassment in the land titles depending to some extent upon these surveys. Great inconvenience and practical injustice, we apprehend, might result from holding that a survey, which has stood for years as the authoritative governmental survey in the office of the surveyor geheral, can at the instance of an adverse claimant be taken up, and, although its investigation results in no action unfavorable to its validity, *512may be held to take date only from tbe time of tliis final determination of the department in the last resort. A title in this way may be postponed, until those who have an interest in keeping up a protracted controversy about it have succeeded in procuring a survey favorable to themselves. A judgment appealed from is nevertheless a judgment until it is reversed; and if not reversed at all, all acts done under it and all liens acquired by it stand good from the date of the' judgment. The survey of Brown was an approved survey in 1884, and the appeal from its approval in 1839 having been dismissed it stands as a survey of that date.
We do not consider it important to express any opinion as to the effect of the secretary’s exclusion of the tract of 1702 acres reserved for the 'military post at Jefferson Barracks. The government of the United States and the inhabitants of Oarondelet being agreed about this, it is not a matter for third persons to contest. We are not called upon to say whether this reservation could of itself have any effect or not; nor is there any thing in this record to show the history of this title.
]qi relation to the alleged discrepancy which is said to exist' between Brown and Rector’s surveys, it is sufficient to observe that the government of the United States has acquiesced in Brown’s, which is said to embrace more land than the other ; and the confirmation to Cerré’s representatives is within either.
Our conclusion is, then, that Brown’s survey, being a legal and approved survey in 1884, must prevail over a title emanating from the United States two years afterwards.
In McGill v. Somes & McKee, 15 Mo. 87, this court held that “ when a survey is made and passes through the examination and receives the sanction contemplated by law, it is conclusive on the government. It is in like manner conclusive upon all persons who claim title to the land under titles originating subsequently to the survey.
It is no longer a question that confirmations under the act of July 4th, 1836, do not relate back to the date of the *513original Spanish concession so as to exclude intermediate grants, but that they take effect only from the passage of the act. The-confirmation of the Carondelet commons was made by the act of June 13,1812, and the location by a survey made at least as early as 1834. “ These laws,” (the acts of 1812 and 1831) says Mr. Justice Catron in Le Bois v. Brammell, 4 How. 464, “ and the acts done by the United States in pursuance of them, we suppose,- made and located the commons title as effectually as a patent could have done, and brought it within the exception of the act of 1836.” In Menard v. Massey, 8 How. 293, the entries and patents of Massey in 1826 and 1827 prevailed over a confirmation under the act of 1836.
We are not to be understood as deciding that the survey of 1834 would not relate back to the incipiency of the title in 1812. There is a class of confirmations, such as were passed upon by the Supreme Court of the United States in Cochran v. West, where the grant, was indefinite, and its confirmation was accompanied with an order of survey, in which the title would of course take date only from the date of the survey. The title to commons confirmed by the act of 1812 is not of this class; but if it was so regarded, still, as it was located two years before the. date of Cerré’s confirmation, it must prevail. Nor is it to be understood, because the date of Eector’s survey has not been adopted as the date of the Carondelet survey, that our opinion is that the executive officers of the government can, after a lapse of twenty-two years, take up a survey for revisal. Twenty years’ possession of land, under a claim 'of .right, gives title here; and although no statute of limitations- runs against the United States, and no definite period appears to be fixed by law within which the acts of subordinate officers may be revised, we are not prepared to say that twenty-two years would be a reasonable delay.
We have refrained from any discussion of the title of Carondelet as it existed under the Spanish government, or as it was brought to the attention of the board of commis*514sioners previous to 1812. For my own part I have never been able to perceive any greater difficulties or less merit in the claim of Carondelet than in those of St. Charles and St. Louis. The great obstacle to its early recognition as a valid title seems to have been created by the blunder of the persons who first presented the claim to the old board of commissioners in estimating its area at six thousand arpens. This mistake is not difficult to account for, if we adopt the suggestion made at the bar in the argument of this cause, that the depth of the common field lots was supposed to be the measure of the northern line. These lots being forty arpens in depth, and the starting point for the western line being fixed at their extremity, and the length of that line being one hundred and fifty arpens, the area produced by the northern and western lines, meeting at right angles, would bo exactly the quantity claimed. But the common field lots did not extend to the river; and just below them the river deflected considerably to the east, and the line of one hundred and fifty arpens, as fixed by Soulard, instead of being made parallel to the river, was made to conform to the lines of Alvarez and Reihl, so that the metes and borrnds included greatly more than the quantity claimed. But these questions can not be regarded now of any practical importance. The act of Congress of June 13,1812, undoubtedly confirmed the commons of Carondelet; the government has pointed out by a survey the locality of the confirmation, and Carondelet has received the survey. This was done at least as early as 1834’, and a claimant under the act of 1836 can not dispute this survey.
Judge Richardson concurring,the judgment of the circuit court is affirmed.