The appellees brought this suit for a league of land in Jefferson county, granted to Joseph Butler. *451They claimed under an instrument purporting to be a deed from Joseph Butler to E. D. Johnson, dated September 16, 1840. Upon the paper is what purports to be the proof of the deed for registration by two of the three subscribing witnesses before the chief justice of Shelby county, and where the seal should be impressed is a red stain, which may be the relic of a sealing wafer. The certificate of proof is dated November 5, 1840, and upon it, as sufficient authentication, the paper was admitted to record in the proper county, on March 23, 1837. The instrument is worn, and, in appearance, aged. The plaintiffs, who claimed under it, produced it at the trial, and it was shown that they obtained it from the party named in it as vendee. The name of the grantee and the date appear to have been written in a different ink and penmanship, and, probably, at a differ-, ent time from the other words in the instrument. The naked eye Joes not discover the change in the date testified about by the experts.
In making the proof upon which such paper gets to the jury, the party offering it proceeds ex parte. If, without considering any other evidence than that produced by him, there is enough to raise an issue of fact upon the genuineness of the document, it is proper for the court to allow the paper to go before the jury, and the issue of fact is then determined by them, after hearing all the testimony on both sides. 49 Tex. 594; Id., 118; 1 Greenl. Ev., sec. 21.
The preliminary proof before the judge is merely an earnest of the issue. What shall be sufficient for this purpose cannot probably be embraced in a definition that would suit the peculiar facts of every case. It would be always proper to admit the paper when the proof is sufficient, if none opposing is offered, to sustain a verdict in favor of the genuineness of the instrument. In some cases it might be proper to admit it on less than this. Without stamping upon the testimony any estimate of the weight we may attach to it, we think the paper purporting to be a deed from Butler to Johnson was properly admitted to go before the jury for them to pass on the question of genuineness.
In this connection it may be remarked that it is not a correct practice for the court to charge the jury that such and such proof constitutes prima facie evidence of the execution or genuineness of the paper in question. In the case of Cox v. Cock, 59 Tex. 524, that charge was not considered reversible error under the facts of the case, and in Holmes v. Coryell, 58 Tex. 689, the error was covered by other instructions, and it evidently did not influence the result. In this ease, the charge given at the request of the appellant very likely *452corrected any misapprehension on the part of the jury, which the court’s instructions may have induced.
With respect to an instrument offered as ancient, the jury should be informed of the conditions upon which the law dispenses with the ordinary methods of proving the execution of private writings, but whether the conditions exist or not, and whether the circumstances proved to corroborate the antiquity and genuineness of the paper are sufficient, after the court has heard enough to make the issue, should be determined by the jury as any other fact, without any intimation of the judge’s opinion. It is entirely proper in our practice for the court to instruct the jury that a fact proved on one side, with no conflicting proof on the other, is an established fact. Again, when the statute declares that certain testimony shall be prima facie evidence of a fact, to instruct the jury to accord to the testimony this weight is allowable. But when, though there may be no conflict upon the circumstances in evidence, their sufficiency to establish another fact is an issue before the jury, the court ought not to prejudice their determination by telling them what might be considered as a prima facie case. This is fixing the weight of a portion of the evidence, and is an unwarranted invasion of the jury’s province.
Courts of chancery have built up a great and wise system of rules to govern themselves in determining questions of fact. They are but so many scales for weighing the different kinds and phases of testimony not allowed to be furnished to the jury. A jury knows nothing of these rules; they search for the truth with the free use of all their faculties, and doubtless find it more generally by instinct, intuition and common sense than they would by any process that would filter or exclude the use of these honest and fearless guides. The finest feature, perhaps, of a jury trial is that it ends in a verdict which is not the result of artificial and technical tests and measures ; and this feature our statute has carefully preserved by forbidding an instruction upon the weight of the evidence.
The defendants claimed the land by a regular chain of transfers from David Brown down to them, and offered in evidence a certified copy, from the records of Jefferson county, of the testimonio or second original of an authentic act of sale from Joseph Butler to David Brown. This paper, as a recorded instrument, was impeached by the statutory affidavit, and, to fulfill the requirements of the common law, the defendants proved a thorough search for the protocol in the office of the county clerk of San Augustine county. The registered testimonio purported to evidence an act of sale before A. Hotchkiss, as judge of the first instance, on December 16, 1835. *453Article 14, of the Provisional Government of Texas, required these judges to turn over to their successors in office the records and papers in their official custody, and their successors were required by the act of December 20, 1836, sec. 33, which created the county recorder’s office as it has substantially ever since existed, to turn over to the county clerks (the recorders) of their several counties their papers and records. The protocol, or first original of the authentic act of sale, should properly have been deposited and preserved in the office which the defendants caused carefully to be searched for it. They proved that they were unable to find this first original.
To show that the testimonio had been acted upon, they introduced and proved the execution of the original deed from E. O. LeGrand to B. B. Wallace (from whom the defendants deraigned title by competent proof), dated August 21, 1841, witnessed by W. B. Ochiltree and F. L. Green, and by them proved for record before the clerk of the county court of Jefferson county, on August 29, 1841, and, on the same day, recorded. The execution of this instrument was attacked by affidavit, and it was proven by Col. F. B. Sexton, who stated that the deed was in the handwriting of B. B. Wallace, the signature to it in the hand of E. O. LeGrand, and the witnesses’ names in their respective hands. The defendants also produced the original record book D, of Jefferson county, and proved that the record of this deed was in the handwriting of A. Calder, the then county clerk, and that the certificate and proof of registration indorsed on the original deed was also in his hand.
They also read in evidence a duly certified copy of a deed from David Brown to E. O. LeGrand, dated December 25, 1840, which was acknowledged by Brown for registration before the chief justice of San Augustine county, on January 11, 1841, and recorded April 16, 1841. An affidavit of forgery was also filed against this deed, and the defendants produced the original book D of the Jefferson county records, and proved the registration in the handwriting of the clerk, A. Calder, and proved by the party making the affidavit of forgery that his statement was based on no personal knowledge. 1 This deed appears of record in the book D, at pages 216, 217, and 218. They showed in the same book, on pages 215 and 216, the registration of the testimonio, purporting to be done by the same clerk, A. Calder. The two instruments were probably copied in the record by the same hand, but we do not find any proof of this in the long bill of exceptions. The defendants also proved that those claiming under this testimonió paid the taxes on the land, from 1848 to 1877.
They, the defendants, introduced an official, translated copy of the *454grant to Joseph. Butler, from which it appeared that A. Hotchkiss, acted as agent of the colonization contractor in recommending the extention of title, and David Brown made the original survey of the land for Butler. The proof thus authorized the admission of the certified copy of the testimonio, if that instrument was duly registered. We now advance to that enquiry.
The courts construed the act of 1836, Hartley’s Dig., art 2754, to authorize the registration of the second original of authentic acts, but held that some proof to admit them to record was indispensable. Wood v. Welder, 42 Tex. 396; Hutchins v. Bacon, 46 Tex. 411. The acknowledgment, alone, of the officer authenticating the transaction, his acknowledgment with proof by him of the handwriting of the grantor, and proof of the handwriting of the instrumental witnesses, have all been held sufficient under that act—McKissock v. Colquhoun, 18 Tex. 150; Secrest v. Jones, 21 Tex. 125; Gainer v. Cotton, 49 Tex. 103; Beaty v. Whitaker, 23 Tex. 527—but some proof in every instance was required.
The act of 1839, Hartley’s Dig., art. 2761, authorized the registration or use in -evidence of a copy of the protocol, duly certified by the officer having the custody of it, but it did not authorize the record, of the testimonio. Lambert v. Weir, 27 Tex. 360. It was said by Judge Hemphill, in Titus v. Kimbro, 8 Tex. 210, that this law ignored, the testimonio, and, in violation of the first principles of evidence, exalts above the second original a mere copy. He also said that in all that early legislation, the testimonio, though really possessing extraordinary authenticity, was either not understood or was strangely disfavored.
Our reports show that a large proportion of the conveyancing prior to 1836, in the territory now embraced in the state of Texas, was done in the form of authentic acts,, and, in originating the policy of requiring the exposure of the condition of land titles upon public records, it is somewhat remarkable that no definite provision was made for the only evidence of his right the interested party was allowed to possess. In the violent change of government and officers, it must have been known, or very shortly afterwards discovered, that a great number of the protocols would never reach even the uncertain custody of the county recorder. Considering the territorial extent of the municipal divisions of Coahuila and Texas, and the rapid formation of new counties after the secession of Texas, some of them, Jefferson county among the number, having a recognized existence without a known beginning, it was inevitable that the protocol, if preserved through all the risks to which it was exposed, would be irremovably, and, *455perhaps, undiscoverably lodged in an office, in many instances, a great number of miles from the land it conveys.
It was natural that the claimants of land under such instruments, fully proved and credited by their own proper vigor, when formally executed under the law in force prior to the revolution, would, upon the establishment of registration offices, seek that sanctuary with their only evidences of title. Construing the statute for themselves, for it was not until Titus v. Kimbro, 8 Tex. 210, that there was any distinct intimation from the courts, some would make such proof of the testimonio as was obtainable, the varieties of which are illustrated in reported cases, while a large number deceived by the high authenticity of the testimonio under former laws, would content themselves with the spreading upon the record the paper as given to them. The failure of the courts for a long time to rule definitely that the registration of the testimonio, without proof, was' unauthorized, may account for the absence of distinct legislation. The question as to the validity of such registration was merely suggested in the case of Smith v. Townsend, Dallam, p. 570; and, in Paschal v. Perez, 7 Tex. 348, and Titus v. Kimbro, 8 Tex. 210, the question was not raised or decided, but a view of the statute of 1836 was indicated, probably inconsistent with the validity of the registration of such papers without some proof. The question was incidently discussed but never really decided until, in the case of Lambert v. Weir, 27 Tex. 364, it was said to be considered as settled since the case of- Titus v. Kimbro. It may have been better, by an early and decisive exposure of the situation, to induce legislation for the protection and preservation of papers, in themselves possessing great probative force, and evidencing valuable and meritorious claims. Since Lambert v. Weir, its enunciation of the law has been repeated in a number of cases. Under all the early acts, the registration of the testimonio, without proof, is determined by an impregnable array of authority to be invalid.
It only remains to ascertain, whether such record is within the scope of the healing act of 1860. Section 2 of that act, Pas. Dig., art. 5021, provides, that “any grant, deed, or other instrument of writing for the conveyance of real estate or personal property, or both; or for the settlement thereof in marriage, or separate property, or conveyance of the same in mortgage, or trust to uses, or on conditions, as well as every other deed or instrument required or permitted by law to be registered, and which shall have been heretofore registered or. recorded, shall be held to have been lawfully registered, with the full effect and consequences of existing laws; provided, the same shall have been acknowledged by the grantor or grantors, before any chief *456justice or associate justices, or clerk of the county court, or notary public, in any county of the late republic, or the now state of Texas, or judge of the department of Brazos, or any primary judge or judge of the first instance in 1835 or 1836, or proven before any such officer, by one or more of the subscribing witnesses thereto, and certified by such officer, whether such acknowledgement or proof shall have been made before any such officer of the county where such instrument should have been recorded or not.” [The sense of the act, as applied to this case, is seen in reading only the words we have italicized.]
The registration of instruments upon proper proofs, but taken before officers in the wrong county, had already been cured, except as to county clerks, by the act of February 5, 1841. Pas. Dig., art. 4977.
It had already been held that a deed acknowledged before a primary judge, by the grantor, was lawfully admitted to record, on the ground that a primary judge, under the act of 1836, was ex officio a notary public. Butler v. Dunagan, 19 Tex. 565. It had also been held that a judge of the first instance, acting with two instrumental and two assisting witnesses, had the same power possessed by a notary to authenticate and give effect to acts of sale. McKissock v. Colquhoun, 18 Tex. 151.
Since the passage of the act of 1860, we have not been able to find the report of any case in which a testimonio, certified in 1835 or 1836 by a primary judge, or judge of the first instance, has been held not properly registered without additional proof. In Hawley v. Bullock, 29 Tex. 216, a conveyance or authentic act of sale to David Brown, executed on May 8, 1835, before A. Hotchkiss, as judge of the first. instance, the genuineness of which it seems was not controverted, it was said, “never has been properly registered,” but it does not appear, from the report of the case, how the instrument was authenticated, or that it was recorded at all in the proper county. The land was in Jefferson county, but the county clerk testified that a search in his records, in 1857, failed to discover the record of the deed or act. Ho testimonio, dated in 1835 or 1836, has since the act of 1860 been held not properly recorded, if registered in the proper county and authenticated by the proper judge, as shown by the reports.
The first and the second original of authentic acts were executed at the same time. The grantor was required to sign the protocol, and whilst the testimonio is in form a contemporaneous copy of the body and names to the original, it is believed that in practice it was usual, or not unusual, for the grantor to sign his own name to both originals. This was likely the case in McKissock v. Colquhoun, 18 Tex. 150.
*457Long before the act of 1860, the protocol had a fixed and unchangeable abode in the office of the county clerk, in the county in which it was executed, regardless of the location of the land. The person interested in it had no right to remove it or change its custody. When preserved and finally deposited in the proper office, the act of 1839 had already made adequate provision for the registration of a certified copy of it in the proper county. Gainer v. Cotton, 49 Tex. 114.
The only instrument connected with authentic acts, in the possession or reach of the private individual, which could have upon it the original signature or certificate of the primary judge, or judge of the first instance, was the testimonio.
The original certificate to the testimonio, a copy of which was offered by defendants, was dated one day before the date of the protocol. Being executed at the same time and both probably signed by the grantor, the discrepancy no more needs explanation than if the certificate had been dated one day after instead of one day before the first original. There was no room, it may here be remarked, for any objection to the certified copy on this ground, and it was properly not considered by the district court. The acknowledgment of an instrument for registration, it has been held, need not be appended to it, but may be embraced in the body of it in the form used in the civil law, in the authentication of acts of sale. Brownson v. Scanlon, 59 Tex. 222.
Prior to the act of 1860, a testimonio was permitted by law to be registered. That, a copy of which was offered by defendants, had been registered, and the registration of it was validated by that act, “provided” it was “acknowledged by the grantor * * * before * * * a judge of the first instance in 1835 or 1836, * * * and certified by such officer. ’ ’
The instrument, as recorded in Jefferson county, states: “In the town of San Augustine, on December 16, 1835, before me Augustus Hotchkiss, judge of the first instance, and the witnesses named at the end, appeared the citizen Joseph Butler, whom I know, and to whose act I give faith, and declares that he grants in real and public sale * * * .” Thus he declares, “* * * and signed the same before me and the witnesses, * * * together with those of my assistance, with whom I authenticate, for want of a notary, in comformity with law.” Then follows the name of the grantor, the witnesses and the officer.
Then follows this certificate:.
“I do hereby certify this to be a true and correct copy of the *458original which exists in this office under my charge. San Augustine, December 15th, 1835.
Witness: Witness: A. Hotchkiss.
S. W. Blount, Thos. J. Stansbuby.”
Cut off this last certificate and there is left just such paper as was held to be properly recorded in the case of Brownson v. Scanlan, 59 Tex. 222. The certificate shows that the instrument recorded was the second instead of the first original. It contains every requisite of an acknowledgment of a deed for registration. It lacks but one essential of a proven deed, and that is the signature of the grantor, if, indeed, it was not signed by him. The old laws dispensed with inquiry as to the grantor’s signature, and only required proof of the signature of the officer.
The decisions had fixed the right of the testimonio to go to record upon proof of the signature of the officer. If the statute, which cannot apply to the protocol, for reasons already stated, has any effect at all, it must be held to give authenticity, without proof, to the signature of the judge of the first instance. It was intended to make valid certain registrations made upon proof insufficient under former laws. The signature of the judge of the first instance, formerly required to be proved, is made to prove itself as to acts done in 1835 and 1836. To give that clause of the act relating to primary judges and judges of the first instance any curative effect, or any effect at all, not already accomplished by older laws and established rules of decision, it must be construed to embrace such instruments as that relied upon by the defendants.
The healing act of 1846, it was held, should be liberally construed, Butler v. Dunagan, 19 Tex. 565, and the same principle of interpretation should be applied to that of 1860; particularly with reference to a record that, in some jurisdictions, would be considered proved by its antiquity. Wharton Ev., sec. 113.
The language of the act is not clearly comprehensive of such documents. They are in the scope of the equity of the statute, the language used is meaningless if it does not embrace them, and we think it was. the intention of the legislature, in the clauses quoted, to silence controversy over the registration of testimonios certified in the troubled and eventful years named in the act. We, therefore, conclude, that the court erred in sustaining the objection to the copy that the original was not properly recorded.
The defendants did not prove the loss of the original testimonio, or any search for it, but the introduction of the certified copy was not *459objected to on this ground. It was probably taken for granted that, after the lapse of so long a time, the failure to produce the original sufficiently proved its loss. This was not a correct assumption. If the original exists and can be found, it should be produced; if it cannot be found, that fact should be proved. This might not be required if the defendants were offering a certified copy of the protocol, as in the case of Shepherd v. Harrison, 54 Tex. 96, and for the reasons there stated. These reasons do not apply when the proof offered is secondary to the testimonio itself.
There was no error in the refusal of the court to consider, as part of defendants’ preliminary proof of the authentic act, the agreement of the plaintiff indorsed on the certified copy. The object, originally, of making it was to avoid an application for continuance. If it had been stated in the face of it that it should apply only to the particular trial, it would have been just as effective to defeat the continuance. The defendants were not prejudiced by the failure of the parties to express the limitation in the face of the paper. Notice of their purpose to contest its use was given long enough before the final trial to enable the defendants to supply any proof made requisite by the exclusion of the agreement.
It was further proved by the plaintiffs, that the agreement was made under a misapprehension of the facts. Botts w. Martin, 44 Tex. 91. Such agreements have no particular sanctity arising from the fact that they are made in the course of a judicial proceeding. They are still contracts that may be set aside if entered into under a mistake of fact. The proceeding to set them aside is summary, and is addressed, in a large measure, to the discretion of the trial court. Hancock v. Winans, 20 Tex. 320.
Independent of this agreement the defendant proved enough, notwithstanding the affidavit of forgery, to raise the issue of fact upon the genuineness of the act of sale, and the certified copy of the testimonio from what, if the testimonio was genuine, was a proper record of it, ought to have been allowed to go to the jury. The case made for the introduction of the copy is very similar in many of its features to that of Holmes v. Coryell, 58 Tex. 687, and on that authority the copy should have been allowed. The corroborative proof, introduced by defendant®, has already been stated as fully as we deem it necessary.
The objection that the act of sale does not identify the land was not well taken. The land conveyed was a league of land granted to Joseph Butler by the “state”; a league (and only one) of land was shown to have been granted to Joseph Butler by the state of Coahuila and Texas; the vendor and vendee both resided in that state [the seces*460sion of Texas had probably not then been practically consummated for all purposes], and they were effecting a transaction before an officer of that state. This, we think, identifies the source of the grant, as the state Coahuila and Texas and the name of the grantee is stated. Brown made the survey for Butler, Hotchkiss recommended the extension of the title, and no other grant was made to Butler. In these lights we read, in the words of the instrument, a sufficient definition of the subject mattér of the transaction.
All the objections that were urged to the reading of the certified copy ought to have been overruled, and the issue, as to the genuineness of the testimonio and the reality of the original act of sale should have been submitted to the jury. The failure to prove that A. Hotchkiss was a judge of the first instance in 1885, the failure to account for the original testimonio and to take the testimony of any assisting witness shown to be yet living, go to the jury along with the flock of circumstances bearing upon the issue to be considered and weighed by them.
The degree of diligence the jury would exact of the party relying on such an instrument, in exhausting the probable sources of corroboration, would properly depend upon the age of the paper and the probative effect of the facts proved. His duty is to prove enough to reasonably satisfy an unbiased mind that the instrument once existed and was genuine. When this measure is filled the jury alone can determine.
There are numerous other assignments of error complaining of the charge of the court, and of the refusal to give special charges requested by the defendants. The most of these relate to the introduction, admissibility and proof of ancient documents, and have already been considered. The others, in view of the reversal of the judgment on the points already noticed, do not require consideration. The judgment is reversed and the cause remanded.
Reversed and Remanded.
[Opinion delivered February 12, 1886.]