of the Commission of Appeals, delivered the opinion for the Court. *105This is a trespass to try title suit. It was brought by W. J. Green and his sister Adelia Cessna, joined pro forma by her husband, to recover what was alleged to be that portion of the A. B. Hardin one-half league of land in Liberty County known as the Mary Conley Bend, a bend in the Trinity river containing 83.9 acres of land. Plaintiffs, who are defendants in error here, complain in their petition of two sets of defendants, the Ross heirs who claim under Champ Ross, immediately, and the Hutcheson and Blanding claimants, all of whom were defendants below. They are plaintiffs in error here. The parties will be designated as in the trial court.
Trial was before the court without a jury. Plaintiffs recovered. The trial court filed findings of fact and conclusions of law in support of the judgment, which was affirmed by the Court of Civil Appeals. 128 S. W. (2d) 477. Writ of error was granted upon application of W. W. Ross and the other defendants, because of the importance of the question involved.
According to the allegations of plaintiffs the land involved, which now lies west of the present channel of the Trinity river, lay, prior to the sinking of the “Mary Conley,” a steamboat by that name, on the east side of the river.
The land is particularly described in plaintiffs’ petition by field notes as being bounded on the north for 73.8 feet by the present channel of the Trinity and on all other sides by the old channel of the river, together with that certain one-half of the old river bed adjacent thereto. The field notes of the tract are called to begin at a point on the west bank of the present channel “said point being in the middle of the old * * * channel, and being located north 73 3/4 west 1333 feet from the southwest corner of the north 1/2 of the A. B. Hardin league.” The succeeding calls are “thence along and in the middle of” the old river bed, delineating a tract in the form of a horseshoe “to a point where the middle of the old river bed intersects the southwest bank of the present channel.” The call thence to the place of beginning is “along the southwest bank of the present channel * * * 738 feet.”
The theory upon which the change of location of the land from one side of the river to the other is predicated, is that in 1872 the river suddenly changed its course and cut a new channel across the mouth of the horseshoe, leaving the land in the neck of the horseshoe within the bounds of the old channel as it was when patented in May, 1831. The general theory upon which the recovery by plaintiffs is predicated is that while the course of the river changed at the mouth of the horseshoe *106at the time'of the avulsion hereafter referred to, there had been no appreciable change in the old channel around the neck since the land was patented.
The trial court found in support of the judgment that when the land was patented, the river, “which formed the west boundary of the A. B. Hardin one-half league, was located as contended for by plaintiffs; and that on account of the sinking of the steamboat, Mary Conley, in the neck of the river bend in 1872 “the stream suddenly left its old bed” and “formed a new channel through which the main flow of the river has continued to the present time.” Specific finding is made that the tract in controversy “still can be identified as the same land that was originally on the east side of the old river channel.”
Some of the trial court’s findings of fact are set out ip the opinion of the Court of Civil Appeals, to which, in the interest of brevity, reference is here made. The remaining findings (omitting the field notes) read:
“A rough sketch of the location of said river, as it was in 1831 from the northwest to the southwest corner of said one-half league is hereto attached and made a part hereof. This sketch also shows how the Trinity River finally cut through the neck of said horseshoe bend known as Mary Conley bend, forming a new channel, and leaving the old channel and the land in controversy on the west side of Trinity River. This change in the course of the river occurred after the sinking of the Mary Conley Steamboat in the upper part of the bend, which, together with overflows, caused the river to cut a new channel through the neck of what is now known as Mary Conley bend. This occurred shortly after 1872, the steamboat having been sunk shortly before 1872.
“4. The west five hundred fourteen (514) acres of the North one-half of said A. B. Hardin 1/2 league, which fronts on the Trinity River, and was originally bounded on the West by the Trinity River, as it flowed from the northwest corner of said A. B. Hardin 1/2 league with the meanders of said river around the old channel now known as Mary Conley Bend, to the southwest comer of said five hundred fourteen (514) acres, said comer being on the division line of the one-half league, was conveyed by A. B. Hardin and all of the heirs of his deceased wife, to their daughter, Jerusha Hardin, and the plaintiffs in this case, W. J. Green and Adelia Cessna, are the only heirs of Jerusha Hardin and her descendents, and are the owners of the title to said property by inheritance from Jerusha Hardin, *107to whom the title to the land on controversy was conveyed in said five hundred fourteen (514) acre tract. The location of said tract is shown on the attached plat.”
The trial court’s conclusion of law are as follows :
“I conclude that the title to the land in controversy was originally vested in A. B. Hardin, and passed by proper conveyance and inheritance to the plaintiffs, W. J. Green and Adelia Cessna. I conclude that the action of the Trinity River in cutting through the head of the Mary Conley Bend and abandoning its old bed, seeking a new bed and the main course of the river changing into the new channel, was an avulsion, and that the West boundary of the A. B. Hardin 1/2 league and the land in controversy is not changed, but still remains in the center of the old abandoned river bed; and that the action of the Trinity River in separating the land in controversy from the rest of its original tract in such a manner that the land in controversy can still be identified, did not change the title to the property or soil so cut off from the rest of said tract, but that title continued to be vested in the owner of the tract before the change in the channel. I further conclude that the plaintiffs are entitled to judgment for all of the land in controversy, their title running to the center of the old abandoned river bed, since such stream is above the ebb and flow of the tide.”
No contention is made but that it is the law that “where a stream, which is a boundary, from any cause suddenly abandons its old and seeks a new bed, such change of channel works no change of boundary and that the boundary remains as it was, in the center of the old channel”; and that when the stream thus “separates a considerable part of one piece of land and joins it to another, but in such manner that it can still be identified, the property of the soil so removed naturally continues vested in its former owner,” just as before the avulsion. New Orleans v. United States, 10 Peters (U. S.) 662, 9 L. ed. 573; State of Arkansas v. State of Tennessee, 246 U. S. 158, 62 L. ed. 638, 38 Sup. Ct. 301; Siddall v. Hudson, 206 S. W. 381.
Both plaintiffs and defendants recognize that the trial court’s judgment, and its affirmance by the Court of Civil Appeals, are based upon the trial court’s ultimate finding that the old river channel around the bend was located in 1831 with respect to the land sued for, as it was at the time of the avulsion.
Defendants contend, however, that the record is devoid of evidence to support the finding stated; that there is no evi*108dence showing what the position of the river was with respect to the land in 1831, when the A. B. Hardin league was granted, or that the land is a part of that survey as alleged by plaintiffs. They point out that plaintiffs, in order to recover, must do so upon the strength of their own title, and urge that inasmuch as they failed to make the necessary proof, the affirmance of a judgment in their favor was erroneous. It is the view of defendants, in connection with their contention as stated, that the land in controversy lay west of the river in 1831, and was then, and has always been, a part of the William Swail league, which was granted on May 12, 1831, the day after the Hardin was granted. They contend also that whatever inference, or presumption in favor of plaintiffs’ contention is created by their testimony, is destroyed by defendants.
Plaintiffs contend they adduced substantial proof of the location of the river in 1831, as well as proof of the position of the land with respect thereto at that time, and that conclusive effect cannot be given to defendants’ evidence; and that for this reason there is no ground for setting aside the judgment of the trial judge and that of the Court of Civil Appeals affirming it.
Plaintiffs’ leading witness on the issue involving the location of the river with respect to the land in 1831 when granted, is Surveyor Compton, who resides in Liberty County, and who, prior to the trial, had spent thirty odd years surveying in that county, was familiar with the Hardin and Swail surveys, and had had occasion to run their respective lines. He testified that he was acquainted with the league lines “on both sides of the Trinity river from their actual location on the ground,” and that the old river bounded the Hardin one-half league on the west; that the river is very crooked throughout the entire county and in the vicinity of the land in controversy; that he had had occasion to survey along the river at certain points in that vicinity and to survey the Jerusha Green 518 acre tract in the Hardin one-half league and had surveyed that tract on the ground as a basis for its division between the heirs of W. B. Green about 1931; that he was acquainted with the land sued for known as the Mary Conley bend and had surveyed it on the ground and that he had prepared a plat “from the field notes I made of the survey of what is known as the Mary Conley bend.” The following is a photostatic copy of the plat referred to:
*109
He testified substantially that the river as noted on the plat, indicates its present course; that from the point marked “N 18 1/2 E. 594 feet,” and following the meander line from there on through the looped lines back to the call “S 62 W,” illustrates, and is the way of defining, the water that stands there; that *110the portion indicated by the looped lines stands in water all the time; that the line through the center of those lines indicates the center line of the lake thus created by the standing water; that the points indicated along the line are calculated from the meanderings of the banks on either side. He states that “of course in some places it is a bluff bank on one side and a sandbar on the other, like where the river makes a bend, and you couldn’t tell if it is the center of the river (the old channel) but those bends are figured to the best of my knowledge and ability as the center of that lake”; that the portion included in the notation above referred to has been filled from deposit of the river after overflows; that it is sandy land with scrub trees growing on it, and that on the east extremity of the course where it is marked “south 62 west 350 feet” there is a similar area; that the bluff bank of the old river can be followed, but that the portion right next to the river is being obliterated every time there comes an overflow and that eventually it will be impossible to tell where the present river left the old channel.
The witness points out in his testimony what he designates as the northwest (on the plat) and southwest (not on the plat) corners respectively of the south 1/2 of the A. B. Hardin league, and points out also a portion of the north line of the league, stating it passes, tracing from the nortwest comer, “where the extreme bend (the bend defining what is now known as the Mary Conley) of the river is and corners back in the bend.” This places the line somewhat north of the bend. He points out also the river front of the league between the points referred to, stating that if the river ran along this old channel, “of course as you come down the river (the old channel) with its meanders you would be bound to include it (the Mary Conley bend) in the A. B. Hardin” within the bounds given in the patent to the Hardin.
Compton testified substantially that the old channel was filled up at its mouth where it leads into the present channel, and that the water cannot get out of the lake in the bend; that he was able to trace the old river bed, which “hasn’t changed at all,” that is, so far as location is concerned; that he had heard, even before he was a surveyor, the history of the bend, and that according to that history the old river ran through the neck of.the horseshoe prior to the avulsion.
The witness pointed out the southwest and northwest corners of the Jerusha Green 514 acre tract, stating in this connection that the land in controversy lay “between the two west corners”, of that tract; that Jeff Green, one of the plaintiffs, *111and William Cessna, went with him when he surveyed the land on the ground about 1931; that he had had occasion to go back on the ground since that time to establish definitely the mouths of the old channel.
The witness used in connection with his testimony a map which he had had in his office as county surveyor of Liberty County since 1912, which had been used by him and others in determining the location of the river. It was a map made or used by the war department in connection with a government survey of the Trinity river through Liberty County in 1899, and there appears on it a notation, “old river survey in 1872 and 1873.” On that map the witness designated the land which he pointed out to be that in controversy.
Plaintiffs supplemented and corroborated the testimony of Surveyor Compton with that of witness whose testimony will presently be summarized. It is the view of plaintiffs that the land sued for was embraced in a tract of 514 acres which passed to Jerusha Green when her father, A. H. Hardin, the original grantee of the Hardin league, joined by all of his surviving children except Jerusha, conveyed the tract to her in 1871; and that the plantiffs, W. J. Green and Mrs. Cessna, inherited the land in controversy from their father, Billy Green, who was the only child born to Jerusha and her husband, Steve Green. The supplemental testimony above referred to relates, among other things, to the uses of, and claims to, the land in the Mary Conley bend over a period of time extending back of the avulsion, and tends, in the manner reflected in the testimony, to identify it as a part of the Jerusha Green tract.
Nev Green (not related to plaintiffs), who was born in 1863, testified in substance that he had seen the Mary Conley as it plied up and down the Trinity “lots and lots of time,” and had seen it run between the Hardin and Swail leagues; that other boats ran on the river at that time, but that the Mary Conley was the largest among them; that it went around the bend where the old river channel now is, which could still be traced; that he had fished from the old hull of the boat after it sank; that he saw it before the river completed cutting through the neck of the horseshoe, leaving an island there; that he knew “Uncle” Stephenson Green who was the husband of Jerusha Green and the father of Billy Green; that he visited with them and knew them before the river cut through, and knew that the land was claimed by Steve Green, and later by Billy; that after the river had cut through he had an agreement with. Billy Green that he was to. take charge of the land *112for him, and did so, and used it as a pasture for cattle and some horses, and built about 250 yards of fence from the old channel to the main river; that he could trace the old river bed where the sand had filled in on each side; that the old river channel, with the standing water in it, was such that it would “turn stock”; that it was several years after the river cut through before he could use the land for pasture because of the difficulty of getting the stock across; that he built a fence about 1884 right after the big overflow in that year, and kept the fence up thirty odd years and rebuilt it in 1910 but did not repair it after that time; that between 1884 and 1910 he kept horses and cattle in there and kept some in there practically all of the time unless the river was too high, during which time he had to take them out; that “Uncle Billy Green claimed the land,” since it was cut off of his tract, by the making of the new channel.
Nev Green testified in substance also that he knew Champ Ross under whom the defendants claim immediately, and that he went with Mr. Ross to look at the land he had purchased in the William Swail league shortly before he purchased it; that he looked after the land for Ross some of the time after he acquired it, and that frequently, while he was in charge, Mr. Ross stayed there with him a week at a time; that he heard him state on occasions that he did not own the land in the Mary Conley bend and that he had nothing to do with it; that it was Green’s. He also testified that Billy Green sold some of the timber in the bend to Flirsch Cunningham; that he (Nev Green) had never claimed to own any of the land, but held it for the purposes stated, under Billy Green; that he told Mr. Ross he held it with the permission of Billy Green and showed Ross the fence that he had built, and that Ross did not at any time claim to him that the land in controversy was located in the Swail league; that he (Nev Green) knew where the Swail league lines were, and that the land was not in the Swail.
The witness “Uncle Henderson” Day, a negro who has lived in the vicinity of the river in that section since the year of his birth (1885) and who worked with cattle at intervals in and around the bend for Mr. Ed Pruitt from the time he was large enough to sit on a horse, testified substantially that while he knew nothing “about the surveys at all,” he did know' where the Mary Conley was sunk and that what is now the Mary Conley bend, “went into” the land involved in this suit; that the river went around it and “left an island there.” That there was a wire pasture fence put up by Nev Green, which was there *11325 or 30 years; that the land was known as Nev Green’s pasture and that the river formed the boundaries of this pasture; that the fence Nev Green built ran from one side of the river to the main river; that Mr. Pruitt would give orders to use the pasture and that Nev Green was with them a great deal of the time they worked there, which was around 35 years ago; that it was the general reputation in the community that the land was claimed by Steve Green and “Uncle Billy” Green, and also Jeff Green, the plaintiff, who was his son; that during the 30 years that he went up to the Mary Conley bend with Mr. Pruitt to work with the cattle Nev Green always had cattle in the bend; that Mr. Pruitt usually had Jim Jackson go up a day or two in advance to look after the fence; that the cattle could not cross the river except at one place “because there was only one place you could go in there and you had to come out that way”; that Nev Green’s fence was up when his (Uncle Henderson’s) wife died and that had been 36 years ago; that the last year he used the pasture was the second year after the war ended; that they used the pasture to put the cattle and horses in it then.
Jim Jackson who was 72 years old at the time he testified stated that he worked for Mr. Pruitt; that he knew the Mary Conley bend and had seen the steamboat Mary Conley; that he saw it in the northeast corner of the bend and that there was a big gully where it sank and at that time “we couldn’t get through”; that he never saw it after the river cut through; that about 1900 he was up there working for Mr. Pruitt; that Mr. Pruitt and the workers camped inside the Mary Conley bend “where the fence went around”; that “we went in there and they called it the Nev Green pasture”; that Mr. Pruitt sent him up there twice a year to see about the condition of the fence; that he saw them on such occasions running and catching calves and branding them in there; that the river kept the cattle in the bend and that was why they put them in there.
Roscoe Davis and Hugh Grimes were also- acquainted with the land in the Mary Conley bend and gave testimony of the same character as that stated.
The above summary of the testimony of plaintiffs witnesses is not substituted for that made by the Court of Civil Appeals, but it is intended that the summary there made shall be considered in connection herewith, and reference is made to the opinion for that purpose. Particular reference is made to the closing part of the summary, which directs attention to the fact that Hugh Grimes, who, about 1933, lived on- the west side *114of the river, and while holding under Mr. Ross on the Swail league, made no effort to take possession of the Mary Conley bend; also to the fact that Mr. Ross in April of that year, in response to a letter from Grimes with reference to timber between the old river and the present channel, wrote that he would not at that time care to enter into any controversy with respect thereto, saying that his information was“ that about 1876 the channel of the river was in the place now called the old river” and that “the channel was changed by cutting across the bend” and that the land between the old river and the new channel” was formerly on the east side of the river, and that he had supposed “that such a change in the river involving a piece of land of that size would not operate to change the title,” and that “the title to the land really belongs to whomever owned it before the change or as acquired by deed under him or inherited from him.” (Italics ours).
Defendants contend that the trial court’s finding that the old channel of the river was located in 1831 where it was in 1872 with respect to the land sued for, has no support in the evidence, unless proof alone of such location in 1872 is a sufficient basis for the inference or presumption of fact that the river was so located in 1831. They contend in this connection that when it is proved that a certain condition existed at a certain time the courts of this State" follow the general rule that the continuance of the proven condition operates only prospectively, citing several cases, including Ralls v. Parish, 151 . W. 1089; Western Union Telegraph Co. v. Hughey, 55 Texas Civ. App. 403, 118 S. W. 1130; and Brenan v. Eubank, 56 S. W. (2d) 513.
We do not sustain this contention. The rule is not so absolute as contended for. The case last cited merely holds that there is no presumption that a bank which is closed by the department of banking for some unknown reason, was insolvent sixteen months prior to the time it was closed. The opinion in the telegraph company case contains the statement that presumptions which arise from proof of a given status “operate prospectively and not retrospectively,” but the status there dealt with is that of an unconscious and dying woman, a status necessarily of a briefly limited duration. The court states that the fact that this condition obtained when her husband arrived was no evidence that she was in the same condition fifteen hours prior to that time when her husband arrived at her bedside.
*115The case first cited is more nearly in point and does not support the contention of defendants, as appears from the following excerpt from the opinion:
“It has been urged by appellants that from the situation of the houses the town of Emma in 1910 cannot be presumed where thus situated in 1891. This is true. The presumption will not prevail retrospectively, but the situation in 1910 was a circumstance which can be looked to for what it is worth. It does not rise to the dignity of a presumption, but it is nevertheless a fact or circumstance which may be entitled to some consideration.”
A case more nearly in point than any of those discussed is State v. Macken et al (wr. ref.), 162 S. W. 1160. The suit involved the title to a small island and a sandbar in the Colorado river in the limits of the City of Austin. The State claimed title for the reason that they were and had been islands situated in' the bed of the river. Macken and others based their claim upon the ground that the land involved was included within the bounds of the Decker grant, alleged to lie along the south bank of the river when granted in 1835. They contended that at that time the stream flowing along the south boundary of the land was not a part of the river, but a mere slough, and that when the Decker grant called to follow the bank of the river it followed the bank of what they claimed was the main river channel, and that therefore the grant included the land within its bounds.
The trial court found that at the earliest time (1839) the land in controversy was known to living witnesses it was separated from the main body of the grant by a slough or depression, which left the main channel just below where Congress Avenue crossés the river, and again connected with the river several hundred yards below. It is obvious that this finding involved an inference extending retrospectively at least four years.
The Court of Civil Appeals was of opinion that the trial court found that in 1835 the land bordered on the channel of the river and was therefore a part of the grant. The eminent Chief Justice (Key) who wrote the opinion, in holding that the finding was supported by the evidence, says:
“It is true that no witness testified as to conditions existing earlier than May, 1839, and it is also true that the oldest map that was produced tending to show those conditions appeared to have been made in 1839. The only witness who testified that *116he saw the land in controversy as far back as 1839 was Mr. John Darlington, who at the time he testified was 92 years of age; and it is not probable that any other witness could have been produced whose knowledge upon the subject extended further back than his. Mr. Darlington’s testimony, together with certain maps introduced by the defendants, and the testimony given by Maj. Walton and Mr. Von Rosenberg as to conditions existing as far back as 1853, and changes that were wrought by the floods of 1867 and 1869, support the findings of fact filed as such by the trial judge; and the facts there found justify the further finding of fact, which was incorporated in the judgment, to the effect that at the time the. Decker grant was made the land in controversy was no part of the river, and therefore was embraced in the Decker grant, which extended to the river. * * * We have already stated that from the facts recited in the court’s findings of fact it was permissible to infer that the same conditions existed four years prior to the time referred to in the findings.”
We are in accord with the view expressed in the foregoing excerpts, as well as with the view of the Court of Civil Appeals in the present case that the decisions of this State recognize the general principle announced in Volume 1, Wigmore, Evidence (2d. ed.), pp. 436-437, sec. 437, that “the interval of time to which any inference will be allowable must depend upon the nature of the thing and the circumstance of the particular case”; and that the following illustration used by the author is an apt exemplification of the principle:
“On landing in New York it can hardly be inferred that the steamer at the next dock has been there for a week; but it may usually be inferred that the dock has been there for some years; while the particular circumstances of appearances and the like will in the last instance affect the length of time to which the inference could' be carried back.”
For the purpose of disproving that the river flowed around the Mary Conley bend in 1831 when the Hardin and Swail surveys were patented, and for the purpose of showing that the land sued for was a part of the Swail survey, rather than the Hardin, defendants introduced, in addition to the oral testimony of their witnesses, much documentary evidence. The documents include the original grant of the Hardin league, and a certified copy of the grant of the Swail, together with what is claimed to be a map of that survey, the certified copy and the map showing to have been prepared by the general land *117office in 1856 and recorded in Liberty County in 1902; two administrator’s deeds, one to James Morgan dated April 7, 1846, together with a plat of the field notes called for in the deed, and one to Reuben D. Palmer, executed about two years later; a certified copy of the general land office official map of Liberty County made in 1862; other early day instruments showing that land claimed to be the land in controversy was referred to and dealt with consistently as a part of the Swail league; a deed from A. B. Hardin and his other children to his daughter, Jerusha, in 1871, purportedly based on a survey and division of the Hardin 1/2 league made by the county surveyor of Liberty County in 1861, showing that the river which constituted the western boundary of the northeast corner of the Hardin survey ran at that time and place practically due north and south, the north and south lines of the tract being called for as of equal length.
Defendants, also for the purpose of disproving plaintiffs’ contentions with respect to the location of the old river channel around the bend back of 1872, direct attention to the testimony of plaintiffs’ witness, Compton, to the effect that the river generally speaking, has changed back and forth from one location to another at different points in its course throughout its known history and that it flows in the vicinity of the land in question through a flat plain of alluvial soil from thirty to forty feet deep, and at certain points, “now and in times past goes just about where it will,” and that the tortious channel, at points along the course of the river had done much shifting through the years in Liberty County. This general testimony cannot, of course, be given conclusive effect as proving a change of course in the river at any particular point, and was, as were the maps and other evidence in the record, given consideration by the trial judge in making his findings of fact.
Defendants urge that the certified copy of the field notes and map made by the surveyor who originally surveyed both the Swail and Hardin Grants in 1831 “strongly suggested, if it did not conclusively establish,” that the land sued for was not a part of the Hardin, but a part of the Swail grant. No field notes of the meanders of the river, however, are given in the field note description of the league. They point to the fact that plaintiffs offered no deed or other document in which the land sued for was ever dealt with expressly as a part of the Hardin survey. They point out also that the administrator’s deed to Morgan in 1846 which contains the field notes of the south one-third of the Swail, do not call for the horseshoe bend, notwithstanding the north line of the tract apparently would *118have crossed both sides of the bend to reach the river; that Compton himself testified that the field notes in the deed indicate that an actual survey of a part of the league was made on the ground. They claim that the west line of the Hardin survey in 1831, as located by Compton, cannot be reconciled with the land office map of 1862, in that on the various maps which Compton used and marked in making the location the river at the northwest corner of Hardin in 1872 appears to be a substantial distance west of the southwest comer of the Hardin, whereas a survey and division of the Hardin 1/2 league in plaintiff’s chain of title calls for the north and south lines to be the same length east and west, thus requiring the river at the northwest corner to be almost due north of where it is at the southwest comer of the north half of the survey, which is inconsistent with plaintiffs’ claim or theory that the horseshoe bend in question opens toward the northeast. In line with the contention of defendants just stated it is claimed that there is a conflict between administrator Abbott’s deed to Morgan in 1846 and his deed to Palmer in 1848.They claim that the only evidence that the land sued for was a part of the Hardin league in 1831, is that with reference to the location of the river with respect to the land in 1872.
Ordinarily, whether an inference has been overcome is a question for the trier of the facts. Of course when the evidence is of such character that only one reasonable deduction can be drawn from it the court may so declare as a matter of law.
Under the record before us, viewing the evidence in the light stated, the trial court’s findings as to the location of the river with respect to the land in controversy, and as to whether it is a part of the Hardin survey as alleged, are based upon factual considerations. We cannot agree that the record reflects no basis in fact for the trial court’s inference that the old river channel was located in 1831 with respect to the land as it was at the time during which the short stretch (738 feet) of the new channel was being cut across the neck of the bend. The inference was warranted.
Defendants contend also that the inference, if indulged, and the prima facie case, if any, in favor of plaintiffs, are destroyed by “compelling evidence to the contrary.” While the documentary evidence offered by defendants is highly persuasive of the correctness of the position taken by them, the evidence as a whole is not such that reasonable minds could not differ as to the effect. In other words we do not accord conclusive effect to the contrary evidence.
*119We are in accord with the holding of the Court of Civil Appeals. Its judgment affirming that of the trial court is affirmed.
Opinion adopted by the Supreme Court April 17, 1940.
Rehearing overruled June 5, 1940.