The opinion of the court was delivered by

The plaintiff, Glenn Warner, and the state of Kansas, intervener, claimed that a certain tract of land in Ford county was an island in the Arkansas river, and, as such, that it was state school land. Warner entered upon the disputed property with intent to perfect his right thereto as a school-land settler. The defendants claimed the tract, or most of it, as riparian owners, and thát they held fee titles thereto. The land in dispute lies mainly between the original government meander lines of the Arkansas river in that locality.

The jury made special findings, all of which were approved by the trial court. The most significant of these read:

Questions propounded by plaintiff.

*346“1. Q. Did any portion of the land' claimed by the plaintiff originate as an island? A. No-.

“2. Q. Did all that portion of the land claimed by plaintiff lying south of the north channel of the Arkansas river, as the said river is designated on the plat filed by the plaintiff, and north of what has been termed the old channel of the river, as indicated on the plat, originate as an island? A. No.”

The jury found specifically that the north bank of the river in that locality had remained unchanged since the time of the government survey in 1868, and that the original meander line on the north bank coincided with a later (Fonda) survey.

Other questions, propounded by defendants, were answered:

“5. Q. Was the so-called south or abandoned channel in existence at the time the U. S. government made its survey of the lands adjoining the river? A. No.

“7. Q. Was the so-called south or abandoned channel cut through or made at some flood stage of the river by the water breaking over the south bank and running over and across the lands adjoining the river, after making of the United States government survey? A. Yes.

“8. Q. If you answer the last preceding question in the negative, then state if the U. S. government surveyed and disposed of the lands north of the so-called south or abandoned channel. A. Government sold certain lots north of so-called abandoned channel.

“9. Q. Are lots 5 and 6, section 36; lots 5, 6, 7 and 8 of section 35, and lots 7 and 8 of section 34, township 27, range 22, or any portion or portions thereof, as originally surveyed by the U. S. government, located north of the so-called south or abandoned channel? A. Yes.”

Judgment was entered for defendants, and the plaintiff and intervener appeal.

The issue in this case was whether the land settled upon by plaintiff was or had been an island- lying between the meander lines of the Arkansas river in Ford county for such duration of time as to give it permanency. It does not show on the government plat nor in the United States surveyor’s field notes made in 1868, so it must be assumed that if an island did exist thereabout in 1861 or 1868, it was ignored as a part of the federal domain, and passed to Kansas as an incident to the state’s sovereign proprietorship of the bed of the stream. (Dana v. Hurst, 86 Kan. 947, 122 Pac. 1041; The State, ex rel., v. Akers, 92 Kan. 169, 140 Pac. 637; Wilson v. Zutavern, 98 Kan. 315, 158 Pac. 231; Corbett v. Cohen, 100 Kan. 348, 164 Pac. 264; *347Brenneman v. Fleming, 101 Kan. 393, 166 Pac. 482; Stogsdill v. Minor, 103 Kan. 790, 176 Pac. 643; Wear v. Kansas, 245 U. S. 154.)

It is not contended by the defendants that they own all the land which is claimed by the plaintiff and the state as school land. What they do contend is that a very considerable part of the disputed tract is theirs by virtue of government patents and mesne conveyances deraigned therefrom. There is considerable evidence in the record tending to show that the government surveyors in 1868 had done loose and inaccurate work in that locality. There was testimony that no government monuments, stones, or charred stakes, could be found south of the river for about twelve miles, and scarcely any north of the river for six miles. It is also suggested that the meander line of 1868 on the south side of the river was a mere paper survey and did not correspond with any possible measurements which could have been made. The meander line north of the river was determinable from permanent monuments, certain rocks on the river bank, and the “Santa Fe Trail” which ran nearby. The want or disappearance of the government monuments had necessitated several later surveys, the so-called “Fonda” survey north of the river and the “Mather” survey on the south side. From a calculation based upon measurements between the river and recognized monuments located many miles south of the river, a considerable surplus of land was disclosed. In apportioning this surplus (doubtless in the best of faith and in the exercise of his best judgment), Mather, one of these later surveyors whose work has been generally accepted and acquiesced in, allotted part of this surplus to lands north of the southern meander line of the river. And but for the fact that the government monuments north of the river were so obvious, and the northern meander line of the river so readily ascertainable, the Mather apportionment of surplus land would have inured to lands north of the river. But as that line could not be shifted, this apportionment of Mather’s could be placed nowhere except in the river bed. Because of this surplus of land, a surveyor who testified for plaintiff and intervener, said:

“Q. How do you account for this land that you have described between the north boundary of the lots and the south channel of the river as it *348now exists? A. Well, I should judge that at the time of the survey, the inaccuracy of the government making these surveys, and being such a surplus east of that in coming up the line, that it has been a large island there and the government did not take it into consideration by not knowing it was there. I never could figure out how they could have such' a vast surplus along the Kiowa and Edwards county line'.”

In Foskuhl v. Herzer, 77 Kan. 809, 91 Pac. '56, the “Fonda” and “Mather” surveys are discussed. But it attaches altogether too much significance to the Mather survey to argue ■that because that survey discloses such a large tract of land in the river bed there must have been an island there which the government surveyors ignored in 1868. This great area of land in the stream bed is partly due to Mather’s division and apportionment of the surplus. If Mather had given a larger proportion of the surplus to the twelve tiers of sections to the south, or had disposed of it at other intervals, the surplus in the river bed would have been less, and the conjectured existence of an island in the river in 1861 or 1868 would not have arisen.

But it is true that the evidence shows that there was an island which existed for some length of time in that locality. It was known by early settlers as Titus’ island. Titus lived on it, and cut and sold hay from it. Titus patented it (findings 8 and 9), and the defendants, or one of them, holds the fee by .a chain of conveyances from Titus. Another curious incident which the evidence disclosed, if true, was that Warner’s house, as a school-land settlement, is located further south than was the settlement cabin of Titus thirty-five years ago. Be that as it may, the plaintiff and the intervener have not established the first primary essential to the maintenance of this lawsuit— the existence of an island unsurveyed by the federal government, lying between the meander lines of the river, to which the proprietorship of the state could attach, so that it might be acquired by plaintiff as state school land.

It is not enough to justify a disturbance of the judgment to point out that the evidence conflicts with the jury’s finding No. 1. Two juries have made the finding that no portion of the land claimed by the plaintiff originated as an island. (Warner v. Snook, 102 Kan. 814, 172 Pac. 521.) At the former trial the judge first set aside that finding as contrary to the uncontradicted evidence, but later he changed his ruling, and held *349that the jury could compare the soil on the disputed tract with the soil of the mainland south of the- river, and could properly conclude from their similarity that the disputed tract did not originate as an island.

This lawsuit is largely another fact case. There was no error in admitting the evidence of the Fonda survey, nor of the government field notes. Defendants would have a grievance if these had been excluded. The fact that the government field notes have been discredited in other litigation by the Mather survey, and that private rights, public roads, etc., have been established thereunder, does not disqualify the Fonda survey and the field notes as admissible evidence on the question whether there is or ever was an island thereabout to which the state may lay claim. While conflicting deductions may be made by giving validity to both the Fonda and Mather surveys, the Fonda survey, which coincides with the indisputable monuments of the first government survey, cannot be ignored, The fact that there is a large surplus of land thereabout to squabble over does not prove that this surplus is island school land. Even if it be conceded that the evidence is insufficient to sustain findings 1 and 2, the appellants cannot prevail. It was for them to prove the contrary to the satisfaction of the jury. This, in two trials, they have failed to do; and this litigation should not be protracted any further.

The judgment is affirmed.