Larson, C. J.

Plaintiff’s .quieting-title action brought against defendants claimed absolute and unqualified legal ownership of real property in Woodbury County, Iowa, described as: “The SE^ of the NE% of Section 28; The NW% of Section 21; The NE% of the SW% of Section 21; The NE% of the NE% of Section 20; The E% of the SW^ of Section 16; The NW[4 of the SW^ of Section 16; The SW1/^ of the NW1^ of Section 16; and 300.68 acres in Section 17, and 133.32 acres in Section 8, all in Township 86, North of Range 47, and all accretions thereto,” by virtue of certain deeds, and by continuous possession thereof. In view of the fact he held a valid tax deed from Woodbury County, Iowa, to the SE]/4 of the NE14 of Section 28, Township 86, North, Range 47, his right thereto was not contested, and the court quieted title to plaintiff therein. No complaint is made of that judgment and thus it is not an issue herein.

The trial court found against the plaintiff and in favor of the cross-petitioners as to the balance of the land and quieted title in them in accordance with divisions set forth in certain disclaimers filed by the cross-petitioners. These lands, the court held, were shown to- be accretion lands as a result of the wiping out and restoration action of the Missouri River.

Plaintiff contends the trial court erred in that determination. He does not, as we understand it, question the applicable law, but his complaint relates principally to the court’s findings of fact under the evidence as disclosed by the record.

I. This ease, being in equity, is triable de novo in this court. It is our duty, therefore, to find the facts. To aid us we are furnished .numerous exhibits, as well as expert testimony, disclosing the past actions of the Missouri River in this three-mile-wide location. It must be recognized that the trial court has had considerable experience in similar controversies and *1381 understands the problems and testimony relating to them. We are, therefore, justified in giving more than casual attention to its findings, especially where, as here, it is borne out by clear, satisfactory and convincing evidence.

Certain applicable rules of law are undisputed and need little or no citation of authority. The presumption of ownership which follows the legal title can be overcome only by evidence which is clear and convincing. A preponderance of the evidence is not sufficient. Thompson v. Thompson, 240 Iowa 1162, 1170, 1172, 39 N.W.2d 132, and authorities cited therein. Each party seeking to quiet title to those involved tracts must succeed or fail on the strength of his own title and not on the weakness of the other. The record title to these lands since 1855 has been in plaintiff and his predecessors, as shown by plaintiff’s Exhibit A. Thus defendant-counterclaimants’ burden was a heavy one. They must show not only the land plaintiff claims was entirely destroyed, but that the restored land was rightfully theirs.

The record is undisputed that this river and its bed moved over to the east or Iowa side in the late 1920’s or early 1930’s, and then moved back to its former or present location in the middle or the late 1930’s. The vital issue which was decided by the trial court in favor of the counterclaimants was whether or not, in making these changes, the river slowly eroded away all the intervening land, and then by an accretion process rebuilt it as it moved back and forth, or whether the change was sudden or by avulsion, at least to the extent that an island or bar which was included in the tract claimed by plaintiff always remained above the high-water mark and never became a part of the river bed. After a careful consideration of the evidence, we agree with the trial court’s determination.

*1382 At the time of the original government plat in about 1850 the river hugged the west or Nebraska side of this watermelon-shaped area and was in about its present location. (See plat drawn from plaintiff’s Exhibit 1.)

Area shaded claimed by plaintiff. Taken from Exhibit 1.

Dotted lines area claimed by cross-petitioners as accretion land.

U. S. Engineer’s office Omaha, Nebraska 1946-1947.

II. Right or wrong, it is well established that lands of a riparian owner are as subject to being lost by the gradual process of erosion by the river as they are of being added to by the process of accretion. We said in Payne v. Hall, 192 Iowa 780, 783, 185 N.W. 912, 914:

“Where the lands of a riparian owner are removed by the gradual process of erosion by the river, the land being no longer *1383 capable of identification, but having been carried away entirely, and the river occupies the identical space formerly occupied by the lands of the riparian owner, the title to the land so occupied by the bed of the river passes from the owner of the land to the state.”

We reaffirmed this pronouncement in Rupp v. Kirk, 231 Iowa 1387, 1391, 4 N.W.2d 264, a case similar to the one at bar, and in Solomon v. Sioux City, 243 Iowa 634, 638, 51 N.W.2d 472. We also said in the Rupp case the character of the soil, the vegetation and trees, as well as other physical facts, on the tract in question, are proper items to consider in the determination of whether this Missouri River in its westward movement at that locality in 1927 cut away formations in front of it from its bed and built an entire bar from the shore line, or most easterly high bank, by the process of accretion. Also as to this view see Bone v. May, 208 Iowa 1094, 225 N.W. 367; Kitteridge v. Ritter, 172 Iowa 55, 151 N.W. 1097; Arnd v. Harrington, 227 Iowa 43, 287 N.W. 292.

III. The doctrine of accretion is well recognized and well established in Iowa. We have often said, to constitute an accretion there must be a gradual and imperceptible addition of soil to the shore line or any adjoining land above high-water mark by the action of the water to which the land is contiguous. Meeker v. Kautz, 213 Iowa 370, 372, 239 N.W. 27, 28; Payne v. Hall and Solomon v. Sioux City, both supra.

IV. “High-water mark” has a definite meaning in our law. It is co-ordinate with the limit of the bed of the water, and that, only, is to be considered the bed which the water occupies sufficiently long and continuously to wrest it from vegetation, and destroy its value for agricultural purposes. Solomon v. Sioux City and Meeker v. Kautz, both supra.

V. With these pronouncements in mind, the plaintiff’s evidence of record title, and the admission by all parties that the Missouri River is a navigable stream, we examine the testimony and the exhibits to determine for ourselves whether the counterclaimants established by clear, satisfactory and convincing evidence that the land claimed by them was accretion land, whether at sometime between 1855 and 1950 the land was entirely washed away, and whether it had become the bed of the *1384 river for a sufficiently long period of time to wrest it from vegetation and destroy its value for agricultural purposes.

The trial court in its decree said: “There can be no question from the evidence that all of the land in controversy in this action was at one time washed away and was later rebuilt.”

While space will not permit us to relate all the evidence regarding the “island”, the vegetation, or the soil composition between the east and west locations of the river in this vicinity over the years, it is well to note the elevation of this high land or island was about five feet above the abandoned channel of the river in the chute or near its eastern high bank. Both of the well-qualified experts, Boy Towl and L. TI. Hart, had walked over this territory and examined the soil. They seemed to agree that after the passage of the river, the made bar land first grew willows and other small vegetation, and later cottonwood trees, which in turn had to be cleared before the ground could be cultivated. Back in 1936, according to plaintiff’s Exhibit 5, Mr. Towl said a small tract about 500 or 600 feet wide on the high part of the island territory showed evidence of such farming.

On the other hand, Mr. Hart said: “In my opinion from the high bank west of the present channel of the Missouri Biver that is all accretion land”, and referred to the land involved in Sections 9, 15, 16, 20, 21, and 22. None of it, he said, was original land.

Mrs. Chloris L. Grubel, a witness for defendants, testified:

“I have been familiar with this land since 1927, at which time the channel was about where it is now; in 1927 or 1928 I walked to the river, the land west of the 1905 high bank was willows, sand, and vegetation; no islands west of 1905 high bank, only sand bar and small willows; shortly after this the river cut back east towards the house and was cutting about two blocks from the house, everything was gone, all you could see was water; * * * don’t know when the channel went back, but all of a sudden it was about where it is today, and sand bars and willows again.”

George Grubel, her husband, said in 1928 when the river came up along the high bank on the east, he could see “nothing to the west except river and sand bar and no islands,”

*1385 Mr. Towl admitted on cross-examination that since the 1950 United States survey all land between the old high bank and the present channel of the Missouri River had been eroded away by the river, including what he designated as an “island”, and that all land lying west of the old high bank is accretion to some land. To the question, “Then the thing you have indicated in your testimony as an island at sometime subsequent to the survey [1850] was eroded away by the Missouri River, is that correct?” he answered, “I think that is true * * *. There is no accretion to the high bank, that is west of the high bank, anywhere but which has all gone-, and made tack by another movement of the river at some other time.” (Emphasis supplied.)

Mr. Hart also testified that in November of 1932 and in the spring of 1933 the main channel of the Missouri River was down at the position of the chute, and the greater part of the pink area (land claimed by plaintiff as designated on the map, Exhibit 1) would be in the channel of the Missouri River, and the same would be true up until 1934 when the river changed and moved westward.

It also appears from plaintiff’s Exhibit 2 that little or no land was above high water in this area in 1930, but that it was nearly all covered by water. All witnesses seem to agree that if any land appeared during that period, it was at that time only a sand bar upon which all vegetation had been destroyed. This being true, it is quite clear that plaintiff’s chain of title to the land had been erased, and that thereafter the restored or made land was property of another or was subject to a claim of adverse possession. Such evidence and testimony more than justified the trial court’s findings herein. Furthermore, we fail to find substantial evidence in the record to the contrary.

VI. The only question which arises in this action not heretofore specifically passed upon by us is as to the right of the high bank owners to claim as accretion land all lands extending to the high-water mark of the river as it presently exists, when at some time before the river became stabilized on the west side of the area, some land first appeared midway between the east and west high banks and was referred to as a fiat sand bar or island. The river continued to recede and, as its channel moved to the west side, land appeared all the way to the high *1386 bank, with the possible exception of some water in the chutes. As these were blocked off above, this water also disappeared. Thus the so-called island was joined to the main high bank and became one body of land above high water extending to the new river bed. As made land, does it all belong to the owner of the high bank land? To this interesting question Mr. Robert M. Underhill-addressed himself in a well-written article in 42 Iowa Law Review 58-62, especially at page 60 where he says:

“When the Missouri River has retreated from the Iowa shore, in times past, it has created sand bars, leaving a depression immediately below the ‘high bank land’ and following the gradual contour of the bank. Waters would occasionally stand or flow in such depression during high stages of the Missouri River. Such water in the depression below the high bank has been referred to as a ‘shute.’ As time passed, this shute would become- closed at its northern end by natural fill, and thereafter water would back up the shute from the lower end, until finally the whole shute would become filled and be generally dry.”

This, he said, has caused us some trouble as to whether the accretion was to the island or to the main high bank land. He points out that courts do recognize that islands can acquire accretions the same as the mainland. We so held in Meeker v. Kautz, supra.

It is upon this theory that plaintiff bases his principal contention. If cross-appellants could not establish the fact that all the land claimed by plaintiff as a result of the government grant had been washed away, and it had appeared that some part of this land was in an existent island above high water over all these years, or that by a sudden change (avulsion) it had never become the bed of the river, then his claim of accretion land to at least the ehute between the island and the main level or high bank would be justified. He did not claim accretion land to the east beyond the chute.

We are convinced there was no substantial evidence of an avulsion here. There was a period of time disclosed when no land remained in this area where vegetation could exist, and we think the testimony strongly supported the court’s finding that all the land between the high bank and the present river channel had at some time been river bed, covered for a sufficient period *1387 to destroy vegetation, and with it plaintiff’s claim of ownership through his chain of title. This is true without the benefit of the presumption against avulsion set out in Kitteridge v. Ritter, supra, 172 Iowa 55, 151 N.W. 1097.

Whether the created land was all by accretion or part by. reliction is not clear. This distinction seems unimportant here, for once the land became the river bed and was later reformed by a continuing action, whether it was by accretion or by a gradual subsidence of the river at that place with no normal return (reliction) makes no difference. Since the United States Government acted to harness and control this river, its location has been gradually stabilized so that a wandering of its bed hereafter seems unlikely. As a result, all this land has gradually become suitable for agricultural purposes.

There is no reason here to distinguish between that part of a surrendered river bed which first appears as a sand bar, sometimes referred to as a reliction, and that area between the bar and the high bank land that fills by accretion. By the river’s continuing action it all becomes a vegetation-covered area above high-water mark. When stabilized, it is then all one parcel of made land contiguous to the riparian land which existed above the high-water mark. In such circumstances it is proper to treat it as accretion to the high bank land. This does not appear to be the situation found in Holman v. Hodges, 112 Iowa 714, 84 N.W. 950, 58 L. R. A. 673, 84 Am. St. Rep. 367, where an island appearing in the middle of the stream was held to be state-owned land which could not be considered riparian owned years later when the mainland accretions reached it.

VII. Plaintiff’s contention that he had established a right to the claimed area by adverse possession has little or no merit. Evidence of posting “No trespassing” signs in 1937 and 1955 and an unsuccessful effort to maintain a shack on that land in 1937, together with his deed, falls far short of his burden to prove actual, continuous, visible, notorious, exclusive, and hostile possession for a period of ten years under his claim of right or color of title. On the other hand, since 1941 or 1942 the cross-petitioners claim possession of these premises by virtue of quieting-title actions involving plaintiff’s grantors, and by virtue of *1388 quitclaim deed, the payment of taxes thereon, attempts to fence the land, and the renting of part for agricultural pursuits.

In view of our conclusion that plaintiff has failed to carry his burden to show ownership of any of these parcels of land by adverse possession, we need not consider the validity of counter-claimants’ contention as to adverse possession or title under the quieting-title actions. Plaintiff’s title of this land had been lost by erosion, and he did not regain it by adverse possession. Counterclaimants were entitled to it by virtue of their ownership of the land along the east high bank of the Missouri Eiver.

It must be noted that the various counterclaimants have filed herein certain disclaimers so that an apportionment of their respective frontages along the present river course, as well as their lot lines, can be definitely established. The trial court properly recognized these agreements and, in quieting title in the various parties, confirmed these boundaries. There is no issue raised as to this determination. We, therefore, affirm the entire judgment of the trial court. — Affirmed.

All Justices concur.