delivered the opinion of the court.
At the conclusion of plaintiff’s testimony, defendant moved for a nonsuit, and assigns as error the judgment of the court in overruling the same, contending upon this appeal: (1) That the findings of facts made by the trial court were not supported by any evidence; (2) that the findings of facts do not support the decree.
The Sandy River branches a short distance above its month; the western fork thereof forming the Little Sandy River. These, with the Columbia into which they flow, embrace an island of about 447 acres, and several small islands south of the Greenleaf line. For several years there has been a bar, opposite the large island in the Columbia, the area and elevation of which has been increased to a large extent by the alluvion brought down by the Sandy River and augmented by that washed by the waters of the Columbia. This bar is about 1,000 feet across from north to south at the widest part, tapering toward the mouth of the Little Sandy River, and being narrower and of very irregular shape at the outlet of the Sandy. This bar, with some other small strips, constitute the basis of the controversy in this case. High water, which at times covers the bar and many low lands in the vicinity, is chiefly *211caused by the rise of the Columbia River, which is some 1,200 to 1,400 miles in length. The large watersheds of this river and its tributaries account for the rise of the water at different seasons of the year. In the regions drained by the Snake River, which is some 1,100 miles in length, and its tributaries flowing from the south, the spring seasons are early. Therefore the melting of the snow and ice, and the early rains, cause a rise in the Columbia which continues for some time, and, when partially abated, is often augmented later in the season by a similar drainage from the north. These conditions lengthen the period of high water in the Columbia, and in this respect this river differs from any other with which we are familiar. Assuming zero as an arbitrary standard fixed as nearly as possible at low-water mark opposite the point in controversy, the elevation of the land between the Holbrook meander line and Greenleaf line varies, commencing at Holbrook line and extending southerly across the widest part of the tract from 3.9 feet above zero to 9.8 feet as shown by the figures on map in evidence. At this place the extreme annual rise of the Columbia is about 22 feet, and the bar is covered with water from 9 to 15 feet every year. It is agreed that the tide does not affect this portion of the river.
The evidence tends to show that this tract, which is referred to in the testimony as a bar, has, since 1882, been nearly the same elevation, except for slight rises and falls. For years steamboats have endeavored to avoid this barrier. Along the Greenleaf line for the greatest part of the distance, there is a bank clearly outlined 10 or 12 feet high with distinct water marks on it, with vegetation above, and sand and gravel below, except for small deposits caused by back water in a few places. This formation could not possibly grow vegetation. The annual recurring spring rises of the *212Columbia River come over the bar during the months of March or April and remain until the 1st or 15th of August. There is practically no vegetation on the same because of the nature of the soil, and because vegetation could not live under water. The bar is changing continually, and vegetation, if it could be started, would be covered with fresh sand or washed away as the current might affect it.
The various estimates of the time the bar is covered with water are from four weeks to six months, and perhaps eight months of the year. Mr. R. S. Greenleaf, who surveyed the tract in 1907, indicates that along his line of survey the bank on one island is nearly vertical, on another, oval; that of the large island gently sloping, being 50 or 60 feet horizontally, the line running with the vegetation at an elevation of about nine feet. The large island is at an elevation of 25 feet above zero. There is a fall of six feet from the foot of the bank to the water’s edge. A rise of five feet of water would cover the north half of the bar, and four feet more would cover the balance. There is a movement of gravel on the bar every year, and out beyond the four-foot line it is mostly quicksand which is unstable like the water itself and changes with every flood. Other witnesses state that the bar is used for pasture; that it has a willow growth, and is like the land up the river for three or four miles; that there is more vegetation on the island than on the bar; that it is not affected by the waters of the Sandy River, and estimate that other bottom land near is lower. Mr. Philo Plolbrook, county surveyor, states that his line is one foot above water and in some places higher; that there are about 75 acres of land not taken in by his survey, which was made January 18, 1907; that he estimates his line to be six or seven feet above zero. The trial judge visited the locus in quo, and made an investigation in October, 1908, before taking the testi*213mony, and again in May, 1909. Elaborate maps showing the location and elevation of the premises and portions of the Columbia and Sandy rivers were introduced in evidence.
Counsel for plaintiff and defendant are in accord in their statements that the ordinary high-water mark on the Columbia River is the line of demarcation between the land which should be measured and the bed of the river. We need only to state the rules of law, as enunciated by this court and others, as the controversy in this case is largely as to the application of principles already adjudicated.
1. Mr. Farnham, in his valuable work on Waters and Water Rights, digest the definitions of high-water mark given in several cases as follows:
“With reference to rivers which are not affected by the tide, the meaning of the term is somewhat more difficult to determine. But the definition which best meets all requirements of the case and which has in effect been adopted by the weight of authority is that “high-water mark” is the point below which the presence and action of the water are so common and usual and so long continued in all ordinary years as to mark upon the soil a character distinct from that of the banks with respect to vegetation as well as with respect to the soil itself. This definition applies both to navigable and to nonnavigable streams. 2 Farnham’s Waters and Water Rights, § 417.
Mr. Justice Thayer in Johnson v. Knott, 13 Or. 308, 310 (10 Pac. 418, 420), said: “The only matter to be decided in such a case is the location of the line indicating high-water mark, and that involves a discrimination between the upland and the bank proper of the stream. The banks of a river serve, of course, to hold its waters within its bed, and the point to which the ' water usually rises, in an ordinary season of high water, I would regard as high-water mark, and that it constituted the *214true meander line. This line is easily observed by an examination of the banks of a river long after the water subsides, and an intelligent jury, when permitted to view the locality, will have no difficulty in detecting it.” This line of high-water mark as a boundary between a riparian owner and the public should be determined by ascertaining where the presence and action of the water are so common and usual and so long continued in all ordinary years as to mark upon the soil of the bed a character distinct from that of the banks in respect to vegetation, as well as respects the nature of the soil itself. Carpenter v. Board of Com’rs, 56 Minn. 513, 522 (58 N. W. 295, 297). Mr. Justice CURTIS in the case of Howard v. Ingersoll, 13 How. 426, 427 (14 L. Ed. 189), states: “That the banks of a river are those elevations of land which confine the waters when they rise out of the bed; and the bed is that soil so usually covered by water as to be distinguishable from the banks, by the character of the soil, or vegetation, or both, produced by the common presence and action of flowing water. But neither the line of ordinary high-water mark, nor of ordinary low-water mark, nor of a middle stage of water, can be assumed as the line dividing the bed- from the banks. This line is to be found by examining the bed and banks, and ascertaining where the presence and action of water are so common and usual, and so long continued in all ordinary years, as to mark upon the soil of the bed a character distinct from that of the banks, in respect to vegetation, as well as in respect to the nature of the soil itself. Whether this line between the bed and the banks will be found above or below or at a middle stage of water must depend on the character of the stream. The height of a stream, during much the larger part of the year, may be above or below a middle point between this highest and least flow. Something must depend also upon the rapidity *215of the stream and other circumstances. But in all cases the bed of a river is a natural object, and is to be sought for, not merely by the application of any abstract rules, but as other natural objects are sought for and found, by the distinctive appearance they present; the banks being fast land, on which vegetation, appropriate to such land in the particular locality, grows wherever the bank is not too steep to permit such growth, and the bed being soil of a different character and having no vegetation, or only such as exists when commonly submerged in water.”
“A fresh water river, like a tidal river, is composed of the alveus, or bed, and the water; but it has banks instead of shores. The banks are the elevations of land which confine the waters in their natural channel when they rise the highest, and do not overflow the banks.” Gould, Waters § 45.
In the case of Land in New Orleans, Called the Batture, 17 Amer. State Papers, p. 91,- there is deduced from the many authorities noted the conclusion said to be most rigorously exact.
“That all is river, or river’s bed, which is contained between the two banks, and the high-water line on them; and all is bank which embraces the waters in their ordinary full tide.”
This most valuable state paper contains further information to the effect that in the Roman or civil law the channel or hollow containing the river was distinguished as the bed and the bank, the river itself being water; thus alveus, aqua et ripa, the bed, water, and bank. All above high-water mark the Roman law considered as “ripa,” bank, and all below as “alveus,” or bed.
2. This distinction appears iif the English law, with this addition, that the bank or margin of the bed of the river, which lies between the high and low-water marks, is called the “beach” or “shore.” There is a difference-*216between the banks of a river and the shores of the sea. The river bank is not subjacent to the river, as are the shores of the sea, which are daily occupied by its accesses in the tide flow. In our rivers, the beach or shore is the actual, as well as the nominal, bed of the river. When the river is at its full flow, be that by the daily flow of the tide, or by the natural increase of its waters, occasioned by rains, floods, and the like, filling its natural bed to its highest reach of flow, it thus marks its high water, while its lessened range of flow by summer heats shows its low-water mark. See Houck, Rivers, § 7.
3. The bank may be rightly defined as that line or ridge of earth which contains the river, holding the natural direction of its course. But if at any time, either from rains or other cause, it has overflowed for a time that line, it does not by such overflow change its banks, for the reason that the overflow is for a time only, while the natural flow is more or less constant. Houck, Rivers, § '8.
4. In 1 Words & Phrases, p. 99, we find this definition:
“Accretion is the imperceptible accumulation of land by natural causes, and the owner of the property to which the addition is made becomes the owner of such ground, as where land is bounded by a stream of water which changes its course gradually by alluvial formation, the owner of the land still holds the same boundary, including the accumulated soil” — citing Inhabitants of New Orleans v. U. S., 10 Pet. 662, 717 (9 L. Ed. 573).
5. It. is the law, as settled in this State, that, where a stream is intended to be meandered by public surveys, the stream, and not the actual meander line as run on the ground, is the true boundary of the riparian owner. Johnson v. Tomlinson, 41 Or. 198, 200 (68 Pac. 406), citing Minto v. Delaney, 7 Or. 337; Weiss v. Oregon Iron & Steel Co., 13 Or. 496 (11 Pac. 255) ; French Live Stock Co. v. Springer, 35 Or. 312 (58 Pac. 102). See, also, Bowlby v. Shively, 22 Or. 410 (30 Pac. 154), and Taylor *217Sands Fishing Co. v. State Land Board, 56 Or. 157, 161 (108 Pac. 126), wherein Mr. Chief Justice Moore defines high-water mark and quotes with approval from the case of Fowler v. Wood, 73 Kan. 511, 549 (85 Pac. 763, 776: 6 L. R. A. [N. S.] 162: 117 Am. St. Rep. 534). “It is not necessary to give a formation on the bed of a river a specific name in order that proprietary rights may attach to it. In many states lands totally or partially submerged are made the subject of grant by the sovereign in order that they may be reclaimed for useful purposes. Islands that arise from the beds of streams usually first present themselves as bars. * * Before it will support vegetation of any kind a bar may become valuable for fishing, for hunting, as a shooting paik, for the harvest of ice, for pumping sand, and for many other well-recognized objects of human interest and industry. If further deposits of alluvion upon the borders would make it more valuable, no reason is apparent why the law of accretion should not apply.”
6. It is contended by counsel for defendant that the court was controlled solely by the question of vegetation in ascertaining the high-water mark or the line of demarcation between the land sold. But from the testimony as well as from the findings themselves we are unable to agree with counsel as to this claim. From the evidence, a portion of which has been set forth, the trial court evidently took into consideration, and in eifect found: (1) That this bar is under water from two to four months; (2) that it is barren sand and gravel by reason thereof; (3) that it is wholly devoid of vegetation; (4) that it is from two to nine feet above low-water mark; (5) that it is a permanent bar consisting ■for the most part of hard, firm sand and gravel, varying slightly from year to year in its general outline, but the surface thereof changes with the recurring periods of overflow, by the erosion of the surface and the depos*218iting of the sand, by the action of the floods and currents; (6) that along the line of high-water mark a bank rises abruptly in most places. The trial court also found that the dividing line or high-water mark is apparent and discernible, and considered the ordinary high-water mark on the bank, and the ordinary low-water mark, the depth of the water, and the rapidity of the stream. The trial judge had a good opportunity to apply the rule above quoted in Howard v. Ingersoll, and to take into consideration all the circumstances and all the natural objects in seeking for and finding this line by the distinct appearances presented on the bank; and to examine the record which the river itself had made, according to the rule indicated in the case of Houghton v. C., D. & M. R. Co., 47 Iowa, 370, 373. In order to ascertain this dividing line, it was necessary to take into consideration all the circumstances and natural conditions, and we think, from the evidence and the findings of fact made by the trial court that this was done.
7. The action was tried by the court without a jury, and we can only examine the testimony to ascertain whether or not there was any competent evidence tending to support the findings made by the trial judge. Salem Traction Co. v. Anson, 41 Or. 562, 570 (69 Pac. 675) ; Glenn v. Dorsheimer, 51 Fed. 404 (2 C. C. A. 309). The trial court reached the conclusion that the bar in question was below the high-water mark, and that it was not raised, by the deposit of alluvion, to the dignity of being an accretion to the mainland. We think there was competent evidence to support the findings, and that they should not be disturbed.
8. Were we to try the case de novo, we should be compelled to remember that the trial court had a peculiar advantage in inspecting the premises at two different seasons of the year, and, as held in Montgomery v. Shaver, 40 Or. 244 (66 Pac. 923), this would lend a *219peculiar weight to the findings. The question asked in the instructions to the jury in Pane Lumber Co. v. United States (C. C.) 55 Fed. 854, 865, as follows:
“Was it that character of land which by action of the water so permanently remaining upon it, when it reaches its high-water mark, would be deprived of its usefulness as land, and become simply what we all know to be the bed of a river?”
Was answered by the trial court in the case at bar to the effect that the tract was so deprived of its usefulness as land, and was a part of the bed of the river. There was no error in overruling the motion for a nonsuit; and we cannot say that the lower court abused its discretion in denying the application for a new trial.
9. It is suggested by counsel for defendant that, before plaintiff can recover on the warranty in its deed, there must be an eviction from the premises by the true owner, or the title must at least be brought in question by the true owner. This action, however, is based upon a contract set forth above, by which the parties stipulate as to the conditions upon which the defendant will make payment to plaintiff, and the latter asserts that these conditions have been fulfilled. The primary question which was determined by the trial court, according to these stipulations, was the number of acres of land conveyed by defendant to plaintiff. We think the complaint states a cause of action.
It follows from these considerations that the judgment of the lower court should be affirmed, and it is ordered.
Affirmed: Rehearing Denied.