Mr. Justice Wolverton
delivered the opinion.
The plaintiffs are the owners of certain lands, through which Cottonwood Creek, a perennial stream, flows in one or more of its branches, the water of which is utilized by them for irrigation purposes. Being riparian owners-, they allege that the defendants, who have no riparian rights, are diverting the water from one branch of the stream above them, to their detriment and injury. The defendants assert, however, that they are also riparian owners upon one branch of the same stream, and that they have acquired an appropriation of three hundred inches of water, miners’ measurement, by adverse user, and base their defense to the plaintiffs’ suit on these conditions. Cottonwood Creek flows southeasterly in a single channel until near the line dividing sections 33 and 34, township 39 south, range 19 east, Willamette Meridian, where in the' space of a quarter of a mile it breaks up into four channels, two of which (the most southerly) are small, and for the most part reconduct the water into one of the other two. The two latter, known as the “North Fork” and “South Fork,” are well defined, and have a perpetual flow of water through the lands of the plaintiffs. At certain times of the year, ever since there has been a settlement in the vicinity, — some twenty years or more, — water has escaped from the South Fork through a depression or swale near the east boundary of the northeast quarter of the southeast quarter of section 34, which at some distance below runs in a well-defined channel, known as the “Old Channel,” which passes through a portion of the defendants’ lands, by reason whereof they claim to be riparian owners with respect to Cottonwood Creek. Whatever appropriation the defendants have acquired has been by diversion of the water through a ditch constructed by them intersecting tliis old channel *55perhaps a half mile below where it leaves the South Foi’k of Cottonwood Creek. The feature of paramount importance is touching the alleged waterway connecting the old channel with the South Fork of Cottonwood Creek. The plaintiffs do not question the defendants’ right to have such an amount of water flow down the course as would naturally if maintained in the condition as it existed originally; but they insist that one or more artificial channels have been constructed, which largely increase and facilitate the flow therefrom, for which the defendants are responsible, and hence they should be enjoined from maintaining them.
A brief reference to the testimony will indicate more clearly the contentions of the respective parties, and enable us to discuss and determine understanding^ their relative merits : Creed Pendleton testifies that he has resided on the creek about twenty-three years, and is acquainted with the old channel; that before there was anything done at the head of it the water flowed down sometimes until April, sometimes May, and sometimes June, owing to the freshet; that where it left the creek there was a kind of sag or swale came round from the bend, and started off from it; that there was no regular channel at all from the head of the creek, which continued in that condition from seventy-five to one hundred yards before any banks were perceptible ; that at first it formed into potholes, which extended for a half mile or more, where a regular channel was cut out, extending on down ; that the water did- not run down the swale or the old channel later than until May before there was a change made, unless it was during the time of high water, and then it would run over for a day or two ; that within the last four years there has been a channel cut to the creek, which witness thinks was done by putting in dams and running the water out, thus *56causing it to wasli, and probably expedited by opening it up, as he had observed shovel marks and dirt that had been thrown out on the banks ; that Bernard used to put in dams some twenty to forty yards above the place where the dam had been lately constructed to turn the'water down the swale, and that witness would tear them out; that the water coming down the creek by its natural flow is not more than sufficient for the purpose of irrigating the land of plaintiffs, except when it is very high; and that the diversion made by the defendants has interfered with their use of it. F. M. Barnum testifies that he has lived on the place on which the water is diverted from Cottonwood Creek, from 1890 to 1893, at which time there was an old swale extending from the creek, — just a low place where the water ran through, probably, when it was high, but that it was never high enough when he was there to run out without damming it out. John Fitzgerald, who lived on the place from 1886 to 1890, testifies that while he was there Bernard’s channel carried water in the spring until, probably, June, and sometimes later ; that where the channel left the creek it was a small ditch, probably a foot wide, but not very deep where the water ran out, and it would get a little larger every year ; that he was at the place in the fall before the trial, and at that time saw another channel some fifty yards lowe.r down, which was considerably deeper than the other ; that there is a kind of head gate, or something, in there now, which is considerably deeper than the other would have been ; that even when the water was quite plenty there was not a large volume went down the channel, — only about what the ditch would carry, — but in the time of flood, it would spread all over the vicinity. On redirect he says that the ditch comes out now on a little higher ground than whére the other rah. W. G. Spencer testifies that he *57lias lived in the near vicinity since 1884 ; that water ran down the old channel in the spring of the year, when it was high, but not in its ordinary stage ; that it continued in that manner until about four years since that when he first knew the creek there was a swale or low place, probably not deeper than a foot from the top of the bank, sodded over, — kind of meadow land, — which extended out, perhaps, eighty to one hundred yards, and there were potholes for one hundred yards more, from which there was a channel continuing lower down ; that at this time there is a ditch cut straight across, something like twenty feet below the swale, connecting the creek with this old original channel, which is, on an average, six feet wide, and probably three feet deep; and that the flow of water through the ditch diminishes it perceptibly in the creek.
A. A. Smith says that generally the water ran down the old channel until about the first of June, during the high water season of the year ; that it seemed that Mr. Bernard was always contending for water, to let it flow naturally, and other parties were stopping it from washing out. J. F. Arthur observed the condition of the old channel since 1884 or 1885, at which time there was no channel where it leaves the creek; that it overflowed in very high water, and ran down into a low swale and formed a channel below ; that the water would run down in the early part of the season, which could not have been later than June, or might have lasted into July, some years; that water does not run down into the channel when Cottonwood Creek is in its normal condition ; that there is now a ditch running out from the creek, fifty or sixty yards long, five or six feet wide, and three or four feet deep, which he first noticed about two years ago ; that .he has seen dams in the stream, but not for the last six or seven years. He says that they seem to have been put in to *58turn the water down the slough, and the effect of the ditch, as it now exists, is to take the water from the creek, and flow it off down the slough. M. D. Fisher testified that the water ran down the slough or channel to Bernard’s place generally until June, sometimes depending on the snowfall in the mountains ; that it was very low this year, and, at the stage it was in Cottonwood Creek in May, he thinks it would not have run down the slough. J. B. Fisher said that the channel running down at Bernard’s place would carry water to along about the middle of May or the first of June. M. E. Spriplin was of opinion that when he first knew the place water would flow down the old channel early in the spring ; that while the snow continued to go off there was more than at any other time, but when the freshets were over the channel would commence to go dry, which depended on the season ; that sometimes it might be until the first of June ; that there was a second rise usually, from the melting of the snow from the mountains, which would be about the middle of May or the first of June, at which time the water would flow down the channel somewhat. B. T. Spriplin saw some shovel marks and dirt thrown out at the mouth of the slough where it comes out fifteen„or twenty feet from the creek. Felix Green said in wet seasons he supposed the water would run in the old channel until the first of June; that is, when there was plenty of snow in the mountains ; that at the head of the channel a ditch running from the main channel had the appearance of being dug out, and in February preceding the trial it had washed out larger ; that Bernard had put a dam in the main creek, and water was flowing from it when it would not have done so under normal conditions.
J. E. Bernard, one of the defendants’ witnesses, testifies that, excepting a small portion at the head, there *59was a plain channel running down nearly to the lake, where the water spread out over the meadow. He continues : “I find there is a washout there. The channel is cut down deeper than when I first noticed the situation there. As well as I recollect, the water ran over the sod, and the small channels came down, and there were two ditches. In fact, didn’t either of them run out into the meadow much, only there may have been a small channel, — very small. But it is the same on each side. For some cause or other, those channels have both extended further up than they were. Like, for instance, here is the washout; they extend further up. My idea of this is about this : That perhaps here one swale would come in about there, and here, a little further down, another, and, as the water would wash, it cut out those and bring them together, — both into one. And there is about where we put the head gate in. * * * The first rise we.have in the early irrigating is from the snow in the foothills, and when that mins off it goes down and the water gets normal; and, as the snow begins to melt in the high mountains (l think that commences about the middle of May, extending up until the first of June), we have a second rise. This water comes down and fills up almost all of those channels ; that is, the water runs enough for irrigating. That is .the time we use most of it. * * * I never knew it to fail in the second rise.” T. E. Bernard has known the channel for nineteen years. He says that water flows in it in the spring of the year, in the average season, up to the first of or middle of June ; he has known it to go dry by the middle of May; that the flow down the channel has not changed during the later years, but was about the same as formerly ; that there was a dam put in Cottonwood Creek twelve or thirteen years ago for the purpose of turning the water down the old,channel, which was torn out soon *60after ; that they irrigate the latter part of April and May; that in the early years there was no channel connecting with the creek cut out of the sod. A. P. Koozer says water would flow down the old channel ordinarily as late as about the first of June. J. E. Bernard, recalled, further states that Cox and Clarke filled up the lower place this spring, but not higher than it was when he first went there, while, on the other hand, they have not attempted to obstruct the other; that they (the Bernards) took it out, and plaintiffs filled it up again the second time ; and that, before any channel had washed through, there was just a low place or depression through which the water ran. Testimony was adduced tending to show that the water ran in the old channel in about the same volume of late years as it formerly did. Some time prior to the commencement of the suit several of the interested parties, representing both sides of the controversy, met and discussed the situation, but there is a disagreement touching the understanding arrived at. Subsequently some of the defendants put in a head gate, the plaintiffs refusing to participate in the work. This was undoubtedly so constructed as to admit of a largely increased flow of water from Cottonwood Creek down the old channel, and was subsequently torn out, for which the plaintiffs do not deny responsibility. The plaintiffs were perpetually enjoined from further damming up or interfering with the flow of the water from the creek into the old channel, from which decree they appeal to this court.
The water does not flow in the old channel upon which the defendants’ lands are located later than June, unless in exceptional years ; its source of supply being from the perennial stream upon which the plaintiffs’ lands are situated, somewhat lower down. Mr. Long, in his work on Irrigation (section 9), commenting upon like conditions, says : ‘ ‘ Every proprietor of lands on the banks of a natu*61ral stream has an equal right to have the water of the stream to continue to flow in its natural course as it was wont to run, without diminution in quantity or deterioration in quality, except so far as either of these conditions may result from the reasonable use of the water for irrigation or other lawful purposes by upper proprietors.” It is thus made apparent that the plaintiffs are entitled to have the water of Cottonwood Creek come down to them undiminished in quantity and unimpaired in quality, except so far as may result from reasonable use by upper riparian proprietors ; and, to the end that they may insure it coming down, they assuredly have the right, when they can not be regarded as trespassers, to maintain the banks of the stream in their accustomed and natural condition, so that the water may not be diverted or carried off. by reason of washouts, rents, or crevasses therein. The rule is well established that a riparian proprietor may protect his premises against overflow of the stream : Pierce v. Kinney, 59 Barb. 56. And it is reasonable that he may also protect the banks, and thereby secure the use of the water which would naturally flow within them. Now, we find that the water has been accustomed to flow out of the South Fork of Cottonwood Creek through the depression or swale described by the witnesses, and thence into the old channel, for many years. It is unnecessary to determine whether this way constitutes a watercourse, or not, in legal parlance, so as to entitle the defendants to the rights and privileges of riparian proprietors. They have used the water which has been accustomed to flow down the old channel, and we do not understand the plaintiffs to dispute their right to use it as it was wont to flow within the banks of the creek in their natural and normal condition. As riparian proprietors, they could claim nothing more. So that the entire question turns upon the amount of water, or stat*62iug it more precisely, the time or season in which it was accustomed to flow from the creek through the depression and into the old channel.
The consensus of the testimony seems to be that until within late years the water flowed out through the course in the early spring after the breaking up of winter, and thence on, perhaps, until the snow melted in the mountains. The time when it ceased to flow varies, as detailed by the witnesses, from the middle of May to the first of July. There has been a dispute or contention between the parties interested, all along, respecting the outflow. The Bernards have on several occasions constructed dams in the creek for the purpose of diverting its waters, and these have been taken out from time to time by the plaintiffs. Within the last two or three years there have been one or two channels cut through the swale to the creek, and whether these have been produced by natural overflow or through artificial means is not entirely clear ; nor does it matter, as the plaintiffs have the right to maintain the banks in their accustomed condition. The witnesses all concur that one of the channels which now connects with the creek is very much larger and deeper than it used to be ; that the head gate as constructed by the defendants puts the conduit down on the level with the bed of the creek, and is of sufficient capacity to absorb a large percentage of the natxxral sti’eam. Trxxe, the plaintiffs and defendants, or individuals representing each side of the controversy, entered into some sox-t of an agreement whereby the head gate should be put in, but this was not carried into effect; and it is claimed by the plaintiffs that the appliance was not constructed or placed in accordance with the understanding, so that, if any such agreement was entered into, it can have little bearing xxpon the question at issue. The plaintiffs have tom oxxt the head gate, bxxt the defendants insist on maintaining *63the channel in the condition in which it has been placed in later years. This they have no right to do, and are only entitled to have the water flow down to them as it was wont to flow down the swale or depression as it existed in its normal state, before the channels were cut through the bank. The plaintiffs are therefore entitled to an injunction restraining the defendants from interfering with the flow of the water in Cottonwood Creek, and from diverting it therefrom by the maintenance of the channel lately opened from the swale or depression into the stream or otherwise, except as it has been accustomed to flow through the natural waterway as it existed in former years. The decree of the court below will be reversed, and one here entered in accordance with this opinion. Reversed .