delivered the opinion of the court.
This is a controversy over the diversion of the waters of a spring, referred to in the record as the “Mountain Spring,” and situated on the land of A. M. McGinnis, in Burk’s Garden, Tazewell county, Va.
In March, 1904, McGinnis and wife conveyed a portion of this land, containing about fifty acres, to Cleveland T. Heninger and others. Later, Heninger acquired several of the shares of the other tenants.
From a diagram in the record, it appears that the fifty-acre tract adjoins the balance of the McGinnis land on the *72west, and is at a lower level than the spring, which was on the northern slope of a mountain. The dividing line between the two tracts is indicated on the diagram as a due north and south line. North of the spring, possibly half a mile, and near the line on the McGinnis side, is a house which, at successive periods, has been occupied by parties who testify in this case. The “Mountain Spring” is described as a medium-sized, constantly flowing spring. A couple of hundred yards, more or less, below this spring and west of the dividing line that is on the McGinnis side, its waters disappear in a small piece of marshy ground. These waters flow-in a defined gulley, or drain, from the spring to this little marsh, or place of disappearance. Below this marsh there is dry ground and sod, and a slight elevation across the course of the gulley. On the other side of this elevated ground is a hollow, starting on the mountain south of the “Mountain Spring,” and running downwards in a somewhat northeasterly course. This hollow starts and runs on the McGinnis land until a point is reached just above and south of the house mentioned, supra. Here the course of the descending hollow crosses the line between McGinnis and the fifty-acre tract, and, about fifty yards further on, it terminates in a sink-hole in said tract. All water coming from the mountain along this hollow, or channel, supra, disappears in this sink.
In 1911, A. M. McGinnis and wife sold and conveyed to J. B. Meek and M. Castle the “Mountain Spring,” with the right to maintain a pipe line therefrom, save “as to the amount of water that would be carried by a half-inch pipe, which was reserved for the benefit of the land on which the spring was situated.” Thereupon the said. Meek and Castle proceeded to wall in the spring, and lay a one-inch pipe therefrom in a ditch along the course of the boundary line between McGinnis’ land and the fifty-acre tract, crossing the water channel above mentioned near the house on the *73McGinnis side. This ditch was dug and the pipe line laid in May, 1911. At the spring, a half-inch pipe was inserted in said line. This pipe ran to a trough on the McGinnis land, and the overflow of the trough flowed by gravitation along the old channel of the spring to the little marsh heretofore described. There it passed below the surface and disappeared.
It is the contention of the complainant, Heninger (appellant here), that the waters of the “Mountain Spring,” after disappearing in the swamp south of the dry sod and little ridge, su/pra, follow a subterranean channel, and appear on the other side of said ridge in the hollow which heads in the mountain. By natural flow in this channel, these waters would then run past the house previously described, and thence upon the fifty-acre tract, where they would be lost in the sink-hole fifty yards or more beyond said house.
In the year 1919, eight years after the diversion of the waters of the “Mountain Spring” by Meek and Castle, Cleveland T. Heninger, part owner of the fifty-acre tract, filed his bill in chancery against said Meek and Castle, and A. M. McGinnis, in which he set up his interest in said tract by reason of his purchase from McGinnis and others, and alleged that the operations of said parties in walling in the “Mountain Spring” and piping away the flow of same had deprived him of water to which he was entitled. Complainant specifically alleged in said bill that after the establishment of said pipe line, he was “wholly deprived of all water on said fifty-acre tract,” and to his very great detriment was compelled to drive his stock, pasturing on said tract, “off said land, a distance of about a quarter of a mile, to water.” Further, that “this interruption of the natural flow of water to which he was entitled, and the consequent loss to his tract, operated to greatly depreciate the value of his land.” The court was asked to issue an injunction *74against the defendants, restraining them from continuing “to divert the flow of water from said spring away from its natural bed and channel * * * and to remove all of said obstructions to said natural flow, and to permit the whole of the natural flow of said spring to flow into and follow the natural bed and channel in which it would naturally flow, upon the removal of the obstructions, except . so much as might be reasonably necessary for the reasonable use and purposes of A. M. McGinnis, to be used by him on his own land upon which said spring was situated.”
The defendants answered this bill, denying—
First: That the flow of water from the “Mountain Spring” “had a fixed, permanent and well-defined channel and bed through which it flowed to where it crossed the boundary line of the fifty-acre tract.”
Second: That “the water from said spring always flowed in a fixed, permanent and well-defined bed and channel by gravitation through and upon the said fifty acres of land.”
The answer alleged further, by way of defense, that the water which, for the greater part of the year, flowed in the hollow heading in the mountain, and thence upon the fifty-acre tract, came from sources other than the “Mountain Spring;” that these sources of supply were springs in the hollow and seepages from the sides; that there was no evidence that the overflow from the “Mountain Spring,” after it sank at the little marsh on the McGinnis land, ever reappeared at any portion of the course of the hollow below said spring; that the flow from this hollow upon the fifty-acre tract had never been constant, but was dependent upon the rains, and was, therefore, irregular and capricious; that the hollow, or water course, was dry at times, both before and after the pipe line was established from the spring, and that the pipe line was in no wise responsible for any interruptions of flow in this channel at any time after its establishment; that the flow of water down said hollow, *75which extends on to the fifty-acre tract, is not and never has been in any way diminished by reason of the removal of the water which respondents removed from the said spring through said pipe line.
The complainant filed certain exceptions to this answer, which were overruled. Thereafter, both parties took depositions, and the case coming on to be heard upon the pleadings and depositions, the court dismissed the plaintiff’s bill. An appeal from, this decree brings the controversy before this court for review. The questions presented are largely questions of fact, as the exceptions appear to have been properly overruled.
[1,2] Conceding that the waters of the “Mountain Spring” flowed by a definite, marked channel to the point where they sank, and thence by a subterranean channel to the mountain hollow, reappearing there in such fashion that they could be identified, and thence proceeded by natural flow in the hollow to the fifty-acre tract, it would be true that McGinnis would be entitled to such portion only of these waters as would be necessary for the reasonable use and purposes of the tract on which the spring was located, and could not dispose of or interfere with the natural flow of the surplus. But if the waters of said spring disappear wholly on the McGinnis land, and do not appear in the hollow below the ridge, save in the form of seepages in the hollow, resulting from percolations through the soil, or if the waters that run in the hollow cannot be identified with the spring waters that disappear on the McGinnis land, then McGinnis would be entitled to dispose at will of the waters of the “Mountain Spring.”
“The only difference in the application of the law to surface and sub-surface streams is in ascertaining the character of the streams. If it does not appear that the waters which come to the surface are supplied by a definite flowing stream, they will be presumed to be performed by the ordi*76nary percolations of water in the soil. * * * A stream, or water course, consists of bed, banks and water, and to maintain the right to a water course, it must be made to appear that the water necessarily flows in a certain direction, and by regular channel, with banks or sides, and having a substantial existence, but it need not be shown that the water flows continually, as it may be dry at times.” Tampa Water Works v. Cline, 37 Fla. 586, 20 So. 780, 33 L. R. A. 376, 53 Am. St. Rep. 262; Roath v. Driscoll, 20 Conn. 533, 52 Am. Dec. 352.
[3] In Gould on Waters (3d ed.), sec. 280, it is said: “Water percolating through the ground and beneath the surface, either without a definite channel, or in courses which are unknown, and unascertainable, belongs to the realty in which it is found.”
The following from a note to the case of Wheatley v. Baugh, 64 Am. Dec. 727, is cited approvingly in Miller v. Black Rock Springs Imp. Co., 99 Va. 747, 40 S. E. 27, 86 Am. St. Rep. 924: “Water percolating beneath the surface without a definite channel, or in courses which are unknown and unascertainable, is not subject to the settled law governing the rights of riparian owners. * * * Waters which thus appear not to be supplied by a definite flowing stream, are presumed to be the result of the ordinary percolations of water in the soil; such a presumption being necessary to obviate the difficulty of determining whether the water flows in a channel. * * * Where percolating water is found, it belongs to the realty where found.”
[4] It is essential for the lower riparian owner who claims that an upper owner has diverted waters to which the former was entitled, to establish that the waters alleged to have been diverted would, without such diversion, flow in natural course upon his land. In other words, the flow on one man’s land must be identified with that on the land of the neighbor.
*77The evidence in this case as to the source and volume of the water flowing by the mountain hollow upon the fifty-acre tract is, in the highest degree, conflicting.
All of the witnesses agree that the “Mountain Spring” has a continuous flow, which is but very little affected by weather conditions; but, with respect to the other facts in issue, they are in acute conflict. Nor is this a conflict of witnesses with unequal opportunities of observation. In large measure, the witnesses testifying have been so situated in the matters of location and occupation, with respect to the premises in question, that their opportunities of observation with regard to the volume of water flowing in the hollow below the' “Mountain Spring” are practically equal. The witnesses for the complainant, with substantial unanimity, state that from year to year -until the spring was piped, enough water flowed in the hollow to supply the wants of the fifty-acre tract. Several of these witnesses lived in the house mentioned, supra. On the other hand, the witnesses of the defendant, with practically equal opportunities of observation—some having lived in the same house, and others in the immediate neighborhood of the spring—state that formerly, as now, the waters in the hollow ceased to run during several months of the year, the flow being dependent upon the character of the season, and that there is at present, taking the flow from month to month, as much water flowing in the hollow, and thence to the sink-hole in the fifty-acre tract, as flowed before the spring was walled in and the pipe inserted. An illustration of the sharp and direct conflict between the witnesses will be afforded by the following extracts from the record placed in immediate juxtaposition. This testimony relates directly to the volume and flow of the water in the hollow at the house, supra, before it entered upon the fifty-acre tract.
The witness for the complainant whose testimony is cited in this connection is Nannie Rhudy; the witness for the defendant is H. Rosenbaum, now chief of police at Graham, Va.
*78Nannie Rhudy.
Q. Do you know the fifty-acre tract and the “Mountain Spring” involved in this case, and which has been spoken of in the taking of these depositions?
A. Yes.
Q. Did you ever live near that spring and get water from it?
A. Got water for years from it. I was raised there.
Q. How old are you?
A. Fifty-eight.
Q. How close to that spring were you raised?
A. About a half a mile from the head of it.
Q. How long did you use the water from that spring?
A. On until I was married. I was married at twenty-one.
Q. How long did you live by it after you were married?
A. Two years. (This witness lived in the same house subsequently occupied by Rosenbaum.)
Q. Where did you get your water during the two years you lived there after you were married ?
A. I got all my water to wash and do my housework out of that spring (i. e., from the water flowing in the channel by the house. The spring was some distance above the house.) We had *79another spring there on the other side of the house that we got our drinking-water from. This water got warm when it ran so far.
Q. How far had that water to run you got to use from the “Mountain Spring?”
A. About a half a mile, I suppose.
❖ i'fi ❖ *
Q. How did you catch the water from the “Mountain Spring’’ for the purpose of using it? Did you dam the branch or have a trough, or how did you get it?
A. We had a trough.
Q. Before you were married, you lived at or near this spring and used this water?
A. Yes, sir; we moved there the year I was bom, and lived there until I was married and two years thereafter.
Q. You used water out of that spring, then, twenty-three years?
A. Yes.
Q. You say you had a trough a half a mile below the spring. What became of the water after it left your trough?
A. It went into a sink-hole on the Heninger land.
❖ % *
Q. About how much water ran from the point where *80it entered the fifty-acre tract to the sink-hole? How much volume of water was there with reference to there being enough water to water stock on that fifty acres?
A. There was plenty of water when I was there. I never knew of any trouble.
Q. State during the twenty-three years that you lived there, whether the flow from that spring to the sink-hole ever dried up or not?
A. No, sir; I never saw it dry.
Q. During the summertime, did you have any scarcity of water or plenty of water?
A. It would get low if there would come a right dry spell. I have known it to get low, but there was always water. Sometimes if would quit running in the trough in the daytime, but it would come in the night good.
Q. There was always water?
A. Always water.
Cross-Examination.
Q. This trough from which you got the water was about a half 'a mile below the spring?
A. Something near that; don’t know the exact distance.
Q. In speaking of the water and water supply, you *81have reference to the place where you got the water at the trough?
A. Yes, sir.
Re-Direct Examination.
Q. Where did all the water that flowed into that trough and on down into the sinkhole come from?
A. From the spring, or along that branch. It all came down along the same way.
Q. By “that spring,” you mean the “Mountain Spring?”
A. Yes, sir.
*78H. Rosenbaum.
What is your age, residence and occupation?
A. Forty-nine' years old; live at Graham, Va.; chief of police.
Q. Did you ever reside in Burk’s Garden, and if so, how far from the “Mountain Spring” in controversy?
A. I lived on the McGinnis place two years.
Q. When was that?
A. I have forgotten the year—about 1900 or 1901; maybe just a little bit later, maybe a year or two later. * * *
Q. During the time you were there, where did the water from the spring run?
A. The water from the spring ran down possibly 200 or 300 yards, and most of it sank. The water from the spring in controversy never did run .down to amount to anything to where we had our watering-trough. There was another spring below this that furnished the water down to the trough (a smaller spring), where our watering-place was.
Q. Then, you lived in the house below the spring, next to what is now known as the Heninger fifty-acre tract?
A. Yes, sir.
Q. During the time you were there, did the water dry *79up in that branch, or gully, that ran by the house on to what is now called the fifty-acre tract, where it sank?
A. It did. I had to take my stock over to the other branch—the other spring.
ifc * * ❖
Q. When you say you had to go to that spring to water stock, to which spring do you refer?
A. I refer to the spring on the right-hand side, where we used to get water for drinking purposes, and at the spring house. (This refers to a spring near the Mc-Ginnis house that did not empty into the mountain hollow, but flowed a little way and sank on the McGinnis land. It was not a source of supply for the water in the hollow.)
Q. About how much of the time was this branch that ran by the house dry?
A. Well, during the hot months of summer we had no water down there except when we had heavy rains.
Q. Was that true every summer you lived there, or not?
A. 'Yes, sir.
Cross-Examination.
Q. About how much of the year was it dry?
A. Well, I can’t say just *80as to the time, but during the hot months of summer.
Q. You can say about how much of the year, can’t you?
A. Maybe four or five months out of the year, except when we. had heavy rains.
Q. Do you mean that the water ran four or five months of the year, or was dry that long?'
A. No, sir; I mean it was dry. In the winter-time we always had water in that branch. We had a watering-trough right at the housg.
Q. Do you know whether that water came from the “Mountain Spring” any part, or from the other spring you mentioned ?
A. Well, there was never any water ran down from that spring (i. e., the “Mountain Spring”) while I was there; there was another spring up there that furnished the water for the trough.
Q. How do you know that no water ran down there from the “M o u n t a i n Spring?”
A. Well, sir, I lived there and was up there every day, and if there was any water running down from that. place I would have seen it. There was a rise in that hollow (i. e., the hollow from *81the “Mountain Spring,” not the hollow west of this spring), where the water could not run over. It didn’t run over that place at all. It sank up there. Now, the water that came from this other spring could have been the same water; I don’t know; I couldn’t trace it underground ; I don’t know nothing about that.
$ í $
Q. Is it not a fact that there is no constant and never-failing spring up that hollow (the mountain hollow) that could furnish the water to flow in the trough oat the house where you lived, except the “Mountain Spring?”
■ A. Why, there are two springs up there, captain. Either one could come to the trough.
Q. Do you testify that both of these two springs are constant and never-failing springs?
A. I do; yes, sir. While I was there they 'were all right.
Re-Direct Examination.
Q. You have stated that the spring or two springs below the “Mountain Spring” which fed the branch that ran by your house were not dry the years that you were there, but the branch was dry.
*82A. Yes.
Q. Do I understand from this that the water from that spring, or those springs during the dry seasons always sank or disappeared before reaching the house?
A. Yes, sir; we didn’t have any water there to use at all in the summer time, neither summer I was there. When there would come a wet spell, that water would run on Mr. Heninger’s place, but it would dry up maybe in a day or so. Maybe there would be a few drops in the trough; just a little, but not enough to water stock at 'all, and then at times it would be dry—just as dry as it is on this floor.
The evidence relied upon by the complainant to establish his case is the evidence relating to the flow of water in the mountain hollow prior to 1911, and the evidence that this flow was greatly reduced after the pipe line was established. It is contended that the coincidence of the reduction of flow in the hollow in the summer months of the year, with the diversion of the waters of the spring by the pipe line, indicates that the spring waters in their natural flow supplied a measurable proportion of the water in the hollow above the McGinnis house. Apart from the evidence that the flow of water in the hollow was greatly reduced after May, 1911, when the water was piped from the “Mountain Spring” by the vendees of McGinnis, there is no sufficient evidence to identify the water in the hollow with the overflow from this spring. Undoubtedly, there *83was flowing water in this hollow during the greater portion of the year, but there were two springs in the hollow, the larger one on the left-hand bank of the hollow going north.. This would be the west bank of the hollow, or the bank opposite to the marsh where the waters of the spring disappeared. Conceivably, this might be point of appearance of the waters of the “Mountain Spring,” though they would be more naturally expected to appear on the east bank of the hollow. But if the waters of the “Mountain Spring,” pursuing a distinct subterranean channel, appeared at this point after sinking at the marsh, it would seem that the flow of this spring, if not as great, would be measurably as constant and regular as the flow of the parent waters. But such was not the case. The evidence shows that the flow of this spring was uncertain and irregular, and at times was so slight as to run but a short distance in the hollow. The other sources of supply in the hollow were the second spring, of which little note is taken by the witnesses, seep-ages along the sides of the channel, and the flow from the head of the hollow due to rains, or melting snows, and persisting during wet seasons.
The defendants put on the stand various witnesses to establish the origin of the waters in the hollow. One of these witnesses was an elderly man named J. C. Kitts, who lived for two years a few hundred yards from the '“Mountain Spring,” and who testifies in great detail on the above line. The following extracts are from his testimony:
“Q. When you lived there twenty-five or thirty years ago, state how the water from the spring ran down the mountain and what became of it.
“A. Well, it came down the mountain—not very far * * * and sank there. It was a low, marshy place.
“Q. Did all of it sink?
“A. Yes, sir, all of it sank.
“Q. Did that water that sank reappear at any time, so far as you were able to tell?
*84“A. No, sir; it never did, so far as I was able to see.
“Q. Do you know where the water came from that ran down the hollow, and ran on to the fifty-acre tract?
“A. There are some wet-weather springs up there that seep out of the bank; one spring that ran about all of the time.
“Q. On which side of the gully (1 e., hollow) was this spring?
“A. On the right-hand side, as you went up the hollow (i. e., the west side).
* * **
“Q. During the two years that you were there, state whether or not water constantly flowed on to the fifty-acre tract, or whether there were times when there was no water flowing on this tract?
“A. There were times when there was no water there. In dry times of the year, there wasn’t any water that ran down on that tract.
“Q. Was that true of both years you were there?
“A. Yes, sir—both years.
Cross-Examination.
“Q. During the two years that you knew it, how much of the time would you say that the water did not run down there, expressed in weeks, months, or days?
“A. Well, something like, I would say, three months out of the year that it did not run down there to amount to anything at all.
* * * H*
“Q. Would there be any time when no water flowed down on the fifty-acre tract, unless it was an unusually dry time?
“A. Well, there were times when there wasn’t any flowing down there at all.
“Q. Would that be when it was unusually dry?
“A. Well, it would be when the weather was dry; it would not be right after a rain.
*85Ht H« H* ❖
“Q. Will you tell me where the water came from that you have testified did flow on the fifty-acre tract, except in the dry weather you have described?
“A. I suppose it came from the seeps along where the branch ran.
“Q. Isn’t it a fact that there is no constant spring that runs all the time anywhere along the hollow leading to the fifty-acre tract, other than the ‘Mountain Spring?’
“A. None that I know of, except the springs along the branch (i. e., hollow) below the spring there.
“Q. Do you call them constant springs?
“A. No. There is one spring up there that I never saw dry. It may not go dry, but it didn’t afford much water.
“Q. Would it have afforded the water that you say ran on the fifty-acre tract?
“A. No, no; there was no water there except them seeps, only in wet weather.
$ $ $ $
“Q. Well, you have just said, when I asked you if that spring would have furnished the water: No, no; that it would not have furnished the water that ran on the fifty-acre tract. Isn’t that correct?
“A. Well, I don’t suppose it would have furnished the water altogether. It would furnish it after a rain. In the wet part of the year, it would furnish the water all along down there.”
This witness, on re-examination, stated that the bulk of the seepages into the hollow, or branch, were on the side of same next to the “Mountain Spring;” that is to say, on the other side of the little ridge from the marsh at which the spring waters disappear. Various witnesses for the defendants corroborate the above statements of the witness, Kitts. There are no waters appearing in the hollow that are in any way sufficiently identified as the waters from the *86“Mountain Spring.” It is posible that the seepages come from that spring, and it is further possible that the spring referred to by Kitts on the right-hand side of the hollow going up may be the reappearance of the waters of the “Mountain Spring.” This possibility is not denied by the witnesses for. the defendants, though they do not consider that the identity of the waters is established. There is one noteworthy fact in the evidence proper to be mentioned in this connection. The pipe from the spring was laid in a ditch that followed the dividing line between McGinnis and the fifty-acre tract, and crossed the hollow from the mountain near the McGinnis house, referred to, supra. This ditch' was dug in May, 1911. From the testimony of the complainant, it would be expected that at this date water would be flowing in this channel upon the fifty-acre tract, since the disappearance, or reduction of this flow is traced to the diversion of the constantly flowing waters of the “Mountain Spring” by the pipe line. At the time the ditch was being dug, the full flow of the waters of the “Mountain Spring” was emptying into the little marsh where they disappeared. Several, if not all, of the parties- who dug this ditch testify that when they crossed the hollow near the McGinnis house, the same was “dry as a bone.”
[5] The trial court was not satisfied that the complainant had identified the water in the hollow with the flow from the “Mountain Spring,” unless it was in the form of seep-ages, or percolations through the soil, or had established by a preponderance of the evidence that the water diverted by the pipe line, had reduced the flow of water upon the fifty-acre tract. This burden was upon the complainant. After a careful review of the conflicting and highly contradictory evidence in the record, we are not able to draw a different conclusion from the evidence from that derived *87by the learned trial court, or to pronounce that conclusion to be erroneous.
If the waters of the “Mountain Spring” appear in the hollow in the form of seepage, or percolating waters only, the complainant is not entitled to relief. Taking the view most favorable to the complainant of the evidence in the record, it is not considered that he has identified the disappearing waters of the “Mountain Spring,” save to the extent of these seepages, or percolations; nor upon the evidence as a whole are we able to conclude that he has demonstrated that the reduction, if any, of the annual flow of the water upon- his land, is traceable to the piping of this spring by the defendants, Meek and Castle.
Upon well-established principles, the decree of the trial court must be affirmed.
Affirmed.