The opinion of the court was delivered by
Mason, J.:P. L. Lynch owns the northwest quarter of section 36, lying about two miles south and a little west of Scott City. The track of the Santa Fe railroad runs nearly north and south about twenty rods east of his east line, and is on an embankment some two and a half feet high. Lynch brought this action against the director general of railroads, who was then operating the road, charging the destruction of fifteen acres of growing alfalfa by the backing up of water caused by this embankment. He recovered a judgment for $1,500, and the defendant appeals.
1. The defendant asks a reversal on the ground that the evidence did not support a finding of the existence of a watercourse on the plaintiff’s land which extended as far east as the railroad, reliance being placed largely upon the opinion in Gibbs v. Williams, 25 Kan. 214. The railroad east of the plaintiff’s land crosses a basin of low ground; Lion creek and its tributaries, flowing eastward, drain a considerable territory, extending some fifteen miles to the west. A witness testified that this creek enters the basin through sections 35 and 36 and into section 31, which would indicate its extension beyond the railroad. Another testified that a person just seeing the place where there is a bridge on the railroad grade might call it the bed of the stream, but he would call it a depression; that the bank at the railroad grade at the point indicated was well defined in places; that it was fairly steep; that the chain of lagoons that runs through the plaintiff’s place are well defined and easy to see; and that' the track goes right through the lagoon, and there was another further west. Another gave testimony to this effect: Lion creek “has its course through the northeast and northwest quarters of section 36. . . . As it comes towards the basin the banks are not as high. The course is very distinct and well marked, however, but the banks are flatter.” He said:
“There is a branch that comes in about a mile west of Lynch’s. The stream from that point on has an almost direct east course. The banks are not steep, *7but the channel is formed of what we might style as lagoons along. Well, they look like buffalo wallows, only they are very long, about two and one-half feet deep. I am not so very well acquainted with the place where the railroad crosses this creek on 36. I am more familiar with where it crosses the wagon roads, both west and east of the railroad. As before stated, the banks are not steep, and the width of the channel, if I would call it that, is something like 100 feet, I guess. I would judge the height of the banks is two and one-half feet. I am acquainted with the course of this creek east of the wagon road for about three-quarters of a mile. It is just a series of what we would call lagoons — long, low lagoons, which form the channel.”
The plaintiff testified:
“I am acquainted with the stream known as Lion creek. Have known it twenty-seven or twenty-eight years. It heads in above Modoc, comes east through my place. There are tributaries west of Modoc and one called Rocky Jaw comes in from the west, could not say how far down, about one mile west of my place. The water comes through Lion creek frequently, ever since I have known it, sometimes once or twice a year and sometimes it may be three or four or five years before the water would come down. • I recall one instance about twenty years ago. It always flowed in the same course. Up near Modoc the banks are very high, in places fifty feet. About a mile west of my place it seems to spread out. It is just like a depression or low place. It is easy to follow it. I never made any measurement as to the height of the banks; I would say something like eighteen inches or two feet. Where the creek starts into my place on the west side I should say it was sixty or possibly eight [eighty] rods; there is a deep place, two or three feet, and it will gradually rise and go on for forty yards, and there will be another place of the same character. These deep places are just connected by a small depression that is grassed over. At the west side of my place I would say that the depression is 100 feet wide, as nearly as I could estimate it, and where the railroad grade comes it is about the same. I would say the banks at the railroad were about two feet high. Right east from the railroad fill it is the same as on the west side. There is a deep depression like that on the west side of my place, then that rises again and dips on towards Coffin’s place, east of me. It continues in this way for about a mile and a half. Prior to the construction of the railroad the water followed the same course. The railroad was constructed in 1909, as I recall it. Prior to the construction of the railroad the water would stand in this creek bed; we would have big rains and the water would stand there for weeks at a time in the creek bed. I don’t remember of its overflowing the surrounding country outside of the creek bed.”
There was other evidence to the same effect and much of a contrary tendency. Considering the evidence for the plaintiff in its most favorable aspect, we think there was enough to carry the case to the jury on the question of the existence of a watercourse.
It is not necessary that the flow should be constant.
“There must be a permanent source of supply, but it is not essential to a *8watercourse that the flow should be constant and continuous. Surface water may be the source of supply, and the flow from that source is necessarily intermittent and somewhat irregular. It is sufficiently permanent if the accumulated surface water flows through a well-defined channel, made by the water flowing with some regularity during the heavy rains which ordinarily occur in that region. . . . To constitute a watercourse it is not necessary that the supply should be from springs, nor that the water should be discharged through a'channel into another watercourse. The fact that the channel of the stream in question grew less distinct and that it practically passed out of sight before the waters reached Dry creek does not argue that the stream lacks the characteristics of a watercourse.” (Brown v. Schneider, 81 Kan. 486, 488, 106 Pac. 41.)
Nor are continuous abrupt banks, and the absence of vegetation from the bed, essential features of a watercourse. (40 Cyc. 556; Railway Co. v. Scott, 71 Kan. 874, 81 Pac. 1131.) In the case last cited it was said:
“The jury found that the plaintiff’s land was not bottom land; that there were no bluffs or gorges on either side of the ravine; that the banks had been plowed across in places; that in certain places vegetation grew in the bottom of the ravine; that where the ravine passed through the land of one McDonald the banks had been plowed and alfalfa sowed, and that he once raised a crop of millet in the z-avine. We do not thizzk that the defendant was entitled to judgment on these special findings notwithstanding the general verdict. Bluffs and gorges are not necessarily essential to a watercourse, nor do we think that the fact that the banks of this ravine were plowed in places, or that occasionally cz-ops matured in parts of it, is conclusive that it was not a watercourse.” (p. 875.)
An earlier opinion contains this language, referring to a situation somewhat analogous to that here presented:
“Of the existence of a natural watez-coiuse, within the terms of that rule, there can be no doubt. An area of countiy of several hundred aci’es (one to two sections, the witnesses state) is drained. The country is bluffy, and deep ravines have been cut in the hills for the outflow of water falling on this surface. In one of the ravines there is a small perpetual flow from a spring, evidently, although the bulk of the water flowing through these ravines is from rain and snow. The configuration of the countiy is such that they deserve protection as natural watercourses. The same may be said. of the ditches running from the foot of these ravines and uniting just north of the wagon road. As to that extending thence to the river, and which is called by different witnesses a drain, a ditch, a slough, a swale, and a depression, the testimony is far from satisfactory; and yet we are constrained to say that there was enough to uphold the verdict. That by the deepening of the ditches along the railroad track the water flowing down tlnough these ravines, ditches, etc., was partially diverted and thrown upon the plaintiff’s field of rye cannot be doubted. Possibly the flood was so great that the land would have been covered anyway. Cei-tainly the railroad ditches helped, if they did not cause *9the overflow. Counsel contends that, because the company owned the fee of the light of way, and because the ditches dug and deepened by it were so dug and deepened for the necessities of its track, its action was proper, and afforded no ground of complaint by plaintiff of its results, and challenges the contrary ruling of the court in these respects. The court was right. A party cannot on his own land, or because of the necessities of his own business, divert the flow of a natural watercourse without paying any party injured the damages he sustains therefrom. This doctrine is clear, and a railroad company hasl no greater rights than any other landowner.” (U. P. Rly. Co. v. Dyche, 31 Kan. 120, 123, 1 Pac. 243.)
2. Reversal is also asked on the ground that some of the special findings are inconsistent with the evidence, some with the general verdict, and some with each other. Omitting two, which will be referred to later, they read:
“1. Was the damage to plaintiff due to rain falling over a large surface of land, which naturally flowed upon and over a depression at one end of the track? Yes.
“2. Was there through section 36 a distinct channel, with well-defined banks, cut through the turf and into the soil by the flowing of the water? Yes.
“3. If you answer the preceding question ‘yes,’ was such channel continuous and of such character clear across the section up to the railroad embankment? Yes.
“4. Did such channel present on casual glance to eveiy eye unmistakable evidence of the frequent action of running water? Yes.
“5. If you answer questions 2, 3 and 4 in the affirmative, describe fully the bed and banks of the watercourse, giving) width of channel, height and slope of banks, and stating whether any considerable part of the) bed was grassed over.
“Starting at west side of section 36, banks 1% to 2% feet; width 50 to 100 feet for about 40 rods; from there east for about 80 rods grassed over; from there to railroad about 1% feet, more or less.
“6. Was there a permanent source of supply of water, or was the flow only occasional and caused by rains? Occasional, by rain.
“7. Had the railroad embankment caused any damage from the time of its construction in 1909 up to the tima of the floods in 1919? No.
“9. How often did water run through section 36? Only when heavy rain fell on watershed.
“11. Was the flood which caused the damage to plaintiff an unprecedented one? No.”
With respect to the challenged findings we think, upon grounds for the most part already indicated, those numbered 2, 3, 4 and 11 are supported by the evidence; that findings 1, 5, 6 and 9 are consistent with the verdict, and that there is no irreconcilable conflict between findings 7 and 11.
3. Upon the issue of the amount of damages two witnesses testi*10fied that the alfalfa land was worth $200 before the flood and $100 after it. Another said such land in alfalfa was worth $250 an acre, and without it not over $100. Complaint is made because two others were permitted to say that the difference in value between such land with and without alfalfa was $100 an acre. This method of showing depreciation has been condemned (W. & W. Rld. Co. v. Kuhn, 38 Kan. 675, 17 Pac. 322), but as there was some competent evidence to the same effect and the defense offered none at all on this matter we regard the challenged rulings as nonprejudicial.
4. In the course of the argument to the jury the plaintiff’s attorney urged them to be very careful in answering the special questions, “because if they did not answer them properly the effect would be to set aside their general verdict.” Counsel for the defendant interrupted him, asking the court to declare a mistrial, and to discharge the jury, which request was refused, and no admonition was given. Reversal is asked on account of the episode. The need of having the special findings harmonize with the general verdict and with each other is not a matter to which the attention of the jury should be directed. (Brick Co. v. Zimmerman, 61 Kan. 750, 60 Pac. 1064; Railroad Co. v. Burrows, 62 Kan. 89, 61 Pac. 439.) The only order asked by the defendant in that connection, however— the discharge of the jury — was too drastic, and it does not seem probable that prejudice resulted.
The railroad was built in 1909 by the Garden City, Gulf & Northern Railway Company, which in 1911 sold it to the Santa Fe. The director general of railroads took possession of it some years later. An instruction whs given that in order to recover it was necessary for the plaintiff to show "that the director general of railroads had notice or knowledge of the fact that such embankment did not contain an ample passageway for such water as might reasonably have been anticipated to flow down said Lion creek.” The two special findings not already quoted read:
“8. What if any notice did the director general of railroads, or any of his predecessor, have that the railroad embankment was a menace to the property of plaintiff? By notifying Santa Fe agent at Scott City and section foreman.
“10. Did the plaintiff, or any one for him, notify the defendant of the danger to his land by reiason of the railroad embankment, and request him to discontinue or remove or change the embankment that caused the danger? Yes.”
The court, on motion of the defendant, set aside these findings, *11and the contention is now made that inasmuch as a material finding upon which the verdict was presumably based, or at all events may have been based, was set aside, a new trial should have been granted. Against this the plaintiff argues that if the defendant had actual knowledge of the obstruction of the watercourse there was no. occasion to give him notice, that there was evidence tending to show he had such actual knowledge, and that it should be presumed the jury found such to be the case. The difficulty with the plaintiff’s argument is that when the jury were called upon to say what notice the defendant had that the embankment was a menace to the property of the plaintiff they did not say he had actual knowledge of it, but specifically that the Santa Fe agent at Scott City and section foreman had been notified, and that the defendant had been requested to change the embankment. In the absence of a special finding on the subject, the jury would be deemed to have found anything supported by the evidence and necessary to uphold the verdict. But with respect to a subject on which a specific finding is made the presumption must be rather that the jury rested their verdict upon the fact which they did find rather than on one which they might have found. ■
“Where the court sets aside a finding which may have been in part the basis of the general verdict, a new trial must be granted unless the remaining findings in themselves require a judgment.” Goff v. Goff, 98 Kan. 201, (syl. ¶ 3, 158 Pac. 26.)
Here the remaining findings do not touch the matter of the defendant’s notice or knowledge.
The judgment is reversed and a new trial is ordered solely upon the issue to which the findings set aside relate.