delivered the opinion of the court.
The first proposition advanced by counsel for defendants is, that the testimony does not establish an appropriation by plaintiff of the waters of Big and Little Spring creeks which antedates the diversion of the waters of Medaño creek by them. If the rights of plaintiff in this respect depended solely upon testimony to establish an actual application of the waters of Big and Little Spring creeks for the purposes of irrigation, this proposition might have some force. Such, however, is not the case. In each of his complaints plaintiff claimed an appropriation of the waters of these streams through Los Ojos, Hull, South Number One and South Number Two ditches. Water district No. 25 embraces the drainage of Medaño and Big and Little Springs creeks. It appears from a decree of the district court of Costilla county entered at the March term, 1895, and introduced in evidence, that under a statutory adjudication of water rights in that district the ditches in question were awarded different priorities, of specific volumes of water, from Big and Little Spring creeks, dating from March 10, 1875, to May 1, 1881. The ditches of defendants were constructed in 1892 and 1894. The water decree was prima facie evidence that appropriations of water for purposes of irrigation had been made from those streams of the dates and volumes awarded the ditches through which *321plaintiff asserts his rights, antedating the rights initiated by defendants.
It is next urged by counsel for defendants that the testimony does not establish that the waters of the Medaño contribute substantially to those of Big and Little Spring creeks, and that if they do, it is only by percolation. At the request of counsel for both sides, the trial judge viewed the premises. Whether or not the knowledge thus obtained should be given the effect of substantive testimony, we do not determine. We must consider, however, that he was thereby better enabled to understand and apply the evidence of the respective parties on the subject under consideration. Our province in reviewing the testimony is to ascertain whether or not the -findings of facts are supported by the evidence. If they are, we cannot interfere by substituting our judgment for that of. the trial court upon the weight of the evidence in the'-case. Colorado Fuel & Iron Co. v. Pryor, 25 Colo. 540. The testimony is very voluminous, and to notice it in detail would be impracticable. Its review1 will, therefore, be limited to a mention of those matters which tend to support the finding of the trial judge, that Medaño creek is the source of supply of Big and Little Spring creeks, for the purpose of ascertaining if there is sufficient legál evidence.to support the finding on this subject.
Medaño creek rises near the summit on the westerly side of the Sangre de. Cristo range, arid flows southwesterly until, .near the mouth of- the canyon from which it issues, it is intercepted by a range of hills-known as the Great Sand Dunes, from which point its course is along the base of those hills on the east side, nearly south until it reaches the south, end where it flows almost west. In this course it unites *322with the Mosca creek, a small stream rising in the same range of mountains- When the water is low it disappears in the sand of the channel at the base of these hills, sometimes at one point and sometimes at another, regulated in a great degree by the volume of water flowing.. The sand hills begin at the base of the main range, and at or near this point are several hundred feet in height, several miles in width and gradually diminish in altitude and width as they extend south, until merged in the San Luis valley. Big and Little Spring creeks rise on the west side of these hills, in a comparatively sandy level immediately adjoining their base. From the point where Medaño creek is deflected in its general course, to -the source of Big Spring creek, is about seven miles southwesterly, and from the point where the waters of the Medaño ordinarily disappear, when high, the source of Little Spring creek is distant about three miles, almost west. The sources of Big and Little Spring creeks are relatively almost north and south of each other, and distant about three miles.
Witnesses state that many years ago a well defined channel of the Medaño was plainly visible to the source of Little Spring creek, and that they had seen water running in this channel very close to the head of the latter creek. One witness in particular states that in 1863 or 1864 he had seen the water of Medaño running to Little Spring creek. Other witnesses state that this occurred again in 1895. Witnesses also state that a well defined depression, like an old channel of a water-bed, extends from near the source of Big Spring creek in a northwesterly direction, for a distance of about two miles, which, if continued in its course, would strike, after passing through, the sand hills, about the mouth of Medaño canyon; that when*323ever the water of Medaño creek diminishes, the waters of Big and Little Spring creeks also decline after a lapse of two or three weeks from the time when the waters of the Medaño become low; that when the Medaño is muddy, the waters of Big and Little Spring creeks would also.be riley, or show discoloration. Witnesses also state that since the diversion by defendants from the head waters of Medaño creek, there has been a marked decrease in the flow of Big and Little Spring creeks. The general drainage of the country, with the exception of the Medaño, is towards the west. According to surveys, the most rapid line of decent from near the mouth of Medaño canyon is in the direction of Big Spring creek.
The theory of plaintiff is that the San Luis valley was at one time a great lake, the eastern shore of which was near the present sources of Big and Little Spring creeks; that when the lake disappeared, the sands of its bed were blown across the valley by the prevailing westerly winds, and deposited upon the wash and drift extending from the base of the main range immediately adjoining the mouth of the canyon of the Medaño, and constituting the divide between that stream .and Sand creek on the north, and Mosca on the south; that before this deposit occurred, the natural channel of Medaño creek extended through this wash and drift in the same general direction of the canyon, which would carry .it on a line with ■ the present channel of Big Spring creek; that the accumulation of Sand in its original channel finally became so great that the waters were unable to longer cut their way through and over it on the surface, but that they continued to flow or percolate through the course sand, gravel and boulders composing .the *324original bed of the creek, and finally emerged as the source of Big Spring creek; that the obstruction of the old channel prevented all the waters from pass-. ing into the valley in that way, and as the sands gradually accumulated and raised, the excess water flowed over them in a general southerly direction along the eastern base of the sand hills, uniting with Mosca creek, and resulting in the creation of Little Spring creek; that the continuous flow of Mosca creek upon the surface obstructed by the sand causes, and that the flow of Little Spring creek is practically on and over the ancient bed of Mosca creek. Scientific writers on the geological formations and conditions of the San Luis valley tend to support this theory. The testimony of the witnesses with respect to the present physical conditions within their own recollection also corroberate it. It is known that the channel which formerly existed from the Medaño to the head of Little Spring creek has in a great measure been obliterated within the past few years, by reason of the accumulation of' sand over its channel; and that the bed of the Medaño along the base of the sand hills is being gradually raised from the same source.
With the exception of the Medaño, after reaching the sand hills, the general drainage of the country is to the southwest. The course of that stream from the point where it intersects the sand dunes is almost at right angles with the course of the gulches and ravines between the canyon of the Medaño and Mosca creek, and at an angle of about forty-five de^ grees with the other principal streams issuing out of the Sangre de Cristo range on that side. The material which forms these hills must have been brought from some other point. That this is true is demon*325strated by the fact'that the process of accumulation of sand on and along them is still in operation. The flow and character of the Medaño is reflected in the Big and Little Spring creeks. Experiments, the results of which were introduced in evidence, indicate that it requires ab..ut the length of time for waters to pass from the Medaño to those streams which intervenes between a change in the Medaño and a corresponding change in them. These streams rise from amphitheatre-shaped basins, the water evidently coming from a stratum of saturated sand. There is no other source than the Medaño (except such as Mosca may contribute to Lillie Spring creek) from which it can be reasonably presumed that the waters of these streams are supplied. There is no other known outlet for the waters of the Medaño. So that, considering the topographical conditions, in connection with other facts stated by the witnesses, we are not required to enter the realm of conjecture in order to determine a connection between the Medaño and the Big and Little Spring creeks, but qn the contrary, the testimony leads irresistably to the conclusion that the Medaño is the source of supply of these streams, that at some time in the distant past the sand hills did not exist, and that during this period the Medaño found its way to the valley practically on a line from the mouth of the canon to the head of Big Spring creek. So with Mosca creek with respect to Little Spring creek. The accumulation of sand in and over the original channels of these creeks, composed, like all mountain streams of sand, gravel and boulders, would not render the old channels altogether impervious to water. The interstices of their original beds would not be filled by the sands accumulating on their surface. *326Water could still find its way along and through their ancient beds. This underground flow would naturally follow the ..respective original channels, bounded by the original banks, and while its passage may, in a great measure, be through sand and gravel, nevertheless, such a stream when well defined, is as much a part of the flow of the water of a creek, to which it contributes as if visible. Such conditions do not present a case of percolating waters, within the meaning of -the law. The subterranean volume of water which finds -its way through the sand and gravel constituting the beds of the streams which traverse the country adjacent to the mountains of this section, are recognized as a part of the waters of the stream to the same extent as though flowing upon the surface. That the surface bed of such a stream may not be visible does not change the rule with respect to this class of .flowing waters.. . Underground currents of water which flow in well-defined and known channels, the course of which can be distinctly traced, are governed by the same rules of law as streams flowing upon the surface. The channels and existence of such streams, though not visible, are “defined” and “known” within the meaning of the law when their course and flow are determinable by reasonable inference. Platte Valley I. Co. v. Buckers Co., 25 Colo. 77; McClellan v. Hurdle, 3 Colo. App., 430; Kinney on Irrigation, §48; Gould on Waters, § 281; Strait v. Brown, 16 Nev. 317.
In this connection it is urged by counsel for defendants that the waters of Sand creek may possibly contribute • to the supply of the waters of Big and Little Spring creeks. Sand creek rises in the same mountain range that the Medaño does, and to the north. From the point where the Medaño is first in*327tersected by the sand hills, and distant about five miles, Sand creek issues from the main range, flows for a short distance along the base of the sand hills on the west, and then continues in a generally southwesterly course over a bed of sand and gravel and through a country more or less sandy on either side. It has a defined channel all the way to the San Luis lakes, but at seasons of low water disappears entirely in the sands of its bed, long before it reaches these lakes. Its general course corresponds with the direction of the canyon through which it flows. It is a well known fact that lying between nearly all the streams ■issuing from the Sangre de Cristo range on the west there exists a defined watershed. That such a shed must have originally existed and been visible between the Medaño and. Sand creek is entirely probable.' For the waters of Sand creek through any subterranean channel to reach either Big or Little Spring creek would require a drainage almost at right angles in direction from the natural drainage of the country. That such a condition would exist is not probable. Besides, experience demonstrates that the volume of water of the stream which disappears in the sands of its bed would, follow the general course directly underneath the surface of such bed, so that it does not appear there are any facts or circumstances from which it can be reasonably inferred that Sand creek in any way contributes to either Big or Little Spring creek above the headgates of plaintiff’s ditches.
It is also urged that a decrease in the flow of these streams may be accounted for by reason of- the fact that many artesian wells have been sunk in the valley within the past few years, and that the water thus drawn off may have depleted the supply of these streams. These wells, with the exception of a few, *328are not,in the near vicinity of either of these streams and all are below and several miles to the west and southwest óf' the -respective springs forming their sources. . If the water supplying these streams in any manner contributes to the supply found in the artesian wells,-we do not see how it would be physically! possible for the wells, situated as they are, to deplete the flow at points distantly removed and above such wells. If water flowing through a stratum of sand is diverted at a point below its source of supply, it would not affect its level or flow above that point any more than a diversion of water from a visible stream upon the surface would affect its flow or volume above the point of such diversion.
It is also urged that injunction is not the proper remedy; that the complaints do not state- causes of action; that there is defect of parties; that the court-below was without jurisdiction, and that the decree is erroneous in unqualifiedly prohibiting the defendants from diverting any water between the dates specified:
The injury which the plaintiff was sustaining by reason of a shortage of the waters of the streams which he is entitled to divert and apply to irrigation uses is continuing in its nature. The defendants are-responsible for this shortage. His rights antedate theirs. An action at law against them by plaintiff' would be wholly inadequate, because it would require a multiplicity of suits to recover the damages which-he might sustain from year to year on account of the shortage of waters from Big and Little Spring creeks.
This is an action to protect priorities already established, and not to determine them- -Plaintiff states a prima'facie case in so far as it relates to an appropriation of water for the purposes of irrigation by pleading the adjudication of water rights in the *329water district in which the streams are situate, and the award to the ditches through which he claims such rights. In pleading this judgment he has substantially complied with section 65 of the civil code, which provides, in effect, that in pleading a judgment of a court of special jurisdiction, it is not necessary to state the facts conferring jurisdiction, but such judgment may be stated to have been duly given or made. It is not necessary to literally follow the prescribed formula; it is sufficient if from the averments it appears that the ultimate fact which may be so pleaded is otherwise stated.
A defect of parties must be raised by demurrer or answer. If objection upon this ground be not so taken, it is waived—Sec. 55 Civil Code. No question of this character was raised in the court below.
These actions are not statutory proceedings to adjudicate water rights; hence, the fact that adjudication proceedings had taken place in the district court of Costilla county did not preclude the district court of Saguache county from assuming jurisdiction of these cases.
Literally construed, the decree does inhibit the defendants from diverting any water between the dates specified. The law, however, reads into the decree that this inhibition only covers such portion of the period between the dates specified that plaintiff has use for the water for the purposes of irriga' tion.
The judgment of the district court is affirmed.
Afirmed