Six actions were brought by the State of Minnesota to determine adverse claims to the beds of certain lakes and connecting streams in Crow Wing County. The state alleged that it was the owner in fee simple of the lands constituting the beds of these lakes and streams. The riparian owners answered and alleged that, since the waters involved were nonnavigable, they were the owners in fee simple of the beds of these waters. The actions were consolidated for trial. By stipulation, the court limited the issue for determination to the navigability of the waters involved, leaving for future determination the boundary lines between lake and stream beds and riparian lands. The importance of these cases lies in the fact that the lake and stream beds involved are underlaid with valuable iron ore deposits. The court found all the lakes and streams involved nonnavigable waters and the title to the beds therefore in the riparian owners, except in the case of Little Rabbit Lake and the channel between it and the Mississippi River (case No. 36,848). Appeals were taken by the state from the orders of the court denying its motions for orders amending the findings of fact and conclusions of law or granting new trials.
The record is voluminous and the facts numerous. Because of the importance of this litigation, it has been deemed desirable and probably necessary to set out the facts more in detail than in the average case, *524with the hope, also, that such procedure will present as clear a picture of the situation as it is possible to do in words.
Rules Of Law Involved
In State, by Burnquist, v. Bollenbach, 241 Minn. 103, 109, 63 N. W. (2d) 278, 283, we said:
“In reviewing findings of fact, the rule in this state prior to the adoption of the new rules of civil procedure has been that the findings must be upheld if they are reasonably sustained in the light of the evidence as a whole. * * * Although now, by virtue of Rule 52.01 [Rules of Civil Procedure], the criterion for review is whether or not the findings are clearly erroneous, there is no practical change in the method of review of findings of fact nor was one intended. 2 Youngquist & Blacik, Minnesota Rules Practice, p. 648.”
As the questions raised by the state in these appeals involve the sufficiency of the evidence to sustain the findings of the court, we must review the record with the above rule in mind.
The ownership of the beds of navigable waters was granted to the State of Minnesota upon its admission to the Union on May 11, 1858. The title to all other lake and stream beds remained in the Federal government. State v. Longyear Holding Co. 224 Minn. 451, 29 N. W. (2d) 657; Bingenheimer v. Diamond Iron Min. Co. 237 Minn. 332, 54 N. W. (2d) 912; State, by Burnquist, v. Bollenbach, 241 Minn. 103, 63 N. W. (2d) 278.
The question of navigability is to be determined by Federal law. As stated in United States v. Holt State Bank, 270 U. S. 49, 55, 46 S. Ct. 197, 199, 70 L. ed. 465, 469:
“* * * Navigability, when asserted as the basis of a right arising under the Constitution of the United States, is necessarily a question of federal law to be determined according to the general rule recognized and applied in the federal courts.”
The Minnesota cases above cited recognized and applied this rule.
In Bingenheimer v. Diamond Iron Min. Co. supra, we reviewed quite fully the Federal cases which have considered the question of navigability. It would therefore seem unnecessary to do so again. After *525such a review, we stated that, under the Federal decisions as we understood them, the rule is that streams or lakes which are navigable in fact are navigable in law; that they are navigable in fact when they are used, or are susceptible of use, in their ordinary and natural condition, as highways for commerce; that ordinary and natural conditions refer to volume of water, the gradients, and the regularity of flow; and that a waterway otherwise suitable for navigation is not barred from that classification merely because artificial aids must make the highway suitable for use before commercial navigation may be undertaken. Before applying this test, we must determine what the facts are to which the test must be applied.
General Description Of The Waters Involved
The Rabbit Lake chain of lakes and streams, also called the Rabbit Lake basin, extends generally in a southwesterly and northeasterly direction, a short distance east of the Mississippi River and some miles to the north of the city of Brainerd. It is about 10 miles in length. At the top of the chain (the north end) lies Rabbit Lake, the largest body of water involved. It has an area of 1,160 acres and consists of 2 so-called bowls, connected by a channel about 700 feet wide and 12 to 15 feet deep. The depth of the lake varies from 0 to 40 feet. Below Rabbit Lake is Clinker Lake, which has an area of 96 acres and a depth of from 0 to 35 feet. Between Rabbit and Clinker is a channel or stream about 500 feet long. Below Clinker is Curley or Turner Lake. We shall refer to it hereafter as Turner Lake. It has an area of 64 acres and a depth of from 0 to 36 feet. Between Clinker and Turner is a connecting stream about one-half mile long. Below Turner lies Mahnomen Lake, the next largest lake, with an area of about 400 acres and from 0 to 25 feet deep. Between Turner and Mahnomen is a stream also about half a mile in length. Below Mahnomen is Pascoe Lake, with an area of 64 acres and from 0 to 15 feet deep. Between Mahnomen and Pascoe is a connecting stream about 800 feet in length. Below Pascoe is Little Rabbit Lake, with an area of 179 acres and a depth of from 0 to 30 feet. Between Pascoe and Little Rabbit is a connecting stream about 1.7 miles in length. At the lower end of Little Rabbit is an outlet stream, about 800 feet in length, which empties into the Mississippi *526River. The stream connections are called Rabbit River.
Besides the lakes and streams in the above chain, there are several lakes and streams that enter into the picture. Again, commencing at the upper end, we find a small lake, called Carlson Lake, which lies a short distance east of Rabbit Lake. A small stream flows from Carlson into Rabbit, the highest recorded flow being 82 gallons a minute. Just below Rabbit, directly east of Clinker and about the same distance from Rabbit, is a small lake called Mud Lake. A small stream flows from it into Clinker. About 2 miles from Mahnomen and to the south and east, lies Serpent Lake, a fairly large lake. A stream flows from Serpent into the upper end of Mahnomen. Just southwest of Mahnomen lies Jeune Lake, a small lake, and directly southwest of Jeune is Portage Lake, also a small lake. Both are landlocked. To the south of Jeune and Portage are Blackhoof and Little Blackhoof Lakes, less than a mile away. A stream flows westerly from these lakes and empties into Little Rabbit Lake.
As the stream from Blackhoof enters Little Rabbit and the court found in favor of the state in holding Little Rabbit navigable, we shall not enter into a discussion as to the navigability of the Blackhoof chain or Little Rabbit.
The Driving Of Logs On The Rabbit River Waters
The state claims that the waters of the Rabbit River chain were used to transport logs to the Mississippi River and that the logs were driven. It produced one witness who testified that he had actually seen this being done. Cornell Sogge, the witness, testified that in 1898, when he was 12 years of age, he had seen men poling logs through a dam a short distance below Pascoe. No other witness ever saw or heard of a log drive on these waters. On cross-examination, he said that he saw the men taking the logs out of Jeune Lake into Mahnomen and that they pulled them over a mining dump. He also said that he saw men shoving logs through a steel culvert. He also said that about the same time he and his brother went by boat from Mahnomen to Little Rabbit and on the trip went through a mining company trestle and that they had no trouble coming back up as there were no rocks or rapids in the stream. The witness was thoroughly discredited. The first *527mine on the Cuyuna Range was started in 1911, and the only mine on Little Rabbit was opened in 1914. The undisputed testimony is that there were both boulders and rapids in the stream between Pascoe and Little Rabbit. Because of old age and illness, Sogge may be excused from testifying as he did, but being thus charitable to him adds no weight to his testimony. The state claims that Sogge’s testimony above conclusively proves that logs were driven down the streams and that the court should have so found. From what has been recited, it does not seem possible that any court could give such testimony sufficient credence to base a finding on it to that effect.
Logging Dams On Rabbit River
The state claims that four logging dams had existed on the stream between Rabbit and Little Rabbit. Roy C. Pascoe, a witness for the state, testified that he saw two logs lying parallel in the stream below Pascoe Lake and a nearby fill. He concluded that what he saw were the remains of a wing dam. Edward D. Rhodes, a witness for defendants, described the same two logs but saw no evidence of a wing dam. He said that the logs were a rough pair of logs, rough knots on them and “hadn’t * * * been roughed off,” extending out from under a road fill. The state also claims that the uncontradicted evidence should have compelled the court to find the existence of a logging dam about 200 feet above the entrance of Rabbit River into Little Rabbit. Witness Pascoe described what he claimed were the remains of a dam at that point. He described them as logs set in the bed of the river at an angle and an old log across the bed of the river. From what he claimed he saw, he prepared a map and drawing of a complete dam. During cross-examination, a comparison of his map with the United States topographical map showed that a dam at that point would have had to be from 425 to 528 feet long to hold back the water because of a wide swamp along the river at that point. Another state’s witness, B. F. Hartman, testified that, when he was 16 years old in 1897, he trapped along the river and saw a dam at the same place. Hartman said that it was a cutaway dam. He was asked:
“Q. Was there any side or any earth wings or dikes?
“A. Always a few splinters where it looked like they were blowed *528out. The dam lay in the bank — logs where the ends laid in the bank.
“Q. Did you see the logs in there?
“A. The pieces of them.”
As stated before, at that point a dam over 400 feet long would be required to hold back the water. The state claims that the testimony of these two witnesses, as partially set out, conclusively proves the existence of a logging dam just above Little Rabbit. The court found otherwise and properly so.
The state also claims two dams between Turner and Clinker and bases its claim on the state’s sanie witnesses Pascoe and Hartman. Pascoe said that in 1948 he found two old rotted logs about 800 to 1,200 feet below the state control dam, a so-called WPA dam. On cross-examination, he said that he was not so sure where he saw the logs and finally said:
“Well, I am sorry if I gave the impression at that particular point there was a dam, I had not intended that. We simply found two logs at that location.”
The state claims that the testimony of Pascoe established a dam at that location. Witness Hartman said that about 50 years ago there were two dams between Clinker and Turner, the one below Clinker being a gate dam. But all that was left of that dam was one log in the bottom of the river. As to the second dam, when asked what kind of a dam it was, he answered: “I don’t recollect. That was pretty well rotted out, but big logs are across the creek yet * * He said that it was not a gate dam but a cutaway dam, and when asked to find the location of the dam he was unable to do so. The court properly refused to find two logging dams between Clinker and Turner.
Serpent Creek is not involved in these proceedings, except what bearing its existence might have upon the watercourse between the upper end of Mahnomen and Little Rabbit. The state claimed two dams on Serpent Creek. The court found that Serpent Creek was never used for driving logs even with artificial means. Without detailing the facts as we have done in connection with the claimed dams on Rabbit River, it is sufficient to say that the court, on the evidence, could very well find that no logging dams existed on Serpent Creek.
*529The court found that a dam existed at a point where the road now crosses the stream between Pascoe and Little Rabbit. It found that the dam was probably used to drive logs out of Little Rabbit into the Mississippi River. In 1888, a power dam was built across the Mississippi at Brainerd, about 9 miles downstream from Rabbit River. At the mouth of Rabbit River, the waters of the Mississippi were raised by this dam about 7 feet according to the testimony of defendants’ expert witness and 4 to 5 feet according to the state’s expert. The government survey of 1869 shows a swamp, what is now an arm of Little Rabbit, with 5 to 8 feet of water. In the area of dead trees, there are depths up to 7 and 8 feet. In the spring of 1950, when the water in the river was the highest ever known, a part of the Brainerd dam went out. A witness testified that at that time he canoed through the channel into Little Rabbit. The channel was only 6 to 8 feet wide, and his paddle touched bottom. Another witness, who drove the Mississippi before the Brainerd dam was built, said that the entrance to Little Rabbit at that time was “kind of a small creek * * * growed up with grass and weeds” and that there was not enough water to drive the logs out. The field notes of the government surveyor showed that on March 3, 1870, the width of the river where it flowed into the Mississippi was .9 chains or 59.4 feet. As the natural flow of the river was small, it is evident that, when spread over a width of about 60 feet, it must have been very shallow. That the dam in question below Pascoe was built to create a head of water to drive logs out of Little Rabbit into the Mississippi seems very probable, and the evidence supports the finding of the court to that effect.
Logging Operations On The Reservation
In 1857, the Indians of the Rabbit Lake Reservation, within which the Rabbit River basin was located, entered into a timber contract with one Morrison, giving him the right for a period of 10 years to cut logs on the reservation. This contract was not approved by the government, but, under some other arrangement, Morrison was permitted to cut timber and did so several seasons, the first one being 1857-1858 and the last one 1863-1864. Altogether he cut over 26,000,000 feet of white and red pine. The logs were floated down the Mississippi to *530St. Anthony, now Minneapolis. In his operations, Morrison used horses and oxen, having, according to a statement made by the Indians, 20 horses the first winter, that of 1857-1858. They also stated that in the winter of 1861-1862 he had four pair of horses “at our camp,” which undoubtedly referred to the Morrison camp near the Indian village. How many horses were used during the other seasons is not disclosed.
The field notes of the government surveyors made in 1868 to 1871 refer to “remnants” of four logging operations. It is assumed that they represent the area of the lumbering done by Morrison. One of the areas was east of and near the Indian village and probably is the one the Indians referred to as “our camp,” at which they said Morrison had four teams of horses stationed during the winter of 1861-1862. This logged area was between one and one-half and two miles from the Mississippi. As all of Morrison’s logs were to be driven down the Mississippi to St. Anthony, it does not now seem reasonable to suppose that Morrison drove these logs down the whole length of the Rabbit chain of streams and lakes, with all the inconveniences connected therewith, if the small streams could be driven at all, when the Mississippi River, the ultimate goal, was so nearby and he had the horses and equipment with which to do the hauling. As this is the only logging operation that could possibly have used the claimed two dams between Clinker and Mahnomen, it is very strong proof of the nonexistence of these two dams. There were also remains, according to the field notes, of old pine lumbering operations directly south of Mahnomen about two and one-half or three miles from Little Rabbit and three and one-half miles from the Mississippi at its nearest point. It is difficult to tell how extensive this logging area was, but it seems to be the smallest of the operations. This cutting, although it is the farthest distance of any of the four from the Mississippi or Little Rabbit, cannot be considered a long haul to either landing. A four- to eight-mile haul is not uncommon, according to the testimony. Furthermore, for most or practically all of the distance there would be a level haul over frozen lakes, Mahnomen and Pascoe, swamps, and the meadow along Rabbit River. It seems fair to assume that the logs from this cutting were hauled to one of the landings indicated, rather than being banked on Mahnomen and then *531driven through the river and Pascoe and the two claimed dams between Pascoe and Little Rabbit. This cutting is the only one that could possibly use these two dams, and, as it seems most improbable that the logs were routed that way, it argues the nonexistence of the dams. The field notes call attention to another cutting of heavy growth with a considerably larger area than the two already mentioned. The west end of this cutting was about half a mile from Little Rabbit. Morrison certainly would not drive the logs from this cutting down Rabbit River, and that he would bank these logs at any point except at Little Rabbit does not seem reasonable. The field notes also disclose remnants of old logging operations near the Mississippi and the angle area between that river and Little Rabbit. No one would claim that Morrison would have used any waters of the Rabbit River chain to float out these logs, unless it would have been Little Rabbit.
The state argues that the logs from the two areas first described were driven from Serpent Lake into Serpent Creek and through the two dams into Mahnomen and then down the rest of the Rabbit River chain through additional dams into Little Rabbit. In view of the short distances to the Mississippi and Little Rabbit and the statement made by the Indians that Morrison used 20 horses the first year he logged (1857-1858); that during the winter of 1861-1862 he had four teams at the camp near the Indian village; and that he undoubtedly used horses the other logging seasons, it does not seem reasonable or practical that he used the Rabbit River chain.
In 1900, one Henry Shields contracted to cut timber on certain tracts of land on Pascoe near the outlet stream. He agreed to bank the logs on “loar Rabbit Lake and drove into Mississippi.” He was thus agreeing to cut logs farther away from Little Rabbit than were the two logging areas last described and to bank them on Little Rabbit, although he was going to log the very shores of the claimed waterway. One witness testified that the dam which he claimed was located below Pascoe was called Shields dam. Apparently Shields, in his logging operations, did not use a dam.
The state in its brief insists that four logging dams existed on Rabbit River and that the trial court disregarded conclusive testimony when it found otherwise. It says:
*532“The evidence conclusively establishes the location of four old logging dams in Rabbit River, * *
Not only do we find insufficient evidence to support the state’s claim of the existence of these dams, but we can find no reason for their existence. It is interesting to note that when government surveyor Ingerson surveyed T. 47, R. 29, in 1871 he made this comment in his field notes:
“There evidently has been a good deal of pine in this Township, of which some has been destroyed by fire and stolen and probably run down the Mississippi.”
This is the township which included one of the areas logged by Morrison. Ingerson evidently did not know of the Morrison contract. This is the only reference to driving logs in the surveyor’s notes and is significant. In neither the survey of Butler nor of Ingerson, the two government surveyors, do we find any reference in their field notes that they saw any remains of old logging dams, although three of the four claimed dams were within 500 feet of the section lines which these surveyors ran, according to state’s exhibit 95. One of the claimed dams was very near the logging camp on Clinker, which was referred to in the field notes.
Physical Changes In Rabbit River Basin
The physical changes which have taken place in the Rabbit River basin during the past half a century make it difficult at the present time to determine the amount of water which the watercourse carried in its original and natural state. In 1911, the first mine was opened near Little Rabbit. Now 16 mines, most of them open pit, are or have been operated along the watercourse. Many of the mines pump water out of the pits into the stream. In 1943, the Youngstown Mines Corporation received a permit to build a dike across the channel between the two bowls of Rabbit and to dewater the east bowl-by pumping the water into the west bowl. For at least five years it pumped at least 15,000 gallons per minute, which water of course had to be carried away by the Rabbit River channel. Then in 1937, the state built two dams, one near Clinker Lake to control the water of Rabbit and the other just below Pascoe to control the water of Mahnomen. All these changes contributed to variations in the water flow and to material *533alterations in the physical characteristics of the stream and riparian lands. Because of the dam, it is apparent that Mahnomen is higher now than it was at the time of the original survey.
Rabbit River Between Rabbit Lake And Mahnomen Nonnavigable
At the present time, the only outlet of Rabbit is the so-called “cutaway” which runs a distance of 500 feet southerly into Clinker. The court found that the cutaway was not a natural outlet but was manmade sometime after the original government survey in 1871. Rabbit Lake was meandered, and the subdivisions of the township within which it lies were surveyed in May and June 1871. At that time the surveyor noted no outlet of Rabbit in his field notes and the plat he made, although his field notes showed the small inlet from Carlson Lake, which inlet, however, was not noted on the plat. On the aerial map, the cutaway seems to be straight, even where it passes through a swamp. It has features of an artificial channel but also those of a natural one. Why it was dug, if dug, and by whom are unanswered questions. There is no evidence of any logging having been done in the vicinity of Rabbit Lake from 1871 to 1900, which would make an artificial channel either convenient or necessary. One expert was of the opinion that the channel was artificial and another that it was natural. Some witnesses testified that it was referred to as a cutaway and considered to have been cut by man. One witness testified that he had heard it considered both ways. As stated, the court found that the channel was manmade and therefore artificial; consequently, that Rabbit was a landlocked lake. Under the rule, there is probably sufficient evidence to sustain the finding. But whether the channel be held to be artificial or natural, such finding is of little materiality because of other facts and findings in the case.
The plats and field notes of the original government surveys were offered by the state and received in evidence. In 1868, one Butler surveyed the north line and the subdivisions of T. 46, R. 29. And in 1871, the subdivisions of T. 47, R. 29, were surveyed by one Ingerson. Neither surveyor noted in his field notes a stream entering Mahnomen Lake from the direction of Rabbit Lake. Nor do the plats made by these surveyors show such a stream. State’s witness Alfred O. Anderson, *534a surveyor, testified that, on the basis of the original surveys, the head-water of Rabbit River was in Serpent Lake. When Butler meandered Mahnomen, he did not show any stream entering the lake where the stream in controversy now enters. In other words, he found no Rabbit River inlet to the lake. Nor is there any mention of an inlet in his field notes. He showed Serpent Creek, a small stream, entering from Serpent Lake to the south. Butler also ran the north line of T. 46, R. 29, in 1869. This is also the south line of T. 47, R. 29, the subdivisions of which were surveyed by Ingerson in 1871. Ingerson noted the stream which came out of Clinker Lake. Whenever and wherever he crossed it, his field notes show it. He showed that the creek never entered Mahnomen but dissipated or disappeared in the large bog north of the lake. This bog is at least 1,000 feet wide at its widest point north of the township line and extends for a distance of three-quarters of a mile. The size of the bog below the township line is not shown. The termination of the stream coincides exactly with Butler’s plat. That Ingerson was not unmindful of the stream is evidenced by the fact that whenever he crossed it below Turner Lake he made note of the fact in his field notes, and his plat shows its disappearance in the swamp at least 400 feet from Mahnomen.
When two official government surveyors, working separately two years apart, find the creek here involved did not in 1869 or 1871 enter Mahnomen Lake, it seems plain that there is sufficient evidence to sustain the finding of the court that the stream did not enter Mahnomen in 1858. The state adopts the only course which would seem to have any possibility of success in combating the fact of the nonexistence of the inlet stream to Mahnomen by attempting to prove inaccuracies in other respects of the surveys made by both surveyors. Many folios of record and brief are devoted to this endeavor. It would be unprofitable to detail all the evidence and arguments pro and con, but an examination of the record and exhibits in the light of the briefs and oral arguments convinces us that the surveyors did a creditable job, and, since it can be demonstrated that in other respects their surveys measure up in correctness with other early surveys, it does not seem possible or probable that both surveyors could be mistaken, either through negligence or fraudulent design, as to the existence of the so-*535called Rabbit River inlet to Mahnomen Lake. It is not a case of faulty surveying in the use of angles and other measurements. It has nothing to do with their skill as surveyors. When they failed to note the claimed outlet from Rabbit and the inlet into Mahnomen, they in effect say that they did not see the claimed streams, and no reason suggests itself why they should not have seen them if they in fact existed at the time. They apparently noted all other streams, even the small rivulet from Carlson Lake to Rabbit. It is not noted on the plat, probably because of its insignificance. Either the outlet or the inlet did not exist or they failed for some reason, not understandable, to record what they saw. The court was fully justified in finding as it did on this feature of the case. A contrary finding on such evidence would be difficult to sustain, unless the court should say that both surveyors were unobserving or dishonest, and there are no indications in the record that such was the case.
Ingerson also meandered Rabbit and noted no outlet in his field notes or plat. The state also attempts to discredit this survey. Because of the fact that the court was fully justified in finding no Rabbit River inlet to Mahnomen in 1858, no navigable stream existed between Rabbit and Mahnomen, and whether the cutaway between Rabbit and Clinker was artificial or natural becomes immaterial. Certainly, the stream which became lost in the big swamp before it reached Mahnomen could not possibly be a navigable stream under the Federal test or any other rule.
Rabbit River Between Mahnomen And Little Rabbit Nonnavigable
We have determined that on the evidence the court did not err in finding the stream between Rabbit and Mahnomen nonnavigable, among other reasons, because of the fact that it was only a watery culde-sac and led nowhere. We will now consider the evidence relating to the stream between Mahnomen and Little Rabbit. At all times .in its natural and original state, Rabbit River could have been nothing more than a very small stream. We can recall no case where a stream so small has been held to be navigable. In 1942, according to state’s exhibit 97, at a point half a mile below Clinker, the stream was 4 feet wide *536and 1.1 feet deep, and about 300 or 400 feet farther on it was 13 feet wide and .8 foot deep. Two witnesses said that they had jumped or stepped across the stream at that point. Exhibit 97 shows that just above the dam below Pascoe it was about 50 feet wide and 5 feet or more deep, practically an extension of the lake. It had no visible current, with a wet bog on one side and a floating bog on the other. But, for a mile below the dam, it averaged 20.4 feet wide and 1.06 feet deep. The shallowest place was six-tenths of a foot deep, and the deepest spot 1-5 feet. The flow was 10,000 gallons per minute. Of this amount, five-sixths or 8,400 gallons were being pumped into the stream from nearby mines. The remaining 1,600 gallons per minute may be considered the natural and normal flow. To a layman, the most understandable evidence as to the size of the stream below Mahnomen is that, up to the time of the dewatering of the east bowl of Rabbit, it passed under the road just below Pascoe through two 18-inch culverts. There is no evidence that these culverts were insufficient to take care of all the water that came down the watercourse during spring thaws and freshets. Not only that, there is no evidence that they were not able also to take care of the additional 8,400 gallons per minute pumped into Mahnomen from six adjacent mines. These small culverts are compelling evidence of the small size of Rabbit River in its original and natural state.
No navigability is claimed for the stream except for the transportation of logs. In the distance between Pascoe and Little Rabbit, the stream curves and meanders between high banks, the curves being so severe that in places the stream nearly runs back upon itself. The size of the stream and its meanders are such that it may not be possible or practicable to drive logs, the only claim of navigability for the stream, even with artificial structures such as dams. In view of the nearness of Mahnomen to Little Rabbit, no one, it would seem, would go to the expense of building dams and to the trouble of transporting the logs through sluiceways down a small tortuous shallow creek when a short haul is available. The Shields contract of 1900 demonstrates that fact. The court found that, because of the small flow and other characteristics, the stream as a whole was not navigable and was not susceptible of being used as a highway of commerce.
*537In United States v. Appalachian Power Co. 311 U. S. 377, 406, 61 S. Ct. 291, 298, 85 L. ed. 243, 252, the court said:
“Both lower courts based their investigation primarily upon the generally accepted definition of The Daniel Ball. In so doing they were in accord with the rulings of this Court on the basic concept of navigability. Each application of this test, however, is apt to uncover variations and refinements which require further elaboration.”
The Daniel Ball definition, 77 U. S. (10 Wall.) 557, 563, 19 L. ed. 999, 1001, has been quoted in practically all the cases. It reads:
“* * * Those rivers must be regarded as public navigable rivers in law which are navigable in fact. And they are navigable in fact when they are used, or are susceptible of being used, in their ordinary condition, as highways for commerce, over which trade and travel are or may be conducted in the customary modes of trade and travel on water.”
The court in the Appalachian Power Company case continues (311 U. S. 407, 61 S. Ct. 299, 85 L. ed. 252):
“To appraise the evidence of navigability on the natural condition only of the waterway is erroneous. Its availability for navigation must also be considered. ‘Natural and ordinary condition’ refers to volume of water, the gradients and the regularity of the flow. A waterway, otherwise suitable for navigation, is not barred from that classification merely because artificial aids must make the highway suitable for use before commercial navigation may be undertaken. * * * The district court is quite right in saying there are obvious limits to such improvements as affecting navigability. These limits are necessarily a matter of degree. There must be a balance between cost and need at a time when the improvement would be useful.” (Italics supplied.)
If the court had found in this case that the stream, although insufficient to transport logs in its original and natural state, still with the use of artificial aids would be navigable, under the Appalachian Power Company case, as we understand it, that would not be sufficient to prove navigability. Unlimited artificial aid is not countenanced under the rule in that case. There is a limit beyond which the improvement of the stream may not go, “obviously” so, the court states, as (311 *538U. S. 407, 61 S. Ct. 299, 85 L. ed. 253) “There must be a balance between cost and need at a time when the improvement would be useful.” Applying that rule here, it is apparent that any investment in dams would be at an excessive cost in comparison with the need. The need for transporting logs from the area near Mahnomen to Little Rabbit is nil, even if logging days would return, because of the nearness to the landings at Little Rabbit and the Mississippi. The Shields contract of 1900 illustrates this. Any cost or expense in constructing dams clearly would outweigh the need. What is said here would apply equally to the stretch of waterway between Rabbit and Mahnomen, if such waterway had been found to exist.
In Wagner v. City of Baltimore, 210 Md. 615, 625, 124 A. (2d) 815, 820, the court, in commenting on the statement in the Appalachian Power Company case relative to the balance between cost and need, said:
“* * * The rule now recognized by the Supreme Court goes beyond the former rule in that waters which, though not navigable in fact in their natural state, may be made navigable in fact without undue cost, are navigable waters.”
Historical Evidence
Historical evidence may be important in determining navigability of waters. If such evidence shows actual navigation in any manner, then navigability is established. In this case, there is no historical evidence sufficient to show navigation or navigability of the waters here involved. In 1836, one Nicollet made a journey covering a large part of what is now northern Minnesota and Wisconsin. He produced a map showing the geographical data which he claims he found. The state claims that the map shows water routes across and through waters here involved and that, upon the basis of the map, the court should have found the waters navigable- Certain dotted lines are found running easterly overland to other lakes and finally overland to Mille Lacs. This, the state’s expert historical witness explained, was not a water route but an overland one. Furthermore, it did not follow the Rabbit River chain and can therefore have no bearing on the navigability of these waters. The early logging operations and Indian travel are being treated elsewhere.
*539The Claimed Trading Post
Certain articles consisting of bones, a fork, at least a dozen damaged clay pipes such as were usually used in Indian trade, and other small items were found some years ago near what appeared to be a small stone wall of clay masonry. The state insists that this evidence conclusively proves the existence of a trading post and attacks the court’s finding that there was not sufficient proof of a trading post at that point. The articles were found a short distance west of the rapids between Pascoe and Little Rabbit. The evidence certainly does not conclusively show that a trading post existed, and the court could very well find that one did not exist. Why a trading post should be located at that point is difficult to understand. If the evidence had been sufficient and the court had found its existence, that fact would have little or no bearing on the question of navigability of Rabbit River, as the point where the claimed trading post was located was some distance from the stream, which at that place was not navigable by canoe or boat because of the rapids and boulders. The location was but a comparatively short distance from Little Rabbit and the Mississippi; a little better than a mile from Little Rabbit and two miles from the Mississippi. The Indians used the portage trail on the east side of the stream. The, existence of the claimed trading post, if it did exist, would add no proof to the question of navigability of Rabbit River.
Rabbit Lake Indian Reservation
In 1855, the Rabbit Lake Indian Reservation was established. In 1865, the reservation was abolished and the land ceded back to the government of the United States. The Mississippi River was the north and west boundary of the reservation. When, in 1858, Minnesota was admitted to the Union, the Rabbit River basin was within an Indian reservation. The Constitution of the United States grants to Congress the power to regulate commerce with the Indian tribes. The argument of the state seems to be that, since Indians lived on the reservation, their comings and goings constituted commerce and the waters within the reservation were therefore navigable waters, or that the fact alone that these waters lay within an Indian reservation was sufficient to make them navigable waters of the United States. The fact that the waters *540here involved were within the Rabbit Lake Indian Reservation can have no bearing on the question whether such waters were navigable under the Federal rule or any other rule. Whatever travel there was by the Indians, within the reservation or outside of it, did not constitute commerce according to the evidence in the case. The evidence shows that there was an Indian village with a population of about 160 adults located at the south end of Rabbit, between Rabbit and Clinker. There, was an Indian trail from the Mississippi to the north end of the west bowl of Rabbit, 1,000 feet away. Presumably the Indians canoed between the trail and the village. From Little Rabbit an Indian trail extended to Mahnomen, with water crossings over Jeune and Portage. Toward the east from Rabbit was an Indian trail which extended as far as Mille Lacs, with several lake crossings. These are the only Indian trails within the reservation as far as any maps show. There is no evidence that the Indians traveled the length of the Rabbit River chain or had a canoe route in that area, and, if they did not use it as a route for travel at all, they of course could not have engaged in commerce over the route. The question here involved is the existence of commerce over these waters and their suitability for commerce. The state claims that there is no evidence to support the finding of the court that the Indian travel was not water travel and that it did not constitute commerce. There is no evidence that the Indians carried any articles of commerce over these waters.
In State, by Burnquist, v. Bollenbach, 241 Minn. 103, 63 N. W. (2d) 278, the evidence showed more extensive Indian travel and activities over Five Lake, the water there involved, than over the Rabbit River chain, and we held that such activities of the Indians or the evidence disclosed provided no basis for finding that a body of water is navigable under the Federal test.
State's Expert Testimony
The state, in its attempt to show that the amount of water that would or did flow through the Rabbit River waterway was sufficient to make it a navigable stream, called an expert witness, Mr. Frellsen. He based all his calculations on a 1,200-foot elevation of Rabbit, which elevation was reached only once within the memory of living men, and *541that was in 1950, when natural water levels were the highest known. In addition, at that time 16,000 gallons per minute were being pumped into the west bowl of Rabbit from the east bowl. It must be obvious that calculations based on an assumed elevation of 1,200 feet, with the additional artificial flow of 16,000 gallons per minute, and without taking into consideration the influence of the control dams, can have little or no value in determining the flow of water in Rabbit River under normal and natural conditions. The elevation of Rabbit for 13 yearly readings within the period of 1914 to 1942 showed an average elevation of 1197.8, the lowest being 1196.6. State’s expert, Mr. Meyer, also made calculations based on a Rabbit elevation of 1198.5, again considerably below the average. Using such elevations as their basis, these experts were of the opinion that sufficient water flowed through the channel'in its normal and natural state for the floating of logs, the only use claimed for the waterway. It is plain that the court was not bound by the opinions of the experts based on such calculations founded on such high elevations. Furthermore, since the testimony shows no channel existed through to Mahnomen from Rabbit at the time the state was admitted to the Union, the expert calculations and opinions are of little or no value in determining the navigability of the stream.
Rabbit Lake
Rabbit Lake, before dewatering, had a water area of 1,160 acres and up to 40 feet in depth. It consisted of two so-called bowls of about equal size, with a channel about 15 feet deep and 700 feet wide connecting them. Each bowl is about a mile across. A small creek coming out of Carlson Lake, with a maximum flow of 82 gallons a minute, flows into the lake at the north end. It has no other inlet. In the south end is the so-called cutaway, already described. Whether the cutaway was natural or artificial, the watercourse disappeared in a large swamp north of Mahnomen, so whatever outlet Rabbit had ended in the swamp. The court found that Rabbit was not navigable by itself or in connection with any other waters. The lake was never used in connection with floating or driving of logs or transportation of any articles in commerce. There is no evidence that any logging was done *542near the shore of the lake, except that in 1900 one Lamere contracted to cut the white and Norway pine on certain tracts of land lying adjacent to the lake and to bank the logs on the Mississippi River. The field notes of the government surveyor, who subdivided the fractional township in which the greater part of Rabbit is located, found considerable hardwood timber, aspen, tamarack, and some pine and much burned-over territory. He notes: “there are many wide stretches almost destitute of timber and many windfalls caused by fire.” Indians had a village south of the lake. There was also a trail which led east from Rabbit. There is no evidence that the Indians used the Rabbit River waterway except as stated. There is no evidence that the Indians engaged in any commercial activities on Rabbit, and the same is true of the white settlers. The lake has been used for recreation, such as boating, swimming, and fishing, and three resorts engaged in the business of renting out cabins and furnishing boats and other recreational equipment were located on the lake. The court found that Rabbit was not navigable by itself or in connection with any other waters, and the evidence sustains the finding.
The navigability of a lake was recently considered by this court in State, by Burnquist, v. Bollenbach, 241 Minn. 103, 63 N. W. (2d) 278. The lake there in question was Five Lake, a meandered lake of about 200 acres, up to 40 feet in depth, and about a mile long and of a maximum width of about half a mile. Five Lake has no inlet or outlet. It has never been used for logging purposes. The trial court found that Five Lake was nonnavigable in 1858, the year Minnesota became a state; that it was not used as a highway for commerce or travel; and that it had never been used for purposes of commerce or travel since the admission of the state to the Union nor has it at any time been capable or susceptible of such use. We held that, although Five Lake has not been used as a highway for useful commerce, it could still be deemed navigable in 1858 if at that time it was in its natural and ordinary condition susceptible to such use. The court said (241 Minn. 120, 63 N. W. [2d] 289):
“* * * The susceptibility of a body of water for use as a highway for useful commerce is determined mainly by two factors: (1) Whether *543the body of water has physical characteristics which permit and are conducive to water travel; and (2) whether the body of water is situated in a location useful to commercial trade and travel.”
As to physical characteristics, Rabbit has sufficient area and depth to comply with the first factor. So did Five Lake. But the location of Rabbit is such that it is difficult to see how it can be useful to commerce. It is not quite as landlocked a lake as Five Lake, but the nature of its outlet is such that, from a practical point of view, there is very little difference. The deadend watercourse had no commercial potentialities. Except as to size, there is very little difference in characteristics between Five Lake and Rabbit. Lakes may be smaller than either Five Lake or Rabbit and still, because of their locations, may be useful for commerce. The usefulness for commercial activities depends on the characteristics of each individual lake, and one cannot say arbitrarily that a lake of so many acres is navigable but that any lake of less acreage is nonnavigable. Navigability of a lake is a fact question, and other factors in addition to its size must be considered. The significant and all-important thread which runs through all the cases dealing with the navigability of a stream or lake is that of commerce, either actual commerce or suitability and susceptibility for use in commerce. The commerce may be large or small, depending on the area furnishing it, but still it must be commerce. A landlocked lake, if sufficiently large to furnish a route of useful commerce within itself between places which have a need, actual or potential, for such route of commerce, of course comes within the Federal test. Also, a small lake may be navigable if so located as to provide a commercial route. The physical capabilities of a lake, together with its location and all surrounding circumstances, determine its navigability. Thus, it will be seen, each lake must be treated as an individual body of water.
In Leovy v. United States, 177 U. S. 621, 632, 20 S. Ct. 797, 801, 44 L. ed. 914, 919, in dealing with the question of navigability of waters of the United States, the court said:
“It is a safe inference from these and other cases to the same effect which might be cited, that the term, ‘Navigable waters of the United States,’ has reference to commerce of a substantial and permanent *544character to be conducted thereon.”
Of course, our situation does not involve “Navigable waters of the United States,” as that term is used and understood, but the idea of navigability includes that of commerce. What has been said of Rabbit applies to Clinker, Turner, Mahnomen, and Pascoe. They are smaller lakes and not situated in locations useful for commerce. They have not been used for commerce and do not provide practical routes for commerce, and no lake connects points between which they would be useful as a practical route for navigation.
Navigability of lakes has been considered in the cases listed in footnote and the lakes involved found nonnavigable.1
Rabbit was never used as a highway of commerce because it afforded no highway or avenue of practical importance for commerce. While the lake is large enough to float a good-sized boat, it is not wide enough or long enough to provide a practical route for the transportation of commodities or passengers in any direction. The beginning and the end of the highway must be such that useful commerce would naturally go between them.
In United States v. Holt State Bank, 270 U. S. 49, 46 S. Ct. 197, 70 L. ed. 465, Mud Lake in Minnesota was held navigable. It was eight miles long and of sufficient depth. It actually was used by the early settlers and merchants for the transportation of supplies and merchandise. It had a location which made it useful for commerce.
In State v. Longyear Holding Co. 224 Minn. 451, 29 N. W. (2d) *545657, Syracuse Lake, with an area of only 33 acres, was involved. The court held it navigable. Not only did it have a location which made it useful for commerce, but it had actually been used for commerce. For many years prior to Minnesota’s admission to statehood, it was part of a commercial route used by Indian trappers and by individual fur traders, as well as agents of the Northwest Fur Company and American Fur Company. It had also been used by timber and lumber interests for moving down forest products and floating down logs.
In Taylor Fishing Club v. Hammett (Tex. Civ. App.) 88 S. W. (2d) 127, 129, the court said:
“* * * Every inland lake or pond that has the capacity to float a boat is not necessarily navigable. It must be of such size and so situated as to be generally and commonly useful as a highway for transportation of goods or passengers between the points connected thereby. It must either alone or in connection with other bodies of water connect points between which it is practical to transport commerce by water.”
These cases illustrate the importance of the existence of commerce or susceptibility for commerce in the determination of navigability of lakes. Susceptibility or availability, of course, depends chiefly on.their location.
Prior Proceedings
The state claims that the defendants here are estopped to claim and prove Rabbit and the cutaway were not navigable under the Federal rule. In 1943, in proceedings following its application for permit to dewater the east bowl of Rabbit, Youngstown Mines Corporation submitted a certain map (exhibit 97) showing the Rabbit River watercourse, and the state bases its claim of estoppel on certain data found on that map. If an estoppel has been created against Youngstown, and we are not considering that question, it could affect only Youngstown and no other party or parties. Youngstown cannot be affected by the result in these cases, since it holds a mining lease from the State of Minnesota covering the iron ore in the bed of the lake and also holds mining leases and numerous easements for drainage from the riparian owners. It is not disputing the state’s title in common with the riparian *546owners, and its claim of neutrality in this controversy is substantiated.
The state also claims that, under the decision of this court in Petraborg v. Zontelli, 217 Minn. 536, 15 N. W. (2d) 174, the question of navigability and naturalness of the cutaway from Rabbit to Clinker is res adjudicata against the defendants in these actions. The Petraborg case involved an action to enjoin the Youngstown Mines Corporation from dewatering the east bowl of Rabbit and constructing a dike across the narrows. The state was not a party. Navigability of the lake under the Federal test was not involved. It was simply an action by riparian owners to enjoin the draining of the lake without their consent. The decision was based exclusively upon the ground that riparian owners had the right to the maintenance of the waters in their natural condition. That right must exist whether a body of water be navigable or nonnavigable. The title to lake or stream beds was not involved. Most of the defendants in the instant cases were neither parties nor privies of the parties in the Petraborg case. The judgment in that case certainly cannot be res adjudicata as to them. The ownership of beds of streams and lakes is quite a different matter from the right to control waters.
As we find no errors that require a reversal, the orders of the trial court appealed from are affirmed.
Affirmed.
United States v. Ladley (D. Idaho) 42 F. (2d) 474; Hodges v. Williams, 95 N.C. 331, 59 Am. R. 242; Lembeck v. Nye, 47 Ohio St. 336, 24 N. E. 686, 8 A.L.R. 578; Cobb v. Davenport, 32 N. J. L. 369; Monroe v. State, 111 Utah 1, 175 P. (2d) 759; McGahhey v. McCollum, 207 Ark. 180, 179 S. W. (2d) 661; Bernot v. Morrison, 81 Wash. 538, 143 P. 104; Snively v. State, 167 Wash. 385, 9 P. (2d) 773; Smith v. State, 184 Wash. 58, 50 P. (2d) 32; Lefevre v. Washington Monument & Cut Stone Co. 195 Wash. 537, 81 P. (2d) 819; State v. Sweet Lake Land & Oil Co. 164 La. 240, 113 So. 833; Hardin v. Jordan, 140 U. S. 371, 11 S. Ct. 808, 35 L. ed. 428; United States v. Oregon, 295 U. S. 1, 55 S. Ct. 610, 79 L. ed. 1267; Harriso6n v. Fite (8 Cir.) 148 F. 781; Toledo Liberal Shooting Co. v. Erie Shooting Club (6 Cir.) 90 F. 680; North American Dredging Co. v. Mintzer (9 Cir.) 245 F. 297; State v. Brace, 76 N. D. 314, 36 N. W. (2d) 330.