delivered the opinion of the court:
These cases were selected by counsel for submission as representative of about seventy pending cases, in all, founded upon the same general facts. They are submitted together but are not consolidated. For these reasons findings of fact are made which are general in their character and applicable to all the cases, followed by findings as to the particular facts in the individual cases. It is thought that the findings very fully and accurately present the whole situation with which we have to deal and an attempt to restate the facts here in condensed form is not only unnecessary but, since a condensation must sacrifice something of valuable detail, is undesirable. We will refer to the facts found as may be necessary in the discussion of the questions involved.
If there is to be recovery in these cases it must necessarily be upon the theory that there was a taking of the plaintiff’s property within the meaning of the fifth amendment to the Constitution. There is much in the record which can only be interpreted as in support of a theory that there could be a recovery for damages, pure and simple, but that matter is hardly for discussion and the cases are to be treated as for a taking.
It is perhaps appropriate and also important, to the end that we may not lose the proper viewpoint and because also *133of the suggestion that, in circumstances stated, the distinction between “ appropriation of land ” and “ damage ” is a “ quibble,” that we consider the primary importance of and necessity for the distinction in this court. The fact that the distinction may not be material in California because the constitution of that State provides against “ damage ” as well as a taking without compensation can be of no force here and the distinction is vital. Perhaps there may sometimes be apparent a disposition to resolve a doubt in favor of a claimant since in no other way can compensation be awarded, but the necessity for the distinction is deep-seated and a proper drawing of the line is of more importance than if it merely involved a liberal interpretation of facts or effects in aid of a deserving claimant. This is because the question to be determined as between a taking and mere damages is in this court jurisdictional. Unless there is a taking within the meaning of the Constitution, implying an obligation to pay, this court is without jurisdiction in this class of cases. There are no presumptions to be indulged in favor of jurisdiction, it can not be assumed if it does not in fact exist, it can not be conferred by consent of parties, it must affirmatively appear, and it is a question for strict construction. The petitions in these cases aver a taking and upon the averment of that jurisdictional fact the court was empowered to hear the cases, but if upon the facts submitted it perchance appears that what the defendant did resulted in the infliction of a consequential injury and did not in fact amount to a taking, it is at once demonstrated that the case is not within the jurisdiction of this court. Therefore, to put a liberal construction on facts appearing to the end that a plaintiff’s injuries may be treated as a taking rather than a consequential damage is but to broaden jurisdiction by liberal construction. If we were proceeding under a constitution like that of the State of California it would not be material that the line was not clearly drawn between a taking and damage, for both are included, but under our Constitution the question is primary and vital.
In Ex Parte McCardle, 7 Wall., 506,-515, it is said:
“It is quite clear, therefore, that this court can not proceed to pronounce judgment in this case, for it has no longer *134jurisdiction of the appeal; and judicial duty is not less fitly performed by declining ungranted jurisdiction than in exercising firmly that which the Constitution and the laws confer.”
And in Reid v. United States, 211 U. S., 529, 539, it is said:
“But jurisdiction is not a matter of sympathy or favor. * * * The courts are bound to take notice of the limits of their authority.”
“The facts upon which the jurisdiction of the courts of the United States rests must, in some form, appear in the record of all suits prosecuted before them. To this rule there are no exceptions.” Ex Parte Smith, 94 U. S., 456. And a general averment in the petition, sufficient for the purpose at that stage of the case, becomes of no avail for jurisdictional purposes if the facts proven do not sustain it but show the subject matter of the case to be of a different general class as to which there is no jurisdiction. Therefore, the necessity for a strict determination whether the subject matter of the case is a taking or a damage and if the latter, a dismissal, not primarily because it is determined that the plaintiff has not suffered by the acts of the defendant, or because the defendant, the sovereign, has not consented to be liable therefor, but because it has not conferred on the court any jurisdiction to adjudicate the subject matter. In the rather recent case of Temple v. United States, 248 U. S., 121 at 131, it'is concluded by the Supreme Court that “ The District Court, instead of rendering judgment for the United States, should have dismissed the suit for want of jurisdiction.” Many other cases recognize the question as jurisdictional.
There are many cases in which the Supreme Court has had before it and has decided questions as to whether the facts showed a taking within the meaning of the fifth amendment or a consequential injury.. A review of them at length does not seem to be necessary for present purposes. The deducing of general principles with reference to a few particular cases should serve our purpose.
From the pioneer case of Pumpelly v. Green Bay Co., 13 Wall., 557, we have in another case deduced the rule that “ an actual and continuing invasion of one’s property by *135superinduced water effectually destroying or impairing its usefulness constituted a- taking under the Constitution though there be no actual conversion to public use.” But it should be added, without necessarily reciting the facts in detail, that in this case the overflow was caused by the erection of a dam across a river which was intended to back up the waters of the river, the result of which, clearly to be anticipated, was to overflow the lands in question.
It is well to bear in mind in the consideration of this class of cases that when the United States, in carrying on its many activities in aid of navigation, builds dams across rivers the very purpose of the structures is to back up the waters and create pools above the dams. The pools thus formed sometimes extend several miles up the river, dependent on the height of the dam and the fall of the river, and extend up tributaries also. It is a simple matter of engineering to prolong a level from the crest of the dam up the stream and its tributaries, and lands, if any, which are below that level will, of course, be overflowed at pool level, and a result so easily ascertainable and so certain to follow is to be regarded as intended. When the Government thus with presumed intention takes the lands of another, the Constitution requiring compensation, the implication of a promise to pay therefor naturally arises. In the Lynah case, 188 U. S., 445, the court, reviewing the authorities, said: “The rule deducible from these cases is that when the Government appropriates property which it does not claim as its own it does so under an implied contract that it will pay the value of the property it so appropriates.”
In the last-named case, in which it was found that there was practically a total destruction of plaintiff’s plantation, and in which it was held that there was a taking, there was a strong dissenting opinion by the present Chief Justice, concurred in by two Associate Justices, which, aside from holding that there was no taking under any construction of the findings, called attention to the fact that the majority of the court had evidently construed the findings as showing that by the construction of the Government’s works the water had been so backed up in the river as to overflow the embankment protecting plaintiff’s lands and thus flood the *136same. There is at best room for the inference that but for this construction of the findings the conclusion of the majority of the court might have been otherwise, since it is nowhere indicated that this assumed fact was not material to the conclusion reached.
In the Heyward case, 52 C. Cls., 87, involving a plantation adjoining the Lynah plantation and substantially the same facts, this court allowed a recovery, expressly basing its conclusion on the authority of the Lynah case, the Chief Justice stating in a concurring opinion that but for that case his conclusion would have been different. This case, on appeal, was affirmed by an evenly divided court. While the Lynah case, of course, still stands as an authority holding that under the facts shown there was a taking, the matters referred to may be proper for consideration as affecting the weight of the authority.
But it is not deemed necessary to review in detail these cases or a number of other well-known cases in which it has been held that there was a taking, some of which have been cited by plaintiff, for it seems to us beyond question, under the facts found, without now referring to them specifically, that the cases before us do not come within any of these cases. Some of them we have considered at length in the opinion of this court and in the concurring opinion of the Chief Justice in County Court of Marion County, W. Va., v. United States, 53 C. Cls., 120. We must conclude from plaintiffs’ briefs that their chief reliance is on the Cress case, 243 U. S., 316, and, since there is, as it appears to us, a tendency, apparent not only in these but in other cases, to misconstrue the facts upon which the holding in the Cress case was predicated, we advert to them briefly.
The facts found by the trial court do not appear in the opinion, and it seems not to have been regarded as necessary to state them in great detail. It is shown, however, that the plaintiff in that case was the owner of land on Whiteoak Creek, a tributary of the Cumberland Liver, and that the United States erected Lock and Dam No. 21 in the Cumberland Liver and thereby subjected 6.6 acres of plaintiffs’ land to frequent overflows of water from the river and damaged it to the extent of one-half its value. The dam in ques*137tion was, of course, below tire confluence of Whiteoak Creek with the river. Its purpose when in operatiop must have been, as in all such cases, to back up the waters of the river and form a “ pool,” and the further result was necessarily to back up the water in confluent stream within the limits of the pool. In the opinion it is said that the land in question is subject to frequent overflows of water from the river, and further, that the findings “render it plain that this is not a case of temporary flooding or of consequential injury, but a permanent condition, resulting from the erection of the lock and dam, by which the land is subject to frequent overflows of water from the river" and referring to the principle that “ overflowing lands by permanent backwater is a direct invasion amounting to a taking,” citing the Pumpelly and Lynah cases, it is held that “ it is the character of the invasion, not the amount of damage resulting from it, so long as the damage is substantial, that determines the question whether it is a taking.”
It thus seems clear that the court was considering and deciding as to the effect of backwater caused by the dam in the performance of its usual and intended functions, for, aside from the forced inference from the language used, there is no reference to flood waters.
Some misapprehension may have arisen as to the facts of the case because it appears that the land in question was not continuously overflowed but was subjected to frequent overflows, and it might be contended that if the overflow was the backwater at pool level it would be continuous. ■ Such would be the case with a fixed dam, provided there was water- enough in the stream above the dam to maintain a pool. But while we have not the facts we venture the •conclusion that the case was dealing with a movable dam.
From the record in other cases presented to this court we know that it was frequently the policy of the Government in erecting dams in streams of the general character of the Cumberland lid ver, to erect, movable dams the wickets of which might lay on the bed of the river when not in use and be raised when it was desired to put the dam in operation and fill the pool. In such a case there would be no backwater when the dam was not in operation and lands within *138pool level would be flooded during such portion of the time as the dam was in operation.
In Chapman v. United States, 53 C. Cls., 203, we had for consideration the effect of just such a dam built across one fork of the Big Sandy Biver and, when in operation, backing the water up a tributary stream. The record showed, and we found as facts the portions of a given year when the dam was and was not in operation, and since it appeared that the dam was in operation and plaintiff’s lands within pool level flooded so much of the time during cropping season that they were useless for agricultural purposes, we gave judgment for the plaintiff. There could be no difference in principle between such a case and one dealing with an overflow caused by the backing up of the water by a permanent dam. In one case as well as the other the results were the natural consequence of the Government’s structure which it must be presumed to have anticipated and intended.
The petitions in these cases aver a taking. A form petition intended no doubt for use in each of the group has attached to each what is termed a “ bill of particulars ” setting up the claim of the particular plaintiff. In No. 32914, Stefano Sanguinetti, the claim is for 40 acres of land at $500 per acre, “ taken by the construction of the Stockton diverting canal,” 20 acres at $1,000 per acre, and residence, barn, outhouses, and other improvements valued at $16,000. In No. 32901, Richard Russell Smith, the claim is for the value of an undivided one-half interest in Oak Grove ranch, containing 1,750 acres, at $200 per acre, $175,000. When the cases were submitted there was attached to the record, in lieu of the original petitions, so-called amended petitions verified May 20, 1919, a date subsequent to the filing of the requests for findings and briefs, which had never in fact been filed in the clerk’s office or noted on the docket or in any proper way made a part of the record. Aside from other changes the bills of particulars set up the claims on an entirely different basis. In the Sanguinetti case the claim is for “ damage to and impairment in value of house,” $7,500, “ damage to and impairment in value of land,” $20,000, and “ to one barn destroyed by floods of 1911,” $800. In addition there is a substitution of heirs as parties. In Smith’s case, No. 32901, *139by this so-called amended petition, Nellie Alice Smith, a sister of Eichard Russell Smith, and the owner of an undivided one-half of the land, is attempted to be joined as a party plaintiff, although any right of action by her was then clearly barred, and the claim is predicated on “ damage done by flood waters of Stockton diverting canal in January to March, 1911, as follows: ” Cost of pumping out lowlands, repairs to levees and ditches, restoring buildings damaged, removing debris and noxious weeds, loss of crops and rent money, and depreciation in value of land. But, of course, the original petitions are the only petitions which are properly in the record.
Of the cases under consideration, one, No. 32913, Silva Sanguinetti, is for the taking of personalty and one, No. 31191, Theodore Infalt, is not for consideration on its merits for the reasons stated in Finding XY. The other two, both for the taking of real estate, are illustrations of different conditions prevailing in the area affected, one, the case of Stefano Sanguinetti, involving real estate lying along the diverting canal and in the acute angled junction between the canal and the Calaveras River, and the other, the case of Smith, involving real estate lying some distance to the north of the Calaveras River and not near the canal, which was confluent with the Calaveras on its south side, and east of the Smith land.
The primary purpose of the canal was, of course, to divert the water flowing down Mormon Slough away from the city of Stockton and the lower reaches of the canal in that city, where it formed a part of the city’s commercial channel, and empty it into the Calaveras River whence it was to find its outlet. Incidentally there was another condition which it must meet. Eastward and northeastward from the, canal and between the Mormon Slough and the Calaveras River was an area miles wide at the canal, widening a few miles northeastward to 6 or 7 miles, and thence narrowing to a point some 12 or 15 miles northeastward from the canal, with a light descending grade toward the canal. This area during heavy rains carried large quantities of water, some falling on the area itself, but in larger part due to overflows from the slough and the river along the upper portions of *140this area, the water flowing for the most part oyer the lower portions of the area in depressions or swales, but at times flowing over other portions of the land and before the construction of the canal temporarily flooding much of this area during heavy rainfall. It is apparent, therefore, that the canal, beside carrying the waters of Mormon Slough to the Calaveras, must also serve to carry the water flowing over this area to the river.
One of the contentions of plaintiff’s counsel, appropriate for reference in this connection, is that it must have been foreseen that “ the waters flowing westward during the normal flood season would considerably exceed the capacity of the diverting canal” and why, he asks, if it had not been so foreseen, should the plan include a levee on the south bank of the canal and “ where else would it throw the waters except upon the lands of Sanguinetti and the others oh the other side of the canal ? ”
It is shown and found that the flood of 1911 was not “ normal,” but was unprecedented, unless tradition was correct as to the flood of 1862, 49 years before, at a time when no part of its effects could be charged to the canal, and two preceding high waters in January, 1911, had been cared for by this canal without damage, so far as appears. But aside from that, can the construction of the levee on the southwest bank of the canal with no levee on the northeast bank be construed as indicating a foreseen lack of capacity in the canal and an intention to overflow the Sanguinetti and other lands ?
The construction of the levee was a part of the general plan, but why? The canal was approximately miles long, 150 feet wide on the bottom, with necessarily sloping banks, and was estimated to require in its construction the excavation of considerably over a million cubic yards of material. It was necessary, in the first place, that there be an accessible dumping ground for this material. To have removed it from the locality would have so aggravated the cost as to probably make it prohibitive. The right of way had been secured of such additional width over the excavation as to permit of its deposit on one of the banks. The effect of its deposit, or a part of it, in a levee on the *141northeast bank on the flow of water over the area above described is so apparent as to need no mention. And aside from all this the conditions to be met certainly indicated that a strengthening of the southwest bank of the canal was good construction. A glance at the map referred to , in the findings will show that Mormon Slough, where intersected by the canal, ran nearly but slightly south of west. The canal left it at an obtuse angle and ran thence northwest. Its determined direction indicates the obviating of any sharp change of direction at either end. But there was necessarily a change the result of which would inevitably caxise waters rapidly flowing from the slough into and through the canal to impinge on the southwest bank. This effect would be aggravated rather than otherwise by the force of the waters flowing into the canal from the area between it and the river. The banks of tl^e canal had a slope of one and one-half on one. The top soil was a light adobe 1 to 1-J feet in thickness superimposed on a red sandy clay without rock therein. It should need no argument to show the necessity for the strengthening of this bank even when the volume of water in the canal was yet within its capacity. Under such conditions it may readily be assumed that erosion of the southwest bank would soon permit gradually increasing escape of water over the lands to the southwest toward Stockton even when the water was below the top of the bank on the other side. But aside from these facts the record shows and we have found that the engineers who made the preliminary investigation and recommended the plan concluded, on the basis of such information as was available, that the canal as recommended would have a greater carrying capacity than would be required for the expected volume of water. And. in fact, there is nothing in the record showing any conditions during years preceding 1911 which would indicate that their conclusions were wrong. Abnormal conditions following, which they did not anticipate,' caused the trouble.
It seems to us quite clear that, independent of the character of the invasion there was no taking of any of the Sanguinetti land, and add to that the necessary conclusion as to the; character of the invasion, together with the fact *142always to be borne in mind, that the land actually devoted to canal purposes was purchased and not taken, and the idea of a taking under the Constitution with an implied promise to pay seems to be absolutely precluded.
What of the Sanguinnetti land was “taken” by the United States when it constructed the diverting canal as demonstrated by its operation during the flood of 1911? It was covered with water for two or three days but the water subsided at once, except as to low spots where there was no outlet and where presumably surface waters collected and stood until absorbed by the earth, and the land remained to the use of the owner as’theretofore. To what extent it had been flooded before we do not know. Water did to some extent at least flow over it, and the construction of the house with its main floor elevated 6 feet might in the absence of other explanation arouse at least a suspicion that even before the construction of the canal surrounding waters were anticipated. Debris was deposited which required labor to remove. There were small deposits of sand and gravel, but there were also deposits of valuable silt. If ditches were thus filled up it was again but a matter of labor to restore them. Some trees were washed out and destroyed by the waters. They could be replaced by the original process, and indeed many trees were set out on this land after this and following floods. All of these things were a damage to the owner to the extent of the labor and cash expenditures necessary to restore former conditions. And so as to the very old barn, not of sufficient stability to withstand such conditions. The house alleged to have been taken remained as it was before, except as to the front veranda steps, occupied during the flood, never invaded by the water, except as to the cellar or basement, and occupied thereafter. There was as to it, of course, inconvenience of access during the flood, and if those conditions were apt to repeat themselves there was probably depreciation in value because of inconvenience of access. But depreciation in value by interference with access does not constitute a taking. Gibson v. United States, 166 U. S., 269.
And so it may be admitted that if there was apparent danger of a recurrence of serious floods there was a conse*143quent depreciation in value of the land not because it was not there to cultivate just the same but because a recurrence of floods meant necessity for the expenditure of labor and money for purposes indicated above. When depositions were taken in these cases and plots of these lands introduced in evidence it was shown that less of these lands were uncultivated than before the flood of 1911. They were not so productive but it is'not shown that they might not have been. Large claims by all these landowners were pending against the United States and it is shown and found that lack of good care, particularly of the fruit trees, was apparent. But however these conditions in the abstract may present themselves, even if, without regard to other necessary elements, results ordinarily to be regarded as damages, pure and simple, might be construed to be a taking of a part of the land because such damages resulted in a depreciation in value, a result always following a damage, at least until it is .repaired, there must be further foundation for an action as for a taking under the Constitution.
We may readily assume it to be settled that the basis of such an action in this court is the implied contract arising out of the appropriation of private property by the United States and the Constitutional provision that it shall not be taken without just compensation. In this connection it may be observed that in considering questions of a “taking” under the Constitution the Supreme Court generally resorts to the use of the word “ appropriate.” While the two words are in a sense synonymous, the use of the word appropriate in relation to governmental acts seems to carry some peculiar significance. . It conveys, as the other does not, an idea of the exercise of a sovereign right to appropriate to its own use some specific thing which it needs for its own proper purposes and does not seem to embrace the idea of a fractional depreciation in value by an inflicted damage to something not in fact appropriated. In the Lynah case the governmental right in question is distinguished from a proprietary right and is referred to as “ its governmental right to appropriate the property of individuals” and is predicated on the right of eminent domain and the principle *144that all private property is held subject to the necessities of government.
In a concurring opinion in the Lynah case Mr. Justice Brown, stating that he saw no reason for holding that there was an implied contract within the meaning of the Tucker Act, expressed the opinion that, irrespective of the question of contract or tort, jurisdiction could be supported under that clause of the Tucker Act which vests this court with jurisdiction of “ all claims founded upon the Constitution of the United States or any law of Congress,” and that claims founded on the Constitution might be prosecuted in this court “ whether sounding in contract or in tort,” but such a theory has never been approved by the Supreme Court as the rule of jurisdiction in this court in this class of cases,
Quoting from many authorities cited in “Words and Phrases,” “ Implied contracts are such as reason and justice dictate and which the law presumes from the relation and circumstances of the parties” and “An implied contract is one which the law infers from the facts and circumstances of the case,” and “Neither an express contract nor one by implication can come into existence unless the parties sustain contract relations, and the difference between the two forms consists in the mode of substantiation and not in the nature of the thing itself,” and “ To constitute either one there must be that conviction, mutuality of will, and interaction of parties generally expressed, though not very clearly, by the term ‘ privity.’ Without this a contract by implication is quite impossible.” And if contract liability is to result from taking of another’s property by the United States is it not essential that there must have been an intention to take? The intention, of course, need not be expressed. It may also be a matter of implication. But it must be fairly inferable from all the circumstances. The inference may be justified when the taking by an overflow is the natural, known, or easily to be ascertained result of a governmental enterprise. As in the case of a dam across a stream erected to create a pool above, the land, if any, which will be overflowed thereby is easily and accurately to be ascertained and before, by modern engineering methods, as easily as after the erection *145of the dam, and such a certain, known, or easily ascertained result must be presumed to have been intended. But eliminate such conditions and substitute an entirely unanticipated result of an authorized Government work, a result not susceptible of advance ascertainment and perhaps due also to abnormal and unanticipated conditions, and there is no room for an implication of intention. And if the implied contract must arise out of the intention, express or implied, to take coupled with constitutional obligation to pay, it must fail for want of an essential element.
In the rather recent case of Temple v. United States, 248 U. S., 121, the United States, in aid of navigation, had done dredging in the North'Branch of the Chicago Biver in what was supposed to be the natural bed of the river or at least by dedication or in some other manner a part thereof, but where in fact the stream had been widened by the plaintiff’s lessee, for his own purposes, by the dredging out of a part of plaintiff’s land.and submerging it to a considerable depth. The Government in prosecuting its work had no knowledge that it was dredging plaintiff’s land. Plaintiff first demanded possession of that part of the submerged land which had formerly constituted a part of his upland, and this demand being refused he instituted a suit to recover the value of that which he claimed had been taken by the Government. Here there was a direct invasion of the land in question and the intention was, of course, to do the very physical thing that was done, but the court said, “ If the plaintiff can recover it must be upon an implied contract ” and, holding it unnecessary to determine whether or not the Government’s claim of a property right in it was well founded, it said:
“The mere fact that the Government then claimed and now claims title in itself and that it denies title in the plaintiff prevents the court from assuming jurisdiction of the controversy. The law can not imply a promise by the Government to pay for a right over, or interest in, land, which right or interest the Government claimed and claims it possessed before it utilized the same. If the Government claim is unfounded, a property right of plaintiff was violated ; but the cause of action therefor, if any, is one sound*146ing in tort; and for such the Tucker Act affords no remedy,” citing Hill v. United States, 149 U. S., 5493.
And in concluding the opinion the learned Justice says,' “The facts preclude implying a promise to pay. If the Government is wrong in its contention, it has committed a tort.”
We are, of course, mindful of the fact that there is no analogy between this case and the instant cases, because here we have no question of title or of full knowledge as to title, but the underlying principle of the Temple case is applicable hero and in all cases of this character. We understand the court to have meant in that case that even though it were established that the property in question was the plaintiff’s property, and even though there was actual invasion thereof, there was no taking for which compensation could be adjudged and no jurisdiction to determine the controversy, because the Government, believing in its own right, could not have intended to take the property of another, there was therefore no proper basis for the implication of a contract and, equivalent to the same thing, the facts, it is said, precluded implying a promise to pay.
Applying the facts found in these cases, particularly to the Sanguinetti lands, perhaps the strongest case of the whole group and for that reason selected as one of those for trial, we find no basis for any implication of an intention to take, even in the minutest degree, or for any implied contract with reference thereto. Facts found as to intention are specifically to the contrary. There was no purpose by the construction of the dam to back up the waters of a stream. The dam built across Mormon Slough was simply to divert the water into a new channel deemed sufficient for the purpose. The Government sent its agents, through whom it must act, to investigate and recommend and, based on such data as was available to them, they recommended a canal of dimensions and capacity which they deemed more than sufficient. And but for abnormal, unprecedented floods, their judgment would probably have been vindicated. While the findings refer to the insufficiency in capacity of the canal as an engineering mistake it was not such a mistake as should reflect on them, for in fact they had no warning of the extreme and abnor*147mal conditions to be met. And if any consideration is to be given to the theory that they planned and recommended a canal of known insufficient capacity and by its construction with a levee on one bank thereof they contemplated and intended the overflowing of the Sanguinetti land, may we not wonder why they did not further conserve Government money and energy by building it one-half the width and accomplishing the same purpose? And when they planned the confluence with the river they surely did not intend that the force and volume of the two streams coming together and not finding a sufficiently rapid outlet should back the water onto the Sanguinetti land, for, had such been the intention, they would scarcely have recommended the dredging of the channel of the river below the mouth of the canal to increase its capacity to carry off the combined volume of water.
As to this feature of the situation, the backing up of the water on-the Sanguinetti land rather than its overflow from the canal irrespective of its backing up, there was a contributing cause, for which the Government was not responsible, in the bridge or trestle of the Southern Pacific Railroad which crossed the Calaveras River about 500 feet below the mouth of the canal. We can not determine, can only speculate as to how much of resulting conditions were due to this obstruction. Possibly, but for it, the whole volume of water would have found a ready outlet. The findings are that it is not shown directly or inferentially that there was any intention to flood the lands in question or that there was any reason why the engineers in charge should expect or anticipate that such a result would follow. They might have gone further, if an inference in.positive form was justified as a finding, and said that the facts and circumstances showed an intention to care for all waters within their intended channels and a well-founded belief that such a result would follow. So far as appears from the record it might reasonably be inferred, if inference as to such a matter was proper or material, that if this canal had been built 40 years before it was it would for all that period have successfully cared for all the water it was required to carry, provided, of course, it had not been permitted to fill up. It happened *148that this extraordinary flood came the first year after its completion. Subsequent floods, following three years after, approximating but not equaling that of 1911, were aggravated by the fact that the canal had been permitted to fill up by deposits therein and its capacity was thus much reduced. In the case of Cubbins v. Mississippi River Commission,, 241 U. S., 351, the rule of limitation as to accidental and extraordinary overflows is discussed and is referred to without repetition. It is in attempted avoidance of the rule of this case that plaintiff presents the theory, already sufficiently discussed, that the construction of the levee on the southwest bank of the canal was in contemplation of the foreseen insufficient capacity of the canal and for the intended purpose of throwing the water on the Sanguinetti lands.
Plaintiff, in the Sanguinetti case, also discusses the Grizzard case, 219 U. S., 180, using it in support of a theory as to “ destruction as a mode of taking ” and, if we understand the contention, it is in part that the fact that the right of way for the canal had been bought and paid for is immaterial. In the Grizzard case a strip of plaintiff’s land lying on Tates Creek, a tributary of the Kentucky Eiver, had been submerged and its use for agricultural purposes destroyed by the erection of a dam across the Kentucky Eiver in aid of navigation. There was’ no question made as to the right to compensation for the submerged land. The question vras as to compensation also for the destruction by the same flooding of an “ easement of access ” by a jrablic road, running over the submerged land to the Tates Creek pike. Distinguishing authorities “touching an injury to land not taken by the construction of a railroad along and upon a public road,” etc., it is said:
“ Since, therefore, there has been a taking of a part of the owner’s single tract' and damage has resulted to the owner’s remaining interest by reason of the relation between the taken part and that untaken, or by reason of the use of the taken land, the rule applied in the cited cases does not control this case,” and
“Whenever there has been an actual physical taking of a part of a distinct tract of land the compensation to be *149awarded includes not only the market value of that part of the tract appropriated, but the damage to the remainder resulting from that taking, embracing, of course, injury due to the use to which the part appropriated is to be devoted.”
Supporting conclusions reached the court cites Sharp v. United States, 191 U. S., 341, in which, however, it was held that there could be no assessment of damages as to a separate tract no part of which was taken, and Bauman v. Ross, 167 U. S., 548, from which quotation is made holding that .“the incidental injury or benefit to the part” (of a parcel of land) “not taken is also to be considered.” The infirmity in the Grizzard case, if there is any, is in the fact that it is largely rested on the two cases referred to and quoted from which were both proceedings in condemnation. In condemnation the usual rule is no doubt as stated, and we do not question holdings by which we are bound as to its application in the class of cases under consideration, but it is to be observed that if the rule of Bauman v. Boss is the rule of this class of cases an assessment of ~benef,ts to the remainder of the tract as well as damages is required. But, however that may be, it is apparent that a recovery of damages to the part of a tract not taken is allowed only because of its relation to the part actually taken and because the result of it is in effect a part of that actual taking. There is no analogy in any of the cases under consideration. The lands involved in the Stefano Sanguinetti case are parts of tracts out of which the canal right of way was taken, but that right of way was bought, conveyed by deed, and paid for. In the Grizzard case and others like it there was an unquestioned appropriation, “ an actual physical invasion ” by overflow of a part of the tract.
The fact of a conveyance suggests another question, not necessary to a determination of the case but proper for consideration. The findings of fact show the procedure in the acquisition of the right of way for the canal. Condemnation was pending against Stefano Sanguinetti when, at some stage of the proceedings not shown, he executed the deed set out in Finding XI, in which the purposes of the grant are fully recited. If the proceedings in condemnation *150had proceeded to final decree an assessment would have been required, first, of the value of the land taken, and, second, of the incidental damage to the remainder of the tract or tracts by reason of the taking, including the damage resulting from the use to which the part taken was to be put and including all damages reasonably to be anticipated as a result of the taking and the use. Plaintiff’s theory is that an insufficient canal capacity in times of extreme high water was foreseen and that a part of the project and a necessary result, evidenced particularly by the construction of the levee on the southwest side, was to cast flood waters on the San--guinetti lands. While we do not agree with this contention, it must necessarily result that if the theory is correct and if the proceedings in condemnation had proceeded to a finality, such result and damages 'to the parts of the tracts not taken would have been included, or would be deemed to have been included, in the condemnation award. A very common exercise of the right of eminent domain by private corporations is the condemnation of rights of way by railroad companies. The juries usually called to assess the damages are always instructed as to the elements of damage which they may assess as inuring to the remainder .of the tract by reason of the taking and also by reason of the use for railroad purposes of the part taken. Lawsuits against the railroad company by the landowner frequently follow. Sometimes the question arises whether damage- complained of is a damage which must be deemed to have been included in the condemnation award, but, that question eliminated, such actions declaring on damages resulting from the construction or operation of the railroad are always actions in tort. We are of the opinion that the deed executed by Sanguinetti pending condemnation proceedings, and so fully expressing the purposes thereof, put him in the same status as if the right of way conveyed had in fact been the subject of condemnation.
But we need not predicate our conclusion on this point, for to us it seems quite clear that this action (Stefano San-guinetti, No. 32914) can not be made to rest on an implied contract to pay for any part of or interest in the land in *151question as for a taking, but is an action in tort, of which we have no jurisdiction.
The action of Silva Sanguinetti, No. 32913, is for the alleged taking of growing crops and personal property on the lands involved in the other Sanguinetti case, and the conclusion reached in that case must apply also in this. If this were not the conclusion we might have for consideration the status, in such a case, of growing crops, the fact that he, a tenant, sues for the whole value of the crops alleged to have been destroyed, although his tenancy was on the basis of a crop rental, proportions not shown, etc. But the general conclusion, as stated in the case of Stefano Sanguinetti, is sufficient for the purpose and renders other questions of no moment.
As to the case of Smith, No. 32901, it is not necessary to enter into much of detail except to supplement the applicable parts of the discussion in the Stefano Sanguinetti case and differentiate in some respects. The Smith lands in their relation to the canal were radically different. They did not touch the canal, but laid on the north side of the Calaveras River, about a mile and a half north thereof at its nearest point and about two, and a half miles, on the opposite side of the river, from the canal. They were lowlands, naturally subject to tidal overflows, were overflowed when acquired by the Smiths, had been reclaimed by reclamation levees, draining, and pumping, and required pumping each year before crops were put in. Further facts as to conditions and what happened during the flood of 1911 are set out in the findings and need not be repeated.
Plaintiff’s theory in the Sanguinetti case, and the same counsel appear in both cases, as to the alleged foreseen incapacity of the canal and the intent manifested by the construction of the levee to cast the excess waters on the San-guinetti lands, can have no application, even if conceded to be correct, to the Smith case. The case, it is said, is concerned alone with the flood of 1911, and is predicated on the alleged fact that “the waters of Mormon Slough conveyed through the diverting canal caused the levees on the north side of the Calaveras to break.” At least that is the only statement we find in plaintiff’s brief as to the cause of the *152result complained of and perhaps the only statement that could be made. This break occurred on a part of the old levee slightly above and opposite the mouth of the canal and above the portion of the levee which had been strengthened by deposits of materials dredged from the river in the work of increasing its channel capacity from the mouth of the canal down toward the San Joaquin.
To be accurate, it is not found that the waters carried by the canal broke the levee, but facts are found as to the backing up of the waters in the Calaveras by the flow of water from the canal, contributed to by the railroad bridge below, and it is found that as a result of the backing up and embanking of the waters against the levee there was a break. But just how the facts as to the break are stated is probably of slight importance.
Plaintiff then contends that there are but two points in’ the case; first, the extent of the Smith damage, and, second, the items on which claimants are entitled to recover, following with a discussion of the evidence as to the loss of a barley crop, hay crop, interest in a crop of onions, lost cash rentals, restoring damaged buildings, repairing ditches and levees, pumping out the water, clearing up debris, and depreciation in the value of the land “ because of the constant menace of the canal,” and ooncludes that there was “ (1) a physical taking of claimant’s crops and (entire or partial) of his buildings, and (2) a like physical taking to an extent of the naked lands which were actually overflowed, and (3) a taking in the form of diminshed value of lands not overflowed.” The basis of the claim thus itemized is evidently the bill of particulars attached to the unauthorized amended petition referred to in the earlier part of this opinion and the claim of Nellie Alice Smith, the owner of an undivided one-half interest in the land, is evidently included, although she was never properly made a party to the action, but so far as the form of the claim of Bichard Bussell Smith is concerned we can regard it as immaterial to the determination of the merits of the case.
Plaintiff’s brief cites to us the Cress and the Grizzard cases, both already referred to, and the Welch case, 217 U. S., 333, *153and distinguishes High Bridge Lumber Co. v. United States, 40 Fed., 738.
The Cress case is cited in support of the proposition that the question is “how much less are claimant’s lands as a whole worth by reason of the canal and the ‘permanent liability’ of parts or all of the same to overflow from that cause.” In fact, we have'not found that there was any such depreciation, but have found that they were sold for a reasonable price, and that it is not satisfactorily shown that they could have been sold for any greater price but for this flood and any “ impending menace ” of floods. But, aside from these findings as to comparative values, we can find in the facts of this case no room for the application of the Cress case except it be, as here, the use of words disconnected from the context and the facts to which they apply. And the Welch case is cited in support of the proposition that—
“ It should now be taken as settled that physical contact of any parts of the lands with the flood waters which have occurred in these past years, and which may be apprehended to occur in future years, is not a condition precedent to an award of compensation respecting that acreage; it is sufficient that the estate at large to which those lands belonged has been injuriously affected in some of its parts or appurtenances.”
We quote the proposition because it is due the plaintiff that we should fully weigh the arguments and authorities submitted, but the writer must admit his inability either to grasp it or apply it. The Welch case involved the flooding of a part of plaintiff’s land by the operation of a dam built across the Kentucky River, the controverted question being as to a recovery also in that connection for the taking or cutting off of access to a right of way by the flooding of the intervening strip. We are unable to apply the case to the one under discussion. We have already referred to the holding in the Grizzard case, for which we find no application. To be entirely frank about the matter, it must be said that plaintiff has failed to point us to any authorities sustaining his view of this case.
The principles hereinbefore discussed as deducible from the well-known authorities cited seem to exclude the idea that there can be any recovery in this case as for a taking. *154It is going too far to even discuss the idea that the Govern-ent when it constructed this canal had or by the remotest possibility could be imagined to have had any intention even remotely to be implied as to the flooding of the Smith lands, and there seems to us no possibility of erecting an implied contract out of the circumstances of the case. If the United States was in manner stated responsible for damage to those lands, it was a tort of which we have no jurisdiction.
The Infalt case, No. 31191, is one of the four originally selected for preparation and trial as illustrative of the group, and was therefore set and taken under submission with the cases already discussed. It might have been dismissed before submission for the reasons stated in Finding XV if the attention of the court had been called thereto and for want of prosecution. Such will now be the order with reference to it. If we are right in'the conclusions reached in the other cases, it would serve no good purpose to give it a separate consideration.
No. 32914, Stefano Sanguinetti; No. 32913, Silva Sangui-netti; No. 32901, Richard Russell Smith: These cases are not within the'jurisdiction of the court, and should be and are dismissed.
Graham, Judge; Hay, Judge; Booth, Judge; and Campbell, Chief Justice, concur.