This is a proceeding to review two Federal Power Commission orders granting a license for the multipurpose New Don Pedro project on the Tuolumne River in California. The petitioners are the Turlock and Modesto Irrigation Districts, the Secretary of the Interior (Secretary), and the State of California (state). The districts, which had jointly applied for the license, complain of terms and conditions attached to the license imposed for the protection of salmon runs. The Secretary and the state, on the other hand, assert that those terms and conditions are not sufficiently far-reaching.

The nonnavigable Tuolumne River rises in the Sierra Nevadas above the Hetch Hetchy Valley. By the time it reaches the San Joaquin Valley floor at La Grange, the river drains an area of about 1540 square miles. From La Grange it runs for some fifty miles to its confluence with the San Joaquin River, with the Turlock district on one side and the Modesto district on the other. The river has extreme variations in annual flow, ranging from 3,204,300 acre feet in 1938 to 600,300 in 1924. A high flow is usual in March to June, followed by very little run-off.

The two districts were formed in 1887. At that time they had a combined acreage *920 of 257,000 acres, roughly two thirds in the Turlock district and one third in the Modesto. Rainfall in this area averages only twelve inches annually, ninety percent of which is in the winter months. The storage and diversion of surface water is therefore essential to successful agriculture. The Tuolumne is the only source of surface water available to the districts.

The districts initially purchased prior appropriative rights, dating back to 1855, to the use of Tuolumne River waters. They supplemented these rights by direct appropriations filed under California law in 1889 and 1890. The districts constructed the La Grange diversion dam in 1894, the off-stream Modesto Reservoir in 1911, and the off-stream Turlock Reservoir in 1914.

Early in the twentieth century, the City and County of San Francisco sought rights of way from the United States in Yosemite National Park so that it could establish storage in the headwaters of the Tuolumne, to support the diversion of water for domestic use in the San Francisco Bay area. The result was the enactment of the Raker Act, 38 Stat. 242 (1913).

In the Raker Act, Congress granted San Francisco the rights of way it needed. This grant, however, was made subject to the observance, by the grantee, of certain enumerated conditions. One of these is that the grantee shall recognize the specified prior right of the Turlock and Modesto irrigation districts. This was the right to receive 2,350 second feet of the natural flow of the Tuolumne measured at the La Grange Dam, “ * * * whenever the same can be beneficially used * * * ” by the districts for the irrigation of three hundred thousand acres of land. Section 9, 38 Stat. at page 246. During a sixty-day period beginning on April 15 of each year, San Francisco is, under this statute, required to recognize the right of the districts to the extent of four thousand second feet.

San Francisco initiated its activity in the upper Tuolumne watershed in the 1901 to 1911 period, posting notices of appropriations of water under California law and conducting surveys. The city intensified this activity following passage of the Raker Act, a system being planned to export ultimately four hundred million gallons a day (mgd) to the Bay area. In 1918, the city completed Lake Eleanor Reservoir, having a capacity of 27,100 acre feet. In 1923 it completed O’Shaughnessy Dam, impounding 206,-000 acre feet in Hetch Hetchy Reservoir.

In the same year the districts constructed the original Don Pedro project, impounding about 290,000 acre feet at a site five miles upstream from La Grange Dam. O’Shaughnessy Dam was altered in 1935 to 1938 and 1949 to a capacity of 360,360 acre feet. In 1956 San Francisco completed Cherry Valley Dam, impounding 268,200 acre feet in Lake Lloyd.

In 1951, the districts made application to the state for water right permits covering storage and diversion for irrigation use necessary to the construction of the New Don Pedro project. The New Don Pedro Dam would be downstream of Don Pedro Dam, and would flood out the latter structure.

It would have a maximum height of 580 feet and would create a reservoir extending twenty-two miles up the Tuolumne having a gross storage capacity of 2,030,000 acre feet. Of this, 310,000 acre feet would be for a minimum pool and 340,000 acre feet would be dedicated to flood control. In addition, 570,000 acre feet would be available as exchange storage in which San Francisco could store water to satisfy the Raker Act entitlements of the districts. This would make it possible for San Francisco to make the necessary diversions to its upstream aqueduct to maintain a dependable supply in its Bay area system. The remaining storage capacity would be utilized by the districts in serving their needs.

The project involves a power plant with a peaking capability of 150,000 kilowatts, such production to be subservient to the operation of the project for the other purposes described above. In addition to the purposes already mentioned, *921 the districts contemplated a recreational development of the reservoir in coordination with interested federal, state and county agencies. It was also proposed that the project be operated so as to provide in the fall and winter months as much water as possible for salmon in the stream below the La Grange diversion dam.

The total cost of the project is estimated at $92,415,850. San Francisco has entered into an agreement with the districts to contribute to the cost of the project an amount to be based on the cost of constructing a dam and reservoir with a storage capacity of 1,200,000 acre feet, exclusive of cost of power facilities and lands. San Francisco’s contribution is estimated at $43,181,790. The districts propose to finance their share of the cost, estimated at $49,334,060, by the sale of bonds.

Lands of the United States in the public domain would be used in the construction and operation of the project. Therefore the districts needed to obtain from the Federal Power Commission a license to construct and operate the project, notwithstanding the fact that the Tuolumne is not navigable. See section 4(e) of the Federal Power Act, 49 Stat. 839 (1935), 16 U.S.C. § 797 (e) (1958); F.P.C. v. State of Oregon, 349 U.S. 435, 442-443 , 75 S.Ct. 832 , 99 L.Ed. 1215 .

The districts filed a joint application for such a license. A proceeding before the Commission was thereupon instituted, and the state immediately intervened in that proceeding. The Secretary and others sought to intervene at a later stage in the proceeding. The latter were granted limited intervention for the purpose of making an opening statement, cross examining witnesses, and filing briefs and exceptions.

After proceedings before an examiner, the latter issued an initial decision. The districts, state, Secretary and the Commission’s staff filed exceptions. After oral argument, the Commission, on March 10, 1964, filed its Opinion No. 420 and order, issuing a fifty-year license to the districts subject to certain terms and conditions. The districts, state and Secretary applied for a rehearing. On May 6, 1964, the Commission entered an order modifying the earlier order, and denying the application for rehearing. This review proceeding was then commenced.

The questions presented here relate, for the most part, to the inclusion, among the terms and conditions of the New Don Pedro project license, of Article 37, pertaining to the release of waters to maintain fish runs below the New Don Pedro Dam. This article provides that, for the first twenty years of the project operation, the districts shall maintain minimum stream.flows in the Tuolumne River at La Grange Bridge (river mile 50.5) for fish run purposes in accordance with a table set forth in the article.

The table lists the number of cubic feet per second (cfs) and the number of acre feet to be released during specified periods of each year, depending upon whether the immediately preceding year is a “normal” year (schedule A) or a “dry” year (schedule B). The total acre feet to be released in a year immediately following a “normal” year is 123,210 acre feet, and immediately following a “dry” year, 64,040 acre feet. A “normal” year is defined as any year ending on October 1 in which the inflow into New Don Pedro Reservoir equals or exceeds one million acre feet. A “dry” year is defined, in like manner, as one in which such inflow is less than one million feet. Schedule B, applicable immediately after a “dry” year, is also made subject to the proviso quoted in the margin. 1

*922 Article 87 also contains the following provision stating what is to happen at the end of the twenty-year period:

The districts contend that Article 37 impairs the irrigation uses of the districts covered by water rights acquired under California law, and that the Commission is without authority to impair those rights. Counsel for the Commis;sion does not, on this review, take serious issue with the districts’ contention that .Article 37 could operate to impair full use of their irrigation water rights. Instead, counsel defends the Commission’s authority, under the circumstances, to attach a condition to the New Don Pedro project which may have that result.

In its opinion and order of March 10, 1964, the Commission declined to pass upon the legal question of the Commission’s authority in this regard because, in its view, the fish releases required under Article 37 are not “ * * * expected to interfere with such rights [the districts’ water rights] during the first 20 years of operation, * * Elsewhere in that opinion and order, and in its order of May 6,1964, the Commission seems to state, rather categorically, that the fish release requirements of Article 37 will not limit the districts’ planned use of its irrigation water rights for at least twenty years.

But in other of its statements in these two orders the Commission discloses that its expectations in this regard might not be realized. The Commission indicates that its judgment in the matter is based in part upon the assumption that the past hydrologic cycle is a generally representative pattern for the future, and that it has taken into consideration the districts’ “average irrigation yield.” The Commission’s last statement on the subject is that, in making fish releases for twenty years as specified in Article 37, the districts “ * * * are giving up considerable value in terms of * * * the unrestricted use of the water for irrigation.”

Still more significant is the Commission’s statement, in its order of May 6, 1964, that the districts’ present irrigation needs are about one million acre feet a year “ * * * and these needs should continue at about 1,000,000 acre feet a year for the foreseeable future. * * * ” Inasmuch as the proviso of Article 37 specifically provides for fish releases, although in reduced amounts, even when the gravity diversion at La Grange by the districts falls below nine hundred thousand acre feet, it would seem that Article 37 specifically contemplates fish releases which may take water the districts will need for irrigation.

The Commission indicates, in its order of May 6, 1964, that if circumstances develop calling for application of the proviso to Article 37, and reduced releases therein called for are not sufficient to meet the districts’ needs:

The Commission thus indicates that it is aware of the great importance to the districts of having water sufficient to meet their irrigation needs, no matter how this may affect the fish release program. But the agency also makes it clear that should such an unanticipated conflict of interest arise, it would deal with the question in the context of the public interest. Presumably this could *923 lead to a result different than would be called for by a strict recognition of the districts’ irrigation water rights unless, when it came to deal with the problem, the Commission should conclude, as a matter of law, that the districts’ rights must prevail.

So analyzing the potentials inherent in Article 37 and the Commission’s announced way of administering it, we are convinced that a determination of the legal question as to the Commission’s authority should have been made by the Commission and must, in any event, be made on this review. As before indicated, both the districts and counsel for the Commission have argued the question on the merits and make no contention that the proceeding should be remanded to the Commission for an initial adjudication on the point.

The districts’ present exercise of their irrigation water rights through utilization of existing facilities is not actually or potentially impaired by the Commission order. But the districts propose to augment their exercise of those rights by the construction and operation of facilities which will include a power project and which will be located, in part, upon the public lands of the United States. It is this circumstance which required the districts to apply to the Commission for a license.

In granting such a license the Commission is required, under section 10(a) of the Federal Power Act, 49 Stat. 842 (1935), as amended 16 U.S.C. § 803 (a) (1962), to consider all beneficial public uses including recreational uses. Under that section, and under section 4 (g) of the Act, 16 U.S.C. § 797 (g) (1958), the Commission is made the “guardian of the public domain.” F. P. C. v. Idaho Power Co., 344 U.S. 17, 23 , 73 S.Ct. 85 , 97 L.Ed. 15 . The issuance of a license can be justified only on the theory of a resulting benefit to the public. Alabama Power Co. v. F. P. C., 75 U.S.App.D.C. 315, 323-324 , 128 F.2d 280, 288-289 .

The districts rely upon section 27 of the Federal Power Act, 41 Stat. 1077 (1920), 16 U.S.C. § 821 (1958), as precluding the Commission from attaching conditions to the license which may impair use of their irrigation water rights, even though such conditions may be appropriate under section 10(a) of that Act. 2 *****8 In Portland General Electric Co. v. F. P. C., 9 Cir., 328 F.2d 165, 176-177 , we held that section 27 did not stand in the way of the Commission in imposing navigation conditions in a license pursuant to section 11 of the Federal Power Act, 41 Stat. 1070 (1920), 16 U.S.C. § 804 (1958).

In our opinion the same principle governs where conditions in a license are imposed pursuant to section 10(a). It is true that in the Portland case the basis of Commission authority is the navigability of the water in question, while here the basis of Commission authority is the fact that public lands will be affected. This might make a difference if the issue presented were whether the consent of California was required with regard to the provisions made for the conservation of anadromous fish. See F. P. C. v. State of Oregon, 349 U.S. 435 , 75 S.Ct. 832 , 99 L.Ed. 1215 . But California makes no such contention here and the districts are as opposed to state restrictions as they are to federal restrictions.

The districts also argue that section 29 of the Federal Power Act, 41 Stat. 1077 (1920), 16 U.S.C. § 823 (1958), providing that nothing contained in that act shall be held or construed to modify or repeal any of the provisions of the Raker Act, precludes the Commission from imposing a license condition which may impair their irrigation water rights.

*924 The districts, however, will continue to receive the flows guaranteed to them under the Raker Act as long as they are content with their present facilities. That act did not give them the right to use the public lands they now wish to utilize in connection with the New Don Pedro project. With regard to those public lands, the districts are in the same position as any other applicant for a license — if they are to use those lands they must accept the reasonable restrictions and obligations attached thereto.

We therefore conclude that the Commission had authority to incorporate in the tendered license a condition which could operate to impair the districts’ full use of their irrigation water rights in some future year. The likelihood that circumstances will occur during the next twenty years which will, in fact, present such a problem seem remote. The prospect that, if such circumstances do arise, the Commission will solve the problem in a manner which will deprive the districts of essential irrigation requirements is speculative. Should such an eventuality come to pass, any such Commission action will be subject to review as to the sufficiency of the evidence to support findings upon which such action is based, and as to whether the action is arbitrary and capricious. But we now hold that the Commission has the legal authority to take appropriate action restricting the use of such irrigation rights, should the occasion arise.

The districts argue that the Commission violated section 6 of the Federal Power Act, 49 Stat. 841 (1935), 16 U.S.C. § 799 (1958), by ordering the issuance of a license which did not state all of its terms and conditions. Section 6 provides, in relevant part, that licenses issued under the Act shall be issued for a period not exceeding fifty years and upon terms and conditions “ * * * expressed in said license.” The section further provides that a license may be altered or surrendered only upon mutual agreement between the licensee and the Commission.

The asserted violation of section 6 is brought about by the provision of Article 37 which requires that, after the first twenty years of operation, the districts shall maintain such minimum stream flow at La Grange Bridge as may be then prescribed by the Commission after further proceedings to be held at that time.

The districts indicate that certainty sufficient to overcome this section 6 objection will be achieved only if the license guarantees them a specific dependable power capacity in kilowatts for the entire fifty-year license period, indicating that this should be not less than 73,000.

In its second opinion and order the Commission made the statement set out in the margin concerning the purpose and scope of any modification of the initial release requirement which might be made at the end of the twenty-year period. 3 In our view Article 37, read in the light of the quoted statement, definitely establishes that whatever curtailment in irrigation and municipal purposes may be found legally and factually supportable at the end of the twenty-year period, no fish release will be required after that date which will impair the continuing economic feasibility of the project.

The section 6 requirement that the terms and conditions of a license be expressed in the license must not be given a construction which is impracticable of application. When the Commission reasonably foresees the possibility that a need may develop years in the fu *925 ture requiring, in the public interest, the imposition of a burden upon the licensee at that time, but either the dimensions of the need or the way of meeting it is not presently ascertainable, the license terms cannot possibly speak with definiteness and precision concerning the matter. Under these circumstances, it is sufficient, under section 6, to include in the license a condition reserving the problem, including the licensees’ rights to test the validity of any future action taken.

The principle applicable here is analogous to that which obtains in determining whether the Commission order must contain specific findings of fact as to future conditions. We dealt with that problem in Portland General Electric Co. v. F. P. C., 9 Cir., 328 F.2d 165, 175 . We there held that it is not arbitrary, unreasonable, or a deprivation of due process to refrain from making findings as to what conditions may be in the future, at least where no present burden is imposed.

In its opinion of March 10, 1964, the Commission indicated that it could not now evaluate with assurance a number of factors which should be considered in determining the kind of fish release program which will be needed during the last thirty years of the license period. 4 We find no basis in the record before us for holding that the Commission has overstated these present uncertainties.

In light of these circumstances the provision of Article 37, giving notice of the prospective reexamination of the fish release problem at that time and outlining procedures to be followed in the interim looking to the development of a proper solution, provide as much precision and definiteness in this regard as is presently practicable. We therefore hold that there has been no violation of section 6 of the Federal Power Act in the respect asserted.

The districts contend that the specification of releases was totally unnecessary in the face of the Commission’s own expectations, as stated in its opinion of March 10, 1964, quoted in the margin. 5

We do not read the quoted statement as meaning that, even without prescribed releases the fish will be assured of sufficient water until 1985. The Commission was discussing the adequacy of water, not the need of imposing release requirements to assure that the available water would be released for fish runs. This is made clear by the Commission’s further statement which immediately follows that quoted in note 5. 6

The districts argue that the Commission’s findings of fact are inadequate for *926 purposes of court review. They urge that the Commission did not say how the releases in the amounts prescribed could be justified, “ * * * it simply summarized in one paragraph the California contentions about salmon requirements (R. 3896) and then stated that its analysis of the record showed California’s studies to be ‘reasonable.’ ” Even then, the districts state, the Commission qualified this by adopting the staff’s “inflow” rather than California’s “natural flow” formula, in language that recognized that California’s claims for salmon were overstated. The Commission makes no response in its brief to this argument about the clarity of the findings.

The rule the districts here invoke is that there must be clarity and completeness in the basic or essential findings on which administrative orders rest. Colorado-Wyoming Gas Co. v. F. P. C., 324 U.S. 626, 634 , 65 S.Ct. 850 , 89 L.Ed. 1235 . Or, as the District of Columbia Court of Appeals recently said in Government of Guam v. Federal Maritime Commission, 117 U.S.App.D.C. 296, 300 , 329 F.2d 251, 255 , “ * * * the reviewing authority is entitled to know — indeed must know — the basic data and the whys and wherefores of the Commission’s conclusions.” 7

The findings of the Commission made in support of its fish releases, as stated in its opinion of March 10, 1964, are entirely sufficient with regard to: (1) the average run of salmon which ought to be protected, (2) the number of spawning salmon which must be provided for each year to protect the average run, (3) the number of square feet of spawning gravel required for each spawning salmon, and therefore the total amount of square footage required.

With regard to the amount of river flow necessary to support spawning called for under these three factors, the table of releases set out in Article 37 represents an adequate and express finding. Nowhere in the report does the Commission state the facts as to how this required river-flow figure was computed. The opinion does, however, indicate that the Commission accepted California’s computations as to this. The only element of the latter computation which the districts appear to question is the low-flow figure for the test period during a survey conducted by the California Department of Fish and Game and the United States Department of the Interior. This low-flow figure is ascertainable from exhibits and testimony referred to in the briefs.

We therefore hold that the Commission findings are adequate for our purposes in reviewing the agency orders. This being true, the adequacy of the findings in other respects is immaterial, especially in view of the fact that, in their petition for rehearing filed with the Commission, the districts did not complain as to the adequacy of the findings. See section 313(b), Federal Power Act, 49 Stat. 860 (1935) as amended, 16 U.S.C. § 825l(b); Utah Power & Light Co. v. F. P. C., 10 Cir., 339 F.2d 436, 438 .

The districts contend that with regard to the annual amount of water releases necessary to accommodate eighty thousand salmon within the area found to be required, the findings of the Commission are not supported by substantial evidence.

This is true, the districts assert, because there inheres in California’s proposed water release schedule, accepted in substance by the Commission, two assumptions of fact which are contrary to the evidence. One of these is the assumption that, during the test period referred to above, the low flow was 160 cubic feet per second, whereas it was actually 130 cubic feet per second. The other is the apparent assumption that there were no substantial accretions to the river which maintain a live stream regardless of releases at La Grange, whereas the fact is to the contrary.

*927 With regard to the first of these points the evidence is sufficiently in dispute so that we are unable to say that the Commission erred in accepting the larger figure. Moreover, the Commission recognized that California’s studies were “conservative” in some respects, and allowed for this by modifying, in the direction of reduced releases, the use to be made of the “normal year” and “dry year” schedules.

Concerning the failure of California to take accretions into account, there is reason to believe that this could not have been an important factor during the October to March period when spawning occurs. Accretions are for the most part the result of application of irrigation water to the lands on both sides of the stream bed and there is little irrigation in progress during the spawning season. The Commission indicated that it was aware that accretions had not been taken into consideration and allowed for it by ordering the modification referred to above.

Having in view the highly technical nature of the fact determinations in question and the judgment factors which were involved, both calling for expertise which the Commission has, and this court does not, we conclude that the findings referred to above are supported by substantial evidence sufficient to withstand attack on this review.

Finally, the districts contend that the evidence does not support the Commission’s finding that enough water should be released during the twenty-year period to accommodate eighty thousand salmon.

The Commission based this determination on its findings that the fish run which ought to be protected averages forty thousand salmon a year, and that to maintain this average the spawning grounds must be large enough to provide for up to eighty thousand adult fish each year, of which forty percent will be female. The districts’ challenge runs to both of these findings.

The determination of the average run to be protected was based upon the best information available as to actual runs for past years. That information is adequate to support a protectible average run of forty thousand if the run of 130,-000 in 1944 and the run of 120,000 in 1940 are included. A protectible average run of about twenty-five thousand fish would be indicated if the years prior to 1950 were excluded. The run was only five hundred in 1961, two hundred fifty in 1962 and one hundred in 1963.

In deciding how many past years should be taken into consideration in determining the protectible average, the Commission was required to make judgment determinations not subject to challenge here unless plainly arbitrary. We find no basis for holding that the Commission was arbitrary in providing for the protection of spawning on a scale which would reverse the downward trend in recent years.

On the other hand if, after a period of five years or so, under the Article 37 release schedule, the actual fish run has dwindled to but a fraction of the forty thousand hypothetical average, the districts ought to have the opportunity to seek a reopening of the fish release terms and conditions. We do not construe the Commission orders as foreclosing this opportunity. As of the present time, however, we conclude that the finding in question is supported by substantial evidence.

The finding that, in order to protect an average spawning run of forty thousand, the spawning grounds must be large enough to provide for up to eighty thousand adult fish, is also supported by substantial evidence.

We turn now to the questions raised on this review by both California and the Secretary although, to some extent, these have already been discussed in dealing with the issues presented by the districts.

The first of these is whether the Commission erred in limiting the release schedule set out in Article 37 to the first twenty years of the fifty-year license period.

*928 At the outset of this argument the Secretary and California call attention to various federal and state statutes manifesting legislative concern with the preservation of fish and wildlife resources. 8 The Secretary also asks us to bear in mind the magnitude of the fishery values involved. 9

These petitioners argue that if the fish resources of the Tuolumne are to be saved, it must be done now by fixing adequate terms and conditions for that purpose which will be effective throughout the fifty-year license period. This is, they assert, the normal procedure, applicants being required ordinarily to demonstrate at the time of the application proceedings that the project will be economically feasible for the entire license period having in view all expenditures necessary to preserve fish resources.

Petitioners cite numerous examples in which investor-owned utilities have been required to make substantial sacrifices, determined in advance, to finance fish facilities, hatcheries and fish ladders. The implication is that the irrigation districts and the City and County of San Francisco ought to receive similar treatment. If this determination is to be postponed for twenty years, they urge, there will then be in existence a new and overpowering circumstance, namely, the physical presence of the multimillion dollar New Don Pedro Dam and appurtenant works, which will prejudice freedom of action in providing for the preservation of fish runs.

All of these and other considerations of like nature were urged upon the Commission. But that agency had to consider the whole picture. While these petitioners say that construction of the new project will present the Commission with a fait accompli which will limit freedom of action at the end of the twenty-year period, the fact is that the Commission is already faced with a fait accompli. The districts and San Francisco are now free to operate the presently-existing river structures in a manner which will not protect fall fish runs. The prospect is that if the new project is not constructed, the entire Tuolumne fish run will be destroyed in the near future.

If this danger is to be averted, the Commission had to devise terms and conditions of a kind which the districts and San Francisco, balancing all of their present rights and future needs, could reasonably be expected to assume. They cannot be required to accept a license and they are not likely to if, in their considered judgment, the terms and conditions attached thereto threaten the economic feasibility of the project or substantially impair water rights assiduously acquired and developed through the years.

The district introduced credible evidence tending to show that, whatever the situation is now, in twenty years when San Francisco increases its diversion for municipal purposes to 295 mgd, there will not be sufficient water for the districts’ irrigation needs if they are required to continue releases on the scale called for in Article 37. This alone was sufficient to justify the Commission in limiting the releases to that period, subject to review in the manner indicated in Article 37.

If, because of insufficient Tuolumne water, further Tuolumne releases should not be ordered after that date, alternatives such as a fish hatchery system may prove feasible, depending on need and economic feasibility. In our view the present form of Article 37 is adequate to *929 enable the Commission to deal with the problem as it then exists and to impose such burdens on those who are in fact and law subject thereto, as the then-indicated solution requires.

Only the districts are licensees and any order the Commission may enter twenty years from now pursuant to Article 37 will therefore necessarily be directed only to them. This does not mean, however, that the districts, as licensees, may be immune from burdens justified because of the value of the New Don Pedro project to San Francisco.

Under section 10(a) of the Federal Power Act, 16 U.S.C. § 803 (a), the Commission is required to measure each proposed project against an objective comprehensive plan for the optimum coordinated development of the entire waterway. See United States ex rel. Chapman v. F. P. C., 345 U.S. 153, 168 , 73 S.Ct. 609 , 97 L.Ed. 918 . Fulfillment of that duty in this case necessarily requires the Commission to consider the benefits San Francisco will derive from the project.

We think the Commission adequately performed this duty, insofar as present purposes are concerned, when it accepted San Francisco’s agreed monetary contribution to the project, arrived at after arm’s-length bargaining between the districts and San Francisco as the measure of the latter’s economic benefit from the project.

Under Article 37, the duty of the licensees with regard to fish releases is fixed for only the first twenty years of the license period, and this Article appears to provide for maximum use of the water available for fish releases or other conservation purposes during that period. The project, as licensed, is engineered to take advantage of the entire storage capability of the area, having in view the natural topographic configuration. The only way in which storage capacity could be increased would be to run a ring dike entirely around the reservoir.

There is thus no reasonable possibility, and indeed no contention, that had specific evidence pertaining to the economic value of the project to San Francisco been received, it might have warranted the Commission in requiring that storage capacity be enlarged for fish release or other conservation purposes by constructing such a dike. Nor is there reason to believe that, during that period, other measures such as the construction and operation of fish hatcheries will be needed.

It follows that, for this initial period, Commission inquiry into the question of the economic value of the project to San Francisco, in all likelihood, would not have enabled the agency to make any more favorable order, insofar as conservation is concerned, than it has already entered.

However, when it comes to the question of determining what needs to be done to protect the public interest during the last thirty years of the license period, a problem which the Commission, under Article 37, has reserved, the Commission will undoubtedly require additional evidence as to the benefits accruing to San Francisco. The Commission orders should not be construed as foreclosing such an inquiry.

With respect to any burdens imposed upon the districts at that time and as may be warranted by the value of the project to San Francisco, the districts will be entitled to reimbursement from San Francisco. Whether or not the districts and San Francisco already have, or hereafter enter into an agreement making clear their respective rights and obligations in such an eventuality, the position in which they have placed themselves before the Commission requires this result. San Francisco must be deemed to have acceded to this principle if it proceeds with its commitments under the New Don Pedro project.

The Secretary further argues that, under section 23(b) of the Federal Power Act, 49 Stat. 846 (1935), 16 U.S.C. § 817 , San Francisco was a necessary joint applicant for the license, and the Commission erred in not requiring it to assume *930 that status in the case. Under that section it is unlawful, among other things, for any person, state or municipality, for the purpose of developing electric power, to construct dams and the like upon any public lands of the United States, except under a license granted by the Commission. The Secretary contends that although San Francisco will not be a legal owner of the project, it is a direct beneficiary thereof, and so comes under the sanction of section 23(b).

We do not believe that section 23(b) is to be so broadly construed. San Francisco not only has no proprietary interest in the project works, but as the Commission found, it has no interest whatever in the power to be developed at New Don Pedro. Any upstream benefits which it may realize, as the Commission further found, “ * * * will be only incidental to the main purpose of its participation, which is to acquire additional municipal water storage.”

As long as the Commission requires the districts, as actual legal owners of the project works, to obtain a license, and as long as the duties and obligations imposed upon those owners, as licensees, take into account all the monetary benefits to be derived by anyone from the project, the public interest is fully protected and the purpose of section 23(b) is achieved. We have indicated above that the Commission must take account of the benefits derived by San Francisco from the New Don Pedro project when it reviews the fish run problem as provided for in Article 37. We have also held that the Commission may at that time impose burdens upon the districts warranted by the benefits derived by San Francisco on the assumption that the latter will reimburse the districts for any such expenditures.

The Secretary and California have presented several additional arguments. Among these are the Secretary’s contention that the Commission erred “ * * * in wholly ignoring water quality and pollution problems,” and in failing to consider the impact which Article 37 might have on the Federal Central Valley Project. 10 The state argues that the selection of a “dry” year definition by the Commission is not based upon evidence or reason, and the Commission failed to provide, in Article 40 of the terms and conditions, sufficient control over the temperature of the water releases.

We have reviewed each of these arguments, and others not mentioned above, and conclude that the Commission did not err with regard to any of them.

The orders under review are affirmed.

. “Provided, that in a •water year when the inflow is less than 750,000 acre feet, the amount of water provided under Schedule B shall he reduced by a percentage of the total acre feet equivalent to the percentage reduction in the gravity diversion at La Grange by the Licensees. For this purpose, the diversion shall be defined as 900,-000 acre feet to Licensees.”

. Section 27 reads:

“§ 821. State laws and water rights unaffected

“Nothing contained in this chapter shall be construed as affecting or intending to affect or in any way to interfere with the laws of the respective States relating to the control, appropriation, use, or distribution of water used in irrigation or for municipal or other uses, or any vested right acquired therein.”

. “Opinion No. 420 clearly states that the releases which the applicants are now required to make will be subject to modification at the end of the 20-year period to assure the continuing feasibility of the New Don Pedro project; there is no assurance, however, that the applicants and San Francisco will thereafter be free to use the waters of river for irrigation and municipal purposes without limit. For the reasons spelled out in Opinion No. 420, the allocation of river waters which will best serve the public interest within the limits permissible under law appropriately should and will be made when and if the need for such allocation arises.”

. The factors mentioned are: vagaries of the weather, possible alternative sources of water for the purpose of preserving fish runs, development of other techniques such as fish hatcheries for preserving fish runs, possible destruction of the Tuolumne fishery by changes in flow downstream from La Grange by causes beyond the control of the districts, possible reduction in the districts’ irrigation requirements due to increased urbanization and industrialization or by the development of more efficient methods of utilizing water, and the possibility of reduced interest rates on the districts’ bonded indebtedness.

. “ * * * we do not seriously question the applicants’ contention that even without such a condition the salmon runs would be preserved longer with New Don Pedro than with existing Don Pedro. On the basis of the record, we would expect that New Don Pedro could be operated without adverse effect on the salmon runs until after 1985, when San Francisco’s upstream diversions reach 295 mgd; whereas without New Don Pedro it appears that the fish would be seriously affected, if not destroyed, after 1968, when San Francisco’s diversions reach 210 mgd.”

. “This evidence, however, does not substantiate the applicants’ contention that no fish water releases should be required in the New Don Pedro license. In our judgment such releases are required as hereinafter prescribed if the projects [sic] is to be found to be best adapted to a comprehensive plan for development of the waterway, since only by making the releases a condition of the license can we be sure that the project will be operated so as to utilize the available water in the best interest of all parties and provide the best plan for comprehensive development for all public uses.”

. See also, section 8(b) of the Administrative Procedure Act, 60 Stat. 242 (1946), 5 U.S.C. § 1007 (b).

. The Secretary and the state cite the following statutes: Fish and Wildlife Act of 1956, 70 Stat. 1119 -1124, as amended, 16 U.S.C. §§ 742a-742j (1958) ; Fish and Wildlife Coordination Act, 48 Stat. 401 -402 (1934), as amended, 16 U.S.C. §§ 661 -666c (1958) ; California Water Code, § 1257; California Fish and Game Code, § 1505.

. Experts testified that, during the period from 1953 to 1959, the Tuolumne River produced annually approximately 463,000 pounds of commercial salmon in California. On the average there are 2,600 commercial salmon fishermen, 1,330 salmon fishing boats, and thirty processing plants for salmon in California.

. The Secretary in his petition for intervention mentioned neither of these contentions.