delivered the opinion of the Court.

The main question in this case is whether California can enforce her minimum wholesale price regulations as respects milk sold to the United States at three military installations1 (Travis Air Force Base, Castle Air Force Base, and Oakland Army Terminal) located within California and used for strictly military consumption, for resale at federal commissaries and for consumption or resale at various military clubs and post exchanges. Milk used for the first two categories of use is paid for with *248appropriated funds, while that used in the clubs and exchanges is purchased with nonappropriated funds. Prior to January 1959, the milk supplies purchased with appropriated funds and used at those installations were obtained as a result of competitive bidding and on terms below the minimum prices prescribed by the Director of Agriculture of California. The Director advised distributors that the State’s minimum price regulations were applicable to sales at Travis. Subsequently bids for milk-supply contracts at Travis were in strict compliance with California’s regulations, the added cost to the Federal Government being about $15,000 a month. Later that year California instituted a civil action in the state courts against a cooperative that had supplied milk at Travis below the state minimum price, seeking civil penalties and an injunction. Thereafter the United States brought this suit in the District Court. The complaint alleged that state price regulation of milk sales at Travis, a federal enclave, was barred by the Constitution, since Travis is subject to the exclusive jurisdiction of the United States.2 It also alleged that such regulation was an unconstitutional burden on the United States in the exercise of its constitutional power to establish and maintain the Armed Forces and to acquire and manage a federal enclave. The complaint asked that a three-judge court be convened.

Meanwhile, the Director of Agriculture of California warned distributors that the California regulation would be enforced at Castle and at Oakland. Bids for milk thereafter received at Castle were all at or above the state minimum price; and accordingly they were rejected. A *249new invitation for bids was issued, and one of those received was below the state minimum. Thereupon California sued the successful bidder for an injunction; and later it sued other like bidders. A similar experience was had at Oakland; bids at or above the minimum were rejected, and a contract with a distributor for a prior period was extended for three months with an estimated saving to the United States of over $30,000. California again instituted suit to enjoin the supplier from selling at below established minimum wholesale prices. The United States amended its complaint to include its purchases at Castle. As respects Oakland the United States commenced a separate action by a complaint substantially identical with the other one; and they were later consolidated.

Appellants denied that these three installations were federal enclaves giving the United States exclusive jurisdiction and that there was any conflict between the state regulatory scheme and the federal procurement policy. Appellants also moved that the District Court stay these actions pending determination of state-law questions by the state courts in the pending actions.

The three-judge District Court refused to stay the proceedings and granted the motion of the United States for summary judgment. 190 F. Supp. 645.. We postponed a determination of jurisdiction to the merits. 368 U. S. 965.

Here, as in United States v. Georgia Public Service Comm’n, post, p. 285, decided this day, the suit was one “required” to be heard by a three-judge court within the meaning of 28 U. S. C. § 1253 and therefore properly brought here by direct appeal. Apart from the question whether the three federal areas were subject to the exclusive jurisdiction of the United States, the issue as to *250whether or not the state regulatory scheme burdened the exercise by the United States of its constitutional powers to maintain the Armed Services and to regulate federal territory was a substantial federal question, as Penn Dairies, Inc., v. Milk Comm’n, 318 U. S. 261, Public Utilities Comm’n of California v. United States, 355 U. S. 534, and United States v. Georgia Public Service Comm’n, supra, make clear. A three-judge court was therefore required even if other issues that might not pass muster on their own were also tendered. See 28 U. S. C. § 2281; Florida Lime & Avocado Growers, Inc., v. Jacobsen, 362 U. S. 73.

II.

The California Act authorizes the Director of Agriculture to prescribe minimum wholesale and retail prices3 “at which fluid milk or fluid cream shall be sold by distributors to retail stores, restaurants, confectioneries and other places for consumption on the premises.” 4 The prohibitions run both against sales and against purchases;5 and both criminal and civil penalties are provided.6 The minimum wholesale prices, promulgated by the Director of Agriculture, have been enforced with respect to sales to the United States, as already noted.

In Public Utilities Comm’n of California v. United States, supra, we held that the federal procurement policy, which required competitive bidding as the general rule and negotiated purchase or contract as the exception, prevailed over California’s regulated rate system. That case, like United States v. Georgia Public Service Comm’n, supra, concerned transportation of commodities. But the federal policy at the times relevant here was the same for procurement of supplies and services. The statutes in effect at the time of the Public Utilities Comm’n of California case are still the basic provisions governing all *251procurement by the Armed Services out of appropriated funds. They require that contracts be placed by competitive bidding, the award to be granted “to the responsible bidder whose bid . . . will be the most advantageous to the United States, price and other factors considered.” 7 There are statutory exceptions, the relevant ones being as follows:

“(a) Purchases of and contracts for property or services covered by this chapter shall be made by formal advertising in all cases in which the use of such method is feasible and practicable under the existing conditions and circumstances. If use of such method is not feasible and practicable, the head of an agency, subject to the requirements for determinations and findings in section 2310, may negotiate such a purchase or contract, if—

“(8) the purchase or contract is for property for authorized resale ;

“(9) the purchase or contract is for perishable or nonperishable subsistence supplies;

“(10) the purchase or contract is for property or services for which it is impracticable to obtain competition;

“(15) the purchase or contract is for property or services for which he determines that the bid prices received after formal advertising are unreasonable as to all or part of the requirements, or were not independently reached in open competition, and for which (A) he has notified each responsible bidder of intention to negotiate and given him reasonable opportunity to negotiate; (B) the negotiated price is *252lower than the lowest rejected bid of any responsible bidder, as determined by the head of the agency; and (C) the negotiated price is the lowest negotiated price offered by any responsible supplier.” 8

The Armed Services Procurement Regulation speaks in unambiguous terms of a policy “to use that method of procurement which will be most advantageous to the Government — price, quality, and other factors considered.” 9 The Regulation states, “Such procurement shall be made on a competitive basis, whether by formal advertising or by negotiation, to the maximum practicable extent . . . .” 10 Whatever method is used — formal advertising or negotiation — “competitive proposals” must be “solicited from all such qualified sources of supplies or services as are deemed necessary by the contracting officer to assure such full and free competition as ... to obtain for the Govérnment the most advantageous contract — • price, quality, and other factors considered.” 11 If advertising for bids is used, the contract is to be awarded “to the lowest responsible bidder.” 12 Moreover, even when advertising for bids is not used, competitive standards are not relaxed. The policy is “to procure supplies and services from responsible sources at fair and reasonable prices calculated to result in the lowest ultimate over-all cost to the Government.” 13 “The fact that a procurement is to be negotiated does not relax the requirements for competition.” 14 “Whenever supplies . . . are to be procured by negotiation, price quotations . . . shall be solicited *253from all such qualified sources of supplies or services as are deemed necessary ... to assure full and free competition ... to the end that the procurement will be made to the best advantage of the Government, price and other factors considered.” 15 The Regulation then specifies 20 separate considerations for the selection of a supplier in case of a negotiated procurement.16 The first of these is a “comparison of prices quoted.” 17

We have said enough to show that the Regulation does more than authorize procurement officers to negotiate for lower rates. It directs that negotiations or, wherever possible, advertising for bids shall reflect active competition so that the United States may receive the most advantageous contract.

While the federal procurement policy demands competition, the California policy, as respects milk, effectively eliminates competition. The California policy defeats the command to federal officers to procure supplies at the lowest cost to the United States by having a state officer fix the price on the basis of factors not specified in the federal law. Moreover, when the supply contract is negotiated because “it is impracticable to obtain competition,” to use the statutory words,18 it is the state agency, not the federal procurement officer and the seller, that determines the price provisions of the contract, if state policy prevails. The collision between the federal policy of negotiated prices and the state policy of regulated prices is as clear and acute here as was the conflict between federal negotiated rates and state regulated rates in Public Utilities Comm’n of California v. United States, supra. In that case we said that the Regulation then existing, which was promulgated under the same Act here involved, “sanc*254tion[ed] the policy of negotiating rates for shipment of federal property and entrust [ed] the procurement officers with the discretion to determine when existing rates'will be accepted and when negotiation for lower rates will be undertaken.” 355 U. S., at 542-543.

Penn Dairies, Inc., v. Milk Control Comm’n, supra, is not opposed. As we noted in United States v. Georgia Public Service Comm’n, supra, Congress, after the Penn Dairies decision and before Public Utilities Comm’n of California v. United States, revised and restated the federal procurement policy. As stated in the House Report,19 “. . . the bill represents a comprehensive revision and restatement of the laws governing the procurement of supplies and services by the War and Navy Departments. It holds to the time-tested method of competitive bidding. At the same time it puts within the framework of one law almost a century’s accumulation of statutes and incorporates new safeguards designed to eliminate abuses, assures the Government of fair and reasonable prices for the supplies and services procured and affords an equal opportunity to all suppliers to compete for and share in the Government’s business.”

The Regulation controlling the Penn Dairies decision stated, as does the present Act, that supplies might be purchased on the open market where it is “impracticable to secure competition.” 318 U. S., at 277. But, unlike the present Regulation, the earlier one declared that such a situation arose “when the price is fixed by federal, state, municipal or other competent legal authority.” Ibid. The earlier Regulation further stated that federal procurement officers should not require suppliers to comply with state price-fixing laws before it was judicially determined whether the latter were applicable to government contracts (id., at 276), a provision which *255the Court said manifested a federal “hands off” policy-respecting minimum price laws of the States. Id,., at 278.

The present Regulation makes no such allowances, contains no such qualifications, and provides for no such exception. Its unqualified command is that purchases for the Armed Services be made on a competitive basis; and it has, of course, the force of law. Public Utilities Comm’n of California v. United States, supra, at 542-543. California’s price-fixing policy for milk is as opposed to this federal procurement policy as was California’s rate-making policy in Public Utilities Comm’n of California v. United States, supra.

Policy-wise, it might be better if state price-fixing systems were honored by federal procurement officials. It is urged that if that were done substandard producers of some suppliers would lose the advantage they may enjoy in competitive bidding. Congress could of course write that requirement into the law. Congress has written into the Act certain provisions of that character. It has required that contractors or manufacturers pay not less than the minimum wage as determined by the Secretary of Labor to be the prevailing wage; that building contractors pay such minimum wages to laborers and mechanics; and that no laborer or mechanic doing any work for contractors and subcontractors on government contracts shall be required or permitted to work more than eight hours a day, unless one and a half times the basic rate is paid for overtime.20 The inclusion of these provisions, aimed as they are at substandard working conditions, shows that Congress has been alert to the problem. Their inclusion makes more eloquent the omission of any like requirement as respects prices or rates fixed by state law.

*256It is argued that the Act of September 10, 1962, 76 Stat. 528, changed the situation. California points to § 2306 (f), which requires contractors to submit cost or pricing data for any negotiated contract, but goes on to lift that requirement where “prices [are] set by law or regulation.” But this provision does not say, even equivocally, that federal procurement officers must abandon competitive bidding where prices are “set by law or regulation.” The Regulation makes competitive bidding the rule, as we have seen. Section 2306 (f) only provides for waiver of “cost or pricing data” under certain kinds of negotiated contracts if the prices of some commodities included in the contract have been “set by law or regulation.” That is to say, as, if, and when the procurement officer is authorized to accept prices “set by law or regulation,” he need not follow the requirements of § 2306 (f) concerning “cost or pricing data.”

California cites but builds no argument around § 2304 (g), also added in 1962. It is now suggested for the first time that § 2304 (g) requires federal procurement to follow state rate-fixing and state price-fixing. It provides in relevant part:

“In all negotiated procurements in excess of $2,500 in which rates or prices are not fixed by law or regulation and in which Jime of delivery will permit, proposals shall be solicited from the maximum number of qualified sources consistent with the nature and requirements of the supplies or services to be procured, and written or oral discussions shall be conducted with all responsible offerors who submit proposals within a competitive range, price, and other factors considered. . . .”

Here again, the new statutory provision does not purport to say when rates or prices “fixed by law or regulation” govern federal procurement. At the time § 2304 (g) was added to the Act, the Regulation which we have dis*257cussed at length was in full force. That Regulation, unlike the one in Penn Dairies, eliminated the earlier provisions which had been construed to manifest a federal “hands off” policy respecting minimum price laws of the States. 318 U. S., at 278. The Regulation in force when this litigation started and in force when the 1962 Act was passed provides unequivocally for competitive bidding “to the maximum practicable extent,” as we have noted. That might well permit procurement officers under some circumstances to purchase at state-fixed prices. But competitive bidding is the rule, not the exception. There is not a word in the legislative history of the 1962 Act21 *258which indicates a congressional policy to uproot the Regulation or to change it. It was, indeed, repeatedly approved. See S. Rep. No. 1884, 87th Cong., 2d Sess.; H. R. Rep. No. 1638, 87th Cong., 2d Sess., Parts I and II; Cong. *259Rec., June 7, 1962, p. 9231 et seg. Four years before the 1962 Act was passed California Comm’n had held that state regulations cannot preclude the Federal Government from negotiating lower rates. This result was not once questioned in the legislative history of the 1962 Act, even though the instant case was being litigated during this entire period. That Act only reflects an effort to provide collateral accommodations as, if, and when federal procurement follows state price-fixing. The mandate of 10 U. S. C. § 2305 (a) is still unequivocal; and the statutory exceptions to competitive bidding contained in § 2304 (a), discussed above, remain unchanged.

The 1962 Act fails to show a congressional purpose to abandon competitive bidding. On the contrary the pur*260pose, as stated in S. Rep. No. 1884, 87th Cong., 2d Sess., was to increase the efficacy of the competitive bidding system then in force.

Not only was the existing Regulation cited repeatedly with approval, but the aim of the Act was described in unambiguous terms:

“In general, the objectives of the changes are—

“(1) To encourage more effort to accomplish procurements by formal advertising;

“(2) To require a clearer justification before certain authorities to negotiate contracts are used;

“(3) To obtain more competition in negotiated procurement;

“(4) To provide safeguards for the Government against inflated cost estimates in negotiated contracts.” Id., p. 1.

The House received an equally unambiguous explanation from the floor manager of the bill:

“[TJhis bill . . . has for its chief purpose, an increase in competitive purchasing. . . . [OJnly 13 percent of purchasing is now done by sealed competitive bidding. That is clearly not enough. Competition must be increased; competition must be had even in negotiated purchasing; and all negotiated purchasing must be further reduced.” Cong. Rec., June 7, 1962, p. 9234.

If there had been a desire to make federal procurement policy bow to state price-fixing in face of the contrary policy expressed in the Regulation, we can only believe that the objectives of the Act would have been differently stated. In sum, the references to rates or prices “fixed by law or regulation” are merely minor collateral accommodations to those situations where, within the limits of the Regulation and the 1962 Act, the federal procurement *261official decides that the practical way to obtain the supplies or services is by following the state price-fixing or rate-fixing system.

California, however, says that whatever may be the federal policy as to purchases of milk for mess-hall use, purchases of milk for resale at federal commissaries stand on a different footing. These commissaries are “arms of the Government deemed by it essential for the performance of governmental functions” and “partake of whatever immunities” the Armed Services “may have under the Constitution and federal statutes.” Cf. Standard Oil Co. v. Johnson, 316 U. S. 481, 485. Purchases for resale at these federal commissaries are made from appropriated funds; and the procurement officers act under the same Regulation when they purchase milk for the commissaries as they do when they purchase it for mess-hall use. California points out, however, that the federal statute provides that where commodities are purchased for resale, they may be procured by negotiation rather than by formal advertising22 — a provision we have quoted above and which was written into the law because purchases for commissaries “are generally not made by specifications but by brand names.” 23 Milk, however, does not fit the category of commodities for which that exception was designed. Moreover, the statutory exception to formal advertising is merely permissive; the procurement officer “may” negotiate for articles to be resold but he is not required so to do. He is free to purchase by formal advertising from the responsible bidder whose bid “will be the most advantageous to the United States.” 24 Whether he negotiates milk contracts or uses competitive bidding is made dependent by the federal statute on his informed *262discretion, not on state price-fixing policies. Moreover, as, if, and when he negotiates, the Regulation, as already-noted, requires price quotations “from all such qualified sources of supplies or services as are deemed necessary by the contracting officer to assure full and free competition ... to the end that the procurement will be made to the best advantage of the Government, price and other factors considered.” 25 And, to repeat, the procurement officer when he negotiates is controlled by 20 separate factors, one of which is “comparison of prices quoted,” 26 and none of which relates in any manner whatsoever to the price-fixing policies of a State.

The fact that the cost of products sold at commissaries benefits commissary purchasers does not make the commissary any the less a federal agency. Cf. Standard Oil Co. v. Johnson, supra. Congress authorizes the payment for commissary supplies from appropriated funds.27 The federal statutes dealing with procurement policies expressly make them applicable to all purchases “for which payment is to be made from appropriated funds.” 28 Congress, to be sure, has provided that commissaries may not use any appropriated funds “unless the Secretary of Defense has certified that items normally procured from commissary stores are not otherwise available at a reasonable distance and a reasonable price in satisfactory quality and quantity to the military and civilian employees of the Department of Defense.” 29 Here again, however, the question of what is a “reasonable price” is left to the discretion of a federal officer. Congress has not *263directed that commissaries be removed from the purview of federal procurement policies; nor has it adopted state price-fixing policies as federal policies when it comes to purchases for commissaries or otherwise.

III.

What we have said would dispose of the entire case but for the fact that some of the milk was purchased out of nonappropriated funds for use in military clubs and for resale at post exchanges. This brings us to the question whether Congress has power to exercise “exclusive legislation” over these enclaves within the meaning of Art. I, § 8, cl. 17, of the Constitution, which reads in relevant part: “The Congress shall have Power ... To exercise exclusive Legislation in all Cases whatsoever” over the District of Columbia and “to exercise like Authority over all Places purchased by the Consent of the Legislature of the State in which the Same shall be, for the Erection of Forts, Magazines, Arsenals, dock-Yards, and other needful Buildings.”

The power of Congress over federal enclaves that come within the scope of Art. I, § 8, cl. 17, is obviously the same as the power of Congress over the District of Columbia. The cases make clear that the grant of “exclusive” legislative power to Congress over enclaves that meet the requirements of Art. I, § 8, cl. 17, by its own weight, bars state regulation without specific congressional action. The question was squarely presented in Pacific Coast Dairy v. Department of Agriculture, 318 U. S. 285, which involved, as does the present litigation, California’s Act and an attempt to fix the prices at which milk could be sold at Moffett Field. We held that “sales consummated within the enclave cannot be regulated” by California because of the constitutional grant of “exclusive legislation” respecting lands purchased by the United *264States with the consent of the State {id., at 294), even though there was no conflicting federal Regulation.

Thus the first question here is whether the three enclaves in question were “purchased by the Consent of the Legislature” of California within the meaning of Art. I, § 8, cl. 17.

The power of the Federal Government to acquire land within a State by purchase or by condemnation without the consent of the State is well established. Kohl v. United States, 91 U. S. 367, 371. But without the State’s “consent” the United States does not obtain the benefits of Art. I, § 8, cl. 17, its possession being simply that of an ordinary proprietor. James v. Dravo Contracting Co., 302 U. S. 134, 141-142. In that event, however, it was held in Ft. Leavenworth R. Co. v. Lowe, 114 U. S. 525, 541, 542, that a State could complete the “exclusive” jurisdiction of the Federal Government over such an enclave by “a cession of legislative authority and political jurisdiction.”

Thus if the United States acquires with the “consent” of the state legislature land within the borders of that State by purchase or condemnation for any of the purposes mentioned in Art. I, § 8, cl. 17, or if the land is acquired without such consent and later the State gives its “consent,” the jurisdiction of the Federal Government becomes “exclusive.” Since 1940 Congress has required the United. States to assent to the transfer of jurisdiction over the property, however it may be acquired.30 In either event — whether the land is ac*265quired by purchase or condemnation on the one hand or by cession on the other — a State may condition its “consent” upon its retention of jurisdiction over the lands consistent with the federal use. James v. Dravo Contracting Co., supra, 146-149. Moreover, as stated in Stewart & Co. v. Sadrakula, 309 U. S. 94, 99-100:

“The Constitution does not command that every vestige of the laws of the former sovereignty must vanish. On the contrary its language has long been interpreted so as to permit the continuance until abrogated of those rules existing at the time of the surrender of sovereignty which govern the rights of the occupants of the territory transferred. This assures that no area however small will be left without a developed legal system for private rights.”

California has had several statutory provisions relevant to our problem under Art. I, § 8, cl. 17. One pertained to acquisition of land by the United States through “purchase or condemnation.” 31 Another con*266cerned land “ceded or granted” by California to the United States.32

Those provisions were codified in 1943, acquisitions by “purchase or condemnation” appearing in one section33 and acquisitions by cession in another.34 Another section of the codification, after stating that California “cedes” to the United States “exclusive jurisdiction” over all lands “held, occupied, or reserved” by the United States “for military purposes or defense,” provides that a description of the land by metes and bounds and a map or plat of the land “shall first be filed in the proper office of record in the county in which the lands are situated.” 35

Most of the transactions creating these three federal enclaves took place between 1942 and 1944, some in 1946 36 and some even later.

*267Whether the United States has acquired exclusive jurisdiction over a federal enclave is a federal question. As stated in Silas Mason Co. v. Tax Commission, 302 U. S. 186, 197:

“The question of exclusive territorial jurisdiction is distinct. That question assumes the absence of any interference with the exercise of the functions of the Federal Government and is whether the United States has acquired exclusive legislative authority so as to debar the State from exercising any legislative authority, including its taxing and police power, in relation to the property and activities of individuals and corporations within the territory. The acquisition of title by the United States is not sufficient to effect that exclusion. It must appear that the State, by consent or cession, has transferred to the United States that residuum of jurisdiction which otherwise it would be free to exercise. ... In this instance, the Supreme Court of Washington has held that the State has not yielded exclusive legislative authority to the Federal Government. . . . That question, however, involving the extent of the jurisdiction of the United States, is necessarily a federal question.”

As already noted, a California statute “cedes to the United States exclusive jurisdiction” over described lands provided a description of the metes and bounds and a map of the land first be filed.37 California earnestly argues *268that “cedes” in that context includes “purchases” and “acquisitions by condemnation.” But the California statutes have consistently drawn the line between acquisitions by cession on the one hand and all other acquisitions on the other. That is the gist of a recent opinion of the Attorney General of California, in which he treats an acquisition by cession as an alternative to acquisition in other ways and rules that when the acquisition is by means other than cession no map of the land need first be filed.38 That seems to us to be the fair meaning of the statutory provisions.

The conditions expressed in the California Acts,39 by which California consented to “the purchase or condemnation” of land by the United States for the prescribed purposes, do not undertake to make applicable to the federal enclaves all future laws of California. Since a State may not legislate with respect to a federal enclave unless it reserved the right to do so when it gave its consent to the purchase by the United States, only state law existing at the time of the acquisition remains enforceable, not subsequent laws. See Stewart & Co. v. Sadrakula, supra; Arlington Hotel v. Fant, 278 U. S. 439. If the price-control laws California is now seeking to apply to sales on federal enclaves were not in effect when the United States acquired these lands,40 the case is on all fours with Pacific Coast Dairy v. Department of Agriculture, supra. There the Court held that the California statutes under which some of the present acquisitions were made granted the United States exclusive jurisdiction over the tracts in question in spite of the express conditions therein contained (id., at 293) and that this price-control law was *269not enforceable on a federal enclave in California because it was adopted “long after the transfer of sovereignty.” 318 U. S., at 294. The United States seeks shelter under that rule, saying California is trying to enforce its current regulatory scheme, not the price regulations in effect when the purchases were made. Yet if there were price control of milk at the time of the acquisition and the same basic scheme has been in effect since that time, we fail to see why the current one, albeit in the form of different regulations, would not reach those purchases and sales of milk on the federal enclave made from nonappropriated funds. Congress could provide otherwise and has done so as respects purchases and sales of milk from appropriated funds. But since there is no conflicting federal policy concerning purchases and sales from nonappropriated funds, we conclude that the current price controls over milk are applicable to these sales, provided the basic state law authorizing such control has been in effect since the times of these various acquisitions. A remand will be necessary to resolve that question, as the present record does not show the precise evolution of the present regulatory scheme.

There also remains another uncertainty concerning the purchases and sales of milk out of nonappropriated funds. There is a dispute over where some of these sales are made. Each of the three enclaves has numerous units acquired at various times, some of which may be subject to “exclusive” federal jurisdiction and some of which may not be. California earnestly claims that some sales out of nonappropriated funds were made on units of land over which the United States does not have “exclusive” jurisdiction. She makes the claim as respects some milk used at Travis, some at Castle, and some at Oakland.

We do not resolve the question but vacate the judgment of the District Court insofar as it relates to purchases and sales of milk made from nonappropriated funds and *270remand the case to the District Court to determine whether at the respective times when the various tracts in question were acquired California’s basic price-control law as respects milk was in effect. If so, judgment on this class of purchases and sales should be for appellants. If not, then the District Court must make particularized findings as to where the purchases and sales of milk from nonappropriated funds are made and whether or not those tracts are areas over which the United States has “exclusive” jurisdiction within the meaning of Art. I, § 8, cl. 17 of the Constitution.

Moreover, the decree must be modified to reflect the change in federal procurement policy as respects producers, already noted.41

Accordingly the judgment is affirmed in part and in part vacated and remanded.

It is so ordered.

whom

I.

I do not doubt that Congress in the exercise of its war power1 could by virtue of the Supremacy Clause 2 provide that an otherwise valid state law affecting the price of milk shall not apply to milk purchased with federal funds for use at these military installations. But I cannot agree that Congress has done so. I am unable to find either in the terms of the relevant legislation or in its history any evidence of a congressional purpose to immunize these federal purchases from the generally applicable California minimum price regulations. The *271California statutes regulating its milk industry are admittedly a valid exercise of that State’s power to legislate for the general health and welfare of its people, and serve the important function of insuring stability in the production and supply of a vital commodity. In Penn Dairies v. Milk Control Comm’n, 318 U. S. 261, the Court emphasized that “[a]n unexpressed purpose of Congress to set aside statutes of the states regulating their internal affairs is not lightly to be inferred and ought not to be implied where the legislative command, read in the light of its history, remains ambiguous.” 318 U. S., at 275. I think that the congressional purpose in the present case is less than ambiguous — that Congress has in fact manifested a presumption and a desire that valid state welfare legislation such as this is not to be undermined by the procurement activities of the Federal Government.3

In the Penn Dairies case the Court held that the State of Pennsylvania could enforce its milk marketing statute against suppliers dealing with the federal military establishment. It was held that the federal procurement legislation then in effect contained no evidence of a policy to override state regulatory legislation of this type. 318 U. S., at 272-275. A different result was reached in California Comm’n v. United States, 355 U. S. 534, where the Court held that California could not apply its law regulating intrastate transportation rates to the carriage of strategic military supplies of the United States. The Court discussed at length the peculiarly burdensome *272effect that the state regulation there involved would have upon the shipment of this kind of freight, stressing the difficulty and delays involved in classifying such goods under existing state tariffs, and the importance to the national security of secrecy and rapid movement. 355 U. S., at 544 — 546. Regardless of any impact on transportation costs, therefore, enforcement of the State’s regulatory scheme was barred because it constituted a direct interference with the performance of a vital federal function. M’Culloch v. Maryland, 4 Wheat. 316. The opinion in the California Commission case also discussed the 1947 Armed Services Procurement Act4 but nowhere suggested that the 1947 Act had changed the law upon which the decision in Penn Dairies had rested. Rather, the Court distinguished the Penn Dairies case on the ground that the Pennsylvania milk marketing statute had not subjected the National Government or its officers to any direct restraints, as did the California legislation. 355 U. S. 543-544.

The Court today abandons that distinction and for the first time suggests that the 1947 Act did in fact change the federal procurement policy in effect at the time of the Penn Dairies decision. I think this novel interpretation of the statute which is the basis of all federal procurement, civilian as well as military5 is incorrect and that *273any doubt which could ever have existed on that score has been laid to rest by the amendment to the 1947 statute enacted at the last session of Congress.6

There is simply no support in any of the pertinent legislative materials for the conclusion that Congress, solely in order to save a few dollars, intended to permit federal agencies to subvert general and nondiscriminatory state *274regulatory measures which promote health, safety, or better working or economic conditions. Indeed, Congress has evidenced a directly contrary intention. Of course, as the decision in the California Commission case demonstrates, state law cannot be allowed to impair fulfillment of appropriate federal functions, be they civil or military; similarly, state measures contrary to national policy cannot be allowed to bind or inhibit federal activities. This case, however, presents no such problems. The only issue is whether Congress has or has not expressed a desire to bypass valid state regulatory legislation in the conduct of federal procurement activities.

The 1947 Armed Services Procurement Act was proposed to Congress jointly by the War and Navy Departments. During World War II, these Departments had run their procurement operations with a relatively free hand under the First War Powers Act, 55 Stat. 838, which authorized placement of contracts without regard to existing provisions of law regulating procurement procedures. The War Production Board had early determined that the traditional method of procurement by advertising for sealed competitive bids was unsatisfactory during wartime, and had adopted the practice of placing contracts by direct negotiation with suppliers.7 When the war ended, the need arose to return to a peacetime system of procurement, and the 1947 bill was introduced to fill this need. At the same time, the military departments thought that the prewar procurement statutes were “woefully inadequate for supplying the tremendously expanding needs for military supplies and equipment,” 8 and that *275“a total reversion to prewar methods would be unfortunate in the extreme and would severely handicap the War and Navy Departments . ...”9

Reflecting this attitude, the Departments stressed three major objectives of the new legislation they proposed:

“1. To modernize peacetime military procurement methods;

“2. To unify the procurement legislation under which the War and Navy Departments do their buying; and

“3. To permit suspension of advertising as a method of procurement upon the declaration of a national emergency.” 10

The third purpose, to provide authority to suspend competitive bidding in a national emergency, was simply intended to eliminate the need for legislation in time of crisis and thus to enable the defense establishment to respond immediately to such emergencies.11 As for the second, prior to the war each branch of the armed services had been governed by its own separate, and sometimes unique, procurement legislation. The 1947 Act was intended to substitute a single statute for all military procurement.12

The proposal to “modernize” the law was primarily a proposal to relax, in certain situations, the very strict rule requiring that almost all contracts be placed through advertised competitive bidding. Experience had shown *276that the formalized ritual of competitive bidding was often unwieldy and uneconomical. For example, competitive bidding was not suited to contracts involving secret projects, nor for contracts involving items for which there was no effective competition between sellers. For these types of procurement, the bill proposed direct negotiation between the Government and available suppliers. The heart of the proposed bill was § 2 (c), now 10 U. S. C. § 2304.(a), which set out a list of 15 specific exceptions to the rule of competitive bidding.13

The bill was reported out and passed in essentially the same form as proposed. Both the House and Senate Reports made clear that the purposes of the bill remained the same. The House Report began by saying, “This bill provides uniform purchase authority for the Army and Navy, and reestablishes the requirement that the advertising-competitive bid method shall be followed by those Departments in placing the great majority of their contracts for supplies and services.” 14 The Report went on to acknowledge that there are “a limited number of situations [in which] the public interest requires that purchases be made without advertising,” and it listed most of the specific exceptions proposed by the War and Navy Departments.15 It was at this point, after describing the purpose to unify procurement laws and to relax the previously rigid advertising requirements, that the House Report summed up by describing the bill as “a comprehensive revision and restatement of the laws gov*277erning the procurement of supplies and services by the War and Navy Departments.” Id., at 6.16

The background of the 1947 Act thus makes it abundantly clear that the “revision and restatement” of law involved in its formulation had absolutely nothing to do with the issue dealt with in Penn Dairies and presented by the case now before us. The dissatisfaction with existing prewar procurement law centered upon its lack of uniformity and its apparent insistence upon the ritual of competitive bidding in situations for which such procedures were unsuited. Neither of these major concerns touched upon the problem presented by the present case — ■ whether federal procurement transactions were to undermine valid state laws regulating price.17

Evidence is not lacking, however, of the attitude of Congress with respect to that problem, and I think such evidence clearly shows that Congress presumed and intended that federal procurement was to be conducted subject to valid state price and rate regulation of otherwise general applicability.

First, it is clear from the Act itself that Congress was not willing to override other important social and economic policies in blind pursuit of the lowest possible purchasing price. The Act commands procurement officers to consider many other factors in addition to price. For instance, § 8 directs compliance with the Walsh-Healey Act, the Davis-Bacon Act and the Eight Hour Law.18 *278And in § 2 (b) Congress declared that a fair proportion of purchases and contracts made under the chapter should be placed with small business.19

Secondly, while the legislative history of the 1947 Act contains only a few references to the specific problem of price-regulated industries, these references clearly reflect an acknowledgment that state price regulations are to apply to suppliers doing business with the Government. The statements in question relate to § 2 (c)(10) of the bill as enacted, now 10 U. S. C. § 2304 (a) (10). The subsection provides that the head of an agency need not employ the advertised bid method when

“(10) the purchase or contract is for property or services for which it is impracticable to obtain competition.”

This exception to the normal bidding procedure was first enacted in the Army Appropriations Act of 1901, 31 Stat. 905. Until 1947 it applied only to Army procurement, and one of the purposes of the Act was to make the exception applicable to all services.20 In explaining the existing law on this subject, Under Secretary Royall of the War Department, chief spokesman for that Department, made the following remarks:

“As to the exception which deals with supplies or services for which it is impracticable to secure competition, this language originally appeared in the act of March 2, 1901 (31 Stat. 1905; 10 U. S. C. 1201), and has been the subject of a number of highly restrictive administrative interpretations. In my opin*279ion, this exception is intended to apply in at least these three situations:

“1. Where the nature of the supply or service is such that only one person can furnish it, for example, a patented or secret article.

“2. Where the price of the supply or service has been legally fixed.

“3. Where the practical circumstances are such that it would be difficult to secure real competitive proposals by means of advertising for formal bids.” (Emphasis added.)21

The Senate Report expressly acknowledged the applicability to federal procurement activities of laws regulating prices:

“The experiences of the war and contracts negotiated since the war in the fields of stevedoring, ship repairs, chartering of vessels, where prices are set by law or regulation, or where there is a single source of supply, have shown clearly that the competitive-bid-advertising method is not only frequently impracticable but does not always operate to the best interests of the Government.” (Emphasis added.) 22

The plain meaning of these references to price regulation is that both Congress and the Departments concerned assumed such price regulation would apply to government purchases. Unless this assumption is made, there would be no reason for believing that competition would be “impracticable” in these areas. For, absent the duty of suppliers to comply with uniform price regulations, it *280would not be “impracticable” to advertise for bids at competitive prices.

Apart from the clear import of these references, it is also significant to note that both the Departments and the sponsoring congressional committees were aware of the fact that governmental price fixing would affect the nature of competition for procurement contracts. Yet not once did any spokesmen for the Departments question or even mention the rule of the Penn Dairies decision, of which they could hardly have been unaware.23 Indeed, they consistently testified that § 2(c) (10) was, as to regulated prices, merely a restatement of the existing law.24

Despite this clear legislative history, it is said that the statutory authorization to “negotiate” in cases where com*281petitive bidding is not appropriate reflects a policy to allow procurement officers to bargain for prices lower than those set by state regulatory agencies.25 If all we had to go on were this provision of the 1947 Act, there might be an arguable basis for an inference of such a federal procurement policy, since the 1947 statute nowhere defined the word “negotiation.” 26 Just last year, however, Congress added an amendment to the Act, in which it defined “negotiation” for the first time. Although the definition generally adopts and implements the ordinary meaning of the word — to bargain for a lower price — it expressly excepts price-regulated transactions. The amendment provides in pertinent part:

“(g) In all negotiated procurements in excess of $2,500 in which rates or prices are not fixed by law or regulation and in which time of delivery will permit, proposals shall be solicited from the maximum number of qualified sources consistent with the nature and requirements of the supplies or services to be procured, and written or oral discussions shall be conducted with all responsible offerors who submit proposals within a competitive range, price, and other factors considered. . . .” 27 (Emphasis added.)

*282In the words of the floor manager of the bill in the House, price-regulated transactions were excepted because they were “instances where it would be futile to have discussions.” 28 In short, it is clear that Congress has now explicitly declared what was adumbrated in the legislative history of the 1947 Act — that federal procurement is to be conducted subject to valid state price and rate regulations of otherwise general applicability.29

While the Court’s opinion discusses this legislative history, I read the opinion as resting primarily on the Court’s reading of certain executive regulations issued under the authority of the procurement law. In this I think the Court errs — for two reasons. First, if I am *283right in the view that the statute recognizes that federal procurement is not to be immunized from the impact of valid state economic legislation, then any regulations to the contrary are completely invalid. Williamson v. United States, 207 U. S. 425, 462; Lynch v. Tilden Co., 265 U. S. 315, 321-322; United States v. Barnard, 255 F. 2d 583, 588-589. Secondly, I think that the regulations upon which the Court relies do not speak with so clear a voice as the Court would have us believe. The Court can find not a single regulation of either general or specific application which says, in so many words, that a procurement officer may in his discretion negotiate a contract in disregard of valid state price regulation.

II.

I agree with the conclusion in Part III of the Court’s opinion that it is not now possible to undertake final resolution of the Government’s claim that the sales of milk involved in this case take place on federal enclaves within the scope of Art. I, § 8, cl. 17, and therefore are immune from state regulation under the rule of Pacific Coast Dairy v. Department of Agriculture, 318 U. S. 285. Even if these military installations are now such federal enclaves, this claim will be moot if the substance of California’s milk regulation scheme antedated the acquisition of exclusive jurisdiction by the Federal Government. The concept of exclusive jurisdiction “has long been interpreted so as to permit the continuance until abrogated of those rules existing at the time of the surrender of sovereignty . . . .” Stewart & Co. v. Sadrakula, 309 U. S. 94, 99. This question of priority cannot be decided on the record before us, and its resolution, therefore, first requires a remand of the case to the District Court. If I am right in my view of the federal procurement law, a finding that state regulation was imposed before these *284military installations became federal enclaves within the scope of the constitutional provision would mean that all sales of milk at issue in this case, regardless of the source of funds, would be subject to the legislation which California has validly enacted to stabilize and make economically sound the business of producing and marketing a commodity vital to the health and welfare of her people.