Petitioner, a taxpayer and voter in Midland County, Texas, sought a determination by this Court that the Texas Supreme Court erred in concluding that selection of the Midland County Commissioners Court from single-member districts of substantially unequal population did not necessarily violate the Fourteenth Amendment. We granted review, 388 U.S. 905, 87 S.Ct. 2106, 18 L.Ed.2d 1345 (1967), because application of the one man, one vote principle of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964), to units of local government is of broad public importance. We hold that petitioner as a resident of Midland County, has a right to a vote for the Commissioners Court of substantially equal weight to the vote of every other resident.

Midland County has a population of about 70,000. The Commissioners Court is composed of five members. One, the County Judge, is elected at large from the entire county, and in practice casts a vote only to break a tie. The other four are Commissioners chosen from districts. The population of those districts, according to the 1963 estimates that were relied upon when this case was tried, was respectively 67,906; 852; 414; and 828. This vast imbalance resulted from placing in a single district virtually the entire city of Midland, Midland County's only urban center, in which 95% of the county's population resides.

The Commissioners Court is assigned by the Texas Constitution and by various statutory enactments with a variety of functions. According to the commentary to Vernon's Texas Statutes, the court:

'is the general governing body of the county. It establishes a courthouse and jail, appoints numerous minor officials such as the county health officer, fills vacancies in the county offices, lets contracts in the name of the county, builds roads and bridges, administers the county's public welfare services, performs numerous duties in regard to elections, sets the county tax rate, issues bonds, adopts the county budget, and serves as a board of equalization for tax assessments.'1

The court is also authorized, among other responsibilities, to build and run a hospital, Tex.Rev.Civ.Stat.Ann., Art. 4492 (1966), an airport, id., Art. 351 (1964), and libraries, id., Art. 1677 (1962). It fixes boundaries of school districts within the county, id., Art. 2766 (1965), may establish a regional public housing authority, id., Art. 1269k, § 23a (1963), and determines the districts for election of its own members, Tex.Const., Art. V, § 18, Vernon's Ann.St.

Petitioner sued the Commissioners Court and its members in the Midland County District Court, alleging that the disparity in district population violated the Fourteenth Amendment and that he had standing as a resident, taxpayer, and voter in the district with the largest population. Three of the four commissioners testified at the trial all telling the court (as indeed the population statistics for the established districts demonstrated) that population was not a major factor in the districting process. The trial court ruled for petitioner. It made no explicit reference to the Fourteenth Amendment, but said the apportionment plan in effect was not 'for the convenience of the people,' the apportionment standard established by Art. V, § 18, of the Texas Constitution. The court ordered the defendant commissioners to adopt a new plan in which each precinct would have 'substantially the same number of people.'

The Texas Court of Civil Appeals reversed the judgment of the District Court and entered judgment for the respondents, 397 S.W.2d 919 (1965). It held that neither federal nor state law created a requirement that Texas county commissioners courts be districted according to population.

The Texas Supreme Court reversed the Court of Civil Appeals, 406 S.W.2d 422 (1966). It held that under 'the requirements of the Texas and the United States Constitutions' the present districting scheme was impermissible 'for the reasons stated by the trial court.' 406 S.W.2d, at 425. However, the Supreme Court disagreed with the trial court's conclusion that precincts must have substantially equal populations, stating that such factors as 'number of qualified voters, land areas, geography, miles of county roads and taxable values' could be considered. 406 S.W.2d, at 428. It also decreed that no Texas courts could redistrict the Commissioners Court. 'This is the responsibility of the commissioners court and is to be accomplished within the constitutional boundaries we have sought to delineate.' 406 S.W.2d at 428—429.2

In Reynolds v. Sims, supra, the Equal Protection Clause was applied to the apportionment of state legislatures. Every qualified resident, Reynolds determined, has the right to a ballot for election of state legislators of equal weight to the vote of every other resident, and that right is infringed when legislators are elected from districts of substantially unequal population. The question now before us is whether the Fourteenth Amendment likewise forbids the election of local government officials from districts of disparate population. As has almost every court which has addressed itself to this question,3 we hold that it does.4

The Equal Protection Clause reaches the exercise of state power however manifested, whether exercised directly or through subdivisions of the State.

'Thus the prohibitions of the Fourteenth Amendment extend to all action of the State denying equal protection of the laws; whatever the agency of the State taking the action * * *.' Cooper v. Aaron, 358 U.S. 1, 17, 78 S.Ct. 1401, 1409, 3 L.Ed.2d 5 (1958).

Although the forms and functions of local government and the relationships among the various units are matters of state concern, it is now beyond question that a State's political subdivisions must comply with the Fourteenth Amendment.5 The actions of local government are the actions of the State. A city, town, or county may no more deny the equal protection of the laws than it may abridge freedom of speech, establish an official religion, arrest without probable cause, or deny due process of law.

When the State apportions its legislature, it must have due regard for the Equal Protection Clause. Similarly, when the State delegates lawmaking power to local government and provides for the election of local officials from districts specified by statute, ordinance, or local charter, it must insure that those qualified to vote have the right to an equally effective voice in the election process. If voters residing in oversize districts are denied their constitutional right to participate in the election of state legislators, precisely the same kind of deprivation occurs when the members of a city council, school board, or county governing board are elected from districts of substantially unequal population. If the five senators representing a city in the state legislature may not be elected from districts ranging in size from 50,000 to 500,000, neither is it permissible to elect the members of the city council from those same districts. In either case, the votes of some residents have greater weight than theose of others; in both cases the equal protection of the laws has been denied.

That the state legislature may itself be properly apportioned does not exempt subdivisions from the Fourteenth Amendment. While state legislatures exercise extensive power over their constituents and over the various units of local government, the States universally leave much policy and decisionmaking to their governmental subdivisions. Legislators enact many laws but do not attempt to reach those countless matters of local concern necessarily left wholly or partly to those who govern at the local level. What is more, in providing for the governments of their cities, counties, towns, and districts, the States characteristically provide for representative government—for decisionmaking at the local level by representatives elected by the people. And, not infrequently, the delegation of power to local units is contained in constitutional provisions for local home rule which are immune from legislative interference. In a word, institutions of local government have always been a major aspect of our system, and their responsible and responsive operation is today of increasing importance to the quality of life of more and more of our citizens. We therefore see little difference, in terms of the application of the Equal Protection Clause and of the principles of Reynolds v. Sims, between the exercise of state power through legislatures and its exercise by elected officials in the cities, towns, and counties.6

We are urged to permit unequal districts for the Midland County Commissioners Court on the ground that the court's functions are not sufficiently 'legislative.' The parties have devoted much effort to urging that alternative labels 'administrative' versus 'legislative'—be applied to the Commissioners Court. As the brief description of the court's functions above amply demonstrates, this unit of local government cannot easily be classified in the neat categories favored by civics texts. The Texas commissioners courts are assigned some tasks which would normally be thought of as 'legislative,' others typically assigned to 'executive' or 'administrative' departments, and still others which are 'judicial.' In this regard Midland County's Comissioners Court is representative of most of the general governing bodies of American cities, counties, towns, and villages.7 One knowledgeable commentator has written of 'the states' varied, pragmatic approach in establishing government.' R. Wood, in Politics and Government in the United States 891—892 (A. Westin ed. 1965). That approach has produced a staggering number of governmental units—the preliminary calculation by the Bureau of the Census for 1967 is hat there are 81,304 'units of government' in the United States8—and an even more staggering diversity. Nonetheless, while special-purpose organizations abound and in many States the allocation of functions among units results in instances of overlap and vacuum, virtually every American lives within what he and his neighbors regard as a unit of local government with general responsibility and power for local affairs. In many cases citizens reside within and are subject to two such governments, a city and a county.

The Midland County Commissioners Court is such a unit. While the Texas Supreme Court found that the Commissioners Court's legislative functions are 'negligible,' 406 S.W.2d, at 426, the court does have power to make a large number of decisions having a broad range of impacts on all the citizens of the county. It sets a tax rate, equalizes assessments, and issues bonds. It then prepares and adopts a budget for allocating the county's funds, and is given by statute a wide range of discretion in choosing the subjects on which to spend. In adopting the budget the court makes both long-term judgments about the way Midland County should develop—whether industry should be solicited, roads improved, recreation facilities built, and land set aside for schools—and immediate choices among competing needs.

The Texas Supreme Court concluded that the work actually done by the Commissioners Court 'disproportionately concern(s) the rural areas,' 406 S.W.2d, at 428. Were the Commissioners Court a special-purpose unit of government assigned the performance of functions affecting definable groups of constituents more than other constituents, we would have to confront the question whether such a body may be apportioned in ways which give greater influence to the citizens most affected by the organization's functions. That question, however, is not presented by this case, for while Midland County authorities may concentrate their attention on rural roads, the relevant fact is that the powers of the Commissioners Court include the authority to make a substantial number of decisions that affect all citizens, whether they reside inside or outside the city limits of Midland. The Commissioners maintain buildings, administer welfare services, and determine school districts both inside and outside the city. The taxes imposed by the court fall equally on all property in the county. Indeed, it may not be mere coincidence that a body apportioned with three of its four voting members chosen by residents of the rural area surrounding the city devotes most of its attention to the rpoblems of that area, while paying for its expenditures with a tax imposed equally on city residents and those who live outside the city. And we might point out that a decision not to exercise a function within the court's power—a decision, for example, not to build an airport or a library, or not to participate in the federal food stamp program—is just as much a decision affecting all citizens of the county as an affirmative decision

The Equal Protection Clause does not, of course, require that the State never distinguish between citizens, but only that the distinctions that are made not be arbitrary or invidious. The conclusion of Reynolds v. Sims was that bases other than population were not acceptable grounds for distinguishing among citizens when determining the size of districts used to elect members of state legislatures. We hold today only that the Constitution permits no substantial variation from equal population in drawing districts for units of local government having general governmental powers over the entire geographic area served by the body.

This Court is aware of the immense pressures facing units of local government, and of the greatly varying problems with which they must deal. The Constitution does not require that a uniform straitjacket bind citizens in devising mechanisms of local government suitable for local needs and efficient in solving local problems. Last Term, for example, the Court upheld a procedure for choosing a school board that placed the selection with school boards of component districts even though the component boards had equal votes and served unequal populations. Sailors v. Board of Education of Kent County, 387 U.S. 105, 87 S.Ct. 1549, 18 L.Ed.2d 650 (1967). The Court rested on the administrative nature of the area school board's functions and the essentially appointive form of the scheme employed. In Dusch v. Davis, 387 U.S. 112, 7 S.Ct. 1554, 18 L.Ed.2d 656 (1967), the Court permitted Virginia Beach to choose its legislative body by a scheme that included at-large voting for candidates, some of whom had to be residents of particular districts, even though the residence districts varied widely in population.

The Sailors and Dusch cases demonstrate that the Constitution and this Court are not roadblocks in the path of innovation, experiment, and development among units of local government. We will not bar what Professor Wood has called 'the emergence of a new ideology and structure of public bodies, equipped with new capacities and motivations * * *.' R. Wood, 1400 Governments, at 175 (1961). Our decision today is only that the Constitution imposes one ground rule for the development of arrangements of local government: a requirement that units with general governmental powers over an entire geographic area not be apportioned among single-member districts of substantially unequal population.

The judgment below is vacated and the case is remanded for disposition not inconsistent with this opinion.

It is so ordered.

Judgment vacated and case remanded.

Mr. Justice MARSHALL took no part in the consideration or decision of this case.

Mr. Justice HARLAN, dissenting.

I could not disagree more with this decision, which wholly disregards statutory limitations upon the appellate jurisdiction of this Court in state cases and again betrays such insensitivity to the appropriate dividing lines between the judicial and political functions under our constitutional system.

I believe that this Court lacks jurisdiction over this case because, properly analyzed, the Texas judgment must be seen either to rest on an adequate state ground or to be wanting in 'finality.' The history of the Texas proceedings, as related in the Court's opinion, ante, at 477-478, clearly reveals that the decision of the Texas Supreme Court disallowing the present county apportionment scheme rests upon a state as well as a federal ground. The state ground—Art. V, § 18, of the Texas Constitution was clearly adequate to support the result. This should suffice to defeat the exercise of this Court's jurisdiction. See, e.g., Department of Mental Hygiene of Cal. v. Kirchner, 380 U.S. 194, 85 S.Ct. 871, 13 L.Ed.2d 753; Herb v. Pitcairn, 324 U.S. 117, 125 126, 65 S.Ct. 459, 462—463, 89 L.Ed. 789.

Nor does this Court have jurisdiction to review the Texas Supreme Court's statement that in reapportioning the county in the future the county commissioners may take into account factors other than population. That holding obviously does not amount to a '(f)inal judgment' within the meaning of 28 U.S.C. § 1257.1 The traditional test of finality of state court judgments has been whether the judgment leaves more than a ministerial act to be done. See, e.g., Pope v. Atlantic Coast Line R. Co., 345 U.S. 379, 382, 73 S.Ct. 749, 751, 97 L.Ed. 1094; Republic Natural Gas Co. v. Oklahoma, 334 U.S. 62, 68, 68 S.Ct. 972, 976, 92 L.Ed. 1212. It is clear that the acts which must be performed in order to bring about a new apportionment of Midland County are very far from ministerial in character, and conceivably might even result in satisfying petitioner's demands without further litigation. For example, since the statement of the Texas Supreme Court regarding nonpopulation factors was merely advisory and not mandatory, the county commissioners might choose to reapportion the county solely on the basis of population, thus leaving petitioner with nothing about which to complain. Since the requirement of finality is an unwaivable condition of this Court's jurisdiction, see, e.g., Market St. R. Co. v. Railroad Comm'n of State of Cal., 324 U.S. 548, 551, 65 S.Ct. 770, 772, 89 L.Ed. 1171, I consider that this case is not properly before us.

On these scores, I would dismiss the writ as improvidently granted.

On the merits, which I reach only because the Court has done so, I consider this decision, which extends the state apportionment rule of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506, to an estimated 80,000 units of local government throughout the land, both unjustifiable and ill-advised.

I continue to think that these adventures of the Court in the realm of political science are beyond its constitutional powers, for reasons set forth at length in my dissenting opinion in Reynolds, 377 U.S., at 589, 84 S.Ct., at 1395 et seq. However, now that the Court has decided otherwise, judicial self-discipline requires me to follow the political dogma now constitutionally embedded in consequence of that decision. I am not foreclosed, however, from remonstrating against the extension of that decision to new areas of government. At the present juncture I content myself with stating two propositions which, in my view, stand strongly against what is done today. The first is that the 'practical necessities' which have been thought by some to justify the profound break with history that was made in 1962 by this Court's decision in Baker v. Carr, 369 U.S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663,2 are not present here. The second is that notwithstanding Reynolds the 'one man, one vote' ideology does not provide an acceptable formula for structuring local governmental units.

The argument most generally heard for justifying the entry of the federal courts into the field of state legislative apportionment is that since state legislatures had widely failed to correct serious mal-apportionments in their own structure, and since no other means of redress had proved available throgh the political process, this Court was entitled to step into the picture.3 While I continue to reject that thesis as furnishing an excuse for the federal judiciary's straying outside its proper constitutional role, and while I continue to believe that it bodes ill for the country and the entire federal judicial system if this Court does not firmly set its face against this loose and short-sighted point of view, the important thing for present purposes is that no such justification can be brought to bear in this instance.

No claim is made in this case that avenues of political redress are not open to correct any malapportionment in elective local governmental units, and it is difficult to envisage how such a situation could arise. Local governments are creatures of the States, and they may be reformed either by the state legislatures, which are now required to be apportioned according to Reynolds, or by amendment of state constitutions.4 In these circumstances, the argument of practical necessity has no force. The Court, then, should withhold its hand until such a supposed necessity does arise, before intruding itself into the business of restructuring local governments across the country.

There is another reason why the Court should at least wait for a suitable period before applying the Reynolds dogma to local governments. The administrative feasibility of judicial application of the 'one man, one vote' rule to the apportionment even of state legislatures has not yet been demonstrated. A number of significant administrative questions remain unanswered,5 and the burden on the federal courts has been substantial. When this has thus far been the outcome of applying the rule to 50 state legislatures, it seems most unwise at this time to extend it to some 80,000 units of local government, whose bewildering variety is sure to multiply the problems which have already arisen and to cast further burdens, of imponderable dimension, on the federal courts. I am frankly astonished at the ease with which the Court has proceeded to fasten upon the entire country at its lowest political levels the strong arm of the federal judiciary, let alone a particular political ideology which has been the subject of wide debate and differences from the beginnings of our Nation.6

There are also convincing functional reasons why the Reynolds rule should not apply to local governmental units at all. The effect of Reynolds was to read a long debated political theory that the only permissible basis for the selection of state legislators is election by majority vote within areas which are themselves equal in population—into the United States Constitution, thereby foreclosing the States from experimenting with legislatures rationally formed in other ways. Even assuming that this result could be justified on the state level, because of the substantial identity in form and function of the state legislatures, and because of the asserted practical necessities for federal judicial interference referred to above, the 'one man, one vote' theory is surely a hazardous generalization on the local level. As has been noted previously, no 'practical necessity' has been asserted to justify application of the rule to local governments. More important, the greater and more varied range of functions performed by local governmental units implies that flexibility in the form of their structure is even more important than at the state level, and that by depriving local governments of this needed adaptability the Court's holding may indeed defeat the very goals of Reynolds.

The present case affords one example of why the 'one man, one vote' rule is especially inappropriate for local governmental units. The Texas Supreme Court held as a matter of Texas law:

'Theoretically, the commissioners court is the governing body of the county and the commissioners represent all the residents, both urban and rural, of the county. But developments during the years have greatly narrowed the scope of the functions of the commissioners court and limited its major responsibilities to the nonurban areas of the county. It has come to pass that the city government * * * is the major concern of the city dwellers and the administration of the affairs of the county is the major concern of the rural dwellers.' 406 S.W.2d 422, 428.

Despite the specialized role of the commissioners court, the majority has undertaken to bring it within the ambit of Reynolds simply by classifying it as 'a unit of local government with general responsibility and power for local affairs.' See ante, at 483. Although this approach is intended to afford 'equal protection' to all voters in Midland County, it would seem that it in fact discriminates against the county's rural inhabitants. The commissioners court, as found by the Texas Supreme Court, performs more functions in the area of the county outside Midland City than it does within the city limits. Therefore, each rural resident has a greater interest in its activities than each city dweller. Yet under the majority's formula the urban residents are to have a dominant voice in the county government, precisely proportional to their numbers, and little or no allowance may be made for the greater stake of the rural inhabitants in the county government.

This problem is not a trivial one and is not confined to Midland County. It stems from the fact that local governments, unlike state governments, are often specialized in function.7 Application of the Reynolds rule to such local governments prevents the adoption of apportionments which take into account the effect of this specialization, and therefore may result in a denial of equal treatment to those upon whom the exercise of the special powers has unequal impact. Under today's decision, the only apparent alternative is to classify the governmental unit as other than 'general' in power and responsibility, thereby, presumably, avoiding application of the Reynolds rule. Neither outcome satisfies Reynolds' avowed purpose: to assure 'equality' to all voters. The result also deprives localities of the desirable option of establishing slightly specialized, elective units of government, such as Texas' county commissioners court, and varying the size of the constituencies so as rationally to favor those whom the government affects most. The majority has chosen explicitly to deny local governments this alternative by rejecting even the solution of the Texas Supreme Court, which held that the present county apportionment was impermissible but would have allowed the new apportionment to reflect factors related to the special functions of the county commissioners court, such as 'land areas, geography, miles of county roads and taxable values,' 406 S.W.2d, at 428, as well as population.

Despite the majority's declaration that it is not imposing a 'straitjacket' on local governmental units, see ante, at 485, its solution is likely to have other undesirable 'freezing' effects on local government. One readily foreseeable example is in the crucial field of metropolitan government. A common pattern of development in the Nation's urban areas has been for the less affluent citizens to migrate to or remain within the central city, while the more wealthy move to the suburbs and come into the city only to work.8 The result has been to impose a relatively heavier tax burden upon city taxpayers and to fragmentize governmental services in the metropolitan area.9 An oft-proposed solution to these problems has been the institution of an integrated government encompassing the entire metropolitan area.10 In many instances, the suburbs may be included in such a metropolitan unit only by majority vote of the voters in each suburb.11 As a practical matter, the suburbanites often will be reluctant to join the metropolitan government unless they receive a share in the government proportional to the benefits they bring with them and not merely to their numbers.12 The city dwellers may be ready to concede this much, in return for the ability to tax the suburbs. Under the majority's pronouncements, however, this rational compromise would forbidden: the metropolitan government must be apportioned solely on the basis of population if it is a 'general' government.

These functional considerations reinforce my belief that the 'one man, one vote' rule, which possesses the simplistic defects inherent in any judicially imposed solution of a complex social problem,13 is entirely inappropriate for determining the form of the country's local governments.

No better demonstration of this proposition could have been made than that afforded by the admirable analysis contained in the dissenting opinion of my Brother FORTAS. But, with respect, my Brother's projected solution of the matter is no less unsatisfactory. For it would bid fair to plunge this Court into an avalanche of local reapportionment cases with no firmer constitutional anchors than its own notions of what constitutes 'equal protection' in any given instance.

With deference, I think that the only sure-footed way of avoiding, on the one hand, the inequities inherent in today's decision, and on the other, the morass of pitfalls that would follow from my Brother FORTAS' approach, is for this Court to decline to extend the constitutional experiment of Reynolds, and to leave the structuring of local governmental units to the political process where it belongs.

Mr. Justice FORTAS, dissenting.

I would dismiss the writ in this case as improvidently granted. The Texas Supreme Court held the districting scheme unlawful under the Texas Constitution. It ordered redistricting. In this difficult and delicate area I would await the result of the redistricting so that we may pass upon the final product of Texas' exercise of its governmental powers, in terms of our constitutional responsibility, and not upon a scheme which Texas itself has invalidated.1

The Court's opinion argues (ante, at 478, n. 2) that the Texas Supreme Court's order is a final judgment because it contemplates no further proceedings in the Texas courts, although it holds the present districting unlawful and requires the Commissioners Court to redistrict. I do not reach this point.

The Court acts now to superimpose its own formula because it disagrees with the standard for redistricting that the Texas Supreme Court states. That standard directed redistricting on the basis of the 'number of qualified voters, land areas, geography, miles of county roads and taxable values.' 406 S.W.2d 422, 428. This standard may or may not produce a result which this Court or I would find constitutionally acceptable. We cannot know in advance how the melange of factors stated by the Texas court would emerge from the mixing machine of the Texas authorities who would deal with the problem. It is clear that the extreme imbalance now prevailing would be eliminated, because the Texas Supreme Court has held it unconstitutional. It might be that the substitute finally worked out would be such that a majority of this Court would not reject it as a denial of equal protection of the laws. After all, at the last Term of this Court, we accepted as passing the scrutiny of the Constitution, the less-than-mathematically perfect plans in Dusch v. Davis, 387 U.S. 112, 87 S.Ct. 1554, 18 L.Ed.2d 656 (1967), and Sailors v. Board of Education of Kent County, 387 U.S. 105, 87 S.Ct. 1549, 18 L.Ed.2d 650 (1967).

The Court, however, now plunges to adjudication of the case of Midland County, Texas, in midstream, apparently because it rejects any result that might emerge which deviates from the literal thrust of one man, one vote. Since it now adopts this simplistic approach, apparently the majority believes that it might as well say so and save Texas the labor of devising an answer.

I am in fundamental disagreement. I believe, as I shall discuss, that in the circumstances of this case equal protection of the laws may be achieved—and perhaps can only be achieved—by a system which takes into account a complex of values and factors, and not merely the arithmetic simplicity of one equals one. Dusch and Sailors were wisely and prudently decided. They reflect a reasoned, conservative, empirical approach to the intricate problem of apply constitutional principle to the complexities of local government. I know of no reason why we now abandon this reasonable and moderate approach to the problem of local suffrage and adopt an absolute and inflexible formula which is potentially destructive of important political and social values. There is no reason why we should insist that there is and can be only one rule for voters in local governmental units—that district for units of local government must be drawn solely on the basis of population. I believe there are powerful resons why, while insisting upon reasonable regard for the population-suffrage ratio, we should reject a rigid, theoretical, and authoritarian approach to the problems of local government. In this complex and involved area, we should be careful and conservative in our application of constitutional imperatives, for they are powerful.

Constitutional commandments are not surgical instruments. They have a tendency to hack deeply—to amputate. And while I have no doubt that, with the growth of suburbia and exurbia, the problem of allocating local government functions and benefits urgently requires attention, I am persuaded that it does not call for the hatchet of one man, one vote. It is our duty to insist upon due regard for the value of the individual vote but not to ignore realities or to bypass the alternatives that legislative alteration might provide.

I agree that application of the Equal Protection Clause of the Constitution, decreed by this Court in the case of state legislatures, cannot stop at that point. Of course local governmental units are subject to the commands of the Equal Protection Clause. Cooper v. Aaron, 358 U.S. 1, 17, 78 S.Ct. 1401, 1408, 3 L.Ed.2d 5 (1958). That much is easy. The difficult question, and the one which the Court slights, is: What does the Equal Protection Clause demand with regard to local governmental units?

Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964), stands for the general proposition that the debasement of the right to vote through malapportionment is offensive to the Equal Protection Clause. It holds that where the allegedly debased vote relates to the State Legislature, a judicial remedy is available to adjudicate a claim of such debasement, and that, subject to some permissible deviation, the remedy is to require reapportionment on a population basis. Although the Court's opinion carefully emphasizes the appropriateness of allowing latitude to meet local and special conditions, 377 U.S., at 577 581, 84 S.Ct., at 1389—1391, its insistence upon the need for general correspondence of voting rights to population has come to be called the one man, one vote rule.2

This rule is appropriate to the selection of members of a State Legislature. The people of a State are similarly affected by the action of the State Legislature. Its functions are comprehensive and pervasive. They are not specially concentrated upon the needs of particular parts of the State or any separate group of citizens. As the Court in Reynolds said, each citizen stands in 'the same relation' to the State Legislature. Accordingly, variations from substantial population equality in elections for the State Legislature take away from the individual voter the equality which the Constitution mandates. They amount to a debasement of the citizen's vote and of his citizenship.3

But the same cannot be said of all local governmental units, and certainly not of the unit involved in this case. Midland County's Commissioners Court has special functions directed primarily to its rural area and rural population. Its powers are limited and specialized, in light of its missions. Residents of Midland County do not by any means have the same rights and interests at stake in the election of the Commissioners. Equal protection of their rights may certainly take into account the reality of the rights and interests of the various segments of the voting population. It does not require that they all be treated alike, regardless of the stark difference in the impact of the Commissioners Court upon them. 'Equal protection' relates to the substance of citizens' rights and interests. It demands protection adapted to substance; it does not insist upon, or even permit, prescription by arbitrary formula which wrongly assumes that the interests of all citizens in the elected body are the same.

In my judgment, the Court departs from Reynolds when it holds, broadly and generally, that 'the Fourteenth Amendment * * * forbids the election of local government officials from districts of disparate population.' Ante, at 478. This holding, literally applied as the Court commands, completely ignores the complexities of local government in the United States—complexities which, Reynolds itself states, demand latitude of prescription. The simplicity of the Court's ruling today does not comport with the lack of simplicity which characterizes the miscellany which constitutes our local governments.

As of the beginning of 1967, there were 81,253 units of local government in the United States. This figure includes 3,049 county governments, 18,051 municipal governments, 17,107 township governments, 21,782 school districts, and 21,264 other special districts.4 These units vary greatly in powers, structure, and function. The citizen is usually subject to several local governments with overlapping jurisdiction.

The Court in this case concedes that in a 'special-purpose unit of government,' the rights of certain constituents may be more affected than the rights of others. It implies that the one man, one vote rule may not apply in such cases. See ante, at 483 484. But it says that we do not here have to confront the implications of such a situation. I do not agree.

I submit that the problem presented by many, perhaps most, county governments (and by Midland County in particular) is precisely the same as those arising from special-purpose units. The functions of many county governing boards, no less than the governing bodies of special-purpose units, have only slight impact on some of their constituents and a vast and direct impact on others. They affect different citizens residing with their geographical jurisdictions in drastically different ways.5

Study of county government leaves one with two clear impressions: that the variations from unit to unit are great; and that the role and structure of county government are currently in a state of flux.6 County governments differ in every significant way: number of constituents, area governed,7 number of competing or overlapping government units within the county,8 form, and means of selection of the governing board,9 services provided,10 the number and functions of independent county officials,11 and sources of revenue.12

Some generalizations can be made about county governments. First, most counties today perform certain basic functions delegated by the State: assessment of property, collection of property taxes, recording of deeds and other documents, maintenance of rural roads, poor relief, law enforcement, and the administration of electoral and judicial functions. Some counties have begun to do more, especially by the assumption of municipal and policy-making functions.13 But most counties still act largely as administrative instrumentalities of the State.14

Second, '(t)he absence of a single chief executive and diffusion of responsibility among numerous independently elected officials are general characteristics of county government in the United States.'15 Those who have written on the subject have invariably pointed to the extensive powers exercised within the geographical region of the county by officials elected on a countywide basis and by special districts organized to perform specific tasks. Often these independent officials and organs perform crucial functions of great importance to all the people within the county.16

These generalizations apply with particular force in this case. The population of Midland County is chiefly in a single urban area.17 That urban area has its own municipal government which, because of home rule,18 has relative autonomy and authority to deal with urban problems. In contrast, the Midland County government, like county governments generally, acts primarily as an administrative arm of the State. It provides a convenient agency for the State to collect taxes, hold elections, administer judicial and peace-keeping functions, improve roads, and perform other functions which are the ordinary duties of the State. The powers of the Commissioners Court, which is the governing body of Midland County, are strictly limited by statute and constitutional provision.19 Although a mere listing of these authorizing statutes and constitutional provisions would seem to indicate that the Commissioners Court has significant and general power, this impression is somewhat illusory because very often the provisions which grant the power also circumscribe its exercise with detailed limitations.

For example, the petitioner cites Art. VIII, § 9, of the Texas Constitution and Article 2352 of the Texas Civil Statutes as granting the Commissioners Court authority to levy taxes. Yet, at the time this suit was tried, Art. VIII, § 9, provided that no county could levy a tax in excess of 80¢ on $100 property valuation. And Article 2352 allocated that 80¢ among the four 'constitutional purposes' mentioned in Art. VIII, § 9 (not more than 25¢ for general county purposes, not more than 15¢ for the jury fund, not more than 15¢ for roads and bridges, and not more than 25¢ for permanent improvements).20

Another example is the authority to issue bonds. It is true, as the majority notes, that the Commissioners Court does have this authority. Yet Title 22 of the Texas Civil Statutes sets up a detailed code concerning how and for what purposes bonds may be issued. Significantly, Article 701 provides that county bonds 'shall never be issued for any purpose' unless the bond issue has been submitted to the qualified property-taxpaying voters of the county.

More important than the statutory and constitutional limitations, the limited power and function of the Commissioners Court are reflected in what it actually does. The record and briefs do not give a complete picture of the workings of the Commissioners Court. But it is apparent that the Commissioners are primarily concerned with rural affairs, and more particularly with rural roads. One Commissioner testified below that the largest item in the county budget was for roads and bridges.21 And, according to that Commissioner, the county does not maintain streets within the City of Midland. The Commissioners seem quite content to let the city council handle city affairs. 'The thing about it is, the city of Midland has the city council and the mayor to run its business, * * * and we have a whole county to run * * *.'

'Theoretically, the commissioners court is the governing body of the county and the commissioners represent all the residents, both urban and rural, of the county. But developments during the years have greatly narrowed the scope of the functions of the commissioners court and limited its major responsibilities to the nonurban areas of the county. It has come to pass that the city government with its legislative, executive and judicial branches, is the major concern of the city dwellers and the administration of the affairs of the county is the major concern of the rural dwellers.' 406 S.W.2d, at 428.

Moreover, even with regard to those areas specifically delegated to the county government by statute or constitutional provision, the Commissioners Court sometimes does not have the power to make decisions. Within the county government there are numerous departments which are controlled by officials elected independently of the Commissioners Court and over whom the Commissioners Court does not exercise control. The Commissioners view themselves primarily as road commissioners. 'The other department heads really have the say in that department. We merely approve the salary. We do not hire anyone in any department in Midland County except the road department. The department heads of the other departments do hire the employees.'22

As the Texas Supreme Court stated, 'the county commissioners court is not charged with the management and control of all of the county's business affairs * * *. (T)he various officials elected by all the voters of the county have spheres that are delegated to them by law and within which the commissioners court may not interfere or usurp.' 406 S.W.2d, at 428. These officials, elected on a direct, one man, one vote, countywide basis, include the Assessor and Collector of Taxes, the County Attorney, the Sheriff, the Treasurer, the County Clerk, and the County Surveyor.23 The County Judge, who is the presiding officer of the Commissioners Court, is also elected on a countywide basis.24 Other county officials and employees are appointed by the Commissioners Court.25

The elected officials are generally residents of the city, probably because of its preponderant vote. A Commissioner testified that 'Every elected official * * * in Midland County today (except the three rural commissioners), and it has been way back for years, has been elected by the people that live here in the city limits of Midland.' Another Commissioner testified that of about 150 employees of the county, only four of those who were not elected lived in the rural precincts. Of all the elected officials only the three rural commissioners lived outside the city limits.26 And, as I have noted, the fifth member of the Commissioners Court, its Chairman, is the County Judge who is elected at large in the county.27 It is apparent that the city people have much more control over the county government than the election of the Commissioners Court would indicate. Many of the county functions which most concern the city, for example, tax assessment and collection, are under the jurisdiction of officials elected by the county at large.28

In sum, the Commissioners Court's functions and powers are quite limited, and they are defined and restricted so that their primary and preponderant impact is on the rural areas and residents. The extent of its impact on the city is quite limited. To the extent that there is direct impact on the city, the relevant powers, in important respects, are placed in the hands of officials elected on a one man, one vote basis. Indeed, viewed in terms of the realities of rights and powers, it appears that the city residents have the power to elect the officials who are most important to them, and the rural residents have the electoral power with respect to the Commissioners Court which exercises powers in which they are primarily interested.

In face of this, to hold that 'no substantial variation' from equal population may be allowed under the Equal Protection Clause is to ignore the substance of the rights and powers involved. It denies—it does not implement—substantive equality of voting rights. It is like insisting that each stockholder of a corporation have only one vote even though the stake of some may be $1 and the stake of others $1,000. The Constitution does not force such a result. Equal protection of the laws is not served by it.

Despite the fact, as I have shown, that many governmental powers in the county are exercised by officials elected at large and that the powers of the Commissioners Court are limited, the Court insists that the Commissioners Court is a unit with 'general governmental powers.' This simply is not so except in the most superficial sense. The Court is impressed by the fact that the jurisdiction of the Commissioners Court extends over the entire area of the county. But this is more form than reality.

Substance, not shibboleth, should govern in this admittedly complex and subtle area; and the substance is that the geographical extent of the Commissioners Court is of very limited meaning. Midland County's Commissioners Court has its primary focus in nonurban areas and upon the nonurban people. True, the county's revenues come largely from the City of Midland. But the Commissioners Court fixes the tax rate subject to the specific limitations provided by the legislature. It must spend tax revenues in the categories and percentages which the legislature fixes. Taxes are assessed and collected, not by it, but by an official elected on a countywide basis. It is quite likely that if the city dwellers were given control of the Commissioners Court, they would reduce the load because it is spent primarily in the rural area. This is a state matter. If the State Legislature, in which presumably the city dwellers are fairly represented 'Reynolds v. Sims), wishes to reduce the load, it may do so. But unless we are ready to adopt the position that the Federal Constitution forbids a State from taxing city dwellers to aid their rural neighbors, the fact that city dwellers pay most taxes should not determine the composition of the county governing body. We should not use tax impact as the sole or controlling basis for vote distribution. It is merely one in a number of factors, including the functional impact of the county government, which should be taken into account in determining whether a particular voting arrangement results in reasonable recognition of the rights and interests of citizens. Certainly, neither tax impact nor the relatively few services rendered within the City of Midland should compel the State to vest practically all voting power in the city residents to the virtual denial of a voice to those who are dependent on the county government for roads, welfare, and other essential services.

I have said that in my judgment we should not decide this case but should give Texas a chance to come up with an acceptable result. Texas' own courts hold that the present system is constitutionally intolerable. The 1963 population estimates relied upon in this case show that the district which includes most of the City of Midland with 67,906 people has one representative, and the three rural districts, each of which has its own representative, have 852; 414; and 828 people respectively. While it may be that this cannot be regarded as satisfying the Equal Protection Clause under any view, I suggest that applying the Court's formula merely errs in the opposite direction: Only the city population will be represented, and the rural areas will be eliminated from a voice in the county government to which they must look for essential services. With all respect, I submit that this is a destructive result. It kills the very value which it purports to serve. Texas should have a chance to devise a scheme which, within wide tolerance, eliminates the gross underrepresentation of the city, but at the same time provides an adequate, effective voice for the nonurban, as well as the urban, areas and peoples.29

Mr. Justice STEWART, dissenting.

I would dismiss the writ as improvidently granted for the reasons stated by Mr. Justice HARLAN and Mr. Justice FORTAS.

Since the Court does reach the merits, however, I add that I agree with most of what is said in the thorough dissenting opinion of Mr. Justice FORTAS. Indeed, I would join that opinion were it not for the author's unquestioning endorsement of the doctrine of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506. I continue to believe that the Court's opinion in that case misapplied the Equal Protection Clause of the Fourteenth Amendment that the apportionment of the legislative body of a sovereign State, no less than the apportionment of a county government, is far too subtle and complicated a business to be resolved as a matter of constitutional law in terms of sixth-grade arithmetic. My views on that score, set out at length elsewhere,* closely parallel those expressed by Mr. Justice FORTAS in the present case.