delivered the delivered theopinion of the Court. This case involves the extent to which the Fourteenth Amendment and the “one man, one vote” principle apply in the election of local governmental officials. Appellants are residents and taxpayers of the Kansas City School District, one of eight separate school' districts that have combined to form the Junior College District of Metropolitan Kansas City. Under Missouri law separate school districts may vote by referendum to establish a consolidated junior college district and elect six trustees to conduct and manage the necessary affairs of that district.1 The state law also provides that these trustees shall be apportioned among the separate school districts on the basis of “school enumeration,” defined as the number of persons between the ages of six and 20 years, who reside in each district.2 In the case of the Kansas City School District this apportionment plan results in the election of three trustees, or 50% of the total number, from that district. Since that district contains approximately 60% of the total school enumeration in the junior college district,3 appel*52lants brought suit claiming that their right to vote for trustees was being unconstitutionally diluted in violation of the, Equal Protection Clause of the Fourteenth Amendment. The Missouri Supreme Court upheld the trial court’s dismissal of the suit, stating that the “one man, one vote” principle was not applicable in this case. 432 S. W. 2d 328 (1968). We noted probable jurisdiction of the appeal, 393 U. S. 1115 (1969), and for the reasons set forth below we reverse and hold that the Fourteenth Amendment requires that the trustees of this junior college district be apportioned in a manner that does not deprive any voter of his right to have his own vote given as much weight, as far as is practicable, as that of any other voter in the junior college district.

In Wesberry v. Sanders, 376 U. S. 1 (1964), we held that the Constitution requires that “as nearly as is practicable one man’s vote in a congressional election is to be worth as much as another’s.” Id., at 7-8. Because of this requirement we struck down a Georgia statute which allowed glaring discrepancies among the populations in that State’s congressional districts. In Reynolds v. Sims, 377 U. S. 533 (1964), and the companion cases,4 we considered state laws that had apportioned state legislatures in a way that again showed glaring discrepancies in the number of people who lived in different legislative districts. In an elaborate opinion in Reynolds we called attention to prior cases indicating that a qualified voter has a constitutional right to vote in elections without having his vote wrongfully denied, debased, or diluted. Ex parte Siebold, 100 U. S. 371 (1880); Ex parte Yarbrough, 110 U. S. 651 (1884); United States v. Mosley, 238 U. S. 383 (1915); Guinn v. United States, 238 U. S. 347 (1915); Lane *53v. Wilson, 307 U. S. 268 (1939); United States v. Classic, 313 U. S. 299 (1941). Applying the basic principle of Wesberry, we therefore held that the various state apportionment schemes denied some voters the right guaranteed by the Fourteenth Amendment to have their votes given the same weight as that of other voters. Finally, in Avery v. Midland County, 390 U. S. 474 (1968), we applied this same principle to the election of Texas county commissioners, holding that a qualified voter in a local election also has constitutional right to have his vote counted with substantially the same weight as that of any other voter a case where the elected officials exercised “general governmental powers over the entire geographic area served by the body.” Id., at 485.

case argue that the junior college trustees exercised general governmental powers over the entire district and that under Avery the State was thus required to apportion the trustees according to popula-on an equal basis, as far as practicable. Appellants that since the trustees can levy and collect taxes, bonds with certain restrictions, hire and fire teachers, contracts, collect fees, supervise and discipline stu-pass on petitions to annex school districts, acquire property by condemnation, and in general manage the operations of the junior college,5 their powers are equiv-for apportionment purposes, to those exercised by county commissioners in Avery. We feel that these while not fully as broad as those of the Midland Commissioners,6 certainly show that the trustees *54perforin important governmental functions within the districts, and we think these powers are general enough and have sufficient impact throughout the district to justify the conclusion that the principle which we applied in Avery should also be applied here.

This Court has consistently held in a long series of cases,7 that in situations involving elections, the States are required to insure that each person’s vote counts as much, insofar as it is practicable, as any other person’s. We have applied this principle in congressional elections, state legislative elections, and local elections. The consistent theme of those decisions is that the right to vote in an election is protected by the United States Constitution against dilution or debasement. While the particular offices involved in these cases have varied, in each case a constant factor is the decision of the government to have citizens participate individually by ballot in the selection of certain people who carry out governmental functions. Thus in the case now before us, while the office of junior college trustee differs in certain respects from those offices considered in prior cases, it is exactly the same in the one crucial factor — these officials are elected by popular vote.

I When a court is asked to decide whether a State is ¡required by the Constitution to give each qualified voter I the same power in an election open to all, there is no dis-I cernible, valid reason why constitutional distinctions I should be drawn on the basis of the purpose of the elec-*55one person’s vote is given less weight through ”j unequal apportionment, his right to equal voting partid- j pation is impaired just as much when he votes for a school | board member as when he votes for a state legislator, f While there are differences in the powers of different f officials, the crucial consideration is the right of each I qualified voter to participate on an equal footing in the j election process. It should be remembered that in cases j like this one we are asked by voters to insure that they | are given equal treatment, and from their perspective the I harm from unequal treatment is the same in any election, | regardless of the officials selected. *4

a election were to be the determining factor in deciding whether voters are entitled equal voting power, courts would be faced with the difficult job of distinguishing between various elections. cannot readily.perceive judicially manageable stand-to aid in such a task. It might be suggested that apportionment is required only in “important” elections, but good judgment and common sense tell that what might be a vital election to one voter well be a routine one to another. In some in-the election of a local sheriff may be far more important than the election of a United States Senator. there is any way of determining the importance of a particular governmental official, we think the of the State to select that official by popular is a strong enough indication that the choice is an one. This is so because in our country pop-election has traditionally been the method followed government by the people is most desired.

we distinguish for appor-purposes between elections for “legislative” and those for “administrative” officers. Such suggestion would leave courts with an equally unman-*56principle since be classified in the neat categories favored by texts,” Avery, supra, at 482, and it must also be rejected. We therefore hold today that as a general whenever a state or local government decides to persons by popular election to perform governmental functions, the Equal Protection Clause of the Fourteenth Amendment requires that each qualified must be given an equal opportunity to participate that election, and when members of an elected are chosen from separate districts, each district must be established on a basis that will insure, far as is practicable, that equal numbers of voters can vote for proportionally equal numbers of officials. It is course possible that there might be some case in which State elects certain functionaries whose duties are so removed from normal governmental activities and disproportionately affect different groups that a popu-election in compliance with Reynolds, supra, might not be required, but certainly we see nothing in the present case that indicates that the activities of these trustees fit in that category. Education has traditionally been a vital governmental function, and these trustees, whose election the State has opened to all qualified voters, are governmental officials in every relevant sense of that term.

In this particular case the man, one ciple is to some extent already reflected in the Missouri statute. That act provides that if no one or more of the component school districts has 33%% or more of the total enumeration of the junior college district, then all six trustees are elected at large. If, however, one or more districts has between 33%% and 50% of the total enumeration, each such district elects two trustees and the rest are elected at large from the remaining districts. *57Similarly, if one district has between 50% and 66%% of the enumeration it elects three trustees, and if one district has more than 66%% it elects four trustees.8 This scheme thus allocates increasingly more trustees to large districts as they represent an increasing proportion of the total enumeration.

statutory scheme reflects to some extent a principle of equal voting power, it does so in a way that does not comport with constitutional requirements. This is so because the Act necessarily results in a systematic discrimination against voters in the more populous school districts. This discrimination occurs because whenever a large district’s percentage of the total enumeration falls within a certain percentage range it is always allocated the number of trustees corresponding the bottom of that range. Unless a particular large district has exactly 33%%■, 50%, or 66%% of the total enumeration it will always have proportionally fewer trustees than the small districts. As has been pointed out, in the case of the Kansas City School District approximately 60% of the total enumeration entitles that district to only 50% of the trustees. Thus while voters large school districts may frequently have less effective voting power than residents of small districts, they can never have more. Such built-in discrimination against voters in large districts cannot be sustained as a sufficient compliance with the constitutional mandate that each person’s vote count as much as another’s, as far as practicable. Consequently Missouri cannot allocate the college trustees according to the statutory formula employed in this case.9 We would be faced with a dif-*58question if the deviation presented in this case resulted from a plan that not contain a built-in bias in favor of small districts, rather from the inherent mathematical complications equally apportioning a small number of trustees among a limited number of component districts. We said before that mathematical exactitude is not required, Wesberry, supra, at 18, Reynolds, supra, at 577, a plan that does not automatically discriminate in favor of certain districts is.

In holding that the guarantee each voter applies in all elections of governmental officials, we do not feel that the States will be inhibited finding ways to insure that legitimate political goals representation are achieved. We have previously up-against constitutional challenge an election scheme required that candidates be residents of certain districts that did not contain equal numbers of people. Dusch v. Davis, 387 U. S. 112 (1967). Since all the officials in that case were elected at large, the right of voter was given equal treatment.10 We have also that where a State chooses to select members of an official body by appointment rather than election, and that choice does not itself offend the Constitution, the fact that each official does not “represent” the same number of people does not deny those people equal protection of the laws. Sailors v. Board of Education, 387 U. S. 105 (1967); cf. Fortson v. Morris, 385 U. S. 231 (1966). And a State may, in certain cases, limit the *59a group or class of people. As we said before, “[vjiable local governments may need many innovations, numerous combinations of old and new devices, great flexibility in municipal arrangements to meet changing urban conditions. We see nothing in the Constitution to prevent experimentation.” Sailors, supra, at 110-111. But once a State has decided use the process of popular election and “once the class voters is chosen and their qualifications specified, we see no constitutional way by which equality of voting power may be evaded.” Gray v. Sanders, 372 U. S. 368, 381 (1963).

judgment below reversed and the case is remanded to the Missouri Supreme Court for proceedings not inconsistent with opinion.

and remanded.

dissenting.

demonstrates, to a degree that no case has, the pervasiveness of the federal judicial intrusion into state electoral processes that was unleashed the “one man, one vote” rule of Reynolds v. Sims, 377 U. S. 533 (1964).

for the apportionment state legislatures, thereby denying States the right take into account in the structuring of their legisla-any historical, geographical, economic, or social considerations, or any of the many other practical and factors that have always been recognized as play-a legitimate part in the practice of politics.

Four years later, in Averyv. Midland County, 390 S. 474 (1968), the “one man, one vote” rule was to many kinds of local governmental units, affecting to an unknown extent the organi-*60integrity of some the country, and constricting the States in the use the electoral process in the establishment of new ones.

And applicable to the various boards of trustees of Mis-junior college system, and the case forebodes, if it does not decide, that the rule is to be applied every elective public body, no matter what its nature.

While I my continued disagreement with them as constitutional holdings (see my dissenting opinions in Reynolds, 377 U. S., at 589, and in Avery, 390 U. S., at 486) — I not think that either of these cases, or any other in this justifies the present decision. I therefore dissent, taking off from Avery in what is about to be said.

I

In Avery the Court acknowledged that "the states' varied, pragmatic approach in establishing governments” has produced “a staggering number” of local governmental units. The Court noted that, “while special-purpose organizations abound . . . , 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.” The Midland County Commissioners Court, the body whose composition was challenged in Avery, was found to possess a broad range of powers that made it “representative of most of the general governing bodies of American cities, counties, towns, and villages,” and the Court was at pains to limit its holding to such general bodies. 390 U. S., at 482-485. Today the Court discards that limitation, stating that “there is no discernible, valid reason why constitutional distinctions should be drawn on the basis of the purpose of the election.” Ante, at 54-55. I believe, to the contrary, that the need to preserve flexibility in the design of local *61serve functions and that must meet particular local conditions, furnishes a powerful reason to refuse to extend the Avery ruling beyond its original limits. If local units having general governmental powers are to be considered, like state legislatures, as having a substantial identity of function that justifies imposing on them a uniformity of elective structure, it is clear that specialized local entities are characterized by precisely the opposite of such identity. From irrigation districts to air pollution control agencies to school districts, such units vary in the magnitude of their impact upon various constituencies and in the manner in which the benefits and burdens of their operations interact with other elements of the local political and economic picture. Today’s ruling will forbid these agencies from adopting electoral mechanisms that take these variations into account.

an arbitrary limitation on the ways in which local agencies may be constituted. The Court concedes that the States may use means other than apportionment “to insure that legitimate political goals of representation are achieved.” For example, officials elected at large may be required to be residents of particular areas that do not contain equal numbers of people, Dusch v. Davis, 387 U. S. 112 (1967); right to vote may be denied outright to persons whose interest in the function performed by the agency is nonexistent or slight, cf. Kramer v. Union Free School District, 395 U. S. 621 (1969); Cipriano v. City of Houma, U. S. 701 (1969); or the State may in many instances abandon the elective process altogether and allow members of an official body to be appointed, without any for the equal-population principle, Sailors v. Board Education, 387 U. S. 105 (1967). Since the Court recognizes the States’ need for flexibility in structuring units, I am unable to see any basis for its selectively *62denying to them one of the means flexibility. If, as the Court speculates, other means will as effective as apportionment in the adaptation local agencies to meet specific needs, presumably those means will also enable the States just as effectively accomplish whatever evils the Court thinks it is preventing by today’s decision. The Court has not shown under the supervision of state legislatures that are apportioned according to Reynolds, flexible methods of apportionment of local official bodies carry any greater danger of abuse than these other means of achieving the desirable goal of specialization. The Court’s imposition this arbitrary limitation on the States can be justified only in the name of mathematical nicety.

I do not believe that, even compelled by the absence of “judicially manageable standards” for the “difficult job of distinguishing between various elections.” Ante, at 55. Before today, the Court's rule was that “one man, one vote” applied only local bodies having “general governmental powers over the entire geographic area served by the body.” 390 U. S., at 485. The Court in Avery professed no temerity about concluding that the Midland County Commissioners Court was such a body. The Court’s mere recitation of the powers of that entity, ante, at 53-54, n. 6, suffices to establish that conclusion. At the same time, it cannot be argued seriously that the Junior College District of Metropolitan Kansas City is the general governing body for the people of its area. The mere fact that the trustees can, with restrictions, levy taxes, issue bonds, and condemn property for school purposes does not detract from the crucial consideration that the sole purpose for which the district exists is the operation of a junior college. If the Court adhered to the Avery line, marginal cases would of course arise in which the courts would face difficulty in determining whether a particular *63governmental powers, but such a determination would be no different in kind from many other matters of degree upon which courts must continually pass. The importance of ensuring flexibility in the organization of specialized units of government, and the uncertainty whether the rule announced today will further any important countervailing interest, convince me that the Court should not proceed further into the political thicket than it has already gone in Avery.

II

case afford a clear indication of the extent to which reasonable state objectives are to be sacrificed on the altar of numerical equality. We are not faced with an apportionment scheme that is a historical relic, with no present-day justification, or one that reflects the stranglehold of a particular group that, having once attained power, blindly resists a redistribution. The structure of the Junior College District of Metropolitan Kansas City is based upon a state statute enacted 1961. Prior to that date, the individual school boards the power to create their own junior colleges, as they do, but there was apparently no authorization for cooperation among districts. The 1961 statute was enacted out of concern on the part of the legislature that Missouri’s public educational facilities were not expand-at a satisfactory rate, see Three Rivers Junior College District v. Statler, 421 S. W. 2d 235, 237 (Mo. 1967).1 *64The provisions of the statute evidence a legislative determination of the most effective means to encourage expansion through cooperation between districts.

The statutory provision for election of the six-man board of trustees, summarized by the Court, reflects a careful balancing of the desirability of population-based representation against the practical problems involved in the creation of new educational units. The statute does not by its own force create any junior college districts; this is left to the initiative of the residents of particular areas who are interested in providing public junior-college education for their children. In recognition of the fact that individual school districts may lack the funds or the population to support a junior college of their own, the state legislature has authorized them to make voluntary arrangements with their neighbors for joint formation of a junior college district. If one of the cooperating school districts greatly preponderates in size, it enters into the arrangement knowing that its representation on the board of trustees, while large, will be somewhat smaller than it would be if based strictly on relative school enumeration.

The features of this system are surely sensibly designed to facilitate creation of new educational bodies while guaranteeing to small school districts that they will not be entirely swallowed up by a large partner. The small districts are free to avoid alliance with a highly populated neighbor, if they prefer to link with enough others of their own size to provide a viable base for a junior college. At the same time, a very large school district is probably capable of forming a junior college on its own if it prefers not to consolidate, on the terms set by statute, with smaller neighbors. On the other hand, large and small districts may work together if they find this the *65most beneficial arrangement.2 The participation, as here, of seven smaller and one larger school district in the joint formation of a junior college district, represents a pragmatic choice by all concerned from among a number of possible courses of action.

that such a voluntary arrangement effects an unconstitutional “dilution” of the votes of residents of the largest school district. When the Court, in Reynolds, rejected a proposed analogy between state legislatures and the Federal Congress, it relied heavily on the fact that state legislative districts “are merely involuntary political units of the State created by statute to aid in the administration of state government.” 377 U. S., at 548. In contrast, the National Government was created by the union of “a group of formerly independent States.” The system of representation in Congress was “conceived out of compromise and concession” between the larger and smaller States. Id., at 574. The system struck down today shares much of this same character of voluntary compromise. It is true that the analogy would be even closer if the legislature had left the school districts free negotiate their own apportionment terms, rather than imposing a uniform scale; but as I read the Court’s opinion today, it would strike down the apportionment this case even if the terms had resulted from an entirely free agreement among the eight school districts. Insistence upon a simplistic mathematical formula as measure of compliance with the Equal Protection *66Clause in cases involving the electoral process has resulted in this instance in a total disregard of the salutary-purposes underlying the statutory scheme.

III

Finally, I find particularly perplexing the portion of the Court’s opinion explaining why the apportionment involved in this case does not measure up even under the “one man, one vote” dogma. The Court holds that the voters of the Kansas City School District, who elect 50% of the trustees, are denied equal protection of the laws because that district contains about 60% of the school enumeration. This is so because the statutory formula embodies a “built-in discrimination against voters in large districts.” Ante, at 57. The Court seems to suggest that the same discrepancy among districts might pass muster if it could be shown to be mathematically unavoidable in the apportionment of the small number of trustees among the component districts; but the discrepancy is not permissible where it simply reflects the legislature’s choice of a means to foster a legitimate state goal. This reasoning seems hard to follow and also disturbing on two scores.

First, to apply the rule with such rigor to local governmental units, especially single-function units, is to disregard the characteristics that distinguish such units from state legislatures. As I noted in my dissent in Avery, 390 U. S., at 488-490, there is a much smaller danger of abuse through malapportionment in the case of local units because there exist avenues of political redress that are not similarly available to correct mal-apportionment of state legislatures. Further, as noted above, the greater diversity of functions performed by local governmental units creates a greater need for flexi*67bility in their structure. If these considerations are inadequate to stave off the extension of the Reynolds rule to units of local government, they at least provide a persuasive rationale for applying that rule so as to allow local governments much more play in the joints.

Such an approach is not foreclosed by the previous cases. In Reynolds, 377 U. S., at 577-581, the Court catalogued a number of considerations indicating that “[s]omewhat more flexibility” might be permissible in state legislative apportionment than in congressional districting. Compare Swann v. Adams, 385 U. S. 440 (1967), with Kirkpatrick v. Preisler, 394 U. S. 526 (1969), and Wells v. Rockefeller, 394 U. S. 542 (1969). The need for more flexibility becomes greater as we proceed down the spectrum from the state legislature to the single-purpose local entity.

The disparities of representation in Avery were of an. entirely different order from those here. In that case, each of the four districts elected one commissioner to the Comissioners Court, despite the fact that the population of one district was 67,906, while those of the remaining three were 852, 414, and 828. I think that the Avery rule, born in an extreme case, is being applied here with a rigidity that finds no justification in the considerations that gave it birth. Cf. Wells v. Rockefeller, 394 U. S., at 553 (White, J., dissenting). In this case, the disparity of representation is relatively minor. Even more important, it is not an unexplained unjustified deviation from equality, see Swann v. Adams, 385 U. S., at 445-446, but reflects an enlightened state policy of encouraging individual school districts join together voluntarily to expand the State’s public junior college facilities.

*68Second, the Court leaves unexplored the premises underlying its conclusion that the apportionment here does not achieve equality, “as far as practicable.” Ante, at 57. Missouri is forbidden to use the statutory formula employed in this case because the percentage categories it creates will, in particular instances, only approximate equality, and because whatever discrepancy exists will always favor residents of the smaller districts. The Court does not suggest how a formula could be devised that would provide a general rule for application to all the various junior college districts but would not share these alleged faults. If a large district falling within a given percentage range were allocated the number of trustees corresponding to the top, rather than the bottom, of the range, that would also produce, on the Court’s theory, a “built-in discrimination” against voters in small districts.

Thus, the result of the Court’s holding may be that Missouri is forbidden to establish any formula of general application for apportionment of trustees, but must instead provide for the improvisation of an individual apportionment scheme for each junior college district after the contours of the district have been settled. But surely a State could reasonably determine that the mechanics of operating such a system would be so unduly burdensome that it would be better to apportion according to a statewide formula. Would not such considerations justify a conclusion that the statewide formula achieves equality “as far as practicable”? While the Court does not discuss the problem, its invalidation of this statutory formula seems to be based on the premise that such practical considerations, like a State’s desire to encourage cooperation among districts, are constitutionally inadequate to justify any divergence from voting “equality.”

*69not, however, spell out any rationale for concluding that such matters of administrative convenience deserve no weight in determining what is “practicable.” This is especially incongruous in light of the Court’s unexplained conclusion that deference can be be given to legislative determinations that the boards should have a small number of trustees and that the trustees in some instances should represent component school districts. Why does the Court not require that the number of trustees be increased from six, in order to reduce the roughness with which equality is approximated? Would a three-man board be unconstitutionally small? Why is the Court willing to accept inequality that derives from a desire to give representation to component school districts, when similar inequality in state legislative districting could probably not be justified by desire to give representation to counties? Cf. Reynolds Sims, 377 U. S., at 579-581; Swann v. Adams, 385 U. S., at 444. If equality cannot be achieved when representation is by component districts, why does the “as far as practicable” standard not require at-large election of trustees? Is there something about these considerations that gives them a status under the Equal Protection Clause that is not possessed by a legislative desire to apportion by a formula of statewide application?

It seems to me that beneath the surface of the Court’s opinion lie unspoken answers to these and other similar questions, questions that I can characterize only as mat-of political judgment. The Court’s adoption of a rigid, mathematical rule turns out not to have saved from having to balance and judge political considerations, concluding that one does merit some weight in apportionment scheme while another does not. The that the courts, rather than the legislatures, now the final arbiters of such matters will continue, I fear, *70after the present decision to be the inevitable consequence of the shallow approach to the Equal Protection Clause represented by the “one man, one vote” theory. The Court could at least lessen the disruptive impact that approach at the local level by approving this relatively minor divergence from strict equality on the ground that the legislature could reasonably have concluded that it was necessary to accomplish legitimate state interests.

I would affirm the judgment of the Supreme Court of Missouri. What our Court has done today seems to me to run far afield of the values embodied in the scheme of government ordained by the Constitution.

dissenting.

I concur fully in the opinion of Mr. Justice Harlan. add this comment to emphasize the subjective quality a doctrine of constitutional law that has as its primary standard “a general rule, [that] whenever a state local government decides to select persons by popular election . . . ,” the Constitution commands that each qualified voter must be given a vote which is equally weighted with the votes cast by all other electors.

The failure to provide guidelines for the Court’s “general rule” is to be applied is exacerbated when the Court implies that the stringent standards of “mathematical exactitude” that are controlling in apportionment of federal congressional districts need not be applied to smaller specialized districts such as the junior college district in this case. This gives added relevance to Mr. Justice Harlan’s observation that “[t]he need for more flexibility becomes greater as we proceed down the spectrum from the state legislature to the single-purpose local entity.” Ante, at 67. Yet the Court has given almost no indication of which non-*71may or may not legitimately be considered by a legislature in devising a constitutional apportionment scheme for a local, specialized unit of government.

can finally apply these “general rules” but in the interim all other judges must speculate as best they can when and how to apply them. With all deference I suggest the Court’s opinion today fails to give any meaningful guidelines.