delivered the opinion of the Court.
This case concerns the congressional reapportionment of the State of Texas.
On June 17, 1971, the Governor of the State of Texas signed into law Senate Bill One (S. B. 1), Tex. Acts, 62d Leg., 1st Called Sess., c. 12, p. 38, providing for the congressional redistricting of the State. S. B. 1 divided the State into 24 congressional districts for the ensuing decennium.1 Based upon 1970 census figures, absolute *785population equality among the 24 districts would mean a population of 466,530 in each district. The districts created by S. B. 1 varied from a high of 477,856 in the 13th District to a low of 458,581 in the 15th District. The 13th District exceeded the ideal district by 2.43% and the 15th District was smaller by 1.7%. The population difference between the two districts was 10,275 persons, and their total percentage deviation was 4.13%. The ratio of the 13th District to the 15th was 1.04 to 1. The average deviation of all districts from the ideal district of 466,530 was .745% or 3,421 persons.2
*786On October 19, 1971, appellees, residents of the 6th, 13th, 16th, and 19th congressional districts, filed suit in the United States District Court for the Northern District of Texas against appellant, the Secretary of State of Texas and the chief election officer of the State. Ap-pellees alleged that the reapportionment of the Texas congressional seats as embodied in S. B. 1 violated their rights under Art. I, § 2, and the Equal Protection Clause of the Fourteenth Amendment.3 They requested an injunction against the use of S. B. 1, an order requiring a new apportionment or the use of a plan submitted with their complaint, or at-large elections. The plan appended to appellees’ original complaint, which came to be called Plan B, generally followed the redistricting pattern of S. B. 1. However, the district lines were adjusted where necessary so as to achieve smaller population variances among districts. Plan B created districts varying from 466,930 to 466,234, for a total absolute deviation between the largest and smallest district of 696 persons. District 12 exceeded the ideal by .086% and District Four was under the ideal by .063%, for a total percentage deviation of .149%. Although the plan followed the district lines of S. B. 1 where possible, in order to achieve maximum population equality, Plan B cut across 18 more county lines than did S. B. I.4
*787A three-judge court was convened. 28 U. S. C. §§ 2281, 2284. On January 10, 1972, several days prior to the scheduled hearing of the case, appellees filed an amended complaint suggesting an alternative plan, which came to be called Plan C. Plan C, unlike Plan B, substantially disregarded the configuration of the districts in S. B. 1. Instead, as the authors of the plan frankly admitted and the District Court found, Plan C represented an attempt to attain lower deviations without regard to any consideration other than population. The districts in Plan C varied in population from 467,173 as a high to 465,855 as a low, a difference of 1,318 persons. The largest district was overpopulated by .139%, and the smallest underpopulated by .145%, the total percentage deviation being .284%. Plan C had 14 districts with greater deviations than Plan B, eight districts with deviations *788equal to those found in Plan B, and two districts with deviations smaller than those in Plan B.5
On January 21, 1972, the District Court heard argument and received into evidence various depositions. The next day, the court announced its decision. Relying upon this Court’s decision in Kirkpatrick v. Preisler, 394 U. S. 526 (1969), the District Court declared S. B. 1 unconstitutional and enjoined appellant from “conducting or permitting any primary or general elections based upon the districts established by S. B. 1.” The District-Court ordered the adoption of Plan C as “the plan of this Court for the congressional districts of the State *789of Texas.” 6 Noting that its order was entered “without prejudice to the legislative and executive branches of the State of Texas to proceed with the consideration and adoption of any other constitutionally permissible plan of congressional redistricting at a called or regular session of the Legislature,” the District Court retained jurisdiction “for the purposes of considering any such plan which might be adopted by the Legislature of the State of Texas until congressional reapportionment is enacted based on the Twentieth Decennial Census to be conducted in 1980.” 7
This Court, on application of appellant, granted a stay of the order of the District Court. 404 TJ. S. 1065 (1972). The 1972 congressional elections were therefore conducted under the plan embodied in S. B. 1. We noted probable jurisdiction of the appeal. 409 U. S. 947 (1972).
*790I
The command of Art. I, § 2, that representatives be chosen “by the People of the several States” was elucidated in Wesberry v. Sanders, 376 U. S. 1 (1964), and in Kirkpatrick v. Preisler, 394 U. S., at 527-528, to permit only those population variances among congressional districts that “are unavoidable despite a good-faith effort to achieve absolute equality, or for which justification is shown.” Id., at 531.8 See also Wells v. Rockefeller, 394 U. S. 542, 546 (1969). Kirkpatrick and Wells invalidated state reapportionment statutes providing for federal congressional districts having total percentage deviations of 5.97% and 13.1%, respectively. In both cases, we concluded that the deviations did not demonstrate a good-faith effort to achieve absolute equality and were not sufficiently justified.
The percentage deviations now before us in S. B. 1 are smaller than those invalidated in Kirkpatrick and Wells, but we agree with the District Court that, under the standards of those cases, they were not “unavoidable,” and the districts were not as mathematically equal as reasonably possible. Both Plans B and C demonstrate this much, and the State does not really dispute it.'9 *791Also, as in Kirkpatrick and Wells, “we do not find legally acceptable the argument that variances are justified if they necessarily result from a State’s attempt to avoid fragmenting political subdivisions by drawing congressional district lines along existing county, municipal, or other political subdivision boundaries.” Kirkpatrick v. Preisler, supra, at 533-534.
The State asserts that the variances present in S. B. 1 nevertheless represent good-faith efforts • by the State to promote “constituency-representative relations,”10 a policy frankly aimed at maintaining existing relationships between incumbent congressmen and their constituents and preserving the seniority the members of the State’s delegation have achieved in the United States House of Representatives. We do not disparage this interest. We have, in the context of state reapportionment, said that the fact that “district boundaries may have been drawn in a way that minimizes the number of contests between present incumbents does not in and of itself establish invidiousness.” Burns v. Richardson, 384 U. S. 73, 89 n. 16 (1966). Cf. Gaffney v. Cummings, ante, at 752. But we need not decide whether this state interest is sufficient to justify the deviations at issue here, for Plan B admittedly serves this purpose as well *792as S. B. 1 while adhering more closely to population equality.11 S. B. 1 and its population variations, therefore, were not necessary to achieve the asserted state goal, and the District Court was correct in rejecting it.12
Appellant also straightforwardly argues that Kirkpatrick and Wells should be modified so as not to require the “small” population variances among congressional districts involved in this case to be justified by the State. S. B. 1, it is urged, absent proof of invidiousness over and above the population variances among its districts, does not violate Art. I, § 2. It is clear, however, that at some point or level in size, population variances do import invidious devaluation of the individ*793ual's vote and represent a failure to accord him fair and effective representation. Appellant concedes this and would locate the line differently than the Court did in Kirkpatrick and Wells. Keeping in mind that congressional districts are not so intertwined and freighted with strictly local interests as are state legislative districts and that, as compared with the latter, they are relatively enormous, with each percentage point of variation representing almost 5,000 people, we are not inclined to disturb Kirkpatrick and Wells. This is particularly so in light of Mahan v. Howell, 410 U. S. 315 (1973), decided earlier this Term, where we reiterated that the Wesberry, Kirkpatrick, and Wells line of cases would continue to govern congressional reapportionments, although holding that the rigor of the rule of those cases was inappropriate for state reapportionments challenged under the Equal Protection Clause of the Fourteenth Amendment.
II
The District Court properly rejected S. B. 1, but it had before it both Plan B and Plan C, and there remains the question whether the court correctly chose to implement the latter.13 Plan B adhered to the basic district configurations found in S. B. 1, but adjusted the district lines, where necessary, in order to achieve maximum population equality among districts. Each district in Plan B contained generally the same counties as the equivalent district in S. B. I.14 Plan C, on the other hand, was based entirely upon population considerations *794and made no attempt to adhere to the district configurations found in S. B. I.15 Both plans were submitted to the District Court by appellees. After deciding that S. B. 1 was unacceptable, the District Court ordered the implementation of Plan C. In announcing its decision, the court said only:
“Plan C is based solely on population and is significantly more compact and contiguous than either S. B. 1 or Plan B. . . . The Court has considered Plans B and C . . . and concludes that Plan C best effectuates the principle of 'one man, one vote’ enunciated by the Supreme Court.”
Appellant argues that, even if the District Court properly struck down S. B. 1, it should have selected Plan B rather than Plan C. Appellees defend the selection of Plan C as an exercise of the remedial discretion of the District Court, although in doing so they argue against a plan that they proposed and frequently urged upon the District Court.
From the beginning, we have recognized that “reapportionment is primarily a matter for legislative consideration and determination, and that judicial relief becomes appropriate only when a legislature fails to reapportion according to federal constitutional requisites *795in a timely fashion after having had an adequate opportunity to do so.” Reynolds v. Sims, 377 U. S. 533, 586 (1964). See also, id., at 584, 586-587; id., at 588-589 (opinion of Stewart, J.). We have adhered to the view that state legislatures have “primary jurisdiction” over legislative reapportionment. See Maryland Committee for Fair Representation v. Tawes, 377 U. S. 656, 676 (1964); Davis v. Mann, 377 U. S. 678, 693 (1964); Roman v. Sincock, 377 U. S. 695, 709-710, 711-712 (1964); Burns v. Richardson, 384 U. S., at 84-85; Ely v. Klahr, 403 U. S. 108, 114 (1971); Whitcomb v. Chavis, 403 U. S. 124, 160-161 (1971); Sixty-seventh Minnesota State Senate v. Beens, 406 U. S. 187, 195-201 (1972); Mahan v. Howell, 410 U. S., at 327. Just as a federal district court, in the context of legislative reapportionment, should follow the policies and preferences of the State, as expressed in statutory and constitutional provisions or in the reapportionment plans proposed by the state legislature, whenever adherence to state policy does not detract from the requirements of the Federal Constitution, we hold that a district court should similarly honor state policies in the context of congressional reapportionment. In fashioning a reapportionment plan or in choosing among plans, a district court should not pre-empt the legislative task nor “intrude upon state policy any more than necessary.” Whitcomb v. Chavis, supra, at 160.
Here, it is clear that Plan B, to a greater extent than did Plan C, adhered to the desires of the state legislature while attempting to achieve population equality among districts. S. B. 1, a duly enacted statute of the State of Texas, established the State’s 24 congressional districts with locations and configurations found appropriate by the duly elected members of the two houses of the Texas Legislature. As we have often noted, reapportionment is a complicated process. Districting *796inevitably has sharp political impact and inevitably political decisions must be made by those charged with the task. See Gaffney v. Cummings, ante, at 753. Here those decisions were made by the legislature in pursuit of what were deemed important state interests. Its decisions should not be unnecessarily put aside in the course of fashioning relief appropriate to remedy what were held to be impermissible population variations between congressional districts.
Plan B, as all parties concede, represented an attempt to adhere to the districting preferences of the state legislature while eliminating population variances. Indeed, Plan B achieved the goal of population equality to a greater extent than did Plan C. Despite the existence of Plan B, the District Court ordered implementation of Plan C, which, as conceded by all parties, ignored legislative districting policy and constructed districts solely on the basis of population considerations. The District Court erred in -this choice. Given the alternatives, the court should not have imposed Plan C, with its very different political impact, on the State. It should have implemented Plan B, which most clearly approximated the reapportionment plan of the state legislature, while satisfying constitutional requirements. The court said only that Plan C is “significantly more compact and contiguous” than Plan B. But both Plan B and Plan C feature contiguous districts, and, even if the districts in Plan C can be called more compact, the District Court’s preferences do not override whatever state goals were embodied in S. B. 1 and, derivatively, in Plan B. “The remedial powers of an equity court must be adequate to the task, but they are not unlimited. Here the District Court erred in so broadly brushing aside state apportionment policy without solid constitutional or equitable grounds for doing so.” Whitcomb v. Chavis, supra, at 161. If there was a good reason for adopt*797ing Plan C rather than Plan B, the District Court failed to state it.
Of course, the District Court should defer to state policy in fashioning relief only where that policy is consistent with constitutional norms and is not itself vulnerable to legal challenge. The District Court should not, in the name of state policy, refrain from providing remedies fully adequate to redress constitutional violations which have been adjudicated and must be rectified. But here, the District Court did not suggest or hold that the legislative policy of districting so as to preserve the constituencies of congressional incumbents was unconstitutional or even undesirable. We repeat what we have said in the context of state legislative reapportionment: “The fact that district boundaries may have been drawn in a way that minimizes the number of contests between present incumbents does not in and of itself establish invidiousness.” Burns v. Richardson, 384 U. S., at 89 n. 16. Cf. Gaffney v. Cummings, ante, at 752; Taylor v. McKeithen, 407 U. S. 191 (1972). And we note that appellees themselves submitted Plan B to the District Court and defended it on the basis that it adhered to state goals, as embodied in S. B. 1, while eliminating impermissible deviations.16
The judgment of the District Court invalidating S. B. 1 is affirmed. The adoption of Plan C is, however, reversed, and the case is remanded for further proceedings consistent with this opinion.
It is so ordered.
Prior to the passage of S. B. 1, the Texas Senate had twice defeated redistricting bills, passed by the House, with total deviations smaller than the total deviation in S. B. 1.
The redistricting of the 24 Texas congressional districts under S. B. 1 follows:
At a subsequent pretrial conference, the Fourteenth Amendment claims were eliminated.
Plan B resulted in the following districting:
District Population Absolute Variance from Ideal
1 466,545 + 15
2 466,565 + 35
3 466,266 -264
4 466,234 -296
5 466,620 + 90
*787District Population Absolute Variance from Ideal
466,285 6 -245
466,336 7 -194
466,704 8 + 174
466,678 9 + 148
466,313 10 -217
466,258 11 -272
466,930 12 +400
466,663 13 + 133
466,437 14 - 93
466,359 15 -171
466,663 16 + 133
466,432 17 - 98
466,520 18 - 10
466,649 19 + 119
466,514 20 - 16
466,753 21 +223
466,707 22 + 177
466,424 23 -106
466,875 24 +345
Plan C resulted in the following districts:
District Population Absolute Variance from Ideal
1 465,986 -544
2 466,817 +287
3 466,835 +305
4 467,108 +578
5 466,258 -272
6 467,023 +493
7 466,336 -194
8 466,704 +174
9 466,678 +148
10 466,303 -227
11 466,569 + 39
12 466,926 +396
13 467,173 +648
14 466,437 - 93
15 466,359 -171
16 465,941 -589
17 466,340 -190
18 466,520 - 10
19 466,154 -376
20 466,654 +124
21 466,875 +345
22 466,707 +177
23 466,167 -363
24 465,855 -675
The District Court’s entire discussion of its reasons for selecting Plan C follows:
“Defendant has not submitted any plan of reapportionment as an alternative to S. B. 1. Plaintiffs have proposed two plans, B and C. Plan B is based on S. B. 1, but has a significantly lower deviation than S. B. 1. Plan C is based solely on population and is significantly more compact and contiguous than either S. B. 1 or Plan B. . . . The Court has considered Plans B and C, as well as the plan submitted by the intervening plaintiffs, and concludes that Plan C best effectuates the principle of ‘one man, one vote’ enunciated by the Supreme Court.”
The District Court's order also granted leave to intervene to Van Henry Archer, Chairman of the Bexar County Republican Party, and others. The intervenors, appellees in this Court, filed a suggested reapportionment plan with their complaint-in-intervention which was rejected by the District Court and is not pressed here. The District Court also retained jurisdiction for the purpose of extending the impending February 7, 1972, filing date for congressional candidates “in the event it is made known to [the District] Court that a called session of the Legislature will include congressional reapportionment.” However, the Governor refused to call a special session of the legislature.
Kirkpatrick v. Preisler “reject[ed] Missouri’s argument that there is a fixed numerical or percentage population variance small enough to be considered de minimis and to satisfy without question the ‘as nearly as practicable’ standard.” 394 U. S., at 530. We concluded, “Unless population variances among congressional districts are shown to have resulted despite such [good-faith] effort, the State must justify each variance, no matter how small.” Id., at 531.
Prior to the passage of S. B. 1, the Texas House twice passed a congressional reapportionment bill with lower deviations. Each bill had a total deviation of 2.5%. Although both bills were ultimately defeated in the Senate, their passage by the House, and indeed their very existence, indicates that it was possible and practicable to construct a redistricting scheme with lower population deviations among districts than those embodied in S. B. 1.
“Appellant earnestly submits that the term 'constituency-representative relations’ is the more accurate term; indeed it is very hard to see why those who are so concerned about representation should stigmatize as a mere euphemism a term which brings in both parties to the representational relationship .... (The assumptions seem to be that while a Congressman may like his job, no constituency can like its Congressman, or care whether he continues to represent it or not — and that no Congressman can possibly learn to know his constituency well enough to serve it better than he can serve another constituency selected for him by, it may be, a young mathematician in Dallas.) Under either name, appellant would defend this motive as entirely proper, if the burden of that defense fell upon him on the facts herein.” Brief for Appellant 72.
It appears that the two plans passed by the House and defeated by the Senate may also have fostered this goal while achieving lower population variances.
Appellant contends that the authors of S. B. 1, and the legislature in passing on the plan, took into account projected population shifts among the districts. Remembering that the congressional districting plan will be in effect for at least 10 years and five congressional elections, the appellant argues that the legislature might properly consider population changes in devising a redistricting plan. In Kirkpatrick v. Preisler, we recognized that “[w]here these shifts can be predicted with a high degree of accuracy, States that are redistrieting may properly consider them.” 394 U. S., at 535. We were, however, careful to note:
“By this we mean to open no avenue for subterfuge. Findings as to population trends must be thoroughly documented and applied throughout the State in a systematic, not an ad hoc, manner.” Ibid.
In the present case, we conclude that Texas’ attempt to justify the deviations found in S. B. 1 falls far short of this standard. The record is barren, with the exception of scattered and vague assertions in deposition testimony, of adequate documentation of the projected population shifts and firm evidence that the alleged shifts were in fact relied upon.
There is also some suggestion that passage of S. B. 1 was preceded by a dispute as to who would fill the Second District congressional seat. The State does not urge this alleged goal as a justification for the deviations in S. B. 1, nor can we tell from this record whether S. B. 1 in fact resolved this dispute.
The court had before it a plan submitted by the plaintiffs-intervenors and, possibly, other plans. Only Plan B and Plan C appear to have been seriously urged by the parties and considered by the court, and only those plans are defended before this Court.
“Plan B, presented by Appellees, merely took the plan of the legislature and adjusted that plan to achieve greater equality to present to the court, in a graphic manner, what the legislature could *794have done if it had been disposed to make an attempt at population equality . . . Brief for Appellees 25.
Appellees’ amended complaint explained Plan C, as follows:
“That had the legislature desired to enact a statute consonant with the mandate of Article I, § 2 of the U. S. Constitution, that is a plan which made each district as compact and contiguous and as nearly equal in population to each other district as practicable, taking into account solely population and not taking into account 'social/ 'cultural/ 'economic’ or ‘other factors’ including preservation of incumbent congressman, it could have enacted a plan the same as or substantially similar to that plan set forth in Exhibit C annexed hereto and herewith incorporated by reference as though set forth at length herein. That such plan is hereinafter referred to as ‘Plan C.’”
S. B. 1 is conceded also to have sought adherence to county lines. While Plan B admittedly cuts more county lines than does Plan C, it also achieves lower deviations. Because both Plan B and Plan C were required to fracture more political boundaries than did S. B. 1, in order to achieve population equality among districts, appellant does not contend that Plan B is unacceptable because of more cutting of county lines.