Roman v Sincock Brief Amicus Curiae
Public Court Documents
June 15, 1964

78 pages
Cite this item
-
Brief Collection, LDF Court Filings. Roman v Sincock Brief Amicus Curiae, 1964. d7fef23c-c39a-ee11-be37-00224827e97b. LDF Archives, Thurgood Marshall Institute. https://ldfrecollection.org/archives/archives-search/archives-item/b3db6478-b6d4-43c5-a1eb-fad71c1793d1/roman-v-sincock-brief-amicus-curiae. Accessed April 22, 2025.
Copied!
No. 307 J n t o JSttjjitmt dfoort of t o ttito it JStatea October Term, 1963 M abel V . R oman, Clerk of the P eace for N ew Castle County, et al., appellants v. R ichard Sincock, et al. ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE BRIEF FOR THE UNITED STATES AS AMICUS CURIAE ARCHIBALD COX, Solicitor General, BRUCE J. TERRIS, Assistant to the Solicitor General, RICHARD W . SCHMUDE, Attorney, Department of Justice, Washington, D.C., 20530. I N D E X Paga 'Opinions below________________________________________ 1 Jurisdiction____________________________________________ 1 Questions presented_________ Interest of the United States. .Statement: 1. The complaint_______ 2. The pre-decision proceedings in the district court. 3. The evidence___________________________________ 6 a. The apportionment before the 1963 con stitutional amendment__________________ 6 b. The apportionment after the 1963 constitu tional amendment_______________________ 8 4. The decision and decree of the district court_____ 10 Argument: Introduction and summary_________________________ 13 I. In determining whether an apportionment of legislative representatives violates the equal protection clause, the basic standard of com parison is the representation accorded qualified voters per capita____________________________ 16 II. The Delaware apportionment creates gross discrimination among voters without a ra tional foundation in policy__________________ 23 III. Even if the apparently capricious apportion ment of the Delaware legislature had an in telligible foundation, the discrimination would violate the equal protection clause unless based upon criteria relevant to legislative apportion ment________________________________________ 32 1. The principle___________________________ 32 2. Application to the instant case_________ 43 (i)714- 431— 63----------- 1 to to t o II Argument—Continued IY. The apportionment of the Delaware legislature violates the equal protection clause by sub ordinating popular representation to the rep resentation of political subdivisions to such a degree as to create gross inequalities among voters and give control of both houses of the paga legislature to small minorities of the people— 45 Y. The injunction ordered by the district court is not an abuse of discretion------------------------------- 56 Conclusion_____________________________________________ 59 Appendix----------------------------------------------------------------------- 61 CITATIONS Cases: Armstrong v. Mitten, 95 Colo. 425, 37 P. 2d 757------- 56 Ashury Park Press, Inc. v. "Woolley, 33 N.J. 1, 161 A. 2d 705________________________________________ 57 Baker v. Carr, 369, U.S. 464------------ 16,23,27,33,37,39,54 Brooks v. State, 162 Ind. 568, 70 N.E. 980--------------- 56 Brown v. Board of Education, 347 U.S. 483------- 52, 53, 54 Butcher v. Trimarchi, 28 D. & C. 2d (Pa.) 537------ 57 Davis v. Mann, 213 F. Supp. 577, pending on appeal, No. 69, this Term_______________________________ 56 Denny v. State, 144 Ind. 503, 42 N.E., 929--------------- 56 Fortner v. Barnett, No. 59, 965, Chancery Court, First Judicial District, Hinds County, Mississippi- 58 Goesaert v. Cleary, 335 U.S. 464------------------------------ 16 Gomillion v. Lightfoot, 364 U.S. 339----------------------- 37 Gray v. Sanders, 372 U.S. 368___________ 37,38,40,41,42 Griffin v. Illinois, 351 U.S. 12---------------------------- 18, 20,43 Legislative Reapportionment In re, 374 P. 2d 66— 57 Lein v. Sathre, 205 F. Supp. 536---------------------------- 57 Magraw v. Donovan, 163 F. Supp. 184---------------------- 57 Maryland Committee for Fair Representation v. Tawes, No. 29, this Term_______ 2,21,23,31, 32,33,39, 46 Mikell v. Rousseau, 183 A. 2d 817------------ ---------------- 57 Moss v. Burkhart, U.S. D.C., W.D. Okla., decided July 17, 1963___________________________________ 58 Nixon v. Condon, 286 U.S. 73--------------------------------- 37 Parker v. State, 133 Ind. 178, 32 N.E. 836--------------- 56 n i Cases— Continued paga Ragland v. Anderson, 125 Ky. 141, 100 S.W. 865___ 56 Rogers v. Morgan, 127 Neb. 456, 256 N.W. 1_______ 56 Scholle v. Secretary of State, 367 Mich. 176, 116 N.W. 2d 350___ ;_______________________________________ 56 Sims v. Frink, 208 F. Supp. 431, pending on appeal sub nom. Reynolds v. Sims, Nos. 23, 27, 41, this Term____________________________________________ 56, 58 Skinner v. Oklahoma, 316 U.S. 535_________________ 43 Sobel v. Adams, 208 F. Supp. 316__________________ 58 South v. Peters, 339 U.S. 276______________________ 35 State ex rel. Attorney General v. Cunningham, 81 Wis. 440, 51 N.W. 724____________________________ 56 State ex rel. Lamb v. Cunningham, 83 Wis. 90, 53 N.W. 35_________________________________________ 56 State v. Warwick, 108 A. 2d 85____________________ 52 Stevens v. Faubus, 354 S.E. 2d 707________________ 58 Stiglitz v. Scliardien, 239 Ky. 799, 40 S.W. 2d 315_ 56 Sweeney v. Notte, 183 A. 2d" 296____________________ 40 Thigpen v. Meyers, U.S. D.C., W.D. Wash., decided May 3, 1963, pending on appeal, No. 381 this Term 56 Toombs y. Fortson, U.S. D.C., N.D. Ga., decided September 5, 1962_______________________________ 58 Wesberry v. Sanders, No. 22, this Term____________ 51 WMCA, Inc. v. Simon, No. 20, this Term_________ 32 Young, ex parte, 209 U.S. 123______________________ 56 Constitutions and statutes: U.S. Constitution: Article I : Sec. 2-----------------------------------------------------------49,50 Sec. 3_____________________________________ 49 Fourteenth Amendment_____________________ 2 3, 4, 6, 11, 14, 18, 20, 21, 22, 23, 28, 32, 36, 40, 43, 46, 48, 52, 53, 59 Statutes: Civil Rights Act, 42 U.S.C. 1983______________ 4, 6 2 U.S.C. 2a___________________________________ 51 28 U.S.C. 1343_______________________________ 3 28 U.S.C. 2201___________________ 28 U.S.C. 2281_______________________________ 4 42 U.S.C. 1988_______________________________ 3 IV Constitutions and statutes— Continued Delaware Constitution: pag9 Article 2, Sec. 2___________________ 4,5, 6,8,10,11,12 Article 16: Sec. 1________________________ 22,57 Sec. 2_______________________________________10,22 Florida Constitution of 1868______________________ 55 Georgia Constitution of 1868______________________ 55 Louisiana Constitution of 1868____________________ 55 Massachusetts Constitution of 1780: Part II, Ch. I : Sec. 2, Art. I ______________________________ 67 Sec. 8, Art. I I _____________________________ 67 South Carolina Constitution of 1868_______________ 54 Virginia Constitution of 1870_____________________ 55 Miscellaneous: Blackstone, Commentaries, pp. 114—116_____________ 7 80 Cong. Globe 3024_______________________________ 53 I Farrand 179--------------------------------------------------------- 69 I Farrand 323--------------------------------------------------------- 72 I Records of the Federal Convention (Farrand ed., 1911)________________________________________ 61 Jit tU djrmrt of t\u Mnltd states October Term, 1963 No. 307 M abel Y . R oman, Clerk of the P eace for N ew Castle County, et al., appellants v. R ichard Sinoock, et al. ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE BRIEF FOR THE UNITED STATES AS AMICUS CURIAE OPINIONS BELOW The opinion of the district court (R. 491-531) is reported at 215 P. Supp. 169. Prior opinions of the district court (R. 66-76, 214-217, 319-332) are re ported at 207 P. Supp. 205 and 210 P. Supp. 395 and 396. j u r i s d i c t i o n The judgment o f the district court was entered on April 17, 1963 (R. 532-533). Notices o f appeal to this Court were filed on May 28 and June 12, 1963, and probable jurisdiction was noted on October 21, 1963 (R. 541-545, 551-554). The jurisdiction of this Court rests upon 28 TT.S.C. 1253. ( i ) 2 QUESTIONS PRESENTED 1. Whether the provisions o f the Delaware Consti tution apportioning the Delaware legislature, which create serious disparities in per capita representation in both houses without any permissible or even ra tional justification and give control o f the legislature to a small minority of the people, violate the equal protection clause of the Fourteenth Amendment. 2. Whether the decree of the district court enjoin ing further elections pursuant to the unconstitutional apportionment was an abuse of discretion. INTEREST OP THE UNITED STATES This is one o f five reapportionment eases pending disposition in which the Court is called upon to formu late under the Fourteenth Amendment constitu tional principles applicable to challenges to malappor tionment o f a State legislature. The Court has heard argument in the New York, Alabama, Maryland, and Virginia cases. The United States has filed its prin cipal brief in Maryland Gommittee for Fair Repre sentation v. Tawes, No. 29, in which we attempted to present a compendious analysis of principles which we deem applicable in the reapportionment field. The instant case raises specific problems in the application of those principles. STATEMENT 1. The Complaint.— On June 5, 1962, the plain tiffs—seven citizens and taxpayers of the United States and the State of Delaware, who are residents and qualified voters of New Castle County, Dela ware—filed a complaint in the United States District 3 Court for the District o f Delaware, in their own be half and on behalf of all persons similarly situated, challenging the apportionment o f the Delaware legis lature (R. 1-28). The defendants, who were sued in their representative capacities as officials charged with the performance of duties in connection with State elections, included the State Election Commis sioner, the members of the Department of Election for each o f the State’s three counties, the members o f the Board of Canvass of the three counties, and three Clerks of Peace (R. 5-7). The complaint al leged deprivation o f rights under the equal protection clause of the Fourteenth Amendment, and asserted that the district court had jurisdiction under the Fourteenth Amendment, 42 U.S.C. 1983 and 1988, and 28 U.S.C. 1343 and 2201 (R. 3). The complaint alleged that the appoidionment of the Delaware legislature resulted in “ invidious dis crimination as to the inhabitants of New Castle Coun ty and the City of Wilmington” (R. 10-14). It alleged that the plaintiffs had the right “ to cast votes [for legislators] that are of equal effect with that of every other citizen of the State of Delaware” ; that the malapportionment of the legislature has operated to deny those rights to the plaintiffs; and that such malapportionment constituted “ arbitrary and capri cious state action which denies to petitioners a full vote in equality with other voters of the State of Delaware by reason of the failure to provide a reason able classification of voters and members of the Gen eral Assembly,” in violation of, inter alia, the equal 4 protection and due process clauses of the Fourteenth Amendment (R. 14-15). Plaintiffs also alleged that they were without any other adequate remedy inasmuch as the legislative apportionment was “ frozen” into the State constitu tion ; that the legislature was dominated by representa tives from the less populous counties; and that it was impossible to amend the constitution or to convene a constitutional convention for the purpose of reappor tioning the General Assembly (R. 16). They asserted that the General Assembly “ has prevented reappor tionment by repeatedly failing to take appropriate restorative action” (R. 13-14). Plaintiffs requested: (1) that a three-judge district court be convened under 28 TJ.S.C. 2281; (2) that the court enter a declaratory judgment that Article 2, Section 2 o f the Delaware Constitution, which established the apportionment of the Delaware legis lature, is unconstitutional under the Delaware Consti tution, the Fourteenth Amendment, and the Civil Rights Act, 42 U.S.C. 1983; (3) that the court enjoin the defendants from conducting or engaging in any elections held pursuant to Article 2, Section 2 of the State Constitution; (4) that the court either apportion the Delaware legislature in accordance with the popu lation of the State or direct that the general election o f November 6, 1962, be conducted on an at-large basis; and (5) that the court retain jurisdiction for the entry of “ such other and further relief as may be necessary and proper” (R. 17-18). 2. The Pre-decision Proceedings in the District Court.—A three-judge district court was convened 5 (R . 29-30). On July 25, 1962, the court entered an order staying the proceedings until August 7, 1962, “ in the hope and expectation that some appropriate action [might] be taken by the 121st General Assem bly” (R. 66-76, 77). 207 P. Supp. 205. The court noted that inasmuch as the Delaware Constitution required the Secretary of State to publish any pro posed constitutional amendment in newspapers in each county at least three months prior to the succeed ing general election (in this case, the general election of November 6, 1962), it would serve no useful pur pose to grant a stay beyond August 7, 1962 (ibid.). On July 30, 1962, the 121st General Assembly ap proved a proposed amendment (Del. H.B. 534) to Arti cle 2, Section 2 of the Delaware Constitution (R. 275- 277; PL Exs. 2 and 8). It could not, however, become effective unless adopted by the next session of the legislature. On August 7, 1962, the district court held that the complaint properly alleged a cause of action within the jurisdiction of the court, and denied motions to dismiss (R. 214-217). 210 P. Supp. 395. The court noted that it had no desire to substitute its judgment for the collective wisdom of the Delaware General As sembly in matters of apportionment of the legislative, but that it had no alternative but to promptly proceed in the matter (ibid.). Some of the defendants ap plied for a further stay of proceedings so that the 122d General Assembly coming into office on January 8, 1963, would have an opportunity to agree to the proposed amendment (R. 219-220). 6 On August 8, 1962, the plaintiffs applied in the district court for a preliminary injunction in order to enjoin the conduct of the November 1962 general election “ until the present discrimination against the Plaintiffs and all others similarly situated is corrected” (R. 224). Thereafter, the court permitted the plain tiffs to amend their complaint (1) to have the proposed constitutional amendment declared uncon stitutional as violative of the Fourteenth Amendment and the Civil Rights Act, 42 TT.S.C. 1983; and (2) to request an order provisionally reapportioning the Delaware legislature (R. 280-282, 316-318). On October 16, 1962, the district court denied both the applications for a preliminary injunction and for a further stay (R. 319-332, 333). 210 F. Supp. 396. After extended pretrial proceedings, the court, on November 27, 1962, entered a pretrial order in which the parties agreed to the accuracy o f a series of ex hibits, statistics, statistical computations, and the like (R . 368^400). In early January 1963, the 122d General Assembly approved the proposed constitutional amendment (Del. H.B. 574) by the requisite two-thirds vote. As a result, the amendment to Article 2, Section 2 became effective on January 17,1963. 3. The Evidence, a. The apportionment before the 1963 constitutional amendment.-—At the time of the adoption of the Constitution o f 1897, Kent County, Sussex County, and “ rural” New Castle County (i.e., the area outside Wilmington) were each apportioned 10 representatives and 5 senators (R. 377). The City of Wilmington, which had considerably more popu- 7 lation than any one o f the rural areas, was given 5 representatives and 2 senators (R. 377). The 35 rep resentative districts generally followed the boundaries o f a “hundred,” a geographical subdivision of a county in Delaware since its founding.1 Many of the representative districts coincided with a particular hundred; others consisted o f a portion of a hundred, or of a hundred and a portion of another hundred, or of two or more hundreds or portions o f hundreds. Wilmington City (apparently within the bounds of the ancient Wilmington Hundred) was divided into 5 representative districts (Court Ex. 2, pp. 57-58; PI. Ex. 23). The 17 senatorial districts, which were like wise specified in the Constitution, were composed either of two representative districts each (the situation ap plicable to Kent and Sussex Counties), or of two or more hundreds or portions of hundreds (the 3d, 4th, 5th, 6th and 7th districts of Hew Castle County), or of portions o f Wilmington City (the 1st and 2d districts of New Castle County). 1 1 In early England, “hundreds” were subdivisions of counties or shires based upon the number of families of freeholders residing there. Each hundred consisted of ten towns or tith- ings, and each town or tithing consisted of ten families of freeholders. Each county or shire was composed of an indefi nite number of hundreds. 1 Blackstone, Commentaries, pp. llT - 116. Hundreds were used as geographic subdivisions in the early days of the colonies of Virginia, Maryland and Pennsylvania. In Delaware, the hundreds are simply ancient subdivisions of the counties established under the authority of William Penn (R. 1019-1020; Court Ex. 2, p. 58). There are 11 hundreds in New Castle County, including Wilmington Hundred, which is coterminis with the City of Wilmington; 9 hundreds in Kent County; and 13 hundreds in Sussex County. 8 The evidence showed that in the six years prior to the institution of this case, about 7 bills providing for reapportionment of the legislature or for calling a constitutional convention for that purpose were in troduced in the General Assembly but were either not reported out of committee or else failed to pass (R. 51-57, 60-65, 271-274, 389; PI. Exs. 6, 7, 8, and 9). b. The apportionment after the 1963 constitutional ameyidment.— The amendment to Article 2, Section 2 of the Delaware Constitution, which became effective on January 17, 1963, increased the size of the Senate from 17 to 21 members and allotted the 4 additional senators equally to Sussex and Kent Counties, thereby giving each of the three counties 7 senators (R. 385; PI. Ex. 8). The sixth senator for Kent County was to be elected at large from the first, second, fifth, seventh, and eighth representative districts of that county, and the seventh senator for Kent County was to be elected at large from the third, fourth, sixth, ninth, and tenth representative districts {ibid.). The sixth senator for Sussex County was to be elected at large from the first through the fifth representative districts of that county, and the seventh senator was to be elected at large from the sixth through tenth repre sentative districts of that country {ibid.). As a result, each voter in Kent and Sussex County would have been entitled to vote for one representative and two senators. As for the House of Representatives, the amend ment provided that each representative district with a population in excess of 15,000 persons was to be alloted an additional representative for each addi tional 15,000 persons or major fraction thereof (R. 384; 9 PL Ex. 8). The boundaries of the original 35 repre sentative districts were not affected.2 The net effect of the amendment in terms of immediate representation in the House would have been to allot an additional 10 rep resentatives to various districts in Hew Castle County, increasing the size of the House to 45 members (R. 384).3 The representation of Kent and Sussex Coun ties would not have been affected. The populations of each senatorial and representa tive district, both before and after the 1963 amend ment, are set forth in the opinion of the court (R. 496- 497, 499, 503-505). The constitutional amendment also provided that in in the event that a constitutional convention was con- 2 Districts receiving additional representatives were to be divided so that each of the new districts would elect one repre sentative. The new districts -were to be “ as nearly equal in population as possible to the other new districts being created” from, the same representative district. The actual reapportionment of representatives and redis tricting of the districts was to be performed by a Redistricting Commission consisting of the Governor and the state chair men of the two political parties receiving the largest vote for governor at the preceding gubernatorial election as advisors to the Governor. The reapportionment and redistricting wais to be completed by the Commission within 120 days following the official reporting to the President of the United States of each decennial census. In the event that the Commission failed to act within the 120 days, any qualified voter, within the fol lowing 30 days, might, by mandamus or otherwise, compel the Governor to reapportion and redistrict the House of Repre sentatives. These provisions, however, relate solely to the reapportionment and redistricting within the existing 35 repre sentative districts. 3 The 10 additional representatives would have been alloted as follows: one each to Districts 2, 5, and 8; two each to Districts 7 and 10, and three to District 6 (R. 384, 503). 10 vened, as provided for in Article 16, Section 2 of the Constitution, the number of delegates and their method of election was not to be affected by the addi tion of representatives as a result of the amendment. For the purpose of any future constitutional conven tion, the representative districts were to elect dele gates on the basis of the apportionment provided by Article 2, Section 2 as it existed prior to the amend ment (R. 387; PI. Ex. 8). Article 16, Section 2 pro vides that the General Assembly, by a two-thirds vote, may submit to the voters at the next general election the question whether a constitutional convention to revise and amend the constitution shall be convened, and, if a majority of the people voting at the election approve, the General Assembly must convene the convention. The number of delegates would continue to be 41, one from each of the original 35 representative districts and 2 elected at large from each of the three counties.4 4. The decision and decree of the district court — On April 17, 1963, the district court, in an opinion by Circuit Judge Biggs, held that Article 2, Section 2 o f the Delaware Constitution, as it existed prior to the 1963 amendment as well as after the amendment, resulted in gross and invidious discrimination against the plaintiffs and others similarly situated, in viola tion of the equal protection clause of the Fourteenth Amendment. The court could find no rational or rea sonable basis for the apportionment under Article 2, Section 2, either before or after the amendment. 215 4 In addition to other witnesses, three political scientists testi fied at length at the trial concerning proper methods of ap portionment (R. 591-706, 709-886, 890-918, 927-1010, 1166- 1295). 11 F. Supp. at 184-189. The court concluded that the matter of reapportionment was basically a legislative function, and that a further opportunity should be given to the General Assembly properly to reappor tion itself in accordance with the Fourteenth Amend ment. In this regard, the court noted that minimal constitutional standards required that at least one house of the Delaware General Assembly be ap portioned on an “ equal population basis” with the overriding aim of “ one vote, one person” ; and that, with one house so apportioned, other cogent and rele vant factors (i.e., history, geography, custom) might be given some weight along with population in appor tioning the other house. 215 F. Supp. at 189-191. However, the weight given to such factors “ should not be thought of as necessarily destroying the founda tion principle of majority rule, or effecting invidious discrimination or irrationality in apportionment.” Id. at 191. Due to the closeness of the 1964 general election, the court gave the General Assembly until October 1, 1963, to enact a reapport.ionment statute. Chief District Judge Wright concurred that Article II, Section 2 of the Delaware Constitution, before as well as after amendment, was unconstitutional, and that at least one house of a State legislature must be apportioned strictly on a population basis. He agreed with the “precatory observation” of Judge Biggs that the upper house must also be based “ substantially upon population * * * so long as the majority prin ciple is adhered to.” 215 F. Supp. at 195. District Judge Layton concurred in that portion of the opinion holding Article 2, Section 2 of the Con 12 stitution, prior to as well as after the 1963 amend ment, unconstitutional as applied to the House o f Representatives. Judge Layton pointed out that, since the 1963 amendment did not contain a severa bility provision, the whole amendment was unconsti tutional because of the House provisions and, there fore, no need existed to consider whether the Senate provisions were constitutional. Judge Layton said that it was constitutionally permissible for a State to pattern one house upon an equal area basis where the other house was apportioned upon a “ strict popula tion basis” inasmuch as such a system would be pat terned upon the Federal Congress. Id. at 196-197. A decree was entered declaring Article 2, Section 2 of the Delaware Constitution to be unconstitutional (R. 536-537). The court reserved jurisdiction to order injunctive or other relief if this appeared nec essary. On May 20, 1963, the court enjoined the de fendants from conducting any special or primary elec tion, or the general election of November 1964, for the General Assembly under the apportionment provi sions of Article 2, Section 2 of the Delaware Con stitution as they existed prior to or after the 1963 constitutional amendment. The court again reserved jurisdiction to make such further orders as it might deem necessary (R . 540). The decree does not em body any of the “ precatory observations” about the apportionment to be made by the Delaware legislature. The defendants thereafter filed notices o f appeal to this Court. On June 27, 1963, Mr. Justice Brennan stayed the district court’s injunction pending dispo- 13 sition of the ’case by this Court. Probable jurisdic tion was noted on October 21, 1963. ARGUMENT INTRODUCTION AND SUM M ARY This is by far the easiest of the reapportionment cases now before the Court on the merits. The ap portionment of the Delaware legislature plainly vio lates any standard of equal protection. Appellants’ elaborate argument that a State may model repre sentation in its legislature after the federal analogy is totally irrelevant. The Delaware legislature is constituted upon a radically different basis from the federal Congress, and the apportionment results in gross discrimination that the federal system has never created.5 Nor is the question raised by the decree * 1 5 For convenience we list here some of the reasons, discussed in the Argument, showing why this case raises no questions of the constitutionality of a State apportionment of seats in a bicameral legislature analogous to the Congress of the United States: 1. The Delaware House o f Representatives is not appor tioned in accordance with population. The federal House of Representatives is so apportioned, subject to the very minor qualification that each State is assured one representative. This basic contrast is evident from the fact that a majority of the representatives in Delaware’s lower house come from districts containing only 28 percent of the population—less than one-third—whereas it takes, under the constitutional ap portionment of Congress, almost half the people to elect a majority of the United States House o f Representatives. 2. The Delaware Senators are not chosen as representatives of the counties. The 21 Senators are chosen one each from 21 districts established solely for the purpose of their election. Each Senator represents his district, not the county. The only point o f resemblance to the United States Senate is that the 714-431— 63------2 14 of the district court, which merely adjudicates the unconstitutionality of the present apportionment. In our briefs and arguments in the previous ap portionment cases, we have submitted and sought to justify four basic principles by which the unconsti tutionality of a State’s apportionment should be de termined. There is no occasion to repeat the argu ment here. In this brief we content ourselves with showing how the four principles apply to this case, supplementing the basic analysis only where neces sary to deal with questions brought out in oral argu ment and inadequately covered in our prior briefs. The starting point, we argued, should be the arith metical comparison of the representation accorded qualified voters per capita. W e discuss the proposition here only to the extent necessary to show that the com parisons of per capita representation are a proper measure o f the political equality guaranteed by the equal protection clause o f the Fourteenth Amendment. W e then turn to the second proposition—that an apportionment which creates gross inequalities in per capita representation without any intelligible founda tion violates the Fourteenth Amendment—and show that the Delaware apportionment, not as an imaginary “ federal plan” but as it actually exists, creates that form of unconstitutional discrimination. legislature, in an effort to create an analogy, created the same number of districts in each county. 3. There is no rational basis for the apportionment of seats in either house of the Delaware legislature. The apportion ment of seats in each branch of Congress is based upon a coherent political philosophy. 15 Since counsel suggest no rationalization of the idiosyncrasies that permeate the apportionment of both houses of the Delaware legislature and none are apparent, it may be superfluous to go further; but out of an abundance of caution we submit that any criteria of differentiation offered by way o f explana tion would be insufficient to justify the discrimination unless relevant to the permissible purposes of legisla tive apportionment. The critical distinction, which we elaborate below, is between (a) rules serving the purpose of making representative government work better, the operation of which may have the collateral consequence of creating per capita inequalities in rep resentation, and (b) rules whose only function is to create classes of voters with preferred political rights giving them voting power disproportionate to their number. Whatever may be true in the first case, the discrimination is unconstitutional in the second. Finally, we show that the Delaware apportionment violates the equal protection clause because, apart from all else, it creates inequalities in per capita rep resentation and submerges the principle of majority rule to a degree so utterly disproportionate to any permissible purposes accomplished as to be arbitrary and capricious. In the Senate there are inequalities in the treatment of voters in different districts within the same county that run as high as 15 to 1. In the House of Representatives there is “ one disparity as high as 12 to 1, several at 11 to 1, 10 to 1, and 9 to 1, [and] many at 8 to 1” (R. 510). A numerical majority of the House comes from districts contain ing only 28 percent of the people. A majority of the 16 Senate comes from districts containing less than 20 percent of the people. The discrimination runs in the same direction in both houses of the legislature. I IN' DETERMINING W H ETH ER AN APPORTIONMENT OF LEGIS LATIVE REPRESENTATIVES VIOLATES TH E EQUAL PROTEC TION CLAUSE, TH E BASIC STANDARD OP COMPARISON IS TH E REPRESENTATION ACCORDED QUALIFIED VOTERS PER CAPITA “ The Constitution in enjoining the equal protection of the laws upon States precludes irrational discrimi nation as between persons or groups of persons in the incidence of a law.” Goesaert v. Cleary, 335 U.S. 464, 466. Baker v. Carr, 369 U.S. 186, 226, shows that this principle is applicable to the apportionment of repre sentation in a State legislature. To show inequalities in representation, the complain ants in the several reapportionment cases, like our selves, have made many statistical comparisons of the per capita representation of the voters of the dis tricts from which representatives were elected. A member of the Court suggested, however, that the re sulting ratios were merely sixth grade arithmetic perhaps lacking relevance to the issues at bar. The comment doubtless was intended to raise the funda mental inquiry, whether the equal protection clause is concerned with uniform apportionment per capita or with some broader concept of equal representation. I f the latter, do figures showing the per capita repre sentation afforded voters in different geographical dis tricts truly reflect the equality or inequality of repre- 17 sentation, or must the focus be widened to encompass other aspects of State government before one can de termine whether the particular governmental system denies some voters equal protection of the law? Our answer is that State action which gives rise to hostile or capricious discrimination in the per capita apportionment of representatives is so related to un equal representation in the larger sense as to violate the constitutional guarantee of equal protection of the law, however fairly other aspects of the State govern ment have been constructed. Deciding whether an unreasonably discriminatory apportionment of representatives violates equal pro tection is a more specific task than establishing an ideal form of representative government. One at tempting to establish an ideal “ representative govern ment” would face, at the threshold, the task of defin ing “ representativeness.” Even if it were agreed that a perfectly representative legislature would be a microcosm of the whole people, reproduced in exact proportions, it would still be apparent that the alloca tion of representatives in exact proportion to the population of the several districts would not alone ensure the objective. Where party loyalties are strong or few issues divide the electorate, skillful gerrymandering of the district lines may enable a mi nority of the people to elect a majority of the legisla ture. Even without gerrymandering, a population that was divided 60-40 on a critical issue might con ceivably choose a legislature having a unanimous view, because the vote could divide 60-40 in every district. 18 Thus, the political scientist planning Utopia or the* statesman organizing a new framework of govern ment must concern himself with such questions as where the district lines will be drawn, with “ wasted votes,” and with whether and how to make the legisla ture reflect the divisions and intensity of opinion with in the individual districts. Sundry plans of propor tional representation and other devices have been di rected at those ends. Nothing we have argued casts doubt upon their constitutionality. Whether the Fourteenth Amendment reaches such questions is not in issue. Any inability to guarantee “ equal repre sentation” in the largest sense would be no better reason for upholding an arbitrary classification in ap portionment than is the inability to offset all the dis advantages from which the outcast pauper suffers at a criminal trial a reason for withholding from him the transcript necessary for appeal that the rich can pur chase. Cf. Griffin v. Illinois, 351 U.S. 12. But if the apportionment of representatives among geographical districts in direct ratio to their popula tion is not enough to guarantee “ fair representation” in the broadest sense, arbitrary and capricious dis crimination in per capita representation will ahnost always defeat “ equality” in any sense of the word. The arithmetical comparisons undeniably measure the comparative influence that the ballot of each voter in different districts can have on the composition of the legislature, and thus on the course of legislation. I f one starts with gross and capricious discrimination at that point, it is virtually certain that the inequality will persist. The diversions, dilutions, distortions, and cross-currents that are felt before the votes cast 19 for representatives are reflected in action on the floor of the legislature, would rarely, if ever, rectify gross inequalities in the ratio of seats to population. No one, to the best of our knowledge, has ever attempted to show that the apparently capricious apportionment of seats in a State like Delaware is intended to offset other inequalities and so ultimately to produce equal representation. The most extraordinary combination o f circumstances would be required. There is not the slightest reason to suppose that Delaware’s per capita divergencies of 12 to 1, 9 to 1 and 8 to 1 as well as the many smaller discrepancies, all scattered as if by a throw of the dice, somehow mysteriously operate to cancel out a hidden gerrymander or other distortions of the per capita vote. It would be the wildest co incidence if the members of the Delaware House o f Representatives from the grossly overrepresented Dis trict 8 in Sussex County, Districts 3, 4 and 6 of Kent County, and Districts 12, 14 and 15 of New Castle County accurately reflected the wishes of those whose candidates lost the elections in the grossly underrep resented districts. It is even less likely that the underrepresented voters of New Castle District 9 are being represented by the representative of an over represented district, such as 14 or 15. Even if that could be the fact, it would be no answer to the com plaint. Those who raised the cry, “ No taxation with out representation” could not have been satisfied by an argument that they were being represented in the English Parliament by Edmund Burke and Charles James Eox. 20 The extent of the legislature’s participation in the law-making process is also irrelevant. The equal pro tection clause does not require a representative legis lature, but it does guarantee all voters against arbi trary discrimination in the incidence of any laws creating a legislature and apportioning its seats. Cf. Griffin v. Illinois, 351 U.S. 12. That the role of the legislature may be confined, or that it may be supplemented by the initiative and referendum, does not affect the problem. For it is no answer to the victims of arbitrary discrimination in respect to the influence of their votes upon the composition of the legislature to tell them that the role of their legislature is less important than in other States, or that there are special ways of making laws in which there is no denial of equal protection. Since the right to be free from capricious discrimi nation based upon no policy is a personal right, it is immaterial whether the discrimination was adopted, or can be changed, by initiative and referendum or some other exercise of the popular will. The central purpose o f the Fourteenth Amendment is to protect minorities. That principle, which is clear in other fields of constitutional law, is controlling here. A simple illustration is enough to put the point beyond dispute. I f a State with 21 identical counties, each containing 60,000 voters, divided each of 20 counties into six equal districts, with each district to elect one representative, but allocated only one represent ative to the twenty-first county, for a total of 121 representatives, it would take representatives elected by 610,000 voters or 48 percent of the population to 21 elect a numerical majority of the legislature. The entire State, except the victims of the discrimi nation living in the underrepresented county, might favor the apportionment. Nevertheless, the capricious discrimination would violate the Fourteenth Amend ment. This is also true when the apportionment is attacked under our third basic proposition as an instance of systematic discrimination based upon criteria that are invidious or irrelevant to any of the permissible purposes of legislative apportionment. I f the dis crimination is based upon grounds condemned by the concept of equal protection, then it matters not that the victims suffer at the hands of a majority of the people. On the other hand, the ease with which an appor tionment can be changed by a majority of the people in a State may be relevant in applying the fourth of the proposed principles—that the equal protec tion clause is violated by an apportionment which sub ordinates the principle of popular representation to other permissible considerations to such a degree as to create gross inequalities in the representation of voters and give control of the legislature to small minorities of the people. Under this head the issue turns upon the reasonableness of the balance the State has struck between the interest in equality of repre sentation and opposing objectives that we assume arguendo to be permissible objectives of legislative apportionment.6 The balance depends upon the degree 6 See Brief for the United States in Maryland Committee for Fair Representation v. Tawes, No. 29, this Term, pp. 24-25. 22 of inequality and submergence of the popular will. I f the popular will can find expression in other out lets, such as the initiative and referendum, or if a majority of the people can readily correct the mal apportionment, then the degree of minority control of the legislature is somewhat less important than would otherwise be the case. Such institutional ar rangements may be relevant to the application of the fourth proposition, therefore, even though they have no bearing upon the others.7 While such questions must be answered in any com prehensive rationale, they are not raised in the pres ent case. A majority of the people of Delaware are unable to reapportion the legislature by initiative and referendum or even a constitutional convention.8 There are no other institutions by which the popular will may express itself in the enactment or repeal of legislation. A statistical comparison o f the per capita represen tation of the voters in the districts into which a State is divided, standing alone, is insufficient to prove violation of the Fourteenth Amendment. The arith metic leaves open all of the legal questions. It is a 7 The availability of methods of constitutional reform may also bear upon the propriety of equitable intervention, but no such question is presented here. 8 The Delaware Constitution may be amended in only two ways: First, an amendment may be passed by a two-thirds vote in both houses of the legislature at two consecutive ses sions. Art. 16, § 1. Second, the General Assembly may, by a two-thirds vote, submit to the voters the question whether to hold a constitutional convention. I f a majority of voters approve, a convention must be called. Art. 16, § 2. Thus, the legislature has complete power to block any amendment of the constitution. 23 means of comparison, and for that purpose it is a useful, if not indispensable, tool. Be it easy or diffi cult, it measures exactly the extent to which the State has introduced a source of discrimination between persons and groups in the influence their votes may have on legislation through participation in the choice of representatives in the law-making body. While the process of measurement answers none of the legal issues, its simplicity should not be allowed to hide or denigrate the basic importance of what is being meas ured—equality or inequality of voting power in the selection of the composite legislature. For unequal voting power almost surely means unequal repre sentation. I I TH E DELAWARE APPORTIONMENT CREATES GROSS DISCRIM INATION AMONG VOTERS W ITH OUT A RATIONAL FOUNDA TION IN POLICY The Fourteenth Amendment’s condemnation of irra tional discrimination as between persons or groups of persons in the incidence of a law is squarely ap plicable to the apportionment of representatives in a State legislature. Baker v. Carr, 369 U.S. 186, 226. See also Brief for the United States in Maryland Committee for Fair Representation v. Tawes, No. 29, this Term, pp. 38-39, 50-51, and cases cited. Under this principle the apportionment o f seats in the Delaware legislature is manifestly unconstitu tional. Taking the House of Representatives first, the district court, in an opinion by an able and ex perienced circuit judge, found as a fact that there was 24 “ no rational basis for the apportionment” (R. 510). This ultimate finding was based upon such subsidiary- findings as that even under the 1963 reapportionment there is (R. 510)— one disparity as high as 12 to 1 and also— several at 11 to 1, 10 to 1 and 9 to 1 as well as— many at 8 to 1. The district court also foimd that there are (Hid.) — vagaries existing in representative districts even contiguous to each other both in rural areas and in the City of Wilmington. The result, the court pointed out, is to increase the power of favored districts to such an extent that a numerical majority of the Delaware House of Rep resentatives is elected by districts having only 28 per cent of the people. The undisputed evidence fully supports the findings. The population of each district, as the districts were constituted by the 1963 amendment, is set forth below (R. 504): 25 Sussex Comity District Population 1 ________________ 9, 641 2 ________________ 6, 439 3 ________________ 9,437 4 _____________ - 5,318 5_________________ 8,140 6__________________ 8, 010 7 ________________ 6,623 8 ________________ 2, 957 9 ________________ 4,271 10___________________12,359 Total____________ 73,195 Kent County District Population 1 _________________ 6, 084 2 __ 17,806 3 ________________ 3, 361 4 ________________ 3,716 5 ________________ 9,125 6 ________________ 2,626 7 ________________ 7,880 8 ________________ 5,168 9 ________________ 4, 956 10___________________ 4, 929 Total__________ 65,651 New Castle County (Wilmington) District Population 1__________________ 7, 814 2A_________________ 16,886 2B 1________________ 16,886 New Cattle Cou/nty—Con. ( Wilmington) District Population 3 ________________ 18,359 4 _______________ 5, 394 5A_________________ 15,244 5B_________________ 15,244 Total__________ 95,827 New Castle County (Outside Wilmington) District Population 6A_______________ 14, 557 6B_______________ 14, 557 6C_______________ 14, 557 6D_______________ 14, 557 7A_______________ 15,836 7B_______________ 15, 836 7C_______________ 15,836 8A_______________ 11,714 8B_______________ 11,714 9________________ 20, 040 10 A _______________ 13,431 10B_______________ 13,431 10C_______________ 13,431 11 _______________ 9,325 12 _______________ 3,401 13 _______________ 5,218 14 _______________ 2, 534 15 _______________ 1, 643 Total__________ 211,619 Total New Cas tle County— 307,446 1 The figures in this table for districts which will be divided under the 1963 amendments assume that the old districts will be divided so the new districts will be exactly equal in population. The governor has, since the opinion of the court below, issued a proclamation forming these districts (Appendix to App. Br., pp. 6a-10a). The actual population figures of these •districts are, as far as we can ascertain, not available. 26 The inequalities are too numerous and too obvious to catalog but it may be permissible to mention the most egregious: (a) District 15 in New Castle County has 10 times the representation of Districts 2A, 2B and 3 in the City o f Wilmington. District 14 has five times the average representation in all seven districts in the City of Wilmington. (b) There are two districts in Sussex County and five in Kent County that have more than three times the representation of each of five districts in the City of Wilmington. (c) On the average, a voter in Kent County has just under twice the representation of voters in New Castle County. A voter in Sussex County has more than 1.6 times the representation. (d ) Within each of the three counties (omitting Wilmington) the inequalities are egregious. In New Castle County the smallest district has 12 times the representation of the largest; in Sussex, the ratio is 4 to 1, and in Kent, 6% to 1. (e) The net effect of the discrimination in favor of minorities is so great that representatives elected by only 28 percent of the people constitute a numeri cal majority of the lower house of the legislature. In the selection o f the Delaware assembly, the dis criminations between persons and groups in the allo cation of voting power are as capricious as they are egregious. There is no meaningful or even consistent policy in the composition of the districts. Although the district lines often correspond to a “hundred” —a subdivision of a county in Delaware (see p. 7, 27 note 1 above)—they frequently differ. While the representative districts in New Castle County outside Wilmington each formerly consisted of a single hun dred, those in Wilmington consisted of two to four wards or portions of wards. In Sussex and Kent Counties, the districts consist of all or part of one, two, or three hundreds. (Even if the district lines were to conform to the ancient “ hundreds,” that fact would not justify the discrimination because the hundreds have no political or governmental sig nificance.) Viewed alone, the 1963 amendment was rational insofar as it provided that each representative district should receive an additional representative for each 15,000 people or major fraction thereof above a base o f 15,000 people. The districts would then be subdi vided. The total effect is to leave the apportionment of seats in the assembly still a crazy-quilt because no changes were made in the districts containing less than 15,000 people. Since there are 25 districts out of a total of 45 in which the ratio of representation is greater even than 1 to 10,000 persons, superimposing a rational rule upon a fraction of an irrational base did little to correct the capricious pattern described above. Appellants offer no explanation of the gross dis crimination in representation in the House of Repre sentatives. The omission speaks volumes, for the case was tried long after the decision in Baker v. Carr and their brief was filed after the briefs in the first group of apportionment cases had focused attention upon the proposition that gross discrimination in per capita representation, resting upon no policy, violates the 28 Fourteenth Amendment. Appellants’ only arguments upon this fundamental issue are that the legislature might have done better if more time had been avail able, and that over a long period the growth of the population may remove the injustices. The first point is obviously unsound: if the discrimination violates the complainants’ constitutional rights, the Court will not render an adjudication that there is no violation because the continuing wrong was initiated in a hurry. I f more time is needed, it can be afforded in framing the decree. The second point is equally un tenable : the possibility that time and changes in popu lation may, at some remote day, eliminate the inequalities is not the slightest justification for toler ating the present deprivation of constitutional rights. The apportionment of seats in the Delaware Senate is equally capricious. The State constitution prior to the 1963 amendment created 7 senatorial districts in Few Castle County, each of which was allocated one senator, and 5 districts in both Kent and Sussex Coun ties, each of which was apportioned one senator, mak ing a total of 17. The 1963 amendment gave Kent and Sussex two additional senators apiece, but it re tained the old senatorial districts and superimposed two new floterial districts on each county.9 Although 9 Rather than disturb the existing districts in Kent and Sussex Counties, the Delaware Legislature gave each county two additional Senate districts by having part of each county elect one o f its additional senators and the other part elect the other, in addition to the Senators from the five existing dis tricts. We have difficulty in following appellants’ assertion (Brief, p. 9) that the districts were formed by dividing the 29 each county is now allocated an equal number of senatorial districts in an apparent effort to justify the gross discrimination in per capita representation by reference to the United States Senate, the differ ences are plain. Delaware’s senators do not repre sent the counties but geographical districts within the counties. The composition of the districts, within the counties, has no intelligible basis, as shown by the following table (R. 385, 505) : State into eastern and western halves, for the new seventh dis trict in Kent County and the new sixth district in Sussex County stretch from the eastern half to the western half of the State. As a result, every voter in Kent or Sussex County votes for two senators: one from the pre-1963 district and one from the new floterial district in which he lives, while in New Castle County a citizen votes for only one senator (R. 386-387, 516). The district court held this discrimination to be unconstitu tional. We pretermit the question because the apportionment is unconstitutional however the question be decided. I f each of the three counties had the same population and the districts within each county contained the same numbers of people, a mathematical argument could be made for the proposition that there was substantial equality. This can best be shown by an example: I f each county had 70,000 people and seven senators, each voter in New Castle would be in a district of 10,000 people and would therefore elect 1/10,000 of a senator. On the other hand, voters in Kent and Sussex Comities would each be in two districts, one with 14,000 people and one with 35,000. They would therefore elect 1/14,000 of a senator plus 1/35,000 of a senator, which is 1/10,000 of a senator. We are not pre pared to say, however, that the mathematical equality means that there is no substantial discrimination. 714-431— 63-------3 30 New Castle County Kent County Sussex County 1 - ___ 52,131 1. __ ___ 23,890 1_________ 16,080 2_______ 43, 696 2.- ____ 7,077 2 ____ 14,755 3 ___ 64, 345 3. 17,005 3_ _ 16,150 4 _ ____ 64, 820 4.- ____ 7,582 4_ __ _ 9, 580 5 _____ 63, 734 5. 10,097 5 ____ 16, 630 6 _ 14, 543 6 1________(46,063) 6 1 (38,975) 7_ _______ 4,177 7 1— _ (19,584) 7 1 ___ ___ (34,220) Total____ 307,446 Total__ 65, 651 Total__ 73,195 1 These are floterial districts which, between them, cover the entire county. See pp. 28-29, note 9 above. Voters in the most overrepresented district in New Castle County have more than three times the repre sentation o f voters in any other New Castle district; they have more than 15 times the representation of the voters in the three most underrepresented districts. Voters in the second most overrepresented district in New Castle County have four times the representation of voters in three other districts. In Kent County dis parities run in the ratio of 2 and 3 to 1. In Sussex County one district has iy 2 times the representation of any other. Every district in Sussex County has six times the representation of the voters in the three most populous districts in New Castle County. The result of the disingenuous effort to make the apportionment o f the Delaware Senate look like the composition of the Senate of the United States was greatly to increase the overrepresentation o f the voters of Kent and Sus sex Counties. We can discover no explanation for the idiosyn crasies in the senatorial apportionment and districting. Appellants suggest no explanation whatever, save the obviously false analogy to the United States Senate. 31 The analogy would be false even if the relationship o f a county to a State were like that o f a State to the United States. In the United States Senate, the Senators represent their States and are chosen by the people thereof. In the Delaware Senate, the Senators come, not from the counties, but from geographical dis tricts within the counties and are chosen by the people of the districts. There is nothing in the composition of the United States Senate remotely comparable to Delaware’s gross and capricious discrimination in the representation of voters in different senatorial districts within a single county. Consequently, there is no need to discuss here whether the so-called “ federal analogy” is pertinent in a State which apportions one branch of the legislature according to population and arranges to have its counties represented in the upper house in the same maimer as the States are represented in the United States Senate.10 The portions of the opinion below dealing with this subject are no part of the decree. It follows that the apportionment of representation in the Delaware legislature denies the complainants equal protection of the law. 10 See Brief for the United States in Maryland Committee for Fair Representation v. Tames, No. 29, this Term, pp. 73-82. I I I EVEN IF TH E APPARENTLY CAPRICIOUS APPORTIONMENT OF THE DELAWARE LEGISLATURE HAD AN INTELLIGIBLE FOUNDATION, TH E DISCRIMINATION WOULD VIOLATE TH E EQUAL PROTECTION CLAUSE UNLESS BASED UPON CRI TERIA RELEVANT TO LEGISLATIVE APPORTIONMENT 1. The principle Since the districting and apportionment of repre sentatives in elections for both branches of the Dela ware legislature apparently has no intelligible founda tion of any kind, it is probably superfluous to discuss which criteria of systematic differentiation might be constitutionally acceptable and which would violate the Fourteenth Amendment. But because some rationalization may be suggested, and because the oral argument of the previous reapportionment cases re fined (but did not change) our analysis, we reiterate the proposition that a discriminatory apportionment violates the equal protection clause unless the differ ences in per capita representation bear a rational relationship to permissible objectives of legislative apportionment. See Briefs for the United States in Maryland Committee for Fair Representation v. Tawes, No. 29, this Term, pp. 39-43, and WMCA, Inc. v. Simon, No. 20, this Term, pp. 20-33.11 11 11 Strictly speaking, the crazy-qnilt test is probably insufficient standing alone. It must be applied in company with an addi tional rule, either implicit or explicit, that separates the per missible and impermissible bases of discrimination. I f a State may prefer farmers over store-keepers in matters o f apportion ment, bankers over wage-earners, port cities over manufacturing centers, cities with a population of between 20,000 and 50,000 over both towns with less than 20,000 and cities with more than 32 33 In those briefs we suggested but did not adequately articulate the critical distinction between permissible12 and impermissible objectives of legislative apportion ment. The distinction is between (a) rules serving the purpose of making representative government work better, the operation o f which may have the collateral consequence of creating per capita inequali ties in representation, and (b) rules whose only func tion is to create classes of voters with preferred political rights disproportionate to their number, which we say are impermissible. A legislature need not be simply a mirror to public sentiment like a Gallup poll. A legisature is a de liberative body, and deliberation requires knowledge. For that reason it might be thought to “make repre sentative government work better” to take steps to ensure that representatives in one house of the legisla ture come from geographical districts small enough for them to know the distinct problems of every 50,000 and so forth, and if the rationalization can be supplied after the event because the Court cannot plumb the legislatures’ true motives, then anyone with sufficient time and imagination could always find somewhere in the jumble of facts and potential policies some rule or compromise between many con flicting rules that would rationalize the apportionment. Com pare the dissenting opinion of Mr. Justice Harlan in Baker v. Carr, 369 U.S. 330, 333-344. In a case like Baker v. Carr or the instant case, there is hardly need to formulate the implicit limitation upon the acceptable kinds of rationalization, for the apportionments are so capri cious and internally inconsistent that the defendants may be required to come forward with any explanation in terms of State policy and judgment will go against them if they fail. 12 As repeatedly pointed out in these briefs, we assume them to be permissible for the purpose of these cases. See Brief for the United States in Maryland Committee for Fair Representa tion v. Tawes, No. 29, this Term, pp. 24-25. 34 locality, whether a unique local industry or a special need for schools or roads and bridges. Nantucket and Martha’s Vineyard in Massachusetts are excellent ex amples. Geographically small districts might also be thought to enable the people to know their representa tives better. Again, a legislature might conceivably be better informed and more representative if the apportionment recognized political subdivisions with an historic coherence and identity of their own. In an earlier age local political subdivisions formed a meeting place for discussion and the interchange of ideas, and the representative who went up to the legislature was speaking for the consensus in that com munity rather than for so many individuals. Even today the town hall or county court house may serve as the nerve center o f a political organism attuned to local needs and opinions. A legislature may also be better informed and more representative, in the sense that it is closer to its constituents, if there is a measure o f continuity and stability in political arrangements; thus a State may limit the frequency with which it reapportions the legislature. Since effective political organiza tions help to make a legislature representative in this sense, a State might desire to conform the boundaries o f election districts to the boundaries of local political subdivisions in order to reduce the number o f geo graphical entities for which a separate organization must be created. The foregoing illustrations are suggestive, not ex clusive. We consider below (pp. 45-48) whether there is any limit upon a State’s freedom in making an 35 apportionment on the basis of rules that serve the foregoing functions. Our point here is that those rules are different from differentiations or classifica tions among voters whose only function is to give farmers more representation than wage earners, ship ping interests more representation than the manu facturing community, city dwellers more than suburbanites, or Protestants more than the Jews. The latter kind of distinction, when applied to po litical rights including representation, is the very antithesis of equality before the law. “ The creation by law of favored groups of citizens and the grant to them of preferred political rights is the worst of all discriminations under a democratic system of gov ernment.” South v. Peters, 339 TT.S. 276, 279 (Jus tices Black and Douglas dissenting). W e find the support for this proposition in four areas. First, the principle that the policy knows no distinction in the political worth of one man as op posed to another, or of one group as opposed to another, is a fundamental strain in our political philosophy. The long appendix to our brief in the Maryland case shows the extent to which the idea of equal repre sentation per capita was accepted in the Constitu tional Convention of 1789 and the ratifying conventions as a principle applicable to any govern ment dealing directly with its citizens. Two quali fications were widely accepted, but neither is relevant here. One was the principle that in a federation of sovereign States the States should have equal repre sentation. That principle is manifestly inapplicable 36 within a State and, at most, it would justify repre sentation by political subdivisions, not the creation of preferred classses. The other qualification was the view that there should be an upper house not directly chosen by the people. This was the theme of the remarks by Governor Randolph which the amicus brief o f the Attorneys General of several States in the Maryland case cites (pp. 36-37) along with a quotation from James Madison in an effort to show that the constitutional convention was not in agreement that State legislatures ought to be appor tioned according to population. Randolph’s views upon the indirect election of the upper branch have nothing to do with the geographical apportionment o f representatives elected by the people. His state ment on the latter point unequivocally supports our reading of the records. Appendix B to the govern ment’s brief in the Maryland case, pp. 51-52. The burden of the quoted passage from Madison is un clear, but he often insisted that representatives should be apportioned in direct ratio to population. Id. at 4, 15-16, 22-24, 27, 32, 38-39, 45, 48, 53, 60-61, 70-71, 117-118. The only proposals to weight apportion ment so as to preserve the political power of the original States were rejected.13 Id. at 40, 50, 53. The thrust of the Fourteenth Amendment is to ward the elimination of all class distinctions in the exercise of political and other civil rights. There are no qualifications upon the equal protection clause. It 13 For the possible convenience of the Court we have set forth in the Appendix an analysis of the historical materials in the brief o f the State Attorneys General. 37 has general application. It has been applied spe cifically to laws dealing with participation in demo cratic self-government. Nixon v. Condon, 286 U.S. 73; Gomillion v. Lightfoot, 364 U.S. 339, 349 (Mr. Jus tice Whittaker concurring) ; Gray v. Sanders, 372 U.S. 368. Nothing could he more inconsistent with the basic spirit of the Amendment than the creation o f special political classes with preferred voting rights or pre ferred representation. There is nothing to the contrary in the history cited by appellants. The constitutions of the States re admitted to the Union after the enactment o f the Fourteenth Amendment were all before the Court in Baker v. Carr, despite appellants’ claims o f original ity, for Air. Justice Frankfurter developed essentially the same arguments in his dissenting opinion in an effort to show that the Amendment is not concerned with legislative apportionment. 369 U.S. 186, 266, 315- 317. The Court found the history unpersuasive, doubtless because it fails to show that Congress either concerned itself with the apportionment provisions of the State constitutions or was aware of their practical consequences. In addition, nothing is more apparent than that the limits o f the guarantee of equal protec tion as applied in 1963 are not to be found in the con stitutions and laws of the several States during the Reconstruction Period. See pp. 52-56 below. Even if it were otherwise, appellants’ argument would be irrelevant to the point that we are now making—that our whole political history rejects the creation o f fa 38 vored political classes as a legitimate function o f elec tion laws or the apportionment of representatives.14 Although practice often fell short of the professed ideal,15 each stage in our history has brough the ideal of per capita political equality closer to reality. As the Court said in Gray v. Sanders, 372 U.S. 368, 379- 381: The concept of “ we the people” under the Con stitution visualizes no preferred class o f voters but equality among those who meet the basic qualifications. * * * * * The conception of political equality from the Declaration of Independence, to Lincoln’s Gettysburg Address, to the Fifteenth, Seven teenth, and Nineteenth Amendments can mean only one thing— one person, one vote. Occasionally, one finds a political figure praising the unique virtue of the rural voter or defending the pre ferred voting power of special interests, but we know of no eminent statesman or political philosopher who has avowed the view it is proper to weight the legis lature so as to assign to the farm interests, the wage 14 We have shown above the irrelevance o f appellants' his torical arguments as an answer to the fundamental defect charged against the Delaware apportionment—that it introduces gross and capricious discrimination into the apportionment of both houses o f the legislature without even an intelligible foundation. See pp. 23-31 above. We show below that the history, even if otherwise relevant, does not support such gross inequalities in both houses of the legislature as exist under Delaware’s current constitution. 15 There is scant support in practice for the extreme discrimi nation in many States today, which is chiefly a consequence o f the more rapid growth o f cities and suburbs. 39 earning interest, the urban interest, or any other class o f voters political power disproportionate to its number. Second, the practice in framing State constitutions bears out this contention. The central principle, quite uniformly, has been to base representation primarily upon population.16 There are qualifications and lim itations but they are nearly all consistent with the basic distinction which we are emphasizing. The lim itations have been addressed, with rare exceptions, to making the legislature more representative by avoid ing the creation of excessively large districts, by recognizing the coherent and independent identity of historic political subdivisions, by preserving a meas ure of stability, etc. No State constitution has ever openly espoused the view that it is a proper function o f apportionment to allocate power to different social or economic classes out of proportion to their numbers. Few have provided for discrimination against popu lous areas merely because they are populous.17 The practice has sometimes been different, but our short comings are not the measure of constitutional protection. 16 A brief survey of State constitutions is set forth in our brief in Maryland Committee for Fair Representation v. Tawes, No. 29, this Term, pp. 30-31. See also the dissenting opinion of Mr. Justice Frankfurter in Baker v. Carr, 369 U.S. 266, 309-318. 17 New York systematically discriminates against citizens in counties having more than 6 percent of the State’s popula tion. Maine, Oklahoma, Pennsylvania and Rhode Island limit the number of representatives a single city or town may have in one or both branches of the legislature. The Pennsyl vania limitation has no practical effect because no city, on the basis o f population, is entitled to more than the limit. 40 Third, it is plain that in closely related areas the equal protection clause prohibits the kind of discrimi nation which we are discussing. No one would argue that a State might deny the vote to bankers, or wage earners, or lawyers, or farmers. It seems equally clear that in electing a legislature it cannot, consistently with the Fourteenth Amendment, give 10 votes to every banker, 5 to every wage earner, 1 to every lawyer, or 50 to every farmer. How then can the State accomplish the same discrimination by giv ing some classes greater representation per capita and others less, in apportioning the seats ? Fourth, following the last point a step further, we submit that Gray v. Sanders is decisive of the point at issue. There, the Court squarely rejected the argu ment that the votes in rural areas could be weighted more heavily than votes in urban areas in choosing a governor (372 U.S. at 379) : Once the geographical imit for which a rep resentative is to be chosen is designated, all who participate in the election are to have an equal vote— whatever their race, whatever their sex, whatever their occupation, whatever their income, and wherever their home may be in that geographical unit. I f the rural voter is not entitled, by reason o f his interests or residence alone, to have the State give him a greater voice in the selection of a governor than an The Rhode Island Supreme Court has held the Rhode Island limitations unconstitutional. Sweeney v. Notte, 183 A. 2d 296. Georgia, Maine and Texas give somewhat less representation per capita as the population of an area increases. 41 urban voter, then the rural voter is not entitled, by reason of that fact alone, to have greater representa tion in the legislature. This is not to say that Gray v. Sanders is decisive upon all aspects of the State ap portionment cases. It is different because a legisla tive apportionment, unlike an election within a given area, may serve functions, such as those we have out lined, the achievement of which requires some depar ture from per capita equality of representation. An apportionment that serves such a purpose is not in valid merely because it happens to result, if measured by per capita representation, in classifications more or less corresponding to economic, occupational, or geo graphic groups. But Gray v. Sanders necessarily holds that the equal protection clause prohibits giving some classes of voters more weight than others on the ground that they have unusual virtue or special importance.18 The point may be clarified by a simple example. Assume a State with four counties, each with a popu lation of 50,000 and also identical in every other re spect save that one county, Poliak County, is pastoral : 18 We have discussed above (pp. 16-23) the reasons for reject ing any distinction based upon an argument that although 'per capita comparisons measure equality in an election within a single constituency they do not truly measure equality or in equality in a representative body. 42 Spritzer County Population _ _ 50,000 Friedman County P opu lation KÔ OOf) Unit Votes__ ____ 10 Representatives __ 10 Unit V otes____ ____ 10 Representatives_____ 10 Barnett County Populat ion _ 50, 000 Poliak County {Pastoral) P opu lation KOj 000 Unit Votes 10 Representatives _ 10 Unit Votes- ___ _ 31 Representatives_ _ 31 Gray v. Sanders squarely holds that the discrimina tion inherent in giving the 50,000 people o f Poliak County 31 unit votes in the election of a governor as opposed to the people of Spritzer, Friedman and Barnett Counties, who receive only 10 unit votes apiece for 50,000 people, cannot be justified by an alleged State policy of fostering the pastoral way of life. It follows that the policy is equally irrelevant as a justification for giving Poliak County 31 repre sentatives in the legislature as opposed to the 10 rep resentatives apiece allocated to the other counties despite their identical size and population. The equal protection clause bars laws which serve only the function of creating o f favored political classes whether it be in the choice of a governor or the selec tion of a legislature. We recognize that such distinctions are not uncon stitutional in every context. In enacting tax or regu latory laws and in granting appropriations or other benefits, a State is free, within reason, to encourage one industry or discourage another, to protect the wage earner against the employer or the employer 43 against the union, etc. It may take into account the special problems of the farmer as opposed to the con sumer and grant exemptions or privileges not accorded to others. But that kind of classification, we submit, is not permissible in respect to political rights. The Fourteenth Amendment guarantees the rich and the poor, the banker and the wage earner, the farmer and the city dweller, like the Catholic and the Protestant, equality before the law defining the opportunities for participation in self-government. Precedent amply sustains the view that differentia tions which may be permissible in one context become irrelevant in another. A man’s income may be rel evant for purposes of taxation, but since it has no bearing upon his guilt or innocence when charged with crime, a State cannot give an appeal to the man who can afford to buy a transcript but deny it to one who is too poor. Griffin v. Illinois, 351 U.S. 12. The nice lines between larceny and embezzlement may be constitutionally acceptable in the ordinary adminis tration of criminal justice but not in relation to so fundamental a human right as procreation. Skinner v. Oklahoma, 316 U.S. 535. 2. Application to the instant case W e can discover no policy to support the capricious discriminations in the apportionment of both branches of the Delaware legislature. A fortiori there is none that can be defended as relevant to the permissible purposes of legislative apportionment. Delaware’s apportionment nowise resembles the State constitu tions that guarantee each county at least one repre- 44 sentatiVe, or provide for an upper branch made up ' of representatives of the counties, chosen by the equal votes of all the county’s citizens. Appellants state (Br. 9) that the two additional senators given to Kent and Sussex Counties by the 1963 amendment were allocated to the eastern and western halves of each county. Since one new dis trict in each county extends across the whole State, we question the accuracy of the assertion. Even if accurate, it is irrelevant. Surely the distinction be tween the eastern and western halves of Kent County is not an adequate justification for constituting the districts so that District 7 with a population of 19,584 has the same representation in the Senate as District 6 with a population of 46,063, a ratio of worse than 2 to 1. And, of course the method of appor tioning the two additional seats given to each of the two counties does nothing to justify the gross inequali ties not only between the counties but among the dis tricts within the counties. Nor is there any indication that senators or repre sentatives are distributed on the basis o f area in an effort to give each part of the State representation in the legislature. The contrary appears clear from the widely differing size of the districts. Moreover, dis tricts large in area often have greatly more popula tion than those much smaller, yet each has only one representative. For example, the second representa tive district of Kent County has well over twice the population and size of that county’s seventh district. The tenth representative district of Sussex County has almost three times the population and size o f that 45 county’s ninth district. The first senatorial district of Kent County has one-third more population than the third district although the former is approximately four times larger. The boundaries of the ancient “ hundreds” (see p. 7 above) play some part in the districting, but it is plain that they neither explain nor justify the in equalities. Sometimes hundreds or portions of hun dreds are combined; sometimes they are subdivided. Furthermore, it is plain that the hundreds have no governmental functions or other meaningful identity that could be supposed to give its representation as such a useful function in making representative government work better. IV TH E APPORTIONMENT OF TH E DELAWARE LEGISLATURE VIO LATES TH E EQUAL PROTECTION CLAUSE BY SUBORDINAT ING POPULAR REPRESENTATION TO TH E REPRESENTATION OF POLITICAL SUBDIVISIONS TO SUCH A DEGREE AS TO CREATE GROSS INEQUALITIES AMONG VOTERS AND GIVE CONTROL OF BOTH HOUSES OF TH E LEGISLATURE TO SM ALL MINORITIES OF TH E PEOPLE Apart, from the fatal defects already discussed, the apportionment o f the Delaware legislature is unconsti tutional because any otherwise permissible objectives that disparities in per capita representation might be supposed to secure are insufficient to justify such gross inequalities and total disregard for the princi ple of majority rule as exist in the selection of rep resentatives to both the Senate and House of Repre sentatives. As we have shown in our previous briefs, objectives that might furnish acceptable justification 714- 431— 63------------4 46 for some variations from equal representation per capita19 become so relatively inconsequential as to render the discrimination arbitrary and capricious where the inequalities are grosser and the principle of majority rule has been totally submerged in both houses of the legislature. See especially Brief for the United States in Maryland Committee for Fair Rep resentation v. Tawes, pp. 46-50, 57-73. Measured by any reasonable standard, the appor tionment o f the Delaware legislature creates gross in equalities among voters and allows a small minority to control the legislature. We have seen above (pp. 24-30) the wide disparities of voting power among persons residing in the various counties and senatorial and representative districts. The disparities are as high as 15 to 1 in the Senate and 12 to 1 in the House of Representatives. There are gross disparities be tween districts within the same county as well as across county lines. A majority of the members of the House is elected by persons residing in districts with only 28 percent of the population. A majority of the Senate is elected by persons residing in districts with only 21 percent of the population—a reduction o f 1 percent 19 Since the instant case, like the State apportionment cases, argued earlier this Term, can be decided without ruling upon whether the Fourteenth Amendment requires substantially equal representation per capita in both houses of a State legislature, we assume here, as we have in those cases that there may be permissible objectives of legislative apportionment that would justify some departure from equality per capita. The asssump- tion is made arguendo reserving further judgment until the issues are presented. 47 as a result of the 1963 amendments. The Delaware legislature considered as whole is worse apportioned than any of the States involved in the State apportion ment cases argued last month. Indeed, measured by the percentage of votes able to elect a majority of the two houses, it is the third worst apportioned legisla ture in the country.20 The discrimination in both houses seems to run almost entirely in favor of the same areas. For ex ample, the two least populous house districts in the State (numbers 14 and 15 in New Castle County), comprise the least populous senatorial district (num ber 7 in New Castle County). The second and third least populous districts in Kent County (niunbers 3 and 4), form the least populous Senate district in that county (number 2). The least and fifth least populous house districts in Kent County (numbers 6 and 9) form the second least populous Senate district in the county (number 4). The least and fifth least populous house districts in Sussex County (numbers 7 and 8) form the county’s least populous Senate dis trict (number 4). Conversely, the 20 most populous house districts (New Castle County numbers 2A, 2B, 3, 5A, 5B, 6A, 6B, 6C, 6D, 7A, 7B, 7C, 8A, 8B, 9, 10A, 10B, 10C; Kent County number 2; Sussex County number 10) are in seven of the eight most populous Senate districts (New Castle County numbers 1, 2, 3, 4, 5; Kent County number 1; Sussex County num 20 The worse apportioned legislatures from this standpoint are Florida and Nevada. Maryland is approximately the same as Delaware. This ranking has been calculated by adding to gether the percentage of the voters needed to elect a majority of each of the two houses of the 49 bicameral State legislatures. 48 ber 5 ) ; only the seventh most populous Senate dis trict (Kent County number 3) is formed of less pop ulous house districts.21 In short, the inequality in each house adds to the inequality in the other. W e submit that, where, as here, a small part of the people is given complete control of the legislature, the apportionment violates the Fourteenth Amendment regardless of any justi fication that can now be advanced. The justifications sometimes put forward by larger States would carry less weight here even if they fitted the apportionment. Delaware is the second smallest State in the country. Seventy-one senators and forty-one representatives would, if apportioned directly on the basis of popula tion, still represent extremely small districts. Kent County, the least populous county, would still be en titled to three senators and six or seven representa tives for an area of less than 595 square miles—an average of less than 100 square miles per representa tive. It is plain that the analogy appellants seek to draw between the apportionment of the Delaware legislature and the Congress of the United States is no answer to the point we have been making. There is plainly no analogy, among other reasons because the actual apportionment of the Delaware legislature bears no resemblance to the constitutional apportion ment of the Congress. 21 In considering the Senate districts, we have not included the Senate districts created by the 1963 amendments in Sussex and Kent Counties which overlap with the earlier districts. 49 First, the Constitution of the United States pro vides an obviously intelligible basis for apportioning Congress. Each State is entitled to two members of the Senate. Art. I, Sec. 3. The members of the House of Representatives are apportioned on the basis of population, except that each State is guar anteed a minimum of one. Art. I, Sec. 2. In con trast, as. we have seen above (pp. 23-31), the apportionment of both houses of the Delaware legis lature is a crazy-quilt, having neither rhyme nor rea son. Both the Senate and House districts vary widely in population without any intelligible ex planation. Second, the limitations on apportionment based strictly on population in the Constitution concerning the United States Congress are relevant to the repre sentative process. They are intended to give each State, i.e., each political subdivision, equal represen tation in the Senate and minimum representation in the House of Representatives. Delaware’s apportion ment has no comparable objectives. Third, even if it were not a crazy-quilt, the Dela ware Senate is not apportioned like that of the United States. While each county has seven senators, the Senate is actually elected from districts within the counties rather than by the counties themselves. These districts are merely arbitrary units formed for election purposes which are unequal in population. Thus, unlike the United States Senate, the Delaware Senate does not consist o f representatives of govern 50 mental subdivisions which are given equal numbers of legislators who represent the subdivision on a cor porate basis. Instead, the Delaware Senate is com posed of representatives from individual districts with no corporate existence or governmental significance apart from elections. Fourth, the apportionment o f the Delaware House o f Representatives is so lacking in relation to popula tion that a numerical majority comes from districts containing only 28 percent o f the people. The history o f the Constitution and its ratification shows that the framers intended that the federal House of Repre sentatives would be apportioned between the States on the basis o f population. While each State is as sured of at least one Representative, it is clear that the framers expected that each State would have enough population to entitle it to one Representative without this guarantee. Thus, the Constitution sets the minimum ratio of Representatives to popula tion at one to 30,000 (Art. I, Sec. 2), which was be low the population of any o f the States at the time. And the first apportionment of the House o f Repre sentatives, which is contained in the Constitution itself (Art. I, Sec. 2), gives more than two representatives to every State except Rhode Island and Delaware, and those two States would have been entitled to such representation on the basis o f their population. Today only four States have less than 1/435 of the population.22 All have over half the ratio and would be entitled to a representative, even without the mini 22 Vermont, Wyoming, Nevada, and Alaska. 51 mum guarantee, under the equal proportions method of allocating representatives which Congress now follows. 2 U.S.C. 2a, Thus, the federal House of Representatives, in effect, is apportioned by the Consti tution among the States almost exactly in conformity with population. It takes representatives from States with almost 50 percent of the population to constitute a majority of the House.23 The apportionment of the Delaware House of Rep resentatives between counties is based only partially on population. New Castle County is guaranteed fif teen representatives and Kent and Sussex ten apiece; thus, 35 of the present 45 representatives are allocated wholly without regard to population.24 Furthermore, the 35 representatives are distributed within the coun ties with no real regard for population, being elected from grossly unequal districts arbitrarily established 23 Unfair districting within the States by the State legisla tures, however, has resulted in having a majority o f repre sentatives elected from districts having only 42 percent of the population. Such unfair districting is probably unconstitu tional. See the government’s brief in Wesberry v. Sanders, No. 22, this Term, pp. 30-35. 24 In addition, the other ten representatives are allocated according to the population of arbitrarily formed election dis tricts rather than on a basis of the population of the county as a whole. Thus, a county could be given additional repre sentation if one of its districts was populous even though the other districts were so unpopulous that the county as a whole would be entitled to no more representation. At present, this has not occurred since only the most populous county, New Castle, has received additional representation under the 1963 amendment. However, if the second representative district of Kent gains less than 5,000 people it will receive an additional representative although the county as a whole is overrepre sented now. 52 for election purposes. The net result is that only 28 percent of the people of Delaware live in districts electing over half the State House of Representatives. The federal analogy, therefore, does not apply, since Delaware does not apportion either house of the legis lature on the basis o f population.25 Nor can anything be found in the history of the readmission of the seceded States that supports the notion that the equal protection clause permits the gross discrimination in per capita representation found in both houses o f the Delaware legislature. In the first place, as the history set forth by the appellants shows, Congress did not even advert to the problem of apportionment in readmitting the Southern States.28 For, at that time, unlike the time 25 Even if the apportionment of the Delaware General Assem bly was factually similar to the apportionment o f Congress, the federal analogy would not apply. The equality o f repre sentation by States in the United States Senate recognized that the new government was in part a continuation of the old confederation o f sovereign States. See our brief in the Mary land case, pp. 73-82. Whether or not the three counties of Delaware were sovereign for three months in 1776 as appel lants contend, they are today “nothing more than political sub divisions of [the] State [s]” with “ restricted powers and duties” delegated by the legislature. State v. Warwick, 108 A. 2d 85, 89 (Del. Superior Ct.). And of course the application of the Fourteenth Amendment depends on present facts and circum stances. Brown v. Board of Education, 347 U.S. 483, 492-493. 20 Appellants state (Br. 67) that, “ the Federal pattern for a legislature was discussed in the course of the debates.” The relevant “ excerpts from the debates” included in appellants’ brief (pp. 112-132) contain no discussion whatever of the apportion ments provided in any o f the State constitutions under exami nation. Perhaps appellants’ are referring to a speech by Senator Sumner (App. Br. 119-121) which discusses the equality o f the States under the 1789 Constitutions, and par- 53 of the adoption o f the original Constitution and again during the last few decades, apportionment was not an issue of major importance. There is no way to know whether, if the issue had been squarely posed, Con gress would have considered that the equal protec tion clause applied to apportionment of State legis latures at all and, if it did, what standards were appropriate. Second, this Court long ago made it clear that the protection afforded by the Fourteenth Amendment cannot be limited by historical research into the con scious intent of its framers. In Brown v. Board of Education, 347 U.S. 483, massive briefs and “ [r]ear- gument [were] largely devoted to the circumstances surrounding the adoption of the Fourteenth Amend ment in 1868. [They] covered exhaustively consider ation of the Amendment in Congress, ratification by the states, then existing practices in racial segregation, and the views of the proponents and opponents of the Amendment.” Id. at 489. Despite this extensive historical analysis, the Court found that history was “ inconclusive” (ibid.) : The most avid proponents of the post-War Amendments undoubtedly intended them to remove all legal distinctions among “ all per- ticularly equal representation in the Senate. However, Sen ator Sumner does not even suggest that this equality among sovereign States is relevant to the apportionment of State legislatures. Indeed, he explicitly stated that he was speaking on an entirely different subject ( “What I have to say today will be confined to a single topic. I shall speak of the validity and necessity of fundamental conditions on the admission of States into the body of the Nation * * * (emphasis in original)). 80 Cong. Globe 3024. 54 sons born or naturalized in the United States.” Their opponents, just as certainly, were antago nistic to both the letter and the spirit of the Amendments and wished them to have the most limited effect. What others in Congress and the state legislatures had in mind cannot be deter mined with any degree of certainty. Furthermore, the Court went on to say that “ [i]n ap proaching this problem, we cannot turn the clock back 'Oto 1868 when the Amendment was adapted, or even in 1896 when Plessy v. Ferguson was written.” Id. at 492. The silence as to apportionment leaves the historical issue in this area even more uncertain. The history that appellants invoke was all brought to the Court’s attention in Baker v. Carr by the dissenting opinion of Mr. Justice Frankfurter. The argument was rejected. Finally, appellants’ history, even if otherwise per suasive, does not sustain the present Delaware appor tionment. At most, the history shows that some de viations from per capita representation in the re admitted States provoked no objection. There is no evidence that Congress knew of them; nor that it knew whether they were large or small. In point of fact, five of the State constitutions, as appellants admit (Br. 57), provided for the apportionment of both houses on the basis of population. In the other five the disparities were small compared to those existing today in Delaware. The South Carolina Con stitution of 1868 provided for a Senate composed of one senator for each county, except Charlestown County was to receive two, and a House of Repre sentatives apportioned according to population except 55 that each county was given one. As a result, 36 and 44 percent o f the people lived in counties electing a majority o f the two houses. The Georgia Constitution of 1868 provided no stand ard for the reapportionment o f either House. The initial apportionment specified in the Constitution grouped three counties into each single-member sena torial district and gave three seats in the House to the six largest counties, two seats to the next 31, and one to the rest. Consequently, 36 and 37 percent of the people lived in counties electing a majority of the two branches. In Florida, the 1868 Constitution apportioned the Senate by districts without stating any standard and gave a representative for each county and an addi tional representative for each 1,000 “ registered votes.’ ’ A majority of both houses could be elected by persons residing in counties with 33 and 27 percent of the people, respectively. The Louisiana Constitution of 1868 provided no standard for apportioning the Senate, but apportioned the House of Representatives on the basis of popula tion with each parish receiving at least one member. Under the initial apportionment specified in the con stitution, 48 and 40 percent of the people lived in counties electing a majority of the two houses. Finally, the Virginia Constitution of 1870 provided no standard for apportionment. The initial appor tionment specified in the Constitution resulted in 44 and 40 percent of the people residing in counties electing a majority o f the legislature. Except for Louisiana, the percent figures are based on the 1870 census because the Civil W ar caused sig nificant relocation of population. In no State did the 56 malapportionment approach the severity of that now existing in Delaware. Y TH E INJU N CTION ORDERED BY TH E DISTRICT COURT IS NOT AN ABUSE OP DISCRETION The district court, after holding unconstitutional the provisions of the Delaware Constitution appor tioning seats in the State legislature, entered a declar atory judgment and later enjoined the appellant election officials from holding elections thereunder. Since, as we have shown above, this determination was correct, the district court was acting well within its discretion in enjoining the existing apportionment. The traditional remedy when a State statute is held unconstitutional is to enjoin further action thereunder. E.g., Ex parte Young, 209 TT.S. 123. The State and lower federal courts have in numerous cases used this remedy in the field of apportionment. E.g., Sims v. Frink, 208 F. Supp. 431 (M.D. Ala.), pending on appeal sub nom. Reynolds V. Sims, Nos. 23, 27, 41, this Term; Davis v. Mann, 213 F. Supp. 577 (E.D. Va.), pending on appeal, No. 69, this Term; Thigpen v. Meyers, U.S. D.C., W.D. Wash., decided May 3, 1963, pending on appeal, No. 381, this Term; Scholle v. Secretary of State, 367 Mich. 176,116 N.W. 2d 350; Parker v. State, 133 Ind. 178, 32 N.E. 836; Denny v. State, 144 Ind. 503, 42 N.E. 929; Brooks v. State, 162 Ind. 568, 70 N.E. 980; Ragland v. Anderson, 125 Ky. 141, 100 S.W. 865; Armstrong v. Mitten, 95 Colo. 425, 37 P. 2d 757; Stiglitz v. Schardien, 239 Ky. 799, 40 S.W. 2d 315; Rogers v. Morgan, 127 Neb. 456, 256 N.W. 1; State ex ret. Lamb v. Cunningham, 83 Wis. 90,53 N.W. 35; State 57 ex rel. Attorney General v. Cunningham, 81 Wis. 440. 51 N.W. 724; Asbury Park Press, Inc. v. Woolley, 33 N.J. 1,161 A. 2d 705, 713-714. Appellants contend (Br. 98-110) that the Delaware legislature should have been given the opportunity to reapportion itself before the injunction was issued. The legislature had already had the opportunity. After indicating doubt about the existing Delaware apportionment, the court in a pre-trial order gave the legislature time to act (R. 66-76).27 28 The deadline was fixed to meet the requirements of Delaware’s election laws (R. 75-76). The legislature, in response, passed a constitutional amendment at two successive sessions.23 In the Senate, the amendment gave the already-over represented counties, Kent and Sussex, two additional members apiece. The existing wide disparities in senatorial districts within the counties (see pp. 28-30 above) were left unchanged and the percentage of people living in districts which elected a numerical majority of the Senate was reduced from 22 to 21 27 Several other courts have likewise delayed determining the constitutionality of State apportionment so the legislature could act. Lein v. Sathre, 205 F. Supp. 536, 540 (D. N .D .); Magraw v. Donovan, 163 F. Supp. 184, 187-188 (D. Minn.) ; MiJcell v. Rousseau, 183 A. 2d 817, 823 (Yt. Sup. Ct.) ; Asbury Park Press, Inc. v. Woolley, 33 N.J. 1, 161 A. 2d 705, 714—715; In re Legislative Reapportionment, 374 P. 2d 66, 71-72 (Colo. Sup. C t.); Butcher v. Trimarchi, 28 D. & C. 2d (Pa.) 537, 542. 28 The Delaware Constitution provides two procedures for amendment, both of which depend on the legislature. First, the' legislature may itself adopt an amendment by passage by two-thirds vote at two consecutive sessions. Art. 16, § 1. Second, the legislature by two-thirds vote may place on the ballot the question whether to hold a constitution convention and, if a majority of the voters so decide, a convention is held. Art. 16, § 2. 58 percent. The amendment gave some additional repre sentation in the lower house to populous districts in New Castle County but left untouched many of the existing disparities between and within counties (see pp. 24-28 above). The percentage o f people living in districts electing a majority of the House of Repre sentatives was raised only from 19 to 28 percent. In short, the Amendment hardly constituted a seri ous effort by the legislature to give fair representation in either house to all the people of Delaware. The district court in these circumstances was entitled to decree a reapportionment, as was done in the Alabama case now before the Court. Sims v. Frink, 208 F. Supp. 431, pending on appeal sub nom. Reynolds v. Sims, Nos. 23, 27, 41, this Term; accord, Moss v. Burk hart, U.S. D.C., W.D. Okla., decided July 17, 1963; Sobel v. Adams, 208 P. Supp. 316, 318 (S.D. Fla.) ; Toombs v. Fortson, U.S. D.C., N.D. Ga., decided September 5, 1962. See also Fortner v. Barnett, No. 59,965, Chancery Court, First Judicial District, Hinds County, Mississippi; Stevens v. Faubus, 354 S.W. 2d 707, 711 (Ark. Sup. Ct.). The court below stopped short of specifying an apportionment. While it indicated guidelines for testing the constitutionality of a State legislative apportionment in its opinion, they are not part of the order. The decree merely enjoins holding a. new elec tion under the unconstitutional apportionment, while allowing the legislature to reapportion itse lf. I t is true that the legislature will not be able before the 1964 elections to pass a constitutional amendment at two consecutive sessions. B u t, since the existing con 59 stitutional provisions have been held invalid, there is no reason why the new apportionment cannot be en acted by statute. No provision in the State constitu tion prohibits apportionment by statute where no valid apportionment provision exists in the constitution it self. Furthermore, the statutory apportionment need be only temporary. By passing a valid constitutional amendment now, the legislature elected in 1964 can adopt it by approving it again. In these circumstances, the district court plainly did not abuse its discretion. The Delaware appor tionment violated the Fourteenth Amendment. The court decreed the narrowest remedy possible to correct an unconstitutional provision o f a State constitu tion— an injunction against its continued operation. The court could provide time for a constitutional amendment without permitting the election of still another legislature in violation of the complainants’ constitutional rights. The State still retains freedom to shape its own system of apportionment subject to the requirements of the Fourteenth Amendment. CONCLUSION For the foregoing reasons, we respectfully submit that the decision of the district court should be affirmed. Ajrchibald Cox, Solicitor General. B ruce J. Terris, Assistant to the Solicitor General. R ichard W . Schmude, Attorney. December 1963. A P P E N D IX 1. In a brief filed as amici curiae in the four State apportionment eases argued earlier this Term, four teen State Attorneys General dispute (pp. 19-32) the government’s historical analysis of the Constitutional Convention of 1787 as supporting the constitutional requirement that State legislatures be apportioned on the basis of population. First, they say (Br. 20-24, 26) that the framers repeatedly analogized apportion ment of the State legislatures and the apportionment of the legislature in the new “national” government; that the apportionment of the State legislatures was fre quently not based on population, such as New Jersey which gave each county a member of the State Sen ate; and that therefore similar apportionments today may properly be analogized to the United States Senate. It is true that the framers did refer to the apportionment of the State legislatures. But, with only one exception, they did not approve apportion ments which gave some voters more representation than others. The amici’s own examples (Br. 22) prove this point. Madison’s statement that “ there was the same rea son for different numbers of representatives from dif ferent States, as from Counties of different extents within particular States” is part o f a speech opposing equal representation from every State in the new legislature. Madison argued that the new nation would no longer be “ a federal one among sovereign States” but a “national Governt.” I Records of the (61) 714—431— 63------5 62 Federal Convention (Farrand ed., 1911), p. 37. The entire speech, insofar as is relevant is : 1 [WJhatever reason might have existed for the equality of suffrage when the Union was a fed eral one among sovereign States, it must cease when a national Govern! should be put into the place. In the former case, the acts of Congs. depended so much for their efficacy on the cooperation of the States, that these had a weight both within & without Congress, nearly in proportion to their extent and im portance. In the latter case, as the acts of the Genl. Govt, would take effect without the inter vention of the State legislatures, a vote from a small State wd. have the same efficacy & importance as a vote from a large one, and there was the same reason for different num bers of representatives from different States, as from Counties of different extents within particular States. Madison refers to State apportionment among coun ties to support his position because they likewise are not entitled to equal representation, regardless of population, in the legislature. The convention, as Madison himself recognized in the Federalist Papers, rejected his view that the new constitution would form a national government and end State sovereignty in favor of adopting a mixed national government. Appendix B to the government’s brief in the Maryland case [hereinafter Appendix B ], pp. 69, 71-72, 77-78. Consequently, Madison’s analogy between the new leg islature and the State legislatures does not apply to the constitution as it actually exists. The second statement of Madison relied on by the amici (p. 22) is that “ [t]he counties in Virginia are 1 This speech, which Strongly supports the government’s position, was unfortunately omitted from the Appendix to our brief in the Maryland case. 63 exceedingly disproportionate, and yet the smaller has an equal vote with the greater, and no inconvenience arises.” This quotation, which comes from the notes of Robert Yates, a convention delegate, is, as we pointed out in Appendix B, p. 24, note 10, almost certainly erroneous. Appendix B contains at least twelve statements by Madison (pp. 4, 15-16, 22-24, 27, 33, 38-39, 45, 48, 53, 60-61, 70-71, 117-118) made at the convention and elsewhere strongly supporting- representation both in the federal and State govern ments based directly on population and attacking other methods of apportionment as unjust. More over, Madison’s own account of his speech (Appendix B, pp. 22-24) contains no such statement. On the contrary, he says that “ Counties of the same States [are] represented in proportion to their numbers” and argues that, since the new government will be supreme, just as the States have been, the legislature must be apportioned on the basis of population as a matter of justice. Thus, again, Madison’s speech supports our position that the framers believed that a legislature in a government operating directly on the people must be apportioned on the basis o f popu lation and that therefore this was the only fair method of apportioning State legislatures. While, as the amici suggest (Br. 21), Madison may have been factually wrong in assuming that the State legisla tures of his time were all apportioned on the basis of population whereas several were not, his view of the only fair method of apportionment is plain. The amici then quote Gunning Bedford of Dela ware that “ [a]n exact proportion in the Representa tion is not preserved in any one of the States.” Ap pendix B, p. 36. Aside from the fact that perfect apportionment based on population is a virtual im possibility for which no one contends, Bedford’s basic 64 argument is that equal representation in the new leg islature is required because there is to be “ a mere con federacy of the States,” n o t11 a perfect consolidation.” Ibid. He refers to the State legislatures, like the mal apportionment of Great Britain, merely to show that all legislatures need not be apportioned by population. He does not praise the unequal apportionment of State legislatures, and implies that, if the new government were to be a consolidation ending the sovereignty of the States, he would not support equal representation for the States. The amici quote (p. 22) a speech by Elbridge Gerry stating that in the State legislatures one branch is often “ somewhat aristocratic” which he said would assure “ a better chance” of refinement in the choice of representatives. I Farrand 155.2 This statement was, however, made in support of election by the State legislatures, rather than the people. It has nothing to do with apportionment since apportionment can be made on the basis of population whatever method of election is used. This is made clear by the fact that Gerry stated at the convention that allowing each State in the Confederation an equal voice was an “ in justice.” Appendix B, pp. 29-30. He agreed to equal representation in the Senate only because a com promise was made to form less than “ a proper na tional plan.” Appendix B, p. 44. The statement by Hugh Williamson referred to by the amici is part o f a speech supporting apportion ment based on population. He (Appendix B, p. 7) : illustrated the cases by a comparison of the dif ferent States, to Counties of different sizes within the same States; observing that propor tional representation was admitted to be just in 2 This quotation was not included in Appendix B since, for the reasons stated above, it is irrelevant. 65 the latter case, and could not therefore be fairly contested in the former. This statement can hardly be used to support an ap portionment giving each county equal representation regardless of population. The reference by Randolph cited by the amici ( I Farrand 26-27) is to various State legislatures which have been feeble and inadequate because they were too democratic. But while he obviously did not want the people directly to control all organs of the government, he did not even suggest that apportion ment should be based other than on population. More over, Randolph was one of the most vigorous sup porters at the convention of apportionment based directly on population. See Appendix B, pp. 51-52. The amici cite (Br. 26) a statement by Hamilton that “ individuals forming political Societies modify their rights differently, with regard to suffrage. Ex amples of it are found in all the States.” Appendix B, p. 28. This was merely a statement of the fact that many persons were not entitled to vote. It was certainly not a defense of apportionment on some basis other than population since Hamilton went on in the speech to attack proposals to give each State the same representation. And earlier in the convention, Hamil ton had stated that equal representation for each State “ shocks too much the ideas of Justice, and every human feeling * * Appendix B, p. 14. The amici omit other references to the apportion ment of State legislatures. Madison argued for the adoption of what is now Article I, Section 4, giving Congress power to regulate federal elections because unfair apportionment of State legislatures might lead to similar results in Congressional districting. Ap pendix B, pp. 63-64. He repeated this contention at 66 the Virginia ratifying convention in attacking the in justice of South Carolina’s discriminatory apportion ment. Appendix B, p. 98. At the Massachusetts ratifying convention, Francis Dana and Rufus King defended Article I, Section 4, in part by pointing to the unfair apportionments of South Carolina, Rhode Island, Connecticut, and Great Britain. Appendix B, pp. 93-94. At the same convention, John Coffin Jones and James Bowdoin praised the State legislatures as generally apportioned by population. Appendix B, p. 94. We, and apparently the amici, have found but one statement at the convention supporting their position.3 Nathaniel Gorham of Massachusetts, in support of the compromise according each State equal repre sentation in the Senate, stated (Appendix B, pp. 17-18) : I f Va. should have 16 votes & Delre. with sev eral other States together 16. those from Virga. would be more likely to unite than the others, and would therefore have an undue influence. This remark was applicable not only to States, but to Counties or other districts of the same State. Accordingly the Constitution of Massts. had provided that the representatives of the larger districts should not be in an exact ratio to their numbers. And experience he thought had shewn the provision to be expedient. 3 Gouverneur Morris of Pennsylvania approved of efforts in his own State to prevent backwoodsmen from having their fair representation in the legislature because he thought that they might attain control. Appendix B, p. 53. This was part of Morris’ argument that the presentation of new States in the national legislature should be limited so that they never could attain a majority. However, Madison and others vigorously opposed the proposal (Appendix B, p. 53), and the conven tion rejected it. 67 Clearly, Glorham was justifying consideration of fac tors other than population in apportioning State leg islatures just as in the new legislature. He does not, however, say that State apportionments can, like the new Senate, give equal representation to each county. For, he referred to the Massachusetts Constitution of 1780 which provided that the Senate would be appor tioned among the districts on the basis of taxes paid (a method of apportionment considered at the time as analogous to population) and that the House of Rep resentatives would be apportioned by giving to each town having 450 voters one member and an additional member for every 225 additional voters, except that each existing town was given at least one. Part II, Ch. I, Sec. 2, Art. I, and Sec. 3, Art. II. Thus, the Massachusetts legislature was apportioned largely in proportion to taxes and population. One statement, made by a delegate having no par ticularly significant part in the convention, is hardly enough to show that the framers believed that the apportionment of State Senates was analogous to the apportionment of the Senate of the United States. This would be so even if the convention contained no other statements on this subject. But, as shown above and particularly in our brief in the Maryland case (pp. 75-80), the whole history of the Constitu tion demonstrates the contrary conclusion. In short, the references to the States legislature apportionments at the Philadelphia convention and the subsequent ratifying convention do not support the proposition that State Senators are analogous to th at o f the United States. The references cited above show that framers overwhelmingly believed that the only fair method of apportioning State legisla tures is according to population. Madison and W il liamson argued that the new legislature should like- 6 8 wise be apportioned by population because they de sired a truly national government ending State sov ereignty and operating directly on the people. But this was not the result of the convention. Instead, the new government was to be a mixture o f a national government and a confederation, with the House of Representatives reflecting the former and the Senate the latter. Since the States are not such a mixture, their legislatures, as the framers plainly believed, must be apportioned on the basis of population. 2. The amici Attorneys General also claim (Br. 27-29) that the federal anology is valid because many of the framers o f the constitutional convention pro posed limiting the democratic features of the new government. This we do not dispute. Our point is rather that those supporting a national government operating directly on the people strongly supported apportionment of the entire legislature, whether it was to be one or two houses, on the basis o f popula tion. This was true even when they opposed election o f the legislature by the people in favor o f election by the State legislatures. The framers cited by the amici made clear that ap portionment of the legislature or a non-population basis was not intended to be one of the permissible checks on control by the people. The amici cite (Br. 27-28, 32, 34, 37) Randolph, Gerry, and Madi son. Not only do none o f these statements even sug gest that apportionment on any basis but population is fair, but, as we have seen above (pp. 61-65), all three vigorously supported apportionment based en tirely on population. The amici quote (Br. 28-29) a series of statements made by proponents of the plan finally adopted of equal representation of States hi the Senate. The quotations state that equal representation was needed 69 to prevent the populous States from dominating and destroying the less populous. It is significant, how ever, that these men believed in representation by States rather than people entirely because they con sidered the new government as either a mere continua tion of the old Confederation although with somewhat enhanced powers or as a mixture of a Confederation and a national government. In the same speech quoted by the amici in which William Paterson of Hew Jersey said he feared that his State “ would be swallowed up” (I Parrand 179), he plainly stated (Appendix B, p. 6) : A confederacy supposes sovereignty in the members composing it & sovereignty supposes equality. I f we are to be considered as a nation, all State distinctions must be abolished, the whole must be thrown into hotchpot, and when an equal division is made, then there may be fairly an equality of representation.4 Later, he said that apportionment based on popu lation “ is right in principle” if “ state distinctions are done away; but those to certain purposes still exist.” Appendix B, p. 12. On several other occasions, he made plain that equal representation by States was necessary to protect State sovereignty. Appendix B, pp. 6, 11, 12. Thus, it is plain that Paterson believed that the legislature of a government operating directly as the people, such as a State, must be apportioned according to population. David Brearly of Hew Jersey agreed with Paterson that equal representation of States was required if the Confederation was to be continued, as he desired, 4 Paterson is referring to a proposal of David Brearly of Hew Jersey that, if the Confederation is to be ended, the States should be abolished, new States should be formed with equal population, and representation in the legislature should be the same for each State (Appendix B, p. 7). 70 and that representation by population was permissible only if a national government were formed (Appen dix B, p. 7) : I f the states still remain sovereign, the form of the present resolve is founded on principles of injustice. He then stated the comparative weight of each state—the number of votes 90. Georgia would be 1, Virginia 16, and so of the rest. This vote must defeat itself, or end in despotism. I f we must have a national gov ernment, what is the remedy? Lay the map of the confederation on the table, and extinguish the present boundary lines of the respective state jurisdictions, and make a new division so that each state is equal— then a government on the present system will be just. The next delegate cited by the amici (Br. 29), Roger Sherman of Connecticut, first proposed the compro mise that the Senate be equally apportioned among the States and the House of Representatives be ap portioned by population. He explained the former as an effort to protect State sovereignty (Appendix B, p. 9 ) : * * * as the people ought to have the election of one of the branches of the legislature, the legislature of each state ought to have the election of the second branch, in order to pre serve the state sovereignty; and that each state ought in this branch to have one vote. The speech in which Alexander Martin of North Carolina states that he feared domination by the pop ulous States begins by suggesting that the new gov ernment must be a continuation of the Confederation (Appendix B, p. 16) : [The States] entered into the confederation on the footing of equality; that they met now to to [sic] amend it on the same footing, and that he could never accede to a plan that would 71 introduce an inequality and lay 10 States at the mercy of Va. Massts. and Penna. Gunning Bedford, as we have seen (pp. 63-64), based his argument for equal representation of States on his view that the “ confederacy of the States,” in which each State is sovereign, must be continued. Oliver Ellsworth of Connecticut supported the Con necticut compromise. He based this position explicitly on the ground that the new government must be a mixture (Appendix B, pp. 30-31) : We were partly national; partly federal. The proportional representation in the first branch was conformable to the national principle & would secure the large States agst. the small. An equality of voices was conformable to the federal principle and was necessary to secure the Small States agst. the large. * * * The existing confederation was founded on the equality of the States in the article of suffrage: was it meant to pay no regard to this antece dent plighted faith. Ellsworth later repeated essentially the same remarks. Appendix B, pp. 32-33. Numerous other framers made similar statements at the Philadelphia convention and the State ratify ing conventions. See the government’s brief in the Maryland case, pp. 75-79. The amici can cite only one statement in response— that by Nathaniel Gorham to which we referred earlier (pp. 66-67). This state ment, in the face of overwhelming contrary evidence, cannot be taken as the sense of the convention. In short, the fact that many of the framers wanted to limit the democracy contained in the new constitution is irrelevant. The fact is that the basis of apportion ment was no part of this effort. Those who wanted to give the States equal representation in the whole legislature or one house did so because they believed 72 the new government must continue the sovereignty o f the States. 3. The amici Attorneys General contend (Br. 30) that the States, at the time of the convention were not truly sovereign. They cite only one statement at the convention in support of this argument. Rufus King' of Massachusetts said that “ [t]he States were not ‘ sovereigns’ in the sense contended for by some” and specified various limitations on their sovereignty. I Farrand 323. However, he went on to admit that the States were in some respect sovereign (id. at 324) : I f the States therefore retained some portion of their sovereignty, they had certainly divested themselves of essential portions of it. I f they formed a confederacy in some respects—they formed a Ration in others. Moreover, King was a vigorous proponent o f repre sentation based on population for the whole legisla ture. See Appendix B, p. 35. Indeed, he was willing to continue the fight even after Madison agreed to the compromise. Appendix B, pp. 58-59. King therefore desired to minimize State sovereignty. But his view was rejected as to one house of the legislature by those who believed that substantial State sover eignty existed and should be continued. U.S. GOVERN VENT PRINTING 0FFICE:!963