(dissenting).
The majority opinions both find that plaintiffs have failed in their proof, i. e., they, have not proved a prima fade case of unconstitutional deprivation of their rights.
*472I disagree and find that plaintiffs have borne their prima, facie burden (Hernandez v. Texas, 347 U.S. 475, 74 S.Ct. 475, 98 L.Ed. 866) and because of the absence of any proof by defendants or in-tervenors they are entitled to judgment declaring the challenged portion of Chapter 980 unconstitutional in violation of the equal protection clause of the fourteenth amendment. Let me premise my reasons with a few concessions.
I concede that there was a total absence of direct proof of any specific intent by the New York Legislature in drawing the lines of any district; I concede that disparity alone in the population of one district compared to another or to a general state or city average is not dispositive; I concede that of itself a district’s lines whether jigsaw, straight, serpentine or otherwise would not be controlling; I concede that some disproportion of numbers of ethnic groups in adjoining districts would not be enough; I concede that the federal courts should ordinarily refrain from entering into “political thickets” and that, it is beyond our competence to suggest- or supervise a remedy for unlawful apportionment. But see Inequities in Dis-tricting for Congress: Baker v. Carr and Colegrove v. Green, 72 Yale L.J. 13 (1962.)
The uncontradicted proof submitted' by plaintiffs, however, establishes a visual figure picture of the end results of the recent redistricting of Manhattan Isle (New York County) as follows:
Manhattan has a population of 1,698,-281 people and is entitled to four congressmen. The census figures of I960' divided the ethnic groups into only two-classes — white and non-white and Puerto Rican. These classes have been counted, and according to the census 1,058,589 or 62.3% are white and 639,622 or 37.7%. are non-white and Puerto Rican.
The district lines as fixed by Chapter 980 created the four districts in question^ with the following make-up:
District Total Population White Population % of District Non-White and Puerto Rican Origin Population of District;
17th 382,320 362,668 94.9%' 19,652 5.1%-
18th 431,330 59,216 13.7% 372,114 86.3%-
19th 445,175 318,223 71.5%' 126,952 28.5%-
20th 439,456 318,482 72.5% 120,974 27.5%.
Total 1,698,281 1,058,589 62.3% 639,692 37.7%;
The following table shows the percent of non-white persons and persons of Puerto Rican origin in each congressional district in relation to the total number of such persons in the entire county:
% of Non-White and Puerto Rican of County
District
17th 3.1%'
18th 58.2%-
19th 19.8%
20th 18.9%'
100.0%
The figure picture of the 17th District shows that the lines as drawn encompass, a population 94.9% white and 5.1,% nonwhite and Puerto Rican. It further shows it has a population of 382,320 people, or between 15.4% and 12% less than-any of the adjoining districts. The 18thi District encompasses a population that, is 86.3% non-white and Puerto Rican- and only 13.7% white. Its population of 431,330 people is 12% more than the-17th and 5% above the state average.
It is my judgment that the only available inference from the above uncontra-dicted figure picture establishes per se a *473prima facie case of a legislative intent to draw congressional district lines in the 17th and 18th Districts on the basis •of race and national origin. To me it fits foursquare with Mr. Justice Frankfurter’s statement in Gomillion v. Light-foot, 364 U.S. 339, 341, 81 S.Ct. 125, 5 L.Ed.2d 110, that the act in question was not an ordinary geographical redistricting measure even within the familiar -abuses of gerrymandering. Although -Justice Frankfurter’s statement referred to the court’s holding that there was -a violation of the fifteenth amendment this statement is equally apposite to the •equal protection clause of the fourteenth amendment under Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873. Cf. the concurring opinion of Mr. Justice Whittaker in Gomillion at 349, 81 S.Ct. at 131, 5 L.Ed.2d 110. The conclusion here is, as in Gomillion, irresistible, tantamount for all practical purposes, to a mathematical demonstration that the legislation was solely concerned with segregating white, and •colored and Puerto Rican voters by fencing colored and Puerto Rican citizens out •of the 17th District and into a district of their own (the 18th).
We assume that had the district lines •of the 17th District been drawn so as to exclude all non-white and Puerto Ricans, or the 18th to exclude all white, my brothers would agree that plaintiffs had established a prima facie case of per se segregation. Gomillion v. Lightfoot, supra. It is acknowledged, however, that plaintiffs’ uncontradicted evidence demonstrates that New York County, an island having 639,692 non-white and Puerto Ricans or 37.7 % of the total population, was redistricted into four congressional districts with one district, the 17th, having only 5.1% non-whites and Puerto Ricans and the 18th with only 13.7% white.
The question then posed is—Does the fact that the congressional district lines decreed by the state legislature for the 17th District to encompass only 5.1% non-white and Puerto Rican and the 18th only 13.7% white as distinguished from 0% so dilute plaintiffs’ proof as to require them to prove more? If so, did they do it when the uncontradicted proof also showed that the 17th District had 15.4% less people than the adjoining 19th District; 14% less than the 20th and 12% less than the 18th. My brothers say “No” and I disagree.
It might very well be that the defendants and intervenors could have offered proof to counteract the inference of racial segregation that plaintiffs proof implies but they did not—and furthermore they chose not to do so. They might have proved all of the factors enumerat-éd by Mr. Justice Frankfurter in Baker v. Carr, 369 U.S. 186, 323, 82 S.Ct. 691, 767, 7 L.Ed.2d 663, that go into the complicated political potpourri of apportionment. They might have proved that the lines were drawn as part of a political compromise between the major political parties to insulate certain sections for “traditional purposes”—but the simple answer is that they did not.
What more need plaintiffs prove? Surely it cannot be argued that they must prove some oral or written statement made by the legislature either in the form of a committee report or from the manager of the bill, or statements from the legislators themselves. It is undisputed that no public hearings were had on the bill and that the only report filed was the interim report of the Joint Legislative Committee on Reapportionment referred to by Judge MOORE. The bill recommended was submitted to the legislature on November 9, 1961, and passed on November 10, 1961, and was signed by the Governor that day. N.Y.Sess.Laws, 2d Extraordinary Sess. 1961, c. 980, §§ 110-112.
Judge FEINBERG and I part company only on the quantum of plaintiffs’ proof. He agrees that the plaintiffs are not required to prove any diminution or dilution of their voting rights. They prove their prima fade case once they show that the district lines were constituted on racial basis but he agrees with Judge MOORE that the plaintiffs have not proved enough—but neither opinion *474tells us how much more or enough of what.
Judge FEINBERG states that the principal area of the inquiry must be the changes brought about by the 1961 redistricting. With this as his premise he points out that the 17th District has approximately only 7% less population than the average for the state and such disproportion does not justify a finding of racial discrimination. I agree.
All I say is, it is a factor or a fact to be considered with all of the others, keeping in mind that the legislature was dividing an island into four districts and such island contained 37.7% nonwhite and Puerto Ricans.
He also suggests that the word picture of figures would infer not discrimination along racial lines but rather that nonwhite and Puerto Ricans live in certain concentrated areas so that district lines encompassing these areas would necessarily include a very high percentage of non-white and Puerto Ricans. This is exactly my point and also the plaintiffs. The pattern of the 18th District lines shows that they were drawn so that any district lines encompassing these areas would necessarily include a very high percentage of non-whites and Puerto Ricans. And, we might add, a very high percentage of whites in the 17th.
In answer to my question — What more need plaintiffs prove ? He says some answers might be — not should be, but might be: (a) Failure to build on prior lines in a rational, logical manner. This presumes that the prior lines were without any constitutional infirmity. In any event, how does one build four districts on foundations of six districts? (b) A greater population disparity. It is suggested that if the plaintiffs had shown a failure to increase the population in the 17th District enough to keep it without a fair approximation of the state average a stronger inference might be drawn that the population was deliberately kept small because adding to it could only increase the non-white and Puerto Rican percentage. The 17th District is 7% below the state average. Would 8% be enough, or 9%, or 10%, etc.? What is a fair approximation? Isn’t it really a question of fact? How do you weigh such questions when a defendant offers no proof? I submit that the scale tips toward the plaintiffs. The City of New York with 7,781,984 people has been divided by the legislature into 19 districts with an average population per district of 409,578. It is true that the New York City average population almost equals the average population per district throughout the state. But why must we make comparisons with the entire 19 districts in the City of New York or the entire 41 districts in the state? We are dealing with Manhattan Island which for all practical purposes is a unique metropolitan area with many well-known river to river cross streets and famous north and south or longitudinal streets. See, for example, the plaintiffs other proof in which they demonstrated by three hypothetical divisions how the island could have been divided into four districts on a logical and rational basis using the natural boundaries or well-known streets and avenues. I agree that such hypothetical districts are not conclusive but they do have some probative value and I think are helpful in pointing up the obvious segregation that the legislature effected, (c) An increase in boundary zigzagging. How much of an increase and how is the number of zigzags measured or counted, and do you compare the zigzagging lines with the lines drawn by the legislature in 1951 or 1941, and do you confine yourself to Manhattan Island or New York City or any district in any part of the state.
I agree that no plaintiff, or for that matter any person on Manhattan Island, has lost or been deprived of a right to vote for Congress or that his vote will not be counted but the parallel to Gomil-lion (concurring opinion.) is clear. There it was a glaring exclusion of Negroes from a municipal district. Here it is a subtle exclusion from a “silk stocking district” (as the 17th is so fre*475quently referred to) and a jamming in of colored and Puerto Ricans into the 18th or the kind of segregation that appeals to the intervenors.
We are told that the fifteenth amendment nullifies sophisticated as well as simple-minded discrimination. In my judgment the New York legislature has attempted, in violation of the equal protection clause of the fourteenth amendment, a sophisticated and subtle discrimination. Accordingly, I would give judgment for plaintiffs that the challenged part of the act is unconstitutional.