with whom INGRAHAM, Circuit Judge, joins (dissenting in part):
On November 17, 1972, a majority of the Judges ordered this case reheard en banc, directing counsel to file through briefs on the issue of “dilution of the franchise of black voters” as raised in the District Court and considered by the panel decision.1 **Clearly, what the en banc court intended to do was to establish some standards as to what is required to “minimize or cancel out black voting strength”,2 particularly in the context of a sparsely settled governmental unit inhabited by less than 13,000 people.
With sincere deference, I fear that the intended objective has eluded the en banc opinion. Regretfully, it leaves me with no plainly discernible idea of what the District Courts are hereafter to do with this troublesome problem. Moreover, I must disagree with the effect of those portions of the opinion which I think I do understand.
The opinion appears to be anchored, in the main, on the recent decision of the Supreme Court in White v. Regester, dealing with conditions prevailing as to the multi-member legislative districts in Dallas and San Antonio. East Carroll Parish, Louisiana, however, is a “far cry” from either of these highly populated metropolitan areas. The Parish Seat, Lake Providence, has a population of 6,183. The remaining 450 square miles of typical rural delta terrain have 6,701 inhabitants, with an average population density of fifteen persons (not voters) per square mile. In East Carroll blacks are in a heavily preponderant majority, not in the minority as were the blacks and Mexican-Americans in Dallas and San Antonio. In White, the Supreme Court considering the totality of the circumstances, was not inclined to disturb the findings of the Three Judge District Court that certain conditions in the cities had led to invidious discrimination against the minority groups. In the instant appeal, the en banc majority holds a non-cancellation, non-minimization finding of the District Court to be clearly erroneous.
Let us now turn to the undisputed facts in the case presently before us.
In round figures, the Parish population is 59% black and 41% white;- th,e voting registration is 54% white and 46% black. The record fails to reflect the number of eligible blacks who have chosen not to register.
With a spread of only eight percentage points between white and black registration, even in an election conducted solely on racial considerations a switch of less than five percent would reverse the expected outcome. Does such a thin line, even in a formerly segregated jurisdiction, amount as a matter of law to impermissible “minimization” or “cancellation” of black voting strength? I would respond in the negative, but the en bane opinion says “yes”; that any other view is clearly erroneous. Of course, divergencies of such size in constitutionally mandated population reapportionments at state and local levels do not call for per se invalidation, White v. Regester, supra.
One of the factors stressed in White v. Regester was that for nearly a hundred years only two minority representatives from Dallas County and only five Mexican-Americans from Bexar County had been elected to the Texas Legislature. I do not know how many had been candidates nor what qualifications they possessed, but in East Carroll the en banc opinion declines to consider the unchallenged fact that in the at-large elections in East Carroll Parish in 1971 and 1972 three black candidates were elected *1310to office, including one who previously had been defeated when running in his own ward. The opinion holds that an appellate court cannot take cognizance of matters not passed upon by the trial court. It was all right for the Supreme Court to consider election results in Dallas and Bexar for the past one hundred years but when it comes to dismembering a little parish we must blind ourselves to facts asserted on oral argument and not in the least disputed. One rule for plaintiffs and the opposite for defendants. I might add that in the eight years I have served on this Court I have frequently seen it go outside the formal record in racially oriented cases for the reason that it would not blind itself to unchallenged facts.
Obviously, the election of three black candidates in at-large elections within two years after the adoption of the plan now under review, and at a time when black registration amounted to 46% and white registration came to 54%, pretty well explodes any notion that black voting strength has been cancelled or minimized. What the en banc opinion condemns as clearly erroneous turns out to be clearly right when put to the test of actual use at the ballot box rather than in the rarified atmosphere of the judicial chamber.
Additionally, in White v. Regester, the at-large candidates for the legislature were not required to live in a particular zone, so that it was possible for every legislator to be elected from outside the minority areas. Indeed, the Three Judge Texas District Court felt impelled to say, and the Supreme Court quoted it, that these legislators were “insufficiently responsive to Mexiean-American interests”.
In East Carroll Parish, however, the candidates were required to run from specifically designated areas.
I believe that White v. Regester was decided on the denial of access to the political process, not on cancellation or minimization of the minority vote. In any event, a decision rendered on the facts peculiar to that case and not shown to exist in East Carroll provides faint precedential support for the en banc decision now about to be rendered.
In East Carroll Parish the black people hold a predominant population majority — 59% to 41%. It necessarily follows that the opinion of the Court in this case can only mean that in formerly “segregated” areas a black population majority is of no consequence in one man-one vote reápportionment cases unless it is also matched by a black registration majority when there is an effort to escape from the pocket borough method of electing county officials. The basic requirement, the fundamental goal, of population equality must yield to registration equality.
I submit that this is a serious distortion of the one man-one vote principle. It has to be premised on the idea that voters will always vote for candidates of their own race, an idea which has repeatedly failed the test of electoral experience, even in East Carroll Parish itself. It also overlooks the absolute certainty that with such a predominantly heavy population majority it is only a matter of time, and not too long at that, when there will be a black registration majority in East Carroll.
The en banc opinion is quite inconclusive as to what the District Court must do on remand. Apparently, however, no choice is left but to direct the division of the Parish into wards (districts). The Balkan must be fragmented into nine Balkans. After it had lost jurisdiction in this case, as set forth in the panel opinion, the District Court approved a redistrieting plan submitted by the plaintiff-appellants by which three of the nine board members would be elected from one ward while the other six wards would be allowed to elect only one each. One ward would elect % of the Board while the other six would elect % each. So far as ultimate power on the Board *1311is concerned a vote in one ward would be worth exactly three times that of a vo.te in any other ward. To me that certainly smells of invidious discrimination and a complete distortion of the one man-one vote rule within a small county area.
The paramount consideration in this case is that in East Carroll Parish, Louisiana, the black race is a definite majority, not a minority.3 Thus, the ultimate question is whether a 7% deficit in black registration cancels or minimizes the black voting strength of those who hold an 18% advantage in population. The en banc opinion says that it does, as a matter of law, and that the affirmative finding of the District Court to the contrary is clearly erroneous.
Hence, we direct a change in the election machinery of a small rural parish, where the black officials elected under the old malapportioned ward plan expressed, as the record shows, a preference for the at-large plan.
I would adhere to what the panel said originally:
“We are unable to see how at-large elections in this small parish could possibly discriminate against its black citizens. They have a commanding majority of the total population. This majority would participate in the election of not one but of every police juror and every school board member in the parish. Every elected official would thus be answerable to all black citizens of the entire parish. On a single-member district plan, if such can be devised, the voters would have a voice in the election of only one member on the police jury or the school board. It would, therefore, be just as easy to say, in the exercise of discretion, that a single district plan in this parish would more clearly dilute the voting power of the blacks than would one in which every voter, black and white, has the same identical voice in the selection of not one, but all members of the elective body. As a matter of fact, the District Judge so found.”
To the extent indicated, I respectfully dissent.
. The panel opinion in this ease (Judge Gewin dissenting in part) is reported, 467 F.2d 1381.
. Burns v. Richardson, 384 U.S. 73, 86 S.Ct. 1286, 16 L.Ed.2d 376; Fortson v. Dorsey, 379 U.S. 433, 85 S.Ct. 498, 13 L.Ed.2d 401.
. The en bane opinion speaks of single-shot voting, run-off elections, and slating candidates, but these were not issues in the court below.