Toombs v. Fortson

ORDER ON MOTION FOR PRE-TRIAL CONFERENCE

PER CURIAM.

The motion filed February 6, 1967, by the defendant Honorable Ben W. Fortson, Jr., for a “pre-trial conference” in this state of these proceedings is hereby denied.

This motion appears to be directed to a desire on behalf of defendant Fortson to provide an opportunity for the State Senate, as one of two chambers of the Georgia General Assembly, once more to justify its present method of apportionment.

In our order of April 1, 1965, we held, “the present plan of Senate apportionment in Georgia and the proposed plan of apportionment for the House of Representatives, separately and collectively, fall short of the mandate of the equal protection clause of the Fourteenth Amendment, as construed by the Supreme Court in Reynolds v. Sims, and related cases.” This order was not appealed by the defendants, and is final and binding both upon this court and upon all parties.

The most recent case by the Supreme Court dealing with the Florida legislature, Swann v. Adams, 385 U.S. 440, 87 S.Ct. 569, 17 L.Ed.2d 501, #136, October Term, 1966, again makes plain the standard that was first announced in Roman v. Sincock, 377 U.S. 695, 84 S.Ct. 1449, 12 L.Ed.2d 620, in which the Court stated:

“The proper judicial approach is to ascertain whether, under the particular circumstances existing in the individual state whose legislative apportionment is at issue, there has been a faithful adherence to a plan of population based representation, with such minor deviations only as may occur in recognizing certain factors that are free from any taint of arbitrariness or discrimination.” (Emphasis added.)

This court cannot give an advisory opinion in advance of the action of the General Assembly in carrying out the previous mandate of the court. However, we think it appropriate to state again that emphasis should be placed upon attaining as nearly mathematically equal representation as is practicable rather than upon seeking to justify deviations within certain arbitrary percentage tolerances.

GRIFFIN B. BELL, Circuit Judge:

OPINION AND ORDER

This court must again consider the question of the apportionment of the General Assembly of Georgia. This time we deal with final reapportionment. The details and past history of the problem will be found in our previous opinions: Toombs v. Fortson, N.D.Ga., 1962, 205 F.Supp. 248; Toombs v. Fortson, N.D. Ga., 1965, 241 F.Supp. 65; Toombs v. Fortson, N.D.Ga., 1966, 275 F.Supp. 128 [opinion dated March 25, 1966], affirmed, Toombs v. Fortson, 1966, 384 U.S. 210, 86 S.Ct. 1464, 16 L.Ed.2d 482.

Our opinion and order of April 1,1965, as amended by the opinion and order of March 25, 1966, required that the Senate and House be apportioned in accordance with federal constitutional standards not later than May 1, 1967 with the apportionment to be effective for the 1968 primaries and general election for members of the Senate and House. In those decisions we approved interim reapportionment which had been accomplished in the Senate in 1962, Ga.Laws, 1962, extra sess., pp. 7, 14; and in the House in 1965, Ga.Laws, 1965, p. 127.

Faithful to our orders, the General Assembly modified both Senate and House apportionment at the regular 1967 legislative session, Ga.Laws, 1967, pp. 159, 187, in an effort to comply with federal constitutional standards. Secretary of State Fortson, acting through the Attorney General of Georgia, promptly moved for approval of the reapportionment plans provided in the legislation. Plaintiffs objected, contending that the legislation fails to comply with the court’s orders and fails to meet constitutional standards. Subsequently, we allowed two interventions, one on behalf of citizens of Douglas County and the other on behalf of citizens of Burke County. After a full hearing in which evidence was introduced in justification of the apportionment plans and after briefs and oral argument, the issue was submitted to the court.¹

By way of preliminary summary we approve the proposed reapportionment in substantial degree. Some changes, however, will be necessary in both the Senate and House plans and these can be made at the 1968 regular session of the General Assembly. The changes in the senate plan will involve shifting four counties, Bryan, Effingham, Jones, and Rockdale. In the House, three additional seats must be found for DeKalb County and one seat for Douglas County which will be constituted as a separate district. Some shifting and recombination of counties will be necessary to provide these four seats and also to reduce population variances in other districts.

The suggested changes will provide apportionment plans whereunder the citizens of Georgia will be afforded substantial equality in legislative representation based on the 1960 census as adjusted to unusual population growth situations. We believe that the apportionment plans, as changed, will meet the equal protection of the law standard of the Constitution. They will then be approved for use pending receipt of the 1970 census which means, as a practical matter, until the 1972 primaries and general election. As we have made clear previously, the General Assembly must be reapportioned after each decennial census.

The disposition of the issues presented is somewhat complicated by the difference in the approaches to reapportionment taken by the Senate and House. The Senate based its reapportionment on 1960 population data but gave some weight to population trends and current population estimates. The House relied entirely on the 1960 census. Both, no doubt, had in mind our previous statement that a variance of more than 15 per cent,would hardly be justifiable. We were searching for a mathematical formula to be used in testing apportionment plans. We were careful then to say that a deliberate built-in variance of 15 per cent would not be proper. We stated the rule to be followed as “ * * * a good faith effort to meet the average [or norm] with departure only where necessary to afford individual representation to as many counties as possible.” Toombs v. Fortson, 241 F.Supp. at 70. We find that such a good faith effort has been made and that there is no deliberate built-in variance in either plan.

No mathematical formula has been set by the Supreme Court in its supervening decisions by way of an allowable tolerance or variance from the one man-one vote doctrine. Rather the requirement continues to be that an apportionment plan must meet the norm as near as practicable and any substantial departure from the norm must be justified. Reynolds v. Sims, 1964, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506; Swann v. Adams, 1967, 385 U.S. 440, 87 S.Ct. 569, 17 L.Ed.2d 501; and Kilgarlin v. Hill, 1967, 386 U.S. 120, 87 S.Ct. 820, 17 L.Ed. 2d 771. We find no departure from the teaching of Roman v. Sincock, 1964, 377 U.S. 695, 84 S.Ct. 1449, 12 L.Ed.2d 620, where the court said:

“* * * the proper judicial approach is to ascertain whether, under the particular circumstances existing in the individual State whose legislative apportionment is at issue, there has been a faithful adherence to a plan of population-based representation, with such minor deviations only as may occur in recognizing certain factors that are free from any taint of arbitrariness or discrimination.” 377 U.S. at p. 710, 84 S.Ct. at p. 1458.

It is clear to us from Swann v. Adams and Kilgarlin v. Hill, supra, that a variance of more than 10 per cent is not to be condoned unless there is strong justification for the particular variance or variances. Indeed, we do not think that a 10 per cent tolerance is acceptable as a matter of course. It must appear, as here, that a good faith effort was made to meet the norm in the first instance and that the result was variances in a small number of districts to the outer limit of 10 per cent because of the policy of affording representation to sub-governmental units.

Having concluded that there was a good faith effort at population-based representation and that there was no deliberate built-in or systematic variance which would indicate arbitrariness or discrimination, we think it proper to approach the plans by using a 10 per cent tolerance test as a maximum acceptable without special justification. In using this test we will also use the population norm based on the 1960 census. It is settled that apportionment based on the decennial census meets the constitutional minimum. Reynolds v. Sims, supra. On the other hand, apparent deviations in representation may be justified on the basis of population growth between the last census and the time of formulating the apportionment plan. Cf. Kilgarlin v. Hill, supra.

Plaintiffs urge that adjustments should be made in the plans of both the Senate and House to uniformly reflect present populations. We conclude that the plans should allow reasonable additional representation to those areas where the present population estimates indicate extraordinarily disproportionate growth. We think this position is supported by considerations of fairness and equity, especially in view of our use of population growth to justify deviations. Population trends and current population estimates do justify the variances in some instances which might not otherwise be justifiable if we were restricted to 1960 data. Our adjustments to reflect these unusual growth situations are also supported by the fact that the plans in question were enacted in an effort to reapportion a General Assembly which was malapportioned to an extreme degree. Beginning in 1962 we have allowed a gradual approach as distinguished from complete and final reapportionment in an effort to maintain legislative stability. Hence it is proper in now considering final reapportionment to take account of population changes.

We have considered all population data available. The parties concede that the Georgia Health Department population estimates are the most accurate available and these reflect populations as of July 1, 1966. They have been adjusted to include December 31, 1966 military population. The Health Department estimates as adjusted, have been modified in only three instances; two in the case of transient military population in Chattahoochee and Richmond Counties, and the other in adjusting the large difference between the July 1, 1965 Bureau of Census estimate which is made of metropolitan populations, and the July 1, 1966 Health Department estimate for DeKalb County. The disparity is so great as to cause us to take a figure in between.

We find state policy in connection with legislative apportionment to be directed toward three goals: One, to meet the federal constitutional standards; two, to maintain stability in the legislative process; and three, to afford efficient and representative government by having compact and contiguous districts with separate representation for as many counties as possible. An added goal has been to reduce the number of seats in the House and the present plan reduces the number from 205 to 195. A reapportionment plan which results in as few changes as possible will enhance the legislative process from the standpoint of stability but, of course, any plan must meet federal constitutional standards. Our interim approval of reapportionment legislation over the years since 1962 has been in the interest of legislative stability and to allow time for trial and error in the realignment and regrouping of counties where multi-county districts were necessary. Reynolds v. Sims, supra, recognized independent representation for political subdivision as being a desirable governmental goal which would warrant some deviation from population-based representation.

We find that the plans before us were formulated in line with these state policies. Members of the Senate and House testified in detail as to the basis of the plans and their testimony shows in general that the state policy was followed and that a balance was sought to preserve a system of representative government within the context of the equal protection clause. The court recognizes that the one man-one vote doctrine can be given perfect application only by state at large elections. The larger the district, the nearer the standard may be approached but the Supreme Court has made it clear that the rule is “as near as practicable” in light of a representative form of government. Reynolds v. Sims, supra. The individual voter can more nearly translate his vote into action through his own , representative or at least through a small number of representatives. Compare the case of Dusch v. Davis, 1967, 387 U.S. 112, 87 S.Ct. 1554, 18 L.Ed.2d 656, where the Supreme Court described the particular allocation of councilmen as a rural-urban detente affording representative government without invidious discrimination.

With these general findings and conclusions in mind, we turn to a separate consideration of the Senate and House plans.

THE SENATE

The 1962 Senate apportionment was based on 54 senatorial districts of one senator each. The 1967 apportionment plan adds two senators to make a total of 56. Using 1960 population data (norm of 70,413), and adding the two additional senators, one to Fulton County and one to DeKalb County, we find the following variances:

CHART I

Percentage Departure from Norm Number of Districts Total Population of Districts

15-20% below 2 114,174

10-15% below 2 124,844

5-10% below 8 516,994

0-5% below 12 832,800

0-5% above 21 1,513,714

5-10% above 7 524,957

10-15% above 4 315,637

Using the same 1960 basis we find variances from a 10 per cent tolerance in the following districts to the extent shown:

but 1966 population estimates (norm of 79,461), we find that a total of 14 districts exceed the tolerance to the extent shown:

CHART II

District Per Cent of Variance

1 —10.99

2 -11.71

3 — 9.88^*

4 +11.89

8 +11.10

15 +12.60

16 +12.68

32 —19.66

33 -18.19

CHART III

Per Cent of District Variance

Chatham County (1, 2, & 3 averaged) —22.94

11 -10.90

14 +27.23

21 —15.05

Richmond County (22 & 23 averaged) +15.16

DeKalb County (41, 42, 43, & 55 averaged) +13.45

44 +42.11

48 +13.77

The maximum variance excluding these nine districts runs from +8.68 to -8.83. These charts demonstrate that the residents of 4 out of 56 districts are overrepresented by 10 to 20 per cent. Those residing in 4 districts are underrepresented by 10 to 15 per cent. The deviation from the average, based on the 1960 census, ranges from -19.66 to +12.68. Approximately 14 per cent of the total state population resides in these under and overrepresented districts. Approximately 26 per cent of the population is in districts under or overrepresented between 5 and 10 per cent, leaving 60 per cent in those districts which deviate from the average by less than 5 per cent. The variances which exceed 10 per cent are all justified on the basis of current population estimates or may be corrected, in the case of the three districts in Chat-ham County, by the inclusion of Bryan and Effingham Counties.

Giving consideration to growth and using the same 10 per cent tolerance test

The maximum variance excluding these 14 districts, using 1966 population estimates, is from +8.24 to -9.13. We proceed to a discussion of each of these districts wherein the variance exceeds 10 per cent whether on 1960 or 1966 populations.

In this connection we must preliminarily consider the growth areas of the state. The 1960 state population of 3,943,116 had increased to 4,449,793 by July 1, 1966 according to Health Department estimates. This increase includes a military population of approximately 100,000 in Georgia as of December 31, 1966. Thus the state population growth, give or take part of the military population which may have been transient, has been about 12 per cent. All counties have shown some growth except Chat-ham, Colquitt, Decatur, Lee, Lincoln, Lowndes, and Taliaferro. A few counties have had growth approaching the state average; Coweta, Fulton, Hall, Spalding, Thomas and Twiggs. Several have had growth in excess of the average and some in spectacular degree as the following chart depicts:

CHART IV

County Pop. 1960 Pop. 1966 Rate of Gain

Clayton 46,365 72,024 55.3%

Columbia 13,423 20.300 51.3%

Rockdale 10,672 15.700 48.5%

Cobb 114,174 163,632 43.3%

Henry 17,619 25.200 43.1%

DeKalb 256,782 360,600 40.4%

Douglas 16,741 22.700 35.6%

Houston 39,154 52,489 34.2%

Jones 8,468 11.300 33.4%

Gwinnett 43,541 56,000 28.0%

Glynn 41,954 53,318 27.0%'

Effingham 10,144 13.700 25.5%

Paulding 13,101 16.200 24.4%

Whitfield 42,109 52,200 23.8%

Bartow 28,267 33,100 17.0%'

Baldwin 34,064 39.700 16.5%

Walker 45,264 51,900 14.6%

Dougherty 75,680 86,008 13.6%

Chattahoochee 13,011 40,201^*

Richmond 135,601 183,013^*

Chatham County, as noted, has not kept pace with the overall growth of the state. The result has been an increase in the degree of overrepresentation which already existed as to that county in the Senate. We disápprove the apportionment plan as it relates to the three districts allocated to that county. The condition can be corrected by shifting Effingham County from District No. 4 and Bryan County from District No. 5 to be included in Chatham County’s senatorial representation.

The underrepresentation of District No. 4 is +11.89 per cent based on the 1960 census and +7.10 per cent on the 1966 estimated population. The under-representation of District No. 5 has increased due to intervening population changes, mainly in Glynn County, from +3.45 to +7.10. Both districts will approximate the 10 per cent tolerance even with the counties removed although the underrepresentation will become overrepresentation.

There is a historical precedent for joining other counties with Chatham in senatorial representation. Bryan, Effingham, and Chatham made up one senatorial district for many years beginning more than one hundred years ago. Const. of 1865, § 576. From 1906 to the original reapportionment in 1962, Chatham and Effingham formed a senatorial district. Ga. Laws 1906, p. 80.

The result of reassigning Effingham County will be to leave a total 1960 population of 68,643 and a 1966 population of 71,400 in the 4th District, and an overrepresentation of -2.51 per cent and -10.15 per cent, respectively. The result of reassigning Bryan County will be to leave a total 1960 population of 66,679 and a population of 78,391 as of 1966 in the 5th Senatorial District. The 5th District will be overrepresented by -5.3 per cent based on the 1960 population and —1.35 per cent on the 1966 estimates. The 1960 total of Chatham, Effingham, and Bryan Counties is 204,669 which population, as of 1966, had decreased to 204,105. The result will be that Districts 1, 2, and 3 averaged and including Effingham and Bryan Counties, will be overrepresented by -3.11 per cent on 1960 population and -14.39 per cent on the 1966 estimates. Meanwhile, Chat-ham County may recoup some of its loss by the 1970 census.

According to the 1960 census the residents of District No. 8 are underrepresented to the extent of +11.10 per cent by the apportionment plan. This came about through the shift of Lanier County from District 7 to District 8 and Echols County from District 8 to District 7. This variance could easily be adjusted by reshifting these two counties. We conclude however, that this will not be necessary in light of the 1966 population estimates. District 7, according to these estimates, is overrepresented by -6.50 per cent instead of being underrepresented by +0.52 per cent based on the 1960 population. District 8 is overrepresented by —0.45 per cent according to the 1966 estimates instead of being underrepresented by +11.10 per cent as of 1960. We hold the variance in District 8 to be justified on the basis of present population.

District 11 is underrepresented by less than 1 per cent (+0.74) when considered in light of the 1960 population of the counties comprising that district. Their growth has been less than the overall state growth and based on the 1966 population estimates they are now overrepresented by -10.90 per cent. We do not believe this present variance to be so extraordinary as to require a departure from the 1960 population basis.

District 14 is comprised of six counties. One of the counties, Chattahoochee, has a large military population. The 1960 census shows the county population as 13,011. The Health Department estimates the 1966 population as 40,201. The evidence discloses that as of December 31, 1966 it had a military population of 43,801. It is clear that a large portion of the increase is in military population. In view of the transitory nature of the military population in the Port Benning area, part of which lies in Chattahoochee County, we think a fair basis would be to attribute only one third of this population increase to this senatorial district. Based on the 1960 census this district is underrepresented by +2.40 per cent. The percentage of variance is still well within a 10 per cent tolerance if we give consideration to the increase in military population in Chattahoochee County which we do. We therefore conclude that the present variance with respect to this district may be and is justified.

Districts 15 and 16 are located in Muscogee County, a two district county. The residents of these districts are underrepresented by +12.60 per cent and +12.68 per cent, respectively, based on the 1960 census. The evidence discloses that Muscogee County’s growth has not kept pace with that of the state at large. The population of the county in 1960 was 158,623. It had increased to only 163,434 by the end of 1966 according to the Health Department. It had a military population of 7,748. The 1960 variances were reduced, according to these estimates, by the end of 1966 to +2.84 per cent, dividing the county population equally as between the two districts. We hold that the 1960 variances of more than 10 per cent are justified in light of the current population of the county.

District No. 21 is composed of four counties, Jefferson, Burke, Emanuel and Jenkins. The overrepresentation in this district has increased from -7.65 per cent on 1960 population to -15.05 per cent in 1966. Each county has shown growth but not in proportion to that of the state as a whole. We conclude that the loss has not been so disproportionate as to require a departure from the 1960 variance which was within the 10 per cent tolerance.

Districts 22 and 23 are allocated to Richmond County. One of these districts is assigned to the City of Augusta and the other to that part of Richmond County which lies outside the city. The city district varies from the norm by only +0.30 per cent according to the 1960 census while the county district varies by -7.72 per cent. The testimony was that these districts presently have about the same population. The problem comes in justifying a present variance of +15.16 per cent as to the districts averaged. Here again we must deal with the large military population at Fort Gordon. According to the evidence that military population was 33,831 as of December 31,1966. We can assume that much of this military population is transitory and therefore we find that the variance based on the present estimated 1966 population may be justified as being well within the 10 per cent tolerance.

DeKalb County was allocated three senatorial districts, (41, 42, and 43) in the 1962 reapportionment. It had a population of 256,782 according to the 1960 census and this meant that the county was underrepresented by +21.69 per cent. This county was allocated an additional senator (No. 55) under the 1967 reapportionment, assuming that the constitutional amendment is adopted. We proceed on the assumption that the constitutional amendment will be adopted. Using the 1960 population, this means that DeKalb County with four senators will be overrepresented by -8.83 per cent. However, based on the Health Department estimated population as of July 1, 1966 the population has increased to 360,600 and the residents of that county will be thus underrepresented by +13.45 per cent even with an additional senate seat. According to the Census Bureau, DeKalb, as of July 1, 1965, had a population of only 323,000. The difference between the Health Department estimates and the census figure, although representing dates twelve months apart, is substantial. We conclude that the true population of DeKalb County lies somewhere in between the two figures and that the present variance does not greatly exceed 10 percent if at all. We hold that the present variance is justified as to the four districts in DeKalb County.

Two senatorial districts are allocated to Cobb County, Nos. 32 and 33. Cobb County is experiencing rapid growth. It had a population in 1960 of only 114,174. The residents of District No. 32 were overrepresented by -19.66 per cent while those of the other were overrepresented by -18.19 per cent according to the 1960 census. The Health Department estimates that Cobb County had a population of 163,632 in 1966. This increase is buttressed by the Census Bureau which gives Cobb County a population of 153,-000 as of July 1, 1965. Dividing the 1966 Health Department estimate equally between the two districts it appears that the variance is an underrepresentation of +2.96 per cent. We find the 1960 variances to be justified in light of the extraordinary population growth.

District 44 is comprised of Clayton, Henry, and Rockdale Counties. These counties are in the Atlanta metropolitan area and have experienced a rapid growth. The variance from the norm according to the 1960 census was +5.88 per cent. Based on the 1966 Health Department estimates this underrepresentation had increased to +42.11 per cent. This can only be classified as an extraordinarily disproportionate growth when compared to the growth of the state at large. We disapprove the plan as it relates to this district. It cannot be justified. The underrepresentation can be greatly decreased by moving Rockdale County which had a 1960 population of 10,572 and which has a Health Department estimated 1966 population of 15,700 to another senatorial district.

District 48 is adjacent but, as will be hereinafter discussed, it is already underrepresented. This, as stated, is also true as to DeKalb County. This leaves only the 45th Senatorial District for Rockdale County. That district is comprised of the counties of Walton, Newton, Morgan, Jasper, Putnam, and Jones at present. It is underrepresented by +5.32 per cent as of 1960 and+0.55 per cent at the present time. If Rockdale is to go to the 45th, one of the counties presently in the 45th must be shifted to an adjacent district. As stated, District 48 is underrepresented and this leaves Districts 18, 24, 25, 26, 27, 28, and 46 for consideration. The desired result of minimum variance, compactness and contiguity can best be accomplished by removing Jones County from the 45th District to be included with Bibb County to which is allocated Districts 26 and 27.

Those citizens of Bibb County residing in District 26 are overrepresented by — 1.3 per cent according to the 1960 census while those in District 27 are underrepresented by +1.97 per cent. According to the 1966 Health Department estimates, the residents of these two districts, averaged, are presently overrepresented by —6.75 per cent. This is based on the estimated population for Bibb County of 148,190 as of July 1, 1966. This is in line with the census estimate as of July 1, 1965 of 149,000.

The addition of Rockdale to the 45th District and of Jones to Bibb County (Districts 26 and 27) will achieve the following results, based on 1966 estimates: District 44 will be comprised of Clayton and Henry Counties with an estimated total population of 97,224. Its 1966 underrepresentation will be reduced from +42.11 per cent to +22.4 per cent as compared to an overrepresentation of -9.17 per cent using the 1960 census. The 1965 Census Bureau estimate for Clayton County is 68,000. (There is no Census Bureau estimate for Henry County.) Using this data the underrepresentation is reduced to +17.3 per cent instead of +22.4.² Rockdale will become a part of District 45 which district will have a total 1960 population of 76,215 and an underrepresentation of +8.27 per cent; a 1966 population of 84,300 and an under-representation of +6.12 per cent. Districts 26 and 27, considered together and including Jones County will have a 1960 population of 149,717 and a current population of 159,490; a 1960 underrepresentation of 6.34 per cent, and a 1966 underrepresentation of -.35 per cent.

This leaves Senatorial District 48 for consideration. This district is comprised of Gwinnett, Barrow, and* Jackson Counties. The residents of this district are underrepresented by +8.68 per cent based on the 1960 census. The current underrepresentation is +13.77 per cent based on the Health Department estimates. Jackson and Gwinnett Counties are separated by Barrow County. In order to preserve contiguity and in view of the population of Gwinnett County, it is only feasible to join Jackson County with another district. The population of Jackson County was 18,499 according to the 1960 census and it now has an estimated population of 19,400. It can be joined to Districts 46, 47 or 49. These districts were all underrepresented according to the 1960 census within a maximum variance of +3.30 per cent. Using 1966 population data, District 46 is overrepresented by -6.83 per cent; District 47 by -8.13 per cent, while District 49 is underrepresented by +0.43 per cent. The removal of Jackson County from District 48 would reduce the present under-representation of +13.77 per cent to an overrepresentation of -10.7 per cent. It would, on the other hand, reduce the overrepresentation of District 46 from -6.83 per cent to an underrepresentation of +17.7 per cent; the overrepresentation of District 47 from —8.13 per cent to an underrepresentation of +16.3 per cent; and the underrepresentation in District 49 from +0.43 per cent to an underrepresentation of +24.4 per cent. Using equitable considerations in the context of equal protection, it is apparent that the net result of these changes would not be such as to greatly improve the present situation among these districts. We conclude that it is proper to justify the variance in District 48 in the absence of extraordinarily disproportionate intervening growth as is the case.

In sum, we approve the reapportionment plan of the Senate except as stated. Our approval is on the assumption that the constitutional amendments creating the two extra Senate seats will be adopted. The General Assembly, from the beginning of the reapportionment litigation, has chosen to formulate its own plans of apportionment instead of leaving formulation to the court. This course has been followed in the interest of maintaining legislative stability through the application of the knowledge and experience of the membership to the task of creating legislative districts. Our approach is to approve the plans as submitted to the extent possible. Once the indicated changes are made with respect to Districts 1, 2, 3, 4, and 5 as they involve Effingham and Bryan Counties, and Districts 26, 27, 44, and 45 as they involve shifting Rockdale from the 44th to the 45 th and Jones County from the 45th to the Bibb County districts, 26 and 27, the entire Senate apportionment plan may be submitted for approval.

THE HOUSE

The assignment of bringing the apportionment of the House within constitutional standards is more difficult. For many years it has been composed of 205 representatives, one of the largest in the nation. The 1967 plan reduces this number to 195. The only other reapportionment of the House since Reynolds v. Sims, supra, which required that both the House and Senate in a bicameral legislature be apportioned on population, was in 1965. We approyed it as an interim measure and our action was affirmed by the Supreme Court. Toombs v. Fortson, supra.

The 1967 plan divides the state into 116 legislative districts with some, as distinguished from the Senate, having multiple seats. In the case of Rockdale County, as was the case in 1965, it is joined with one of the three DeKalb County districts. Again using a tolerance of 10 per cent we find the situation as follows:

CHART V

1960 Census — House

Percentage Departure No. of No. of Reprefrom Norm Districts sentatives Population

10-15% below 19^* 31 547,888

5-10% below 12 17 317,504

0-5% below 24 47 931,387

0-5% above 23 38 787,176

5-10% above 28 37 795,603

10-15% above 10 25 563,558

It is to be seen that 1,111,446 Georgians reside in the districts which deviate more than 10 per cent from the average. This is 28 per cent of the total 1960 population. Approximately the same number reside in the districts which deviate from 5 to 10 per cent from the average or norm while the balance, approximately 44 per cent, reside in the districts which deviate less than 5 per cent.

The specific districts which exceed the 10 per cent tolerance, together with their population, the number of seats involved; and the degree of variance is reflected on the following chart:

CHART VI

District No. No. of Seats Variance Population-1960

1 3 —11.10 53,930

3 . 3 -13.36 52,556

4 1 +11.48 22,542

5 1 -14.24 17,341

9 3 +13.96 69,130

12 1 -10.41 18,116

14 2 —13.46 35,000

18 2 +12.17 45,363

20 1 -11.80 17,835

23 3 —10.05 54,564

24 1 -12.87 17,619

29 1 +11.46 22,538

30 1 —12.09 17,777

31 2 —10.78 36,082

34 2 -12.46 35,404

38 1 -13.61 17,468

39 2 +10.44 44,663

45 1 —11.90 17,815

46 2 -14.92 34,407

49 2 +13.56 45,924

51 2 +13.85 46,044

52 1 —13.46 17,500

59 1 -11.37 17,921

66 2 -15.38 34,219

70 1 —10.91 18,015

71 2 -15.14 34,319

DeKalb and Rockdale Counties (73, 74, 75 with 12 seats averaged) +10.18 267,354

The above chart reflects that 29 districts having a total of 56 seats deviate more than 10 per cent from the average. Excluding these 29 districts the maximum variance is from +9.52 per cent to -9.87 per cent.

Having in mind the large portion of the total population which is outside the 10 per cent tolerance together with this number of districts and seats, it is clear to the court that there must be some corrective action with respect to the House plan. Some but not all of the variances can be justified. This conclusion is based on the variances which caused the plans in Swann v. Adams and Kilgarlin v. Hill, supra, to be disapproved. In addition, the plan must be adjusted to compensate for unusual growth situations.

The evidence before us, based on the 1966 Health Department population estimates indicates that 8 districts having a total of 17 seats are now within 10 per cent of the average due to intervening population changes. We find the variances with respect to these 8 districts, 3, 4, 9, 18, 31, 39, 49, and 51, to be thus justified. The variances in these districts now range from -7.39 per cent to +5.48 per cent. This leaves 21 districts with a total of 39 seats to be considered.

We begin with the 3 districts (73, 74, and 75) and 12 seats which are allocated to DeKalb and Rockdale Counties jointly. These districts, averaged, are underrepresented to the extent of +10.18 per cent based on the 1960 census. The picture at the present time is one of extraordinary disproportionate growth when compared to the state growth. See Chart IV. The 1960 population of DeKalb County had increased from 256,782 to 360,600 by July 1, 1966 according to the Health Department or to 323,000 as of July 1, 1965 according to the Census Bureau. The population of Rockdale County had increased from 10,572 to 15,700 by July 1, 1966 according to the Health Department. No Census Bureau estimate is available for Rockdale County. Based on the Health Department estimates the underrepresentation in these districts had increased from +10.18 per cent to +37.42 per cent. Adjusting the Health Department estimates on DeKalb County somewhat in view of the Census Bureau estimate, as we did in the case of the Senate plan, it nevertheless appears that these counties are now grossly underrepresented. We find that it will be necessary to modify the House plan to add three additional representatives to these counties, Districts 73, 74, and 75, making a total of 15. This will bring the present variance within the 10 per cent range.

As Chart IV, supra, indicates, the growth in Clayton County has also been extraordinarily disproportionate. Some relief must be afforded because of this growth. Clayton County is now joined with Fayette County to make up District 23 with 3 representatives. Spalding County lies adjacent as a separate district, No. 34, with two representatives. The 1960 population of Clayton County increased from 46,365 to 72,024 as of July 1, 1966 according to the Health Department. This growth is supported by the Census Bureau estimate of 68,000 as of July 1, 1965. The population of Fayette County had increased from 8,199 to 8,500 according to the Health Department. The overrepresentation of Clayton and Fayette based on 1960 populations of -10.15 per cent has now become an under-representation of +17.62 per cent. The overrepresentation of Spalding County as of 1960 of -12.46 per cent had increased to —14.76 per cent by 1966. Fayette County should be removed from the 23rd District and assigned to the 34th District with Spalding, leaving only Clayton County in the 23rd with 3 representatives and with the 34th to continue with 2 representatives. The result will be an underrepresentation in the ease of Clayton County based on 1966 data of +5.21 per cent and in the case of Spalding and Fayette, on 1966 data, an underrepresentation of +3.86 per cent.

Cobb, Paulding, and Douglas Counties form the 116th District. There are all high growth rate counties. See Chart IV. Intervenors from Douglas County, seeking a separate district for that county, introduced evidence which demonstrated that the county is experiencing many problems pertaining to urban growth and which require local legislation. This county until recently was mainly a rural county. It is adjacent to Fulton County, in the metropolitan Atlanta area, and additional growth can be anticipated. The population of Douglas County has increased from 16,741 in 1960 to a 1966 population of 22,700 according to the Health Department. The 1966 norm is 22,819. We agree that Douglas County is entitled to a separate representative. This will leave Cobb and Paulding Counties in the 116th District with 7 representatives. These counties are entitled to keep the 7 representatives in light of their growth. Together they will have a 1960 overrepresentation of -10.0 per cent but a 1966 underrepresentation of +12.6 per cent. Douglas County will have an overrepresentation of -17.4 per cent on 1960 population but a 1966 over-representation of only -0.52 per cent. The sum of the changes as between these 3 counties is that one additional representative must be found to supply the representative to Douglas County.

As will be discussed in some detail, and as has already been pointed out, there are other counties which have experienced sharp growth but we do not think a case has been made for additional representation for any of them at this time. We come then to the proposition of finding the 4 additional seats, 3 for the DeKalb and Rockdale districts, and the other for Douglas County.

Chatham presently has 9 representatives with each being in a separate district, Nos. 85 through 93. The population of Chatham County, as was mentioned in connection with the Senate plan, has declined from 188,299 in 1960 to 183,705 as of July 1, 1966. The Health Department 1966 estimate differs by some 8,000 from the Census Bureau estimate of 192,000 on July 1, 1965. In either event, based on 1966 norm or 22,819, it appears that Chatham County is entitled to only 8 representatives at this time. The norm has increased due to the state population' growth and the reduction in the size of the House. It would need a population of 205,371 or thereabout to sustain 9 representatives. Even using the 1965 Census Bureau estimate, Chatham County would be underrepresented by only +5.18 per cent with 8 representatives. Using the 1966 Health Department estimate it will be underrepresented by +0.63 per cent. We hold that Chatham County must give up one of its representatives and this will make one seat available of the 4 needed in the DeKalb, Rockdale, -and Douglas change.

The same situation is true in Muscogee County. It is allocated 3 districts with a total of 8 seats. Its 1960 population of 158,623 had increased to only 163,434 as of 1966 according to the Health Department. This population will sustain only 7 seats at present. The overrepresentation of -1.94 per cent with 8 representatives in 1960 had increased to an overrepresentation of —10.47 per cent in 1966. With 7 representatives the county will be underrepresented by +12.0 per cent on 1960 population but only by +2.32 per cent based on 1966 population. We conclude that Muscogee County must give up one representative for use in the DeKalb, Rockdale, and Douglas change.

One of the needed representatives should come from the north Georgia area. District 5 and 12 exceed the tolerance based on the 1960 census by -14.24 per cent and -10.41 per cent, respectively. The trend in each is toward more over-representation: -20.68 per cent and -15.86 per cent, respectively, as of 1966. Neighboring District No. 13 has also fallen behind based on current data from a 1960 overrepresentation of -9.05 per cent to -16.30 per cent. The same is true of the 6th District—from -6.39 per cent to -12.79 per cent. The population trends are established by the evidence and it is clear that there will be some loss in representation in this area of the state after the next census. The over-representation at this time can hardly be justified in light of the need elsewhere. It can be eliminated by a new grouping of the counties in Districts 5, 6, 11, 12, and 13. The result of the new grouping will be compact and contiguous districts and the release of one legislative seat for use elsewhere in the DeKalb, Rock-dale, and Douglas situations.

In the regrouping Dawson County should be moved from the 5th to the 11th District with Hall and Forsyth Counties. The result-will be a 1960 population of 65,499 and a total 1966 population of 72,200. The new 11th District with 3 representatives will be underrepresented by +7.98 per cent on I960 populations and +5.47 per cent on the 1966 estimates.

Aside from Dawson, Hall, and Horsy th Counties which will make up the new 11th District, a combination or regrouping of the seven counties in the 5th, 6th, 12th, and 13th Districts is to be considered. These are the counties of Union, Lumpkin, Towns, White, Rabun, Habersham, and Stephens. They can be combined into one district with three representatives. Such a seven county district will be above the present five county maximum House district under the plan [Nos. 47 and 65], but the underrepresentation of such a district based on 1966 estimated population of 69,187 would be only +6.06 per cent as compared to an underrepresentation based on 1960 population (72,600) of +14.06 per cent.

As an alternative available to the General Assembly, Union and Lumpkin Counties which make up the remainder of the 5th District may be combined with Towns and White of the 6th District into one district. The combined group of Union, Lumpkin, Towns, and White will have a 1960 population of 25,244 and a 1966 population of 26,400. One representative would be allocated to this new district and the underrepresentation based on the 1960 population would by +24.76 per cent but down to +15.7 per cent on 1966 population estimates. This excessive variance may be justified under the circumstances as a temporary measure since this underrepresentation should be under 10 per cent in light of the population trend, by the time of the 1970 census. Rabun County, now in the 6th, would be grouped with Habersham and Stephens Counties which make up, respectively, the 12th and 13th Districts, so as to form one district consisting of Rabun, Habersham, and Stephens. Under this alternative, two out of the four districts of 5, 6,12, and 13 would be eliminated. A district formed of Rabun, Habersham, and Stephens would have a combined 1960 population of 43,963 and a 1966 population of 46,200. Two representatives would be allocated to this district and the underrepresentation based on the 1960 population would be +8.73 per cent but only +1.23 per cent on the 1966 estimates.

As will be seen, one representative is to be taken from a recombination of the 64th, 65th, and 66th Districts in south Georgia for the DeKalb, Rockdale, and Clayton deficit. This is, nevertheless, a further alternative to taking the representative from this group of north Georgia counties and that is to take the representative, instead, from the southwest Georgia area comprising Districts 69, 70, and 71. The three districts combined (5 representatives) are overrepresented, as is the case in north Georgia, to the extent of one representative on either 1960 or 1966 figures. As combined they would make a five county district but the district would have a larger land area than the north Georgia group of seven counties and also a larger population. The variance in the combined southwest Georgia district would be an underrepresentation of +12.83 per cent on 1960 populations and +2.78 per cent on the 1966 estimates as compared to the north Georgia seven county group variance of +14.06 and +6.06 per cent, respectively. In any event, this is a choice for the General Assembly to make between leaving north Georgia overrepresented or southwest Georgia overrepresented and between taking two representatives from south Georgia or one from north and one from south Georgia.

The overrepresentation in the 66th District which is composed of Ware County with two representatives was -15.38 per cent in 1960 and that overrepresentation had increased to —22.43 per cent by 1966 according to the Health Department estimated population. Three representatives are allocated to the 64th District which is comprised of Tift and Colquitt Counties. The overrepresentation there of —5.16 per cent had increased by 1966 to -14.40 per cent. Cook County can be removed from the 65th District, a five county district, and added to the 64th District and leave this district with three representatives and an underrepresentation of +3.86 per cent according to the 1966 estimates. The remainder of the 65th District consisting of Berrien, Atkinson, Lanier, and Clinch Counties can be combined with the 66th District, Ware County, still leaving a five county district, and this will make a district of a 1960 total population of 64.087 and a 1966 estimated population of 66,900. These two districts combined now have four representatives and this number can be reduced to three with a resultant 1966 overrepresentation of -2.27 per cent. This will provide the fourth representative for allocation because of the DeKalb, Rockdale, and Douglas change.

District No. 1 is made up of Dade and Walker Counties with three representatives. These counties lie adjacent to Chattanooga, Tennessee. Walker County is experiencing growth somewhat above the state average. See Chart IV. The overrepresentation of —11.10 per cent in this district based on the 1960 census had decreased to -10.61 per cent by 1966 and continued growth is anticipated. We think the departure from the tolerance is thus justified in this instance.

This same situation exists in District 24, a one representative district composed of Henry County. This county is overrepresented by -12.87 per cent on the 1960 population but it is experiencing rapid growth. Presently it is underrepresented to the extent of +10.43 per cent but this variance does not warrant additional representation at this time.

The population of Burke County was 20,596 in 1960 and it now has an estimated population of 21,440. It has been included in the 39th Legislative District consisting of Burke, Jenkins, and Screven Counties. The citizens of Burke County have intervened with the complaint that the county should be allocated a separate House district in view of the nearness of its population to the norm and particularly in view of the fact that Emanuel County, with a 1960 population of 17,815 and an estimated 1966 population of 18,-500 was awarded a separate district. Nothing was said of its next door neighbor, Jefferson of the 38th District, which exceeds the variance even more than Emanuel. We think fairness dictates that Burke County be given a separate district with one representative. This can be accomplished by joining Jenkins and Screven Counties, now in the 39th, with Bulloch and Effingham of the 46th District. The 46th is overrepresented by -14.92 per cent on 1960 population and -14.33 per cent on 1966 population. Jenkins and Screven will bring a representative with them to go with the two now assigned to the 46th but the resulting district will be overrepresented on 1960 population by only -6.62 per cent and on 1966 population by -5.34 per cent.

The above changes and justifications, in the case of Districts 1 and 24, will obviate the 1960 census deviation of more than 10 per cent in Districts 1, 5, 12, 24, 34, 46, 66, 73, 74, and 75. This leaves a total of 10 districts with 12 representatives to be discussed. These are not easily justified.³

These 10 districts, Nos. 14, 20, 29, 30, 38, 45, 52, 59, 70, and 71 have a total of 12 seats. They have a combined 1960 population of 216,188 or 5.45 per cent of the state total. Of this total all but 22,538 in District No. 29 is overrepresentation.

The following chart reflects the scope of the variances in these 10 districts. Districts 29 and 30 will be discussed separately, They adjoin and a solution after the 1970 census is apparent. The other 8 districts are overrepresented on the 1960 population and the overrepresentation had increased in every case by 1966.

CHART VII

District Percentage of Population Overrepresentation^* 1960 1966 1960 1966

71 (Thomas) —15.14 —17.39 34,319 37,700

38 (Jefferson) —13.61 —20.68 17,468 18,100

14 (Banks Franklin Hart) —13.46 —21.99 35,000 35,600

52 (Treutlen Wheeler Montgomery) -13.46 -21.12 17,500 18,000

30 (Putnam Hancock) -12.09 -19.37 17,777 18,400

45 (Emmanuel) -11.90 —18.93 17,815 18,500

20 (Elbert) —11.80 —19.80 17,835 18,300

59 (Wayne) -11.37 —15.86 17,921 19,200

70 (Grady) -10.91 —18.05 18,015 18,700

29 (Jasper Jones Twiggs) +11.46 +14.55 22,538 26,138

District 29 is composed of Jasper, Jones, and Twiggs Counties and is underrepresented by +11.46 per cent based on the 1960 population and +14.55 per cent on the 1966 estimate. District 30 adjoins and is composed of Putnam and Hancock Counties with a 1960 overrepresentation of -12.09 per cent and a 1966 overrepresentation of -19.37 per cent. The growth of Jones County has been substantial. Chart IV, supra. It now appears that the situation as between these districts could be imposed slightly by combining Jasper with Putnam and Hancock Counties as District 30 with Jones and Twiggs being left in the 29th. This would have the effect of Jasper, Putnam, and Hancock with a 1966 population of 24,600 being underrepresented by +7.81 per cent. Jones and Twiggs would have a 1966 population of 19,938 and would be overrepresented -12.5 per cent. This trend is established but the improvement in variances would not be such as to require the switch of Jasper County at this time.

With respect to the other 8 excessive deviations, we will discuss Districts 14 and 20 together. They are substantially overrepresented as are neighboring Districts 17 and 19. The 15th District, Gwinnett County, was underrepresented by +7.67 per cent on the 1960 population and that underrepresentation had increased to +22.70 per cent in 1966. We are not requiring extra representation for Gwinnett at this time since it would be necessary to join another county with it but the trend is in that direction. District 18, Clarke County, in the same area of the state has also shown some growth. It may be necessary to attach one or more counties to Clarke and to Gwinnett in order to achieve an equitable result in this area of the state. The overrepresentation now so clear in Districts 14, 17, 19, and 20 must be eliminated after the 1970 census to accord the needed representation to these or other growth counties.

Districts 38, 45, and 52 are also overrepresented with the trend being toward more overrepresentation. This is likewise true of neighboring Districts 28, 37, and 53. Richmond and Columbia Counties are growing and there will doubtless have to be a rearrangement in these districts to find additional representatives if needed in the Richmond-Columbia area. The 1970 census will make reliable information available as to this growth. The present data is complicated by the Fort Gordon population in Richmond.

District 59, Wayne County, is outside the tolerance in overrepresentation and its overrepresentation is increasing. The 68th District, Glynn County, is adjacent and its underrepresentation has increased from +3.74 per cent in 1960 to +16.83 per cent as of 1966. We note that the combination of these two districts at this time would result in a total population of 72,518 on the 1966 estimate with 3 representatives and an underrepresentation of 5.93 per cent. We will not require this to be done at this time but the trend is established. Glynn will be entitled to more representation after the 1970 census.

The 47th District is composed of 5 counties and the current military population of one, Chattahoochee, as discussed in connection with the Senate plan, indicates a gross underrepresentation in that district. It is overrepresented based on the 1960 census and we will leave the matter as is pending more reliable information as to population.

The population trend indicates that Houston County and possibly Dougherty may be entitled to additional representation after the 1970 census and the same may be said of all counties in the Atlanta metropolitan area and on the periphery of that area. We have already discussed several districts from whence needed representatives may be obtained. Districts 70 and 71, compromised of Grady and Thomas Counties, respectively, as heretofore noted, are in this group of districts which exceed the tolerance. The overrepresentation had increased by 1966. The underrepresentation in neighboring Districts 55, 56, and 69 had also increased well past the tolerance. These are apt districts to supply the representatives needed in the growth areas either now or after the 1970 census but the present population data does not dictate that representation be transferred at this time.

The 1970 census result may well be one of large districts or with counties divided as between districts. The present situation is that only the 71st (Thomas County) out of the 10 districts with which we are concerned would deviate from the average more than 10 per cent on 1960 population and a 205 member House. Our review of the House plan is based on a House of 195 representatives and this decision is rendered on the assumption that the number will not be increased.

We justify the variances in these 10 districts pending the 1970 census on the reasoning that the overrepresentation is not required elsewhere to correct an excessive variance on 1960 population or because of underrepresentation caused by extraordinarily disproportionate growth since 1960. We justify them also because they are not so great when combined populationwise (approximately 5 per cent of the 1960 state population) as to vary the entire state apportionment picture, taken as a whole, from being substantially with a 10 per cent range of deviation from the average.

In sum, the following changes will be required in the House plan. The DeKalb-Rockdale County districts (73, 74, and 75) must be allocated 3 additional representatives. Douglas County must be constituted as a separate district and afforded one representative. The 4 needed representatives will come, one each, from the north Georgia area or in the alternative, from the south Georgia area as herein described; Chatham County, Muscogee County; and from the recombination of Districts 64, 65, and 66 in south Georgia. Fayette County must be removed from the 23rd District and put in the 34th District with Spalding with 2 representatives. This will leave Clayton in the 23rd District with 3 representatives. Jenkins and Screven Counties must be combined with Bulloch and Effingham with 3 representatives. Burke County will be constituted as a separate district with 1 representative.

Once these changes are made, the Senate plan and the House plan may be presented for approval. A final judgment may also be presented.

1 . Some two weeks after the hearing two citizens of DeKalb County attempted to intervene. Their motion to intervene is denied. All questions asserted by them which are not frivolous are being pursued by the original plaintiffs in the class action.

* Included since Districts 1, 2, and 3 are in Chatham County and the county as a whole is overrepresented by more than 10 per cent.

* It is impossible to calculate the rate of growth of Chattahoochee and Richmond Counties precisely in light of their 1966 military populations, some of which may be transient, of 38,701 and 37,513, respectively.

2 . Additional relief is being afforded Clayton County under the House reapportionment plan. See pp. 25-26, supra.

* Districts 66 and 71 with variances of included in this , group. -15.38 and -15.14 respectively are

3 . While we are not at this time requiring the division of any county as between districts, it is not amiss to say that this would not be an onerous requirement. The current inclusion of Rock-dale County with one district in DeKalb County is a precedent. This system should be followed in future apportioning where it is necessary to-sveid excessive variances.

* Includes District 29 which is underrepresented.