The issues raised on appeal pose important questions regarding the propriety of certain procedures established by the Secretary of Commerce and the Director of the Bureau of the Census for the taking of the Nineteenth Decennial Census of the United States (1970 Census).

Two of the plaintiffs below, the City of Philadelphia and James G. Fulton, a Pennsylvania Congressman, appeal from an order of the district court denying, after a hearing, their requests for injunctive relief and entering a declaratory judgment in favor of appellees, 319 F. Supp. 971. The case was heard below on a stipulation of facts and an unchallenged affidavit of Dr. Conrad F. Taeuber, Associate Director of the Bureau of the Census, filed on behalf of appellees.

Specifically, appellants challenge as violative of Art. I, § 2, the Fifth and the Fourteenth Amendments to the Constitution, and the Census Acts of the United States, the action of the Bureau of the Census whereby college students, members of the Armed Services stationed in the United States, and inmates of institutions were enumerated in the 1970 Census as inhabitants of the respective states in which their colleges, military bases, or institutions were located, and not, say appellants, the states which such individuals considered their "legal residence for all purposes other than the census." Appellants assert that, as a result of this method of enumeration, the population of their political subdivisions has been underestimated since more people in the above categories leave than enter these particular subdivisions. It is claimed that as a consequence: (1) certain political subdivisions will be denied their proper share of various funds allocated by both the Federal Government and the Commonwealth of Pennsylvania according to the federal census, (2) these subdivisions will be denied proper representation in the House of Representatives and the Legislature of the Commonwealth of Pennsylvania, and (3) voters in these subdivisions will have the weight of their votes in federal and local elections improperly diluted.

Appellants also challenge the action of the Bureau of the Census whereby military and other government personnel serving outside the United States are not enumerated in any specific political subdivision, although they are enumerated in the state where they maintained their former residences.

In their request for relief in the district court, appellants sought, inter alia, to restrain the Department of Commerce from adopting and promulgating the compilations obtained from the taking of the 1970 Census, to restrain future use of the allegedly invalid enumeration procedures, and to compel a retaking of the 1970 Census.

Art. I, § 2, as amended by Section 2 of the Fourteenth Amendment to the Constitution, provides:

"Representatives shall be apportioned among the several States according to their respective numbers, counting the whole number of persons in each State, excluding Indians not taxed." U.S.Const. Amend. 14, § 2.

"The actual Enumeration shall be made within three Years of the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct." U.S.Const. Art. I, § 2.

Congress has consistently provided for the contemplated enumeration. The Census Act contained in Title 13 of the United States Code governed the 1970 Census. It provides in pertinent part:

"(a) The Secretary [of Commerce] shall, in the year 1960 and every ten years thereafter, take a census of population * * * as of the first day of April, which shall be known as the census date.

"(b) The tabulation of total population by States as required for the apportionment of Representatives shall be completed within eight months of the census date and reported by the Secretary to the President of the United States." 13 U.S.C. § 141.

Congress, however, did not legislate concerning the details of the enumeration but instead delegated this authority to the Secretary of Commerce. 13 U.S.C. § 5. In addition, Congress permitted the Secretary to delegate his authority to the Bureau of the Census, an agency within the Department of Commerce. 13 U.S.C. § 4. After the enumeration is completed, the President is required to transmit to Congress a statement showing the "whole number of persons in each State" as ascertained by the census and the number of representatives to which each state is entitled. 2 U.S.C. § 2a.

The question with which we are concerned is the proper interpretation of the phrase "whole number of persons in each State" as it appears in the Constitution and the various acts of Congress. The phrase has been historically interpreted by the Bureau of Census to embrace those persons whose usual place of residence is a particular state on the date of the census, as well as those persons present in the state who have no usual place of residence. This interpretation is derived from the language of the First Decennial Census Act adopted in 1790:

"Be it enacted, That every person whose usual place of abode shall be in any family on the aforesaid first Monday in August next, shall be returned as of such family. * * *" 1 Stat. 101.

The 1790 Act also required the enumeration of persons "without a settled place of residence" and persons "occasionally absent at the time of the enumeration, as belonging to that place in which he usually resides in the United States." Id. There are no similar provisions in the present United States Code. However, it has been stipulated that the following criterion was used by the Bureau of the Census to determine usual place of residence for the 1970 Census:

"Persons are enumerated at the place in which they generally eat, sleep and work, with persons who are temporarily absent for days or weeks from such usual place of abode being counted as residents of their usual place of abode."

We believe that the criterion adopted by the Bureau is a historically reasonable means of interpreting the Constitutional and legislative phrase "whole number of persons in each State."

Before we consider whether this criterion has been reasonably applied to the categories of people which appellants' claims have brought into issue, we must resolve appellants' claim that the district court erred in placing the burden upon them to demonstrate that the application of this criterion lacked a rational basis. Appellants rely on Swann v. Adams, 385 U.S. 440, 443-444, 87 S. Ct. 569, 572, 17 L. Ed. 2d 501 (1967), in which the Court placed the burden upon the State "to articulate acceptable reasons for the variations among the populations of the various legislative districts" for the purpose of the apportionment of the State legislature. They reason that since the apportionment of the House of Representatives depends upon the method of enumeration of persons in the 1970 Census, "the burden of proof [is upon] the Census Bureau to justify any deviation from a standard of population equality once Appellants establish such a deviation exists." We reject this contention since we believe that the reasoning upon which it is based is circular. The only way in which appellants can demonstrate that a deviation from population equality will exist in congressional districts as a result of the 1970 Census is to prove that the application of the Bureau's criterion used to enumerate the "whole numbers of persons in each State" is not consistent with the Constitution or the Census Act. Hence appellants' burden below was to demonstrate that the Bureau failed to apply the proper criterion in a reasonable manner or its application lacked a rational basis.1 Compare N.L.R.B. v. Hearst Publications, Inc., 322 U.S. 111, 64 S. Ct. 851, 88 L. Ed. 1170 (1944); Lewes Dairy, Inc. v. Freeman, 401 F.2d 308, 316 (3d Cir. 1968), cert. denied, 394 U.S. 929, 89 S. Ct. 1187, 22 L. Ed. 2d 455 (1969); Davis, Administrative Law Treatise, §§ 30.05, 30.11 (1958).

Prior to the 1950 Census, college students were enumerated as inhabitants of the states in which their parents maintained their homes. However, in the census for 1950, 1960, and 1970, college students were enumerated as inhabitants of the states in which they were living while attending college. The change resulted from the recommendation of a study conducted in 1948 by the Bureau's Technical Advisory Committee on General Population Statistics which reviewed the sharp increase in the number of college students. It was found that under the procedure used prior to the 1950 Census, the enumerations failed to reflect a number of college students who were not reported at either their parental homes or at their colleges. The Bureau of the Census determined that the best way to assure that all college students were enumerated was to count them in the states where they lived while attending college. Another reason for the change was to eliminate the inconsistency which resulted from the practice of not enumerating college students in the place where they usually eat, sleep, and work while this standard was uniformly applied to other major categories of people.

Appellants assert that if an individual college student indicates that he feels a particular connection or attachment to the state of his parental home, registers to vote in that state, and accordingly regards it as his home, the Bureau should consider these facts in determining his residence for the purpose of the census enumeration. We think, on the contrary, that the Bureau is entitled to limit its inquiry to the objective facts as to where the student chooses to generally eat, sleep, and work — the state of his college rather than the state where his contacts are substantially reduced. The enormous task which the Bureau must complete in enumerating people according to each state can reasonably necessitate the use of a definite, accurate and verifiable standard. Nor is our view altered by the fact that a college student may be designated by the state of his parental home as a resident or domiciliary and permitted to register there for voting purposes.2 Such matters are peculiarly for the state to determine unless they are otherwise regulated by the Congress, U.S.Const. Art. I, § 4, Oregon v. Mitchell, 400 U.S. 112, 91 S. Ct. 260, 27 L. Ed. 2d 272 (1970), or in conflict with the Constitution, Carrington v. Rash, 380 U.S. 89, 85 S. Ct. 775, 13 L. Ed. 2d 675 (1965). In contrast, formulation of the procedures to be used in taking the federal census is a matter exclusively covered by the authority of the Congress, United States ex rel. City of Atlanta v. Steuart, 60 App.D.C. 83, 47 F.2d 979 (D.C. Cir. 1931). Therefore, we are not in this case presented with the question nor do we express an opinion as to whether a college student can be denied the right in federal elections to vote in the state where his college is located, or whether the state of his parental home can properly permit him to vote there although his usual place of abode for purposes of the federal census has become his college state. Cf. Evans v. Cornman, 398 U.S. 419, 90 S. Ct. 1752, 26 L. Ed. 2d 370 (1970); Carrington v. Rash, 380 U.S. 89, 85 S. Ct. 775, 13 L. Ed. 2d 675 (1965).

Appellants assert that the method used by the Bureau to enumerate college students was adopted without an adequate foundation and was an abuse of discretion since the Bureau did not conduct a statistical survey to determine the actual number of college students who return to the state of their parental home after the completion of their college course. Art. I, § 2 provides for an enumeration after every term of ten years. The enumeration must therefore of necessity determine the "whole number of persons in each State" at a particular moment every ten years. Congress previously expressed a policy which would discount mere temporary movements of individuals as a basis for shifting their state for enumeration purposes. But we do not think an interpretation limiting the application of such an assumed policy to a period substantially less than that required to obtain a college education is unreasonable.

Appellants also claim that the Bureau used an improper factor in determining the state in which college students should be enumerated. They say that the Bureau considered the fact that college students increase the financial burden on the local governments of the college locations as a result of the increased demand for public services. Appellants claim that by taking cognizance of such a factor in making a decision, the Bureau deliberately followed a course of action calculated to influence the allocation of federal funds which are based upon census enumerations. Appellants did not offer any evidentiary support for this position. However, they argue that the affidavit given by Dr. Conrad Taeuber, Associate Director of the Bureau of the Census, which appears in the record, shows that the Bureau considered this as a factor in reaching its conclusions. Dr. Taeuber's affidavit describes the Bureau's reasons for enumerating persons of various categories in specific states. In what appears to be an argument in support of the Bureau's position, as distinguished from a factor considered in reaching the particular position, Dr. Taeuber indicates that the presence of certain classes of persons in a particular state increases the demand in that state for public services and that financial aid is given to political subdivisions on the basis of total population. Therefore, assuming arguendo that such a factor was irrelevant, we think it does not appear that it was a factor in the Bureau's decision to enumerate college students as they did.

We think that the decision of the Bureau concerning the enumeration of college students was reasonable. Once a person has left his parental home to pursue a course of study at a college in another state which normally will last for a period of years, it is reasonable to conclude that his usual place of abode ceases to be that of his parents. Such students usually eat, sleep, and work in the state where their college is located.3

Using the "usual place of abode" standard in the case of members of the Armed Forces stationed in the United States, the Bureau's practice has been to enumerate them at the specific location where they are living at the time of the census. Thus, they are most often enumerated in the states where their military bases are located. For the same reasons we expressed in the case of college students, we think the decision of the Bureau possessed a rational basis and that appellants did not show the use of an irrelevant factor.

The 1970 Census, for the first time, included in the calculation of "whole numbers of persons in each State" all American military and civilian personnel of the Federal Government stationed outside the United States and their dependents who are with them. These people are allocated to the state of their "home of record" — as it appears on military records or as they reported it on the 1970 overseas census report form — but not to any political subdivision thereof. Thus the people in this category are enumerated for purposes of apportionment of representatives among the states, but not for redistricting within the states. This change was prompted by a recommendation of the House Subcommittee on Census and Statistics which demonstrated concern over the large number of Federal Government employees serving overseas at the time of the 1970 Census. It was also determined that this procedure would not result in any state having its congressional representation changed since the estimated 2 million persons in this category are less than 1% of the total population and have approximately the same state-by-state distribution as the resident population. H.R.Rep. No. 91-1314, 91st Cong., 2d Sess. (1970), at 4.

Appellants do not challenge the decision to allocate these people back to the state of their "home of record" but insist that the Bureau should allocate them to a particular subdivision of the state. Apparently, the Bureau did not make such an allocation because it felt that this would be an impossible task. See H.R.Rep. No. 91-1314, supra at 4. Whether or not this explanation is persuasive, the fact remains that the census is used to determine the total number of representatives to which a state's population entitles it, and neither the Constitution nor the Census Act demands the allocation of persons to any particular subdivision of a state. Furthermore, appellants have failed to demonstrate that they will suffer any injury as a result of the Bureau's procedure. As noted above, the distribution of federal overseas personnel is approximately the same as the resident population and would not affect the percentage of representatives to which a particular subdivision is entitled. We therefore conclude that appellants have failed to impugn the Bureau's exercise of discretion.

Persons confined to institutions where individuals usually stay for long periods of time, such as penitentiaries or correctional institutions, mental institutions, homes for the needy or aged, or hospitals for the chronically ill, are enumerated as residents of the state where they are confined. People in this category, as distinguished from, for example, those temporarily in a hospital for a short duration, often have no other fixed place of abode, and the length of their institutional stay is often indefinite. As with college students and members of the Armed Forces stationed in the United States, we think that the decision of the Bureau as to the place of counting institution inmates has a rational basis.

Finally, appellants argue that the Bureau abused its discretion by not adopting a procedure to enumerate Americans who are studying or working overseas and who are not government employees. The Bureau takes the position and the district court found that there is no way of accurately locating these people. Appellants have not alleged any cognizable injury resulting from the procedure. In any event appellants have not presented us with any reason to conclude that the justification stated by the Bureau is without a proper basis.

Appellants, in their Constitutional challenge, claim that the procedures adopted for the enumeration of people with concurrent residences violate the principle of "one man-one vote." They claim that if an individual is enumerated in the state where he, for example, attends college, and votes in the state of his parental home, the weight of the votes of the electors in the college state will be increased and the weight of the votes of the electors in the state of his parental home will be proportionately decreased.

It is clear that the House of Representatives must reflect equal representation for equal numbers of people. This assures that, as nearly as is practicable, one man's vote is worth as much as another's in each congressional election. Thus, the Constitution demands that for the purposes of congressional elections, congressional districts are to have an equal number of people. Wesberry v. Sanders, 376 U.S. 1, 84 S. Ct. 526, 11 L. Ed. 2d 481 (1964). Since we have found the Bureau's enumeration method to be proper, the President will report to Congress the correct number of representatives to which each state is entitled in the House of Representatives, 2 U.S. C. § 2a(b), and the states will be able to draw their congressional districts in a manner which conforms with the requirements of the Constitution. If the representation of a state is increased in the House of Representatives because there is, for example, a large number of college students in that state, there is nothing in this record to show that the increase is not justified.4 We are not required to decide appellants' contention that present census procedures will result in an improper dilution of the vote. Whether a person enumerated in a particular state will be permitted to vote in that state, or may be permitted to vote in another state is, as we have already indicated, a matter ordinarily for the state to determine, not the Secretary of Commerce or the Bureau of the Census.

The judgment of the district court will be affirmed.

Judge ADAMS concurs in the result.