(dissenting).
Congress committed to an expert administrative agency, not to the courts, the responsibility both of protecting public safety and health and of accommodating energy needs and environmental interests in the licensing and regulation of commercial nuclear power plants. At the time of the administrative proceedings we are reviewing, that agency was the Atomic Energy Commission.1 We of course owe the expertise of an agency responsible for regulating the use of nuclear power the same deference we owe the expertise of other agencies, even though we have the normal human dis*531trust of such an awesome force as nuclear reaction and fear that the consequences of a miscalculation may be catastrophic. Congress no doubt had the same distrust and fear, but it nevertheless chose the policy of providing for the development of nuclear power as a source of energy and relied upon the agency’s expert judgment to protect society’s interests. Congress limited the courts’ role to review of the agency’s actions in accordance with the Administrative Procedure Act. I think that in this case the court has overstepped the limits of that role and substituted its judgment for that of the agency.
The court’s decision turns primarily upon the meaning of the words, “the nearest boundary of a densely populated center containing more than about 25,-000 residents,” which appear in the Commission’s regulation dealing with reactor site criteria. (10 C.F.R. § 100.11(a)(3) (1974).) The court holds that these words refer to the political boundary of a city and not, as the agency construed them, “that portion of the population center at which the dense population starts.” In rejecting the argument that the words refer to a political boundary, the Atomic Safety and Licensing Board, whose order was affirmed by the Appeals Board, noted that the regulation also provides,
“In applying this guide [that the population center distance be at least one and one third times the distance to the low population zone boundary], due consideration should be given to population distribution within the population center,” .
and concluded that,
“giving due consideration to the low population density of those areas closer than 4.5 miles to the Bailly reactor, principally industrial areas of the City of Portage, the provisions of Part 100 [of the regulations] with respect to population center distance are satisfied.”
The Atomic Energy Commission’s interpretation of its own regulation is entitled to “that respect which is customarily given to a practical administrative construction of a disputed provision.” Power Reactor Development Co. v. Electrical Union, 367 U.S. 396, 408, 81 S.Ct. 1529, 1535, 6 L.Ed.2d 924 (1961). “[S]ince the meaning of the language is not free from doubt, we are obligated to regard as controlling a reasonable, consistently applied administrative interpretation ..” Ehlert v. United States, 402 U.S. 99, 105, 91 S.Ct. 1319, 1323, 28 L.Ed.2d 625 (1971). The only suggestion in the court’s opinion in the case at bar that the agency has not consistently interpreted the disputed regulation is the reference to Southern California Edison Co. (San Onofre Station), ALAB-248, RAI—74-12, 957 (Dec. 24, 1974). The Commission’s position in that case, however, that city boundaries are not controlling when densely populated areas extend outside those boundaries, is entirely consistent with its position here. In both cases the location of populated areas rather than of political boundaries is viewed as controlling.
I do not understand why that is not the most sensible interpretation of the regulation, as well as the one to which we owe respect as the agency’s expert interpretation. Since the purpose of the regulation is to separate the reactor site from densely populated areas, we would expect the draftsmen to be concerned with the location of those areas rather than political lines. Nowhere in the regulation is there any reference to a city, village, town, or other political subdivision. Instead the terms used are “population center” and “densely populated center” (10 C.F.R. §§ 100.3(c), 100.-11(a)(3)(1974)), evidencing an intention to describe de facto concentrations of population rather than political subdivisions. The term “boundary” is used with reference to such a concentration or center, as the agency correctly concluded.
The court’s discussion of the evidence concerning population distribution, which is found near the end of Part III of the opinion, appears to be intended to illus*532trate the difficulty of finding the boundaries of a de facto population center, rather than an alternative holding that, even if the boundary referred to is that of such a center, the agency’s finding as to the location of the boundary is not supportable. Nevertheless, implicit in the court’s discussion is a rejection of the agency’s findings on this point, which I think are supported by substantial evidence, as required by section 10 of the Administrative Procedure Act (5 U.S.C. § 706(2)(E)), under which our review is conducted.2 It is true that the map used in the presentation of the evidence on population center distances was five or six years old. But there was ample additional evidence, which we would surely find sufficient if we were reviewing the judgment of a district court or an order of another agency, that the land within the two-mile population center distance, and indeed beyond, is not densely populated; 3 that its present and probable future development is principally industrial; 4 and that extensive residential development would be unlikely anywhere within the two-mile zone because of the industrial zoning, the industrial uses already there, and the availability of land elsewhere which is more desirable for residential use. It was for the agency and not this court to weigh and appraise that evidence.
Other grounds put forward in Part III of the court’s opinion seem to me to be unpersuasive. The outer border of the census enumeration district, which no one contends is the boundary referred to in the regulation and which, within the two-mile radius, coincides with the city limits of Portage, has no relevance to population density; census enumeration districts often include large tracts where no one at all lives, as is the case here.5 *533The regulation cannot reasonably be interpreted to mean that other small residential communities near the proposed site are to be deemed within the densely populated center even though they are separated from it by substantial distances.
The subjects discussed in Part IY of the opinion are not set forth as grounds for the court’s decision, and I shall not, therefore, lengthen this dissent by discussing them in detail. It is enough to say that the location of this plant in relation to others in the area is a matter peculiarly within the province of the Commission; that we are in no position to judge whether the agency’s rulings on other license applications on entirely different records of which we have little knowledge are consistent with its ruling here; that the agency carefully con-' sidered the environmental impact on the National Lakeshore and its findings on that subject find ample support in the evidence; that Congress did not give the Assistant Secretary of the Interior a veto power over the Commission’s decision of issues entrusted to it by Congress for decision; and that we cannot reverse the Commission for applying its regulations instead of proposals and views put forward by some of its staff but not adopted by the Commission. In this part of the opinion, as in Part III, the court has invaded the function of the Commission.
. The Energy Reorganization Act of 1974 (88 Stat. 1233) transferred “all the licensing and related regulatory functions of the Atomic .Energy Commission,” including the functions of the Atomic Safety and Licensing Board Panel and the Atomic Safety and Licensing Appeal Board, to the newly-created Nuclear Regulatory Commission. (§§ 201(f) and (g) of the Act.) Despite what is said in the introductory portion of Part IV of the majority opinion, I take it there can be no dispute that the standards of review are the same for the orders of the old and the new commissions.
. Petitioners appear not to have preserved for ■ judicial review the sufficiency of the evidence on this point. They did not challenge the sufficiency of this evidence in their 78-page brief before the Appeal Board, though they later attempted to do so in a supplemental brief, which the Board refused to allow them to file. Cf. United States v. Tucker Truck Lines, 344 U.S. 33, 37, 73 S.Ct. 67, 97 L.Ed. 54 (1952).
. It is noteworthy that the parties opposing the agency before the court, while criticizing the map as too old and too vague, do not contend that the record shows, or that there are in fact, densely populated areas within the two-mile zone. The reason for this is suggested by the 1970 Census, of which we may take judicial notice. About 95 per cent of the land within the City of Portage that is also within the two-mile radius appears to be north of the Penn Central (formerly New York Central) railroad track. As we read the 1970 Census, it shows that the census taker found no residents in that area. The remaining 5 per cent of the land which is within both Portage and the two-mile radius represents what appears to be less than 5 per cent of the land area in the next census block to the south of the railroad. A total of 81 residents were found in that entire census block. See U.S. Bureau of the Census, Census of Housing: 1970 Block Statistics Final Report HC(3)-68 Chicago Ill.-Northwestern Indiana Urbanized Area (1972).
As to the argument that the pink areas on the map have no probative value, we note testimony in the record that “the definition of pink tinted areas here by the Geological Survey is that population density which is too sense [obviously a misprision for ‘dense’] to indicate the houses by themselves,” because if individual houses were indicated “theré would be no opportunity to indicate the other significant landmarks.” This testimony is consistent with U.S. Dep’t of Interior, Geological Survey, Topographic Map Information and Symbols (1972), which explains that “red tint” is used to represent urban areas, in which only landmark buildings are shown. The nearest periphery of the nearest pink area was approximately four and one-half miles away from the proposed reactor site.
. The evidence shows that the bulk of the land within the two-mile radius and not within the National Lakeshore is zoned for industrial use and, to the extent it is developed, is devoted primarily to industrial uses, and that all of the small part of Portage within the two-mile radius is in the industrial category, which explains why few, if any, persons live in that area. See note 3, supra.
. According to the 1970 Census there were no residents in the area about two and one-half miles long and over one mile wide running generally west from the proposed site along the shore of the lake to Burns Waterway. That area, which includes all but about 5 per cent of the part of Portage that is within the two-mile radius (see note 3, supra), is the part of Census Enumeration District 719 (to which the majority refers) nearest to the site.
Census enumeration districts are “small population areas averaging about 250 housing *533units and are defined by the Census Bureau” for use in “the collection and tabulation of population and census data.” U.S. Bureau of the Census, 1970 Census Users’ Guide pt. I, 87-88 (1970). Criteria used in determining enumeration districts are “(1) the estimated population size of the ED should constitute an adequate ’ enumerator workload; and (2) the enumeration district must fall within the boundaries of certain areas for which the results are to be tabulated, i.e., tracts, cities, minor civil divisions, etc.” Id. at 88. These districts are “administrative areas” and represent “a work assignment for a single enumerator.” They “may range in size from a city block to several hundred square miles, but usually encompass from 750 to 1,500 persons.” Id. at 22.