MEMORANDUM OPINION

WILL, District Judge.

The plaintiff seeks a permanent injunction against the construction of two bridges over the Illinois River at a point between Valley City and Florence, Illinois, known as Napoleon Hollow. The proposed bridges would be part of a four-lane expressway designated FAP 408 (Federal Aid Primary Route), which would extend from Decatur, Illinois, to the Mississippi River, a distance of approximately 145 miles. On June 26, 1980, after a hearing and after finding that substantial questions existed as to the adequacy of the Environmental Impact Statement, the application of section 4(f) of the Department of Transportation Act, 49 U.S.C. § 1653(f) and section 106 of the National Historic Preservation Act of 1966, 16 U.S.C. § 470f, and the propriety of using funds appropriated for the Highway Bridge Replacement and Rehabilitation Program established pursuant to 23 U.S.C. § 144(a), we entered a preliminary injunction against construction of the bridges and against construction of FAP 408 beyond the point it had reached by that date, approximately four miles east of the proposed site for the bridges.

By agreement, rather than having the case remanded to the Department of Transportation for further administrative proceedings, the parties then compiled and, with the Court's permission, supplemented the administrative record. The issues raised in the plaintiff's amended complaint have been briefed on the basis of that record, which now consists of more than 1200 pages plus 29 appendices, some of which are multi-volume and some of which exceed one hundred pages in length. As might be expected, the record is archival in its scope and length and is occasionally esoteric, containing detailed descriptions and evaluations of the biological, environmental, archeological, and historic characteristics and significance of the area and the extent to which they would be affected by construction of the bridges as proposed. Nevertheless, the essential facts can be stated fairly briefly.

I. FACTUAL BACKGROUND

The Illinois River Valley

In the area between Valley City and Florence, the west bank of the Illinois River is bounded by rocky bluffs overburdened with loess soils capable of sustaining trees and vegetation. These bluffs are 100 to 150 feet high, but at several points are broken by wooded ravines of varying sizes and configurations. See A.R.App. 4 at 28; A.R.App. 5, Vol. I at 6-7. In a few locations quarries have been established. A.R.App. 4 at 28. Because of the difficulty which the bluffs present for construction of two bridges and a segment of highway adjacent to the western end of the bridges, choosing a natural or a man-made gap in the bluffs for the location of the bridges (and the highway's western approach to them) would facilitate construction. See A.R.App. 4 at 17.

The Wade Property

When this action was filed, the proposed bridge and highway plan contemplated that the highway to and from the west end of the bridges would be built on a portion of the property owned by Sam Wade and his sister Juliet Wade. The Wade property consists of 190 acres of farm land, pasture, and woodland approximately one mile west of the Illinois River. That property was settled by the plaintiff's ancestors during the 1830's, and contains a house, two "basement barns," and several smaller outbuildings. The house is limestone construction, dates from the late 1830's or early 1840's, and is significant as one of the earliest settlements in Pike County and as an example of an early nineteenth century stone farmhouse. A.R. at 455; A.R.App. 20 at 8-9. On July 31, 1978, the Wade property was determined by the Keeper of the National Register of Historic Places to be eligible for inclusion in the National Register of Historic Places. See preface to A.R.App. 20. Since such a listing gave the property the protections afforded by section 4(f) of the Department of Transportation Act, 49 U.S.C. § 1653(f), and section 106 of the National Historic Preservation Act of 1966, 16 U.S.C. § 470f as well as the regulations of the Department of Transportation ( 23 C.F.R. § 771.135), the Illinois Department of Transportation (IDOT) has, since this action was filed, revised the highway plan to avoid construction of any part of the highway on the Wade property. The Pike County Conservation Area and Napoleon Hollow

Between the Wade property and the west bank of the Illinois River lies the Pike County Conservation Area ("the PCCA"). The PCCA consists of approximately 862 square acres, is owned by the State of Illinois, and is operated by the Illinois Department of Conservation. The PCCA was not purchased by the Illinois Department of Conservation until 1970, which was after the federal authorities had already approved the proposed corridor on August 18, 1969. The PCCA is also subject to the requirements of section 4(f) and the related statutes and regulations. Both the earlier proposed highway alignment and the current alignment as modified to avoid the Wade property call for construction of the highway through a ravine, known as "Napoleon Hollow," in the middle of the PCCA, thereby destroying a portion of the PCCA and cutting the remainder into separate and smaller areas.

Defendants initially challenged the plaintiff's standing to bring this lawsuit and we denied that challenge in a memorandum opinion dated January 28, 1982.

Since the filing of this action, several "biota" studies of the flora and fauna of Napoleon Hollow and of other locations identified as possible highway crossings of the Illinois River have been undertaken by various authorities, with particular reference to the extent that certain endangered species or threatened species inhabit or use Napoleon Hollow or any of the other locations suggested as alternative bridge and highway sites. Most prominent among the species whose use of Napoleon Hollow (and its environs) has been watched and studied is the American Bald Eagle.

As IDOT itself notes, "there is strong evidence of continued, uninterrupted use of the Pike County Conservation Area by bald eagles as a wintering site." A.R.App. 7 at 1-2. Such evidence consists in part of the observations of recent study groups, the observations of Robert Smith, a tenant farmer occupying part of the PCCA, A.R. at 927, and reports from 50 to 70 years ago of eagles nesting in the area. A.R.App. 19 at 9; A.R.App. 7 at 4-6.

The Bald Eagles

Bald eagles generally inhabit and migrate through two areas of North America, the Pacific Coast up into Alaska, and the valleys formed by the Mississippi and Illinois Rivers. A.R.App. 19 at 2. During the winter months, Illinois is host to approximately 14% of the North American bald eagle population: More bald eagles winter in Illinois than in any of the other lower 48 states. A.R.App. at 1.

A group or a "community" of eagles will return year after year to a particular wintering area as long as a sufficient supply of food and sufficient protection from the elements remain available. A.R.App. 19 at 2-3. These "roosting" sites are used both by particular communities of eagles as their winter habitation and by other eagles en route to or from other winter roosting sites if severe weather occurs during migration. Id. at 3. There is some evidence that maintenance of the population of bald eagles at its current level requires a chain of roosts along the rivers which serve the eagles as migration routes. Id. at 5.

Nocturnal roosts and the features thought to be important for a suitable wintering site are discussed in several places in the administrative record. See A.R. at 930, et seq.; A.R.App. 7 at 4 and 7; A.R.App. 13 at 35-36; see also A.R.App. 19 at App. II. Trees used as nocturnal roosts along the Illinois River are on "east-facing slopes that are protected from the prevailing westerly winds." A.R. at 928. Roosts are usually large trees, and include oaks, cottonwoods, sycamores, and silver maples. Id. A severe weather nocturnal roost requires greater protection from the wind than does a fair weather nocturnal roost. A.R.App. 7 at 4. Although a roost tree must be sheltered from the wind, it must also have an "open branch system" which faces the "flyway of the ravine." A.R.App. 13 at 35. In other words, the shelter must be easily accessible from the flyway or common flight pattern.

Also important are the physical characteristics of the ravine, particularly its depth, the steepness of its slopes, and its general direction. Given the prevailing westerly winds in the area, a general direction of NE-SW or E-W is preferred. Id. The nature and amount of human activity in the area are also relevant to a ravine's suitability as a roosting site. A.R.App. 7 at 7.

The eagles' tolerance of nearby human activity is a matter of some debate. Automobile traffic is considered less disturbing to eagles than is the presence of pedestrians, particularly if cars do not slow down or stop so that their occupants can look for eagles. A.R.App. 7 at 9-10; A.R.App. 19 at 12-13. There are instances of known eagle activity near other bridges and highways in Illinois.

At the major eagle wintering area at Lock and Dam 19 on the Mississippi River, one feeding site is as close as 400 feet, several feeding, eating and loafing sites are within 1500 feet, and many others are within a mile of the Keokuk Bridge, which carries an estimated average of 6600 vehicles per day. Moreover many of these same eagles use Mink Island as a mild weather night roost. Mink Island is within one-fourth mile of Warsaw Road, which is used by an estimated average of 2300 vehicles per day. The Oak Valley Eagle Refuge, a night roosting area near Lock and Dam 14 on the Mississippi, is within one-eighth to three-eighths of a mile from State Route 84 which has an estimated average of 5500 vehicles per day.

A.R.App. 7 at 10.

However, the noise which accompanies major construction might be enough to drive away the eagles if construction occurs during the period when the eagles are present. Apparently, a similar disturbance caused eagles to leave another location

in 1966 when loggers were cutting trees near the mouth of Eagle Valley in southwestern Wisconsin. Even though the loggers were 1/2 mile from the roost and were not visible from the roost itself the eagles would not use the roost because of the constant noise of the chain saws and tractors operating during the day. Within one week after the loggers had begun their work, the number of eagles in the winter use area had dropped to just one or two. Even during the following winter few eagles used the roost or the entire use area as compared to the previous six winters.

A.R.App. 19 at 12. The defendants have agreed to limit construction to the period from March 1 to November 15 to avoid most of the period when eagles inhabit the area. See A.R.App. 7 at 8.

The various discussions of eagles' tolerance of human activity suggest that whether the eagles would stay in this area if the bridge and highway are built is likely to depend on how close the traffic would be to the eagles' remaining roosting, feeding and loafing sites, and upon the visibility of those sites from the highway. With regard at least to the location of roosting sites, such a determination appears to be very difficult to make in advance; as indicated below, the various reports disagree as to how many nocturnal roosts are or are likely to be within Napoleon Hollow. Additionally, these reports disagree as to the locations of the sites which they identify as nocturnal roosts, whether those sites are within the path of the highway, and, if not within that path, the distance between the roosts and the highway.

Archeological Sites from the Middle Woodland Period

On November 7, 1977, archeological sites known as the Napoleon Hollow Village and associated Russell Mound Group were found to be eligible for the National Register of Historic Places. A.R. at 720-18. These archeological sites date from what is known as the "Middle Woodland" period, circa 100 B.C. to 450 A.D. The village site covers 30 acres between the base of the bluffs and the west bank of the Illinois River and extends into the mouth of Napoleon Hollow. A.R. at 492; A.R.App. 16 at 4. The Russell Mound Group overlooks the village site and consists of 26 burial mounds and two possible burial knolls. A.R. at 492; A.R.App. 16 at 4. The proposed highway alignment would destroy much of the village site and many of the burial mounds. See A.R. at 489-88, A.R.App. 16 at 15.

The Russell Mound Group first came to the attention of archeologists in 1882 when Judge John G. Henderson of Winchester, Illinois, described one of the mounds (Naples Mound No. 8) and artifacts found in it in the Smithsonian Institution Annual Report for 1882. A.R. at 492. However, the village site and most of the other burial mounds were not discovered until the 1970's, during surveys of the region by the Northwestern University Archeological Program (the N.U.A.P.). Id. at 489. Although the N.U.A.P. has performed test excavations at the village midden and at some of the burial mounds, id. at 488, very little excavation and data recovery have yet occurred. Dr. Jane Buikstra of the N.U.A.P. estimates that one group of six burial mounds alone (the Elizabeth mounds) contains about 180 human skeletons. A.R.App. 16 at 6.

Archeologists hypothesize the existence of a trade network stretching at least from Wyoming (obsidian) through the Great Lakes region (copper) to the Gulf Coast of Florida (marine products) during the Middle Woodland period. Id. at 9. One theory is that certain sites located along major river arteries in the Midwest served as trade centers. Id. Although six Middle Woodland mortuary sites are known to have been located in the lower Illinois River valley, the Russell Mound group and the Napoleon Hollow village site comprise the only one of these possible trade centers known to archeologists that has not yet suffered major destruction at the hands of modern man. Id. The Mormons and This Area

The general environs of the PCCA and the Wade property, and in particular one of the burial mounds in the Russell Mound Group, also have some significance in the history and theology of the Church of Jesus Christ of the Latter Day Saints, better known as the Mormon Church. In 1834, the Mormon prophet Joseph Smith led about 150 of his followers from Ohio to Missouri. Phillips' Ferry, where Smith and his group crossed the Illinois River, was located near Valley City. The Mormons continued along a road, now known as "Church Hollow Road," part of which forms part of the northern boundary of the Wade property. Of the Mormons' Ohio to Missouri route, known as "the Zion Trail," the only part which remains in a condition substantially similar to its condition in 1834 is a segment of Church Hollow Road. A.R. at 751-50, 745-42, 738, 668-67. The current proposed alignment for FAP 408 would cut through this portion of Church Hollow Road, and force a "rerouting" or elimination of that part of the road which borders the Wade farm. See A.R.App. 20 at 12-13 and compare maps on pages 5, 7 and 15.

Apparently after crossing the Illinois River and before proceeding west along Church Hollow Road, Joseph Smith and some of his followers happened upon one of the burial mounds mentioned above (Naples Mound No. 8, the same mound which later was the subject of Judge Henderson's study). Smith wrote in his book, History of the Church:

Our enemies had threatened that we should not cross the Illinois river, but on Monday the 2nd we were ferried over without any difficulty. The ferryman counted, and declared there were five hundred of us, yet our true number was only about one hundred and fifty. Our company had been increased since our departure from Kirtland by volunteers from different branches of the Church through which we had passed. We encamped on the bank of the river until Tuesday the 3rd.

During our travels we visited several of the mounds which had been thrown up by the ancient inhabitants of this country The — Nephites, Lamanites, etc., and Finding this morning I went up on a high mound, of near the river, accompanied by the Zelph brethren. From this mound we could overlook the tops of trees and view the prairie on each side of the river as far as our vision could extend, and the scenery was truly delightful.

On the top of the mound were stones which presented the appearance of three altars having been erected one above the other, according to the ancient order: and the remains of bones were strewn over the surface of the ground. The brethren procured a shovel and a hoe, and removing the earth to the depth of about one foot, discovered the skeleton of a man, almost entire, and between his ribs the stone point of a Lamanitish arrow, which evidently produced his death. Elder Burr Riggs retained the arrow. The contemplation of the scenery around us produced peculiar sensations in our bosoms: and subsequently the visions of the past being opened to my understanding by the Spirit of the Almighty, I discovered that the person whose skeleton was before us was a white Lamanite, a large, thick-set man, and a man of God. His name was Zelph. He was a warrior and chieftain under the great prophet Onandagus. * * *

On July 5, 1979, various other archeological sites known as the Burnt Hill Multiple Resource Area (Burnt Hill or the Burnt Hill MRA) were also determined to be eligible for inclusion in the National Register of Historic Places. A.R. at 839-25. The Burnt Hill MRA covers a band approximately 23 miles long by 1.3 miles wide, between the town of Barry, Illinois, and the bluffs west of the Illinois River. A.R.App. 17 at 2 and 6. Located within that area are 68 prehistoric sites. Some are thought to be small camp sites; others may be ancient villages. Id. at 3. In contrast to the Napoleon Hollow village site and the Russell Mound Group, which are thought to date from the Middle Woodland period (100 B.C. to 450 A.D.), the Burnt Hill sites are thought to represent the gamut of prehistoric North American time periods, with the greatest number of sites being from the "Late Archaic" period, 2500 to 1000 B.C. Id. at 4. The original alignment would have taken approximately 1.3 square miles from the Burnt Hill MRA and would have destroyed or endangered 47 of its prehistoric sites. Id. at 6 and 14. The current alignment, which by-passes the Wade farm (and which is labelled as "Alternate 3" in Appendix 17) apparently would still take 15 of the Burnt Hill sites.

The Napoleon Hollow village and associated Russell Mound Group, and the Burnt Hill MRA, by their eligibility for inclusion in the National Register of Historic Places are subject to the same statutory and regulatory protections to which we referred previously in connection with the Wade property and the PCCA. IDOT proposes to mitigate in part the harm to these archeological sites through "preservation or conservation where prudent and feasible" and, alternatively, through data recovery. A.R.App. 17 at 14; see A.R.App. 16 at 14. With regard at least to the Burnt Hill MPA, however, IDOT conceded, though before modification of the alignment to avoid the Wade farm, that "[t]he complete excavation of all known significant cultural resources . . . which will be adversely impacted may be impossible due to the constraints imposed by limits of time and available funding." A.R.App. 17 at 14; see also A.R.App. 16 at 25. Apparently, even after modification, this will still be true.

The Bridges

The highway alignment calls for two separate bridges, one to carry westbound traffic and the other to carry eastbound traffic. Each bridge will have roadways forty feet wide, accommodating two twelve foot lanes with a ten foot shoulder on the right and a six foot shoulder on the left. A.R.App. 16 at 1-2.

As originally planned, the separate spans were to be approximately parallel and from 600 to 700 feet apart. As IDOT stated in a so-called "mitigation report" prepared in December of 1978, on the west side of the river, "[t]he split alignment . . . locates the freeway along the gently sloping gradient of Napoleon Hollow, holding unsightly rock cuts to a minimum. A rest stop located on the north side of the freeway has been incorporated into the conservation area with an overlook which will provide users with a scenic view of the Illinois River." A.R.App. 16 at 1.

The construction plan was later modified to narrow the median strip between the eastbound and westbound segments of the highway which are to be built in Napoleon Hollow, such that the proposed right-of-way width will be approximately three hundred feet. See A.R.App. 17 at 1. The modified plan also eliminated the rest area which would have been constructed within Napoleon Hollow and the PCCA. Id. IDOT has stated that it "anticipate[s] that the bridges will be erected by segments" in such a way as to reduce the areas on the river bank needed for actual construction of bridge materials and thus reducing the harm to areas near the shoreline. A.R.App. 16 at 2. IDOT proposes to buy land immediately south of the PCCA and "annex" that property to the PCCA as an acre for acre replacement of the portion of Napoleon Hollow that would be taken by the proposed alignment for FAP 408. A.R.App. 4 at 79-80. As indicated, the defendants have also agreed to limit construction to the period from March 1 to November 15 in order to avoid most of the period when eagles inhabit the area. See A.R.App. 7 at 8. As further mitigation of the harm to the PCCA, IDOT has stated that the segment of the freeway which would cut through and divide Napoleon Hollow will contain "structures [evidently tunnels] . . . large enough for game and man to pass through." Id. at 80.

U.S. 36-54 and Its Bridge at Florence, Illinois

Approximately 4 and 1/2 miles south of Napoleon Hollow, U.S. 36-54, designated FAP 757, crosses the Illinois River just north of Florence, Illinois. FAP 757 is a two-lane open access highway, and roughly parallels the proposed alignment for FAP 408. According to IDOT, U.S. 36-54 is inadequate for its present volume of traffic and has a high accident rate. A.R.App. 20 at 1; A.R.App. 4 at 63. The volume of traffic in this area is expected to increase over time. A.R.App. 20 at 1.

The prevailing view of state and federal officials is that construction of a limited access freeway in this area of Illinois will enhance its economic development, in particular that of Quincy, Illinois, a city of approximately 45,000 people. A.R. at 1058. See A.R.App. 17 at 1 and 17. FAP 408 would also serve as a link in the freeway network between Indianapolis and Kansas City.

IDOT noted in 1978 that the bridge at Florence was unsafe and in need of immediate rehabilitation. A.R.App. 16 at 3. That bridge was built in 1930 and must be raised in order for some river boats to pass beneath it. Plaintiff's Exhibit 14 at I-2 and I-35. In 1980, the state rehabilitated the Florence bridge, thereby adding approximately 20 years to its useful life. See Plaintiff's Exhibits 15 and 17. The record appears not to indicate the cost of this rehabilitation, but that cost was estimated to be "a large sum of money (probably a few million dollars or more)." Plaintiff's Exhibit 14 at I-35; see generally Plaintiff's Exhibit 15 and 31. No federal funds were spent on this bridge rehabilitation project. Defendants-Intervenors' Brief at 36 n. 15.

Assuming that FAP 408 is built along the alignment proposed for it, with two bridges over the Illinois River at Napoleon Hollow, the state has no plans to abandon, demolish, or close the Florence bridge. The bridges planned for Napoleon Hollow would not join and carry over the Illinois River the existing FAP 757 (U.S. 36-54). On the contrary, U.S. 36-54 would continue to carry traffic, though it is anticipated that the volume of its traffic might decrease substantially. See A.R.App. 2 at 32; A.R. App. 5, Vol. I at 62; Defendants-Intervenors' Brief at 36.

Alternatives to the Existing Proposed Alignment for FAP 408

During the 1960's, the state held public hearings and commissioned or conducted various feasibility studies and location studies regarding construction of a limited access four-lane freeway in west central Illinois. Three possible general routes were identified:

Route A — from Decatur to Springfield to Jacksonville to the Mississippi River along U.S. 36-54;

Route B — along Route A, above, as far as Jacksonville (about 20 miles east of the Illinois River) but then following Illinois Highway 104 (approximately parallel to and 15 miles north of U.S. 36-54) to the Mississippi River;

Route C — from Springfield to Quincy along Illinois Highway 125 and U.S. 24 (generally 10 to 15 miles north of Route B).

According to IDOT, the decision that Route A was the preferred alignment for FAP 408 was based upon population demand, construction costs, reduction of traffic on existing major roads, cost benefit ratio, and service to a greater number of work, shopping, and business trips. Defendants-Intervenors' Brief at 4; see A.R.App. 2 at 28-35.

In a letter dated August 18, 1969, the federal government (via a Division Engineer in the Bureau of Public Roads, the predecessor of the Federal Highway Administration) concurred in the selection of Route A to the extent of determining that "additional corridor hearings will not be required" but indicating that it remained "understood that design studies and hearings must be held in order to determine the final design of the supplemental freeways." A.R. at 10; see A.R.App. 4 at 1. Thus, federal and state authorities initially committed themselves to a freeway route which would track existing U.S. 36-54 at least to the extent of being in the general vicinity of U.S. 36-54.

While the Location Study addressed possible alignments within Route A with some specificity, see A.R.App. 2 at 9 and at 39, and recommended that the highway be constructed North of U.S. 36-54 and cross the Illinois River between Valley City and Florence, see A.R.App. 2, Map 18, the need for further study of specific alignments was implicit in the Location Study and is explicit in the tentative federal approval of Route A given on August 18, 1969.

The state then began a Design Location Study (A.R.App. 5, Vols. I, II and III) limited to the Jacksonville to Barry, Illinois, segment — which requires a crossing of the Illinois River — of the proposed freeway, and, in conjunction with that study, prepared a combined Environmental Impact Statement/4(f) Determination (the EIS/4(f) Statement) (A.R.App.4). These studies considered various alternatives to the alignment which would cross the Illinois River at Napoleon Hollow. The alternatives considered were the following:

(1) developing existing two-lane U.S. 36-54 into either a limited access four-lane freeway or a full access four-lane freeway;

(2) keeping the current proposed alignment except in the area immediately to each side of the Illinois River, and crossing the Illinois River at

(a) Florence, at or near the site of the existing U.S. 36-54 bridge;

(b) the Blue Creek Basin;

(c) the Flint Creek Basin, just South of Valley City (where the Norfolk and Western Railroad crosses the river);

(d) a ravine about 0.4 miles North of Napoleon Hollow but still within the PCCA:

(3) not building the freeway at all.

Not to build the freeway was deemed imprudent given the volume of traffic and the high accident rate on U.S. 36-54, and given the perception that the lack of a four-lane freeway in west central Illinois was economically disadvantageous. See A.R.App. 4 at 61-62, 65. The conversion of U.S. 36-54 into a four-lane freeway was also rejected as imprudent. According to both the EIS/4(f) Statement and the Design Location Study, this alternative would require extensive severance of valuable farm property, would require relocation of homes and commercial buildings, and would leave much of the existing U.S. 36-54 as frontage abutting other residences and commercial buildings which did not have to be relocated. A.R.App. 4 at 15 and 61; A.R.App. 5, Vol. I at 6-7. Expansion of the existing U.S. 36-54 would also necessarily present at least some of the same problems presented by the alternative of crossing the Illinois River at Florence.

Two reasons were given for rejecting a crossing of the Illinois River at Florence. First, the incorporation of the existing "lift" bridge (of U.S. 36-54) into a freeway (which requires bridges high enough to allow river boats to pass under them without interruption to highway traffic) was deemed not feasible. Second, the amount of rock cuts (in the bluffs) required for a crossing at Florence was deemed imprudent compared to the amount of rock cuts required by a crossing at Napoleon Hollow. A.R.App. 5, Vol. I at 9.

The magnitude of rock cuts at Blue Creek Basin was a factor in the determination that a crossing at that location was also imprudent. A.R.App. 4 at 23. Additionally, the Blue Creek Basin alignment would have required either "destruction or alteration" of the dam built at the western end of the Blue Creek Valley to form Lake Pittsfield. Id.

The EIS/4(f) Statement noted that "the Flint Creek Valley is wide and somewhat long," but rejected that valley as imprudent for the location of the FAP 408 bridge:

Unfortunately, the southern edge of the valley is utilized for the existing alignment of the Norfolk and Western Railroad which was realigned for a new crossing of the Illinois River in 1958. Any crossing of the Illinois River through Flint Creek Valley would necessitate a costly relocation of the Norfolk and Western Railroad amounting to over one half mile in addition to introducing extra curvature for the railroad. It would also result in additional damage to the ecology of the area, by introducing extra fill for the freeway and the railroad. An additional amount of adverse travel to the motoring public would also result from such a crossing. This additional distance is almost one mile longer than a crossing through Napoleon Hollow.

A.R.App. 4 at 23.

Finally, the EIS/4(f) Statement rejected, apparently as not feasible, a crossing through one of the ravines north of Napoleon Hollow but still within the PCCA:

South of Flint Creek there are two very short abrupt draws penetrating the bluff in addition to the confines of the valley of Napoleon Hollow. The scale of the valley of Napoleon Hollow with its gentle gradient and width is compatible to the geometrics of this proposed facility while an alignment through either one of the short draws would penetrate an extremely small opening much shorter and narrower than required for a freeway facility of this type. Any other location through the bluffs would be unfeasible for reasons of cost and environmental damage.

A.R.App. 4 at 23.

Thus, the state concluded that no "feasible and prudent alternative" existed to the use of Napoleon Hollow for a freeway bridge or bridges over the Illinois River. On May 15, 1972, the U.S. Department of the Interior concurred in this conclusion, as it appeared in the Draft EIS/4(f) Statement. A.R.App. 4 at 125. On September 18, 1972, the Chief of the Regional FHWA approved the Final EIS/4(f) Statement, embodying the same conclusion, A.R. at 107-05, and on October 25, 1972, the Acting Federal Highway Administrator sent the EIS/4(f) Statement to the Secretary of the U.S. Department of Transportation for approval. A.R. at 110-08. Eventually, after considerable internal departmental questioning as to its adequacy, on April 30, 1974, the Secretary approved the EIS/4(f) Statement. A.R. at 134, 126-23.

Rock Cuts and Costs

As indicated, the principal reason given for rejection of alternative locations (Florence, Valley City/Flint Creek, and Blue Creek Basin) as imprudent was the magnitude of the rock cuts compared to the rock cuts which would be required for a crossing at Napoleon Hollow, and the additional costs associated with the comparatively greater amounts of rock cuts. Also mentioned were the additional costs which would be incurred in moving the railroad tracks (Valley City) or in displacing a comparatively greater number of homes and businesses along U.S. 36-54 than would be displaced by constructing two bridges at Napoleon Hollow. A particularly distressing inadequacy of this record, despite all its volumes, is the lack of any meaningful, objective, and quantified comparison of either the amounts of the rock cuts or the financial costs of placing the FAP 408 bridges at the various proposed locations.

Some rock cutting will apparently be required at whatever valley or hollow is chosen as the location for the bridges and their western approach. The record indicates that the original alignment through Napoleon Hollow and the Wade farm would have required the cutting of an estimated 80,000 cubic yards of rock. A.R.App. 20 at 12. The current alignment (developed in 1982) cuts a more narrow swath through Napoleon Hollow, eliminates the rest area originally included in the design, and avoids the Wade farm by going north of it, but requires double the estimated amount of rock cuts, 160,000 cubic yards. Id. Interestingly, the current detour around the Wade farm was chosen over another possible alternate — bisecting the Wade farm and taking 9.06 acres but sparing the buildings — which would have required cutting only 79,000 cubic yards of rock, 1000 cubic yards fewer than the amount required by the original alignment. Id. at 17. The record contains several references to "the magnitude of the rock cuts" which would be required for a crossing at Flint Creek/Valley City, Blue Creek Basin, or Florence, but none of those references estimates numerical amounts or costs. See, e.g., A.R.App. 2 at 39; A.R.App. 4 at 61; A.R.App. 5, Vol. I at 8-9. The contexts of these references imply that the "magnitude" of the rock cuts at these locations is greater than the presumably lesser amount of rock cuts at Napoleon Hollow. These references were made at times before modification of the Napoleon Hollow alignment so as to avoid the Wade farm, a modification which doubled the amount of rock cutting required for the crossing at Napoleon Hollow. In any event, there would appear necessarily to be differences between the amounts of rock cutting required at the various alternative sites. How substantial those differences currently are, however, the record falls to disclose.

The record is similarly deficient in any meaningful comparison of the costs of crossing the Illinois River at the various breaks in the bluffs in this region. A study completed in 1966 estimated the cost of a Jacksonville-to-Kinderhook segment of FAP 408 along the Napoleon Hollow alignment at $50,163,000.00. A.R.App. 2 at 46 (Table 18). An alternate segment having the same termini but tracking U.S. 36-54 and crossing the Illinois River at Florence was estimated as costing $53,105,000.00, id., somewhat less than 7% greater than the cost of the Napoleon Hollow alignment. The same table listed "annual costs" for the two alignments as $2,474,300.00 for the Napoleon Hollow route and $2,602,000.00 for the Florence route, thus estimating Florence as about 5% more expensive than Napoleon Hollow. Inflation having been persistent for the past 17 years, these figures have long since lost any reliability. Surprisingly, they are the only such figures which the record contains.

Consistent with the above, data as to the numbers and kinds of buildings which would be taken by the various alignments which were considered exist in the 1966 study (A.R.App. 2 at 9, Figure 1) and in the Design Location Study completed in 1971 (A.R.App. 5, Vol. I at 107, 110 and 116), but have not been revised. The record does not reveal any meaningful comparison of the costs of the compensation required for the property which might be taken by the various alternate alignments. Nor does the record contain any estimate of the cost of moving the Flint Creek railroad tracks so that both the freeway and the railroad tracks would fit in the Flint Creek Basin or what it would cost to bridge, overpass or underpass the railroad tracks.

The 1972 Design Location Study (App.5) contains some comparison of the rock cuts and costs of constructing bridges over the Illinois River at Napoleon Hollow and at the ravine 0.4 mile north of Napoleon Hollow. According to this study, a highway through this smaller ravine would require

up to 600 feet wide cuts in loessial soils and rock averaging 30 to 60 feet in depth. The amount of disturbance caused by this gouging of the terrain is a negative environmental factor to the area involved and would involve extensive procedures to control erosion during and after construction.

A.R.App. 5, Vol. I at 72. The Design Location Study does not estimate the width or depth of soil and rock cuts required by the Napoleon Hollow alignment or quantify, in cubic measurements, the excavations required by these two alignments. That study merely states that more rock cuts and soil excavation would be required at the smaller ravine than at Napoleon Hollow. Id.

The Design Location Study also notes that bridges at Napoleon Hollow would not have to be quite as long as bridges at the ravine 0.4 mile north of Napoleon Hollow. Id. at 139. This study estimated the cost of two Napoleon Hollow bridges at $16,291,720 and the cost of bridges at the smaller ravine at $17,967,000, a difference of $1,685,880. It is not entirely clear from the report whether this difference is due solely to the increased length of bridges at the ravine north of Napoleon Hollow, or is also based upon the expectation that a greater amount of rock cuts would be required for construction at the smaller ravine. Nor is it clear how the later elimination of the rest area from the proposal for FAP 408 would, if at all, affect the comparative amounts of rock cuts and costs at these two alignments. The Design Location Study also estimates that the total cost of the Napoleon Hollow alignment is less than the total cost of an alignment through the smaller ravine north of Napoleon Hollow. However, the report gives conflicting statements as to that amount. See A.R.App. 5, Vol. I at 140 (a difference of $1,923,000) and at 152 (a difference of $832,000).

The Design Location Study does not discuss the difficulties or costs of construction at Florence, Valley City, or the Blue Creek Basin, except to the extent of noting that crossings at Florence and at Blue Creek had been rejected as "not feasible," the former "because the existing lift bridge could not be utilized and because of the tremendous amount of rock cuts," the latter "because of the magnitude of bed rock overburden." Id. at 9. The Design Location Study was limited to a "corridor width of approximately one mile," id. at 10, stretching from Jacksonville to Barry. Although this study discusses the difficulties and costs of several "alternates," it discusses only one "alternate" crossing of the Illinois River (the ravine 0.4 mile north of Napoleon Hollow). Other "alternates" relate to various segments of the Jacksonville to Barry route, and do not affect directly the choice of where to construct bridges over the Illinois River. The "Biota" Studies

After the filing of this action, the parties and the Court agreed to supplement the administrative record with various "biota" studies, some of which were prepared by the defendants' experts or under their auspices, and others of which were prepared by the plaintiff's experts. Many had been written before the filing of this lawsuit. We also appointed, with the parties' permission, as our own independent expert, Dr. Thomas Dunstan of Western Illinois University, to study the use by bald eagles of Napoleon Hollow and the alternative crossings at Florence and Valley City. We directed Dr. Dunstan to report to us his findings and conclusions, as well as his opinion as to the probable effect of the proposed highway upon the eagles' use of these areas. See A.R.App. 13 at 1.

The Dunstan Report

Dr. Dunstan's study (the Dunstan report) identifies four "major types of [eagles'] activities" that are generally observable, and attempts to locate which parts of the study area are significantly used for which activities. Additionally, the report notes the existence of "flight corridors" linking these areas. Id. at 11-12. The four identified types of activity are (1) "foraging" for food, (2) eating, obviously closely associated with "foraging" but often occurring after transporting a captured fish or animal to another location, (3) resting or loafing, and (4) "night roosting." See A.R.App. 13 at 9. Dr. Dunstan concluded that for foraging and resting, the eagles "used the region from Napoleon Hollow downstream past the southern tip of Big Blue Island . . . more often than other portions of the study area." Id. at 22. Dr. Dunstan found that use of the Valley City area for foraging and resting was substantial, but that use of the area near the Florence bridge was insignificant. Id.; see generally, id. at 12-21.

Dr. Dunstan was able to discover only one location, within the study area, which is definitely identifiable as a night roost. Id. at 27. That location is not within the direct path of the FAP 408 revised alignment. Id. at 27 and 52. Given the difficulty of locating a night roost even in areas where eagles can be observed during the day, however, the pinpointing of one roost cannot be taken as establishing the absence of other possible roosts in the same area.

The Dunstan report also describes what are thought to be the features of a good roosting habitat. Id. at 35-36. The report evaluates the various ravines in the study area in terms of the extent to which they have those features. Interestingly, the Dunstan report states that

Napoleon Hollow, [Ravine] # 5, is too broad and shallow and does not have good physical shape. The number of potential use trees is low and an occupied farmstead near the mouth of the hollow is a human disturbance problem.

The Dunstan report also notes the characteristics of "diurnal resting areas" and of the particular trees most likely to be used for daytime resting. Id. at 41-42. The report determines that "the bluff area just downstream of Napoleon Hollow to the end of the bluff line at Big Blue Creek provides almost all the diurnal resting sites and potential sites of the area." Id. at 42. The report concludes that the Napoleon Hollow alignment would eliminate three "potential diurnal resting trees" and also might (at least during the construction period) affect the use of a fourth. Id. at 52.

The report concludes that construction of the bridge and highway at Valley City would have little if any effect upon the eagles, id. at 48 and 51, and that the Florence alignment would have no adverse impact on the eagles. Id. at 55. Dr. Dunstan's conclusion as to the effect upon the eagles of bridges and a freeway at and through Napoleon Hollow is less definite. He concludes that construction of the revised alignment would not "directly" affect "known night roosts," but that two "associated hollows" which he identifies as having the attributes of desirable roosting locations might be adversely affected by the alignment. Id. at 52. Dr. Dunstan also concludes that the proposed route would take three "potential diurnal resting trees," and would affect the eagles' use of "the flight corridor down Napoleon Hollow towards the river, in effect, forcing the eagles to fly between two spans of traffic or abandon the flight route." Id. at 52-53.

Dr. Dunstan does not state that in his opinion the eagles would leave the area if the FAP 408 bridges and freeway are built as proposed. The effect of this alignment upon the eagles, according to Dr. Dunstan, remains unpredictable:

The construction of twin bridges and roadways would definitely change the overall "wildness" aspect for the area. This could cause an adverse impact on some unknown portion of the eagle population, but the magnitude of the impact is difficult to assess. Knowledge about this aspect of eagle ecology is not well documented. It is doubtful that there would be no impact on wildness, and it is more likely that the impact would be negative.

A.R.App. 13 at 54.

The INHS Report

Pursuant to the Endangered Species Act, 16 U.S.C. § 1531, et seq., IDOT commissioned the Illinois Natural History Survey (the INHS) to conduct a biological inventory for the purpose of identifying federally listed threatened and endangered species likely to be affected by the construction of FAP 408. See A.R.App. 8 at 1; note 3, supra. The INHS completed its draft report (Appendix 8 to the Administrative Record) in October of 1980, and completed its final report (Appendix 9 to the Administrative Record) in December of 1980.

The area studied by the INHS includes Napoleon Hollow and is bounded on the north by the Norfolk and Western railroad tracks which pass through the Flint Creek Basin, but does not include the land which would be taken by the Florence alignment. A.R.App. 8 at 2. The INHS did not consider the presence of the bald eagle in the study area because another IDOT report was to focus exclusively on the bald eagle. A.R.App. 8 at 7; A.R.App. 9 at 6.

The INHS report concluded that

[n]o plants on the federal list of endangered species are known or likely to occur in the project area. Seven species of animals on the federal list of endangered species could possibly occur in the project area. . . . Of the[se] seven . . . the two freshwater mussels (Higgens eye pearly mussel and pink mucket pearly mussel) would occur only in the Illinois River.

A.R.App. 9 at 6. The INHS report also concluded that the whooping crane "is not likely to occur in the project area except, perhaps, as an extremely rare stray." Id. at 7. The INHS determined that the peregrine falcon might use the study area during migration, but is not likely to nest within or near the area. Id. at 8; A.R.App. 8 at 9-10. The INHS report concluded that there have been no records of Indiana bats in Pike County, A.R.App. 9 at 11, but that although the gray bat can be found in Pike County, the only caves in the study area suitable for hibernation by gray bats are in the Blue Creek Canyon. Id. at 9.

The INHS report also assessed the presence in the study area of plant and animal species on the Illinois list of endangered and threatened species. See generally A.R.App. 9 at 12-67. The INHS recommended further study of the area with particular reference to its use by the Illinois chorus frog. The INHS also recommended study of the vicinity of the Florence alignment.

The WIU Report

Another report (Appendix 12) was prepared for IDOT in November of 1981 by the Department of Biological Sciences and the Institute for Environmental Management of Western Illinois University. The WIU report studied "an area approximately four miles either side of the Illinois River from one-half mile below the U.S. 36 bridge at Florence to one-half mile above the railroad bridge at Valley City." A.R.App. 12 at I-1.

The WIU report evaluated the Valley City, Napoleon Hollow, and Florence alignments in terms of the vegetation, mammals, birds, reptiles and amphibians, and aquatic invertebrates which would be affected by highway construction. The overall conclusion of the WIU report is that, in terms of environmental consequences, Napoleon Hollow is the least desirable location for a highway, although differences among the corridors are "slight" and "an impartial observer could conclude differently." Id. at C-3.

IDOT's Bald Eagle Report

The IDOT report which focuses upon the bald eagle and its use of Napoleon Hollow is Appendix 7, written some time after March of 1980. As indicated, the report notes the existence of "strong evidence of continued, uninterrupted use of the Pike County Conservation Area by bald eagles as a wintering site," A.R.App. 7 at 1-2, and describes past records and investigations of the eagles' use of the area. Id. at 4-6.

The report also discusses the qualities of nocturnal roosts, both fair weather and severe weather ones, id. at 4, and the degree of eagles' tolerance of human activity near their roosts. Id. at 10. The report concludes that "[t]he most heavily used feeding area in the vicinity of the Pike County Conservation Area appears to be around the north end of Blue Creek Island, approximately 1.5 miles south of the bridge site."

The report notes evidence that both Napoleon Hollow and the ravine immediately to the south of Napoleon Hollow contain severe weather roosts. Id. at 4-5. However, the report appears to adopt the opinion of Carl Becker, Illinois' Endangered Species Program Coordinator, that in view of the features which are desirable for a suitable severe weather roost, five other nearby ravines are superior to Napoleon Hollow (and three others are equal to it) for severe weather night roosting. Id. at 7; compare A.R.App. 7 at 3-4 with A.R. at 929-928. In this regard, Becker's opinion is similar to Dunstan's. Becker and IDOT base their determination "in part on the presence of a livestock farm and farmstead and a county road within Napoleon Hollow. Furthermore, Napoleon Hollow is a relatively open valley that does not afford as much protection from the elements as several other nearby ravines." Id.

The IDOT report concludes that it is clear that the PCCA is indeed a bald eagle wintering area, but recommends further study of the importance of the area to the eagles, stating that "the precise impacts of the proposed projects on the eagles using this area cannot be determined until such studies are completed." Id. at 12. Despite this urging of further study, however, the report continues that

it is also clear that there are adequate data on which to base a conclusion that the availability of suitable alternative habitat within the Conservation Area and nearby for feeding, diurnal loafing, fair and severe weather nocturnal roosting indicate [sic] a reasonable probability that the proposed project will not have a significant harmful effect on eagles wintering in the vicinity of the Conservation Area. It is even more clear that the proposed project would not appreciably decrease the likelihood of the survival and recovery of the American bald eagle or any distinct segment of its population.

"An Impact Analysis [of FAP 408] on Avian Species" was prepared by Pam F. Gibson and Richard S. Sandburg, and was included in the administrative record as Appendices 24(A) and (B). Gibson and Sandburg criticize the other reports for, among other things, failing to discuss the more subtle ways in which the bridge may change the eagles' environment. According to Gibson and Sandburg, Napoleon Hollow currently has a "strong spiralling updraft" which "is used repeatedly by wintering Bald Eagles." A.R.App. 24(A) at BD-3. Gibson and Sandburg imply that this updraft enhances the suitability of Napoleon Hollow as a roosting site. They state that such an updraft is not present at Valley City or Florence, and note that the effect upon the Napoleon Hollow updraft of construction of a bridge is uncertain. Id. Gibson and Sandburg also suggest that the bridge will be a hazard for the eagles because some eagles may collide with it, a possibility which apparently is enhanced by the frequency of comparatively dense fog and mist in Napoleon Hollow (vis-a-vis Florence and Valley City). Id. at BD-1 through BD-4, and at NSI-19 through NSI-21. Gibson and Sandburg also fault the other studies for failure to discuss the possible effect of construction upon food sources of various species and possible collateral effects for other species further along in the food chain. Id. at NSI-20 and 21, and at BE-15.

The Ingram Report

Another report, written by Terrence N. Ingram, Executive Director of the Eagle Valley Environmentalist, Inc., was also admitted to the administrative record. Ingram based his report upon his observations at Napoleon Hollow on six days in January and February of 1982, and upon his research, conducted over the past 20 years, into the habits of bald eagles. See A.R.App. 19 at i-iii; id. at 19.

The Ingram report discusses the importance of maintaining a chain of "wintering" sites along migration paths, the difficulty in identifying nighttime roosts, and eagles' intolerance of nearby activity; we have drawn upon his report as well as the other reports in our discussions of these topics. Ingram also notes the existence of reports from 50 to 70 years ago of eagles nesting near the Big Blue watershed (south of Napoleon Hollow), and suggests that if the population of eagles continues to expand (as it has, though slightly, over the past few years) eagles might return to the Big Blue watershed area. A.R.App. 19 at 9.

As do Gibson and Sandburg, Ingram concludes that Dunstan understates the degree to which the eagles use Napoleon Hollow. Id. at 22-24. In contrast to Dunstan's conclusion that Napoleon Hollow contains one severe weather nocturnal roost, Ingram concludes that Napoleon Hollow probably contains two such roosts and possibly a fair weather nocturnal roost as well. Id.

Ingram concludes that construction of a highway through Napoleon Hollow "will destroy the only real severe weather roost in the area." Id. at 27. He advocates keeping the highway at least 1/2 mile away from Napoleon Hollow, and recommends its rerouting to the north, because of the possibility that the Big Blue watershed might be a suitable wintering site if the eagle population expands.

Funding

The U.S. Department of Transportation and the FHWA have allocated funds for construction of the FAP 408 bridges at Napoleon Hollow from the Highway Bridge Replacement and Rehabilitation Program, 23 U.S.C. § 144 and accompanying regulations, on the theory that these bridges would "replace" the Florence bridge which carries U.S. 36-54 over the Illinois River. See A.R. at 30021. The attorney for the federal defendants has informed us that these funds remain available and earmarked for the proposed Napoleon Hollow bridges. See Transcript of Proceedings on February 24, 1983.

II. THE AMENDED COMPLAINT

The plaintiff's amended complaint now contains seven counts on which she seeks declaratory and injunctive relief. In Counts I and II, the plaintiff alleges that funding the FAP 408 bridges under the Bridge Replacement and Rehabilitation Program violates 23 U.S.C. § 144 and the regulations thereunder because the FAP 408 bridges will not be a replacement for any existing bridge. The amended complaint further alleges (Count III) that the failure of the EIS/4(f) Statement to consider all the alternative routes or the presence of endangered and threatened species of wildlife in the area chosen for the proposed bridges and highway renders that Statement deficient under the National Environmental Policy Act, 42 U.S.C. § 4321, et seq., and its regulations (Count III), the Federal-Aid Highway Act, 23 U.S.C. § 101, et seq. and its regulations (Count IV), and the Department of Transportation Act, 49 U.S.C. § 1651, et seq., and its regulations (Count VII). Count VI alleges that the defendants failed to make an adequate 4(f) statement as required by section 138 of the Federal-Aid Highway Act, 23 U.S.C. § 138.

The plaintiff also alleges that approval of the bridges and the highway route violates the Department of Transportation Act (Count IV) and the Federal-Aid Highway Act (Count V), which prohibit approval of highway projects which require the use of publicly owned land from a public park, recreation area, or wildlife or waterfowl refuge, or from any historic site if there is a feasible and prudent alternative to the taking of the protected property. 49 U.S.C. § 1651(b)(2) and 23 U.S.C. § 138.

III. THE FUNDING ISSUE

As indicated, the plaintiff first challenges the funding of the FAP 408 bridges out of funds available from the Highway Bridge Replacement and Rehabilitation Program (the Bridge Program) if the bridges are to be constructed at Napoleon Hollow. The purpose of that program is stated explicitly in the statute:

Congress hereby finds and declares it to be in the vital interest of the Nation that a highway bridge replacement and rehabilitation program be established to enable the several States to replace or rehabilitate highway bridges over waterways, other topographical barriers, other highways, or railroads when the States and the Secretary finds that a bridge is significantly important and is unsafe because of structural deficiencies, physical deterioration, or functional obsolescence.

Section 144 directs the Secretary to consult with the States regarding the inventory and classification of bridges, and in assigning priorities for "replacement or rehabilitation." Id. at (b) and (c). "Replacement" of a bridge is to involve a "comparable facility," id., and the Secretary is to give "consideration to those projects which will remove from service those highway bridges most in danger of failure." Id. at (d).

The regulations which govern administration of the Bridge Program are found at 23 C.F.R. §§ 650.401, et seq. Those regulations define "eligible projects," for which funds from the Bridge Program may be used, as:

(a) General. Deficient highway bridges on all public roads may be eligible for replacement or rehabilitation.

(b) Types of projects which are eligible. The following types of work are eligible for participation in the Highway Bridge Replacement and Rehabilitation Program (HBRRP), hereinafter known as the bridge program.

(1) Replacement. Total replacement of a structurally deficient or functionally obsolete bridge with a new facility constructed in the same general traffic corridor. A nominal amount of approach work, sufficient to connect the new facility to the existing roadway or to return the gradeline to an attainable touchdown point in accordance with good design practice is also eligible. The replacement structure must meet the current geometric, construction and structural standards required for the types and volume of projected traffic on the facility over its design life.

(2) Rehabilitation. The project requirements necessary to perform the major work required to restore the structural integrity of a bridge as well as work necessary to correct major safety defects are eligible except as noted under ineligible work. Bridges to be rehabilitated both on or off the F-A System shall, as a minimum, conform with the provisions of 23 C.F.R. Part 625, Design Standards for Federal-aid Highways, for the class of highway on which the bridge is a part.

(c) Ineligible work. Except as otherwise prescribed by the Administrator, the costs of long approach fills, causeways, connecting roadways, interchanges, ramps, and other extensive earth structures, when constructed beyond the attainable touchdown point, are not eligible under the bridge program.

23 C.F.R. § 650.405(a), (b) and (c). The regulations also establish a procedure for projects for which replacement or rehabilitation funding is sought:

(a) Consideration shall be given to projects which will remove from service highway bridges most in danger of failure.

* * * * * *

(2) Whenever a deficient bridge is replaced or its deficiency alleviated by a new bridge under the bridge program, the deficient bridge shall either be dismantled or demolished or its use limited to the type and volume of traffic the structure can safely service over its remaining life. For example, if the only deficiency of the existing structure is inadequate roadway width and the combination of the new and existing structure can be made to meet current standards for the volume of traffic the facility will carry over its design life, the existing bridge may remain in place and be incorporated into the system.

23 C.F.R. § 650.411(a) and (c)(2).

Both the statute and the regulations define "rehabilitation" as "the major work required to restore the structural integrity of a bridge as well as work necessary to correct major safety defects." 23 U.S.C. § 144(m); 23 C.F.R. § 650.403(c). Neither the statute nor the regulations, however, define the term "replacement."

The parties argue at great length over whether funding of two bridges at Napoleon Hollow under the Bridge Program will violate 23 U.S.C. § 144 and its applicable regulations. Their arguments reduce, however, to what is meant by "replacement" as that term is used in the statute and the regulations.

The plaintiff notes that no bridge has ever existed over the Illinois River at Napoleon Hollow. This is not surprising, since no highway has previously existed along the alignment chosen for FAP 408. The bridge nearest to Napoleon Hollow is at least 4 and 1/2 miles away, on U.S. 36-54 just north of Florence. U.S. 36-54 and its bridge at Florence will continue in operation after completion of FAP 408, though there is some indication that the designation "U.S. 36" may be transferred to FAP 408 once that freeway is completed. See Plaintiff's Exhibit 21 (letter dated October 19, 1979, from Governor Thompson to State Senator Demuzio: "The Route 36 marking is being relocated to the new Central Illinois Expressway as usable sections are completed.") Indeed, the Florence bridge was recently rehabilitated, extending its useful life for at least another twenty years. Nor will the proposed Napoleon Hollow bridges be constructed so as to join physically and carry over the Illinois River existing U.S. 36-54 which now physically joins and is carried over that river by the Florence bridge. From these facts, the plaintiff concludes that FAP 408 and its proposed Napoleon Hollow bridges will be a new (separate) highway with new (separate) bridges, an alternative to but not a replacement for U.S. 36-54 and its Florence bridge.

The defendants argue that the regulations and the statute permit "functional" replacement of bridges, and that the construction of FAP 408 with two bridges at Napoleon Hollow comes within the concept of functional replacement. Under such a theory, it becomes relevant that it is a reasonable expectation that FAP 408, if constructed along its proposed alignment, would siphon off a substantial amount of the traffic now carried by U.S. 36-54.

In analyzing the parties' arguments over whether the proposed Napoleon Hollow bridges may be funded out of the Bridge Program, we bear in mind certain fundamental principles of statutory construction. "[S]tatutory construction `must begin with the language of the statute itself' and `[a]bsent a clearly expressed legislative intention to the contrary, that language must ordinarily be regarded as conclusive.'" Bread Political Action Committee v. F.E.C., 455 U.S. 577, 580, 102 S.Ct. 1235, 1237, 71 L.Ed.2d 432 (1982); FBI v. Abramson, 456 U.S. 615, 635, 102 S.Ct. 2054, 2066, 72 L.Ed.2d 376 (1982) (O'Connor, J., dissenting). "In short, the plain language of . . . [the statute] controls its construction, at least in the absence of `clear evidence' . . . of a clearly expressed legislative intention to the contrary.'" Bread PAC, 455 U.S. at 581, 102 S.Ct. at 1238. The rules of construction which apply to statutes apply also to regulations. Rucker v. Wabash R.R., 418 F.2d 146, 149 (7th Cir. 1969).

As further discussion will show, nothing in the statute indicates that the word "replace" was intended to convey anything but its plain, ordinary meaning. Apparently, it was deemed so obvious and clear that no statutory definition was necessary and none was included. We take judicial notice of the plain meaning of "to replace," commonly defined as follows:

to put back again in (or into) a place; to take the place of, become a substitute for (a person or thing); to fill the place of (a person or thing) with or by a substitute; to provide or procure a substitute or equivalent in place of (a person or thing).

VIII Oxford English Dictionary 468 (1970). As this definition and others establish, "to replace" has a precise meaning. It means to put something in the place of something else. In view of the plain and ordinary meaning of "to replace," the fact that a substantial number of people might use FAP 408 rather than FAP 757 (U.S. 36-54) if both highways and their separate bridges were to be in operation at the same time would make them alternatives to each other rather than the former a replacement for the latter.

The regulations contain some language which arguably supports the defendants' position that the word "replacement" as used in those regulations is sufficiently flexible in meaning to include a concept of "functional replacement." Section 650.405(b)(1) states that "replacement" of a structurally deficient or functionally obsolete bridge shall be "with a new facility constructed in the same general traffic corridor." (Emphasis added.) The defendants argue that the more than 4 and 1/2 miles between Napoleon Hollow and Florence is narrow enough to place both locations within the same general traffic corridor.

Both sides refer to various highway manuals and to some of the studies, undertaken in connection with the planning of FAP 408, which use the terms "corridor" or "sub-corridor" to refer to bands or swaths of territory, all having varying widths. Some of these other uses of the term "corridor" (or "sub-corridor") support the defendants' broad definition of the term while others support the narrow construction which the plaintiff seeks to give it. None of these references to other uses of "corridor" are very persuasive as to what that term means in the context of these regulations, and the transfer of a semantic dispute from "replacement" to "corridor" is not very helpful.

More significant is the expectation that FAP 408, if built along the proposed alignment, would draw much of the through traffic which now travels FAP 757. That, it is anticipated, many motorists would tend to view FAP 408 and U.S. 36-54 as substitutes or alternatives for each other would incline us to accept the defendants' contention that the two routes are within the same "general traffic corridor."

In the context of the regulations, however, acceptance of the defendants' position that Napoleon Hollow and Florence are within "the same general traffic corridor" does nothing to establish that section 650.405 embodies the broad concept of "functional replacement" for which the defendants argue. Immediately after reference to construction "in the same general traffic corridor," section 650.405(b)(1) states that "[a] nominal amount of approach work, sufficient to return the gradeline to an attainable touchdown point in accordance with good design practice is also eligible [for funds from the Bridge Program]." The logical inference from this discussion of "approach work" is that to be a "replacement bridge" the new bridge must ultimately be connected physically with the highway to which the old bridge is connected and the new bridge must carry that highway over the same river (or other obstruction) which was traversed by the old bridge.

This conclusion is consistent with the language of section 650.411(c)(2) (1982), which describes one exception to the requirement that the "replaced" bridge be dismantled or demolished after the opening of the "replacement bridge":

(2) Whenever a deficient bridge is replaced or its deficiency alleviated by a new bridge under the bridge program, the deficient bridge shall either be dismantled or demolished or its use limited to the type and volume of traffic the structure can safely service over its remaining life. For example, if the only deficiency of the existing structure is inadequate roadway width and the combination of the new and existing structure can be made to meet current standards for the volume of traffic the facility will carry over its design life, the existing bridge may remain in place and be incorporated into the system.

Under the example in subsection (c)(2), the "replaced" bridge may continue to be used where the old bridge and a newly constructed bridge are incorporated into a larger system, such as a set of twin spans, one of which is the old "replaced" bridge. That, however, is not the current plan for FAP 408 which contemplates the construction of new twin spans and the continuing separate operation of the old bridge some 4 1/2 miles apart.

The state argues that the Federal Highway Administration, by changing the language of subsection (c)(2) in December of 1979 to permit "use [of the replaced bridge] limited to the type and volume of traffic the structure can safely service over its remaining life" intended the regulations to permit the kind of "functional replacement" which would encompass funding the construction of the FAP 408 bridges at Napoleon Hollow while letting the Florence bridge continue to carry traffic on existing U.S. 36-54.

Some history of the development of the regulations governing the Bridge Program is helpful to an evaluation of the state's argument. The regulations in effect before March 15, 1979 (applicable only to a "replacement" program) clearly prohibited use of a "replaced" bridge by vehicular traffic. See 23 C.F.R. § 650.406(b)(2) (1978) (eff. August 14, 1974). On March 15, 1979, new regulations were implemented in response to passage of the Surface Transportation Assistance Act of 1978, P.L. 95-599, 92 Stat. 2702 (eff. Nov. 6, 1978). Consistent with that Act, the regulations of March 1979 permitted rehabilitation as well as replacement of old bridges, and created the exception noted above, to the closing of a replaced bridge where the "replaced" bridge had been incorporated into a larger replacement bridge or was to remain as one of a pair of bridges.

Among the public comments on the regulations of March 1979 were urgings for a more flexible approach to the use of old "replaced" bridges. The FHWA summarized those comments and responded as follows:

Regulations should indicate that when a deficient bridge is replaced, the existing bridge may be left in place to service recreation vehicles such as bicycles, mopeds and snowmobiles.

Discussion: The FHWA feels that whenever a deficient bridge is replaced or its deficiency alleviated, the existing bridge under certain conditions should be allowed to remain in place if safe and practical for any purpose. Section 650.411(c)(2) has been revised to so indicate.

44 Fed. Register 72112 (December 13, 1979), reprinted in A.R. at 1131.

We cannot agree with the defendants-intervenors that section 650.411(c)(2) permits the funding of two bridges at Napoleon Hollow for FAP 408 while the Florence bridge, the stated object of "replacement" will remain open as improved and continue in its current vehicular use for at least twenty more years. Although the reference in subsection (c)(2) to "use limited to the type and volume of traffic the structure can safely service over its remaining life" is arguably broad, its sole illustrative example is quite narrow. Similarly, although the FHWA used broad language ("the existing bridge under certain conditions should be allowed to remain in place if safe and practical for any purpose") in its discussion of its revision of subsection (c)(2), the comments which the agency recognized as prompting the revision were those which were narrow in scope (maintenance of the "replaced" bridge for recreational uses). Here, the Florence bridge will continue to carry all types of vehicular traffic though presumably in lesser volume.

We note also that the FHWA, when it implemented the December 1979 revision of subsection (c)(2) explicitly stated that the revision was not a "significant" one:

Note. — The Federal Highway Administration has determined that this document does not contain a significant regulation according to the criteria established by the Department of Transportation pursuant to E.O. 12044. The amendments contained herein simply clarify existing policy and procedures.

* * * * * *

In light of the opportunity provided for public comment on the final rule, and because these amendments were addressed in the comments received, it has been determined that publication of these amendments for notice and comment could not reasonably be anticipated to result in the receipt of useful information. The FHWA has also determined that it is in the public interest to issue these amendments without a 30-day delay in effective date in order to provide immediate clarification of current policy and procedures.

44 Fed. Register at 72113 (December 13, 1979), reprinted in A.R. at 1130. In light of this disclaimer and the absence of any period for public comment following publication of the December 1979 revision, we cannot accept the argument that this revision — the only possible source of support for the position that the regulations now take an expansive view of "replacement" — would permit the use of Bridge Replacement and Rehabilitation Program funds for the construction of twin spans at Napoleon Hollow as well as the continued use of the Florence bridge.

The contention that FAP 408 and the twin spans in Napoleon Hollow can be considered a functional replacement of FAP 757, the present U.S. 36-54 and the Florence bridge, is highly doubtful. FAP 757, a two-lane open access road, is classified as a minor arterial highway and, in the Florence area, as an open access service highway, the lowest primary route classification. FAP 408 is a four-lane, median strip limited access highway classified as a freeway, the highest primary route classification. The former is intended primarily to serve local traffic, the latter primarily to serve long-distance, through traffic. It is difficult to see how FAP 408 can, by any reasonable interpretation, be considered a functional replacement for FAP 757. Nor can the two new bridges proposed for Napoleon Hollow as an integral part of FAP 408 be considered functional replacements for the Florence bridge which, as previously indicated, will continue in operation for at least twenty more years as part of FAP 757.

It seems clear that FAP 408 is a new superhighway in an area not previously served by such a freeway. FAP 757 has been and will continue to be either a minor arterial highway or, in populated areas such as Florence, an open access service highway. FAP 408 is no more a replacement of FAP 757 than the airplane is a replacement of the automobile. It follows that the twin spans which will carry FAP 408 over the Illinois River are not a replacement, functional or otherwise, for the Florence bridge.

Even if we accepted the state's argument that subsection (c)(2) incorporates a notion of functional replacement elastic enough to make the defendants' proposal for the FAP 408 bridges at Napoleon Hollow eligible for federal funding under the Highway Bridge Replacement and Rehabilitation Program, we would have to conclude that the regulations had impermissibly expanded the scope of the governing statute. Thus far, our analysis of the meaning of "replacement" has focused upon the use of that term in the regulations rather than in the underlying statute. Assuming, arguendo, that the state is correct in interpreting "replacement" as used in the regulations to include the proposed two new Napoleon Hollow bridges in the concept of "functional" replacement such an interpretation must also be consistent with the statute, for it is impermissible for administrators to attempt to extend or supplement legislation by regulation. Ernst Ernst v. Hochfelder, 425 U.S. 185, 213-14, 96 S.Ct. 1375, 1391, 47 L.Ed.2d 668 (1976), rehearing denied, 425 U.S. 986, 96 S.Ct. 2194, 48 L.Ed.2d 811.

The rulemaking power granted to an administrative agency charged with the administration of a federal statute is not the power to make law. Rather, "`it is the power to adopt regulations to carry into effect the will of Congress as expressed by the statute.'"

While the language of the regulations offers some slight support for the defendants' "functional" interpretation of "replacement," the statute itself offers no support for that position. As indicated, the statute repeatedly refers to "replace or rehabilitate" (or to forms of those words), 23 U.S.C. § 144(a), (b), (c), (d), (g), (h) and ( l), but defines only the term "rehabilitate." Id. at (m). This disjunctive linking of "replace or rehabilitate" suggests that the statute does not permit the same bridge to be the subject of both rehabilitation and replacement, at least during the period of "useful life" procured through rehabilitation. This may explain why IDOT did not seek Bridge Program funds for the rehabilitation of the Florence bridge.

That subsection (d) refers to "replacing . . . [an eligible] bridge with a comparable facility does not indicate that "replace" carries anything but its plain meaning. The final sentence of subsection (d), to the extent that it casts any light upon the meaning of "replace" in this statute, suggests that its plain meaning rather than a broader "functional" meaning is intended. That sentence states that "[i]n approving projects under this section, the Secretary shall give consideration to those projects which will remove from service those highway bridges most in danger of failure."

In other words, the Secretary is to give priority to projects which will eliminate the bridges which are most in need of replacement or repair. What is significant for this case is the assumption implicit in this final sentence of subsection (d) that such replaced bridges will be closed and removed from service, an assumption which supports the plaintiff's rather than the state's interpretation of the kind of "replacement" which may be funded under section 144.

The state's inability to overcome the presumption that "replace," as used in section 144, carries its plain meaning makes inquiry into the legislative history of the statute unnecessary. Nonetheless, the parties' briefing and our own examination of the legislative history as to what Congress intended by "replacement" confirms the conclusion that the plain meaning of that term must control in our review of the decision to fund the FAP 408 bridges under section 144.

What is now the Bridge Program, as set out in section 144 and the applicable regulations, was enacted in 1970 solely as a highway bridge replacement program as part of the Federal Aid Highway Act of 1970. See P.L. 91-605, Title II, § 204(a), printed in 84 Stat. 1713, 1741-42 (eff. Dec. 31, 1970). Senator Randolph, the author of the bridge replacement program and then Chairman of the Senate Public Works Committee, introduced his bill, which became the law establishing the bridge replacement program, with the explanation that:

Congress is aware — and we are going to have to be much more aware — of the inadequate and the unsafe condition of many of the highway bridges in this country. Frankly, there is an inability on the part of the States to embark upon extensive replacement programs of these antiquated bridges. . . . Therefore, in this legislation we have provided $150 million a year for a program of bridge replacement. . . . We are not going to construct new bridges, but are going to replace unsafe and inadequate bridges, those which were not constructed for the present flows of traffic.

116 Cong.Rec. 34730 (Oct. 2, 1970) (emphasis added). Both the House Report (concerning a similar provision for which the Senate bill ultimately was substituted) and debates on the House floor reflect a concern about the safety and adequacy of old bridges following the collapse in 1968 of the Silver Bridge at Pleasant Point, West Virginia and the subsequent closing (and imposition of restrictions on the use) of other bridges. See House Rep. No. 91-1554 at 30 and at 55; U.S. Code Cong. Admin.News 1970, p. 5392; 116 Cong.Rec. 38951-52, 38958. Legislative history contemporaneous with the consideration and passage of the bill which led to creation of the bridge replacement program not only does not support but negates an intent that "replacement" would include the construction of new bridges on new highways.

The plaintiff's interpretation of "replacement" is consistent with the policy of making federal funds available for improving the quality of bridges on existing highways yet not dissipating those scarce funds for the construction of new bridges on new highways. The danger presented by the deterioration of many of our nation's existing bridges was after all "the evil" which the bridge replacement program was designed to remedy. See Church of the Holy Trinity v. United States, 143 U.S. 457, 463, 12 S.Ct. 511, 513, 36 L.Ed. 226 (1892). The inadequacy of the nation's network of limited access freeways had certainly been identified long before the collapse of the Silver Bridge focused Congressional attention on the condition of the nation's bridges, but the need for more freeways was addressed by passage of the Federal-Aid Highway Act of 1956. See § 108(a) of P.L. 84-627, 70 Stat. 374, 378; see also 23 U.S.C. § 101(b) (West Supp. 1982). There is no indication in the legislative history of 23 U.S.C. § 144 that enactment of a bridge replacement program was deemed necessary or even relevant to fulfillment of the purpose to construct freeways expressed in 23 U.S.C. § 101(b).

The state's argument that the legislative history of section 144 indicates that Congress intended a broad concept of "functional replacement" for the Bridge Program rests primarily on statements by Representatives Howard and Findley. In September of 1978, eight years after the enactment of the legislation creating the Bridge Program and during debate on the provisions of what became the Surface Transportation Assistance Act of 1978, Rep. Howard, then Chairman of the House Committee on Public Works and Transportation, specifically referred to "the Florence bridge" as one which should be considered for federal funding:

One of these is the existing Florence Bridge located on a heavily traveled Federal-aid primary route connecting central and western Illinois. The entire existing highway route is inadequate to meet traffic demands and is being replaced with a new facility. As part of this project, the bridge at Florence spanning the Illinois River must be replaced. It is both structurally deficient and functionally obsolete.

124 Cong.Rec. H 32318 (Sept. 28, 1978). Only comments made contemporaneously with actual consideration and passage of a bill are helpful in ascertaining Congressional intent. When Rep. Howard made this comment in 1978, the debate and consideration of Congress focused on the concept of "rehabilitation," not that of "replacement." A replacement program had been in existence since 1970, and passage of the 1978 Act did not affect the use of the term "replacement" in section 144.

Indeed, the 1978 Act merely added the concept of rehabilitation to the program, made the program available to a wider range of highways, and increased the appropriations and the percentage of the federal share in the federal/state financing ratio. More significantly, however, Rep. Howard's ambiguous comment does not amount to an endorsement of the kind of project which the defendants have in mind. All Rep. Howard said was that the Florence bridge must be replaced. Given the condition of the Florence bridge in 1978 ( see A.R.App. 16), it was reasonable to think that it was a good candidate for replacement with funds from the Bridge Program and ought therefore to be considered for that program. His statement did not indicate that he thought that the Florence bridge and U.S. 36-54 would stay in operation while its "replacement," two new bridges, would be constructed over 4 1/2 miles away on another, as yet unconstructed, freeway. His statement did not attempt to give any nuances to the word "replace."

Representative Findley's comment in 1981, eleven years after the bridge replacement program was enacted, is considerably more precise:

MR. FINDLEY: Mr. Chairman, I would like to clarify the intent of Congress concerning two aspects of the highway bridge replacement and rehabilitation program that have come under question on a bridge replacement project in central Illinois. The State intends to use Federal bridge funds to replace the existing U.S. 36 bridge in Florence, Illinois with a new structure to be located 4 1/2 miles away near Valley City, and to retain the existing bridge to serve local traffic. Although the Federal Highway Administration has approved this action, the project's eligibility for Federal bridge funds has been questioned because of the distance between the existing bridge and its replacement and because of the decision to retain the existing facility.

While the new bridge is in a slightly different location, it is part of an overall replacement of existing U.S. 36 in that area with a new freeway. The new bridge will serve the same general traffic corridor as the existing bridge and is therefore a functional replacement for it. It was decided to allow the old bridge to remain open to local traffic in order to make best use of highway facilities and to meet overall system needs.

I want to clarify that the FHWA properly understood the intent of Congress when it approved the Florence bridge replacement project. Congress intended that the term "replace" include the construction of a new facility in the same general traffic corridor, and not necessarily in the exact same spot. It further intended that the replaced structure be allowed to remain and carry traffic if the Secretary of the U.S. Department of Transportation determines that the combined use of the replaced and new facilities was necessary to meet system needs. As a sponsor of this bill and chief sponsor of the 1978 Highway Act that broadened the bridge program and provided the funds made available by FHWA for replacement bridge structures, do you agree with this clarification of congressional intent and its application to the new Florence bridge project?

MR. ANDERSON (of California): Yes, I agree with your statement of congressional intent concerning replacement bridge language and with the assessment that the Florence bridge project in central Illinois is an eligible use of Federal highway bridge replacement and rehabilitation funds.

127 Cong.Rec. H 6592 (Sept. 24, 1981). This too fails to meet the requirement that to be part of the legislative history of the enactment, a statement must be contemporaneous with its consideration and passage. No provision pertaining to the bridge replacement program was in debate in 1981 when Rep. Findley made his comment; the legislation under consideration was a lump-sum appropriation for all functions of the Department of Transportation, passed as the Department of Transportation Act of 1981. See 94 Stat. 1681. It should also be noted that these statements were made a year after this suit was filed challenging the use of Bridge Program funds for the Napoleon Hollow bridges.

Nor is Rep. Anderson's agreement with Rep. Findley's statement a part of the legislative history relevant to the intent of Congress in enacting the bridge "replacement" program. Such subsequent "clarifications," when made by a Congressman in whose district the project is to be constructed and the funds spent, or even when made by a legislator significantly involved with the earlier enactment — a status apparently not held by Rep. Anderson insofar as the 1970 Act is concerned or we assume that he would have been identified as such — "provide an extremely hazardous basis for inferring the meaning of a congressional enactment." Consumer Product Safety Comm'n v. GTE Sylvania, Inc., 447 U.S. 102, 118 n. 3, 100 S.Ct. 2051, 2061 n. 3, 64 L.Ed.2d 766 (1980).

Such history does not bear strong indicia of reliability . . . because as time passes memories fade and a person's perception of his earlier intention may change. Thus, even when it would otherwise be useful, subsequent legislative history will rarely override a reasonable interpretation of a statute that can be gleaned from its language and legislative history prior to its enactment.

Finally, even assuming that the regulations as amended in December of 1979 permit federal funding of a new bridge and continued operation of the "replaced" bridge, a conclusion which, as previously indicated, we find inconsistent with the provisions of the 1979 amendment, we find no basis for the defendants' argument that Congress, either by silence or through appropriations, has ratified such a change in the regulations. The only legislation concerning the Bridge Program passed after the December 1979 modification of 23 C.F.R. § 650.411(c)(2) was the Department of Transportation Act of 1981, a lump-sum appropriation for all functions of the Department of Transportation. See 94 Stat. 1681.

Congress may, of course, ratify what it might have authorized, and appropriations acts may constitute ratification. Ex parte Endo, 323 U.S. 283, 303 n. 24, 65 S.Ct. 208, 219 n. 24, 89 L.Ed. 243 (1945). "But the appropriation must plainly show a purpose to bestow the precise authority which is claimed. We can hardly deduce such a purpose . . . where a lump appropriation was made for the overall program for the Authority and no sums were earmarked for the single phase of the total program . . . involved." Id. The Florence bridge is merely one of many proposed bridge projects. Additionally, the Bridge Program is only one of many programs administered by the Department of Transportation and funded en grosse by the Department of Transportation Act of 1981.

Congressional silence is hardly a promising basis for implying ratification of a regulation. "It is at best treacherous to find in congressional silence alone the adoption of a controlling rule of law." Girouard v. United States, 328 U.S. 61, 69, 66 S.Ct. 826, 829, 90 L.Ed. 1084 (1946). Unless Congressional repudiation of a specific practice has been sought and denied, to rely on administrative practice, of which there is no indication that Congress was aware, is to elevate Congressional inaction to positive legislative decision. See Baltimore Ohio Ry. v. Jackson, 353 U.S. 325, 330-31, 77 S.Ct. 842, 845, 1 L.Ed.2d 862 (1957); see generally, J.W. Hurst, Statutes in Court 158-63, 207-16 (1970), and cases discussed therein. Where a statute is ambiguous, some deference is accorded to administrative practice reflecting decisions at the policy making level which have existed for many years. United States v. Shreveport Grain Elevator Co., 287 U.S. 77, 84, 53 S.Ct. 42, 44, 77 L.Ed. 175 (1932); see Hurst, supra, at 160-63 and cases discussed therein.

The record in this case does not present a basis upon which to infer Congressional ratification of the defendants' interpretation of the post-December 1979 version of 23 C.F.R. § 650.411(c)(2) as permitting the kind of "functional replacement" represented by the current proposal for funding of the FAP 408 bridges. In the first place, the statute is not ambiguous. In addition, the current version of subsection (c)(2), and the interpretation given to it in this case by the defendants, have, at most, been in existence only slightly more than three years. In fact, the parties have cited no other instance where a similar interpretation has been made.

Even the decision that subsection (c)(2) has expanded the scope of "replacement" so that the Florence bridge may be replaced by new twin bridges and yet remain in operation to its pre-replacement extent has been made only at the divisional and regional levels of the FHWA.

Interestingly, some of these same officials, prior to December 1979 concluded that this same plan (as far as the bridges are concerned) did not qualify for the replacement program because of the distance between Florence and Napoleon Hollow and because the Florence bridge would remain in operation after construction of its "replacement," see A.R. at 30015-13, 30007 and 1063. However, the modified regulation which they interpret to permit this plan was presented in the Federal Register as an insignificant change in the existing regulations. Such a record hardly constitutes Congressional ratification of an administrative interpretation of a regulation where that interpretation is such a significant departure from the language of the statute and the intent of Congress at the time that the program under which the regulation was promulgated was enacted.

If the interpretation of the statute and regulations which the defendants urge is valid, most, if not substantially all, bridges to be constructed as part of new limited access superhighways or freeways will qualify for funding under the Bridge Replacement and Rehabilitation Program. Almost all the interstate highways and freeways parallel existing two or even four lane open access roads. They cross most of the same rivers, railroads and roads which the old highways crossed. The old highways are not abandoned although much of the through traffic which previously utilized them is diverted to the new freeway. The bridges on the old highways remain and, if necessary, are repaired and rehabilitated. Since the passage of the Federal-Aid Highway Act of 1956, printed in 70 Stat. 374, et seq., Congress has provided substantial funds for the construction of limited access freeways and superhighways including any necessary bridges, underpasses and overpasses. See 23 U.S.C. § 101(b) (West Supp. 1982) and notes contained in West Supplement to 23 U.S.C.A. § 101. It has also provided funds for the replacement and rehabilitation of existing unsafe bridges under the Bridge Program since it was enacted in 1970.

It is not within the power or prerogatives of the Department of Transportation to allocate funds appropriated by Congress to repair or replace unsafe bridges on existing highways to the construction of new bridges on new freeways. Yet this is precisely what the defendants here seek to do.

IV. THE ENVIRONMENTAL ISSUE

As indicated, the plaintiff also contends that the decision to build bridges and a freeway through Napoleon Hollow violates the National Environmental Policy Act (NEPA), 42 U.S.C. § 4321, et seq., the Federal-Aid Highway Act, 23 U.S.C. § 101, et seq., particularly section 138, the Department of Transportation Act, 49 U.S.C. § 1651, et seq., particularly section 1653(f), and the regulations applicable to all of the foregoing statutes. The requirements of NEPA are more general and less rigorous than those of either of the other two Acts on which the plaintiff relies, and, accordingly, a statement as to the NEPA requirements will establish a background for a comparative description of the more rigorous standards set by the Federal-Aid Highway Act and the Department of Transportation Act.

NEPA requires that every recommendation or report on "major Federal actions significantly affecting the quality of the human environment" include "a detailed statement by the responsible officials." 42 U.S.C. § 4332(2)(C). That statement must describe, among other things, "the environmental impact of the proposed action," 42 U.S.C. § 4332(2)(C)(i), "any adverse environmental effects which cannot be avoided should the proposal be implemented," 42 U.S.C. § 4332(C)(ii), and "alternatives to the proposed action." 42 U.S.C. § 4332(2)(c)(III).

In dictum, the United States Supreme Court has noted that although "NEPA does set forth significant substantive goals for the Nation," its "mandate to the agencies is essentially procedural." Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519, 558, 98 S.Ct. 1197, 1219, 55 L.Ed.2d 460 (1978), quoted with approval in Strycker's Bay Neighborhood Council, Inc. v. Karlen, 444 U.S. 223, 227, 100 S.Ct. 497, 499, 62 L.Ed.2d 433 (1980) (per curiam); see Strycker's Bay, 444 U.S. at 229, 100 S.Ct. at 500 (Marshall, J., dissenting). Vermont Yankee and Strycker's Bay emphasize that the purpose of NEPA is "to insure a fully informed and well-considered decision." 435 U.S. at 558, 98 S.Ct. at 1219; 444 U.S. at 227, 100 S.Ct. at 499. The court's role in reviewing a decision which is subject to NEPA is to determine that the agency has followed NEPA's procedural requirements, Strycker's Bay, 444 U.S. at 227, 100 S.Ct. at 499, and "to insure that the agency has taken a `hard look' at environmental consequences; [the reviewing court, however,] cannot `interject itself within the area of discretion of the executive as to the choice of the action to be taken.'" Kleppe v. Sierra Club, 427 U.S. 390, 410 n. 21, 96 S.Ct. 2718, 2730 n. 21, 49 L.Ed.2d 576 (1976), quoting National Resources Defense Council v. Morton, 458 F.2d 827, 838 (D.C. Cir. 1972); see Strycker's Bay, 444 U.S. at 227-28, 100 S.Ct. at 499; National Wildlife Federation v. Lewis, 519 F. Supp. 523, 531 (D.Conn. 1981), aff'd, 677 F.2d 259 (2d Cir. 1982). Thus, a court's substantive review of the agency's decision is a comparatively narrow one, limited to whether that decision was arbitrary and capricious under the Administrative Procedure Act, 5 U.S.C. § 706. Grazing Fields Farm v. Goldschmidt, 626 F.2d 1068, 1072 (1st Cir. 1980). The admonition of Strycker's Bay, restated, is that "[t]he court should only assure itself that the agency has given good faith consideration to the environmental consequences of its actions and should not pass judgment on the balance struck by the agency among competing concerns." 626 F.2d at 1072.

Our inquiry under NEPA then is whether the EIS (here, the EIS/4(f) Statement) adequately addressed the environmental concerns expressed in 42 U.S.C. § 4332(2)(C) so as to provide the administrative decision makers with a "hard look" at the environmental consequences of the proposed action in light of the advantages and disadvantages of alternative actions, and whether the agency complied with NEPA's procedural requirements. Kleppe, 427 U.S. at 410 n. 21, 96 S.Ct. at 2730 n. 21; Strycker's Bay, 444 U.S. at 229, 100 S.Ct. at 500 (Marshall, J., dissenting). An agency's decision, based upon an adequate EIS, to proceed with a project to which NEPA applies may be overturned as inconsistent with NEPA only if that decision was arbitrary or capricious. See 5 U.S.C. § 706(2)(A); Strycker's Bay, 444 U.S. at 229, 100 S.Ct. at 500 (Marshall, J., dissenting); Grazing Fields, 626 F.2d at 1072.

The 1972 Final EIS/4(f) Statement, if considered apart from the rest of the administrative record, woefully fails to address significant environmental aspects of the land targeted for the FAP 408 bridges and highway. Nowhere in that Statement is the presence of the American Bald Eagle in the PCCA in general and in Napoleon Hollow in particular even mentioned. That omission alone renders this initial EIS deficient for failure to address "the environmental impact of the proposed action," 42 U.S.C. § 4332(2)(C)(i), and "any adverse environmental effects which cannot be avoided should the proposal be implemented." 42 U.S.C. § 4332(2)(C)(ii). Additionally, the EIS does not mention the possible presence of other endangered or threatened animals, fish and plants in the vicinity of the proposed freeway and bridges. Nor is there any evaluation based on an adequate study of air pollution, water pollution or noise pollution which might result from routing the bridges and freeway through Napoleon Hollow. Moreover, although the EIS contains a general reference to "the potential archeological value" of the area affected by the proposal, no mention is made of any of the Russell Mounds, the Napoleon Hollow village site, or the Burnt Hill Multiple Resource Area. Notwithstanding all of these omissions, this EIS/4(f) Statement was approved by the Federal Highway Administrator, the Secretary of Transportation, the U.S. Department of the Interior, the Illinois Department of Conservation and the Illinois Department of Transportation.

The defendants defend this EIS on the grounds that most of what is now known about the "biota" and the historical and archeological significance of the PCCA has been discovered since that EIS was written, and that very little literature concerning this area existed before 1972. The defendants also argue that this EIS represents "the state of the art" in environmental impact statements in 1972.

As we have stated, there can be little doubt, and the defendants do not seriously dispute, that this EIS, standing alone, is deficient and did not serve the purpose which NEPA intends such statements to serve. NEPA imposes upon agencies an affirmative obligation to investigate if information is not conveniently at hand.

NEPA requires each agency to undertake the research needed to adequately expose environmental harms. . . . Clearly, the "detail" must flow from research conducted according to the statute's mandate. . . . If no . . . information is available, then defendants must see to it that the necessary research is conducted.

Pertinent information found in an administrative record but not incorporated in an environmental impact statement generally cannot bring into compliance with NEPA an EIS that by itself is inadequate. Grazing Fields Farm v. Goldschmidt, 626 F.2d 1068, 1072 (1st Cir. 1980). Here, however, the parties have stipulated to the supplementing of the administrative record, by inclusion in that record of various "biota" studies and other reports (such as the 1982 EIS/4(f) Statement regarding the Wade property), for the specific purpose of avoiding remand of this case in the event that formal review of the EIS revealed that it was inadequate, a determination which we have made on several occasions and now confirm. The NEPA issue before us is therefore not confined solely to the adequacy of the EIS. Rather, the issue is whether the administrative record, as supplemented by its various appendices, sustains under NEPA the decision to build the FAP 408 bridges through Napoleon Hollow as not "arbitrary or capricious."

Based on our review of the entire record, and with particular reference to the various "biota" studies which supplement the administrative record, we regard the contention that the supplemented administrative record adequately addresses "the environmental impact of the proposed action," and sets forth the "adverse environmental effects which cannot be avoided should the proposal be implemented," 42 U.S.C. § 4332(2)(C)(i) and (ii), as a defensible one. The record now describes, fairly comprehensively, the species of flora and fauna which can be found in the PCCA and in the environs of the alternative locations for a freeway crossing of the Illinois River. The record now examines, perhaps as thoroughly as can realistically be expected, possible adverse environmental consequences of the construction project, and attempts to discern those consequences which are unavoidable.

Although in a piecemeal fashion, the supplemented administrative record does provide the basis for a "hard look" at the environmental consequences of building this freeway and its bridges at either Napoleon Hollow or at certain other locations. Almost all of these studies and reports recommend against construction through Napoleon Hollow and view the other alternatives as preferable to the current alignment. However, almost all of these studies and reports also agree that the consequences of construction at Napoleon Hollow are unpredictable, given the new routing avoiding the Wade farm, the narrowed median, the elimination of the rest area from the original design, the defendants' agreement to limit construction to the period from March I through November 15, and the other mitigation features which the defendants identify.

The eagles, for example, might continue to winter in this area despite some chipping away of their habitat and despite the introduction of freeway traffic through part of their habitat. Both Dunstan and Becker conclude that other nearby ravines appear to be superior to Napoleon Hollow as wintering sites. A.R.App. 13 at 39; A.R.App. 7 at 7. Ingram notes evidence that the Big Blue watershed area was once a wintering site for eagles. A.R.App. 19 at 9. The risks, of course, are that the construction of the freeway through this area will so alter the balance of factors which makes the area attractive to the eagles that the PCCA will no longer be a suitable winter refuge, thereby possibly moving the species one step closer to extinction. Given the difficulty of predicting the consequences of constructing the freeway according to current plans, we cannot conclude that, based on the supplemented administrative record, a decision to approve the current alignment would be arbitrary or capricious under the NEPA standards even though we might well come to a different conclusion as to its wisdom. An administrator assessing the "balance of competing interests" as identified in the supplemented administrative record might reasonably conclude that the advantages of the project outweighed the risks that serious environmental disadvantages would occur.

That a decision to construct the FAP 408 alignment through Napoleon Hollow could be upheld under NEPA on the basis of the administrative record as supplemented does not, however, resolve the issues arising from the applicability of the Department of Transportation Act and the Federal-Aid Highway Act to the proposed site for this freeway. The substantive requirements of the Federal-Aid Highway Act and the Department of Transportation Act are more specific and more rigorous than those of NEPA. Section 4(f) of the Department of Transportation Act, 49 U.S.C. § 1653(f), and section 138 of the Federal-Aid Highway Act, 23 U.S.C. § 138, are "clear and specific directives." Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 411, 91 S.Ct. 814, 821, 28 L.Ed.2d 136 (1971). In identical language, both section 4(f) and section 138 provide that

the Secretary shall not approve any program or project which requires the use of any publicly owned land from a public park, recreation area, or wildlife and waterfowl refuge of national, State, or local significance as determined by the Federal, State, or local officials having jurisdiction thereof, or any land from an historic site of national, State, or local significance as so determined by such officials unless (1) there is no feasible and prudent alternative to the use of such land, and (2) such program includes all possible planning to minimize harm to such park, recreational area, wildlife and waterfowl refuge, or historic site resulting from such use. . . .

"Feasibility," as used in sections 4(f) and 138, focuses upon what "sound engineering" makes possible. Overton Park, 401 U.S. at 411, 91 S.Ct. at 821. For the exception to these statutes to apply, "the Secretary must find that as a matter of sound engineering it would not be feasible to build the highway along any other route." Id. "Prudence," as used in these sections, connotes more than "a wide-ranging balance of competing interests." Id. An alternative is "prudent" if it does not present "unique problems." Id. at 411 and 416, 91 S.Ct. at 821 and 823. "[O]nly the most unusual situations are exempted." Id. at 411, 91 S.Ct. at 821; see Monroe County Conservation Council, Inc. v. Volpe, 472 F.2d 693, 700 (2d Cir. 1972); National Wildlife, 519 F. Supp. at 534.

Of course, where only NEPA is applicable, " Vermont Yankee cuts sharply against the . . . conclusion that an agency, in selecting a course of action, must elevate environmental concerns over other appropriate considerations." Strycker's Bay, 444 U.S. at 227, 100 S.Ct. at 499. But where the area affected by an agency's decision is protected by section 4(f) or section 138, the agency bears a substantially higher burden than that of merely considering the environmental consequences. As Justice Marshall stated in Overton Park:

It is obvious that in most cases considerations of cost, directness of route, and community disruption will indicate that parkland should be used for highway construction whenever possible. Although it may be necessary to transfer funds from one jurisdiction to another, there will always be a smaller outlay required from the public purse when parkland is used since the public already owns the land and there will be no need to pay for right-of-way. And since people do not live or work in parks, if a highway is built on parkland no one will have to leave his home or give up his business. Such factors are common to substantially all highway construction. Thus, if Congress intended these factors to be on an equal footing with preservation of parkland there would have been no need for the statutes.

Congress clearly did not intend that cost and disruption of the community were to be ignored by the Secretary. But the very existence of the statutes indicates that protection of parkland was to be given paramount importance. The few green havens that are public parks were not to be lost unless there were truly unusual factors present in a particular case or the cost or community disruption resulting from alternative routes reached extraordinary magnitudes. If the statutes are to have any meaning, the Secretary cannot approve the destruction of parkland unless he finds that alternative routes present unique problems.

As with review of an agency's decision pursuant to NEPA, review of an agency's decision under section 4(f) or section 138 that no feasible and prudent alternative to the taking of a protected area exists is governed by the "arbitrary or capricious" standard of the Administrative Procedure Act, 5 U.S.C. § 706(2)(A). Overton Park, 401 U.S. at 416, 91 S.Ct. at 823. "[T]he reviewing court must be able to find that the Secretary could have reasonably believed that in this case there are no feasible alternatives or that alternatives do involve unique problems." Id.

The plaintiff contends that no 4(f) statement which meets the standards of the statutes and the judicial decisions interpreting them has been prepared with respect to the feasibility and prudence of a number of alternatives to the Napoleon Hollow crossing. Specifically, she urges that the EIS/4(f) Statement and even the augmented record lack detailed information, are conclusory, and do not demonstrate factually that, in the language of Overton Park, supra, the possible alternative routes involve "unique problems," "truly unusual factors," or "costs or community disruption" of "extraordinary magnitudes." Accordingly, plaintiff asserts, the finding by the Secretary of Transportation that there are no feasible or prudent alternatives to Napoleon Hollow was arbitrary and capricious.

The defendants insist that the administrative record as augmented pursuant to agreement presents sufficient facts to warrant an administrative determination (1) that no feasible or prudent alternative exists to the Napoleon Hollow crossing and (2) that all possible planning to minimize harm to Napoleon Hollow and the Pike County Conservation Area has been done. Accordingly, they contend that the record as augmented meets the requirements of sections 4(f) and 138 and the cases interpreting them and a finding to that effect was not arbitrary and capricious.

Mindful of the admonition that an agency is not required "to ferret out every possible alternative, regardless of how uncommon or unknown that alternative may have been at the time the project was approved," Vermont Yankee, 435 U.S. at 551, 98 S.Ct. at 1215 (discussing, however, the requirements of NEPA), we confine our analysis to the defendants' evaluations of the alternatives identified in the administrative record as supplemented and the 1972 Final EIS/4(f) Statement, including Florence, Valley City/Flint Creek, the Blue Creek Basin, and the other (and smaller) PCCA ravine 0.4 mile north of Napoleon Hollow.

As indicated, both sections 4(f) and 138 prohibit the Secretary of Transportation from approving "any program or project which requires the use of any publicly owned land from a . . . wildlife and water-fowl refuge of national, State or local significance . . . or any land from an historic site . . . unless (1) there is no feasible and prudent alternative to the use of such land. . . ." Our first inquiry, then, must be as to whether there are any feasible alternatives to the proposed bridges and highway through Napoleon Hollow. The record clearly indicates that, so far as "feasibility" is concerned, as "a matter of sound engineering," FAP 408 could cross the Illinois river at a number of alternative locations other than Napoleon Hollow. Various alternatives were considered and rejected not for engineering reasons but for various other reasons such as (1) the severance of valuable farm land which would result, (2) the disruption and dislocation of businesses,(3) the greater amount of rock cut and the cost thereof necessary for two bridges as part of a four-lane highway with a slope meeting design standards including the necessary grade for semi-trailer traffic, (4) the necessity for relocation of approximately one-half mile of the Norfolk and Western Railroad tracks and the additional costs attendant thereto as well as to the lengthening of the highway by almost one mile, (5) potential damage to the ecology through introduction of extra fill, etc.

Although we do not have current cost estimates, Napoleon Hollow is, considering cost and ease of construction, probably the most feasible and prudent of the possible sites for the FAP 408 bridges and freeway to cross the Illinois River. That fact, however, does not warrant its selection. The applicable criterion is not which is the most feasible and prudent route but whether or not there are other feasible and prudent alternatives, for portions of a wildlife preserve or historic site may be utilized for bridges and highways only if no such alternatives exist. So far as engineering feasibility is concerned, there certainly are other feasible routes. The question then is one of prudence.

Three of the possible alternative locations for a freeway bridge over the Illinois River (Florence, the Blue Creek Basin, and the PCCA ravine north of Napoleon Hollow) were rejected as "imprudent" because each would require "a tremendous magnitude" of rock cutting compared to the amount of rock cutting required for construction at and through Napoleon Hollow. Neither the 1972 Final EIS/4(f) Statement nor the augmented record, however, estimate the amount of rock cutting which would be required at either Napoleon Hollow or any of the alternative locations. It is impossible to tell what, if any, factual support exists for the conclusion that the "magnitude" of the rock cutting at each of the alternatives would be "tremendous" compared to the magnitude of rock cutting at Napoleon Hollow.

Indeed, the only estimate of the amount of rock cutting required to build the Napoleon Hollow alignment is contained in the 1982 EIS/4(f) Statement concerning the Wade property. That Statement indicates that the Napoleon Hollow alignment before re-routing to the north of the Wade property would have required 80,000 cubic yards of rock cutting. A.R.App. 20 at 12. The current Napoleon Hollow alignment around the Wade farm apparently requires double the amount of rock cutting, 160,000 cubic yards. The "tremendous magnitudes" of the rock cutting required at the alternative locations, had those amounts been estimated quantitatively in the 1972 Final EIS/4(f) Statement, might provide an illuminating comparison against the information now provided in the 1982 EIS/4(f) Statement about the additional rock cutting required in avoiding the Wade property. The ineluctable conclusion from the decision to reroute the FAP 408 alignment north of the Wade property rather than to modify the path of the freeway so as to avoid the buildings on the Wade farm, despite a difference in amounts of necessary rock cutting of over 100% of the lower amount, obviously is that doubling the amount of rock cutting from 80,000 cubic yards to 160,000 cubic yards would not, in itself, render any of the alternative locations for a crossing of the Illinois River "imprudent."

Significant differences in the amounts of rock cutting necessary for construction would undoubtedly affect the estimated costs of construction at the various alternative locations. What little cost data the 1972 Final EIS/4(f) Statement provides were so out of date even in 1972 when that Statement was written and in 1974 when it was approved by the Secretary of Transportation as to be virtually meaningless. The 1972 Design Location Study estimates the costs of building bridges at Napoleon Hollow and at the ravine 0.4 mile north of Napoleon Hollow, but it gives conflicting figures for the "total costs" of those two alignments. Furthermore, it is unclear how the elimination of the rest area from the proposal for FAP 408 would affect any comparison of rock cuts and costs for the two crossings discussed in the Design Location Study. Additionally, neither the 1972 Final EIS/4(f) Statement nor the rest of the administrative record provides much quantitative factual support for the conclusions that the necessity of shifting the railroad tracks to accommodate the Valley City/Flint Creek Basin alignment and the number of homes and businesses along U.S. 36-54 which would be taken by the Florence alignment render those respective alignments so extraordinarily costly as to be imprudent.

The 1972 Final EIS/4(f) Statement rejects the Valley City/Flint Creek alignment as imprudent because railroad tracks cross the river through that valley and would have to be moved before the freeway could be constructed there. (A.R.App. 4 at 23, reprinted in Part I, infra, of this opinion.) The Statement refers to the relocation of the railroad tracks as "introducing extra curvature for the railroad," but does not indicate the extent of that "extra curvature" or what significance, if any, the extra curvature would have. It does not even discuss the possibility of bridging, overpassing, or underpassing the railroad and the cost thereof.

The conclusion in the EIS/4(f) Statement that construction of the freeway through the Flint Creek Basin "would also result in additional damage to the ecology of the area by introducing extra fill for the freeway and the railroad" suffers from a similar lack of factual support and analysis. Finally, the observation that the Valley City alignment would be almost one mile longer than the Napoleon Hollow alignment is insufficient in itself to render the Valley City alignment imprudent.

The administrative record contains some information from a 1966 study comparing the takings of property required by the Napoleon Hollow alignment and the U.S. 36-54 alignment. A.R.App. 2 at 9, Figure 1. The 1972 Final EIS/4(f) Statement repeats some financial data and comparative "Benefit-Cost Ratios" taken from that 1966 study. A.R.App. 4 at 62. Although the Design Location Study (1972) contains tables from which some information about the numbers of buildings (and their values) which would be taken by the alternative alignments can be discerned, see A.R.App. 5, Vol. I at 107, 110 and 116, neither the EIS/4(f) Statement nor the augmented administrative record as a whole contains any systematic, coherent discussion of how the severance of property required by the Florence alignment would affect the total cost of a crossing at Florence in comparison with the cost of the Napoleon Hollow alignment.

The conclusion of the EIS/4(f) Statement that the U.S. 36-54 alignment would take a greater number of commercial and residential properties than would the alignment at Napoleon Hollow is obvious since the latter is a wildlife preserve, but under Overton Park, supra, that an alternate alignment which avoids taking section 4(f) — protected property is more costly and requires greater commercial and residential severance than does an alignment which takes the protected property does not establish that the former is imprudent. Such differences are a question of degree, and, in accordance with Overton Park, such a difference must be of extraordinary magnitude if it is to justify the taking of section 4(f) — protected property. Neither the 1972 Final EIS/4(f) Statement by itself nor that Statement plus the remainder of the augmented administrative record contains sufficient factual information upon which the U.S. 36-54 alignment could reasonably be deemed so extraordinarily more expensive and disruptive than the Napoleon Hollow alignment as to be "imprudent."

Incorporation of the existing lift bridge at Florence into a larger freeway bridge system at that site was deemed "not feasible." Assuming that, as a matter of sound engineering, the current Florence bridge could not be modified to be part of (or one of the spans of) a double freeway bridge, it does not necessarily follow that a freeway crossing north of the existing bridge in the vicinity of Florence (as opposed to one utilizing the old bridge) is "not feasible." Constructing two new freeway bridges in the Florence area might, as a matter of sound engineering, still be "feasible." The wisdom of such action (particularly in light of the recent rehabilitation of the existing lift bridge) goes not to whether it is feasible, but to whether it is prudent.

It is clear from the record here that construction of twin spans across the Illinois River at Napoleon Hollow and the construction of the connecting four lane freeway through the Pike County Conservation Area will involve both environmental and archeological losses and hazards. The defendants concede this. While the State of Illinois has, since this action was filed, proposed a number of mitigating steps for which the responsible state officials deserve commendation, all involved acknowledge, as previously discussed, that there inevitably will be some archeological and environmental loss with the possibility of substantially more.

The policy behind sections 4(f) and 138 is to avoid such losses and risks unless no feasible and prudent alternatives are available. We have already noted that from an engineering point of view, which is the applicable criterion, a number of alternatives are feasible. Alternatives are prudent, the Supreme Court held in Overton Park, supra, unless there are "truly unusual factors present in a particular case or the cost or community disruption resulting from alternative routes reached extraordinary magnitudes." Referring to sections 4(f) and 138, the Court summarized, "If the statutes are to have any meaning, the Secretary cannot approve the destruction of parklands unless he finds that alternative routes present unique problems." 401 U.S. at 413, 91 S.Ct. at 821.

The record before us unfortunately fails to contain current estimates of the differences in cost between the various alternatives, the volume of rock cutting at each location, the number of residences or businesses which would be adversely affected, the extent of community disruption, etc. As a result, it is impossible to determine that there are no prudent alternatives to Napoleon Hollow.

The description in Overton Park of the factors usually present in highway construction and the analyses of the application of sections 4(f) and 138 are equally applicable here. Given the comparative historic, archeological and environmental significance of Napoleon Hollow and of Overton Park, it seems obvious that the standards to be applied to the former and its possible alternatives should be at least as protective as were those enunciated in the Overton Park decision. The Supreme Court said:

"It is obvious that in most cases considerations of cost, directness of route, and community disruption will indicate that parkland should be used for highway construction whenever possible. . . . there will always be a smaller outlay required from the public purse when parkland is used since the public already owns the land and there will be no need to pay for right-of-way. And since people do not live or work in parks, if a highway is built on parkland no one will have to leave his home or give up his business."

The Court went on to say that "if Congress intended these factors to be on an equal footing with preservation of parkland there would have been no need for the statutes." 401 U.S. at 412. It concluded, "Congress clearly did not intend that cost and disruption of the community were to be ignored by the Secretary. But the very existence of the statutes indicates that protection of parkland was to be given paramount importance." 401 U.S. at 412-13, 91 S.Ct. at 821. Accordingly, it held that sections 4(f) and 138 prohibited the approval of highway routes through publicly owned parklands, recreation areas, wildlife and waterfowl refuges or historic sites designated as of national, State or local importance unless no feasible and prudent alternative was available, i.e., unless every alternative route involved unique problems, truly unusual factors, or the cost or community disruption reached extraordinary magnitude.

It may be relevant at this point to state that we have no doubt about the desirability of constructing a four-lane limited access freeway through west-central Illinois in an east-west corridor defined generally by Decatur, Springfield, Jacksonville, and the Quincy-Hannibal, Missouri, area. The need for such a thoroughfare is apparent from the record. We are surprised that this so-called Central Illinois Expressway has been more than twenty years in the planning and construction. It first came to our attention some two and one-half years ago when the instant suit was filed.

Neither the undoubted need for FAP 408 or the fact that it has already been long delayed, however, justifies ignoring the standards which Congress has provided for balancing the paramount interest of protecting areas such as the Pike County Conservation Area with its unique and priceless environmental, historic and archeological characteristics against the importance of providing needed highways. Nor does it justify an executive department's utilizing funds which Congress has provided for one purpose, replacing and rehabilitating dangerous or obsolete bridges, to building two new bridges as part of a new four-lane limited access freeway while the bridge the two new spans allegedly replace is repaired and retained as a continuing segment of the same two-lane open access highway of which it has long been a part.

CONCLUSION

In light of the foregoing, we conclude that the record fails to establish that there is no feasible and prudent alternative to the construction of FAP 408 through Napoleon Hollow and that the determination by the Secretary of Transportation that no such alternatives exist was, therefore, arbitrary and capricious. We also conclude that funds appropriated by the Congress for use under the Highway Bridge Replacement and Rehabilitation Program may not properly be utilized to construct the proposed two new bridges across the Illinois River at Napoleon Hollow. Finally, we conclude that, in light of the augmented record, the conclusion to build the bridge and freeway at and through Napoleon Hollow was not, under the National Environmental Policy Act, 42 U.S.C. § 4321, et seq., arbitrary and capricious.

Accordingly, we enjoin the defendants and their agents and employees or persons contracting with them from funding and constructing two bridges at Napoleon Hollow for FAP 408 with funds from the Highway Bridge Replacement and Rehabilitation Program, 23 U.S.C. § 144, and from taking property from Napoleon Hollow or the Pike County Conservation Area for the FAP 408 highway and bridges. An appropriate order will enter. The Court will retain jurisdiction to modify or vacate the injunction in the event a proper source of funds and an adequate statement under sections 4(f) and 138 are secured.