Perini Corp. v. United States

Cowen, Chief Judge,

delivered the opinion of the court:

The central issue in this contract case is whether a substantial variation from an estimated quantity in contract specifications constitutes a changed condition within the meaning of the standard Changed Conditions article, when the facts establish that the variation was foreseen by the contractor and should reasonably have been anticipated by the Government. That question, and a host of related ones, have generated protracted proceedings before the contracting officer, the Corps of Engineers Claims and Appeals Board, and in this court.1 What is unique here is that the Government rather than the contractor is the party claiming that a changed condition existed. For the reasons discussed below, we reject the Government’s position. We hold that a material variation from an estimated quantity, here consisting of an overrun in the amount of water to be pumped from a speci*771fied area, did not constitute a changed condition within the circumstances of this case and therefore that plaintiff is entitled to be paid the stated contract price for the entire amount pumped.

I

The instant contract is one of a series contributing to the completion of the Jim Woodruff Dam on the Apalachicola River near Chattahoochee, Florida. In order of time, it was the third major contract for the work at the site. The first was for the erection of an earth overflow dike to be built on the east bank of the river by the Shepherd Construction Company; the second, still being performed by plaintiff2 at the time the invitation for bids on the instant contract was announced, was for the construction of a lock and fixed crest spillway.

The contract here involved the construction of a gated spillway, powerhouse and switchyard. The work was to be performed in two separate cofferdam stages in order to avoid interrupting the riverflow and to permit navigation during the construction period. The first cofferdam stage included the major portion of the gated spillway and switchyard embankment construction; the second, the powerhouse and the remainder of the spillway and switchyard structures. Both the gated spillway and the powerhouse were to be located in the river bottom but were to be constructed in the dry within the cofferdams. These requirements meant that water would have to be pumped from the worksite.

In November of 1951, an invitation for bids for the contract was issued. It requested that the bids be submitted on the basis of a price schedule containing 153 pay items, most of which were based on a unit price, but some called for lump sum bids. Included were Bid Items 2 and 102 which required the submission of separate unit prices for pumping units of 1,000 gallons of water per unit for the first and second stage cofferdam areas respectively. The Unit Price Schedule stated an estimated quantity of 323,000 units for Item 2 and 302,000 units for Item 102.3 The contract also contained the *772usual clauses requiring the bidders to make their own site inspection.

Plaintiff bid 40 cents per unit for Items 2 and 102. Despite the fact that these unit prices greatly exceeded those of the other bidders, plaintiff’s total bid of $13,907,379.60 was the low bid, and plaintiff was awarded the contract at that price. The contract was dated January 18,1952; the notice to proceed was received by plaintiff on February 15,1952.

Large quantities of water were encountered almost immediately during the first stage work. By August 12, 1953, pumping from that area had already exceeded the estimated 323,000 units by more than 100 percent. On that account, plus a report from a Government geologist that similar conditions would lead to even greater excesses with respect to the unstarted second stage area, defendant’s Resident Engineer asked plaintiff for proposals and comments in view of the “quite high” unit prices for pumping. When plaintiff demurred, defendant suggested that it assume some of the pumping. The plaintiff by letter of September 10, 1953, objected to this proposal. Subsequently, plaintiff suggested a revised dewatering plan for the second stage cofferdam.

On May 17, 1954, defendant again requested plaintiff to agree to a reduction in the contract price for pumping. Plaintiff again refused. Then, in June, plaintiff was asked to accept a substitute price of 15 cents per unit for the second stage. Plaintiff was also informed that if it failed to agree to the proposed price substitution, defendant would issue a change order reducing the contract unit price. Plaintiff declined to make any change in the contract price, pointing out that “to the present time no pumping whatever has taken place with respect to the Second Stage Cofferdam.”

Work began in the second stage area. By the end of November 1954, plaintiff had pumped 1,137,620 units above the 302,000 units estimated for this stage. One month later, the contracting officer issued Change Orders Nos. 19 and 20. Change Order 19 dealt with the first stage area which had been completed one year earlier and for which plaintiff had already been paid the contract price. The order asserted that a changed condition existed in the subsurface conditions and added a new item to the Unit Price Schedule, *773which provided for pumping 502,501 units, the exact number of units plaintiff had pumped above the original estimated quantity. For the pumping of that quantity, the change order fixed the price at 25 cents per unit instead of the 40 cents bid by plaintiff.

Change Order No. 20 stated that a changed condition existed in the subsurface conditions of the second stage area and set forth a temporary unit price of 10 cents per unit (instead of 40 cents) pending final determination “of such new and final price.” It also raised the estimated quantity of water to be pumped from the second stage to 2,000,000 units. On March 28, 1955, a supplement to Change Order No. 20 fixed the “new and final price * * * for all quantities” of water pumped from the second stage area “in excess of the original estimate of 302,000 units” at 8 cents per unit, and increased the estimated quantity of such item to 2,150,000 units.

The contract as extended was completed on September 25, 1955. As of that date, plaintiff had pumped from the first stage area 825,501 units of water instead of the 323,000 units originally estimated. From the second stage area, plaintiff had pumped 6,600,818 units instead of the 302,000 units estimated.4

An explanation of this enormous overrun entails a review of the facts relating to the water-carrying potential of the rock that had to be excavated for the foundations of the structures, as well as the water conditions that prevailed in the foundations themselves. As previously stated, the contract was to be performed in the dry on the river bottom of the Apalachicola Eiver. The river bottom consisted of four layers of various materials and strata. In descending order, tihe first layer was composed of an overburden of sands and gravels. The second layer, known as the Tampa formation, was a limestone stratum of the rock which plaintiff was to excavate for the powerhouse and gated spillway foundations. This formation was highly porous and permeable, possessing *774a large number of interconnected channels, pits and Cavities. The third layer, known as the Suwannee formation, consisted of limestone and dolomite rock of very high porosity and permeability. At the uppermost part of this formation immediately below its contact with the Tampa formation was a thin layer of dense limestone known as the D-zone. Because of its density, the D-zone usually operated as an aquaclude or partial obstacle to the passage of water from the Suwannee to the Tampa formation. The fourth layer, known as the Ocala formation, was another porous limestone stratum containing a more highly developed system of interconnected solution channels, cavities and pits than either the Tampa or Suwannee formations.

Since the contract required construction in the dry in a river bottom composed of such material, it was obvious the contract work would meet with great difficulties in terms of the amount of water to be pumped.5 In point of fact, the Tampa, Suwannee and Ocala limestone formations at the site were part of a well known and much publicized artesian aquifer system which underlies almost all of Florida and which supplies approximately 90 percent of the public and private water supply for that state.

An equally important cause of the water pumping problem was the fact that the several formations in the site area were not horizontal. Rather, commencing upstream, they dipped in a downstream direction. As a result, each formation in dhinglelike fashion was progressively exposed to the river water at a point of intersection upstream of the site, allowing water to enter the formations. Such continual upstream recharging produced an artesian water condition at the site, for, after the river water entered the formations, it then passed through the interconnecting cavities and channels in the limestone rock into the cofferdam area.

The D-zone, the aquaclude which was expected to prevent the large flow of water into the cofferdams, failed for several re'asons. Since it was located below the Tampa formation, the *775D-zone could not prevent the river water from recharging this formation. Neither could it prevent artesian water, apart from the river recharging in the Tampa formation itself, from flowing into the protected area. Moreover, the D-zone did not extend as an unbroken layer over the entire worksite, either because of natural causes or exploratory drilling and excavation work. Therefore, wherever the D-zone was absent, the water under 'artesian pressure would escape from the underlying Suwannee formation.

By far the greatest source of water in the second stage area, however, was an ancient river channel located to the east of the open end of the second stage cofferdam. Filled with highly permeable river sediments and cut into the rock underlying the switchyard area, the channel’s depression cut through the Tampa formation, the D-zone of the Suwannee formation, and into the Suwannee itself. The absence of the D-zone in this area meant that the water under artesian pressure in the Suwannee formation would escape by flowing horizontally through the formation until it reached the edge of the old river channel. At that point the water would pass through the porous sands and gravels filling the channel and then flow into the second stage cofferdam area through its unprotected side on the east bank. The old channel proved to be the source of 60 to 70 percent of the water pumped from the second stage area.

II

Plaintiff timely appealed Change Orders Nos. 19 and 20 to the Corps of Engineers Claims and Appeals Board.6

The Board held that the material variations in estimated quantities of water experienced were not due to the discovery of subsurface, physical conditions at the site materially differing from the description of those conditions as set forth in the contract plans and specifications. Moreover, the Board did not find the existence of unknown physical conditions of “an unusual nature differing materially from those ordinarily *776encountered and generally recognized as inhering in work of the character provided for in the drawings and specifications.” Instead, the Board held, as a matter of law, that a material variation in the estimated quantity of a unit price item, in and of itself, supports an application for relief by either party under the “Changed Conditions” article. It further held that the mutual ground upon which the parties contracted with respect to the amount of water to be encountered was provided only by the estimated quantities stated in the contract, and that such estimated quantities could not be the subject of a unilateral mistake, because neither plaintiff nor any other bidder had the right to substitute his own figure for that provided in the contract.

Having found that it was impossible to estimate the quantity of water to be discovered with any degree of accuracy, that all other subsurface data militated against the contract estimates, and that no effort was made by the Government to invoke the Changed Conditions article until after work on the second stage had begun, the Board made no adjustment for the overrun on the first stage area.

Again, in view of the uncertainty at the time the contract was entered into regarding the amount of water that would have to be pumped from the second stage, the Board held that an overrun of 150 percent of the estimated quantities was not a condition materially differing from those indicated. The effect of this decision was to allow the Government an equitable adjustment in the unit price in excess of the 150 percent overrun. As a result of the Board’s decision, plaintiff was paid the contract price of 40 cents per unit for the first 302,000 units and for an additional 450,000 units (the 150 percent overrun allowed) from the second stage. For the remaining units, plaintiff was paid 8 cents per unit.

In this action plaintiff seeks recovery based on the contract price of 40 cents per unit for all of the units pumped in the second stage during the contract period as extended, less the amount paid.

In its first counterclaim, defendant alleges that the reasonable cost of the pumping was considerably less than the 8 *777cents per unit paid plaintiff, and it seeks recovery for the difference between 8 cents per unit and the reasonable cost of the pumping.7

After the trial commissioner’s report was made and after the briefs and exceptions of both parties had been filed, defendant moved for leave to amend its answer to assert a third counterclaim, in which it alleges that the Board’s decision, allowing plaintiff the 150 percent overrun at the contract price on the water pumped from the second stage area, was erroneous, unauthorized, and contrary to law. Defendant seeks to recover the difference between the 40 cents per unit paid for the 453,000 units of overrun and plaintiff’s actual cost. We ordered that the question of whether defendant should be permitted to file its third counterclaim at such a late date and also the question as to whether it is entitled to recover thereon be argued, along with the other issues in the case.

Ill

We deal first with defendant’s contention that the decision of the Corps of Engineers Claims and Appeals Board, with respect to the overrun in the second stage area, is final and conclusive as a determination of fact, unless it is found to be arbitrary, capricious, or not supported by substantial evidence. Were the question here one of simple fact, as is the situation in many cases arising under the Changed Conditions article,8 defendant’s point would be well taken, but this case involves more than that — it involves the proper legal standard to be applied under that article. Since the ultimate determination of liability turns on a proper interpretation of the provisions of the contract, a question of law is pre*778sented which we are free to answer independently of the Board’s decision.9

As previously stated, the Board held that a material variation in the estimated quantity of a unit price item, in and of itself, constitutes a changed condition. That conclusion is a decision on a question of law that falls within the purview of Section 2 of the Wunderlich Act.

IY

Written in the disjunctive, the Changed Conditions article sets forth two alternatives as the basis for the existence of a changed condition. The first allows an adjustment for the discovery of “subsurface and/or latent physical conditions” which differ materially from those shown on the drawings or indicated in the specifications. This clause has no application here since the specifications were replete with warnings of the water-bearing capacity of the rock strata in the worksite area. For example, Part IY Section 1, Paragraph 1-02 (g), dealing with “Foundation Leaks,” warns that “[sjprings, leaks, and/or aquifers probably exist in the foundation for the gated spillway, powerhouse, and switch-yard, that will require grouting for their control * *

Subparagraph (e) of Paragraph 2-03 of Part IY, Section II, as amended, also reads:

(e) Exploratory', Excavation. It is very likely that fissures, cracks, joints, clay or disintegrated rock pockets and the like may be encountered that will require excavation below the normal foundation grades, of the concrete structures and the core trench of the switchyard embankment.

Likewise, Part IY, Section III, Paragraph 3-05 (h), Cofferdam Grouting, states that “[sjprings, leaks, and/or aquifers probably exist in the foundations for the gated spillway and *779powerhouse and switchyard and connection to existing overflow dike that will require grouting for their control * *

Also, the various core borings at the site of the work, set forth in the logs of the contract drawings and inspected by plaintiff, indicated that the rock in the areas encompassed by both the first and second stage cofferdams possessed a high degree of cavitation in the form of numerous solution pits, channels, cavities and other openings in the limestone. The logs also reported low core recovery and loss of drill water during the core drilling operation, two factors indicating cavities and other openings and revealing that the rock formations would be heavily water bearing. More important, included in the contract drawings was a core boring geologic profile, which reported the existence of the ancient river channel and which revealed that it cut through the Tampa formation and the D-zone into the Suwannee limestone formation. The existence of the old channel as shown by the profile indicated that large quantities of artesian water under considerable pressure would have access to the area to be protected by the second stage cofferdam.

While performing the earlier lock and fixed crest spillway contract, plaintiff had taken various three dimensional photographs to which the attention of bidders for the instant contract was directed by the contract specifications. This Same specification, Part IV, Section II, Paragraph 2-03 (e), as amended, stated that the conditions in the photos “are expected to be very similar in the foundations for the structures to be constructed by this contract * * The photos showed the kind and type of rock encountered, the fissures, cavities, solution channels and pits that had been found in such rock, the presence of artesian water emerging from the rock, and the extensive wellpoint and deepwell dewatering equipment that was required to cope with the water. The conditions ultimately encountered within the excavated area during performance of the work under the contract were in fact very similar to the conditions depicted in the photographs.

In short, it cannot be said that the subsurface or other conditions experienced in the performance of the contract differed materially from those shown on the drawings or indicated in the specifications. As the Board itself found, *780the contract documents, exclusive of the Government’s quantity estimates, “quite accurately foretold the conditions which were actually experienced.”

It is the second half of the operative language of the Changed Conditions article upon which the defendant places its principal reliance in this case. First, it argues that a substantial variation from an estimated quantity, without more, constitutes an “unknown physical condition of an unusual nature differing materially from those ordinarily encountered” and, therefore, amounts to a changed condition. We reject this argument in view of our consistent holdings that to qualify as a changed condition, the unknown physical condition must be one that could not be reasonably anticipated by the contractor from his study of the contract documents, his inspection of the site, and his general experience, if any, as a contractor in the area.10 This court has not held that a substantial variation from the estimated quantities, standing alone, is sufficient to establish the existence of a changed condition under Article 4.

The cases that have applied the Changed Conditions article to substantial variations from the quantity estimated in the contract have done so on the basis that such a material deviation was not reasonably foreseeable. They deal with special situations where (1) the contractor could not have verified the estimated quantities from the contract documents and his investigation of the site; (2) he had no opportunity to investigate the conditions at the site or for other reasons had the right to rely on the Government’s estimate, or (3) both *781parties labored under a mutual mistake as to the accuracy of the Government’s estimates.11

Defendant’s second and main contention is that a mutual mistake of fact existed here, because neither party contemplated nor was able to foresee a material deviation from the estimated quantity of water that was to be pumped.

Although the Board found that it was impossible to estimate the quantity of water to be pumped with any degree of accuracy, our finding, following a de novo trial, is that it was not possible to make a close estimate of the amount of water to be pumped. The latter finding is controlling.

The difficulty of making an accurate estimate of the quantity of water that would be encountered under the conditions that were known to exist was plainly indicated by the specifications, drawings, core borings, and logs referred to above. In addition, two sections of the specifications warned prospective bidders that the risk inherent in that difficulty would fall upon them.

Paragraph GC-3, entitled “Site Investigation and Representation,” provides that:

The Contractor further acknowledges that he has satisfied himself as to the character, quality and quantity of surface and subsurface materials to be encountered insofar as this information is reasonably ascertainable from an inspection of the site, including all exploratory work done by the Government, as well as from information presented by the drawings and specifications made a part of this contract. Any failure by the Contractor to acquaint himself with all the available information will not relieve him from responsibility for estimating properly the difficulty or cost of successfully performing the work.

Paragraph SC-3 of Part III, “Special Conditions of the Specifications,” states that the contractor will be required to complete the specified work in accordance with the contract and at the contract prices, “whether it involves quantities greater or less than the estimated amounts so listed.”

*782Implicit in such language was the admonition that the contract estimates were approximations only and that pumping in excess of the stated quantities could be contemplated and should be considered by the contractor in computing its bid.12 Thus, these two paragraphs of the specifications, coupled with other provisions of the contract, provided ample indication of the potential water problems that would have to be faced.

Therefore, the test to be applied in deciding this case is whether the parties, from the information available to them, should reasonably have expected a material variation in the contract estimates of the water that would have to be pumped. By a clear preponderance of the evidence, plaintiff has shown that both parties should reasonably have anticipated that a substantially greater quantity of water would have to be pumped than the 302,000 units estimated by the defendant for the second stage area and that plaintiff did in fact expect such an ¡overrun and relied on its ,own appraisal of the situation in submitting its bid. Several facts are especially telling.

Plaintiff’s project manager, before submitting the bid on the instant contract, visited the site of the Shepherd Construction Company’s erection of the overflow dike some 1,000 feet from the eastern limit of the proposed switchyard under the contract here in dispute. Although this work was being performed on dry ground inshore from the river rather than on the river bottom and at a considerable elevation above the river, the manager observed that Shepherd was encountering a large quantity of water, mostly from artesian water flowing under pressure from the rock exposed by the excavation.13 Consequently, he concluded that large quantities of ground and artesian water existed in the area.

The experience gained by plaintiff from its own prior lock and fixed crest spillway contract put plaintiff on notice of the water-pumping hazards which would be encountered. *783That contract also required that the work be done in the dry; however, it contained no estimated quantity of water to be pumped on the grounds that “[quantities cannot be estimated and will be determined by job requirements.” By addendum, defendant subsequently inserted in that proposed contract a fixed price of one cent per unit of 1,000 gallons of water pumped. Once work was started on the contract, large amounts of water flowed into the worksite from the numerous cavities in the rock connected to the river and from the highly porous and permeable Tampa formation which underlay the site. It became economically unfeasible for plaintiff even to meter the water pumped at the one cent rate. As a result, plaintiff stopped metering after it reached 4,636,002 units. By its own calculations, plaintiff pumped an additional 1,500,000 units of unmetered water. Although there is no evidence that plaintiff ever attempted to make an accurate estimate of the quantity of water to be pumped from the second stage area, plaintiff fully expected that it would be required to pump more water for the instant contract with its site in the bed of the river than it had to pump under the prior contract performed on the higher elevation on the river bank. It is difficult for us to see how any contractor, with the same experience in the area, could have avoided that conclusion.

Prior to bidding on the instant contract, plaintiff’s project manager spoke to the Government geologist assigned to the Jim Woodruff Project about the geologic and water conditions at the proposed project site. The geologist expressed the opinion that the D-zone would act as an aquaclude to prevent the passage of water, particularly artesian water rising up from the underlying rock. So long as the foundations for the contract structures did not penetrate the D-zone, or where the D-zone was not broken or had not been eroded away by the river, the geologist believed that grouting would sufficiently retard the Inflow of water into the cofferdam areas so that the inflow could be handled by normal pumping. One of the geologic profiles, however, showed the ancient river channel and its penetration of the D-zone and so the geologist concluded that when construction would take place in this area, a serious artesian water condition *784would be encountered because of the absence of the D-zone.

After the bids were opened, defendant’s contracting officer telephoned Mr. Perini to inquire why plaintiff had bid as much as 40 cents per unit for pumping and why its bid for cement grouting increased as the quantity of cement used increased. Mr. Perini replied that plaintiff wished to avoid a recurrence of its unfortunate experience on the first contract, where it was paid only one cent per unit for pumping instead of being paid for grouting at a higher rate. He also stated that on the instant contract plaintiff proposed to force the Government to make a more realistic choice between the comparative costs of pumping and grouting and that plaintiff’s bid was designed to insure that it would be adequately compensated for the dewatering of the site under either method.

Plaintiff’s original dewatering plans for the first and second stage cofferdam areas, submitted more than 8 months before any pumping was performed under the contract, provided sufficient pumping capacity to handle the quantities of water ultimately encountered in each stage.

When plaintiff’s awareness of the general geologic and water conditions prevailing at the site, including the well known fact that the various limestone formations there were part of one of the most prolific and widespread artesian aquifers in the world, is added to the facts recited above, it becomes clear that plaintiff realized that the Government’s estimate was far too low.14

The estimates for Bid Items 2 and 102 were based upon calculations made by an engineer, who was at that time defendant’s Resident Engineer on the lock and fixed crest spillway contract and who later became the Resident Engineer for the instant contract. In form, the method he employed to arrive at the 323,000 and 302,000 units was a reasonable one. Pie first took the number of units ’he 'assumed had been pumped during the spillway contract and divided it by the number of months the pumping was performed to determine *785the number of units per month pumped. He then divided the quotient by the total area in square yards comprising the lock and spillway areas to arrive at an inflow factor of number of units per square yard per month. He nest multiplied this factor by the area in square yards and the number of months estimated pumping time for each stage to determine the estimated units to be pumped in each stage of the instant contract.

However reasonable the formula may have been in outline form, the use of inaccurate figures and the failure to make proper 'adjustments and allowances for factors that should have been taken into account produced a grossly erroneous result. (See findings 23, 24 and 25). The engineer made virtually no allowance for the unmetered water pumped from the lock and fixed crest spillway area, an amount estimated at 1,500,000 units.15 His figures for the areas for the lock and fixed crest spillway, as well as for the first and second stage areas, were inaccurate. The same was true of the actual and estimated monthly pumping figures. And, with regard to the lock and fixed crest spillway contract, no adjustment was made in attempting to compute a typical inflow rate to eliminate those months when little or no pumping was performed.

One of the most serious errors in defendant’s estimate was the failure to make any allowance for the head differential of the water (level of water outside the cofferdam minus water level inside the cofferdam) between the lock and fixed crest spillway in the first contract and the gated spillway and powerhouse in the instant contract. The difference, particularly with respect to the second stage, was substantial to an extent that would foreseeably affect the amount of water to be pumped.

However, the most egregious error made was the failure to make any allowance in the second stage computations for the existence of the ancient river channel lying in the switch-yard area.

But for some of the mistakes described above, the basic formula used by defendant’s engineer would have provided a very accurate estimate of the water actually pumped from *786tbe first stage area and would liave increased tbe estimate for tbe second stage to 8,967,072 units.

Even sucb a revised calculation excludes tbe large quantity of tbe unmetered water that was pumped on plaintiff’s first contract and does not adequately allow for tbe important contribution that tbe existence of tbe old river channel would foreseeably make to the amount of pumping required in tbe second stage protected area.

Except for tbe large quantity of unmetered water on tbe first contract, which defendant made no effort to ascertain, there was available to defendant all of tbe knowledge and information which plaintiff bad at the time its bid was prepared. In addition, defendant bad the benefit of tbe “Definite Project Keport,” which was not furnished to plaintiff and other bidders. That report set forth tbe dimensions of tbe old river channel, and stated that in a conference held in tbe Office of tbe Chief of Engineers on April 29,1947, “it was decided that a steel sheet piling cutoff wall would be used in lieu of a grout curtain in this trough.” However, tbe cutoff wall was omitted from tbe contract. Instead, tbe contract drawings required plaintiff to provide a U-shaped cofferdam for tbe second stage with the open side facing and immediately adjacent tbe old river channel. Thus, tbe contract made no provision for a barrier across tbe open side to prevent the substantial inflow of water that foreseeably would emanate from tbe old river channel. Tbe existence of the old river channel was plainly shown by tbe contract borings and was well known to the Government geologist whom plaintiff consulted. Tbe significance of defendant’s error in this regard is demonstrated by tbe fact that tbe old river channel was a source of 60 percent to 70 percent of the water pumped from tbe second stage area. Sixty percent of tbe difference between tbe actual and estimated quantities of water pumped from tbe second stage area amounts to 3,779,290 units of pumping.

Our conclusion that tbe material deviation from tbe estimated quantity of water to be pumped did not result from a mutual mistake of fact is buttressed by our findings, which show that a considerable portion of tbe water pumped from tbe second stage area was due to change *787orders issued by defendant or to directions it gave plaintiff during tbe performance of tbe contract. For the purpose of sealing channels, cavities, or other passages in the rock to prevent the inflow of water into the protected area, the contract made provision for grouting to be performed by plaintiff. The amount of pumping to be done was dependent upon the extent and effectiveness of the grouting, but the specifications left the determination of the amount of grouting to the discretion of the contracting officer and further provided that the location of the grout holes and the amount of pressure grouting required to stop the leakage would be determined by the contracting officer from time to time as the work progressed. In view of the larger exposed area in the second stage work and its proximity to the old river channel, it was reasonable for the parties to assume that the more grout used, the greater would be the volume of cavities sealed. Notwithstanding the existing conditions, the defendant ordered the use of less grout per linear foot of grout hole and per square foot of cofferdam in the second stage area than it did in the first. Moreover, the defendant used a less efficient method by requiring the plaintiff in the second stage area to limit the number of batches that were injected into each grout hole, by progressively ordering the addition of sand to the grout in order to save cement, and by reducing the pressures at which the grout was injected into the rock. But for these procedures which were ordered by the defendant, the amount of pumping in the second stage would have been materially reduced.

By various change orders which included performance by plaintiff of additional work, the Government extended the contract time by 317 days. Pumping was performed on 173 days of the extended contract time, and this necessarily resulted in a greater amount of water pumped. The average number of units pumped per day during the contract period, as extended, was 6,310. When this figure is multiplied by the additional 173 days of pumping, the product indicates that 1,095,263 units of pumping were due to the additional period for which the contract was in force and during which pumping occurred. When the *788amount of additional pumping so attributed to the 173 days is subtracted from the total amount pumped by plaintiff under the instant contract, the remainder is very close to the total number of units which plaintiff pumped during the performance of its first contract.

Thus, the situation presented by the facts before us is one where the estimated quantities contained in the contract were grossly understated by the Government. The contractor recognized that the estimate was unduly low and did not rely upon it. In view of all the information it had available to it, the Government should also have anticipated that plaintiff would encounter and) have to pump more water than was pumped under the prior contract. It is in these circumstances that the Government asked us to find that a changed condition existed and that it is entitled to relief therefor under the Changed Conditions article. We decline, because it has never been the purpose of that article in the contract to protect a party from the results of its own miscalculations. We have so held in many cases involving claims by disappointed contractors.16 The same rule must be applied to the Government, for it is no more entitled to use its faulty estimate as a basis for invoking the benefit of the Changed Conditions article than is a careless contractor who makes an improvident bid.

V

In summary, we hold that the criteria to be applied in determining the existence of a changed condition constitute questions of law, and that since the facts of this case establish that a material variation from the erroneously computed estimated quantities was foreseeable, the defendant is not entitled to an equitable adjustment in the contract unit price on the ground that a changed condition existed. Judgment is therefore entered for plaintiff in the amount of $1,870,661.76, which represents the contract price of 40 cents per unit of *789water pumped during the contract period, as extended, less the amount that has already been paid to plaintiff.

As a result of our decision, the question as to whether defendant should be permitted to file its third counterclaim at such a late date is no longer of any significance in this case. Accordingly and without intimating what action we would take on a motion to file such a belated counterclaim in a case where the delay might be prejudicial to the plaintiff, it is ordered that defendant’s motion for leave to file its third counterclaim is granted, and that defendant’s first, second, and third counterclaims are hereby dismissed.

FINDINGS OF FACT

The court, having considered the evidence, the report of Trial Commissioner 'Saul Richard Gamer, and the briefs and argument of counsel, makes findings of fact as follows:

1. Plaintiffs are a joint venture consisting of Perini Corporation (formerly B. Perini & Sons, Inc.), a corporation of the Commonwealth of Massachusetts; Walsh Construction Company, a corporation of the State of Iowa; Ralph E. Mills Company, a corporation of the Commonwealth of Kentucky; and Blythe Bros. Company, Inc., a corporation of the State of North Carolina. All of the plaintiffs have, for long periods of time, been engaged in heavy construction work of the type involved in this case. Perini Corporation was the sponsor of the contract herein involved and was in direct charge of all of the joint venture’s operations under the contract. For convenience, plaintiffs, as a joint venture, will hereinafter sometimes be referred to as the “plaintiff.”

2. The contract in this case, entered into with the Army Corps of Engineers, involved the construction of a gated spillway, powerhouse, and switchyard, all to form a part of the Jim Woodruff Dam on the Apalachicola River in Florida. The dam site on the river is near the northwest comer of the town of Chattahoochee, Florida, about 1,000 feet downstream of the confluence of the Chattahoochee and Flint Rivers in Gadsden and Jackson Counties, Florida, and Decatur County, Georgia. The dam runs east-west across the flow of the river, which is from north to south.

*7903. The purpose of the work to be performed under the contract consisted of the further development of the site as a project to provide flood control, navigation, and power on the Apalachicola River and its tributaries. The contract herein was, in order of time, the third major contract for the work at the site. The first had been let to the Shepherd Construction Company for the construction of an earth overflow dike on the east bank of the river beginning at a point approximately 1,000 feet from the river and extending approximately 2,400 feet in an easterly direction. The second was a contract for the construction of the lock and fixed crest spillway, which contract had been let to, and was being performed by, plaintiff at the time it received the invitation for bids on the instant contract. This work was performed in the dry within two earthen cofferdams. The fixed crest spillway was constructed on the west bank of the river and the lock was contiguous to it and adjacent to the river.

4. (a) Under the terms of the instant contract the work was to be performed in two separate cofferdam stages in order to avoid interrupting the riverfiow and to permit navigation during the construction period. Both the gated spillway and the powerhouse were to be located in the river but were to be constructed in the dry within the cofferdams.

(b) The work to be performed under the first stage consisted of diverting the flow of the river from the western half of its normal channel by constructing a roughly “U”-shaped cellular steel sheet pile cofferdam from the west bank of the river to approximately midstream, the arms of said cofferdam extending inshore to connect into the river wall of the lock by earth dikes. Following the construction of the cofferdam, there was to be constructed the first 13 monoliths of the gated spillway, after which the cofferdam was to be removed. In addition, the first stage of work included the construction of a major portion of the switchyard embankment on the east bank of the river, as well as the remaining portion of the overflow dike connecting the eastern end of the switchyard embankment with the westernmost end of the overflow dike constructed by the Shepherd Company. This portion of the first stage work did not require cofferdam protection.

(c) The second stage work consisted of (1) constructing a similar type of roughly “U”-shaped cofferdam across the *791eastern half of the river (using the steel sheeting extracted from the first stage cofferdam) in order to divert the river-flow through the portion of the permanent structure completed on the first stage, (2) constructing the remaining four monoliths of the gated spillway, the powerhouse, and the remaining portion of the switchyard embankment needed to effect closure with the portion of the switchyard embankment completed under the first stage.

(d) Both cofferdams were of similar construction and consisted of a series of earthfilled circular cells. Each cell consisted of 126 steel sheet piles, was approximately 50 feet in diameter, and was connnected to the cell next to it by two steel sheet pile walls, called connecting diaphragms, in the shape of arcs. Each cofferdam had an open end of the “U” without cells where it met its respective bank of the river. The west side of the first stage cofferdam was formed by the west bank of the river, and the oast side of the second stage cofferdam was formed by the east bank.

5. On November 21,1951, the Mobile District of the Corps of Engineers issued invitations for bids for the work under the instant contract, to be opened January 8, 1952 (later postponed to January 16, 1952). At this time plaintiff was performing work within the fixed crest spillway and lock cofferdams (under the second contract) and was pumping water from both so that work could be performed in the dry.

6. (a) The invitation requested bids to be submitted on the basis of a price schedule containing 153 pay items, most items being based on a unit price, some however being based on a lump sum. The schedule generally divided the various bid items for the work into two categories pertaining to the first and second stages, i.e., “Gated Spillway”, being the first stage, and “Powerhouse and Switchyard”, being the second stage. Items 1 through 74 were placed under the heading “Gated Spillway”, and Items 101 through 179 were placed under the heading “Powerhouse and Switchyard.”

Bidders were required to submit separate unit prices for pumping a unit of 1,000 gallons of water for the first and second stage cofferdam areas to be set forth as Bid Items 2 and 102, respectively. They were also required to include the cost of furnishing, installing, maintaining and removing *792the unwatering system for each cofferdam area in separate lump sum bid items, namely Bid Items 1 and 101, respectively. In this connection, Paragraph 1-05 of Section 1 of Part IY of the Specifications, as amended, provided:

1-05. Payment : (a) Gfeneml. Payment for constructing and maintaining the cofferdams and other protective structures, * * * for removing steel sheet piling to be furnished by the Government from cofferdam and transporting and storing that piling * * * for furnishing, installing, maintaining, and removing the unwatering system: * * * and for all other items required for the construction, maintenance, and removal of cofferdam structures and keeping the protected areas in an un-watered condition for which payment items have not been provided will be made at the lump sum price for Item No. 1, “Cofferdam Protection for First Stage Construction, Dam,” or Item No. 101, “Cofferdam Protection for Second Stage Construction, Powerhouse”. * * * Payment for pumping water from within the protected areas, resulting from leakage of the foundation, from rains, from water used by the Contractor for curing and cutting concrete and for cleaning and treating the foundation will be made under payment Item No. 2, “Pumping from Protected Area for First 'Stage Construction, Dam,” and Item No. 102, “Pumping from Protected Area for Second Stage Construction, Powerhouse.” Payment will be made on the 1,000 gallon basis as measured 'by meters or other approved means to be installed on the discharge pipes from the protected areas. Measurement for payment of water pumped will begin with the first unwatering operations and will be continued as long as pumping operations are considered necessary by the Contracting Officer for efficient construction operations. The Government reserves the right to specify when pumping operations shall be increased, decreased, or discontinued. No payment will be made for pumping from the protected areas after the time fixed for the completion of the contract plus any duly authorized extensions thereof. * * *

(b) Item 2 of the Unit Price Schedule for the gated spillway covered “Pumping from Protected Areas for First Stage Construction, Dam”, defendant’s “Estimated Quantity” therefor being set forth as 323,000 units of 1,000 gallons each. Item 102 of the schedule for the “Powerhouse and Switch-yard” covered “Pumping from Protected Area for Second Stage Construction, Powerhouse”, defendant’s “Estimated *793Quantity” therefor being set forth as 302,000 units of 1,000 gallons each.

(c) Paragraphs SC-1, 2 and 3 of Part III, “Special Conditions” of the Specifications provided:

SC-1. Commencement, Prosecution, and Completion : The Contractor will be required to commence work under this contract within 30 calendar days after date of receipt by him of notice to proceed * * * and to complete the entire work not later than 1,000 calendar days after the date of receipt of notice to proceed. * * *
SC-2. Liquidated Damages: (a) Dwnages. In case of failure on the part of the Contractor to complete the entire work within the time fixed in the contract or any extensions thereof, the Contractor shall pay the Government as liquidated damages the sum of $500.00 for each calendar day of delay until the entire work is completed or accepted.
$ ‡ ‡
SC-3. Items oe Work: A brief description of each item and the estimated quantity thereof are shown in the schedule attached to the hid form and listed in Article I of the contract. Within the limit of available funds, the Contractor will be required to complete the work specified herein in accordance with the contract and at the contract price or prices, whether it involves quantities greater or less than the estimated amounts so listed.

7. Since the proposed work was to be performed in the dry, and a substantial portion of the project was to be built in a river bottom, the water carrying potential of the rock that had to be excavated for the foundations of the structures, as well as the water conditions that prevailed in the foundations, were matters of importance both as to the quantity of water that would have to be handled in order to perform the work in the dry, and the difficulty that such water would impose upon the performance of the work. Defendant, through exploratory drilling, had obtained specimens of rock from the project site, as it had for the earlier lock and fixed crest spillway contract, all of which were made available for examination by bidders on the instant contract.

8. Although it was recognized that it was not possible to foretell with anything approaching exactitude the amount of water that would be required to be pumped, nevertheless plaintiff’s responsible officials, for various reasons set forth *794below, reasonably concluded that defendant’s estimates of 323,000 and 302,000 units set forth for the pumping Items 2 and 102, respectively, were unduly low.

9. It was common knowledge by informed persons, and well known 'by both plaintiff’s and defendant’s responsible officials, that the rock at the site, especially the layer known as the “Tampa” formation into which plaintiff would have to excavate and in which the required foundations would rest, was of high water-bearing quality.

The various materials and strata at the site were, in the order descending from the bottom of the river, known to be as set forth below:

(a) The first layer was overburden consisting of bedded, sandy clays, silts, and sands and gravels which lay uncon-f ormably upon the underlying rock.

(b) Underlying the overburden was an earthy limestone stratum, being the Tampa formation, which belongs to the Miocene period. The thickness of this formation at the site ranged generally from 15 to 30 feet. This was the rock in which plaintiff was to excavate for the powerhouse and gated spillway foundations. The formation was highly porous and permeable, possessing a large number of channels, pits, and cavities, many interconnected. Such pits, cavities and channels result from the soluble nature of the materials composing the stratum. They were created by the erosive and solution effect of artesian waters passing through the rock, and are therefore sometimes referred to as “solution pits”, “solution cavities”, or “solution channels.” The top elevation of the Tampa formation was uneven, varying, at the site involved in the instant contract, from approximately 0.0 feet to 25 feet above sea level.

(c) The third layer was a stratum of limestone and dolo-mites estimated to range between 150 and 200 feet in thickness. This stratum is known as the Suwannee formation and belongs to the Oligocene period. The contact between the Suwannee formation and the overlying Tampa formation is an irregular one generally marked by the presence of light gray clay beds and occasional sand lenses. At the uppermost part of the Suwannee formation immediately below the contact between the Tampa and Suwannee formations, there existed a relatively thin layer of dense sandy limestone the *795thickness of which varied from 8 feet to as little as 1.3 feet, referred to as the “D-zone.” The D-zone, because of its density, acts in many places as at least a partial obstacle, or “aquaclude”, to the passage of water from the Suwannee to the Tampa formations. The Suwannee formation also possessed numerous interconnected channels and cavities and, apart from the D-zone, had a higher permeability and porosity than the Tampa formation. The Suwannee formation also contained a so-called “coquina” limestone composed of shells and remains of animals (“microcoquinas”). The “co-quina” rock is a highly water-bearing part of the Suwannee limestone. It has very high porosity and permeability and has the capacity to deliver large quantities of water.

(d) The fourth layer, known as the Ocala formation and belonging to the Upper Eocene period, consisted of a stratum of cream-colored, very porous, permeable limestone, sometimes dolomitized, and estimated to approach 300 feet in thickness. The Ocala formation possessed a more highly developed system of interconnected solution channels, cavities and pits (such pits and cavities also sometimes being referred to as “vugs”) than either the Tampa or Suwannee formations because a greater volume of artesian water flowed through the formation. The Ocala formation also contained some of the very permeable coquina limestone.

10. The Tampa, Suwannee and Ocala limestone formations at the site were each part of the well-known and much publicized artesian aquifer system which underlies almost all of Florida. It is known as the most prolific and widespread artesian aquifer in the world. This aquifer is the source of approximately 90 percent of the public and private water supply for the State of Florida.1

*79611. The Tampa, Suwannee (including the D-zone), and Ocala formations in the area of the site were not horizontal, but instead, commencing upstream, dipped in a downstream direction, with the result that each formation intersected the river upstream of the site where they were progressively exposed in shingle-like fashion, and were recharged with river water which passed through the numerous solution passages and channels in each formation to gain access to the excavated area at the site. The upstream recharging of the rock produced an artesian water condition at the site.

While all three formations contained artesian water under considerable pressure, the artesian water pressure in the Suwannee formation was greater than that in the Tampa and the pressure in the Ocala formation being even greater than in the Suwannee.

12. (a) Prior to construction of Jim Woodruff Dam, the Mobile District of the Corps of Engineers had made an exten*797sive study concerning the proposed project. The report with respect thereto, originally published on August 1, 1946, and revised March 15,1948, was entitled “Definite Project Report on Jim Woodruff Dam.”

(b) Appendix II of Volume II of the report, entitled “Geology”, described the Tampa formation as follows:

15. Stratigraphy and structure. * * *
16. The Tampa limestone is characterized by a wide range of lithologic and textural differences, varying from a soft, granular, earthy limestone through stages of dense, crystalline, fossiliferous, silicified, and dolo-mitic limestone. The variations of this formation are so irregularly distributed that correlations of the many weakened zones and stratigraphic horizons are extremely difficult. Weathering, leaching, oxidation, and solution are in evidence throughout the area investigated, and alternate hard and soft zones were penetrated by the borings. * * *
19. Although the Tampa limestone formation is extremely variable lithologically, it is rather uniform structurally. The beddmg is essentially horizontal, though dipping gently downstream. * * * The relationships between the structural weaknesses of the rock and the presence of cavities and channels are not definitely established, although the assumption is that they are, or were formerly, related. The majority of the solution cavities encountered are filled, although water-bearing open cavities were encountered in some of the borings.

(c) To the east of the proposed powerhouse structure there was an ancient river channel in existence which had been cut into the rock underlying the switchyard area. Prior to bidding on the instant contract, plaintiff was aware of the existence of this former channel. The channel was filled with river sediments consisting principally of sands and gravels possessing a high degree of permeability. It had resulted from the solution and erosion of the relatively soft limestone rock. Its depression cut through the Tampa formation, the D-zone of the Suwannee formation, and into the Suwannee. It was located just east of the open end of the proposed second stage cofferdam, thereby affording an opportunity for water to enter such cofferdam. Paragraphs 22,23, and 24 of *798said appendix made certain references to foundation conditions at the site, including the channel, as follows:

22. Foundation conditions and proposed treatment under each structure, general. * * * It is evident * * * that the same general foundation conditions prevail throughout the site, with the variations in depth of required excavation being the prime exception.
28. With the extreme variations in lithology, degree of leaching, solution, and erosion, an irregular foundation grade is inevitable for the contemplated structure. This is particularly the case in those areas in which the present and former river channels have been able to erode and dissolve the underlying rocks. Although a “saw-tooth” floor of excavation is inherent, the average required excavation to suitable foundation is regarded as not being excessive for a geologic formation of this character and age.
24. The majority of the solution cavities encountered are filled and do not appear to be very extensive laterally. This cavity filling is often a heavily leached, calcareous clay which in most cases has either settled or partly washed away, leaving open portions usually at the top. In effect, this type cavity is nothing more than an extremely leached zone or seam. Other cavities are filled with alluvial material, silt, sand, mica flakes, the original material having been completely removed and replaced. The treatment of these cavities by grouting and dental methods will be a considerable, but not a prohibitive item. Cut-off curtain grouting will be required to render the foundation impermeable; * * *.

(d) Paragraphs 27, 28, and 32 of said appendix referred to the spillway, powerhouse, and switchyard embankment, including additional references to the former channel, as follows:

27. Spillway. As all but the western 300 feet of the spillway section is to be constructed in the present river channel, the depth to suitable foundation material will be substantially greater than that required for the lock and the fixed-crest overflow section. In the bed of the river the top of rock has been considerably lowered by erosion and solution, and is covered by an average of 20 to 30 feet of alluvial sand, silt, and gravel. It is in this, and the adjacent powerhouse section, that deeply incised channels are to be found, and no doubt the greatest variations in the excavation grade. * * * Unlike *799most of the abutment and flood plain areas, there appears to be only a very thin zone of residuum or badly leached limestone before sound rock is encountered. Thus, once the alluvial covering is stripped, sound f oun-dation rock should be available within a few feet. This does not preclude the possibility of cavity mining and dental treatment, as unsound zones are apparent at depths well below otherwise suitable foundation rock. The establishment of a cut-off curtain is likely to entail somewhat extensive remedial treatment.
28. Powerhouse. The powerhouse will be situated in the present river channel near the left bank. Originally this structure was placed on the left bank, but an extensive solution channel or trough in the rock underlying this location indicated the infeasibility of designing and placing any concrete structure across this area. * * * This area is covered by some 30 to 40 feet of alluvial sand, silt, and gravel. * * *
32. Switchyard embankment. The switchyard is located on the left bank immediately adjacent to the powerhouse and will constitute an enlargement of the overflow dike section. It extends about 500 feet along the axis, nearly half of which includes the extensive solution channel or trough referred to in paragraph 28. Here the depth to foundation rock is at a maximum for the entire site, and the minimum top-of-rock elevation is some 80 feet below mean sea level. This deep trough is believed to have resulted from solution and scour of the soft Tampa limestone formation along a former river channel which has subsequently filled with alluvial sand and gravel as the river shifted or changed its course. This degression might well represent the former bed of the Flint Fiver at a time when its juncture with the Chattahoochee may have been downstream from its present confluence. This trough appears to be entirely due to normal solution and attrition rather than to faulting or other structural disturbance, as no break in the lithologic sequence of the formation is apparent. It is believed to be continuous, with the long axis approximately parallel to the flow of the river, for a considerable distance upstream and downstream. * * * The trough varies from about 200 feet wide at zero elevation to about 50 feet wide at elevation 70 feet below mean sea level and its walls slope on angles of 35° to 45°. Geologic section F-F, drawing 22, exhibit 1, shows the details of the trough along its short axis. In a conference in the Office, Chief of Engineers, 29 April 1947, it was decided that a steel-sheet-piling cut*800off wall would be used in lieu of a grout curtain in this trough. * * *

(e) With, respect to the cofferdams, Paragraph 85 of Appendix II stated:

35. Cofferdams. It is believed that the control of cofferdam leakage and unwatering will be quite a formidable task and will necessitate precautionary measures to alleviate the water problems during construction. The water table is high throughout and, although there is a noticeable lag, the overburden is sufficiently pervious to permit a fluctuation with the river. If only the overburden leakage were involved it could perhaps be effectively handled in the flood plain areas by continuous pumping with the consequent lowering of the water table. However, in order to complete the somewhat extensive rock excavation required, cavities which are potential aquifers with connections to the river will be encountered. * * * In fact, all indications point to a difficult cofferdam problem which is likely to entail substantial expenditures to overcome.

(f) The Definite Project Report was not made available to bidders for either the lock and fixed crest spillway or the instant contract.

13. The Shepherd Construction Company’s work for the construction of the overflow dike began approximately 1,000 feet from the east bank of the river at the eastern limit of the proposed switchyard under the instant contract, and extended in an easterly direction away from the river. Thus, the work was to be performed, without cofferdams, on dry ground inshore from and at a substantial elevation above the river.

To gain knowledge with respect to the preparation of the bid for the lock and fixed crest spillway contract, Albert Berry of the Perini Corporation (who subsequently served as project manager for the joint venture in its performance of such contract, as well as of the instant contract), visited the site of Shepherd’s work on several occasions during March, April, and May 1949, and observed the overburden excavation and dewatering problem presented in constructing the impermeable core for the overflow dike. Berry observed that Shepherd was encountering a considerable quantity of water. A substantial quantity of artesian water flowed *801under pressure from the rock exposed by the excavation. In addition, such quantities of water flowed from the overburden into the excavated areas as to require the use of substantial amounts of dewatering equipment and the sandbagging of the slopes of the excavated areas. Although plaintiff did not know the actual quantity of water pumped by Shepherd (nor is such quantity shown by the record herein) , plaintiff concluded, as a result of Berry’s observations, that large quantities of ground water and artesian water existed in the area of that work. Defendant’s estimate of the quantity of water that would have to be pumped by Shepherd during the performance of the contract for the overflow dike was 3,000,000 units of 1,000 gallons each.

14. In the spring of 1949, following the investigation of Shepherd’s operations and priorato bidding on the lock and fixed crest spillway contract, Berry sought and obtained information concerning the geologic and water conditions at the project site from C. O. Nickell, geologist assigned to the Jim Woodruff Project by the District Engineer’s Office, Mobile, Alabama. Nickell and Berry examined certain geologic profiles contained on the contract drawings. These profiles extended across the entire site and depicted the geological conditions not only in the area of the work under said contract, but also those in the area of the future work for the gated spillway, powerhouse and switchyard project. Nickell and Berry discussed the various zones of rock shown by the profiles. Nickell stated that the layer of sandy limestone, referred to on the profiles as the “D-zone”, was the most impervious rock underlying the site and that it would function as an aquaclude which would prevent the passage of water, particularly artesian water rising up from the underlying rock, in all areas where the excavation for the foundations for the structures did not penetrate through the D-zone, or the D-zone was not broken or had not been eroded away by the river. The profiles showed the D-zone as generally extending across the site between elevation —8 and —18. Nickell believed that if the D-zone was not penetrated or had not been eroded away, grouting would retard the inflow of water into the cofferdammed areas sufficiently so that the inflow could be handled by normal pumping.

*802One of the profiles showed the former river channel and its penetration of the D-zone. Nickell concluded that when construction would take place in this area a serious artesian water condition would be encountered because of the absence of the D-zone.

15. (a) The invitation for bids on the lock and fixed crest spillway contract was issued on February 15,1949. No provision for a payment item for pumping or for grouting was contained therein. However, defendant, by addendum, subsequently inserted in the proposed contract a fixed price of 1 cent per unit, a unit consisting of 1,000 gallons of water pumped. There was no estimate of quantity set forth by defendant for this item, it being stated for a number of items, including this one, that “Quantities cannot be estimated and will be determined by job requirements.” Plaintiff did not feel that the 1 cent price would be sufficient to reimburse it for its pumping costs. Accordingly, the difference between the 1 cent contract price for pumping a unit of water and plaintiff’s anticipated costs was distributed to the other pay items.

(b) In performing the lock and fixed crest spillway contract, plaintiff pumped approximately 6,000,000 units from the cofferdam areas involved therein. Numerous cavities in the rock which were connected to the river were encountered. The Tampa formation which underlay the site of the project was highly porous and permeable and possessed a large number of interconnected cavities, solution pits, open solution channels, and other openings. Under the terms of the contract, plaintiff, in order to receive payment for the pumping item, was required to furnish, install and maintain suitable meters to measure the quantity of water pumped. However, because of the cost of maintaining such meters, and of piping the water to them, it was not economically feasible for plaintiff, at the 1 cent rate, to meter all the water pumped.

(c) Of the total quantity pumped, plaintiff metered 3,-660,850 and 975,152 units from the protected areas for the lock and for the fixed crest spillway, respectively, or a total of 4,636,002 units. In addition to the units actually metered, plaintiff pumped approximately 1,500,000 additional units which it did not meter. Defendant’s responsible officials knew plaintiff pumped approximately 4,500,000 metered units *803plus an additional quantity of unmetered water, although, the record does not indicate that these officials knew or attempted to ascertain how much unmetered water plaintiff pumped.

16. The specifications on which bids were requested for the instant contract made various references to the water condition that would probably be encountered at the site, as well as the need for pumping. For example, Part IV, Technical Provisions, Section 1, Cofferdam, Paragraph 1-02, “Protective Structures”, Subparagraph (g) ilFoundation Leaks'1'' provided in part:

(g) Foundation Leaks. Springs, leaks, and/or aquifers probably exist in the foundation for the gated spillway, powerhouse, and switchyard, that will require grouting for their control before the areas to be inclosed by the cofferdams can be unwatered and/or kept unwatered by a reasonable amount of pumping. * * * it is not the intent of these specifications that the cofferdam or adjacent foundations be grouted until there is no leakage into the protected areas as it is expected that the Contractor will perform a reasonable amount of pumping. The Contractor shall clearly understand however, that the Contracting Officer will decide when grouting is to be performed or when pumping will be done and his decision shall be final. * * *

Also, Subparagraph (e) “Exploratory Excavation” of Paragraph 2-03, “Excavation”, of Section II, “Clearing, Excavation, Foundation” of said Part IV, as amended by Amendment No. 1, provided in part:

(e) Exploratory Excmation. It is very likely that fissures, cracks, joints, clay or disintegrated rock pockets and the like may be encountered that will require excavation below the normal foundation grades, of the concrete structures and the core trench of the switchyard embankment. * * *

Subparagraph (h), Cofferdam Grouting, of Paragraph 3-05, Grouting, of Section III, “Exploratory Drilling, Grout Pipe, Drilling Grout Moles, Grouting, and Removed of Objectionable Materials from Below Foundation Grade by Washing”, of said Part IV, provided:

(h) Cofferdam Grouting. Springs, leaks, and/or aquifers probably exist in the foundations for the gated *804spillway and powerhouse and switchyard and connection to existing overflow dike that will require grouting for their control before the areas to be inclosed by the cofferdams can be unwatered and/or kept unwatered by performing a reasonable amount of pumping or work can be performed in the switchyard area. * * * the Contractor will prepare and submit for the approval of the Contracting Officer, a plan for grouting the cofferdam structures but it should be clearly understood by the Contractor that such cofferdam grouting will be performed only if considered necessary by the Contracting Officer in such amounts and in such limits as he may designate to reduce leakage through the rock into the areas to be protected by the cofferdam. It is not the intent of these specifications that such grouting will be continued until there is no leakage through the rock and no attempt will be made to obtain such results. Cofferdam grouting will be performed only to the extent considered necessary to reduce inflow into the protected areas from the rock to a reasonable amount, as determined by the Contracting Officer, that can be removed by pumping equipment of reasonable size and amount. The actual locations of the grout holes and the amount of pressure grouting that may be required to render the rock within the protected areas reasonably free from leakage are unknown and will be determined at the site as construction work progresses. Cofferdam grouting will be performed with cement grout and intrusion grout composed of cement, flyash, sand, fluidifier, and water in proportions directed by the Contracting Officer.

17. (a) Prior to the invitation for bids for the instant contract, defendant conducted various core borings at the site of the work and set forth on the proposed contract drawings logs indicating the information that it had obtained. These logs indicated that the rock in the areas encompassed by both the first and second stage cofferdams possessed a high degree of cavitation in the form of numerous solution pits, channels, cavities and other openings in the limestone. In some instances the logs also reported low core recovery and loss of drill water during the core drilling operation. The low percentage of core recovery indicated the presence of cavities and other openings in the rock, as well as the degree *805to which the rock had weathered and became fractured. The smaller the percentage of recovery, the greater the number of openings or fractures. Loss of drill water indicated that the drill had intersected solution channels, fissures or other openings in the rock which allowed the water to escape.

(b) From the extensive cavitation and the existence of numerous solution pits, channels and other openings in the rock which appeared in the boring logs, plaintiff reasonably concluded at the time of bidding that the rock in the cofferdam areas was porous and would be heavily water bearing.

18. (a) The contract drawings also set forth a core boring geologic profile which revealed the existence of a “V”-shaped trough, or valley, in the subsurface rock line underlying the western half of the switchyard area [between Station 40+25 and 38+50] along the axis of the dam immediately east of the proposed powerhouse area, the trough extending in a north-south direction paralleling the present river channel. This depression in the rock line was the hereinabove referred to “old river channel”. The core logs indicated it to be filled to approximately elevation +30 with permeable sands and gravels, and from approximately +30 to +55 with impermeable sandy clays. The profile indicated that the old channel was approximately 250 feet wide where the core borings were taken, with the lowest point being at elevation —73.5, or approximately 125 feet below the surface of the existing ground in the switchyard area, and approximately 73 feet below foundation grade of —0.5 feet for the powerhouse structure next to it. (However, another core drill hole set forth on the contract drawings located in the old river channel some 30 feet downstream from the axis of the dam showed an even lower point, i.e., elevation —77.7.) The channel was shown as cutting through the Tampa formation and the D-zone, and penetrated well into the Suwannee limestone formation.

(b) The existence of the old channel as shown by the profile indicated that large quantities of artesian water under considerable pressure would have access to the area to be protected by the second stage cofferdam. The water could gain entrance through the open and unprotected side of the *806area adjoining the switchyard area. This would be accomplished by artesian water of the Tampa and Suwannee formations escaping through the open fissures in the rock forming the sides of the channel, and rising up through the permeable sands and gravels filling the channel to the point where it could flow through the unprotected side of the second stage cofferdam. The channel also provided a means for river water to move around and into the inshore ends of the second stage cofferdam.

(c) No provision was made in the contract for the construction of a complete cutoff across the old river channel to prevent river and ground water upstream of the project site from flowing in a north-south direction (parallel to the present river channel) through the permeable sands and gravels in the old river channel into the switchyard area. Instead the contract drawings called for a grout curtain wall in the rock from the east wall of the proposed powerhouse structure and extending east into the channel but only to a point slightly west of the center of the channel (Station 39 + 65). Thereafter the grout curtain wall in the rock began again in the switchyard area at a point (Station 38+70) which was also in the river channel but east of the center of the channel. This left a gap in the grout wall (between Stations 39+65 and 38+70) of approximately 95 feet in the center of the channel. This gap in the grout curtain in the rock (between Stations 38+70 and 39 + 65) was referred to in Paragraph 3-01 of Part IV, Section III, of the Technical Provisions of the Specifications as follows:

* * * except that as shown on the drawings no drilling and grouting for a grout curtain will be required in the switchyard between approximate Stations Nos. 38+75 and 39+65; * * *.

In addition to the partial grout curtain cutoff in the rock, the only direct old channel overburden cutoff shown in the drawings was the switchyard core trench which began a.t almost the east end of the channel (at Station 38+00) and thence extended east (away from the channel). Thus a gap of approximately 268 feet existed in the overburden between the eastern edge of the powerhouse (to the west of the old channel) and the switchyard core trench (to the east of the *807channel) in which there was no provision for a cutoff in the overburden. These gaps would be expected to funnel water moving through either the overburden or the rock upstream and downstream of the switchyard area into the old river channel.

Nor did the contract drawings make any provision for the construction of a cutoff wall in the overburden on the east bank which would connect the inshore ends of the upstream and downstream arms of the second stage U-shaped cofferdam and thereby prevent the water originating from the old river channel and switchyard area from flowing through the overburden into the second stage area. Thus, water in the switchyard area which came up through the old river channel had unimpeded access to the area which the second stage cofferdam was 'attempting to protect.

(d) From the existence of the old river channel underlying the proposed switchyard area, plaintiff, at the time of bidding, reasonably concluded that large quantities of water would have to be pumped from the second stage area which would originate from the channel.

19. (a) By Paragraph SC-7 (a) of the 'Specifications, bidders were to examine the physical cores of the rock that were retained at the time the borings were taken, said provision reading:

SC-7 (a) Subsurface Investigation. Subsurface and foundation conditions have been investigated by bor-ings at the locations indicated on the drawings. The subsurface conditions as shown on the drawings are approximations and the nature of the material and depth of strata are not guaranteed. Cores taken from most of the holes are available for inspection at a warehouse building at Chattahoochee, Florida. Prospective contractors shall inspect the cores and form their own opinions as to the type and character of material to be encountered. It shall be expressly understood that the Government will not be responsible for any deductions, interpretations, or conclusions made by the Contractor from the boring data and that the interpretations of subsurface conditions as contained on the drawings are only diagrams of assumed geologic conditions, and are not to be used by the Contractor as a reliable basis for preparing his estimate.

*808(b) Plaintiff accordingly examined tbe cores and concluded that they did not differ in any material respect from the descriptions and information set forth in the logs contained in the contract drawings.

20. (a) Paragraph 6 of the Invitation for Bids provided:

6. Bidders should carefully examine the drawings and specifications, visit the site of the work, and fully inform themselves as to all conditions and matters which can in any way affect the work, or the cost thereof. Should a bidder find discrepancies in, or omissions from, the drawings, specifications, or other documents, or should he be in doubt as to their meaning, he should at once notify the Contracting Officer and obtain clarification prior to submitting any bid.

(b) Paragraph GC-3 of the Specifications provided:

GC-3. Site Investigation and Representations: The Contractor acknowledges that he has satisfied himself as to the nature and location of the work, the general and local conditions, particularly those bearing upon * * * river stages, * * * physical conditions at the site, * * * the character of equipment and facilities needed preliminary to and during the prosecution of the work and all other matters upon which information is reasonably obtainable and which can in any way affect the work or the cost thereof under this contract. The Contractor further acknowledges that he has satisfied himself as to the character, quality and quantity of surface and sub-surface materials to be encountered insofar as this information is reasonably ascertainable from an inspection of the site, including all exploratory work done by the Government, as well as from information presented by the drawings and specifications made a part of this contract. Any failure by the Contractor to acquaint himself with all the available information will not relieve him from responsibility for estimating properly the difficulty or cost of successfully performing the work. The Government assumes no responsibility for any understanding or representations made by any of its officers or agents during or prior to the execution of the contract, unless (1) such understanding or representations are expressly stated in the contract and (2) the contract expressly provides that the responsibility therefor is assumed by the Government. Representations made but not so expressly stated and for which liability is not expressly assumed by the Government in the contract shall be deemed only for the information of the Contractor.

*80921. Plaintiff’s officials, including Perini, president of the Perini Corporation, and Berry, made the prebid site investigation for the joint venture. A detailed site investigation was also made by plaintiff’s engineers.

22. (a) In the course of performing its work under the earlier contract for the lock and fixed crest spillway plaintiff took various three dimensional photographs to which the attention of bidders under the instant contract was directed by the above-mentioned Paragraph 2-03 (e) of Section II of the Specifications relating to Exploratory Excavation, as follows:

(e) Exploratory Exomation. * * * Three dimensional pictures taken by the Contractor for constructing the lock and fixed-crest spillway showing the conditions that prevailed at foundation level in that work and the type of excavation that was required are available for inspection at the Resident Engineer’s Office, Chatta-choochee [sic], Florida. As conditions are expected to be very similar in the foundations for the structures to be constructed by this contract, the prospective Contractors shall visit the site and see the pictures to determine for themselves the type of excavation that will be classified as “Exploratory Excavation”, and to decide the best manner for performing such work. * * *

(b) The pictures showed the kind and type of rock encountered, the fissures, cavities, solution channels, and pits that had been found in such rock, the presence of artesian water emerging from the rock, and the extensive wellpoint and deepwell dewatering equipment that was required to cope with the water. The conditions ultimately encountered within the excavated area during performance of the work under the instant contract were in fact very similar to the conditions depicted in the photographs.

23. The estimated quantities of water to be pumped from the first and second stage areas under Bid Items 2 and 102 were based on an estimate made in October or November 1951 by Walter C. Knox, who at the time was Resident Engineer on the lock and fixed crest spillway contract and later became the Resident Engineer on the instant contract. Knox based his estimate on the experience gained from the lock and fixed crest spillway contract. The method he employed and the approximate figures he arrived at (within 5-10 percent) *810in using such, method were as follows: He assumed that on that project 4,365,412 units of water had been pumped over a period of 30 months, or 145,524 units per month. Pie then assumed that the lock and fixed crest spillway areas combined were 128,667 square yards. He next divided 145,524 units by 128,667 square yards to arrive at an “inflow factor” of 1.13 units per square yard per month. He then multiplied this factor by 23,100 square yards, representing the area of the first stage of the instant contract, and then again multiplied by 12 months of estimated pumping time for the first stage, to reach an estimate of 313,000 units of estimated pumping for the first stage (1.13X23,100X12=313,000). For his estimate for the second stage pumping quantity, Knox multiplied said inflow factor of 1.13 units by 24,300 square yards, which he assumed was the area of the second stage work, to arrive at an estimated monthly pumping average, and then multiplied this result by 10 months of estimated time for performance for the second stage work, to arrive at a total of 274,500 units for the second stage (1.13 X 24,300 X 10=274,500).

After Knox completed these estimates, the results were communicated to the Mobile District Office where the estimates were increased slightly and inserted in the bid schedule. The basis for the increase is not shown by the record.

24. (a) Knox’s basic pumping estimate method was a reasonable one. However, corrections of certain inaccuracies in his figures, as well as the making of certain justifiable adjustments or allowances, all as set forth below, would have produced estimates which would have been far more accurate than the estimates which were arrived at, especially the estimate for the second stage. Actually, 825,501 units were pumped in the first stage, instead of the 323,000 estimated, and 7,564,610 were pumped in the second stage, instead of the 302,000 estimated (the actual period of time during which such total second stage amount was pumped being considerably longer, however, than the period of time upon which the 302,000 estimate was based).

(b) No allowance was made for unmetered water pumped from the lock and fixed crest spillway area. At least approximately 1,500,000 unmetered units were pumped by the *811end of tbe pumping period on such, contract (February 1952). Knox’s estimate was made November 1951, well prior to such end, at a time when he could not arrive at any such accurate total figure. However, by October 1951 he did know that plaintiff was not metering all the water it was pumping from the lock and fixed crest spillway protected areas. As of that date, he assumed, without checking with plaintiff, that there was approximately 27,050 units of such unmetered water. This estimate was far too low. If a more accurate estimated allowance for such unmetered water were included, on some reasonable basis, up to the estimate date, the inflow factor would have been increased.

(c) The 128,667 square yard figure for the protected areas of the lock and fixed crest spillway was too high. That contract called for the use of earthen cofferdams, or mounds of earth, to provide protection against the river. These cofferdams were from 50 to 70 feet wide at the base. The measurement for the protected area should have been made from the inside base or toes of the cofferdam. Within such base or toes there were only 24,000 square yards in the fixed crest spillway area, and 61,000 square yards in the lock area, a total of 85,000 square yards, instead of the 128,667 square yards that Knox had used.2 Thus, Knox arrived at a smaller rate of inflow factor because the division of a larger square yardage into the amount of water pumped would produce a smaller inflow factor. Correction of the size of the protected area to 85,000 square yards from 128,667 would have increased the inflow factor from 1.13 to 1.71 units per square yard per month, a 52 percent increase.

(d) The 30-month figure in Knox’s estimate as the pumping period on the lock and fixed crest spillway contract, i.e., from July 1949 to October 1951, both inclusive, was incorrect. This period is only 28 months. The use of 30 months instead of 28 contributed toward producing an inflow rate which was too low.

(e) A more accurate or typical inflow rate would be obtained by eliminating months during which no pumping at *812all occurred, or in which relatively small and inconsequential amounts of pumping were performed as compared with the months immediately prior to the time when substantial pumping commenced or after which substantial pumping ceased.

Pumping for the lock area began on June 14, 1949 and ended on May 23, 1952. Excluding the pumping which was performed during the periods June 14, 1949 to March 31, 1950, and March 1, 1952 to May 23, 1952, because either no pumping was performed therein or because the quantity pumped each month in those periods was so much smaller than the quantities pumped each month in the remaining period, the period of substantial pumping for the lock began on April 1, 1950 and extended until February 29, 1952, a period of 23 months. With respect to the fixed crest spillway, all of the pumping occurred within the same time interval of 23 months mentioned with respect to the lock, with substantial pumping taking place during the 18-month period from August 15, 1950 to February 15,1952. These areas, like the first and second stages on the instant contract, may appropriately be considered separately, instead of computing them together, as Knox did, because they were maintained in an unwatered state for different periods of time and the amount of inflow into each area was therefore different. Furthermore, since the lock area was closer to the river and to the work areas for the instant contract and therefore more representative, the inflow rate for the lock area would be the more pertinent. Thus, 23 months rather than 28 months would have been a more realistic appraisal of the pumping time for the lock and fixed crest spillway areas (which would have involved some estimating as to future pumping, since the 23-month period extends through February 1952, while Knox made his estimate in October or November 1951). The use of this smaller pumping period would have further increased the unit of inflow per month factor used by Knox.

(f) In his estimate, Knox used 12 months as the dewater-ing time for the first stage of the instant contract and 10 months for the second stage, a total of 22 months. Plaintiff, in its original job estimate, used 11 and 12 months respectively as the effective pumping periods for the first and sec*813ond stages respectively, a total of 23 months. Plaintiff’s and defendant’s estimates were not far apart (1 month) on the first stage. However, on the second stage, plaintiff’s 12-month estimate (as against defendant’s 10 months) was more realistic in that it allowed more time than on the first stage for the foreseeably more difficult second stage dewatering operation. Defendant instead allowed more time (and a greater unit estimate) for the first stage pumping. Not only was the second stage work adjacent to the old river channel, but it also involved a larger cofferdam and more concrete, with increased formwork therefor, for the construction of the powerhouse.

(g) The 23,100 square yard first stage cofferdam protected area figure used by defendant to which it applied the 1.13 unit per square yard per month inflow factor was too large. Actually (as could be computed from the contract drawings), there were 19,513 square yards in the first stage protected area.

Also, the 24,300 square yard second stage cofferdam protected area figure used by defendant, to which it applied said inflow factor, was too small. Actually (as could also be computed from the contract drawings), there were 28,870 square yards within the confines of the second stage cofferdam.

(h) Knox made no allowance in the second stage area computations for the existence of the old river channel lying in the switchyard area and the important contribution it would foreseeably make to the pumping operation in that area. (Actually, the old river channel area was the source of 60 to 70 percent of the pumped water that did enter the second stage area, or more than the entire amount that entered from the 28,870 square yard second stage area itself.) In view, among other considerations, of the contract borings, which revealed the depth of the channel, i.e., penetrating the D-zone and extending into the water-bearing rock formations at the site, of its lying immediately adjacent to the open end of the cofferdam, and of its being filled with pervious overburden, this omission was probably the most serious error contributing to the inaccuracy of defendant’s second stage pumping estimate.

*814One reasonable method that could have been used to make an allowance for the water contribution of this adjacent switchyard area would have been to include some portion of it from which the interception of water would be necessary to protect the second stage area itself, as a part of the overall area to be dewatered, despite the fact that this additional area itself was not to be excavated. For this purpose, it would have been reasonable to have included a roughly rectangular area of 29,670 square yards within the switchyard area and lying just outside the open end area of the second stage cofferdam (the western side of which would extend from one arm of the cofferdam to the other, the north and south sides extending directly east). This yardage, added to the 28,870 square yard second stage cofferdam figure, would make a total of 58,540 square yards of area that would be considered as requiring dewatering, thus making a second stage operation approximately three times greater than the area to be unwatered in the first stage.

(i) In arriving at his estimate, Knox made no allowance in his inflow factor for a difference in head between the lock and fixed crest spillway, and the gated spillway and powerhouse, sites. This difference would foreseeably affect the amount of water to be pumped. By referring to the contract drawings, Knox could have determined at the time he prepared his estimate in October or November 1951 that the head differential for the protected areas of both the first and second stages of the instant contract would be significantly greater than for the lock and fixed crest spillway protected areas. The lock and fixed crest spillway construction was to be performed entirely on the west bank of the river where normal ground water levels existed. However, both the first and second stages of the instant contract were to be constructed within cofferdams in the bottom of the existing river where the differential head was naturally greater. In addition, the differential head acting on the second stage cofferdam was further increased since, under the contract requirements, plaintiff, in order to maintain navigation in the river through the lock during the second stage construction, would *815have to raise the elevation of the upstream pool to as much as +70 ( + 65 being the minimum) after the remaining half of the river had been cut off.3

On the feed crest spillway, the differential head amounted to approximately 34.6 feet (average upstream pool water level of 51 feet minus average foundation excavation elevation of +16.4 feet), and on the lock approximately 31.3 feet (same water level minus average foundation excavation elevation of +19.7 feet). On the first stage of the instant contract, it was known from the drawings that the differential head would amount to approximately 50.8 feet (average upstream pool water level of 55 feet minus average foundation excavation elevation of +4.2 feet), and on the second stage to about 70.5 feet (navigation water level of 70 feet minus average foundation excavation elevation of —0.5 feet).

25. (a) Had defendant, in employing Knox’s “inflow factor” theory and method (which were basically reasonable) to arrive at first and second stage estimates of amounts of water to be pumped, employed the more accurate figures, and applied the adjustments to take account of at least certain of the additional factors set forth in finding 24, the estimated figures produced, set forth below, would not have been as inaccurate.

Considering first the feed crest spillway and lock contract experience:

(1) The areas actually enclosed by the cofferdams for the fixed crest spillway and for the lock were 24,000 and 61,000 square yards, respectively (finding 24 (c)).

(2) The effective pumping periods, representing the times when significant pumping was performed, were 18 months and 23 months for the feed crest spillway and the lock areas, respectively (finding 24 (e)).

(3) The metered quantities of water pumped from the feed crest spillway and lock areas during the aforesaid effective pumping period for each area was 973,222 and 3,510,186 *816units (of 1,000 gallons), respectively.4 For this purpose, and considering Knox’s lack of knowledge as to the amount thereof, no allowance is made for unmetered water.

(4) The differential heads for the fixed crest spillway and the lock were approximately 34.6 feet and 31.3 feet, respectively (finding 24 (i)).

(5) Dividing the total quantities pumped (par. 3) by the applicable effective pumping period (par. 2), the figures of 54,068 and 152,617 units of inflow per month are derived for the fixed crest spillway and lock areas, respectively.

(6) Dividing the units of inflow per month factor for each area (par. 5) by its respective area in square yards (par. 1), yields 2.253 and 2.502 units per months per square yard for the fixed crest spillway and the lock areas, respectively.

(7) Dividing the inflow factors, expressed in terms of units per month per square yard for each area (par. 6) by the respective differential heads (par. 4), yields inflow factors of .0651 and .0799 units per month per square yard per foot of differential head for the fixed crest spillway and the lock areas, respectively.

Considering next the known factors, at the time of making the estimate, for the first and second stages of the instant contract:

(8) The first stage protected area was to be 19,513 square yards (finding 24(g)). The second stage protected area, to which should be added a reasonable part of the switchyard area overlying the old river channel which would contribute inflow into the second stage area (and the water from which would have to be intercepted to protect the second stage area), would total 58,540 square yards (finding 24(h)).

(9) An estimate of the effective pumping periods for the first and second stages of 11 and 12 months, respectively (being the same periods as were used by plaintiff in its original job estimate), would be reasonable (finding 24(f)).

*817(10) The contract, drawings and specifications indicated that the differential heads acting on the first and second stage areas would be approximately 50.8 feet and ±70.5 feet, respectively (finding 24 (i)).

(11) Under the formula, it next becomes necessary to derive an appropriate figure representing the inflow factor expressed in terms of units per month per square yard per foot of differential head, as was done on the previous contract (par. 7). For this purpose it would be proper to use an appropriate inflow factor from the two developed on the fixed crest spillway and lock contract and to project it to the instant contract after adjusting it, however, for the greater differential heads applicable to the first and second stages on the instant contract. Such inflow factors were .0651 for the fixed crest spillway and .0799 for the lock protected areas (par. 7). However, it would be more accurate to use the inflow factor for the lock for application to both the first and second stage areas rather than the inflow factor for the fixed crest spillway area because the lock, being situated closer to the river than the fixed crest spillway, would reasonably be regarded as approximating more closely the probable inflow factor for the first and second stage areas.

(12) Applying the lock inflow factor to the differential heads acting on the first and second stages, respectively, produces an inflow of 4.059 units per month per square yard for the first stage area (50.8X.0799), and 5.633 units per month per square yard for the second stage area (70.5 X.0799).

(13) Determining the inflow in terms of units per month for the first and second stage areas necessitates multiplying the units per month per square yard figures (par. 12) for each stage by the area of each stage (par. 8), yielding an inflow of 79,203 units per month for the first stage area (4.059X19,513), and 329,756 units per month for the second stage area (5.633X58,540).

(14) Finally, the total estimated quantity of water to be pumped from the first stage area, resulting from the estimated inflow, is computed by multiplying the estimated 79,203 units per month inflow (par. 13) by the 11 months of estimated effective pumping time for such area (par. 9), *818yielding the figure of 871,288 units. Tbe estimated amount for the second stage is similarly computed by multiplying the estimated 329,756 units per month inflow (par. 13) by the 12 months of estimated effective pumping time for such area (par. 9), yielding the figure of 3,957,072 units.

Thus, the total quantity of pumping from the first and second stage, areas arrived at by this estimate is 4,808,305 units.

(b) Had this more accurate application of the basic Knox formula and method been employed, it would have predicted the amount of actual pumping that was done much more closely, although it would still have failed to predict the actual amount that was pumped from the second stage area by a wide margin. For the first stage, 825,501 units were actually pumped. The estimated amount of 871,233 units using the revised method would thus have provided a quite accurate estimate. For the second stage, however, 7,564,610 units were actually pumped, which was much greater than the 3,957,072 units which even the revised method would produce. On both areas, the revised method would have predicted total pumping of 4,828,305 units, whereas a total of 8,390,111 units was actually pumped.

(c) Further refinements in the revised method would, of course, produce even more accurate results. As shown, even the revised method takes no account of the large quantity of unmetered water that was pumped on the first contract. Moreover, and perhaps more important, the addition of 29,670 square yards of the switchyard area to the second stage protected area in an effort to provide some appropriate compensation for the important contribution to the second stage pumping which the old river channel would foresee-ably make is probably too conservative a method since it adds such area to the second stage area and is thereafter treated like the other areas insofar as the existence of a D-zone is concerned. The inflow factors for the first contract are premised on the existence of a D-zone, as are those for the first and second stages of the instant contract. However, the old river channel penetrated the D-zone which acted as at least a partial aquaclude in the other areas, thus foreseeably resulting in a greater inflow from this area than from comparable square yardage in the other areas. An additional allowance *819could justifiably be made in the second stage estimate for the substantial quantity of artesian water which could be expected to flow into the second stage area after emerging from the rock in the old river channel where the D-zone, or aqua-clude, did not exist.

(d) There is no showing that plaintiff used any method similar to Knox’s, as adjusted, refined or otherwise, in estimating the amount of water to be pumped from the first and second stage areas. Nor is there any satisfactory proof as to whether plaintiff even attempted, upon some rational formula or other basis, to make such an estimate or if so, what such formula or basis was. If there was such an estimate or formula, plaintiff failed to advise defendant of either.

(e) The Invitation for Bids contained the following additional (seefinding20(a)) provisions:

3. The right is reserved, as the interest of the Government may require, to reject any and all bids, to waive any informality in bids received, and to accept or reject any, or all items of any bid, unless the bidder qualified such bid by specific limitation.
*****
12. Modification prior to date set for opening bids. The right is reserved, as the interest of the Government may require, to revise or amend the specifications and/or drawings prior to the date set for opening bids. Such revisions and amendments, if any, will be announced by an amendment or amendments to this Invitation for Bids. Copies of such amendments as may be issued will be furnished to all prospective bidders. If the revisions and amendments are of a nature which requires material changes in quantities or prices bid or both, the date set for opening bids may be postponed by such number of days as in the opinion of the District Engineer will enable bidders to revise their bids. In such case, the amendment will include an announcement of the new date for opening bids.

26. (a) Plaintiff reasonably concluded that the Items 2 and 102 pumping estimates made by defendant were too low on the basis of (1) its observations of the water condition that, had been encountered by the Shepherd Company during construction of the overflow dike; (2) the experience it had gained under the lock and fixed crest spillway contract; (3) *820its realization that both tbe first and second stage cofferdams under the instant contract would be subjected to greater head differentials of water than those which prevailed at the lock and fixed crest spillway; (4) its awareness of the general geologic and water conditions prevailing at the site as a result of various articles and other published information that had come into its possession (finding 10); (5) Berry’s discussion with the Government’s geologist, Nickell, which included a consideration of the old river channel (finding 14), and (6) its conclusion, contrary to defendant’s, that more water would be pumped under the second stage than under the first, the second being larger in area and subject to a greater differential head.

(b) Plaintiff anticipated that the quantity of water it be required to pump would exceed the approximately 6,000,000 units it pumped under the earlier lock and fixed crest spillway contract. Plaintiff’s dewatering plans for the first and second stage cofferdam areas, submitted more than 8 months before any pumping was performed under the contract, provided sufficient pumping capacity to handle the quantities of water ultimately encountered in each stage.

(c) Actually, it would not be possible to make a close estimate of the amount of water to be pumped. One important reason therefor was that the specifications left the amount of grouting that would be performed to control the “springs, leaks and/or aquifers” to the discretion of the contracting officer, whose decision with respect thereto was to be final (par. 1-02 (g), finding 16), and further provided that “The actual locations of the grout holes and the amount of pressure grouting that may be required to render the rock within the protected areas reasonably free from leakage are unknown and will be determined at the site as construction work progresses” (par. 3-05 (h), finding 16). Naturally, the more effective grouting that would be performed to control the water inflow, the less pumping would be required. One obvious factor the contracting officer would be expected to weigh in this connection would be the comparative grouting and pumping costs, a decision that could intelligently be made only by viewing actual conditions as they developed at the various locations on the site.

*821Other factors which made the quantity of water existing in the rock speculative and based on unknown contingencies and variables were the sizes of the channels and cracks therein in the particular location, the number which were water bearing (instead of being filled), the amount of interconnection, and the water-bearing capacity of the old river channel.

Further, defendant, by change orders, which included the performance by plaintiff of additional work, extended the contract time by 817 days, which necessarily caused an increased period of time during which pumping took place, resulting in a greater total amount of water pumped.5

27. Plaintiff’s bid for Items 1, 2, 101, and 102 was computed and arrived at as follows:

(a) Prior to bidding on the contract, plaintiff first had prepared two separate estimates of the direct cost of the work under each bid item in the unit price schedule, including Item Nos. 1, 2, 101 and 102. One estimate (the “Job estimate”) was made, under Berry’s supervision, by the job personnel then on the lock and fixed crest spillway contract, and the other (the “Framingham estimate”) was made at plaintiff’s home office in Framingham, Massachusetts.

(b) (1) The Job estimate for Bid Items Nos. 2 and 102 (the items covering the pumping of the first and second stage cofferdams, respectively) gave consideration only to certain costs which, for the purpose in question, were considered as direct costs of performing such work. The particular costs which were thus considered were direct labor for operating and repairing the pumps, insurance and Social Security tax expenses applicable thereto, and such materials and supplies as fuel, power, parts for pump repairs, discharge lines, and hoses. For pumping the first stage under Item No. 2, this estimate computed the direct cost to be $120,447, which was then divided by 323,000 units (the quantity of water defend*822ant had estimated would be pumped from the first stage under that item), to arrive at 37.8 cents per unit of 1,000 gallons as the direct unit cost. For pumping the second stage under Item No. 102, the estimate computed the direct cost to be $122,258, which was then similarly divided by 302,000 units (the quantity defendant had estimated would be pumped from the second stage under that item) to produce a direct cost of 40.5 cents per unit.

The same equipment was used in estimating the unit costs for both the first and second stages.

(2) The Job estimate as so prepared was then reviewed by a group of employees from plaintiff’s home office and certain adjustments made. With respect to the estimated costs under Item No. 2, reductions in some of the individual estimates were made to arrive at a total direct cost of $97,397, which, when divided by the 323,000 units, produced an estimated direct unit cost of pumping of 30.15 cents per unit. As to Item No. 102, reductions were made which lowered the estimate to $98,708, which, when divided by the 302,000 units, produced an estimated direct unit cost of pumping of 32.68 cents per unit.

(3) The rated capacities of the pumps, as set forth on the Job estimate, totaled 18,850 gallons per minute, or 27,144,000 gallons per 24 hour day (18,850X1,440 (i.e., 24X60)). Under actual field conditions of operating efficiency, considering the need for repairs, the avoidance of straining, etc., it is not expected that more than approximately 25 percent of rated capacity will be realized. On this basis, the pumps on which the Job estimate was based would be used to provide pumping capacity of approximately 6,800,000 gallons (or 6,800 units) per day (27,144,000x25% = 6,786,000). Furthermore, about half the pumps were to be used to provide only standby service, so that actually pumping capacity of only 3,400,000 gallons (or 3,400 units) per day was provided for. As it turned out, this would not have been sufficient to handle the quantities of water which were actually experienced during the performance of the second stage pumping, the pumping of which went as high as 20,000,000 gallons (or 20,000 units) per day on 2 days, and over 18,000,000 gallons (or 18,000 units) per day on many others. For instance, during the period April 10 through June 10, 1955, there were *823only 6 days when less than 18,000,000 gallons were pumped.

The pumping operations for the second stage cofferdam began on July 6, 1954 and ceased on July 27, 1956, a period of 24.68 months (as against the 12 months plaintiff estimated). Even assuming all the pumps would be working with none in merely standby service, the 6,800 unit per day pumping capacity would, except for 1 day, have been insufficient every day during the approximately 14% month period from July 27,1954 through October 18, 1955.

(c) (1) The Framingham estimate for Bid Items Nos. 2 and 102 was similarly based on a calculation of certain costs which were considered, for the purpose, as direct costs of performing such work in each stage, divided by the number of units estimated by defendant. However, the particular costs so considered were considerably less than were included in the Job. estimate. Only the items of fuel, power, and some materials and supplies were included. Among other things, no allowance was made for labor to operate the pumps, or for labor to perform repair services thereon (including Social Security and insurance expenses on such labor). As a result, the Framingham estimate arrived at an estimated direct unit cost of 5.6 cents per unit of 1,000 gallons for the first stage under Item No. 2, and 7.2 cents for the second stage under Item No. 102.

(2) Furthermore, the Framingham estimate was based on using a so-called wellpoint system to accomplish the de-watering hi each stage which was even more inadequate to handle the quantity of water that would have to be pumped than were the deepwell pumps upon which the Job estimate was based. Even making no reduction for standby, the well-point system had the capacity, under field conditions, of pumping only approximately 2,400,000 gallons (or 2,400 units) per day (9,504,000 rated capacity X25% =2,326,000 gallons, or 2,326 units).

(3) As did the Job estimate, the Framingham estimate used the same pumping equipment and capacity for the first stage as for the second.

(d) Upon completion of the Job and Framingham estimates, and prior to submitting its bid, the principal representative of each of the joint venturers constituting plain*824tiff met at Chattahoochee, Florida, to reconcile the results of the two estimates of direct cost and to determine what amount should be bid on each of the bid items (including the unit bid prices for pumping under Items Nos. 2 and 102) and what the total bid should be.

The Framingham estimate had calculated a “Total Direct Cost” in dollars and a “Direct Unit Cost”, based on defendant’s estimated quantities, for each of the 153 bid items. The J ob estimate was not that complete, omitting estimated direct costs (and unit costs) for certain items. Furthermore, the Framingham estimate was a still further complete one in that, to its “Total Direct Costs” of $10,880,732 for all 153 items, there had been added the estimated indirect costs, the amounts that should be included for various contingencies, and the amount of the proposed profit. No such figures appeared in the J ob estimate. The contingencies provided for in the Framingham estimate related to such items as “Flood Damage”, “Labor”, materials costs, and freight costs. No contingency provision as such was set forth for any liquidated damages due to an overrun on contract time. During the preparation of plaintiff’s proposal, its representative had advised the Mobile District Office that plaintiff believed that the work under the contract could not be completed within the 1,000 days specified. However, the representative was notified that the proposed contract time could not be changed.

The total of the “Total Direct Costs” and the indirect costs, contingencies, and profit (which estimated profit was set forth as $1,608,784) calculated on the Framingham estimate equaled $14,269,361, which, after an adjustment of an estimated $175,000 as salvage value of certain items, gave a net proposed bid of $14,094,361, or $3,213,629 over and above the $10,880,732 of total direct costs.

(e) The joint venturers, using the complete Framingham estimate, as above set forth, as the basis upon which to proceed in calculating their bid, decided to reduce the proposed figure of $14,094,361 by approximately $200,000. Reductions in the estimated costs of certain indirect expenses (not specifically set forth in the record) were then apparently made which totaled $186,981.40, and which left the figure of $13,907,379.60 as the final amount agreed to be bid.

*825It then became necessary to distribute the difference between the total estimated direct costs as calculated on the Framingham estimate and the agreed bid price among the 153 individual pay items. No set formula was used in distributing these amounts among the various items, the considerations varying from item to item.

As to Items 1,2,101, and 102, the joint venturers recognized that the Framingham estimate of direct costs appeared to be, insofar as Items 2 and 102 were concerned, too low, being 5.6 cents per unit for Item 2, totaling $17,997 for the 323,000 estimated units, as against the original Job estimate of 37.3 cents per unit, and 30.15 cents per unit for such estimate as adjusted (which would total $97,397), and 7.2 cents per unit for Item 102, totaling $21,822 for the 302,000 estimated units, as against the original Job estimate of 40.5 cents per unit, and 32.68 cents per unit for such estimate as adjusted (which would total $98,708).

In their “Total Direct Cost” calculation of $10,880,732 for bid computation purposes, the joint venturers retained the low $17,997 and $21,822 calculations for Items 2 and 102 as components thereof. Actually, the Framingham direct cost estimate for the related Items 1 and 101 (which included furnishing, installing, maintaining and removing the pumps) so far exceeded those calculated therefor by the Job estimate, as adjusted, that Framingham’s total for the four items was greater than the Job estimate’s by $118,983. Various costs omitted from pumping Items 2 and 102 (such as labor to operate the pumps), which produced the low Framingham 5.6 cents and 7.2 cents estimated unit costs, were obviously included instead in the related “Cofferdam Protection” Items 1 and 10-1 (in all probability as a cost of “maintaining” the pumps). These four items were physically grouped together and totaled in both the Framingham and Job estimates. The Framingham total for the four items, which was included in the direct costs totaled by the joint venturers, was $1,020,-451. The Job estimate total for these items was only $901,468. Because of the differences of opinion as to what costs should be included in what item, the total of the four items would be the important figure so far as the bidder was concerned.

In preparing their bid, the joint venturers, although retaining the Framingham figures for the four items as their *826cost base over which to spread an appropriate portion of the $8,026,647.60 indirect cost, contingency and profit figure, decided to increase the Items 2 and 102 unit prices to 40 cents per unit, thus making the price for these items closer to the Job estimate, as amended, and raising substantially the Framingham unit price. Since the record indicates such “spreading” was not done on any formula basis, all of the considerations entering into the selection of the exact 40-cent figure are not clear. While on Items 2 and 102, such bid figure was closer to the Job estimate figures, yet on Item 1, upon which Framingham estimated direct costs of $461,786 as against the Job estimate of only $427,250, the joint ven-turers decided to bid the still higher amount of $550,000, an amount closer to Framingham’s. The following represents both cost estimates for the four items (as shown by the estimate sheets), and the amounts the joint venturers decided to bid with respect thereto:

Thus, insofar as total dollar amounts are concerned, in using the Framingham total direct cost estimates as a base, the joint venturers were accepting an overall larger one for the four items despite the smaller Framingham Items 2 and 102 components, and in raising the unit cost of said components to 40 cents, they were bringing it more closely in line with the higher Job estimate calculations, thus further protecting themselves.

The most important consideration in raising the unit price to as high as 40 cents was to cause the Government to do more grouting and therefore to require plaintiff to do less pumping. Plaintiff’s experience on the lock and fixed crest spillway contract, with its low 1-cent contract unit price for pumping, was that defendant found it less expensive to re*827quire plaintiff to pump tban to grout, resulting in plaintiff’s being required to pump an enormous amount of water at such a low price as to make it unprofitable for plaintiff even to undertake the expense of metering a substantial amount of water in order to obtain some payment therefor. Plaintiff felt that defendant’s grouting estimate was also too low. Thus, plaintiff wished to guard against the situation it had encountered in the lock, where it felt not enough grouting was done at plaintiff’s pumping expense. Plaintiff felt its 40-cent bid price was sufficiently high to prevent a recurrence of this situation. With such a price, plaintiff felt defendant would be put to a realistic choice to grout or to pump water but in either event plaintiff would, under the bid prices for grouting and pumping, be adequately reimbursed.

(f) When the members of the joint venture compared the Job estimate with the Framingham estimate, they concluded that the direct cost of 30 cents per unit of 1,000 gallons, as set forth in the Job estimate, was a realistic estimate of the direct unit cost of pumping. They also decided that such unit price should be used as a base to which overhead, profit, and indirect expenses should be added in arriving at the bid price.

(g) On January 16,1952, plaintiff submitted its bid in the total amount of $13,907,379.60 (par. (e)) which amount constituted the figure for all of the items based upon unit prices multiplied by the estimated quantities (as estimated by defendant) where the item called for such a unit price, plus the lump sum prices on units for which there was no unit price. The executed contract set forth such bid (and contract) figure as the “Grand Total for Gated Spillway, Powerhouse, and Switchyard (Total of Items Nos. 1 through 74, inclusive and Items Nos. 101 through 179, inclusive).”

The “Unit Price” and “Estimated Amount” bid on Items 2 and 102 (as well as Items 1 and 101) were the same as those set forth in the last two columns of the table in paragraph (e).

28. At all times pertinent herein, defendant had sufficient appropriated funds to pay for all the pumping plaintiff performed at the unit price set forth in the contract.

29. Bids were opened on January 16, 1952, and on January 17, 1952, Colonel Wilson, then defendant’s District En*828gineer in Mobile, Alabama, and who was also a contracting officer, telephoned Perini and advised that plaintiff was low bidder. However, he inquired why plaintiff had bid as high as 40 cents per unit for pumping,6 and why plaintiff’s bid for cement to be used in grouting increased as the quantity used increased (instead of the price decreasing as the quantity increased). As to the 40-cent price, Perini stated that plaintiff wished to avoid a recurrence of the situation encountered on the lock contract on which it was paid only 1-cent per thousand gallons for pumping, a rate which defendant fixed, and with defendant consequently preferring to pay for pumping at that rate rather than to do more grouting at a much higher rate; that on this contract, plaintiff proposed to require defendant to make a more realistic choice as to the comparative costs, with plaintiff’s being adequately compensated in either event; and that, if plaintiff was awarded the contract, defendant would have to pay for unwatering the site, and could not expect the contractor to do so. As to the grouting cement, Perini explained that plaintiff had found that, as it grouted at increased foundation depths, the cost increased so that in this particular instance, the cost did not decrease as the quantity increased. At the greater depths, the grout would have to be pumped against greater pressures.

30. The instant contract was thereafter entered into between plaintiff and defendant for the performance of the *829work at plaintiff’s bid of $13,907,379.60, which instrument was dated January 18, 1952, and was numbered DA-01-076ENGr-1141. Article 1 of the contract provided that the work should be started within 30 days after date of receipt of notice to proceed, and be completed not later than 1,000 calendar days after date of receipt of such notice. The notice to proceed was received February 15,1952. The original 1,000 days of contract time accordingly expired November 12,1954. By various change orders issued pursuant to the terms of the contract, either during performance of the work or after the work was completed, the contract time was extended 317 calendar days, thereby making the revised completion date September 25, 1955. The change orders granting the time extensions were issued for various causes, some of which were directly attributable to the quantity of water encountered, but none of the time extensions were granted expressly for the quantity of water pumped. No additional time was ever granted plaintiff because of changed conditions.

31. Plaintiff did not succeed in completing the work to the point where it was accepted as substantially complete until April 8, 1957, 562 calendar days after September 25 1955, the contract completion date as extended. It was accordingly assessed liquidated damages of $500 per day for 562 days, totaling $281,000. This amount has been paid.

32. The encountering of more water than had been estimated operated to extend the period of time which would be required for performance of the work because of the effect it had on other items of work that had to be performed in the area. For example, the work in connection with the preparation of the foundation surfaces for concreting was affected because plaintiff was required to place a greater number of small pneumatic pumps throughout the area, the effect of which not only impeded the orderly and efficient performance of the work, but also made the work more costly. The presence of the excess water also necessitated the construction of additional sumps in which to collect the water.

In instances where water was encountered flowing from cavities in the rock in the foundation areas, the only way the water could be controlled sufficiently to permit placement of the concrete in the dry was to pump directly from each such cavity through a pipe inserted therein. Prior to pouring *830concrete in the area, plaintiff was required to fill the cavity with crushed stone and cover the stone with tar paper. After the concrete had been placed and allowed to set, plaintiff had to return and grout the cavity.

In other areas where the foundation could not be dried sufficiently to place the concrete, plaintiff used strips of sheet metal so placed as to restrict the flowing water to a narrow area leading away from the foundation area. After concreting up to the strips, plaintiff would then return and grout to seal off the flow of water. Such operations impeded the foundation preparation and placement of concrete.

The encountering of a greater amount of water than estimated also limited the type of equipment which could be used to perform the excavation, as well as the locations in which such equipment could be operated. Instead of being able to use faster and more economical power shovels operating directly in the excavation area, plaintiff was, in order to have a dry stable place to work, compelled to rely on cranes possessing long booms with dragline or clamshell buckets operating from berms extending around the perimeter of the excavated area, or from the top of the cofferdam cells, all of which had the effect of making the excavation a slower and more inefficient operation.

The additional quantity of water also meant that in order to control the water, plaintiff continually had to relocate the deepwell and pneumatic pumps as the work progressed.

33. Articles 4 and 6 of the Contract relating to changed conditions and disputes provided:

4. Changed Conditions. — Should the contractor encounter, or the Government discover, during the progress of the work subsurface and/or latent physical conditions at the site materially differing from those shown on the drawings or indicated in the specifications, or unknown physical conditions of an unusual nature differing materially from those ordinarily encountered and generally recognized as inhering in work of the character provided for in the drawings and specifications, the contracting officer shall be notified promptly in writing of such conditions before they are disturbed. The contracting officer shall thereupon promptly investigate the conditions, and if he finds that they do so materially differ the contract shall be modified to provide for any *831increase or decrease of cost and/or difference in time resulting from such, conditions. If the parties fail to agree upon the adjustment to be made, the dispute shall be determined as provided in Clause 6 hereof.
❖ * * ❖ *
6. Disputes: Except as otherwise provided in this contract, any dispute concerning a question of fact arising under this contract which is not disposed of by agreement shall be decided by the contracting officer, who shah reduce his decision to writing and mail or otherwise furnish a copy thereof to the contractor. Within 30 days from the date of receipt of such copy, the contractor may appeal by mailing or otherwise furnishing to the contracting officer a written appeal addressed to the Secretary, and the decision of the Secretary or his duly authorized representative for the hearing of such appeals shall be final 'and conclusive; provided that, if no such appeal is taken, the decision of the contracting officer shall be final and conclusive. In connection with any appeal proceeding under this clause, the contractor Shall be afforded an opportunity to be heard and to offer evidence in support of its appeal. Pending final decision of a dispute hereunder, the contractor shall proceed diligently with the performance of the contract and in accordance with the contracting officer’s decision.

34. Because the contract time was extended only to September 25, 1955, payment for pumping after that date was, in accordance with Paragraph 1-05 of Section 1, Part IV, of the Specifications (finding 6(a)), not claimed by plaintiff, nor was any such pumping paid for by defendant.

35. Paragraph 1-02 of Section 1 of the Specifications made the following provision for two phases of construction:

* * * the permanent work shall be constructed in two stages in order to provide means for passing the river flow and navigation during the construction period, consequently two cofferdams will be required. * * *

As each cofferdam area was successively enclosed by a cofferdam and unwatered, they became known as the first stage cofferdam and the second stage cofferdam.

36. (a) On February 8, 1952, plaintiff submitted for approval, pursuant to Paragraph 1-03 of the Specifications, a drawing showing its proposed layout of the cofferdam and the unwatering system for the first Stage, which defendant approved with qualification on March 11, 1952. The draw*832ing, captioned “Proposed Cofferdam for First Stage Construction of Gated Spillway”, was accordingly revised and resubmitted, and on March 24, was returned with defendant’s approval subject to a revision which was made.

(b) The drawing indicated use of a deepwell pump at each of the 13 monoliths on alternate sides of the monoliths downstream and upstream (13 deepwell pumps), and two sets of two lines of wellpoints (which are customarily driven into overburden, as distinguished from rock), up on the west bank of the river in the open end of the cofferdam connecting with the earth dikes. In addition, there was shown a line of well-points located along both the upstream and downstream edges of the gated spillway monoliths in the western half of the first stage area. The individual capacities of each of the deepwell pumps were not given, Note 3 on the plan stating only “Capacity of deep wells 300 gpm to 1350 gpm.” Note 1 also stated “All deep wells to extend 20 feet below proposed foundation grade.” The number of wellpoints, their spacing, depth or capacity was not given, Note 2 stating only “Well points to be spaced as deemed necessary.”

37. (a) The first stage cofferdam was constructed in accordance with the approved plan, and at no time thereafter did defendant complain about the manner or methods employed in its construction, or questioned its adequacy or effectiveness.

(b) The pumping system installed by plaintiff for de-watering the first stage area consisted of a combination of deepwell pumps, Moretrench wellpoint systems and open pumping. In this stage, plaintiff installed the bowls of the deepwell pumps at least 6 to 8 feet or more below the foundation design grade at each location by inserting them in holes drilled into the rock in order to insure that the system would function properly and would lower the water to foundation grade. No complaint was ever made by defendant concerning the dewatering system, or the manner in which it was installed, in the first stage area.

38. (a) The pumping operations for the first stage cofferdam began on October 2, 1952 and ended on November 23, 1953, a period of 13.71 months.

(b) Plaintiff pumped 825,501 units of 1,000 gallons each in the first stage area. The amount of water so pumped *833was paid for at the contract unit price of 40 cents per unit without protest from defendant at the time such pumping was performed.

39. (a) With its letter of February 8, 1952 (finding 36 (a)), plaintiff also submitted a schematic drawing for the second stage cofferdam. By subsequent letter of July 21, 1952, plaintiff submitted a drawing entitled “Proposed Cofferdam for 2nd Stage Construction, Gated Spillway and Powerhouse” and which plaintiff stated to be its “final design for the second stage cofferdam.” The dewatering system was described by the letter as follows:

The general unwatering scheme includes deepwell pumps placed in each monolith on alternate sides of the monolith, downstream and upstream. The capacity of these pumps will vary between 300 GPM and 1500 GPM. The pumps will be set in the rock after drilling of holes and they will be installed 20 feet below the proposed foundation elevation of the individual monolith that will be established by your office. Within the area included by the earth dikes it is planned to install a wellpoint system, as indicated, of sufficient capacity to dewater this area. This wellpoint system will be either one, two or three stages depending upon the conditions encountered and the necessity for such systems. Initial dewatering of the cofferdam after closure will be accomplished by open pumping using various size pumps as necessary.

The wellpoint system referred to was indicated on the drawing to be on the east bank of the river (the switchyard area) in the open east end of the second stage cofferdam. Three generally parallel lines of such wellpoints were indicated, the lines varying in length, one being approximately 260 feet long, another approximately 400, and the third approximately 800. As was the case concerning the first stage dewatering system, the capacity of the individual deep-well pumps was not indicated, nor was the number of well-points, their depth, spacing, or capacity. The three notes referred to in finding 39 (b) appearing on the first stage plan were also contained in substantially identical language on the second stage plan.

The basic concept of placing deepwell pumps at alternate sides of each monolith and of using wellpoints “as deemed necessary” at the open ends of the cofferdam on the river*834bank was the same for both the first and second stages. Nineteen deepwells were indicated for the second stage in comparison to 13 for the first stage. Such 50 percent increase corresponded to the sizes of the areas enclosed. The second stage cofferdam area (28,870 square yards) was approximately 50 percent greater than the first (19,513 square yards).

■With corrections unrelated to the dewatering plan, the Eesident Engineer, on March 13,1953, approved this plan.

(b) Three lines of wellpoints on the east bank, totaling approximately 1,460 feet, as shown on plaintiff’s drawing, would, if the individual wellpoints were spaced 2% feet on centers, and with each wellpoint having the capacity to handle 30 gallons per minute, have been able to handle the amount of water which subsequently came from the switch-yard or old river channel area.

(c) The plan for the cofferdam was consistent with the design shown on the pertinent contract drawing (5/1). The design of the cofferdam had the requisite margin of safety against overturning and was consistent with the standard procedures for the type of work involved. Plaintiff’s cofferdam plan was stamped to indicate qualified approval by the Resident Engineer, and was returned with a letter dated March 13,1953, which stated :

This office does not require the transmittal of any additional details or designs of this cofferdam since the responsibility for the cofferdam design is yours.

Thereafter, the second stage cofferdam was constructed in conformity with this plan.

40. On March 19, 1952, plaintiff prepared a 47-page document (entitled “Detail on Proposed Progress Schedule * * *”) essentially for its internal use to indicate anticipated progress payments under the 153-pay items. Under each pay item the total value, the unit quantity, the unit price and the performance period were indicated. On Items 2 and 102, plaintiff used only the Government’s water estimates of 323,000 and 302,000 units for the first and second stages respectively, indicated monthly anticipated progress payments for pumping water based upon such estimates, and also indicated the total amount to be received by plaintiff for pumping water to be the 40-cent bid price multiplied by defendant’s *835water estimates, in the same manner as was set forth in plaintiff’s bid.

41. As a part of the first stage work, plaintiff completed the remaining portion of the overflow dike on the east bank that had been partially constructed under the earlier Shepherd contract. From the portion completed by Shepherd, plaintiff’s portion extended west to the easternmost edge of the switchyard embankment. From there, it extended eastward for 1,000 feet to connect up with the portion constructed by Shepherd, the west end of which was approximately 1,000 feet east of the east bank of the existing river. This acted as an eastward extension of the dam. This construction involved excavation of a “core trench” through the permeable overburden to and into rock and filling the excavation with an impermeable material. The western end of the trench extended approximately 200 feet further east than the overflow dike and extended into the switchyard embankment to the eastern edge of the old river channel. During excavation, large quantities of water were encountered and pumped by plaintiff. In dewatering the core trench excavation, because of the water condition encountered there, plaintiff found it necessary to concentrate the bulk of its pumping equipment near the west end of the excavation and into the center of the old river channel. A total of approximately 165 wellpoints, 7 deepwell pumps, and 6-8 small air operated pumps, was used in plaintiff’s dewatering operation of the overburden in the core trench area. It was necessary to install all seven of the deepwell pumps near the west end and to keep them in constant operation in order to complete the excavation of the overburden in this area. A pronounced increased flow through the sand and gravel, and considerable slumping of the excavation slopes on the west end of the core trench excavation occurred on July 25, 1953, during a high water stage in the river even though the east bank of the river was approximately 300 feet away and west of the area. On two occasions, with the river at a lower level, considerable slumping of the overburden slopes at the west end of the core trench occurred due to the heavy water flow from the sand and gravel. This occurred during two short 15-25-minute power failure periods which prevented pumping. To *836control the water, plaintiff then drove sheet piling along the west end of the core trench and added a Moretrench pump as well as 25 wellpoints in this area. However, the water still could not be controlled until plaintiff redrove the sheet piles to rock. Water pumping operations on the core trench ceased on August 25,1953.

The work involved in pumping the water from the core trench was part of a lump sum item and was, therefore, not metered. On August 4, 1953, however, plaintiff installed a meter on the discharge line to observe the relationship of pump location and river level on the discharge rate.

42. Defendant’s geologist, F. L. Estep, observed the water conditions plaintiff was encountering in the core trench area and concluded that the permeable sand and gravel encountered in the west end of the core trench extended across the old river channel. He further concluded that the un-watering problem in the east bank area of the second stage cofferdam could well be proportional to the exposed area of sand and gravel there located as compared to the sand and gravel which was exposed in the comparatively limited area of the core trench excavation. He reported this to Knox, the Resident Engineer, who instructed him to so inform the Corps of Engineers Construction Division at the District Office in Mobile. A memorandum dated September 11,1953, prepared by Estep and signed by Knox, reported the foregoing facts to the Mobile District Office under the subject “Dewatering Switchyard Core Trench Area and Expected Relation to Dewatering Costs of Second Stage Diversion and Suggested Treatments.” This memorandum contained, among others, the following conclusions:

3. It is, of course, obvious from the above that the greater part of the water was pumped from the westernmost part of the area. The type materials encountered during excavation made for such condition. * * * To the west of Sta. 37+00 sand and small gravel were encountered on top of rock * * *. Every indication is that this same stratum continues westward to and under the river, increases in thickness and in gravel content with corresponding increased depth to top of rock. * * *
4. * * * pumping of water in the switchyard area * * * is, in the opmion of this office, indicative of the conditions expected to be encountered along the east side *837of the second stage diversion even though the river be diverted some 320 ft. upstream from the axis line and 145 ft. upstream from the upstream limits of the upstream retaining wall. This reasoning is based on the assumption that the “old river channel” continues northward past and to the east of the last cofferdam cell which is to be installed in this area * * *.
5. No estimate of the quantity of water that will have to be pumped from this overburden in order to dewater the second stage diversion area is submitted at this time. Based on the quantity pumped from the limited area and higher level for dewatering the core trench the quantity would be extremely high. For this reason it is proposed that some method for control of the water, other than pumping, be resorted to. * * *

The memorandum made two recommendations regarding the handling of water in the second stage cofferdam emerging from the old river channel: (a) the cost and feasibility be investigated of driving to rock a north-south steel sheet pile cutoff wall across the open east end of the cofferdam which would close the opening between the easternmost cells of the upstream and downstream arms; and (b) the feasibility be investigated of grouting the sands and gravels of the old river channel along the east end of the cofferdam with a so-called “bentonite” grouting. It was suggested that a field test of such bentonite grouting be performed at the worksite to determine the feasibility of its use.

43. In the meantime, and during the dewatering operations of the first stage cofferdam and the core trench, Knox, on August 12,1953, wrote to plaintiff the following letter:

As you are aware the quantity for Item 2, Pumping from First Stage Cofferdam Area, has already exceeded the estimated quantity by more than 100%. Your unit price of 400 for items 2 and 102 is quite high. It is more than 100% higher than either the Government estimate or any other bid received.
It appears to this office that in the second stage work, if the estimated quantity of 302,000 units is exceeded that a reduction in the unit price, for the quantity in excess of 302,000 units, would be in order.
Your proposal and comments would be appreciated.

By letter of August 18,1953, plaintiff replied that it could not find in the specifications or contract any provision making such a proposal necessary, and requested the Resident En*838gineer to refer to tbe pertinent paragraph, which, he felt applied. By letter of August 20, 1953, defendant replied that it had no specific paragraph in mind in requesting plaintiff’s proposal and that defendant’s request had been made “in the interests of economy and for the reasons” set forth in its August 12, 1953 letter. The letter further stated that it was assumed that plaintiff did not intend to submit such a proposal; that, if this was so, a confirmation in writing was requested; and that:

In this event what comments would you offer if the Government decided to perform a portion of the pumping in excess of the estimated quantity ?

By letter of August 28,1953, plaintiff requested defendant to clarify its above-quoted inquiry, and on September 1, 1953, defendant replied that “In stating that the Government might decide to perform a portion of the pumping in excess of the estimated quantity it is anticipated that after the estimated quantity has been reached the Government will not be able to take over 100% of the pumping since there will still be some pumping that will have to be performed by your forces in the foundations themselves.”-

On September 10, 1953, plaintiff responded and objected to defendant’s taldng over any part of the pumping. It stated that it had been plaintiff’s “judgment that the pumping requirements would be greatly in excess of the estimated quantity in contract Items 2 and 102, and we brought in pumping equipment to the job accordingly. To take away from us a portion of the pumping would adversely affect our unit cost for the item, although the ultimate effect would depend upon various factors including the amount and character of the pumping taken over by the government. Further, in the event the government takes over a portion of the pumping, we would have to insist upon reimbursement for any increased cost and loss of profit resulting from this action.”

44. By letter of November 6, 1953, plaintiff forwarded to the Resident Engineer a revised proposed dewatering plan for the second stage cofferdam. This revised plan no longer indicated use of 19 deepwell pumps at the monoliths. Instead, a line of 14 deepwell pumps was to be placed at the *839top of the east bank between what had formerly been the first and second lines of wellpoints which were to have been used “as required.” Note I was added as follows: “Deep Well Pumps To Be Installed And Operated Where Ever [sic] Job Conditions Require.” Wellpoints were indicated to be used on the excavation slopes “as required”, and Note 3 stated “Wellpoints To Be Spaced As Deemed Necessary.” The revised plan contained no additional information to what was contained in the original plan concerning the number and capacity of the wellpoints. In each reference to expected wellpoint location on the drawing, the phrase “as required” appeared. Note 1 as to the extension of deepwells to 20 feet below proposed foundation grade was retained.

By letter of November 24, 1953, the Resident Engineer: acknowledged receipt of the plaintiff’s second stage revised dewatering plan and stated:

* * * A comparison of this drawing with the approved JW2-30 drawing indicates that quite a few revisions have been made.
With reference to notes 1 and 1 on the revised drawing, at the present time we do not feel that it would be possible to permit the installation of deep wells which wül penetrate the “D Zone” of the foundation. It is, therefore, requested that the depth and location of any deep wells be discussed with this office prior to installation.

45. (a) By a January 25,1954 “Indorsement” to a memorandum (dated May 11, 1953, pertaining to grouting), the Construction Division Engineer recommended to the Office of the Chief of Engineers that the price for Items 2 and 102 be reduced to 10 cents by the issuance of a change order under Article 4.

(b) By a memorandum dated February 16, 1954, to the District Engineer (a copy of which was, seemingly by error, sent to plaintiff), Knox stated in part :

2. To begin with I’ll admit that I cannot follow the legal minds in their devious meanderings and interpretations of what seems to me to be clear and specific language. The 5th Ind indicates that the “Legáis” think they are clearly within their rights to make a change in price under the terms of Article 4. Personally I can’t see that any existing physical condition has changed. *840As it turned out, our estimate of quantities to be pumped was way off, yes, but no change in physical conditions has occurred. Certainly no “* * * subsurface and/or latent conditions at the site materially differing from those shown on the drawings or indicated in the specifications, or unknown physical conditions of an unusual nature differing materially from those ordinarily encountered and generally recognized as inhering in work of the character provided for in the drawings and specifications, * * *” were encountered. Therefore I cannot see that Article 4 is applicable. However, I bow to the superior intuition and percipience of the legal mind.

Knox speculated whether defendant would not be laying itself open to claims by plaintiff for increased prices in items that underran if the proposed change order was issued, pointing out that plaintiff had already indicated that concrete and excavation for the gated spillway would be considerably less than the amount shown in the contract quantities, a decrease which plaintiff was claiming was “far in excess of normal amount and seriously affects the price bid by the contract for this work” and for which it was claiming compensation. Further, Knox stated:

5. Paragraph SC-3 of the specifications states in part that “* * * the contractor will be required to complete the work specified herein in accordance with the contract and at the contract price or prices, whether it involves quantities greater or less than the estimated amounts so listed.” If we reduce unit prices for overruns then under this paragraph he certainly can increase prices for underruns.

Knox concluded as follows:

6. I freely admit that I think we are paying too high a price for the pumping. Likewise we are paying too low a price for some of the other items in the contract. I don’t think there is any question but that the contract bid is not a balanced bid. However, we accepted the bid and awarded the contract on the bid as submitted.
7. To sum it up I feel that if we reduce the pumping price (claiming changed conditions, when actually there are no changed conditions but an overrun in an item that is almoS impossible to estimate accurately) that any monetary advantage we gain is more than offset by the money that we will almost surely lose when the contractor claims increases on items that will underrun, *841using our own arguments against us. Wouldn’t we be better off leaving things alone and then if he still comes in at the end of the job trying to claim increases for underruns, threaten him at that time with decreases for overruns?

46. In the meantime, the completion of the second stage cofferdam had been effected on May 12, 1954, the work therefor having been commenced on September 14, 1953. The cofferdam consisted of 28 circular cells with connecting diaphragms constructed of interlocking steel sheet piling. Each cell was 50 feet 2 inches in diameter, with the tops of the sheets at approximately elevation +75.

47. (a) At a meeting on May 17,19'54, in the office of the District Engineer in Mobile, Alabama, between representatives of plaintiff and defendant, plaintiff was again requested to agree to a reduction in the contract price for pumping. Plaintiff refused to so agree.

(b) On June 16, 1954, Lieutenant Colonel Harrison, Acting District Engineer, telephoned plaintiff’s Project Manager Berry and requested plaintiff to accept a substituted price of 15 cents per unit of 1,000 gallons for water pumped under Item 102, stating that if plaintiff would not agree to the proposed price substitution, defendant would issue a change order reducing the price. Berry stated that plaintiff would not agree to the offered reduction. On the same date, June 16, 1954, Lieutenant Colonel Harrison wrote plaintiff confirming such conversation. The letter read as follows:

Confirming telephone conversation this date between your Mr. Berry and the undersigned regarding the matter of equitable adjustment of the unit price for “Item No. 102. Pumping from Protected Area for Second Stage Construction, Powerhouse,” under your contract No. DA 01-076-ENG-1141, the price of $0.15 per unit of 1,000 gallons is offered as an equitably adjusted price.
You are requested to give this matter your prompt attention and advise this office in writing as soon as possible, but not later than 23 June 1954, of your acceptance or rejection of this adjusted amount.
In the event a reply is not received by that date a modification in the nature of a change order under paragraph 4 of the contract will be prepared and submitted to you for signature. This modification will provide for the payment under Item No. 102 of a unit price which *842has been determined to be fair and reasonable, after due consideration of all facts and circumstances, including but not limited to the excessive overrun in quantities of Item No. 2, work of identical nature as Item No. 102.

At this time, plaintiff had not as jet performed any pumping or excavation with respect to the second stage cofferdam.

(c) On June 21,1954, plaintiff replied to defendant’s letter of June 16, 1954, pointing out that “to the present time no pumping whatever has taken place with respect to the Second Stage Cofferdam”, declining to accept 15 cents as the unit price for pumping 1,000 gallons, and protesting any action which would change the contract price of 40 cents per unit of 1,000 gallons of pumping to be performed in the second stage cofferdam. Plaintiff relied on Paragraph SC-3 of the Specifications which provided that “* * * the Contractor will be required to complete the work * * * at the contract price or prices, whether it involves quantities greater or less than the estimated amounts so listed”, and stated “Under this provision, it is clear that the contract price for pumping is to prevail independently of whether the quantities of pumping exceed those stated in the estimate.” The letter further stated:

(a) Your reference to Article 4 of the contract indicates that your unilateral change order will be based upon alleged changed conditions. We deny that there has been encountered any subsurface physical condition materially different from those indicated by the contract documents which would have any affect upon the amount of pumping that is to be performed under this contract.
(b) We do not see how you can contend for a changed condition on this contract in view of the pumping that was performed on our first contract for the Lock and Fixed Crest Spillway. In the performance of that work, which was accomplished within the same general area as the present contract, we performed according to measurement 4,636,002 units of 1,000 gallons each, and in addition to this performed approximately 1,500,000 units which were not measured and not paid for.
(c) Under Article 4 of the contract, when the Contractor encounters what he considers to be a changed condition he shall direct the Contracting Officer’s attention thereto in writing before such condition is disturbed. The obvious purpose of this requirement is to enable both parties to be fully acquainted with the physi*843cal aspects of the suspected changed condition. We have now received from you indication that you believe a changed condition exists, and in order to properly protect ourselves it is essential that we know just what this physical condition is supposed to be.
We accordingly request that in your reply to this letter you specifically advise us what physical subsurface condition affecting the pumping has now been encountered which differs materially from the information set forth in the contract drawings or in the specifications.

48. (a) After considering the Estep-Knox memorandum of September 11, 1953 (finding 45), defendant’s officials at the Mobile District Office concluded that the recommended sheet pile wall running north and south across the east bank would not, for various reasons, be feasible. The recommended bentonite grouting test was performed at the work-site. However, it was concluded it would also not be feasible to attempt to grout the sands and gravels of the east bank. Other possible methods of controlling the flow of water from the old river channel into the second stage cofferdam were also considered by defendant at this time, but after investigation these methods were also rejected.

(b) After further consideration, defendant decided to build two sheet pile cutoff walls or diaphragms east-west across the old river channel. These walls, which were extensions from the upstream and downstream cofferdam arms, were to run parallel to the axis of the dam and to each other for approximately 400 feet eastward (the upstream wall extending 400 feet and the downstream wall 440 feet), and were to be driven to the top of rock. Their purpose was to reduce and restrict the area of the old river channel from which water could flow into the second stage cofferdam by preventing river water from flowing around the inshore ends of the upstream and downstream cofferdam arms through the permeable sands and gravels of the old river channel into the second stage cofferdam. Similarly, they would serve to prevent the artesian water rising in the old river channel overburden outside the walls from flowing into the second stage cofferdam.

(c) The contract for the construction of the walls was awarded to the Hardaway Contracting Company on June 21, *8441954. Hardaway started work on the upstream sheet pile wall on July 5, 1954. Plaintiff’s pumping to unwater the second stage cofferdam commenced the following day, July 6,1954. Hardaway completed the upstream wall on August 18, 1954. Thereafter, Hardaway immediately started work on the downstream wall, completing it on September 20, 1954.

(d) The walls did stop the bulk of the water coming from the sources they were designed to cut off. However, they could not prevent artesian water, emerging from the Tampa and Suwannee formations in that portion of the old river chamiel between the upstream and downstream walls, from flowing through the east bank of the switchyard area into the second stage cofferdam area. As it turned out, the completion of the Hardaway walls did not cause a drop in the daily amount of water plaintiff was then pumping. Approximately the same amount was pumped after such construction as had been pumped prior thereto.

49. On December 16, 1954, a meeting was held between representatives of plaintiff and defendant, including the contracting officer, to further consider the matter of the reduced pumping price proposed by defendant. Plaintiff’s representatives were informed that it was defendant’s intention to issue a change order reducing the unit price of pumping for amounts in excess of those estimated. Plaintiff’s representatives protested this proposed action and inquired what subsurface conditions defendant was claiming were encountered which were materially different from those indicated by the contract documents. Plaintiff was advised that the overburden in the old channel turned out to be more permeable than defendant had anticipated.

50. ( a) By letter dated December 27,1954, Change Orders Nos. 19 and 20, both dated December 16, 1954 (and signed by the District Engineer, who was also the contracting officer) , were sent to plaintiff. The letter advised that in the event the determinations set forth in the change orders were not 'acceptable, plaintiff could consider “same as formal decisions of the Contracting Officer from which appeals may be taken within the time and in accordance with” the Disputes Article (Clause 6 of the contract). Provision was *845made on both change orders for plaintiff to indicate its acceptance.

(b) With respect to Change Order No. 19, the contracting officer asserted that a changed condition existed in the subsurface conditions of the first stage construction area (completed November 23,1953) which had required “the pumping of additional quantities of water greatly in excess of those originally estimated” under Item 2. A new pumping item was added to the Unit Price Schedule, Item 2a, entitled “Pumping from Protected Area for First Stage Construction, Dam in Excess of the Original Quantity of 323,000 Units (1,000 Gal. Unit).” The “Estimated Quantity” for this new item was set forth as 502,501 units of 1,000 gallons each. This represented the exact number of units plaintiff had pumped over and above the estimated quantity set forth in the unit price schedule for Item No. 2. The change order fixed the price for pumping such excess of 502,501 units at 25 cents per unit (instead of 40 cents), totaling $125,625.25. No time extension was allowed.

(c) With respect to Change Order No. 20, the contracting officer asserted that a changed condition also existed in the subsurface conditions of the second stage area “requiring the pumping of additional quantities of water greatly in excess of those originally estimated” under Item 102. The order further stated that “it is impossible at this time, when information and experience available are insufficient, to arrive a/t an accurate unit price for the pumping of such excess quantities in the Second Stage Construction and the adjustment of additional time, if any, to be allowed the contractor”, and that “final determination of such new and final price and ■time must be deferred until a date not later than 1 January 1955 * * *.” The order recited, however, that to make progress payments pending final establishment 'of a new unit price, “it is necessary to establish an interim tentative price to be paid prior to 1 January 1955, for the quantities pumped in the Second Stage Construction which are in excess of the originally estimated quantities.” This order similarly added a new pumping item, Item 102a, to the unit price schedule entitled “Pumping from Protected Area for Second Stage Construction Powerhouse in Excess of the Original Quantity of 302,000 Units (1,000 Gal. Unit).” The “Estimated Quan*846tity” for this new item was set forth, as 2,000,000 units. The order fixed the price for pumping such estimated 2,000,000 units at 10 cents per unit, totaling $200,000.

51. (a) On January 18, 1955, plaintiff, by letter to the District Engineer — contracting officer, acknowledged his letter of December 27, 1954, enclosing Change Orders Nos. 19 and 20 and stated that “we do not agree that changed conditions have been encountered * * * which would affect the amount of pumping from protected areas and we protest such a determination.” The letter enclosed appeals to the Secretary of the Army with respect to the change orders, plaintiff treating the change orders as decisions by the contracting officer, with the appeals thus being taken within 30 days thereof. Plaintiff’s two appeals, one from each change order, also dated January 18, 1955, were accordingly forwarded to the Secretary of the Army by the District Engineer. On January 18,1955, plaintiff also sent copies of its two appeals directly to the Secretary of the Army.

(b) In plaintiff’s appeal from Change Order No. 19, plaintiff stated, among other things, that the first stage pumping was completed as long ago as November 25,1953, that plaintiff had therefore “long since been paid 40 cents for all such pumping”, and that the change order would now require plaintiff to repay to defendant $75,375.15 (15 cents per unitX 502,501 excess units); that a changed condition within the meaning of the contract did not exist because the specifications (Par. 2-03 (e)) “warned of the existence of fissures, cracks and joints in the rock which would permit the infiltration of water into the protected area”, as did Paragraph 1-02 (g), which “informed bidders that springs, leaks and aquifers would be encountered”; that the borings showed the rock to possess such characteristics as solution pits and channels as would make it “highly conducive to the entry of water”; that, upon excavation, the rock was found “to possess all of these qualities” and that no material difference was therefore indicated; that the previous experience on the lock and fixed crest spillway contract, under which plaintiff pumped over 6,000,000 units, demonstrated that the water situation on the current contract was not the result of an “unknown condition differing materially from those ordinarily encountered and generally recognized as inhering in *847work of the character provided for in the plans”; that defendant never claimed the amount pumped under the previous contract constituted a changed condition; that, while the quantity of water pumped under the previous contract was “plain notice to the Government and to ourselves that work could not be performed within protected areas at this site without pumping large quantities of water”, nevertheless, as of the date of the appeal, “the quantity of pumping thus far pumped under the present contract for both first and second stages of construction has not even closely approached the approximate amount of 6,136,000 units pumped under the first contract”; that, in view of the earlier experience, plaintiff never considered the estimated quantity of 323,000 units to be “realistic” and that “we fully expected to pump quantities of water far in excess of this amount” for which plaintiff was “legally justified in expecting to receive the contract price of 40 cents per unit for all pumping performed in excess of this quantity” on the basis of Paragraph SC-3 of the Specifications obligating plaintiff “to perform the work required at the contract unit prices whether it involves quantities greater or less than the amounts estimated”; that a review of the events showed defendant had attempted to change the pumping price “in advance of dewatering the second stage area and before the Government was in a position to know or to contend that any increased amount of pumping could be due to rock or to subsurface conditions in that area which differed in any way from those indicated by the drawings and specifications”; and that plaintiff refused to sign the change order and was treating it as a unilateral decision.

(c) The appeal from Change Order No. 20 pointed out that up to November 26,1954, plaintiff had pumped 1,137,620 units over the 302,000 estimated for the second stage, for which it had been paid 40 cents per unit, so that it would now be required to repay to defendant $341,286 (1,137,620 X 30^). In all other respects, it was similar to the appeal from Change Order 19.

(d) On January 25, 1955, plaintiff returned Change Orders Nos. 19 and 20 unexecuted by it and stated that it was treating them as unilateral decisions from which appeals had been taken to the Secretary of the Army.

*84852. (a) By letter of March 29, 1955, the District Engineer and contracting officer forwarded to plaintiff Supplement No. 1 to Change Order No. 20 and advised that plaintiff might “treat the supplement as a formal decision of the Contracting Officer and appeal therefrom as was done on Change Orders Nos. 19 and 20 * * The supplement, dated March 28, 1955, fixed “a new and final price” for Item 102a “for all quantities” from the second stage area “in excess of the original estimate of 302,000 units pumped” at 8 cents per unit, and increased the estimated quantity of such item from 2,000,000 units to 2,150,000 units. Such 8-cent price was based upon a study of plaintiff’s pumping costs. However, the actual calculation thereof is not shown by the record.

53. On April 18, 1955, plaintiff, by letter to the District Engineer-contracting officer, protested Supplement No. 1 to Change Order No. 20, stating “the terms of this supplement are not acceptable to us * * Plaintiff denied the existence of changed conditions, returned the supplement unsigned, and enclosed its appeal of the same date addressed to the Secretary of the Army. The appeal, which was also dated April 18,1955, incorporated by reference “every statement set forth in” plaintiff’s appeal dated January 18, 1955, from Change Order No. 20. On April 18, 1955, plaintiff also sent to the Secretary of the Army a copy of such appeal.

54. On the basis of payments previously made for water pumped prior to the issuance of Change Orders 19 and 20, and Supplement No. 1, deductions were effected by defendant in monthly pay estimates 34 through 42 resulting in a repayment by plaintiff of $608,999.35.

55. (a) By letter of April 29', 1955, the contracting officer forwarded to plaintiff his Findings of Fact, bearing the same date, on plaintiff’s appeal from Ohange Order No. 19. The letter advised that plaintiff could, within 15 days, respond to the findings. The findings stated, among other things, that the “rock and other materials on which the cofferdams were to be founded were known to be pervious” and that “pumping was provided for to prevent leakage from impending construction operations”; that with respect to defendant’s estimate of the quantity of water that would have to be removed by pumping, “definite data for such an estimate were not available, as the foundation material is not *849uniformly pervious, its permeability varying with the number and flow of solution channels it contains, depth of excavation and height of river stage”; that “the amount of anticipated leakage and resulting pumping for each cofferdam was estimated on the basis of subsurface exploration and on the available knowledge of foundation conditions”; and that “the estimated quantities set forth in the unit price schedule were based upon the data and factors known and assumed at the time of the preparation of the contract.” The findings further stated:

It became apparent during the construction within the first stage cofferdam that the materials underlying the river channel and adjacent portions of the foundation were considerably more pervious than the information available at the time of the preparation of the plans and specifications had indicated. Further, the materials comprising the foundation area were found to be of a higher degree of permeability and porosity than anticipated. The materials underlying the first stage cofferdam have proven to be so pervious that it was necessary to pump 825,501,000 gallons of water, or approximately 255% of the estimated quantity of 823 million gallons.
In spite of the fact that such unexpected conditions, or changed condition, were encountered, the consequence thereof was of such magnitude that the increased pumping requirements could not have been anticipated as a contingency provided for in * * * Clause 4 [Changed Conditions] or to which the method of price adjustment as set forth in said Clause 4 would be applicable. Further, that because of the consequences of such conditions as were discovered it was necessary in the protection and prosecution of the work that the contractor be ordered to pump an additional quantity of water in the amount of 502,501 units in excess of the contract estimated quantity or any of the contingencies contemplated under Clause 4 and that such additional quantity of 502,501 units must be considered as coming within the purview of * * * Clause 3 [Changes and Extras]. It is the further finding of the Contracting Officer, therefore, that a new unit price for such excess be fixed in accordance with the provisions of said Clause 3 as an equitable compensation for such additional work.

(b) By letter of the same date, the contracting officer also forwarded to plaintiff his Findings of Fact on plaintiff’s appeal from Change Order No. 20 and Supplement No. 1 *850thereto. The findings were similar to those pertaining to Change Order 19. It was found that there were such changed conditions as to justify a separate unit price for excess pumping, “the quantity of water to be pumped in connection with the Second Stage Cofferdam construction being so greatly in excess of the originally estimated quantities as set out in the contract, that it was impossible at the time of the preparation and submittal of the first part of Change Order No. 20 to determine an accurate unit price to be paid for such excess quantities * * that, on “completion of study of actual costs incurred by the contractor the Contracting Officer determined that a fair and reasonable final unit price for such excess pumping was 8‡ per thousand gallon unit for an estimated 2,150,000 units * * * in excess of the original estimated quantity of 302,000 units.”

56. (a) By letter of May 7, 1955, plaintiff responded to the contracting officer’s findings on plaintiff’s appeal from Change Order 19. Plaintiff stated, in part, that it did not consider the findings “to constitute a factual finding within the meaning of the contract” and that it “is obviously not intended to be a comprehensive factual treatment of the controversy between us * * *”; that the water condition encountered had been indicated by the contract documents and was consistent with the water condition under the prior lock and fixed crest spillway contract as well as with plaintiff’s observations of the work performed under the Shepherd Company contract for the construction of the overflow dike; and that plaintiff therefore denied “that the quantity of water encountered should not have been reasonably anticipated or that it was in fact encountered because of the existence of a changed condition.”

(b) On the same day plaintiff 'also wrote a similar letter to defendant with respect to Change Order 20.

57. (a) Cofferdam grouting for the second stage, which was performed simultaneously with the driving, and before unwatering, was started on March 30, 1954 and completed July 17, 1954. The purpose of this grouting was to seal off the channels, cavities and other passages in the rock so that water could not pass under the cofferdam into the protected area.

*851(b) The grouting operation consisted principally of drilling a casing into the rock, installing a riser into the rock, and pumping grout through the riser. All of this grouting was done only in such quantities and at such locations as defendant ordered.

58. The area exposed by the second stage work was substantially larger than the first stage work. Excluding the area in midstream which was common to both the first and second stage cofferdams due to their overlap, the enclosed square footage of area in the first stage work was 175,617 square feet, while that exposed by the second stage work was 259,830 square feet.

59. (a) Because of the larger exposed area in the second stage work and its proximity to the old river channel, more leakage and infiltration could be expected in such stage than in the first. In addition, the degree of deterioration of the rock increased from a westerly direction to an easterly direction, from which it could be fairly deduced that the rock would be poorer and more permeable in the second stage area than it was in the first. In view of the nature of the rock being grouted, particularly its system of ‘interconnecting channels and cavities, it was reasonable to assume that the more grout that would be used, the greater would be the volume of the cavities that would be effectively sealed off.

(b) Notwithstanding these conditions, defendant ordered the use of less grout per linear foot of grout hole and per square foot of cofferdam area in the second stage area than in the first. In the first stage area 1,021 cubic feet of grout were used per hole, or 31.7 cubic feet per linear foot of hole, whereas in the second stage area, 800 cubic feet of grout were used per hole, or only 23.2 cubic feet per linear foot of hole. Considering the entire area of the cofferdams, in the first stage area 1.49 cubic feet of grout were used per square foot of cofferdammed area, whereas in the second stage only 1.24 cubic feet of grout per square foot of cofferdammed area was used.

Had defendant ordered the same quantity of grout per linear foot of hole, or per square foot of cofferdam area, to be used in the second stage area as it had ordered used in the first, the amount of pumping in the second stage would have been materially reduced.

*85260. (a) The grouting operation on the first stage was more effective in reducing pumping than was the grouting operation on the second stage. On the first stage operation, each hole was started with a so-called “neat” grout consisting of one part Portland cement and five parts of water. Such neat grout possesses a relatively low viscosity and therefore is capable of penetrating the openings in the rock more effectively at lower pressures than a grout containing sand. Thereafter, as the grouting proceeded under constant pressure the grout mix was gradually thickened in consistency by the addition of sand and other additives until the grout became heavily sanded. During performance of the first stage grouting, defendant permitted 500 batches of grout to be pumped into a grout hole in one continuous operation. After such batches had been pumped into a grout hole, plaintiff was ordered to stop in order to allow the grout to solidify, after which defendant ordered further drilling and grouting of the hole.

(b) The manner in which defendant required plaintiff to grout the second stage area resulted in the grout holes becoming “slugged,” i.e., only the immediate area of the hole itself was grouted. This was less desirable than letting the grout flow into the adjacent channels and solution areas in the rock in order to cut off the water that was running through the rock. Such slugging of the grout holes in the second stage area was accomplished by three separate methods of procedure:

(1) Defendant limited the number of batches that were injected in each grout hole in one continuous operation first to 200, and then to 100, instead of the 500 batches which had been permitted in the first stage operation. This reduction in the number of batches from 500 to 100 for the second stage did not represent an improvement of the grouting operation over the first stage, but was simply intended to reduce the use of grout. However, the primary object of the grouting program was to seal off the water by filling the openings in the rock. The greater the volume of grout injected in a grout hole, the greater will be the penetration of grout into the rock. Thus, the reduction of the number of batches of grout in the second stage resulted in a less effective grouting job than that accomplished in the first stage. While it saved grout, it *853failed to retard the flow of water up through the floor of the cofferdam as effectively.

(2) In. an attempt to save cement, defendant progressively ordered the addition of sand to the grout more quickly, with the result that the viscosity of the grout became so great that it could not enter the smaller openings in the rock. Thus, the grout sometimes stopped flowing before it had actually sealed off the cavity.

(3) Defendant reduced the pressures at which the grout was injected into the rock. On the second stage, while the grouting operation on each hole was started with “neat” grout, which was thickened by the addition of sand, plaintiff was not permitted to maintain a constant injection pressure for as long as the grout would move, but instead was required to diminish the injection pressure which slugged the hole. The result was that the grout was not forced into the openings in the rock to the extent it would have been had the higher injection pressure been constantly maintained. Subsequent examination of the grouted rock after it had been excavated from within the limits of the foundations for the permanent structures showed that the grout had not properly penetrated the fissures and cavities.

61. (a) The original contract time was 1,000 days. Notice to proceed was received on February 15,1952 fixing the original completion date as November 12, 1954. In connection with seven “Modification-Change Orders”, some of which involved additional work (such as the construction of the sheet piling cutoff wall at Cell S-8), plaintiff received total time extensions amounting to 317 days, thus extending the original completion date to September 25,1955. Since Paragraph 1-05 of the Specifications provided that no payment would be made for pumping from the protected areas after the time fixed for the completion of the contract plus any duly authorized extensions thereof (finding 6 (a)), plaintiff received no pay for any pumping beyond September 25,1955.

Of the 317 days of time extensions granted by the change orders, no pumping was performed during 144 days. Thus, pumping was performed on 173 days more than on the original contract time basis. Since the total amount of water pumped is directly related to the length of time pumping is going on, the increased project time, during a large part of *854which pumping had to continue, importantly contributed to the total amount of water plaintiff was required to pump on the contract.

(b) With respect to the second stage area, plaintiff, up to July 27, 1956, when all pumping operations ceased, pumped 7,564,610 units. However, within the contract time as extended, i.e., September 25, 1955, plaintiff pumped from the second stage area 6,600,818 units.

62. (a) As the excavation for the second stage progressed downward into the rock, numerous water-bearing solution channels, aquifers, springs, fissures and boils were observed in the rock, which allowed water under pressure to come in from the sides of and from the bottom of the rock cut. On or about September 21,1954,19 springs originating from the foundation rock were counted within an area 2 feet wide and 6 feet long. At another place in the foundation on the same day, water was also observed coming out of the rock in a 14-inch by 3-inch stream, 4 feet below the top of rock. On the same date, there was also in operation a boil which produced approximately 200,000 gallons per day, as well as one which produced close to 100,000 gallons of water per day. On November 30, 1954, two boils discharged quantities of water from the rock with sufficient force to cause the water to spout 6 or 8 inches into the air.

(b) The water which entered the second stage area was both of artesian origin arising from deep aquifers in the rock, and of river origin, coming therefrom by passing through the numerous channels and fissures in the rock. It was not river water entering directly through or under the cofferdam, since the cofferdam had been driven to rock, thus sealing off the flow of river water through overburden. The water passed through a series of interconnecting cavities and channels in the limestone rock. The interconnecting character of the various channels and cavities in the rock beneath the work area was demonstrated by readings from certain observation wells installed by defendant prior to and during the performance of the work, which readings, located in various portions of the work area, showed that the water rose and fell within a short period of time with corresponding changes in the water level of the river itself. This indicated that the limestone underlying the area was extremely porous and *855permeable and permitted the river water to flow through the rock with great rapidity. It also indicated that the Tampa, Suwannee and Ocala formations underlying the site, which are known aquifers, are exposed further up the river and are consequently recharged and influenced by the river levels.

(c) For several reasons, the relatively thin layer of sandy limestone, referred to as the D-zone, and existing in the uppermost layer of the Suwannee formation immediately below its contact with the Tampa formation, was not a wholly effective aquaclude. It could not function to prevent artesian water in the Tampa formation from flowing into the protected area because it was located below such formation. Similarly it could not prevent river water from recharging the Tampa formation some distance upstream of the site and then flowing through the interconnected solution channels, cavities and other openings in the Tampa, thus ultimately entering the cofferdam areas. In addition, the D-zone did not extend as an unbroken layer over the entire site of the work. The absence of the D-zone in any area meant that the water under artesian pressure would escape from the underlying Suwannee formation and rise until it reached an elevation which would equalize its artesian pressure. The most significant area in which the D-zone did not exist was, as stated, in the old river channel. The absence of the D-zone in this area meant that the water under artesian pressure in the Suwannee formation, could, if it found its way through an interconnection, and elude the grout curtain, escape by flowing horizontally through the formation until it reached the edge of the old river channel, at which point it would emerge into the permeable sands and gravels filling the channel and then rise up through these porous sediments to the point where it could then flow into the second stage cofferdam area through its unprotected side on the east bank. The effect was to create an artesian reservoir immediately adjacent to the second stage area. There were, furthermore, several other areas in which there were openings through the D-zone which allowed artesian water from the underlying Suwannee formation to rise up through the permeable Tampa formation and enter the cofferdam areas without first having to enter the old river channel. These openings resulted from natural causes as well as from excavation work and ex*856ploratory drilling. There were, in addition, some interconnected cavities present in some areas in the D-zone itself which would allow water to percolate through.

63. Plaintiff’s two appeal letters of January 18,1955, with respect to Change Orders 19 and 20 (finding 74(a)) as well as its appeal letter of April 18,1955 with respect to Supplement No. 1 to Change Order 20 (finding 77), were duly forwarded to the Corps of Engineers Claims and Appeals Board, and were all consolidated as “Eng C & A No. 865” and plaintiff’s appeal letter of September 17, 1954, with respect to the remedial diaphragm (finding 68(a)), was similarly so forwarded and numbered 971. Both numbered appeals were consolidated for hearing, and hearings were conducted in Washington, D.C., from July 10 to July 24, 1956, at which 1,421 pages of testimony were taken and a large number of exhibits introduced. By agreement, on Appeal No. 865 only the question of whether there was a changed condition was heard, it being understood that no evidence would be presented concerning the sufficiency of the prices established by Change Orders 19 and 20, as supplemented (25 cents and 8 cents per unit for the first and second stages respectively for all units over the amounts defendant had estimated for each stage).

64. (a) On December 13, 1957, the Corps of Engineers Claims and Appeals Board rendered its decision on plaintiff’s appeals. As to Appeal No. 865, the Board found the following material anl pertinent facts, and arrived at the following conclusions: that, considering (1) the specifications (with its warnings of fissures, cracks, etc., in the foundation rock “all of which would be conducive to infiltration of water into the work area in formidable quantities”), (2) the core borings (“showing numerous solution pits, channels and cavities in the limestone” which indicated “a high degree of porosity and foretold * * * a high degree of water intrusion into the work area”), (3) the core boring logs (showing the existence of the old river channel, “with its permeable material, thus providing access for river water moving around the cofferdam and into the protected area,” foretelling “the intrusion of very large quantities of water”), (4) the Government’s Definite Project Report (with its reference to cofferdam unwatering being expected to constitute “quite a *857formidable task”), (5) the experience of the Shepherd Company on its prior contract, (6) prebid discussions between plaintiff’s and defendant’s representatives concerning the old river channel and its breach of the D-zone as well as other factors “which militated against effectiveness of the D-zone as an aquaolude,” and (7) plaintiff’s experience on its own previous contract, it could not be concluded that “the material variations in estimated quantities” were the result of “the discovery of subsurface, physical conditions at the site materially differing from the description of these conditions set forth in the contract plans and specifications”; that “if we examine (exclusive of the estimates themselves) various data, descriptions, prior experiences and other warnings of subsurface, physical conditions to be encountered at the site, we must conclude that they quite accurately foretold the conditions which were actually experienced”; that defendant’s contention that sands in the old river channel proved to be more permeable than it originally anticipated could not be accepted as a basis for declaring a changed condition because there was no proof (1) that tests of such permeability had been made prior to the invitation for bids, (2) that bidders were informed thereof, or (8) that defendant relied thereon in formulating its estimates; that, however, without determining whether defendant’s instant pumping estimates were “based on sound engineering principles,”7 it was plain, being “one point on which the parties agree,” that, “it was literally impossible to estimate in advance and with any degree of accuracy the amount of water to be pumped,” this “statement” being one which “appears in the contracting officer’s findings of fact, and it is frequently alluded to by appellant” (referring to appellant’s brief before the Board) ; that this would be so because “an estimate which would depend upon such imponderables as the duration of the un-watering operation, the permeability of surrounding soils, the frequency and size of artesian sources likely to be intersected, the efficiency of the cofferdams, river stages, and the like, could not be scientifically determined on the basis of *858sound engineering principles”,*8 that, in any event, defendant, “using assumed time and rate of inflow factors * * * came up with estimates which were far lower than would be indicated by the experience on two prior contracts, and missed forecasting actual conditions ‘by the proverbial mile’ ”; but that nevertheless, as a matter of law, a “material variation in the estimated quantity of a unit price item, in and of itself support[s] an application for relief by either party under the ‘Changed Conditions’ article.” On this “final and pivotal issue,” the Board held:

All of the data on subsurface conditions, detailed as it was, foretold nothing more than that, “formidable”, “great”, “substantial”, even “vast” quantities of water would 'be encountered, to use words which appear in the record. These are relative words and they do not purport to provide a precise and mutual ground upon which the parties (and other bidders) could be expected to meet. That meeting place, for better or for worse, is provided only by the estimated quantities set forth in the bidding documents and in the contract.
They represent the translation, in effect the distillation, of all other available subsurface data, and they are offered to all bidders and treated by the Government as facts assumed for the purpose of bidding. If the contrary were true, each bidder would be approaching the same contract on a different and undisclosed basis, and the whole concept of competitive bidding would be destroyed in the chaos which followed.
In this case, for example, appellant was overall low bidder on the basis of estimated quantities printed into the bid form, even though his unit price for pumping was higher than that of others. Would he still have been low bidder had the awarding official applied the quantities which appellant privately had in mind when he submitted his bid, for example, the 2,390% overrun actually experienced on stage two? Having not been furnished with an abstract of bids, we cannot 'say, but the uncertainty and inequity of such a procedure is readily ■apparent. The contractor is not expected to secretly and independently make his own guess as to the quantities of water to be encountered, substitute that guess *859for tbe official guess appearing in the contract, and conclude that the Government is bound thereby.

The Board relied on Peter Kiewit Sons’ Co. v. United States, 109 Ct. Cl. 517, and Chernus v. United States, 110 Ct. Cl. 264, 75 F. Supp. 1018, in support of its conclusion and rejected plaintiff’s contention that these cases were distinguishable “in that they involved mutual mistakes of fact as to the estimated quantities, whereas only the Government was mistaken in this case, appellant having ignored the figures provided in the Invitation for Bids”, stating:

There can be nothing but mutuality of agreement, and hence mutuality of error, concerning anything as clear-cut and as clearly set forth as were the estimated quantities in this case. They could not possibly be the subject-matter of a unilateral mistake, because neither appellant nor any of the other bidders had the right to privately substitute his own figure for that provided, and thus to destroy the “meeting of the minds” essential to any agreement. It is significant that Kiewit and Ohernus both invoke the “Changed Conditions” article in affording relief.
They proceed on the theory that the parties have contracted on the basis of a common understanding, namely, the estimated quantities. When subsurface conditions (in this case2 the quantity of water) are discovered which differ materially from that understanding, the article is for application. Here the rock data was less significant than the estimates, because we are concerned not with an unanticipated rock condition, but rather with an unanticipated quantity of water. Section 1-02 of the specifications anticipated that after construction of the cofferdams and after grouting, a “reasonable” amount of pumping would be required. The estimated quantities defined what would be considered “reasonable”. By that standard, amounts actually encountered were unreasonable.

The Board accordingly held that the “ ‘Changed Conditions’ article is by its terms available for the relief of either party, and that, on the facts presented, it is available to the Government in this case.” It further held, however, that the contracting officer erred in applying the Changed Conditions article “from the very first gallon of water exceeding the original estimates” since the “article itself speaks of conditions ‘materially’ differing from those indicated, and some *860degree of materiality is required before the article comes into play.” The Board therefore held:

In this case, the percentage of overrun which should be permitted prior to adjustment, would obviously be comparatively large because of circumstances unique to this case. Firstly, it has already been agreed and established that it was virtually impossible to estimate the quantity of water to be encountered with any degree of accuracy. It was anybody’s guess, within a very wide range. Secondly, all other subsurface data provided here militated against the estimates actually made, and tended to dilute their reliability. Finally, although a 155% overrun had occurred on stage one, no effort had been made to invoke the “Changed Conditions” article, and it was not mentioned until the stage two work was launched, indicating to some extent the Government’s views on a material overrun under the circumstances. The size of the overrun on stage two is also significant, being 2,390%.
All of the foregoing considered, it would appear that a 150% overrun would be permissible before adjustment in unit price would be in order. What that adjustment should be, was not put in issue before the Board. The appeal is therefore denied, and remanded for necessary action not inconsistent with the foregoing.

(b) As to Appeal No. 971, the Board held, on the basis of the following material and pertinent facts, as follows: that, “when the excavations reached rock, it was possible to see water entering through various solution channels, aquifers, springs, fissures and boils in the sides and bottom of the rock cut” and that “this water was artesian in origin, and not river water entering directly through or under the cofferdam”; that “between 60% and 70% of the water pumped from the second stage area infiltrated from the switchyard area, which was at the landward end of the work site and not within the protection of the cofferdam”; that under a proper interpretation of the contract, the piling was only to be driven to “assumed top of rock” and not to “assumed top of sound rock”, to which only the permanent structures were to extend, and that this conclusion was buttressed by the fact that the Government-furnished piling could not as a practical matter be driven to sound rock; that the piling was in fact driven to the “top of rock” and that “evidence developed at a supplemental hearing following extraction of the piling, fortifies *861this conclusion.” The Board accordingly sustained this appeal.

(c) Such findings of fact as are set forth above and which the Board included in its decision were, on the basis of the proof before it and the record in this court, supported by substantial evidence and were not arbitrary, capricious, or grossly erroneous.

65. As a result of the Board’s decision, the contracting officer, by Change Order No. 74 dated February 3,1958, made the following payments to plaintiff concerning Appeal No. 971:

Cost of remedial wall_ $87, 500. 80
Pumping performed during contraction — 102,940 units at $.08_ 32,235.20
Liquidated damages — 8 days extension at $500 4,000. 00
Pumping during 8 days extended time — 69,470 units at $.08_ 5, 557. 60
Total paid_$129,293. 60

In addition, he granted an 8-day extension of time for the construction of the remedial wall.

66. The Appeals Board arrived at 2,390 percent as the percentage of overrun in pumping for the second stage by subtracting 302,000 units, the estimated quantity of pumping for such stage, from 7,522,530 units, the actual quantity pumped from the second stage as of July 11,1956 (the month during which the Board was conducting its hearings, and apparently the latest figure submitted by the parties, the bulk of the pumping having been completed at that time), then dividing the result by 302,000 units. In so arriving at this percentage the Board compared the 302,000 units of pumping which the defendant originally estimated for the second stage, which quantity was necessarily based upon the assumption that the entire contract would be completed within 1,000 days, i.e., November 12, 1954, with the 7,522,530 units of water that plaintiff actually pumped from the second stage area as of July 11,1956. As shown, the contract completion date as extended (by 317 days) was September 25, 1955 (finding 31). As of that date 6,600,818 units had been pumped from the second stage area. Plaintiff did not complete the project by such date of September 25,1955 and kept *862on working, including some pumping of the second stage cofferdam, the pumping continuing until July 27,1956, when all pumping actually ceased, and by which time the total of 7,564,610 units had been pumped from such stage. Final acceptance of the project was, however, not made by defendant until April 8,1957, a period of 1,879 days after plaintiff commenced work on February 15, 1952.

67. (a) Following the Board’s decision, plaintiff was paid the Item 2 contract price of 40 cents per unit for all the 825,501 units pumped in the first stage area, totaling $330,-200.40. The deductions that had been effected by defendant after the issuance of Change Order No. 19 representing the difference between 40 and 25 cents for the 502,501 units pumped in excess of the originally estimated quantity of 323,000 units were all refunded to plaintiff, and no claim is made with respect thereto in these proceedings by either plaintiff or defendant.

(b) The total quantity pumped from the second stage area during the contract time as extended was 6,600,818 units. Plaintiff was paid the Item 102 contract price of 40 cents per unit for the estimated quantity of 302,000 units and, in accordance with the Board decision, for a 150 percent overrun, for a total of 755,000 units making a total payment at 40 cents of $302,000. For the remaining 5,845,818 units pumped from the second stage during the contract performance time as extended, plaintiff was paid at the Item 102(a) rate of 8 cents per unit, which defendant had established by its Supplement No. 1 to Change Order No. 20, the payment therefor totaling $467,665.44.

On the theory that there was no changed condition within the meaning of the contract, plaintiff claims 32 cents per unit, being the difference between the 40-cent contract price and the 8 ,cents paid for such 5,845,818 units, totaling $1,870,661.76.

(c) In addition, plaintiff pumped from the second stage 963,792 units subsequent to the contract period as extended for which it was not paid and during which period liquidated damages were applied.

68. In submitting its bid, plaintiff did not rely upon the Government’s estimates of the quantities of water to be *863pumped but made its own. appraisal of the situation and expected that it would be required to pump a substantially greater quantity of water than the stated estimated quantities. On the basis of the information that was available to it at the time its estimates were made, the Government should have anticipated that the contractor would pump more water under the instant contract than was pumped under plaintiff’s earlier contract. The parties did not enter into the contract in mutual reliance upon the accuracy of the Government’s estimates and there was no mutual mistake of fact with respect to the reliability or accuracy of such estimates.

CONCLUSION OF LAW

Upon the foregoing findings of fact, which are made a part of the judgment herein, the court concludes as a matter of law that plaintiff is entitled to recover and judgment is therefore entered for plaintiff in the amount of one million eight hundred seventy thousand six hundred sixty-one dollars and seventy-six cents ($1,870,661.76). The court further concludes that defendant is not entitled to recover on its counterclaims, and defendant’s first, second, and third counterclaims are hereby dismissed.

At the trial before our commissioner, both parties proceeded on a de novo basis and no objection was made by either to the offer of evidence beyond that received by the Corps of Engineers Claims and Appeals Board. After all proof was closed, defendant objected to the consideration of the de novo evidence on the basis of the decision of the Supreme Court in United States v. Carlo Bianchi & Co., 373 U.S. 709 (1963). However, because of the time and effort expended in the trial of the case in this court, defendant has expressly waived any right to have the case confined to the record before the Board or to have the case remanded to the Board. (Defendant’s brief, page 1, n. 1.)

Plaintiff is in fact a joint venture of four corporations. Por convenience, the joint venture will be referred to as plaintiff.

How these estimates were arrived at Is discussed in Part IV infra.

Plaintiff actually pumped 7,564,610 units by the time it ceased all pumping operations in the second stage area. Since the contract provided for payment only for pumping performed within the contract period plus duly authorized extensions, plaintiff’s claim is limited to the 6,600,818 units pumped by September 25, 1955.

In the “Definite Project Report on Jim Woodruff Dam” published in revised form in 194-8 and not made available to contract bidders, paragraph 35 of Appendix II warned that “all indications point to a difficult cofferdam problem which is likely to entail substantial expenditures to overcome.”

Plaintiff also presented to tlie Board a claim for extra ■work performed on the second stage cofferdam. The appeal -was sustained, the claim was paid, and it is no longer in issue, although it was made the subject of defendant’s second counterclaim. See footnote 7.

In its second counterclaim, tie defendant alleges that the Board decision in Appeal No. 971 (relating to additional work on the cofferdam for which plaintiff’s claim for additional costs was paid) was arbitrary, capricious, and not supported by substantial evidence. Defendant asked for a judgment for the amount paid on the claim. Since the trial commissioner’s report in this case was filed, defendant has abandoned its second counterclaim.

See National Concrete & Foundation Co. v. United States, 170 Ct. Cl. 470, 475 (1965); Lenry, Inc. v. United States, 156 Ct. Cl. 46, 49, 297 F. 2d 550, 551 (1962); Ivy H. Smith Co. v. United States, 154 Ct. Cl. 74, 83 (1961).

“Simply because an ‘equitable adjustment’ is involved does not automatically project the case into tile category of that kind of contract dispute to -which administrative finality attaches. The fundamental issue is the basic nature of the controversy. When a decision concerning the allowability of an equitable adjustment turns on the proper Interpretation of contract provisions, then what is ultimately involved is a question of law.” Kaiser Ind. Corp. v. United States, 169 Ct. Cl. 310, 330-31, 340 F. 2d 322, 333-34 (1965). See also United Contractors v. United States, 177 Ct. Cl. 151, 162, 368 F. 2d 585, 596 n. 5 (1966).

E.g., Amino Bros. Co. v. United States, 178 Ct. Cl. 515, 524, 372 F. 2d 485, 490 (1967); United Contractors v. United States, 177 Ct. Cl. 151, 167-68, 368 F. 2d 585, 599 (1966); Jefferson Constr. Co. v. United States, 172 Ct. Cl. 650, 661, 348 F. 2d 968, 972-73 (1965); Kaiser Ind. Corp. v. United States, 169 Ct. Cl. 310, 328, 340 F. 2d 322, 332-33 (1965); Hunt & Willett, Inc. v. United States, 168 Ct. Cl. 256, 262, 351 F. 2d 980, 983-84 (1964); Hoffman v. United States, 166 Ct. Cl. 39, 49, 340 F. 2d 645, 651 (1964); S.T.G. Constr. Co. v. United States, 157 Ct. Cl. 409, 416 (1962); Appalachian Floor Co. v. United States, 144 Ct. Cl. 11, 17 (1958); Carlo Bianchi & Co. v. United States, 144 Ct. Cl. 500, 504-05, 169 F. Supp. 514, 516 (1959), vacated and remanded on other grounds, 373 U.S. 709 (1963); Carman v. United States, 143 Ct. Cl. 747, 751-52, 166 F. Supp. 759, 762 (1958); John A. Johnson Contracting Corp. v. United States, 132 Ct. Cl. 645, 653, 132 F. Supp. 698, 701-02 (1955); Tobin Quarries, Inc. v. United States, 114 Ct. Cl. 286, 333, 84 F. Supp. 1021, 1022 (1949).

Chernus v. United States, 110 Ct. Cl. 264, 267, 75 F. Supp. 1018, 1019 (1948); Schatt Constr. Co. v. United States, 173 Ct. Cl. 836, 841-42, 353 F. 23 1018, 1021 (1965) (dicta). Cf., Peter Kiewit Sons’ Co. v. United States, 109 Ct. Cl. 517, 522, 74 F. Supp. 165, 167 (1947).

See Brawley v. United States, 96 U.S. 168, 171-72 (1877); Hirsch v. United States, 104 Ct. Cl. 45, 57 (1945); Thompson v. United States, 130 Ct. Cl. 1, 124 F. Supp. 645 (1954); Russ Mitchell, Inc. v. United States, 121 Ct. Cl. 582 (1952); Brock v. United States, 84 Ct. Cl. 453, 459 (1937); Morris & Cumings Dredging Co. v. United States, 78 Ct. Cl. 511, 517-18 (1933).

There Is no record of the actual quantity of water pumped by Shepherd; however, defendant’s contract estimate for the overflow dike project was 3,000,000 units.

One other important factor, fully discussed in findings of fact 24 (i) and 26(a), from which plaintiff concluded that the Government’s estimate was grossly understated was the difference in head differentials of water (average upstream pool water levels minus average foundation excavation elevations) between the lock and fixed crest spillway contract and the one here in dispute, especially with respect to the second stage area.

In fairness, it should be noted that the Government did not know that 1,600,000 unmetered units had been pumped. But it did know that an additional quantity above the metered figure had been produced, and it did not attempt to ascertain from plaintiffs the approximate amount of unmetered water pumped.

E.g., Leal v. United States, 149 Ct. Cl. 451, 463, 276 F. 2d 378, 384-85 (1960); Hirsch v. United States, 104 Ct. Cl. 45, 57 (1945); Ranieri v. United States, 96 Ct. Cl. 494, 506-07, cert. denied, 317 U.S. 690 (1942); General Contracting Corp. v. United States, 96 Ct. Cl. 255, 278 (1942). See also Thompson v. United States and Russ Mitchell, Inc. v. United States, supra, n. 12.

Tlie artesian water conditions in the area of the site of the project herein involved was described in well-known material published prior to the time the instant contract was let, including publications by the Florida Geological Survey. One such publication, “Information Circular No. 3”, issued in February 1950, was a paper prepared jointly by H. H. Cooper, Jr., District Engineer, U.S. Geological Survey, Tallahassee, Florida, and V. T. Stringfleld, Senior Geologist, U.S. Geological Survey, Washington, D.C., for presentation at a meeting of the Florida Section of the Soil Conservation Society of America. Prior to bidding on this contract, plaintiff had obtained a copy of and was familiar with this circular. The circular, at page 1, under “Occurrence of Ground Water”, states:

The ground water in Florida may be divided into two classes: (1) that generally known as the artesian water, which occurs in the extensive *796limestone aquifer of Eocene, Oligocene, and Miocene age that underlies almost all of Florida, the coastal area of Georgia, and the southernmost parts of South Carolina and Alabama; and (2) that occurring in several younger aquifers of relatively small areal extent, which consists chiefly of limestone, coquina, sand, and gravel in formations that range in age from Miocene to Recent.

The Miocene age encompasses the Tampa limestone formations; the Oligocene age includes the Suwannee limestone formation; and the Eocene age includes the Ocala limestone formation, all of which were hnown water carriers.

The circular also contained a map which had contours superimposed to show the piezometric surface in Florida and southeastern Georgia, i.e., the contour lines on such drawing showing the approximate “height, in feet, to which water will rise with reference to mean sea level in tightly cased wells that penetrate the principal artesian aquifer * * According to this map, artesian water in the limestone aquifers underlying the damsite under the present contract would rise in tightly cased wells to elevation +70, or approximately 70 feet above sea level. As the top of rock at the site of the work was at approximately elevation +15, and as sea level is 0.0, such artesian water would have risen for a distance of 55 feet above top of rock — indicating a heavy artesian flow. Elevation +70 was also only 5 feet below the tops of the cells for the second stage cofferdam at elevation +75.

Other publications concerning the underground water conditions in Florida included material specifically addressed to plaintiff’s experience on the prior work on the dam project. An article in the Engineering News-Record of June 14, 1951, a periodical widely read in the construction industry, stated, with respect to the work which plaintiff was performing under the earlier contract for the lock and fixed crest spillway, and which appeared some 6 months prior to the letting of the instant contract in question:

Porous limestone foundations. — under the entire site of the lock and spillway is Tampa limestone. It varies from soft, granular and earthy to dense and crystalline. It contains many weakened zones, weathering, leaching, oxidation and solution channels are prevalent. Below the Tampa limestone is a relatively unweathered and impervious layer of limestone ; below that is Suwannee limestone, another highly-waterbearing rock.

The record does not indicate how Knox arrived at his 128,667 square yard figure. However, if the measurement is made from the center or crest of the cofferdam, resulting in the inclusion of half of the cofferdam base, a figure close to Knox’s 128,667 square yards is obtained.

Specification, Par. 1-02 (b)(2), entitled Elevations, Protective Structures of Section I, Cofferdam, and Par. 10-03, entitled Functional Operations, of Section X — Gated Spillway, and “Headwater Curve” Chart and General Notes, on Drawing 5/1 of the contract drawings.

As is true of computing the effective pumping periods (see finding 24(e)), the calculation of these figures, or figures approximating them, at the time Knox made his estimates in the fall of 1951 would have required some estimating, since these figures represent amounts pumped through February 29, 1952, for the loch area and through February 15, 1952, for the fixed crest spillway area (id.).

Paragraph 3 of plaintiff’s “Reply to First Counterclaim and First Affirmative Defense” alleges:

* * * Plaintiffs allege that under no circumstances can the amount of water [to] be pumped from a cofferdammed area founded on porous rock in the bottom of a river in a known area of artesian water be accurately determined in advance of the performance of the work, and that plaintiffs and defendant understood this condition at the time bids were invited by defendant. Plaintiffs further allege that both parties understood that the amount of water to be pumped would be greatly influenced by the length of time that the project remained unwatered, as well as upon the amount of grout which defendant ordered plaintiffs to install in the underlying porous rock. * * *

There were three bidders. The Abstract of Bids showed in part:

With respect to the Government Estimate and the Winston and Tompkins bids, there is no showing what, if any, interrelationship there may have been between such items and Items 1 and 101.

Before the Board, plaintiff took the position that a correct application of the factors utilized by Knox in preparing his estimates, as well as a consideration of factors ignored, should have produced a total estimate of 2,200,000 units on both stages (i.e., 400,000 units for the first stage and 1,800,000 units for the second) instead of the '625,000 estimated.

The Board summarized the contentions plaintiff had made before it. Among them was the following:

d. It was impossible to estimate approximately the amount of water to be pumped. Failure to estimate approximately a quantity which it is impossible to estimate approximately, cannot be regarded as the basis of a claim for “changed conditions.”