This case under the Wunderlich Act, 41
U.S.C. §§ 321, 322, comes before the court on cross-motions for summary judgment, and on the plaintiffs request for review of a report and recommended decision by Trial Judge John P. Wiese, filed by him pursuant to Rule 166(c), and affirming a decision of the Armed Services Board of Contract Appeals (ASBCA). It has been submitted on the briefs and oral argument of counsel.
Plaintiff excepted only to the trial judge’s recommended dispositions respecting Claim Items 9, 3.c, lO.d, and 11, and has offered no opposition to the recommendations of the trial judge on other items. Before us, oral argument was confined by both sides to Item 11.
Upon consideration thereof, the court agrees with the trial judge’s recommended decision respecting all claim items, and hereby affirms and adopts the same as the basis for its judgment in the case.
Item 11 demands that the equitable adjustment for changes should incorporate an allowance for the cost of *284borrowing money to finance added work required by the constructive change orders on which such adjustment was allowed by the ASBCA, the amount thereof, absent interest, being $479,641. The Board denied this item as does the trial judge and as do we, but the trial judge alters the Board’s grounds somewhat. He takes the view in substance that the evidence before the Board, read most favorably to plaintiff, fails to establish a nexus or causal connection between the expenditures by plaintiff and the borrowings by plaintiffs corporate parent to provide cash for all its activities and those of its subsidiaries. He refers to a decisional trend by the ASBCA to allow interest not actually paid as such, "as an element of profit based upon the imputed value of the equity capital.” Plaintiff before us argued that the full scope of this new doctrine would make the nexus evidence found missing by the trial judge wholly irrelevant. But the case was not presented to the Board on any such theory, and the issue is not without its difficulties for us in view of the ancient doctrine disallowing interest against the Government, in the absence of express statute or contractual provision. E.g., Mescalero Apache Tribe v. United States, 207 Ct. Cl. 369, 518 F.2d 1309 (1975), cert. denied, 425 U.S. 911 (1976). Our decision here rests solely on the insufficiency of the showing before the Board as to plaintiffs and plaintiffs parent’s actual debt policy, how its debt structure was made up in the relevant period, and what connection extra work on constructive changes might have with that structure, all of which were things we could not judicially know and were essential to support Item 11 on the theory on which it was actually prosecuted. Nothing said or done in this case should be taken as intimating how interest or imputed interest should in general be dealt with in computation of equitable adjustments, where a contractor has financed additional work without contemporary reimbursement for it, nor what proof should be requisite to sustain a claim of that sort. This case stands on its own peculiar history. We note, however, that in a decision contemporaneous with this, another panel of this court grapples with the problem and establishes a precedent with respect to it. Framlau *285Corp. v. United States, ante at 185.
It is, therefore, concluded that the Board decision was supported by substantial evidence, was not arbitrary or capricious, and was not contrary to law. The plaintiff is not entitled to recover, and its motion for summary judgment is denied. The defendant’s motion for summary judgment is granted and the petition is dismissed.
OPINION OF TRIAL JUDGE
Wiese, Trial Judge: In a contract with defendant dated May 3, 1962, plaintiff (also referred to herein as Librascope or the contractor) agreed to design and fabricate a Data Processing Subsystem which was to form the computer nucleus of a world-wide Air Force information gathering, recording and dispensing system — a logistics management tool to be known as the 473-L System. Plaintiff has alleged that in the performance of this contract, the Government ordered work beyond that required by the terms of the agreement, the costs of which are claimed to be compensable under the Changes clause of the contract. Following an unsuccessful presentation of its claims to the contracting officer, plaintiff appealed to the Armed Services Board of Contract Appeals,1 requesting that it be compensated in the amount of $3,918,000 on account of the alleged changes. The appeal was allowed in part and denied in part by the Board, which after reconsideration,2 awarded the contractor the sum of $479,641. Relying now on both sections of the Wunderlich Act,3 plaintiff appeals to this court certain of the claims that were denied by the Board. We affirm the Board’s denial of these claims.
I. Facts
The contract for the Data Processing Subsystem portion of the 473-L System resulted from a two-step advertised *286procurement procedure. The proposal that plaintiff submitted in response to the Government’s request for technical proposals was incorporated into the contract. The original contract covered only what was termed the Interim Operational Capability (IOC) configuration of the Data Processing Subsystem, that is, one-half of a proposed paired system which, in its entirety, was to be referred to as the Complete Operational Capability (COC) configuration. The Government retained and eventually exercised its option to order the complete system.
Librascope had previously provided a computer system to the Federal Aviation Administration (FAA) which it believed to be similar to the Government’s contemplated 473-L System. Accordingly, in the instant situation, Librascope’s proposal envisioned use of this FAA computer as a baseline from which to develop the subsystem for the 473-L program. Also, that proposal relied significantly on the use of off-the-shelf components, a practice'which was encouraged by the Air Force.
The 473-L System was made up of three major hardware components: the Data Processing Subsystem (DPSS) which was to receive, process, store and dispense information received from other components of the system, the AUTODIN which was an existing world-wide Air Force communications network, and the Integrated Console Subsystem (ICSS) which was to be used to display information received from the Data Processing Subsystem. It was understood by all parties that a crucial aspect of contract performance would involve successful integration of these three major elements.
A number of other contractors were involved in the development of the 473-L System as a whole. In addition to contractors that supplied hardware elements, there were those that provided software and others who advised the Government on different technical aspects of the system. The Data Processing Subsystem was the first part of the 473-L System to be procured, although considerable work had previously been done by the Government (as well as other contractors) in preparation for the work of the Data Processing Subsystem contractor.
*287Given its scope and complexity, effective administration of the procurement demanded the participation of several Government agencies and private contractors. Thus, overall responsibility (i.e., the contracting officer) was located in the System Program Office of the Electronics Systems Division (ESD) of the Air Force Systems Command; Rome Air Development Center (RADC), another command in the Air Force, was assigned responsibility for providing technical engineering consultation to the 473-L Engineering Division at the Electronics Systems Division (ESD); MITRE Corporation (MITRE), a nonprofit organization, provided technical assistance in electronics design to the System Program Office; International Business Machines Corporation (IBM) was placed under contract to provide technical assistance to the Electronics Systems Division; finally, International Telephone and Telegraph Corporation (ITT) was contractually responsible for the Integrated Console Subsystem portion of the 473-L System.
The Data Processing Subsystem contract was essentially completed in the fall of 1964, more than a year later than scheduled. Additional facts are provided as part of the discussion of each of the several claims that follow.
II. The Contractor’s Claims4
Claim 6.e
Elimination of Duplexer
In plaintiffs original technical proposal of February 19, 1962, the COC installation (that is, the total paired system) that was depicted included, as a part of the Data Processing Subsystem, a freestanding piece of equipment referred to as a duplexer. The purpose of the duplexer was to provide the COC configuration with a switching capability that would permit either of the two central processing units to function with either of the two mass memory units. In the COC equipment, as finally installed, a freestanding duplex-er was not included. Plaintiffs claim is that elimination of the duplexer had been unilaterally ordered by the Government and that, because of this action, it became necessary *288to substitute a more complex (a.nd more expensive) distributed switching system. Recovery of the additional costs associated with this claimed extra effort was denied by the Board on the ground "that the change was not directed by any official of the Government authorized to do so.” (73-2 BCA at 48,385.) The Board’s decision is correct.
In the development of its proof at trial, plaintiff offered no direct evidence — either in the form of documentation or oral testimony — to support the contention that the Government had ordered the elimination of the duplexer. Those of its witnesses who spoke on this subject testified only to the fact of the change but they were without firsthand knowledge as to the reasons for that change. Significantly, the very individual who had been identified as the recipient of the Government’s change order — plaintiffs technical director — was never questioned on this point even though he was available at the hearing and had testified to other contract matters during the course of the hearing.
The point is pressed now that the Board’s decision was wrong because it ignored the testimony adduced during the cross-examination of the acting director of the Government’s System Program Office. The question had been put to this witness whether he had informed plaintiffs technical director that the duplexer would not meet the contract requirement because it was a unique piece of equipment. The answer that was given was: "This is probably so. Device and not function, again I differentiate.” It is this answer which plaintiffs brief identifies as a Government "rejection” which forced plaintiff to redesign the system to eliminate the freestanding duplexer.
The distinction that was drawn by the witness between uniqueness in device and uniqueness in function is a differentiation that draws upon two separate contract provisions. Functional uniqueness was prohibited by virtue of that requirement in the specifications which stated that "[n]o set of equipment that is used to satisfy these [major] functional areas [i.e., data processing, input/output and storage] shall be unique in the 473L system so that its failure would cause complete failure of the entire 473L system.” To satisfy this prohibition, the equipment had to *289have a "multi-thread” capability, meaning, that it had either to be internally duplicated (parallel capabilities within the same housing) or to be externally duplicated (paired with a like piece of equipment). Uniqueness of device, on the other hand, has reference to that provision in the contractor’s technical proposal which, in describing the intended multi-thread design, said that "[n]o single item of equipment is unique to the complete duplexed system.” In other words, in achieving the multi-thread capability, a design constraint was the requirement that the COC system reflect a paired identity of IOC systems. The two IOC systems were to be mirror images of each other and together they would constitute the COC system.
It is not clear from plaintiffs briefs whether it views the Government’s alleged rejection of the duplexer as one stemming, in the first instance, from the basic requirement for functional redundancy (i.e., the need for multi-thread capability) or from the subordinate provision that the COC system should comprise two essentially identical IOC systems. What is clear, however, is that neither argument can prevail.
As to the first of these arguments, the testimony leaves no room for doubt concerning the Government’s position. The duplexer was internally "duplexed,” meaning that, as built, it was functionally redundant and therefore satisfied the basic requirement of multi-thread capability. There is no question that the Government understood this. The same witness upon whose testimony plaintiff now attempts to rely said unequivocally that "[t]he duplexer, freestanding or otherwise, provided the function that met the requirements of contract.”
At the same time, however, the duplexer was essential only to the COC configuration and, in this sense then, it was "unique” — it was not a piece of equipment common to both the IOC and COC configurations. This too the Government’s witness recognized; thus, his answer on cross-examination: The duplexer was unique in "[djevice and not function, again I differentiate.”
But, to have identified the duplexer as a unique device falls considerably short of saying that the Government had demanded its elimination. The testimony, in fact, is *290otherwise. The witness stated that "Librascope of their own volition chose to incorporate those functions [of the duplexer] in the main frames in more strict compliance with some numerical aspects of the contract that two times IOC equals COC.” That self-directed incorporation, which the witness further identified as "a legitimate design choice on the part of Librascope,” was never convincingly contradicted by any of plaintiffs direct proof. Thus, in the final analysis, what plaintiff would now have us do is read into the record a conclusion that it involuntarily abandoned the initially planned use of a freestanding duplexer because the Government had condemned it for being a unique piece of equipment.
The approach is futile. To begin with, it is far from clear that the uniqueness of the duplexer was ever a matter of more than passing comment on the part of the Government. The Government witness readily conceded, for example, that aside from the duplexer, there probably were other unique pieces in the COC equipment. And this he referred to as a "transient consideration” — an unavoidable condition resulting from the fact that the IOC and COC configurations had represented separate procurements entered into at two different points in time. The duplexer fit this situation — it was a link between the two halves of the COC configuration.
Furthermore, even if we read the testimony on which plaintiff relies as indicative of some sterner measure of Government disapproval, it is still not enough to say that the Government directed plaintiff to take the step it did. Had there been such a directive one could reasonably have expected to see, as part of the proof offered in this case, some contemporaneous evidence of the position that plaintiff now asserts, namely, that the duplexer was not among those items of equipment subject to the uniqueness prohibition — in other words, some evidence of protest. However, proof of that sort was never offered. On the record before it, the Board reached the correct result.
*291 Claims 1.b and 6.a
Change from Fixed to Variable Pairing of Buffer Processors
Plaintiffs original design for the COC configuration, which was incorporated into the contract, included ten L-119 Buffer Processors (L-119s) — small digital computers used to provide necessary changes in the transmission rate of information traveling between the subsystems. Eight of these ten L-119s were located between the two central processing units of the Data Processing Subsystem and the AUTODIN, that is, four L-119 buffer processors were physically and functionally attached to each of the two central processing units in the COC configuration.5 Each such set of four included two L-119s which could send messages from the central processing unit to the AUTODIN and two L-119s which could receive messages for the central processing unit from AUTODIN. Thus, each central processing unit had two send-receive pairs; one was designed to act as a back-up for the other in order to provide the "multi-thread” capability required by the contract.6
This aspect of plaintiffs design underwent change early in the relationship between the parties. The initial configuration of eight L-119s in the DPSS/AUTODIN interface was retained in plaintiffs written description of the Data Processing Subsystem dated April 1962 (the April Subsystem Description) which had been submitted to the Government before the contract was signed. However, in the August 1962 Subsystem Description, transmitted to the Government after contract signing, there were shown only four L-119s in the DPSS/AUTODIN interface — one send-receive pair for each central processing unit rather than the two pair as had initially been proposed. There is no dispute that this was the first time that defendant was *292made aware of plaintiffs intention to reduce the originally planned number of L-119s.
The contractor’s August Subsystem Description, as well as each of its previous program descriptions, had indicated Librascope’s intention to create a fixed link between the send-receive pairs of L-119s, that is, that each "send” L-119 would be able to operate with only one certain "receive” L-119. However, as a result of a meeting that took place between the parties on November 7-8, 1962, plaintiff changed its design to permit variable pairing of the four remaining L-119s so that each of the two "send” L-119s could be paired with either of the two "receive” L-119s.
Librascope has claimed that at the November 7-8 meeting, the Government unilaterally ordered plaintiff to make the change to variable pairing and that this constituted a compensable change. Claims 1.b and 6.a are for the costs of this alleged change. Defendant, on the other hand, maintains that plaintiff agreed to provide the variable pairing capability in return for the Government’s willingness to accept only four L-119s in the DPSS/AU-TODIN interface. The Board sustained the Government’s position. It concluded that there had been an agreement reached between the parties pursuant to which plaintiff agreed to provide variable' pairing as consideration for the Government’s willingness to accept the reduction of L-119s from eight to four. "The Government,” said the Board, "was contractually entitled to receive * * * [the initial] configuration or one of similar reliability. They [the parties] agreed upon a compromise whereby the appellant would furnish only four L-119’s, but would add a variable pairing- switching function to provide the necessary reliability.” (73-2 BCA at 48,381.)
In its challenge to the correctness of the Board’s decision, plaintiff raises, among other points, the contentions that (i) it did not have a contractual obligation to adhere to the initial configuration of eight L-119s, so long as it met contract performance requirements; (ii) that the four L-119s were functionally equivalent to the initial configuration of eight L-119s because the reduction in the number of L-119s was accompanied by an increase (a doubling) in *293their transmission speed; and (iii) that it did not have a contractual obligation to demonstrate any minimum reliability as a condition to the Government’s acceptance of the equipment. It is argued that these considerations, coupled here with the absence of any documentation evidencing the agreement found by the Board, demonstrate that plaintiff had no reason to effect the compromise agreement attributed to it by the Board and allegedly substhntiate that the change to variable pairing was nothing more than an extra-contractual requirement unilaterally imposed by the Government.
Perhaps in another context these arguments would spell out a strong case. Here, however, they do not. The only question at this point is whether the Board’s decision was based upon substantial evidence, and, on that score, there can be no doubt. Not only did the Board have before it the testimony of a Government witness who had attended the November meeting (and who claimed that an agreement had been reached at that meeting) but the record also contained plaintiffs own status report, prepared in April 1963, which recited the substance of the agreement in virtually the same terms alleged by the Government.7 Moreover, to this proof, there would have to be added the equally telling fact that plaintiff produced not a single witness to refute the Government’s testimony that an agreement had been reached at the November 7-8 meeting, nor any witness who could account for the change to variable pairing on some ground other than the agreement which had been claimed. Plaintiffs witnesses knew only that their superior in the Librascope organization had told them to make the change. But that superior himself did not testify.
Given this state of the record, the competing considerations that plaintiff now relies upon can detract little from the Board’s decision. It does not matter, for example, *294whether the Board was right or wrong in viewing the parties’ agreement as one grounded upon equipment reliability considerations, i.e., that variable switching was resorted to in order to provide system reliability essentially equal to the initial eight L-119 configuration. For even if plaintiff is correct in now saying that it was not contractually required to demonstrate any minimum degree of reliability (a point we presently accept only for argument’s sake)8, this would not impair the basic correctness of the Board’s decision, namely, that the parties did reach an agreement. That much, at least, the record unequivocally showed.
Nor is there anything to be gained by arguing, as plaintiff also does, that an agreement between the parties would have prompted the Government to formalize the arrangement in writing, as had, in fact, been done in like instances during performance of the contract. Clearly, the lack of such formalization cuts with equal- force against plaintiff, for, if the variable switching mechanization did indeed portend the cost implications that plaintiff now assigns to it (the change having been introduced, according to plaintiffs argument, at an advanced point in the state of completion of the L-119s), then one can (indeed, must) question why plaintiff itself did not insist upon a written change order from the contracting officer. The absence of such a move on plaintiffs part suggests that what is claimed now as a change was really no change at all, or at least not one that had been unilaterally imposed.
The lack of specificity in plaintiffs proof is not bettered by the only documentary evidence that it offered in connection with this claim, namely, a copy of a trip report that was prepared by plaintiffs contract manager following the November meeting. That report contains the statement that "[i]t is required that the L-119’s be switched in pairs and singly * * Plaintiffs argument emphasizes this "[i]t is required” language, saying that this *295reflects the then contemporaneous unilateral action that was ordered by the Government.
The argument is weak. The language in question is, at best, ambiguous. It is impossible to say whether the author is indicating that the Government required the variable switching capability, that the specifications required it in order to meet the multi-thread capability, or that plaintiffs chosen design required the variable pairing. Certainly the document does not refute the testimony of the Government’s witnesses that plaintiff agreed to make the change in return for defendant’s willingness to accept four L-119s in the DPSS/AUTODIN interface. Assessed separately or in toto, plaintiffs arguments fail to show a lack of substantial evidence in favor of the Board’s decision. The decision therefore stands.
Claim 6.d
Electronic Typewriter Interface Changes
A. Direct Access from Electronic Typewriters into Central Processing Units — Plaintiffs original design called for the path from the electronic typewriters of the Integrated Console Subsystem to the central processing units of the Data Processing Subsystem to be routed through two L-119s, one for each of the two central processing units in the COC configuration. In the equipment as delivered to the Government, the interfacing L-119s had been eliminated — the access from the electronic typewriters into the central processing units was direct. Plaintiffs claim is that this change was one which the Government unilaterally imposed.
It is not disputed that the Government was responsible for bringing this change into being. A finding to this effect was made by the Board. Instead, what remains in controversy at this point is the Board’s conclusion that plaintiff had agreed to do this work — eliminate the interface buffers and provide direct access into the central processing units — at no additional cost to the Government in return for defendant’s acceptance of a reduction in the number of L-119s from ten to four.
*296This last point mentioned, that is, the reduction in the number of L-119s, addresses the same subject previously discussed (Claims l.b and 6.a); hence, the argument plaintiff now makes repeats some of the points that it earlier raised. Thus, again we meet the contention that there is no contractual evidence, that is, no written documentation, to support the agreement found by the Board. Again, too, the argument is raised that plaintiffs economic self-interest would never have counseled in favor of agreeing to the "trade-off’ which the Board concluded the evidence had established.
The arguments are no more persuasive now than in their earlier setting. Aside from the fact that they cut with equal favor in the Government’s direction — a point we noted earlier — there is also another and more substantial point which plaintiff overlooks, namely, the testimony of its own witness. While describing changes that had occurred in the DPSS/ICSS interface, plaintiffs technical director mentioned, among others, the high-speed channels that had been introduced on the electronic typewriter path. This point then led to the following colloquy between the Board member and the witness:
Is this what has been characterized as that claim which resulted in the elimination of the L-119 and made the channel direct from the electric typewriters to the computer, the central processor?
The Witness: Yes, I think at that time we did some trading with the SPO [System Program Office], In exchange for the high-speed channel, we eliminated one L-119 [one L-119 for each of the two paths] and I think we also agreed not to argue whether we were supplying four or eight L-119’s on the communication [this last refers to the subject matter of claims l.b and 6.a].
The previously quoted testimony, the substance of which was repeated again in the course of the Government’s later cross-examination of the same witness, is not dealt with in plaintiffs present arguments. There the point is simply ignored and other arguments are pursued instead. However, these other arguments, though examined, need not be identified here. It is enough to say of them that they neither detract from nor minimize the testimony elicited from plaintiffs witness. That testimony, which was certain*297ly not inherently improbable nor contradicted by any other testimony or circumstances disclosed in the record, fully supports the conclusion that was reached by the Board.
B. Direct Access from Central Processing Unit into Electronic Typewriter — By the terms of the contract, Librascope was required to provide the Government with a specification setting out data relevant to the eventual interface between its own subsystem, the DPSS, and the ICSS, the subsystem whose design and manufacture was to become the responsibility of another contractor. Because of Librascope’s delay in developing this interface specification, the System Program Office decided to appoint an industry-Government advisory group — referred to as SIDG (Subsystems Integration Design Group) — to consider the interface problems likely to be encountered, to develop solutions to these problems and, in the overall, to coordinate development of the DPSS/ICSS interface specification. The SIDG was a composite group; its industry members came from Librascope, IBM, ITT, and MITRE, and its Government members came from RADC and ESD.
As a consequence of the meetings that were held by this advisory group and the interchange of ideas that resulted therefrom, certain changes were introduced into Librascope’s originally envisioned DPSS/ICSS output interface. Among these was a change in the message path from the central processing unit to the electronic typewriter. Initially what had been proposed was a message path routed through the mass memory. What finally was delivered, however, was a system in which communication from the central processing unit to the electronic typewriters was accomplished by direct access.
Librascope sought compensation for this change on the theory that the Government, acting through the SIDG, had unilaterally ordered this changed configuration in the mass memory output interface. The Board rejected the claim saying, "[t]he change * * * was not one directed by the Government. The record does not even disclose who originated the suggestion; whether it was the appellant, the Government, or one of the other contractors. All that is disclosed is that someone at the first SIDG meeting made this as a suggestion * * * [and that] [t]here was a general *298consensus that this was an excellent solution, and appellant implemented it.” (73-2 BCA at 48,387.)
The contractor attacks both aspects of the Board’s conclusion. It contends, first of all, that the individual responsible for the change (meaning, in this situation, the individual whose recommendation it was that first prompted the SIDG’s discussion of the matter) was the Government’s representative from RADC — an individual who served not only as chairman of the advisory group but also as technical assistant to the director of the entire 473-L program. Secondly, plaintiff contends that, by virtue of the authority vested in the SIDG, the actions taken by that group became, in effect, actions taken by the contracting officer.
We agree with the Board that the evidence does not disclose whose recommendation it was that ultimately brought about the change in question. However, given plaintiffs main argument, that it was the advisory group as a whole that was vested with change-directing authority, it becomes unimportant to pass upon the actions of its individual members.
There is no question that the advisory panel recognized that, in its definition of system interface requirements, it might hit upon solutions to technical problems that would impact upon the boundaries of the then-approved design. Nor is there any question that plaintiffs technical director considered SIDG’s technical solutions as the equivalent of directives — functions that had to be implemented because as he put it, "[i]f we didn’t build the interface that way [meaning, per SIDG solutions], we would never deliver the equipment.” As plaintiff sees the situation, these two factors — the recommendations of SIDG on the one hand and their implementation by Librascope on the other— establish plaintiffs right to recover. They do not.
To begin with, the advisory panel did not operate in the freewheeling fashion that plaintiffs brief seems to suggest. There was not a disregard by SIDG of the contractual implications of the action that it was generating. On the contrary, in the course of his testimony, the Government’s SIDG representative — the chairman of the group — emphasized that, at their first meeting, it was made known to all *299participants that SIDG was an advisory body only and as such, no SIDG member would have authority to bind his respective organization, nor would any member have authority to direct the implementation of any technical decisions resulting from SIDG’s deliberations.
Furthermore, even though SIDG did not have a clear perception as to the contractual actions that might be necessary in order to formalize its technical decisions, the important point is that there was an awareness, as the Government witness said "that there would have to be some sort of a corporate government agreement formalized, which would pick up all of the information generated by the SIDG.”
Plaintiffs own witness (its representative at the SIDG meeting) did not understand the situation any differently. While he disputed the point that there had been any initial discussion respecting limitations of authority, he nevertheless understood the SIDG meetings to be technical discussions without authority for commitment. He also acknowledged that, with respect to design changes recommended by SIDG that were subsequently incorporated by Librascope in its equipment, "it was a common understanding that one way or another there would be shakeouts with regard to cost and scope [meaning, a resolution by contract means of any additional costs associated with design changes].”
As anticipated by the above-quoted testimony, contractual follow-up to SIDG’s actions did take place. On October 16, 1962, the System Program Director forwarded to Librascope a draft of the interface specification that had been generated by the SIDG meetings. The accompanying transmittal document requested the 473-L contractor and subcontractors to review the draft specification "to determine whether the design constraints implied therein are within the scope of existing contracts and signify acceptance and exceptions to the 473L SPO by TWX * * * followed by a signed letter of acceptance.” Librascope responded on October 24, 1962. Its telegram of this date stated that it was "TO SIGNIFY CONCURRENCE WITH SUBJ. [interface] DESCRIPTION WITH FOLLOWING EXCEPTIONS.” Of the four exceptions that were noted, *300none dealt with the present claim — the interface between the central processing unit and the electronic typewriter. Neither did the telegram make any reference to any demand for additional compensation based upon "out-of-scope” work.
Given the facts recited above, it is apparent that SIDGsponsored technical changes were not changes unilaterally imposed upon Librascope. The contractor had a voice in the matter from the very beginning of the interface discussions. Furthermore, even if Librascope felt itself obliged to go along with SIDG’s recommendation in order to sidestep any schedule delays that might otherwise have resulted, it was certainly not denied the opportunity to point out, and claim additional compensation for, any changes that transgressed the parameters of its own design. The fact that Librascope did not raise its present claim as an exception to the interface specification is an action that speaks for itself. Thus, while the Board’s conclusion that the Government did not order the change is perhaps a more abrupt answer than the underlying circumstances require, it still fits the facts enough to be correct. The Government endorsed the technical feasibility of SIDG’s recommendations, submitted them for evaluation to plaintiff (for technical as well as contractual assessment) and plaintiff, in turn, accepted them. In the last analysis then, it was plaintiffs concurrence with SIDG’s recommendations, rather than a Government order, that led to the implementation of the change in question.
In an effort to undo its earlier-expressed acceptance of the interface specification, plaintiff now points out that it never submitted, as a follow-up to its acceptance telegram of October 24, 1962, the signed letter of acceptance that had been requested by the System Program Director. From this omission we are presumably being asked to conclude that plaintiff never agreed to implement the interface specification on a no additional cost basis.
The point is irrelevant. Plaintiff points to nothing in the record to warrant the conclusion that either party intended to postpone a binding contractual commitment until evidence of their understanding had been formalized by way of a signed letter of acceptance. The Government’s *301letter of October 16, 1962, asked Librascope to signify its acceptance and exceptions by telegram, this then to be followed by a signed letter of acceptance. In its telegrammed response, Librascope noted its concurrence. These manifestations of assent were sufficient in themselves to conclude an agreement and contract law would not inhibit such effect merely because the parties may also have manifested an intention to prepare and adopt a later written memorial of their agreement. Restatement (Second) of Contracts § 26 (Tent. Drafts Nos. 1-7, 1973).
Claim 6.c
Addition of Interrupts to Input/Output Interface and Mass Memory Interface
Plaintiff had proposed, and its contract required, that the DPSS equipment have an interrupt capability in certain situations. An interrupt is a feature that allows the execution of a computer program to be suspended in order to permit the computer to become immediately responsive to the reception of new material. The interrupt is necessary to accommodate an outside signal or permit a response to the detection of an error in data transfer. Iri Claim 6.c plaintiff seeks additional compensation for certain interrupts that were included in the system. Since separate factual circumstances are involved in each situation, it is necessary to deal with the problems on an individual basis.
A. Integrated Console Subsystem Message Available Interrupt — In the system configuration as initially envisioned, plaintiff had planned to mechanize an interrupt capability in that L-119 buffer processor which linked the central processing unit of the Data Processing Subsystem with both the electronic typewriters and the digital data links of the AUTODIN communication system. This initially planned interrupt was to be referred to as "input-any,” meaning, in effect, that the central processing unit could be signalled to receive a message, transmitted through the "linking” (our word) L-119 buffer processor, originating from any of the electronic typewriters of the Integrated Console Subsystem or any of the digital data links of the communication system.
*302As discussed in Claim 6.d, this linking L-119 was eliminated and in its place was substituted a high-speed channel that provided direct access to the central processing unit. Because of this change in configuration, it became necessary to revise and enlarge the interrupt capabilities in order to accommodate both the Integrated Console Subsystem’s electronic typewriters and the digital data links of the AUTODIN communication system. The configuration that emerged was described by plaintiffs witness as comprising the following: an "L-119 Message Available Interrupt,” an "electronic typewriter (ET) input” or "ICSS Display Console Message Available Interrupt,” an "L-119 input” or "L-119/ET Interface Not Busy Interrupt,” and, finally, a "major” interrupt (as opposed to the first described interrupts which are termed "detail” interrupts), known as the "I/O (input/output) Interface Interrupt.” Of these four interrupts, the three last listed are claimed to represent extra-contractual effort for which additional compensation remains owing.
In its treatment of this particular claim, plaintiff has related the origination of the L-119/ET Interface Not Busy Interrupt (one of the four interrupts which the witness had described) to circumstances other than the design modification that brought about the direct access, high-speed channel earlier mentioned. The reason for this departure from the witness’ testimony is not clear. At any rate, the present claim is limited to two interrupts — the ICSS Display Console Message Available Interrupt and the master I/O Interface Interrupt. And, as to these, what plaintiff claimed was that the same set of circumstances which supported its right to recover for the change to direct access between the electronic typewriters and the central processing units likewise supported its right to recover for the additional interrupts which that change had occasioned.
The Board denied the claim saying that, "[t]he interrupt changes made necessary by the change to direct communication between the CPU [Central Processing Unit] and the ET [electronic typewriters] were merely details of that transaction.” (73-2 BCA at 48,389.) And, since plaintiff was found to have agreed to provide direct access at no *303additional cost, it was therefore not entitled to any further compensation for implementing the interrupt changes attendant to that basic change.
We agree with the Board’s conclusion. In its presentation of this particular set of interrupt claims, plaintiff claimed the same causal connection which the Board eventually relied upon, that is, given the basic change to direct access between the two subsystems — the integrated console and the data processor — the interrupt mechanization changes now in question had necessarily to follow. In other words, the changes were, as the Board called them, "details” of the basic change. Given this connection, and also the Board’s conclusion that the provision for direct access was, in and of itself, a matter of agreement between the parties, it would seem to follow that that agreement intended also to include the necessary changes to the interrupt configuration. At least, as presented, the claims allow that conclusion for nothing in the proof says otherwise.
B. L-119/ET Interface Not Busy Interrupt and L-119 Output Message Available Interrupt — The second series of interrupt changes in the central processing unit interface for which plaintiff asks recovery are those referred to as the L-119/ET Interface Not Busy Interrupt (identified in section "A.” above) and the L-119 Output Message Available - Interrupt. Neither of these interrupts was specifically called out in plaintiffs proposal; both are claimed now to represent changes that plaintiff was obliged to include in its final design. The principal argument made below was that this alleged additional work had been required by IBM, a parallel contractor charged with providing technical assistance to the Government who, according to Librascope’s argument, came to acquire the status of a representative of the contracting officer.
The Board found no merit in this claim. With respect to the L-119/ET Interface Not Busy Interrupt, the Board noted that there was general testimony, on plaintiffs part, to the effect that this interrupt had been included at IBM’s request and Government testimony saying the interrupt was required by the terms of the contract. However, the Board also made the observation (one with which we entirely agree) that the evidence was "too scant to permit *304detailed findings.” (73-2 BCA at 48,389.) Despite the limitations in the proof, the Board concluded that information signaling the completion of an input/output operation had to be conveyed to the central processing unit in some manner, meaning, that some form of interrupt capability was necessary. It further concluded that, since this particular interrupt capability occurred in the interface between the communication system (AUTODIN) and the data processing system (as opposed to the interface between the integrated console and the data processing subsystem), the interrupt feature required mechanization by hardware means rather than by software. Expressed in contract terms, this hardware requirement meant that the interrupt feature was plaintiffs responsibility rather than the responsibility of the contractor charged with development of the software program. (The difference between hardware and software is explained below.) As to the L-119 Output Message Available Interrupt, the Board’s conclusions were the same: the interrupt capability was required by MIL-D-27113 and, since it too was located in the AUTODIN interface, it likewise had to be provided by hardware.
With respect to plaintiffs argument claiming IBM to have been the moving force responsible for requiring the addition of the interrupts, the Board simply held that the evidence failed to establish that the L-119/ET Interface Not Busy Interrupt was the product of a directed change. As to the other interrupt, which plaintiff had likewise attributed to IBM’s demands, the Board made no conclusion save the point noted above, i.e., that the work was required by the contract’s terms.
In the argument which plaintiff now presents, the question is no longer whether the interrupt capabilities in issue were required by the terms of the controlling specification, namely, MIL-D-27113. This much of the Board’s decision the contractor does not quarrel with. Rather, as here refined, the question is whether these capabilities had to be provided by hardware means or could, instead, be satisfied by software. As had been explained in the testimony of plaintiffs technical director, a hardware means of interrupt mechanization is required *305whenever there is a need to command the computer’s immediate attention, that is, the need to go from a program in execution to an external program unique to the specific cause of the interrupt. Software interrupt mechanization, on the other hand, is a programmed or preplanned interrupt system whereby the computer, at specified time intervals, communicates with input indicators to determine whether there are any messages awaiting processing. It was further explained — again by plaintiffs witness — that all so-called synchronous interfaces required hardware interrupt. And, among these synchronous interfaces, the witness included the "L-119 to Autodin” as well as "output to Autodin.” It was upon this last point — the identification of the AUTODIN interface as a hardware-requiring interface — that the Board relied. It is this same point which the contractor now challenges.
Plaintiff contends that the requirement for hardware mechanized interrupts in synchronous interfaces applied only to data transmission channels and not to interrupts of the type in question. The latter, claims plaintiff, were "status-seeking” interrupts, meaning, that their purpose was only to question the availability of the L-119 for communication rather than be involved in data transmission. Thus, according to the argument, there was no necessity for hardware; the same interrupt capability might as readily have been accomplished through programming (i.e., software) in which case the contractual responsibility rested elsewhere than with plaintiff.
It is impossible to decide whether this argument is right or wrong. The testimony upon which plaintiff relies simply does not make the point with any clarity, if, in fact, it makes the point at all. That testimony was often incomplete and often confusing and, on the record as it stands, we simply cannot say that the Board’s conclusion lacked the support of substantial evidence.
Furthermore, even if the evidence could have permitted the above-discussed point tó be resolved in plaintiffs favor, the claim would still fail. There was no evidence of any substance to support the principal argument that had been made below, namely, that the interrupt changes had been ordered by IBM or that that company, in its capacity as *306technical assistant to the Government, had come to dominate plaintiffs contract efforts.
First, with respect to the L-119/ET Not Busy Interrupt, plaintiffs only evidence of IBM involvement was a chart, admittedly prepared by an IBM employee, showing this particular interrupt. However, there was no evidence saying that this configuration had. been proposed — much less ordered — by IBM. Thus, for all we know, the chart might just as easily have reflected IBM’s understanding of plaintiffs proposed design. Indeed, there is more than a casual basis for such speculation. For one thing, the chart was part of a series of notes identified (in the written material itself) as having been "obtained” during a meeting at Librascope on October 29, 1962 through November 6, 1962. Then too there is the fact, mentioned earlier, that this particular interrupt had been described by plaintiffs witness as an interrupt connected with the change to direct access between the electronic typewriters and the central processing units — a change with which IBM had had no connection whatsoever. In short, what we have is a situation in which the evidence fails to establish the point for which it was offered while, at the same time, admits to inferences that undercut plaintiffs position.
As to the second of the interrupts claimed here, the L-119 Output Message Available Interrupt, the proof is no better. The evidence consisted solely of a memorandum from IBM to the 473-L System Program Office requesting the inclusion of an interrupt capability equal to that provided by the L-119 Output Message Available Interrupt. The witness through whom this document was introduced at the hearing did not indicate that there was any causal relationship, exhibited by this correspondence, between IBM and the System Program Office on the one hand, and the contractor’s inclusion of the requested interrupt capability on the other. Nor, it might be added, did any other witness provide such testimony. In fact, we have no evidence that Librascope was ever furnished a copy of this letter. All that was said was that the interrupt was not part of plaintiffs original design and that, in the opinion of the witness, the interrupt was not required by the terms of the contract. Beyond doubt, this was not *307enough to attribute the change to IBM and, thereafter, indirectly to the Government.
C. Display Depleted Interrupt (End of Transfer Interrupt for Mass Memory) — There is no dispute that the contract specification, MIL-D-27113, required the equipment supplied by plaintiff to have the capability of interrupting the program upon completion of a display output by the mass memory. The contract designated this function as one which would be fulfilled by an interrupt and plaintiffs proposal indicated that such an interrupt would be provided. However, that proposal did not specify whether this Display Depleted Interrupt would be accomplished with hardware or with software. Plaintiff claims that it had originally planned to provide the function with software, that it was not precluded from doing so by the contract specification, and that the Government unilaterally ordered it to provide a hardware interrupt for this function. Plaintiff asserts that this is a compensable change.
Struggling with a record that can only be considered obscure and confusing, the Board found that the specifications were not clear as to whether the interrupt function required the use of hardware or could have been accomplished with software. It also found that the Government interpretation was better (from a technical standpoint) than plaintiffs and that, in any event, Librascope had waived its right to claim a compensable change in that it had agreed with the interpretation it now attributes solely to the Government. The Board’s view of the facts of this claim is correct, and the result it reached — a denial of the claim — was proper.
There is no dispute that the requirement for a hardware-mechanized Display Depleted Interrupt originated with the Government. Likewise, it is not disputed that, in asserting this position, the Government was pressing for nothing more than what it believed was required by the governing contract specification, MIL-D-27113.
But, on plaintiffs side of this issue, there existed less unanimity of thought. As the Board noted, one of Librascope’s engineers (a witness at the trial) was of the view that the interrupt function was not at all required by *308the governing specifications, that is, neither hardware nor software had to be provided. Another of plaintiffs witnesses, Librascope’s technical director, conceded the existence of the requirement but, at the same time, wás of the view that, contractually speaking, the requirement could be satisfied by software rather than by hardware means. He allowed, however, that while the change from software (the approach Librascope had plannedfto hardware (what the Government required) represented a major change, the difference in contract interpretation underlying the two approaches "was arguable.”
At any rate, this difference in contract interpretation notwithstanding, Librascope did proceed to do the work according to the Government’s interpretation, that is, provide the interrupt capability by hardware means. And, so far as this record reveals, the work was done without notice to the contracting officer that Librascope considered the effort involved to be extra-contractual. There was no protest from any authorized corporate spokesman.
The contractor’s failure to protest, while perhaps not an outright bar to the claim, is nevertheless an evidentiary consideration which, in these circumstances, takes on controlling significance. For, on this record, with the evidence as sparse as it is, Librascope’s decision to do this admittedly "arguable” work without protest could mean that it came eventually to agree with the Government’s position, or, alternatively, that it had chosen, for reasons of its own, not to press the point. The record simply does not tell us why the issue was never carried to the level of the contracting officer. All we know is that plaintiffs technical director, while disagreeing with the Government’s position, conceded the point was arguable and conveyed these thoughts to Librascope’s contract spokesman. There the evidence ends. The work was done and nothing more was said until after all work had been completed. Thus, the facts admit of two possible conclusions: that Librascope ultimately accepted the Government’s interpretation as correct, or, as the Board saw the situation, that Librascope deliberately waived its right to assert a claim under the Changes article of the contract. In either case, the contractor’s conduct, though purely negative, manifested a *309purpose or intent upon which the Government was entitled to rely. The right to press a claim was thereby extinguished.
Claim 9
Reliability Report
Plaintiff's contract required that it prepare a report setting forth reliability predictions for the equipment that it was supplying to the Government. A preliminary version of this report was submitted on June 27, 1962. On July 31, 1962, the Government rejected the report. Its criticisms included the facts that the calculations had not been based upon the latest version of the statistical guidelines referenced in the contract and also that the calculations had not included the peripheral equipment — the so-called "off-the-shelf’ items which had been obtained from commercial suppliers. The Board upheld the Government’s position on each of these points. The contractor raises them again here; we agree with the Board.
A. The first branch of this two part claim concerns Librascope’s allegation that it was required, by the Government, to prepare its reliability predictions by reference to a document other than the one which had been specified for that purpose in the contract. This alleged change in reference documents necessitated extra effort for which recovery is now being sought.
The contract provided that the contractor should calculate reliability predictions for the equipment in accordance with the Armed Services Technical Information Agency (ASTIA) document, AD-148868 Reliability Notebook (also referred to as the RADC Reliability Notebook). But, as to this documentary reference (as well as others which the contract listed) it was also stated that the applicable document would be "the issue in effect on date of invitation for bids * *
In the preparation of its preliminary reliability report,' Librascope used the document called out in the contract, namely, RADC Reliability Notebook AD-148868. As indicated, the Government disagreed with the contractor’s use of this document because it did not represent what the *310Government considered to be the latest version of that document, namely, the revision dated December 31, 1961. However, this revision bore a different identifying number — PB-161894 rather than AD-148868 — and it had been issued by a different agency of the Government, being a Commerce Department document rather than an ASTIA document.
Because of this change in identification number and issuing source, the contractor now makes the argument that the two documents were not the same — one was not an updated version of the other. Thus, according to this view of the matter, there was only one RADC Reliability Notebook which the contractor was contractually obliged to follow, namely AD-148868, the one which the contractor did use. To round out this argument, the contractor also says that there was no evidence in the record to permit the conclusion arrived at by the Board — that PB-161894 was the controlling document.
The argument is without substance. In its letter advising the contractor of the disapproval of the preliminary reliability report, the Government spelled out that "[fjuture calculations should be based on data from the current RADC Reliability Notebook, dated 31 Dec 61.” This criticism was repeated when the Government received, and again rejected, the contractor’s revision of the preliminary reliability report. Once more attention was called to the fact that "[t]he latest revision of the RADC Notebook is 31 December 1961, and should be used in place of the 30 October 1959 data.” Not only did plaintiff not question the correctness of the Government’s instructions on this matter but, during the course of the trial, plaintiffs technical director specifically stated that he agreed with the bases for rejection which the Government had pointed out. Furthermore, during cross-examination of this witness, the question was specifically asked whether the RADC Reliability Notebook that was in effect at the time of the invitation for bids was Revision 2, dated December 1961. The witness answered the question by saying, "I think so.”
If there had been any doubt as to the correctness of the point which plaintiffs own witness had conceded, the matter could readily have been resolved simply by intro*311ducing into evidence the two documents whose relationship the contractor now questions. However, that evidence was not made a part of the record. Thus, all we now have is simply argument, unsupported by proof, that the documents were distinct rather than different versions of the same basic source material. That is not enough to prevail. The evidence as it stands supports the Board’s conclusion that the contractor was contractually bound to use the document which the Government had insisted upon.
B. The second part of this claim concerns the contractor’s allegation that it was not required, by the terms of the contract, to include reliability figures for off-the-shelf equipment such as electronic typewriters, line-printers, card readers, punch controls, and magnetic tape transports. The Government had required this work, and, as noted, the Board sustained the Government. The Board based its conclusion on the fact that the specifications made no distinction between equipment manufactured by plaintiff and equipment it might purchase from others; reliability was addressed only in terms of the equipment as a whole.
The contractor’s argument against the correctness of the Board’s decision raises two points. First, it is pointed out that at the bidders’ conference (pre-award), the Government had actively promoted the use of off-the-shelf items and that the contract, as executed, carried this purpose forward. (A paragraph in the contract recited that "[i]t is desired that the Data Processing equipment be off-the-shelf equipment.”) Plaintiff goes on to point out that bidders had been told that off-the-shelf equipment could be used even if such use entailed a deviation from applicable military specifications. In view of these circumstances — the announced preference for off-the-shelf items which could be accommodated despite the equipment’s failure to meet all applicable military requirements — plaintiff allegedly concluded that reliability predictions for such components would not be required. The reason for' this conclusion on plaintiffs part is not explained but, at any rate, it was because of this conclusion that plaintiff claims that the scope of the reliability predictions set out in its technical proposal was limited strictly to equipment it manufactured *312itself. It did not propose to furnish reliability predictions on any off-the-shelf equipment it might also be supplying. Thus, plaintiff relies upon the scope of its own proposal, and indirectly upon the acceptability to the Government of off-the-shelf equipment, to support the argument that reliability predictions were contractually required only for its own equipment.
There are two answers to plaintiffs argument. First, with respect to the off-the-shelf items, the notes of the bidders’ conference make clear that the Government’s preference for off-the-shelf equipment was not meant to imply a sacrifice of equipment reliability considerations. In fact, just the opposite was true. The point was several times made that off-the-shelf equipment could be used provided it was built according to best commercial practices and exhibited satisfactory reliability characteristics. By way of example, the Government explained:
* * * A non-mil [i.e., non-military, meaning off-the-shelf equipment] console built to best commercial practices will be acceptable if there is a good indication that it will pass in Category I testing. This is the functional adequacy test of the equipment. If there is such a console built and you have statistical data of at least 6 months duration showing the actual reliability expected of the machine we will accept this in proposals as an indication of passing Cat. I tests. [Emphasis supplied.]
Given the above statement (one of several answers of like import made by the Government at the bidders’ conference), it is impossible to understand how Librascope could have come to the conclusion that use of off-the-shelf equipment, whether modified or not, should relieve it of the contractual obligation to furnish reliability predictions for such equipment. It might be added also that, quite aside from what was said at the bidders’ conference, the same conclusion could have been reached simply from the contract language. That part of the contract which stated the Government’s preference for off-the-shelf equipment (the clause previously quoted) went on to say that, if the proposed use of off-the-shelf items should necessitate tradeoffs in specification requirements, then "an explanation of their effect [i.e., the trade-offs] on the overall system is *313required.” Most certainly, this language should have told the contractor that the Government was as much interested in the performance characteristics of off-the-shelf equipment as it was in the contractor’s own equipment. Accordingly, to restrict reliability considerations only to the latter would make no sense at all.
As to the other part of the argument, namely, the claim that Librascope’s technical proposal restricted reliability predictions only to its own equipment, this too is without substance. In its technical proposal, Librascope had stated that "[t]he Contractor agrees to the inspections and tests required by paragraph 4 of MIL-D-27113” and its "reliability program will comply with MIL-R-27542.” These specifications, MIL-D-27113 and MIL-R-27542, were among the specifications referenced in the proposed contract which accompanied the Government’s request for proposals and, concerning these specifications, the Government had advised all bidders (at the briefing conference) that adherence would be mandatory — the specifications represented contract requirements. These same specifications set out, among other things, the requirement for reliability predictions. And this requirement was not limited, as the contractor would now have it, simply to contractor-manufactured equipment. The requirement was addressed to the system as a whole.
For example, paragraph 4 of MIL-D-27113 stated that the reliability of the equipment was to be tested in accordance with MIL-R-26474 and the latter specification, in turn, identified the requirement for reliability estimates in terms of "the total final equipment configuration.” MIL-R-27542 said the same thing, but said it even more clearly. It recited that "[w]here other equipment, such as Government furnished or associate contractor supplied equipment are integrated in the system to provide a complete operational system, the contractor shall use known or estimated reliability values for these equipments.”
In view of the representations made in its technical proposal that the reliability program would comply with MIL-D-27113 and MIL-R-27542, Librascope pursues an irrelevancy to argue, that since its proposal had only identified self-made equipment as the subject of required *314reliability predictions, its contractual obligation should be narrowed accordingly. The proposal was general, not specific or detailed, and if Librascope really meant to restrict its obligation, it should have said so in no uncertain terms. Whether, in that event, the proposal would have remained sufficiently responsive to the Government’s requirements to be considered acceptable is, of course, another matter. The point is, however, that the proposal said nothing in this regard. Thus, given the "outline” nature of the proposed along with the specific statements promising compliance with the applicable reliability specifications, the Government was entitled to assume that the proposal intended full compliance with all that those specifications entailed. In short, when the Government insisted that reliability predictions include the off-the-shelf equipment, it was insisting upon nothing more than what the contract and an honest reading of the proposal both required.
Claim 10.d
Preparation of Category I Test Plan
As part of its work under the contract, plaintiff prepared and implemented a program of tests designed to ensure that the equipment met Air Force requirements. This testing program, referred to as the Category I Test Plan, is the subject of plaintiffs Claim lO.d. Plaintiffs contentions are that, under the contract, its responsibility was limited to identifying and carrying out the tests that should be conducted and that the responsibility of actually developing and writing the test program represented extra-contractual effort which it was obliged to assume because of the Government’s orders. It contends too that this alleged extra work on its part was actually the responsibility of another contractor, IBM, to whom this work had been assigned by way of a separate contract.
Essentially, the same arguments were made below and all were there rejected. The Board decided as follows: (i) that the mentioned IBM contract pertained only to Category II testing and therefore was of no useful purpose in determining the scope of plaintiffs contractual obliga*315tions with respect to Category I testing; (ii) that, contrary to plaintiffs contention, paragraph 5.23 of Exhibit "A” (the basic statement of work) did. not impose upon the Government responsibility for preparing the Category I Test Plan; and (iii) that, during the course of performance, the conduct of the parties was such as to convincingly demonstrate that each regarded the preparation of the test plan to be plaintiffs responsibility. We agree with the Board’s several conclusions.
Addressing first the contention that the work in question was actually IBM’s responsibility rather than plaintiffs, there is no proof at all of this point. Plaintiffs brief does quote a paragraph of the IBM contract but we have no way of knowing whether the bracketed explanations — what plaintiff refers to as "internal notations supplied” — that plaintiff has added to the text of the quoted paragraph actually represents a correct reading of that paragraph. It would be for us nothing more than guesswork to accept the position that, in speaking of testing, the reference in the IBM contract to "OTC/IOC/COC phase of the system” actually identifies a requirement for Category I subsystem testing. Nor is our understanding bettered by reading the IBM contract in its entirety. So far as we can discern, that contract speaks only of Category II testing; if it does say more than this, then plaintiff has failed to prove it.
As to the language in plaintiffs own contract, the Board recognized that the relevant paragraph, paragraph 5.23, did not admit of any conclusion on the point in issue save the fact that that paragraph did not place upon the Government the responsibility for the development of the test program. (That paragraph simply said that "[t]he contractor shall conduct the necessary Category I testing * * * [and] [t]he 473L System Program Office retains responsibility for proper test design, conduct, evaluation and direction of these tests.”) The Board was of the view, however, that any deficiency or vagueness in the language of the contract was remedied by a course of conduct in which both parties acted from the common premise that the development of the test program was as much the contractor’s responsibility as was the identification and conduct of the tests themselves. In other words, the Board *316adhered to the familiar rule that contract provisions later brought into dispute should be accorded that meaning which is reflected in the parties’ conduct during performance and while their relationship was harmonious.
It would serve little useful purpose now to restate all the facts upon which the Board rested its conclusion. It will suffice to point out that plaintiff initiated the development of the test plan without Government prompting and, over a course of many months, undertook to develop that plan in depth. There was much correspondence and discussion between the parties on this matter, and during all this time no protest was ever raised by plaintiff claiming, as it does now, that it was being asked to do more than what its contract obligated it to do.
The contention raised in plaintiffs brief that it had initially agreed, "essentially as a courtesy to defendant,” to specify the tests that would be required for acceptance of the equipment is pure invention. Nothing in the record, be it testimony or documentary evidence, supports this assertion of limited responsibility gratuitously assumed. Equally devoid of substance is the argument that no negative inference should be drawn from the fact that plaintiff "did not continue to object” to the expansion of the test plan requirements. This too is fabrication; the real point is that plaintiff never once objected. Also in this connection, it should be noted that there is nothing in the record to support the argument that, at a meeting between the parties on December 11-12, 1962, the Government ordered plaintiff to prepare a Category I Test Plan. The notes of that meeting are devoid of any reference to a Government order to plaintiff; so too is the transmittal letter that accompanied these notes. This last says simply that "[tjhese notes may be used as a guide in writing the referenced paragraphs of the outline which was discussed.” The extensive examination of the parties’ actions and their correspondence that appears in the Board’s opinion fully supports the conclusion that the preparation of the Category I Test Plan was understood by Librascope to be its own responsibility.
*317Claim 3.c
Testing to Improve Reliability of Magnetic Tapes
This claim concerns the contractor’s contention that it was required to expend considerable time and money in an attempt to improve the reliability of the magnetic tapes— off-the-shelf items that were included as a part of the system’s so-called peripheral equipment. The Board denied the claim for failure of proof. Specifically, the Board said "[t]he documents show that the appellant made a substantial effort to improve the reliability of the magnetic tapes. They fail to show that this was at the insistence of the Government.” (73-2 BCA at 48,398.) We agree with the Board’s result but not with the reasoning supporting it.
Our difficulty with the Board’s decision begins first with the fact that, in connection with a different but related claim (Claim 9), the Board apparently acknowledged the very point which it judged to be missing from the contractor’s proof on the present claim, namely, that reliability testing was imposed upon the contractor and that this testing did include the peripheral equipment. Thus, in connection with that earlier claim, the Board, after finding that the contract did not include any minimum reliability figures (also referred to as mean-time-between-failure rates or MTBF rates), then went on to say:
* * * The Government ordered appellant to test the equipment to imposed MTBF criteria as a part of the Category I acceptance tests * * *. Appellant and the Government agreed finally upon the criteria to be met and the test methods. It [Librascope] did not agree that the peripheral equipments should be included in the tests. [73-2 BCA at 48,397; transcript citations omitted; emphasis supplied.]
Though not stated directly, the point plainly implicit in the above-quoted paragraph is that the Government’s order to test the equipment to imposed MTBF criteria included the peripheral equipment.
This same conclusion is also evident in what the Board had to say in regard to one of several related contentions included in plaintiffs Claim lO.d (the claim for the preparation of the Category I Test Plan). As stated by the *318Board, it was the contractor’s position that the "Reliability Testing Annex” to the Category I Test Plan (one of several appendices to the basic test plan) represented, in and of itself, an extra-contractual undertaking on the theory that the reliability requirements set forth in the test plan reflected the application of MTBF figures and reliability testing which the Government had unilaterally imposed. Simply put, the contractor’s argument was that the reliability test plan should not have been required for there was no contractual requirement for reliability testing in the first place. Of this claim, the Board said:
* * * This allegation with regard to increased reliability requirements was the subject of Claim 9. It [Librascope] should be allowed to recover under this claim for including the increased requirements in the Test Plan. We stated that this might be extra work as to peripheral equipments, but the record affords no method of allocating that part of appellant’s total cost of the test plan to the peripheral equipments. [73-2 BCA at 48,419; emphasis supplied.]
What is significant about each of the two above-quoted paragraphs, and the reason we cite them here, is that what the Board saw as enhanced reliability requirements imposed by the Government, i.e., its alleged imposition of minimum MTBF figures, had a direct impact upon peripheral equipment reliability — that equipment had to be tested to prescribed values and a plan for the testing of that equipment had to be written. Thus, if we accept the Board’s premise that the contract did not originally include MTBF figures and that the Government unilaterally imposed such figures upon plaintiff (and also their demonstration through testing) then the Board would clearly have been wrong in saying, in regard to the present claim (Claim 3.c), that the contractor’s proof failed to show that the reliability testing of the magnetic tapes was done at the Government’s insistence. On the facts of this case, the imposition of MTBF figures necessarily implied that all equipment in the system would be subject to reliability testing; indeed, this is the very point which the contractor now argues.
*319And it is this, then, which brings us to the second and more essential point in the Board’s decision. We disagree with what appears to have been the Board’s premise, namely, that the Government overstepped its contractual rights by requiring the contractor to meet and to demonstrate certain MTBF levels.
By the terms of the contract, plaintiff was, in fact, required to meet a specified mean-time-between-failures rate. True, the contract did not provide this information in terms of specific numbers but it did reference a mathematical formula from which the contractor could calculate the MTBF rate applicable to its equipment, this based on the number and types of components in the design.
The relevant documents were MIL-D-27113 (this was the basic military specification for the Data Processing Subsystem for the 473-L System) and MIL-R-26474. The former stated, in paragraph 3.2.1.5., that minimum reliability requirements "shall be in accordance with MIL-R-26474 except the part failure rate from the minimum mean-time-between-failures (MTBF) shall be obtained from the ASTIA AD-148868 (section 8).” The latter, MIL-R-26474, provided a formula for the calculation of minimum mean-time-between-failures. This latter specification also stated that:
When the detailed equipment specification [in this case MIL-D-27113] requires compliance with this specification, the final design and construction of the equipment shall provide as a minimum, the degree of reliability specified herein. The following requirements [which included MTBF values derived by formula] establish minimum levels which the contractor shall be required to provide and demonstrate.
The above-noted documents, together with the testimony of the Government’s witnesses on this issue, show that the contract not only mandated that the contractor meet reliability requirements but also that it indirectly specified the governing MTBF values. To the extent the Bpard held otherwise, its decision was erroneous as a matter of law. Denial of recovery for reliability testing of the magnetic tapes is affirmed on the ground spelled out above — such testing represented a contract requirement.
*320There is one last point that warrants comment. In its brief before the Board the contractor had argued that only MIL-D-27113 represented a contract requirement whereas MIL-R-26474 was claimed to represent nothing more than "a design objective or goal.” The Board did not address this issue (at least not in the claims presented in this appeal) nor did plaintiff raise the point here. However, in view of our treatment of this claim and our reliance upon MIL-R-26474, it is appropriate to give a brief answer to this point.
The argument is based, in part, upon the response to a question raised during the bidders’ conference. On that occasion, it was explained by the Government that those technical specifications, such as MIL-D-27113, which were specifically called out in Exhibit A (the basic statement of work) "will be contract requirements,” whereas those specifications that were referred to only in the technical specifications but not in Exhibit A — for example, MIL-R-26474 — were "to insure adequate design considerations and component quality for the elements that make up each equipment.” It is on the basis of this answer or explanation, coupled with the absence of any mention of MIL-R-26474 in its technical proposal, that the contractor made the argument before the Board that it was not obliged to meet the requirements spelled out in MIL-R-26474.
The argument is frivolous. The contention that MIL-R-26474 was intended to serve simply as a design objective or goal rests on a strained interpretation of the answer which the Government gave at the bidders’ conference. The contract required adherence to MIL-D-27113 and that necessarily meant adherence to all the specifications that it incorporated. To read the Government’s answer otherwise is to diminish the very purpose which the Government had identified with those incorporated specifications — the assurance of adequate design and component quality.
As to the contention that the contractor’s proposal did not specifically mention or promise compliance with MIL-R-26474, this point is also without substance. In Librascope’s technical proposal, under the heading of "Quality Assurance Compliance,” it was stated that "[t]he Contractor agrees to the inspections and tests required by paragraph 4 of MIL-D-27113.” The latter, i.e., paragraph 4 *321of MIL-D-27113, specifically stated that the equipment was to be subjected to preproduction tests including reliability testing performed "in accordance with MIL-R-26474.”
Claim 10.e
This claim was called out in plaintiffs petition but the briefs which plaintiff submitted offer no discussion of this claim. We assume, therefore, that the claim has been abandoned.
Claim 11
Interest Costs
The contractor claimed that, as part of an equitable adjustment in contract price, it was entitled to the reimbursement of interest costs associated with borrowings allegedly undertaken to finance the changes in issue. The Board denied the claim on the ground that a "necessity for borrowing” had not been shown. We agree with the Board’s result, but, in light of the argument that plaintiff raises here, find it necessary to expand upon what the Board has said.
In the development of its case before the Board, plaintiff showed that all the cash that it required in the performance of the instant contract was borrowed from its parent corporation, General Precision, Inc. and this parent, in turn, received its cash requirements from its own parent corporation, General Precision Equipment Corporation. It was further shown (and the Board so found) that General Precision Equipment Corporation supported9 the cash demands necessary for all three operations (i.e., its own separate activities as well as the activities of its subsidiary and those of Librascope) by borrowing from banks and other lending institutions. The operative relationship among the three corporations was described by the Board in these terms:
*322Appellant [Librascope] would be advanced money by General Precision Inc. for all its operations to use as operating capital at the beginning of the year without regard to specific jobs or anything of that nature * * *. In addition, the cash generated by Librascope Division would not be retained for its own use but rather returned and then received again from the corporate parent. * * * A similar practice was in effect between General Precision Incorporated and General Precision Equipment Corporation. [73-2 BCA at 48,425; transcript citations omitted.]
Thus, because of the pass-through arrangement described previously, the ultimate source of Librascope’s funds was the borrowings of General Precision Equipment Corporation. And, in light of that relationship, the Board observed that "it is interest on the borrowings of General Precision Equipment Corp. that we are concerned with under this claim.” (73-2 BCA at 48,425.)
In the argument that we are now asked to consider, the contractor invokes a rule to which this court and the contract boards have adhered: interest, absent its specific prohibition by contract or applicable procurement regulation, is recoverable as part of an equitable adjustment where the proof shows it to be a cost incurred in connection with monies required to be borrowed in order to finance a change in the contract. Traditionally, two types of factual situations have been encountered: those involving a specific loan undertaken in order to finance the changed work, see, e.g., Bell v. United States, 186 Ct. Cl. 189, 404 F.2d 975 (1968); Aeronca Manufacturing Corp., ASBCA No. 9173, 69-2 BCA ¶ 7811, and those involving contractors whose business practices included a course of borrowing from lending institutions. In the latter situation, the allowance of interest depended upon a specific showing that the course of borrowing was affected by the change in question, i.e., that apart from the normal borrowing pattern, there was a necessity to borrow specifically due to the change in question. Aerojet-General Corp., ASBCA No. 17171, 74-2 BCA ¶ 10,863. Though apparently not spelled out in the Board decisions, we perceive the reason for the last-mentioned requirement to be the need to provide for an identity of treatment between the contractor who relies *323upon his own equity and one who borrows generally. As to each, interest is a recoverable cost only to the extent that such interest is shown to be attributable to the need to finance additional work (or other extra costs incurred because of Government action) not contractually required at the outset of performance.
In this case, as we have already noted, the Board denied the interest claim because the proof failed to show a necessity for borrowing. Two points are said to invalidate this conclusion. First, it is claimed that the contractor’s necessity for borrowing is established by the fact that Librascope was entirely dependent upon outside funds. Consequently, any increase in Librascope’s cash needs necessarily spelled out a corresponding need for an increase in its borrowing. Second, it is claimed that essentially the same constraints operated at the level of General Precision Equipment Corporation — the ultimate funding source for Librascope. That company’s outstanding indebtedness during the 3-year period in question, 1962-64, was in the area of $50,000,000 annually10 and, in view of that magnitude of borrowing, it is claimed that no idle funds would have existed with which to support Librascope’s own substantial cash needs. We are asked to take judicial notice of the fact that in large corporations "[ijncome in the form of cash is virtually immediately used to reduce a company’s short term debt or invested in short term securities such as treasury notes. Cash is simply not allowed to remain idle because of the same considerations involved in this claim: it is recognized that interest charges áre a significant cost of doing business.” In essence then, the second argument is that when Librascope had to borrow, its funding source also had to borrow.
The first argument is factually correct but legally irrelevant. Librascope’s own necessity for borrowing is of no concern in this case because it was, in reality, nothing more than a cost center — it operated without any funds of its own and retained none of the profits that it generated. *324Its financial dealings were entirely intra-corporate; the interest obligation that it incurred was, in the last analysis, the obligation of General Precision Equipment Corporation.
The contractor’s second argument requires more discussion. In its decision, the Board said that it was "apparent that the cash flow position of General Precision Equipment Corp. is such that its 'course of borrowing’ could not possibly have been affected by the changed work.” (73-2 BCA at 48,425.) Accordingly, the Board found that the requisite necessity for borrowing had not been demonstrated. Although we agree with the Board’s result, we cannot endorse the reasoning behind it.
So far as we can tell, there is nothing in the record that would have permitted the Board to say what it did about the cash flow position of General Precision Equipment Corporation. Indeed, the absence of any proof on this point is fairly evident from the fact that the contractor now asks that we take judicial notice of the fact that large corporations do not maintain idle cash. However, even if we were to enlarge the proof in the manner suggested, it would not overcome the real problem in this record, namely, the lack of any meaningful evidence concerning the course of borrowing that was followed by General Precision Equipment Corporation. For example, the record does not disclose the nature of the various transactions that comprised the total of that company’s annual indebtedness (its course of borrowing), it does not disclose the company’s cash position or the changes in that position during the 3-year period in question nor does it disclose the manner in which Librascope’s capital demands were satisfied by General Precision Equipment Corporation— whether by cash (surplus), by new borrowings or simply by extending past borrowings through added interest payments. Without this type of information, it would be for us nothing more than speculation to accept the contractor’s position — that every increase in total corporate cash demand generated a corresponding increase in interest owed. The record as it stands leaves unanswered the point that matters most: that reasonable business needs dictated a necessity for borrowing in order to finance the compensa*325ble changes that occurred during Librascope’s performance of its 473-L contract.
The contractor tells us that this case is on all fours with the decision in Aerojet-General Corp., supra, where interest was recovered for the increase in borrowings brought on by the Government’s failure to timely make available contractually-required progress payments. Perhaps in broad outline the claimed identity does exist but the cases part on their proof. That case offered what this case does not: specific and detailed evidence showing in positive fashion that, as a consequence of the Government’s action, the corporation’s short term borrowing needs were increased. Had we such information in this case, the result would obviously be different.
One final observation is in order. The argument for the recovery of interest that the contractor presented to the Board and to this court was based upon the traditional notion that interest represents a cost item whose allowance as part of an equitable adjustment requires proof of a specific loan or proof of a necessity for borrowing. Thus, while we note that the Armed Services Board of Contract Appeals has recently taken what appears to be a new direction in the treatment of interest claims, see New York Shipbuilding Co., ASBCA No. 16164, 76-2 BCA ¶ 11,979 (allowing interest as an element of profit based upon the imputed value of the equity capital which the contractor had used in financing the changes in issue), our present decision intends no comment upon that recent Board decision. The arguments made there were not raised here and our treatment of the interest issue in this case was limited accordingly. The interest claim is denied because the necessity for borrowing by General Precision Equipment Corporation was not proven.
Claim 12
Claim Preparation Costs Including Attorneys’ Fees
In this claim the contractor seeks recovery, by way of an equitable adjustment in contract price, for the attorneys’ fees, technical consultants’ fees, and in-house personnel costs that were incurred in connection with the prepara*326tion and documentation of the claims for equitable adjustment that it presented to the contracting officer. The Board denied this claim, saying:
It is clear from the facts that the costs in question were not incidental to performance of work or alleged changes but were incidental to a claim. That the claim was to the contracting officer for an equitable adjustment rather than to this Board for the same relief is not significant. [73-2 BCA at 48,426.]
We agree with the Board’s decision.
The contractor does not dispute the proposition that legal fees associated with the prosecution of a claim before a contract board are not recoverable. Such fees fall within the statutory prohibition of 28 U.S.C. § 2412 (1970). Except as otherwise provided by statute, that section permits costs to be taxed against the United States, "but not including the fees and expenses of attorneys” in any civil action brought by or against the United States or any agency or official of the United States acting in an official capacity. Mindful of this restriction, the contractor contends here that the legal fees and other claim-related fees that it seeks are not litigation-connected but are, instead, expenses incidental to contract performance. In support of this contention, the contractor relies upon the decision in Allied Materials & Equip. Co., ASBCA No. 17318, 75-1 BCA ¶ 11,150. That case, says Librascope, permitted recovery of legal fees incurred in connection with the preparation and presentation, to the contracting officer, of a request for equitable adjustment. And, in so doing, argues the contractor, the Allied decision implicitly overruled such earlier decisions as Power Equipment Corp., ASBCA No. 5904, 1964 BCA ¶ 4025, aff’d on reconsideration, 1964 BCA ¶ 4228, which viewed requests for equitable adjustments (and the attorneys’ fees associated therewith) as the equivalent of claims prosecution rather than contract performance.
It is not disputable that the Board in Allied did allow the recovery of the attorneys’ fees connected with an equitable adjustment request made to the contracting officer. But, to say that in doing so the Board thereby intended to or actually did depart from prior law is to misread the *327decision. The equitable adjustment involved in that case addressed a situation where the contractor’s entitlement to the requested relief was well-nigh conclusive — the equitable adjustment was intended to compensate the contractor for the added performance costs attributable to the Government’s acknowledged failure to have made available the facilities and tooling that had been promised at the outset. Joined with this was the equally important fact that the request for equitable adjustment occurred in the midstream of contract performance and had, as one of its principal objects, the development of a new production schedule to accommodate the delays which the Government’s own actions had provoked. In short, the objective of the equitable adjustment in Allied was twofold: to pay the contractor what was owed in a situation where Government liability was not disputed and to ascertain the course of future performance with a view to securing the objective for which the contract had been let in the first place. The Board in Allied saw the situation no differently. It said:
In the instant case with the sale of the Government-owned plant and the machine tools involved there was never a question of the Government’s liability under the supply contract. The request for an equitable adjustment was on its face meritorious and, of course, it was incumbent in the first instance for the appellant to propose the details of the adjustment with respect to a new delivery schedule to accommodate the delay provoked by the Government action, to propose the price adjustment because of the subjective impact of such action upon the appellant, to propose a new plan of production and to alter the related terms of the contract. * * * That the appellant under the circumstances retained an attorney to present the adjustment was a prudent business decision and a reasonable one. * * *
Whatever the demarcation line may be between the ordinary interchanges. between a supplier and the Government as a customer which have inherent differences in point of view and a claim against the Government, we are satisfied in the facts before us that conflict between the parties as to the equitable adjustment never became so disputatious as to reach the level of a claim against the Government * * *. [75-1 BCA at 53,087.]
Plainly, the situation presented in the instant matter differs significantly from that encountered in the Allied *328case. Here, the claims for equitable adjustment were not presented to the contracting officer until all work had been completed, they addressed no situation in which Government liability was clear or apparent and, in content, they offered nothing that could reasonably be considered as benefiting the contract purpose. Judged both from the standpoint of the time of their submission and the purpose of their submission, Librascope’s requests for equitable adjustment were not performance-related; they bore no beneficial nexus either to contract production or to contract administration. Accordingly, the attorneys’ fees are not recoverable. As to the other claim preparation costs which plaintiff also seeks, these too are not allowable and for the same reason — they bear no relation to contract performance.
Claim Concerning Allowability of Plaintiff’s Requested Costs
Librascope’s final claim is that in determining quantum on the claims in issue in this appeal, we should adhere to the costing approach that the contractor relied upon rather than upon the adjustments to that approach which the Board used on those claims which it allowed. Since we find that the contractor is not entitled to recovery on any of the claims raised by this appeal, the contractor’s disagreement with the Board’s pricing technique becomes a moot question.
Singer-General Precision, Inc., Librascope Group, ASBCA No. 13241, 73-2 BCA ¶ 10,258.
Singer-General Precision, Inc., Librascope Group, ASBCA No. 13241 (June 5, 1974) (unpublished).
41 U.S.C. §§ 321-22 (1970).
For the sake of consistency, the contractor’s claims are identified in this opinion by the same claim numbers that were used in the proceedings before the Board.
Two of the ten L-119s are not relevant to this claim.
Plaintiffs witness testified at the hearing that the original design, i.e., four pairs of L-119s in the DPSS/AUTODIN interface, with two of them acting as backup devices for the other two, was developed in response to the requirement for multithread capability.
That report, dated April 15, 1963, states:
"Later it was shown that only four modules [L-119s] will be required (vice eight) to satisfy the system requirements. In consideration for this decrease in hardware Librascope has provided a redesign of remaining system elements pertaining to the L119’s and also has provided the major effort in interface definition with AUTODIN.”
Although it is unnecessary to discuss plaintiffs reliability argument, it should be noted that there is a significant difference between the absence of a contractual obligation to demonstrate reliability (the point plaintiff argues) and an obligation to design and deliver a system which could be expected to perform at certain reliability levels.
The word "supported” is the Board’s own. In light of the evidence, we take this to mean that some portion of total corporate cash demand derived from borrowing. The record does not support the idea that all of General Precision Equipment Corporation’s cash requirements stemmed from borrowing.
The specific figures found by the Board were as follows: 1962 - $50,522,000; 1963 - $60,228,000; and 1964 - $49,844,000. The exhibit from which these figures were taken does not show the nature of the indebtedness, i.e., whether long term or short term. The exhibit reports only that these figures included "mortgages payable.”