OPINION
NETTESHEIM, Judge.Having survived several motions for summary judgment that would have preter-mitted plaintiff’s day in court, see order entered on January 18, 1985 (motion for summary judgment; bench ruling); United Construction Co. v. United States, 7 Cl.Ct. 47 (1984) (motion for summary judgment), this case has been tried.1 The issue is whether the failure of an asphalt construction project was due to government specifications or lack of compaction and insufficient preparation of the subgrade. The parties have stipulated that recovery shall be in the amount of $250,887.83 if plaintiff prevails.
This case deserves to be put in a proper context by recognizing the significant and extensive efforts of counsel in achieving comprehensive factual stipulations. Counsel also showed a responsible approach to the evidence as it was developed, exemplified by defendant’s withdrawing its counterclaim during the course of trial. Nonetheless, defendant faced a formidable challenge when its expert witness candidly stated that he had no experience in asphalt design, other than in connection with Safeway markets 17 or 18 years ago; that he had not consulted before on a project involving the placement of asphalt for roadways or parking lots, such as the type of project involved in the litigation; and that this type of project is “definitely outside” his general line of work. This witness was *259also confronted with his- preliminary report of 1985 that pointed to a lack of quantitative evidence to show that the subgrade was poorly compacted at the time of placement. The report also suggested that the Army Corps of Engineers would have difficulty in showing that the design of the pavement was sufficient. Not surprisingly, the credibility of witnesses has played no small part in shaping the findings of fact.
The operative events took place between 1978 and 1980, justifiably prompting defendant to try to dispose of the case by motion before trial. One key player, Doyle Shelton, the Government’s construction representative who served as a site inspector, in the long interval before trial became the victim of age and illness, as revealed by his deposition testimony. See Deposition of Doyle Shelton, June 15,1985, at 29. The delay between completion of the contract and filing suit was not attributable to actions of the Government, and the parties were advised that testimony reflecting poor memory would not be considered in plaintiff’s favor. The documentary evidence, buttressed by testimony of percipient witnesses, helped fill the voids of faded memories, and the record includes sufficient evidence to assess whether plaintiff has proved its case by a preponderance of the evidence.
FACTS
The parties agreed to 95 stipulations of fact which are incorporated herein by reference. The stipulations are affixed to the official copy of this opinion, but will not otherwise be reproduced.
In 1978 plaintiff United Construction Co., Inc. (“plaintiff”), was awarded a contract by the United States Army Corps of Engineers (the “Corps”) for the construction of bituminous surfaced (asphalt) and aggregate surfaced roads, eleven asphalt parking areas, concrete boat ramps, and a water distribution system at Harry S. Truman Dam and Reservoir in Missouri. The contract amount was $2,887,661.80, and performance originally was to be completed between August 1978 and April 1979. The three asphalt parking areas involved in this case are the Long Shoal boat ramp parking lot, the Long Shoal maneuver area (collectively referred to as “Long Shoal” unless otherwise indicated), and the Berry Bend boat ramp parking lot (“Berry Bend”).
Three types of contemporaneous records were maintained on plaintiff’s contract performance. Danny Locke, plaintiff’s on-site quality control officer, prepared performance Quality Control Reports (“QCR”s); the Corps maintained Daily Logs of Construction; and Bob Gene Johnson, a former superintendent for plaintiff, kept a daily diary. It was stipulated that plaintiff’s QCR’s and the Corps logs generally do not distinguish between the Long Shoal boat ramp parking lot and the Long Shoal maneuver area, nor do Mr. Johnson’s diary and most of the testimony. As a result the findings for the most part do not distinguish between the two sites.
The parties agree that “cut” refers to the excavation of material on a hillside to a specified grade, and “fill” refers to the placement of material from a cut area or a borrow area into compacted lifts of a specified height or grade. It was also stipulated that beginning in August 1978 the Government’s Daily Logs of Construction recorded the dates and nature of preliminary work in clearing and grubbing the sites, excavation (excavating fill from a cut or borrow area), and placement of the embankment (placing material to bring the ground surface to design grade requirements for placement of asphalt).
The Long Shoal parking lot was largely a fill area on the lower portion of the lot, and both the Long Shoal maneuver area and Berry Bend were entirely fill. For each site the Corps designated borrow areas, or areas from which fill was to be obtained. Some of the fill for the Long Shoal parking lot was obtained from the upper-cut portion. The remainder of the fill for the Long Shoal parking Jot, as well as the fill for the maneuver area and Berry Bend, was borrow from the area that ultimately became Truman Reservoir.
*260Section 8 of the contract specifications entitled “Formation of Embankment” provides:
8.1 General: Embankments shall be placed and compacted in horizontal layers or lifts. Trees, logs, stumps, rubbish, standing or matted brush, matted roots, sod, unsuitable materials, frozen materials, or any other deleterious material or substances shall not be placed in embankments. Holes left from removal of existing structures, and grubbing operations, shall be backfilled and compacted, unless otherwise specified. Additional lifts shall not be placed on embankment where the previous lift, contains frozen materials. Equipment of such weight or so loaded that it might cause damage to culverts or other structures shall not be permitted to work over or immediately adjacent to such structures until reasonable possibility of damage thereto no longer exists. Where embankments, regardless of height are placed against hillsides or existing embankments, either of which have a slope steeper than one vertical to six horizontal, the existing slope shall be benched or stepped as the new fill is brought up. The depth of the benches or steps shall be approximately equal to the lift being placed. In the event rock is encountered when benching, the benching operation shall be omitted, and the overburden stripped to firm rock line to permit fill to be placed in direct contact with the exposed rock surfaces.
8.2 Earthfills: Each layer or lift or material shall not exceed 8 inches in thickness loose measurement and shall be compacted to not less than the required density before the next layer is placed thereon. Blade graders and bulldozers shall be used on each lift to remove mounds and ridges caused by dumping operations and to obtain uniform thickness prior to and during compacting.
8.3 Combination earth and rock-fills: Layers or lifts formed of rock, shale, or other hard materials and earth materials that cannot be satisfactorily compacted in layers of 8 inches, shall be placed in uniform layers not to exceed 24 inches in thickness (loose measurement). The larger stones shall be well distributed with the voids filled with smaller stones or earth. Rocks or boulders too large to permit placing in a 24-inch lift shall be broken as necessary or so placed that proper compaction is obtained around such boulders. Placement methods shall be such that all voids are filled.
The fill was placed in six- or eight-inch lifts under the supervision of John W. Bled-soe, plaintiff’s vice-president in charge of grading. Plaintiff compacted each six- or eight-inch lift as the embankment was constructed and graded, so that the surface would not hold moisture before the crew left the site for the winter.
In October 1978 plaintiff completed constructing the embankments, including rough grading, and work was suspended. The parties stipulated that final preparation for laying asphalt and the actual laying of asphalt cannot be completed during winter months. Plaintiff’s president James William Supica, Sr., was of the view that it was a good idea to build an embankment and leave it over the winter in order to allow it to equalize and adjust to its new environment.
Beginning on May 10, 1979, plaintiff began preparing for the placement of asphalt. This procedure consists of working the sub-grade until it is sufficiently firm to accommodate placement of the base course. Section 13 of the contract specifications provides:
RECONDITIONING OF SUBGRADE shall be accomplished prior to laying the bituminous base course. Any ruts or soft yielding spots that may appear in the subgrade, any areas having inadequate compaction, and any deviation of the surface from the requirements specified shall be corrected by loosening the affected areas, removing unsatisfactory material and adding approved material where required, reshaping, and recom-*261pacting to line and grade and to the specified density requirements.
At this point rainfall data takes on a critical role. Section 12 of the contract specifications entitled “Weather Limitations” provides:
Bituminous base and surface course shall be constructed only when the underlying surface is dry and when the weather is not rainy. Bituminous courses shall not be constructed when the atmospheric temperature is below 40° F. In addition, unless otherwise authorized by the Contracting Officer, bituminous mixtures shall not be placed unless the ambient temperature has been above 32° F. for a continuous period of 36 hours prior to placement.
In recording rainfall on plaintiff’s QCR’s, Mr. Locke took his data from a gauge at plaintiff’s job trailer at Clinton, Missouri, approximately 17 miles north of the subject work site. Corps construction inspector Dale A. Bestgen2 testified that he completed his Daily Logs of Construction based on the QCR’s. Mr. Johnson, plaintiff’s superintendent during the 1979 grading and placement of asphalt, recorded in his diary whether or not it rained on site. At trial defendant relied on the United States Department of Commerce Rain Gauge in Warsaw, Missouri, located approximately five miles east of Long Shoal and seven miles east of Berry Bend. All witnesses who had lived and worked in the area viewed rainfall as a sporadic phenomenon tending at a given time to be present in one area and absent in a proximate area. However, given the importance of rainfall in this case, it is necessary to have an official and proximate rain gauge, and the one at Warsaw fulfills this purpose.
In May 1979 the Warsaw gauge registered 1.10 inches of rain on May 3 and .12 inch on May 4. Plaintiff began blue-topping the subgrade (cutting the compacted fill to grade) at Long Shoal on May 10. On May 11, 1.09 inches of rain were recorded. Fine grading continued on May 14-18. On May 14 and 16, .06 and .08 inch were recorded, respectively. The gauge reported 1 inch of rain over May 19-21. On May 22-25 plaintiff continued fine grading. On May 27 .04 inch was recorded. On May 29 the fine grading was completed.
The contract specifications directed that the subject areas be constructed by placing two two-inch lifts, or layers, of asphalt base course directly on the subgrade and one two-inch lift of asphalt surface course on top of both lifts of the base course.
Plaintiff commenced placing the asphalt base course on the Long Shoal parking lot on May 31, 1979. During the preceding nine days (May 22-30), the Warsaw gauge recorded only .04 inch of rain on May 27. Plaintiff continued laying the base course at Long Shoal between May 31 and June 2. On June 1 the Warsaw gauge recorded .01 inch of rain. There was rain on every day between June 5 and 10, ranging from .02 to .95 inch. The base paving continued on June 6 and 7; fine grading resumed on June 11 and 12; .24 inch of rain was recorded on June 13, with .16 inch on June 14; paving continued on June 14 and 15; there was .07 inch of rain on June 17; and paving at Long Shoal was finished on June 18.
Mr. Johnson testified that as of June 1 he was getting slippage on the material in certain areas (it would not lay properly); that it was pumping water from beneath; and that the base course was unstable. His diary reflects: “Talked to Corps Engr. about lay base in 4" lift.” Mr. Johnson’s testimony is supported by the contemporaneous notations in his diary. Corps inspector Bestgen did not dispute Mr. Johnson’s testimony that the base course initially failed on June 1, although Mr. Bestgen’s log for that date reflects no incident regarding asphalt. As noted in Mr. Johnson’s diary, Messrs. Supica and Bledsoe *262were present on June 1, and they testified about what they observed. The Corps representative D.J. Bell, who was also present, did not testify.
Mr. Johnson’s diary and Mr. Bestgen’s log confirm that on June 2 plaintiff hauled base asphalt until the laying machine broke down. Mr. Johnson’s diary for June 5 reflects that there was rain the previous night and that it was too wet for laying asphalt. June 6 was an important date. Mr. Johnson reported that base asphalt was being laid on Long Shoal, that the Corps was out looking at the base asphalt, and that the two-inch first course was slipping on the subgrade. Mr. Bestgen’s log (he signed the log, although the pertinent portion was written by a Chuck Fowler), says that plaintiff dug out and removed broken-up asphalt in the first base course and one-inch of the subgrade in four different soft spots and replaced the spots with asphalt base. More precisely, the parties stipulated that on June 6 at the Long Shoal parking lot 2,180 square feet of the first two-inch base course were removed from four different soft areas, because the asphalt had cracked or rutted randomly and the pavement had broken up. It was also stipulated that where the pavement had broken up, moisture was observed in the subgrade.
The testimony and notes do not dovetail at roughly this point. Mr. Bestgen was on site at the Long Shoal parking lot on two dates, but he does not remember which. He recalls that on one occasion plaintiff’s paver became stuck during placement of the first two-inch lift: “The subgrade was pumping and yielding quite a bit under the pressure of the trucks and under the paver and it was my opinion above the optimum moisture content, [because] it was very wet. No free moisture appeared, but it was very wet.” Mr. Bestgen directed plaintiff to stop and use a motor blade to dig out deficient areas and to “wind-row,” or dry, them. Mr. Bestgen returned on the second day and observed plaintiff’s crew trying to dry and recompact the lift with a steel drum roller, although it was customary, in his opinion, to use a sheepsfoot roller. According to Mr. Bestgen, a steel drum delaminates a lift, whereas a sheeps-foot penetrates it.
On June 7 Mr. Johnson’s diary noted that it showered in the morning and that asphalt operations started at 9:00 a.m. He recalled that the rain did not hold up the operation, because the crew had to use the water truck from time to time to add moisture. Mr. Johnson testified that, after the rain on June 8, the subgrade was prepared for laying asphalt on June 11-12. His diary reflects that due to rain it was too wet to lay asphalt on June 13, that on June 14 the asphalt base at Long Shoal was completed, and that on June 15 the crew began laying the surface course of Long Shoal, which continued on June 16 and was finished on June 18. The Corps logs are consistent with this testimony and Mr. Johnson’s diary entries.
Plaintiff graded Berry Bend intermittently between May 30 and June 18, 1979. Grading proceeded on every day other than June 3, 8-10, 12, and 17. Rain was recorded on five of the days when grading was underway. Base paving was accomplished between June 19 and 21. The finish course was not attempted. The highest rainfall of the month was recorded on June 20. The Warsaw gauge recorded rain on June 1 at .01 inch; June 5, .92 inch; June 6, .02 inch; June 7, .33 inch; June 8, .18 inch; June 9, .95 inch; June 10, .51 inch; June 13, .24 inch; June 14, .16 inch; June 18, .07 inch; and June 20, 1.67 inches. On June 5 when .92 inch was recorded, Mr. Bestgen noted on the Daily Log of Construction, “Sprinkled a little off and on[;] did not affect work.” On June 20 both Messrs. Locke and Johnson and the Corps Daily Log of Construction noted that the rain (the Corps Log says “last night’s rain”) had postponed laying asphalt at Berry Bend until 12:00 p.m. and that, due to rain, no other work was done other than installing a water line at another site.
The asphalt base course on the Long Shoal parking lot was placed in separate two-inch lifts. Before the asphalt was laid *263at the Long Shoal maneuver area and Berry Bend, Mr. Shelton directed plaintiff to apply MC-70 prime, an asphalt oil material, to stop the moisture. Subsequently, the Corps directed plaintiff to place the base course in one four-inch lift at the Long Shoal maneuver area and Berry Bend.
On June 21 Mr. Johnson’s diary reflects that the asphalt continued to be laid and that Corps inspector Shelton came in to look at the grade and condition of the ground. The diary notes: “Doyle Shelton marked areas to remove on big parking lot at Berry Bend[;] [a]lso told us to remove 3" or 4" below subgrade and replace with asphalt.” Mr. Johnson reported for June 22: “Moisture [came up] in small parking lot [the “Berry Bend beach lot”] from the bottom and made soft areas____” Plaintiff has made no claim with respect to the Berry Bend beach lot, but it is a factor in this litigation. Mr. Shelton, according to Mr. Johnson’s diary notations for June 22, inspected the Berry Bend beach lot and instructed Mr. Johnson that it was plaintiff’s responsibility to repair it. The diary for June 23 reflects, as Mr. Johnson testified, that he spoke with Mr. Supica about the Berry Bend beach lot and that it was decided that a ditch would be dug above the lot to intercept water before it reached the lot. Mr. Supica testified that his response to the problem involving excess moisture was to dig intercept trenches beginning at the Berry Bend beach lot and continuing on all subsequent sites. Billy J. Cheatham, the Corps Resident Engineer for the Harry S. Truman Construction Office since February 11, 1979, later instructed plaintiff to fill the trenches because they did not provide adequate drainage.
On rebuttal Mr. Supica took issue with Mr. Cheatham on the basis of a contour chart depicting the location of trenches and topography of the sites, demonstrating that the trenches daylighted, i.e., drained. The only intercept trench impacting this litigation is one at the Berry Bend beach lot, because it was in place during the time plaintiff was reworking this subgrade per Mr. Shelton’s instructions. On this basis it cannot be inferred that Mr. Shelton’s diligent monitoring of plaintiff’s fine grading and laying of asphalt at the Berry Bend beach lot assured proper preparation of the subgrade and grading, because the pavement at Berry Bend beach lot may not have failed due, in part, to the intercept trench. Mr. Johnson testified that this ditch caught moisture. Moreover, as reflected in the Corps log of June 22, the Corps determined that the subgrade was bad at the Berry Bend beach lot before any asphalt was laid. The problems at Long Shoal and Berry Bend arose during and after the placement of asphalt.
Mr. Johnson’s diary continues on June 25, noting the digging of the intercept trench above Berry Bend beach and the reworking of that lot. On June 26 more moisture came up from the subsoil in the Berry Bend beach lot, forcing plaintiff to rework the area again.
There was rain on June 28, and Mr. Johnson’s notation for June 29 is “wet,” recording that patching was continuing at Berry Bend. On July 2 and continuing on July 5 and 11, the patch areas were drained and dried by reworking the subgrade with rollers and a blade. On July 11 the Corps directed plaintiff to backfill areas to the top of the subgrade with fill instead of asphalt. On July 12 Mr. Johnson noted that Corps representatives were looking at the asphalt and the moisture coming into the parking lots. Mr. Johnson’s diary says that Mr. Cheatham instructed that no more asphalt be laid until Anderson Engineering tested the subgrade. On July 18 plaintiff began laying a four-inch base course on the Berry Bend beach lot. This preparation was completed on July 19, and plaintiff commenced work on the surface course. The surface course was finished on July 20. The Corps logs are consistent with Mr. Johnson’s diary.
Section 9 of the contract specifications, which sets forth the compaction requirements, provides:
9.1 General: Materials, except materials consisting wholly or primarily of rock, and all suitable materials used in *264backfilling shall be compacted to at least 95 percent maximum density. The top 6 inches of undisturbed soil in the sub-grade shall be compacted to at least 95 percent of maximum density____
(Emphasis added.)
The parties stipulated that Anderson Engineering conducted the asphalt base tests during May and June 1979, “as required by the contract.” Because plaintiff maintains that the material was too rocky to test, it is assumed that plaintiff acceded only to the manner of testing. Mr. Locke said that the field work for Anderson’s tests was proper. Four of the six tests taken at the Berry Bend boat ramp revealed asphalt densities of less than the 95 percent required by the contract. Two of the seven compaction tests taken at the Long Shoal maneuver area, and one of the nine compaction tests at Long Shoal parking lot revealed asphalt densities of less than 95 percent. On June 22, 1979, and July 11, 1979, the Corps tested three samples from the subgrade at Berry Bend, all of which showed subgrade densities below 95 percent. The June 22, 1979, test was repeated on June 26, 1979, and it revealed a density above 95 percent.
Mr. Locke’s QCR’s reflect consistently from August 17, 1978, that he considered the material too rocky to test, although it was the tests that he performed which formed the basis of his opinion that the material was too rocky. The parties stipulated that both plaintiff and the Corps took density tests in September and October 1978 that showed densities of 95 percent or above. Messrs. Locke and Bledsoe testified that Corps representatives, specifically Fred Hurst, told them that the subgrade material was too rocky to test. Mr. Hurst, who testified by deposition, was the assistant chief of the laboratory at Harry S. Truman Dam during 1979. It was his job to make check tests at the instruction of the job inspector, usually before asphalt was laid. His testimony describes the type of test generally as what the parties stipulated was a “sand-cone density test.” For this test material is scooped from an area of compacted material and weighed, and the hole is filled with uniform sand of a known density. By calculating the weight of the removed material (accounting for moisture content) in relation to the volume and weight of the sand, the density of the removed material (prior to its removal) is determined. Although Mr. Hurst recalled that the area near the three sites was rocky, he could not recall whether the rocky condition of the soils prevented density testing. “There are some of them sites I know was too rocky to take tests, but I couldn’t tell you whether it was Long Shoal or not.” Deposition of Fred Hurst May 23, 1985, at 8. The bottom line is that Mr. Hurst’s memory failed him, not surprisingly since he was testifying over six years after the fact. The issue whether Mr. Hurst said anything that would bind the Corps is to be resolved against plaintiff.
Mr. Locke presented himself as an intense witness and was cautious to give precise testimony. His QCR’s state unvaryingly that the material was too rocky to test. It is not disputed that Mr. Locke gave his QCR’s to the Corps, and there is no evidence that the Corps quarrelled with Mr. Locke’s opinion of the rock content. Plaintiff and the Corps were taking tests in 1978, although it is unclear whether Mr. Locke or a Corps employee was doing the work (see Daily Log of Construction Aug. 28, 1978; Sept. 1, 6, 13, and 14, 1978). The Corps did rely on these tests (whoever took them) to approve the work at several sites. Testing did not resume in 1979 until after the asphalt failed at Long Shoal. Although the Corps did not waive the requirement that the subgrade be tested unless it was too rocky, it was on notice throughout the fall of 1978 and from May-July 1979 that Mr. Locke deemed the subgrade too rocky to test.
The court’s on-site inspection conducted during the first day of trial sheds scant light on the composition of the subgrade, even though the material that was used as fill was observable. Defendant’s expert witness easily pulverized some of this material between his hands.
*265Mr. Locke testified that there were two types of tests that could have been taken. One was the sand-cone density test. The American Society for Testing and Materials Standard (“ASTM”) “Standard Test Method for Density of Soil in Place by the Sand-Cone Method” restricts the test to particles less than two inches in diameter. Mr. Locke testified that the particles at the sites exceeded this range. The Procter or Laboratory Compaction Test also measures the moisture-density relationship. Section 5 of the ASTM “Standard Test Methods for Moisture-Density Relations of Soils and Soil-Aggregate Mixture” provides:
If 30% or more of the sample is retained on a s/4-in. ... sieve, then none of the methods described under these methods shall be used for the determination of either maximum density or optimum moisture content.
Mr. Locke testified that the material was too rocky under this standard. Messrs. Supica, Johnson, and Bledsoe also considered the material too rocky to test. Defendant does not dispute that the ASTM standards tell whether a test can be made, and it is found that the standards do not contemplate that a sample is pulverized before it is tested. Moreover, beginning on August 4, 1980, during reconstruction of the areas, compaction tests were taken frequently. The Corps log for August 11, 1980 reflected, “The contractor took 2 compaction tests today[,] but I don’t think you will be able to tell much from them because of the rock in them____”
Dr. Roy Leonard, testifying for plaintiff, was qualified as an expert in geotechnical engineering, which involves investigations and advice on anything constructed in or on soil or rock. Dr. Leonard’s background includes extensive experience in and consultancies regarding pavement design. He testified based on his on-site observations that the rock content was “pretty high—approaching 50 percent.” According to Dr. Leonard, over 30 percent of the soil would have been retained on a 3/4-inch screen. He also ruled out the sand-cone density test because ten to 50 percent of the soil on site had sizes greater than two inches.3 Defendant’s witness Dr. Allen W. Hatheway, eminently qualified as an expert in engineering geology, did not give an opinion as to whether the soil was too rocky to test.
Plaintiff’s testimony on this issue overcomes defendant’s. The material was too rocky to test. This finding does not relieve plaintiff from complying with the specifications prescribing the manner and degree of compaction; it merely removes the onus of satisfying the recognized field and laboratory tests. Nor does this finding imply that the subgrade material was not “soil.” Section 5 of the specifications entitled “Utilization of Excavated Materials” indicates at section 5.1 that “suitable materials” include “rock materials.”
Mr. Cheatham directed plaintiff to remove and replace the asphalt at Long Shoal and Berry Bend on September 18, 1979, and removal of the asphalt at Long Shoal commenced on August 4, 1980. In May and June 1979, the Warsaw gauge recorded 3.49 and 5.88 inches total rainfall, respectively. In August and September 1980, when the repair work was completed, total rainfall was 5.58 and 2.07 inches, respectively. On August 5, 1980, the Warsaw gauge recorded 1.98 inches of rain, followed by .40 inch on August 6. Plaintiff began grading at Long Shoal on August 13 and 14. On August 15, 1.74 inches of rain were recorded, with .38 inch on August 16. Grading resumed on August 18-22, during which there were .03 inch of rain on August 18 and 1.05 inches on August 21. Plaintiff graded from August 25-27 and also laid the base course on August 28-29. There was no rain during this period. On September 2, when plaintiff began the finish course, there was .68 inch of rain; there was no rain on September 3 and 4; the finish course was completed on September 4.
*266After the asphalt had been removed beginning on August 4 or 7, 1980, grading commenced at Berry Bend on August 18-20, with .03 inch rain on August 18. On August 15 there had been 1.74 inches of rain and .38 inch on August 16. Grading continued on August 24. It resumed on September 2 when there was .68 inch of rain and was completed on September 5, with 1.1 inches of rain. The base and finish courses were laid on September 6, 8-10, with no rainfall.4
Plaintiff presented an exhibit also based on data at Warsaw plotting precipitation in 1979 and 1980 against average annual rainfall. Between January and June 1979, the rainfall was above average until mid-July. Throughout the period during which Long Shoal and Berry Bend initially were graded and paved, the rainfall was above average by approximately .066 inch. Between January and mid-May, the rainfall was over two inches above average. In contrast, the precipitation from mid-April through the end of December 1980 was markedly below average. During the period Long Shoal and Berry Bend were reconstructed, the rainfall was over ten inches below average.
Defendant elicited on cross-examination of Robert J. Bowerman, an engineer/estimator employed by plaintiffs related United Bridge Co., between May 10 and June 16, 1979, when Long Shoal was graded and paved, there were 5.66 inches total rainfall, in contrast to 6.26 inches total when Long Shoal was reconstructed between August 4 and September 4, 1980. When Berry Bend was originally graded and paved between May 30 and June 21, 1979, there were 5.06 inches total rainfall, in contrast to 3.99 inches during reconstruction between August 8 and September 10. Nonetheless, plaintiffs witnesses testified that a typical rainfall in the area is sloughed off and not absorbed. Even though it rained only slightly less in August 1980 than in June 1979, for example, the Corps log for August 11, 1980, reflects that the “subgrade looks good and tight. However, it could stand a little water on it.”
In reworking and preparing the sub-grade during reconstruction in 1980, plaintiff used a disc, a piece of heavy equipment that blades into the earth vertically and churns it up. Section 9.2 of the contract-specifications governing “Moisture content” provides:
Before rolling is started, each layer shall be dried by aeration or have moisture added as necessary to permit compaction to the required density. The moisture content shall be essentially uniform throughout the lift. Water shall be added to the soil by an approved method. Soils that contain an excess amount of moisture shall be permitted to dry, or shall be disked, harrowed, bladed, and aerated as necessary. If the top or contact surfaces of a partial fill section become too dry to permit suitable bond with additional fill to be placed thereon, the dried materials shall be loosened by scarifying or disking to such depths as will reach through the dried materials. The loosened material shall be moistened to an acceptable moisture content and recompacted in accordance with the applicable requirements of these specifications.
(Emphasis added.)
During September-October 1978, Mr. Bledsoe’s crew used a Hyster Embankment Compactor with a scarifier in compacting the embankment. A scarifier is attached to a motor grader or blade and loosens up elevated spots or aerates the subgrade soil. It contains 18-inch teeth if placed on the rear of the machine or six- to ten-inch teeth if placed on the front. After the freeze/thaw cycle, Mr. Johnson’s crew in May 1979 moved in to regain compaction, utilizing blading equipment, the Hyster Embankment Compactor with a blade (not a scarifier), and water trucks. Next they established blue-top staking, or tolerance staking for the subgrade, using a bulldozer *267with a blade, a scraper, and the embankment compactor. A blade without a scarifier was used to correct soft yielding spots. Plaintiff used a CMI Fine Grade Trimmer, which has a pulverizing or scarifying action, for fine grading. The blade without the scarifier was used to dry the soil when the problem with moisture developed on June 1, 1979. Mr. Johnson viewed a blade and a disc as comparable; Mr. Bledsoe, who agreed at trial, had testified on deposition that he would use a disc if the sub-grade were wet.
Messrs. Johnson and Supica testified that Mr. Shelton inspected and accepted the sub-grade for laying of the pavement at Long Shoal. Defendant so stipulated, preserving its position that Mr. Shelton did not have authority to waive contractual provisions. Defendant asserted that Mr. Shelton was ignorant of all pertinent facts when he gave his approval. Mr. Johnson’s diary for May 31, 1979, reflects that Mr. Shelton inspected the Long Shoal parking lot and approved the grade. According to Mr. Shelton’s deposition, he checked the grade at Long Shoal and although he could not ascertain the compaction, “[T]he grade was correct for evenness and uniformity, and was firm.” Shelton Dep. at 16. Mr. Shelton was on leave for six days before the date on which the base course was laid at Berry Bend, although he testified that he had accepted the subgrade at that site. Mr. Johnson’s diary is silent concerning any inspection of the grade at Berry Bend.
General Provision 10 of the contract entitled “Inspection and Acceptance” provides:
(a) All works ... shall be subject to inspection and test by the Government at all reasonable times and at all places prior to acceptance. Any such inspection and test is for the sole benefit of the Government and shall not relieve the Contractor of the responsibility of providing quality control measures to assure that the work strictly complies with the contract requirements. No inspection or test by the Government shall be construed as constituting or implying accept-ance____
$ })c $ }jc $ 4c
(f) Unless otherwise provided in this contract, acceptance by the Government shall be made as promptly as practicable after completion and inspection of all work required by this contract, or that portion of the work that the Contracting Officer determines can be accepted separately. Acceptance shall be final and conclusive except as regards latent defects, fraud, or such gross mistakes as may amount to fraud or as regards the Government’s rights under any warranty or guarantee.
(Citation omitted.)
Mr. Shelton’s testimony on deposition generally was vague, but he specifically recalled approving the subgrades at Long Shoal and Berry Bend. Although it is found that no Corps representative waived the requirements of the specifications by approving the subgrade, there is no evidence that the subgrade was not properly compacted—either during construction of the embankment or in reconditioning the subgrade before placement of asphalt. Because the material composing the subgrade was too rocky to test, the subgrade could not have passed muster by testing. Mr. Shelton could have gauged compaction in some other manner, but said that he did not. During May-June 1979, the subgrade had been subjected to heavy traffic before paving. For example, the Hyster Embankment Compactor with a blade, which was also used in constructing the embankment, weighs almost 54,000 pounds. The asphalt failed when a 24,000-pound Hyster Steel Wheel Roller was rolled over it to achieve compaction. The testimony is unequivocal that the heavy equipment passing over the subgrade did not cause any failure in the subgrade until asphalt was laid and compacted. No water was visible in the sub-grade at Long Shoal or Berry Bend to anyone who testified or wrote reports in May-June 1979, until after paving operations had begun.
*268Plaintiff’s president, Mr. Supica, with approximately 34 years of experience in highway construction, was an ardent witness who took pains not to stretch his testimony where his memory was not precise. Although a partisan, Mr. Supica showed himself to be a responsible contractor in the sense that “responsible” means an individual who takes pride in completing a contract and turning over a job well done. Plaintiff had completed $20 million in roadway construction projects for the Corps within the Harry S. Truman Dam and Reservoir, half of which were completed before 1979. Mr. Cheatham, as Resident Engineer who represented the Contracting Officer, was Mr. Supica’s counterpart in the Corps. He testified as a conscientious manager, but he was not on site at Long Shoal until after the parking lot failed. Mr. Supica had the more credible testimony about whether the intercept trenches could drain and whether the ditch that the specifications called for at the Long Shoal parking lot was in place. Of all the witnesses, Mr. Supica also had the most comprehensive command of the comparative uses and values of construction equipment. For example, he testified persuasively that plaintiff’s use of a Hy-ster Embankment Compactor in compacting the subgrade prior to placing asphalt in 1979 was preferable to a sheepsfoot roller because the former is heavier and consequently can produce a more compactive layer.
On July 12, 1979, B.A. Anson, one of the construction representatives or inspectors who completed and signed the Corps logs during the work at Long Shoal and Berry Bend, toured, along with other Corps personnel, the parking lots at Long Shoal and Berry Bend. The group sought to determine the reasons for the faults and defects “[i]f due to subgrade, placement or mix.” Mr. Anson wrote: “Great evidence [of] ... free water between subgrade and black base____ Evidence of pumping action in compacting base____”5 Frank Bader, then Chief of Dams and Foundations for the Corps, accompanied Mr. Anson on the inspection. In September 1979 Mr. Bader was asked to look at the lots with problems. He produced a report dated October 5,1979, recording his inspection and conclusions, stating in pertinent part:
3. Inspection Procedure and Results: The appearance of all of the parking lots except the three in question was excellent. There was no evidence of cracking or surface depressions. In order to investigate and compare the sub-grade under both acceptable and unacceptable parking lots, 3-inch diameter auger holes were drilled by the contractor through the swimming beach parking lot at Berry Bend (one of the acceptable lots). Five or six holes were drilled at the boat launching parking lot at Berry Bend and seven or eight holes at the large Long Shoal boat launching parking lot. (Two of the unacceptable lots.) No holes were drilled at the other unacceptable lot at Long Shoal. At the time the holes were drilled, no free water was found in any of the holes, some of which were in fill only, some penetrated through the fill into the natural foundation and some were in cut sections only. The subgrade material were variable (clay, gravelly sandy clay, clayey sandy gravel, gravel, and weathered bedrock). Although auger holes do not permit a very reliable estimate of density, the following general observations were made by observing the drilling and inspecting the sides of the hole. All three holes in the Berry Bend swimming beach lot contained material that seemed to be well compacted and had water contents near optimum or slightly above. The holes in the Berry Bend boat launching lot were in general more moist, particularly in the fill areas, but probably within 3% over optimum. Three holes in fill at the lower (east) end of the Long Shoal boat launching lot were probably more than 3% over optimum water content and the density seemed to be somewhat low. On 27 September, one hole approximately 35 feet *269west of of the SW corner of the Long Shoal boat ramp had free water on the bottom and side of the hole.
4. Conclusions: The existing problems at the three unacceptable parking lots appear to be the result of poor sub-grade preparation rather than a design deficiency. The three unacceptable lots were the first lots surfaced in this contract and there had been rainfall during this surfacing period that may have contributed to the problem by softening the subgrade. Once it became apparent that these failures might continue, the Resident Engineer insisted that no surfacing was to be done until the subgrade was formally approved. Since that time there have not been any more unacceptable lots. Whether or not future problems will be encountered in the presently acceptable lots is unknown. Based on this inspection and observations by FC-HT personnel onsite at the time of earthwork construction and at the time of placement of the base course asphalt, the existing parking lot failures were the result of improper placement of fill, lack of compaction and insufficient subgrade preparation. We have no evidence to support a design change to correct apparent construction deficiencies, in these parking lots when additional lots were paved under conditions supervised by the RE and found to be satisfactory.
In June 1980 Mr. Bader was asked to take samples from the three failed sites, which was accomplished on July 14-16, 1980. His report dated September 9, 1980, states that “15 borings were drilled, in a shelby-tube sampling.” By this procedure a thin-walled metal tube with a sharpened edge is pushed into the soil under constant hydraulic pressure to obtain an undisturbed sample of earthen material. Mr. Bader wrote, “These samples are not truly undisturbed samples because the large amount of gravel in the samples makes sampling difficult and many of the samples were either compressed or had some loss of material during drilling.” Density tests were run on 14 samples. Laboratory compaction tests were run on a composite sample, since there was insufficient material to run a test on each sample. Because of these difficulties with the samples, Mr. Bader concluded, on the basis of his visual examination and the density and compaction tests, that evidence was present of “trends” indicating that the material under the “bad” areas was more moist and “softer” than the material under the “good” areas. Mr. Bader confirmed his earlier conclusion that the problem at the three sites was due to “poor subgrade preparation, primarily insufficient compaction and excessive moisture.”
Mr. Bader admitted that the Shelby tube samples were disturbed. Moreover, the laboratory compaction tests were performed because the technicians were of the view that the material was less than 30 percent gravel of larger than %-inch size. Although Mr. Bader is entitled to rely on his technicians’ judgment on the suitability of the materials for testing, the testimony of other witnesses has been found to be more probative. Mr. Bader agreed that if moisture had been present and was dissipated as of July 1980 when the samples were taken, the densities would be affected. It is found that Mr. Bader’s tests indicate “trends.”
On March 6, 1980, Tom Onuma, of the Corps Contract Administration Branch had issued his report faulting the soil mix and the “very wet winter-spring conditions, with no provision for draining the subsoil moisture and the [contractor’s] haste to get on the lots to finish the paving.”
The reconstruction effort was not different than the work in 1979, except insofar as plaintiff used a disc for aerating, not a motor grader with a scarifier or a blade, and a sheepsfoot roller, not a Hyster Embankment Compactor. One altered procedure was Mr. Cheatham’s requirement that the subgrade be approved in writing before commencement of paving. There was no breaking up, rutting, or cracking of the asphalt as had occurred in 1979. In fact, although the Corps has done nothing to maintain these areas, the site inspection *270confirmed that with the exception of the Long Shoal parking lot near an interior earth island, no cracking occurred other than at joints, which is to be expected.
The contract was completed in 1980, but plaintiff did not submit a certified claim to the Contracting Officer until March 8, 1984. In denying defendant’s first motion for summary judgment, the court ruled that defendant could attempt to show that plaintiff delayed unreasonably in submitting its certified claim as a basis for dismissal of the complaint. United Construction Co. v. United States, 7 Cl.Ct. at 53-54; accord Lacoste v. United States, 9 Cl.Ct. 313, appeal dismissed, No. 86-1002 (Fed. Cir. June 13, 1986); see S.E.R. Jobs for Progress, Inc. v. United States, 759 F.2d 1, 5 (Fed.Cir.1985) (recognizing defense of laches in contract case). However, the issue was not included in the parties’ joint statement of issues to be tried, and no testimony was elicited on this subject. The court nonetheless has applied the presumption that testimony besotted by lapsed memory would generate inferences against plaintiff’s case.
DISCUSSION
As the parties observed in stating the issues upon which they requested findings and legal conclusions, resolution of this case turns on the facts. When the Government orders a structure to be built and prepares the project’s specifications, it implicitly warrants that if the specifications are complied with, satisfactory performance will result. J.D. Hedin Construction Co. v. United States, 171 Ct.Cl. 70, 76, 347 F.2d 235, 241 (1965). The contractor is to rely upon indicated methods in order to ensure its performance. Natus Corp. v. United States, 178 Ct.Cl. 1, 371 F.2d 450 (1967). However, it is the contractor’s responsibility to provide quality control measures to assure that the work strictly complies with the contract requirements. Fortec Constructors v. United States, 760 F.2d 1288, 1292 (Fed.Cir.1985); see also Ordinance Research, Inc. v. United States, 221 Ct.Cl. 641, 609 F.2d 462 (1979). If the specifications prove defective or impossible to perform, and the contractor has sustained its burden of complying with the specifications, the Government must compensate the contractor for additional costs of performance. Ordinance Research, 221 Ct.Cl. at 670, 609 F.2d at 479; Hol-gar Manufacturing Corp. v. United States, 175 Ct.Cl. 518, 525, 360 F.2d 634, 638 (1966).
In J.D. Hedin, the density of the soil and its properties were known at the time the specifications were drawn. These exact conditions were encountered, and the contractor was unable to drive piles properly. The necessity for the subsequent change, according to the court, resulted from the Government’s failure to evaluate properly the known subsurface conditions. Any independent investigation by plaintiff would have disclosed the same conditions that the Government found.6 The inquiry in this case is to pinpoint the primary cause of the breaking up, rutting, and cracking of the pavement at Long Shoal and Berry Bend in June 1979.
Plaintiff takes the position that its construction practices in the field were sound and adequate and that the cause of the failure was a saturated—or nearly saturated—subgrade, which, in turn, was caused by the clayey component of the soil that retained moisture during the spring runoff. Dr. Leonard, plaintiff’s expert, testified that clays control the properties of the materials even though the composition was predominantly gravel. Overall, he classified the soil, according to the Unified Soil Classification System, as “GC” (clayey gravel) with dominant low permeability (“CL”) and some high permeability (“CH”). The low permeability of this soil type, in Dr. Leonard’s view, was consistent with the retention of water. The bearing capaci*271ty of clayey gravel is largely determined by its moisture content. In Dr. Leonard’s experience with design and construction of roadways, the Corps design was adequate and would have worked under most conditions; however, it was risky given the weather and soil conditions that existed in late May and June 1979. The design did not provide adequate drainage of subsurface moisture in areas where the cut of the slope met the fill, because the clayey gravel fill impeded the flow of moisture. Although the Long Shoal maneuver area and Berry Bend were all fill, both sites were on the downhill slope of a ridge and exhibited the same transition between the cut and fill.
According to Dr. Leonard, the asphalt failed under the weight of the paving equipment and rollers due to excessive water pressure in the saturated soil. Shallow rock entraps the water and does not allow dissipation of water pressure; water cannot dissipate below into impermeable rock; the asphalt seals the subgrade and further impedes dissipation of water pressure. The water tends to move vertically, because as the roller moves, it chases laterally moving moisture and forces it upwards. Tension cracks develop in the asphalt as the subgrade loses its bearing capacity.
Dr. Leonard discounted rainfall on or near the dates of grading or laying asphalt. He viewed what happened during the first two or three months prior to construction as more important because the high rainfall during that period would account for the saturated subgrade. A total-strength asphalt construction does not provide for drainage, according to Dr. Leonard, so he would have recommended an open-graded base course. The Asphalt Institute disagrees with Dr. Leonard that total asphalt construction was not indicated. Dr. Leonard opined that the plans, to the extent they were intended to be implemented in late May and June 1979, should have allowed for an uphill ditch or a pavement design with a crushed rock base.
Defendant charges that plaintiff has not met its burden because it constructed without any problem the eight additional parking areas called for by the contract utilizing the same design. Moreover, during reconstruction when the amount of rainfall was approximately the same or greater, no problems developed. It is a fact that Long Shoal and Berry Bend were the first sites constructed on this contract, but they also utilized the first total-strength asphalt design used by the Corps at Truman Reservoir. Plaintiff’s construction practices are challenged because in 1980 plaintiff waited longer after rainfall to finish the subgrade and lay pavement and used a disc, which, in defendant’s view, was the best method for providing maximum aeration. As defendant put it, plaintiff performs contracts well, but no one is perfect; plaintiff got off in the wrong direction by working too soon after rain; and closer supervision by the Corps reformed plaintiff’s construction practices.
Dr. Hatheway’s theory of causation, as defendant’s expert, for the pavement failures at Long Shoal and Berry Bend was that the source of the moisture or saturated condition was in the prepared subgrade near the surface. According to Dr. Hathe-way, the compacted cut and borrow making up the fill at the sites is more impermeable than residual soil, or is soil that is natural to a site. Infiltrated water from snowmelt and rainfall that comes into a site from higher elevations cannot saturate a previously compacted subgrade. Because this infiltrated water cannot reach or saturate the near surface, subsurface moisture was not the source of the moisture near the surface. Rather, the rainfall that fell on the site caused the soil to lose its bearing capacity. Dr. Hatheway would not agree with Dr. Leonard and the Asphalt Institute that there is a vulnerability in the areas where the subgrade changes from cut to fill. He conceded that with a predominately clay material, most rain would shed and drain from a compacted subgrade.
Ultimately, the experts’ theories were somewhat reconcilable, in that they agreed that subsurface water could be trapped in compacted soil, and that CL and CH soil *272was “unforging” (Dr. Leonard) and “deserves respect” (Dr. Hatheway). Both experts also agreed that the phenomenon of failure was the same in that under the weight of paving equipment and rollers, the excess pore water, or the excess water in pores of the soil than can be accommodated by mineral grains that are in contact with each other, could not dissipate, so that the excess water in the soil forced apart the soil particles.
Dr. Leonard’s theory of causation is found the more convincing. He alone had the experience to testify to the impact of subsurface moisture on asphalt construction. In addition, his reliance on the timing of the construction in late spring and early summer in a year of above-average rainfall accounts for the effect of this amount of rainfall on the subgrade, whereas Dr. Hatheway’s theory does not. Moreover, Dr. Leonard examined the sites on July 1, 1980, before the failed asphalt was removed and reconstructed, whereas Dr. Hatheway’s first visit to the sites was in 1985 after reconstruction. Therefore, Dr. Hatheway’s theory is inherently less verifiable than Dr. Leonard’s. During his 1980 visit, when he tested the subgrade material to determine both its classification and rock content, Dr. Leonard observed that the roller marks in the failed asphalt were not uniform. Lack of uniformity, according to Dr. Leonard’s common-sense explanation, is inconsistent with pour compaction.
Defendant says that the base course at Long Shoal did not fail until June 6 when plaintiff laid asphalt the day after rain. The facts do not support this contention. The various records kept by Messrs. Locke, Johnson, and the Corps indicate that the last rainfall before commencement of laying the base course at Long Shoal on May 31 was on May 21. The Warsaw gauge indicated .04 inch on May 27, which no one argues is consequential. The Locke, Johnson and Corps records show no rain for that date. Thus, the failure evident on June 1 occurred after asphalt had been laid following a nine-day dry period. Although plaintiff laid the base course at Long Shoal on June 6 and 7 after a .92 inch rainfall on June 5, .02 inch on June 6, and .33 inch on June 7, the problems with breaking up had been noticed as early as June 1. Prior to June 6, asphalt had been laid only on May 31 and June 1 and 2, and the rain recorded since May 21 was .04 inch on May 27 and .01 inch on June 1. The parties stipulated that over 2,000 square feet of asphalt had failed and were removed from soft spots on June 6. It is found that the base course at Long Shoal failed before June 6.
Fundamental to defendant’s theory of causation is that plaintiff was performing subgrading and paving too soon after rainfall in 1979 and reformed its practices during the 1980 reconstruction. In August 1980, when Long Shoal was reconstructed, plaintiff began laying pavement six days after the last reported rainfall, and no rain was recorded until the base course was completed. In 1980 plaintiff graded before paving during dates of recorded rainfall higher than in 1979. Pavement was laid at Berry Bend in 1979 during one day on which 1.67 inches of rain were recorded, whereas in 1980 no rain was recorded on any of the four days during which pavement was laid. The grading work at Berry Bend in 1980 was accomplished on dates on which it rained less than in 1979.
Defendant’s argument is by no means insubstantial, especially since more rain fell during reconstruction in 1980 than 1979. However, defendant fails to account for the fact that the base course at Long Shoal failed the day after it was laid in 1979 following a nine-day period with no appreciable rain. Dr. Leonard’s testimony persuasively explained that the timing of construction is decisive in a region where rains are quick and rainwater runs off, as other witnesses observed. The summer months’ rainfall in this region does not impact construction operations for roadways. In the spring or early summer, rainfall may not affect operations, but the subsurface moisture content due to prior above average rainfall, coupled with freeze/thaw action, might, and did, in this case.
*273Although any approval of the subgrade that Mr. Shelton gave at Long Shoal and Berry Bend would not excuse a failure to comply with the contract specifications, the parties stipulated that he directed the placement of a primer coat on the subgrade at the Long Shoal maneuver area and Berry Bend. This supports an inference that the problems with the subgrade were not noticeable until paving operations commenced. Moreover, Mr. Shelton rejected the subgrade at Berry Bend beach on June 22 when moisture came up from the sub-grade. The problem ultimately was remedied by reworking the subgrade, yet plaintiff had also put in a trench that intercepted underground moisture. It is true that plaintiff constructed the subsequent eight parking areas under the same specifications without incident and that they stand today, but they were not graded and paved in May-June 1979.
Plaintiff did not use a disc in 1979 either in constructing the embankment or preparing the subgrade for asphalt. However, defendant’s principal evidence that the scarifier and blade plaintiff used in 1979 did not do the job that the disc did in 1980 was Dr. Hatheway’s testimony. He said that a disc is the best method of aerating the subgrade and that he felt that the method of scarification was insufficient. Mr. Bestgen, the Corps inspector, explained that a disc usually is used when the contractor is preparing the entire subgrade and a scarifier is used for spot correction. In 1979 plaintiff used a scarifier for aerating the entire subgrade. Section 18 of the contract specified reconditioning the sub-grade by removing soft or yielding spots before laying asphalt, and plaintiff was obligated to do this work before paving operations began. In 1979 the CMI trimmer with scarifying action was used in fine grading, and a blade was used for soft spots.
Given his experience in road construction, the testimony of Dr. Leonard is more probative that disking and scarifying are alternate methods. In fact, contract specification section 9.2 provides that dried material in the embankment should be loosened by scarifying or disking, implying that the equipment is equivalent for purposes of adding moisture. One could also infer further that the equipment also is equivalent for aerating soil, but it is unnecessary to do so.
There is no basis to find that plaintiff was required to use a disc or a scarifier on the blade to remove soft spots before paving. Section 13’s requirement for reconditioning the subgrade by removing soft yielding spots in the subgrade does not specify the equipment to be used. Section 9.2 gives the contractor the option to use a blade. It requires that soil containing excess moisture “shall be disked ... bladed, and aerated as necessary____”
Mr. Supica testified that he walked the subgrade before pavement was laid at Long Shoal and Berry Bend and that it was dry. Mr. Johnson gave similar testimony. The only evidence the Government presented to the contrary was the testimony of Mr. Bestgen that one morning when the base course at Long Shoal was being laid, the subgrade was saturated and the base course that had been laid was rutted. Although he could not remember the date on which the incident occurred and he did not record it in his log, Mr. Bestgen, who was a careful witness, is believed. His testimony establishes that once the base course failed, plaintiff continued laying pavement even though the subgrade by that time was “very wet.” Section 12 of the specifications requires that asphalt be laid only when the underlying surface is dry. Although plaintiff did not comply with this specification in this one instance, the implication does not follow that there were other instances of noncompliance.
Finally, defendant argues that when Mr. Cheatham imposed more supervision (including the requirement that written approval of the subgrade be obtained before asphalt was laid), plaintiff was able to perform the contract successfully according to specifications. Unquestionably, a profound difference exists between the Corps Daily Logs of Construction in July 1979 and Au*274gust and September 1980 compared with those of May and June 1979. During late May and up to mid-June 1979, the logs merely described the weather conditions and indicated the type of work being performed at the various locations. On June 5, 1979, when grading was continuing at Berry Bend, the log reflects “sprinkled a little off and on[;] did not affect work.” On June 8, when no work was performed, the log indicates, “Rainfall heavy[;] made conditions [too] muddy and adverse for outdoor work,” and no work was accomplished that day. The log for June 13,1979, reads, “No work due to last [night’s] rain.” On June 20, 1979, the log says that work on laying the asphalt base course at Berry Bend was started at 12 noon, although no other work was done because it rained the previous night. The log for June 22, 1979, shows that plaintiff was directed to rework the subgrade of the Berry Bend beach lot because it was “bad and had to be reworked.”
It was on or after this date that the Corps increased its supervision of plaintiff's activities noting, for example, on July 9, 1979, the areas were drying out due to prior rains. The log for July 13, 1979, reads “Drying out subgrade in east parking lot of State park____ Drying out in Berry Bend.” The logs for August and September 1980 list all equipment on the site, hours worked, each type of employee, detailed weather information, and a step-by-step chronology of work done at different sites, including identifying inspectors and other Corps representatives on site. Plaintiff had made its claim for an equitable adjustment when the Corps construction inspectors began preparing detailed descriptive logs. Thus, the fact that the 1980 logs indicate repeatedly that plaintiff was disking and aerating the subgrade dry is not indicative that the notation “grading parking areas” in 1979 means that plaintiff was not allowing the areas to dry sufficiently.
A finding cannot be made that prior to the failures of the three lots that the Corps did not exercise supervision. In fact, when the Corps had problems with the subgrade, they were noted in the Corps logs and plaintiff was notified. The August 28, 1978 log reflects that the upper end of the east ramp had to be reworked completely because the base course was too thin; that two lab compaction tests had failed; and that the contractor was notified of the deficiencies. On September 1, 1978, the log states that the east ramp had been retested for compaction of base course and that the surface, as reworked, passed well within the specifications. On September 6, 1978, tests were taken on the west ramp and failed. On September 13, two compaction tests were run on the Sterett Creek boat ramp and failed. The log noted, “High on moisture in base rock____ Also informed contractor that concrete cannot be poured until gradations on base rock are updated.” On September 14, 1978, tests on base rock at Sterett Point and Thibalt Point failed. The September 15, 1978 Corps log states, “John Bledsoe had a meeting with Doyle [Shelton] today requesting the go ahead on pouring concrete Monday morning although the tests had failed. Approval was given and submission was made to lower the required % for compaction.” On October 4, 13, and 15, 1978, the Corps rejected loads of concrete.
Militating against the Government’s theory that without keen supervision plaintiff’s construction practices were hasty or otherwise inadequate, are two overriding themes: 1) Plaintiff’s witnesses testified credibly that the subgrade was compacted properly; it did not fail until after the pavement was laid; and the test data relied on by the Corps were based on unreliable samples. 2) Although the 1980 construction period was wetter than that of 1979, rainfall during the former period was below average and was above in the latter. Plaintiff’s theory is more plausible that rainfall would impact the moisture content of the subgrade during late spring and early summer, but would have no role in construction later in the summer.
Plaintiff has mustered a preponderance of evidence to show that its construction practices were not the cause of the *275pavement failure and that use of a gravel base for construction during this time of year would have avoided the problem. Plaintiff failed to comply with section 12 of the specifications in laying pavement on a wet surface on one occasion. Given Mr. Bestgen’s testimony and the court’s on-site inspection and examination of photographs depicting the Long Shoal parking lot from various viewpoints, the court can approximate the area in which asphalt had been laid when Mr. Bestgen halted the operation (there being no data and amount of asphalt with which to correlate his testimony). On this basis an appropriate adjustment of ten percent has been derived. Cf. Branning v. United States, 784 F.2d 361, 364 (Fed.Cir. 1986); Salem Engineering & Construction Corp. v. United States, 2 Cl.Ct. 803, 808 (1983) (citing cases); Dee Hong Lue v. United States, 150 Ct.Cl. 655, 667, 280 F.2d 849, 856 (1960).
CONCLUSION
Based on the foregoing, the Clerk of the Court will enter judgment for plaintiff in the amount of $225,799.05, plus interest in accordance with the Contract Disputes Act, 41 U.S.C. § 611 (1982), from March 8, 1984, until payment thereof. Judgment shall also enter dismissing defendant’s counterclaim:
IT IS SO ORDERED.
No costs.
. A third motion was withdrawn before decision.
. Mr. Bestgen was one of the three individuals who prepared the Corps Daily Logs of Construction during the relevant period. He signed logs from May 15—June 8, 1979; July 19—August 22, 1979; August 9, 1980; and August 21—Sep-tember 17, 1980 (not all of this last group are in evidence).
. Because the test data on gradation was missing and the results were inconsistent, Dr. Leonard discredited tests commissioned by plaintiff from the Kansas City Testing Laboratory in August 1980.
. The inconsistencies between the stipulations and DX 125 during the reconstruction phase have been resolved in favor of the stipulations.
. An allegedly bad mix of asphalt has been rejected as a causative factor by both parties, so Mr. Anson’s notes on the mix have been deleted from the quotation.
. In early 1978 John H. Hoyt, then a Corps field geologist, logged borings of borrow at Long Shoal and Berry Bend. Because these borings were not taken over a sufficient period to measure water levels, their only pertinence to this litigation is that they recorded the dominant material as gravel clays, clay gravels, and clays, which are classified as "GC."