Board of Contract Appeals General Services Administration Washington, D.C. 20405 ____________________ DENIED: May 17, 2001 ____________________ GSBCA 14430-TD ROXCO, LTD., Appellant, v. DEPARTMENT OF THE TREASURY, Respondent. James A. Pemberton of King & King, Washington, DC, counsel for Appellant. David H. Brunjes, Office of Legal Counsel, Federal Law Enforcement Training Center, Department of the Treasury, Glynco, GA, counsel for Respondent. Before Board Judges BORWICK, HYATT, and DeGRAFF. DeGRAFF, Board Judge. Roxco, Ltd. entered into a contract with the Department of the Treasury for the construction of two buildings. During performance, two disputes developed concerning the masonry blocks that Roxco was to install. After Treasury denied Roxco's claim for an adjustment to the contract price, Roxco filed this appeal. The Board held a hearing on June 27-29, 2000. As explained below, the appeal is denied. Findings of Fact In July 1994, the Department of the Treasury solicited bids for the construction of two classroom buildings at the Federal Law Enforcement Training Center in Glynco, Georgia. Glynco is near Brunswick, Georgia, which is fewer than five miles from the coast of the Atlantic Ocean. Treasury asked for bids based upon using calcium silicate masonry (CSM) blocks and bids based upon using precast concrete panels for the exteriors of the buildings. Exhibits 1, 2.[foot #] 1 A bidder could, if it wished, submit only one bid, for either the CSM blocks or the precast concrete panels. Exhibit 1; Transcript at 300-01. Bidders had the opportunity to ask questions about the specifications and the drawings contained in the solicitation, and Treasury amended the solicitation in response to questions that it received. Transcript at 300. On September 30, 1994, Treasury awarded a contract to Roxco, Ltd. for construction of the buildings using CSM blocks. Exhibit 1. Roxco is located in Brandon, Mississippi. Exhibit 1. Ben Turnage is the owner and chief executive officer of the company. Mr. Turnage visited the job site once very near the beginning of the contract and again very near the end of the contract. Transcript at 401-02. Ronald Robbins was Roxco's president. Mr. Robbins visited the job site approximately four times between December 1994 and October 1995. Transcript at 389. Patrick Leech was Roxco's vice president and he first became involved in the classroom construction project in July 1995. He visited the job site approximately two times. Transcript at 12, 30. In early March 1995, Roxco appointed William J. Mahrenholz as its quality control manager for the classroom building project. Mr. Mahrenholz's duties, as assigned by Roxco, included checking all materials before they were installed and being present on the site during all phases of the construction. Mr. Mahrenholz was at the job site every day, except for six weeks in early 1996. Exhibits 23, 26; Transcript at 413-14, 435. On June 22, 1995, James Hulbert arrived at the job site to assist Mr. Mahrenholz. Mr. Hulbert was at the site until late December 1995. Transcript at 164-66. All of the Treasury employees who were involved in the administration of the contract worked at the training center in Glynco. Treasury's contracting officer was Donald Smithson. Exhibit 1; Transcript 218. Montford Reed was the project manager until the autumn of 1995. Transcript at 183-84, 458. When Mr. Reed retired, Robert Jones became the project manager. Transcript at 458-59. Robert Tatum was Treasury's construction representative for the entire project, and he was at the job site every day. Transcript at 441-42, 449-50. ----------- FOOTNOTE BEGINS --------- [foot #] 1 All references to exhibits are to Appeal File exhibits. ----------- FOOTNOTE ENDS ----------- CSM Block Coursing Section 04220 of the specifications stated that the CSM blocks were to be 3-5/8 inches deep, twenty-four inches long, and twelve inches high. The joints between the blocks were to be 3/8-inch wide. Some of the blocks were to have smooth faces and others were to have rough faces. Roxco was to lay the CSM blocks in a running bond, with the vertical joints in each course of blocks centered on the blocks in the courses above and below. Exhibit 1 at 272, 274. The contract drawings set out the dimensions of the buildings. Exhibits 2, 3.[foot #] 2 No bidders asked any questions about the dimensions of the buildings, the bond that was specified, or the coursing of the blocks. Transcript at 300. All of the contract drawings were prepared by Saxelbye, Powell, Roberts & Ponder, Inc. (Saxelbye), a firm of architects and planners from Jacksonville, Florida. Exhibit 2. Saxelbye prepared the drawings under an existing delivery order contract with Treasury. Transcript at 298. Contract drawings A3.1 and A3.2 showed the elevations and dimensions of the buildings, but did not depict the CSM blocks. Drawing A3.2 contained cross- references to detail B on drawing A3.6, which showed the face of a wall section constructed with CSM blocks. Detail B/A3.6 showed the blocks laid in a running bond, with the vertical joints in each course of blocks centered on the blocks in the courses above and below. Detail B/A3.6 did not show a bonded corner, in which one block overlaps another so that the blocks interlock. Exhibit 2; Transcript at 489. The detail depicted full-length (twenty- four-inch) blocks and half-length (twelve-inch) blocks on the face of the wall. Detail B/A3.6 also contained a cross-reference to detail 8 on drawing A3.6. Detail 8/A3.6 showed CSM blocks at the outer end of a wing wall and said to cut the blocks to course as shown in the detail. Detail 8/A3.6 showed that the wing wall was thirty inches wide and that some courses laid along that width at the outer end of the wing wall would contain two fifteen-inch long blocks. Detail 8/A3.6 also showed bonded corners. Exhibit 2. Mr. Leech considered the dimensions of the buildings to be a "fundamental," "glaring" error, because not all of the walls were multiples of two feet in length. Transcript at 90. He explained that if the length of a wall is a multiple of two feet and a block is two feet long, "you would cut one block in two pieces and start with a half a block at each end of the course." Transcript at 40. Mr. Leech said that the drawings did not tell Roxco what to do at the end of a course of blocks when it encountered walls that were not multiples of two feet in length. ----------- FOOTNOTE BEGINS --------- [foot #] 2 There are no material differences between the two buildings. Transcript at 462. Therefore, we used the drawings for building number 1 as the basis for our findings of fact. ----------- FOOTNOTE ENDS ----------- He stated that there would be "left over dimensions" if a wall that was not a multiple of two feet long was built using blocks that were two feet long. Transcript at 97-98, 104. According to Mr. Leech, "You can only get a running bond throughout the whole field if the length of your wall is dimensioned around the dimension of the module that you're building with." Transcript at 105. Otherwise, he said, "you don't know what to do . . . . You can't accomplish it." Transcript at 105. Mr. Robbins said that it would have been "impossible" to build the walls according to detail B/A3.6, and that the detail does not show what Treasury wanted because it does not show a bonded corner. Transcript at 323-24, 375. Mr. Robbins said that reading detail B/A3.6 together with the detail to which it referred, detail 8/A3.6, provided him with "useless information." Detail 8/A3.6 made sense to him, and the other detail did not. Transcript at 375-76. In his view, it was fairly obvious that the details were not consistent. Transcript at 568. James Gulde testified on behalf of Treasury as an expert witness in masonry. Transcript at 473-74. He explained that a "course" is a horizontal row or a layer of masonry units in a wall, and "coursing" is achieving the building of a course. In order to achieve adequate coursing, the mason building the wall must know three things: the size of the masonry units, the type of bond required, and the dimensions of the wall. Transcript at 482-83. Because the original contract documents provided all three of these pieces of information, there was no question in Mr. Gulde's mind that a qualified mason using those documents could have built the walls of the classroom buildings. Transcript at 484. Mr. Gulde did not agree with Mr. Leech that the dimensions of the buildings presented a glaring defect in the contract documents. He said that many buildings are not designed with the measurements of the masonry units in mind, and that masons "work with funny dimensions all the time. Some are not as advantageous as others, but as a mason you figure out how to make the wall work." Transcript at 492-93, 498. In preparing an estimate, a qualified masonry contractor would take into account the actual dimensions of the building and would "figure out ahead of time how many cuts you have." He said that every job "has some issues," and the masonry contractor's job is to determine how to address those issues. Transcript at 492. Mr. Gulde explained that when a mason builds a wall, standard practice is to lay the corners first and then fill in the courses in between the corners. Before the corners are laid, however, the mason determines where cuts will be needed and usually lays a "dry course" of masonry units, using the bond required. Transcript at 497-98. Whenever the width of a masonry unit is not half the length of the unit, as is the case with the CSM blocks, the mason needs to install specially-sized units or needs to make special cuts to the blocks in order to make the blocks bond around a corner. Transcript at 490-91. Mr. Gulde did not find the difference between detail B/A3.6, which did not show a bonded corner, and detail 8/A3.6, which showed such a corner, to be an error. He explained that 99.9 percent of all masonry requires a bonded corner. In his opinion, detail 8/A3.6 which showed the bonded corner elaborated upon the more general depiction shown in detail B/A3.6. Transcript at 530-31. On February 6, 1995, Roxco asked Treasury for added details concerning the CSM blocks. Roxco said that detail B/A3.6 showed a running bond with a twelve inch piece at the corner, while detail 8/A3.6 showed a 3-5/8 inch piece at the corner. Roxco also said that one wall was twenty feet and two inches long, and another panel was five feet and two inches long. Regarding the wall and the panel, Roxco asked Treasury "which corner gets the 2" piece[?]" Exhibit 16A. Mr. Reed did not understand why Roxco was confused by the drawings. Transcript at 208-09. Roxco's question concerning which corner would end in a two-inch piece of block concerned Mr. Reed because he understood Roxco to be saying that its masons would build the wall that was twenty feet and two inches long by laying one whole CSM block at the beginning of a course, laying nine more whole blocks along the course, and then inserting a two-inch piece of block at the end of the course. In Mr. Reed's experience, however, masons build walls by first laying the corner stones and then laying the blocks in between the corners. Mr. Reed was also concerned about Roxco's question regarding the panel that was five feet and two inches long, because the "panel" was actually the distance between two points on the drawing and was unrelated to the masonry work. Mr. Reed had doubts about the competence of Roxco's masonry workers, and he concluded that Roxco needed some help in order to accomplish its masonry work. Transcript at 189, 208-12. On March 20, 1995, Roxco told Treasury that exterior masonry work would begin soon and asked for a response to its February 6 questions. Exhibit 29. On March 24, Mr. Reed sent a telefacsimile to Saxelbye. The cover sheet contained this text: "1. RFI #14. 2. Need coursing layout for CS masonry. 3. Contractor says he needs answers on elevator re-submittal." Exhibit 31. Saxelbye sent Treasury a set of drawings labeled reissued and dated April 7, 1995. Exhibits 2. Saxelbye did not charge Treasury for these drawings. Mr. Smithson did not recall why no charge was made, but he did not believe that much effort was required in order to generate the drawings. Transcript at 229- 30. Treasury received the drawings on April 14, and forwarded them to Roxco on April 17. Exhibits 36, 37, 38. Reissued drawings A3.1 and A3.2 showed CSM blocks on some sections of the elevations of the buildings.[foot #] 3 The blocks (or portions of blocks) at opposite ends of a course were nearly the same length. Both drawings contained cross-references to detail 1, which had been added to reissued drawing A3.1. Detail 1/A3.1 showed a typical corner constructed with CSM blocks. The detail showed a bonded corner, and blocks at the ends of some courses were cut so that the remainder of the blocks in those courses could be laid in a running bond, with their vertical joints centered on the blocks in the courses above and below. The blocks shown in detail 1/A3.1 were consistent with the blocks shown in the drawings of the elevations. Reissued drawing A3.2 contained cross-references to detail 1/A3.2, which had been added to show the face of a typical wing wall constructed with CSM blocks laid in a running bond and with bonded corners. Detail 1/A3.2 showed the wing wall from a perspective different from that shown in detail 8/A3.6. Exhibit 2. On April 18, 1995, Roxco pointed out that the drawings did not show any control joints for the CSM block construction. Exhibit 43. A control joint is constructed in a masonry wall in order to control the direction of potential cracks. Transcript at 194. On April 24, Treasury asked Saxelbye to provide some guidance regarding the placement of control joints. Exhibit 39. Treasury provided Roxco with drawings A3.1 and A3.2 labeled reissued and dated April 27, 1995, to show where Roxco should place the control joints. Exhibit 2. On May 2, Roxco told Treasury that the April 27 drawings did not provide details concerning either the coursing at the control joints or cuts to the blocks at the control joints. Roxco asked Treasury for a sketch regarding these issues. Exhibit 43. Treasury provided Roxco with drawing A3.2 labeled revised and reissued and dated May 5, 1995, which added detail 3 to show the cuts at a typical control joint. Exhibit 2. A course of masonry is symmetrical if the blocks (or portions of blocks) at opposite ends of a course are nearly the same length. Transcript at 109, 120, 504-07. Mr. Leech explained that whether a mason tries to achieve symmetrical coursing depends upon the project. If a masonry wall is being constructed in the interior of a warehouse, for example, it might not be important where the joints are located between the masonry blocks. The classroom buildings were "monumental" projects, and the locations of joints on exterior walls are important in such projects. Transcript at 99-100, 110. Mr. Robbins testified that if he were laying blocks in a running bond, he would make the courses symmetrical because "that is to me a better way of doing it." Transcript at 371. Mr. Gulde said that symmetry is part of building a wall. Transcript at 505. ----------- FOOTNOTE BEGINS --------- [foot #] 3 Drawing A3.6 was also reissued, but it did not contain anything different from the original contract drawing. Exhibit 2. ----------- FOOTNOTE ENDS ----------- On May 12, 1995, Treasury asked Roxco for a price proposal to cover a change order for labor, material, supervision, and equipment necessary to provide "control joints and symmetrical coursing." Exhibit 59. According to the contracting officer, symmetrical coursing "was required all along," and the May 12 request covered whatever changes would be needed to the symmetrical coursing due to the insertion of the control joints. Transcript at 232-33. Mr. Robbins explained that whenever a feature, such as a control joint, is added to a building, some adjustment will have to be made to the coursing in order to ensure that the masonry work has a "decent appearance" at that feature. Transcript at 325-26. Roxco's June 6 price proposal asked for $133,150.93, which included the cost of 672 cuts to CSM blocks at control joints, 1124 cuts to CSM blocks at the stair towers, 672 cuts to CSM blocks at the wing walls, 496 cuts to CSM blocks at the corners of the buildings, and 3564 additional CSM blocks. Exhibit 59. Treasury decided that it was responsible for the cost of the control joints and that it would negotiate a contract modification to reimburse Roxco for that work. Treasury concluded that it had not directed Roxco to use the coursing pattern shown in the reissued drawings. Because Roxco had not asked any questions about the CSM blocks before bidding, Treasury assumed that Roxco must have had some coursing pattern in mind when it bid, so it decided to tell Roxco to use whatever coursing pattern it intended to use when it submitted its bid for the project, subject to the contract's Material and Workmanship clause. Exhibit 64. The Material and Workmanship clause required Roxco to perform the contract work in a skillful and workmanlike manner. Exhibit 1 at 742. In an August 18, 1995 letter to Roxco, Treasury recounted an August 16 discussion concerning Roxco's June 6 price proposal. Treasury said that it intended to reimburse Roxco for the control joints, including the 672 cuts, and stated that Roxco's former project manager had agreed with Treasury that $6082 was a fair amount for that work. Treasury understood that Roxco wanted to meet further to discuss other CSM coursing issues. Treasury stated that it was not directing Roxco to use the coursing scheme set out in any of the drawings, and that Roxco must have known how to place the blocks when it submitted its bid, because it did not ask any questions about the blocks before bidding. Exhibit 111. On September 13 and 19, 1995, Roxco provided Treasury with price proposals numbered 14, 15, and 16, for making 672 additional control joint cuts and supplying and installing the material used to create the control joints. The total amount Roxco requested in the price proposals was $13,162.43. Exhibits 117, 119. On September 25, 1995, Roxco asked for a change order to cover the work required by the coursing scheme set out in drawings A3.1 and A3.2. Roxco said that Treasury s August 18, 1995 letter directed Roxco to use the coursing scheme detailed in those drawings. Roxco also said that it was obvious that Saxelbye believed that the original contract drawings were inadequate, or else it would not have issued subsequent drawings. Exhibit 121. Treasury responded that Saxelbye prepared the subsequent drawings due to the requests from Roxco and by bidding, Roxco agreed that it could build the project with the documents originally provided. Exhibit 122. Roxco began laying the CSM blocks in October 1995. Exhibits 123, 125, 350; Transcript at 86. Roxco did not install all of the blocks exactly as shown on the reissued drawings. Exhibits 2, 530 at R1-6 and R1-17; Transcript at 502-03. In Mr. Gulde's opinion, the mason "did a pretty good job of carrying out the intent of the running bond and achieving basically what the government wanted with the least amount of cuts." Transcript at 504. The parties signed modification 10 to the contract on June 11 (Roxco) and June 14 (Treasury), 1996. The modification stated that Treasury accepted Roxco's September 13 and 19, 1995 proposals numbered 14, 15, and 16 for "exterior masonry control joint work" in the amount of $13,000. The modification, which dealt with a number of issues in addition to the control joints, increased the contract price and extended Roxco's performance period by seven days "as a result of these changes." The modification contained this release: In consideration of the modification(s) agreed to herein as complete equitable adjustments for the cost increase confirmed by this modification, the Contractor hereby releases the Government from any and all liability under this contract for further adjustments attributable to such facts or circumstances and giving rise to this modification, with no exceptions. Exhibit 1 at 7-9. On October 18, 1996, Roxco sent a letter to the contracting officer asking for a contract modification in an unspecified dollar amount, due to problems with coursing. Roxco stated that its request for a modification was based upon a change to the requirements of the contract. Roxco said that the installation of the CSM blocks had been on hold from February 6, 1995, through August 18, 1995, due to Treasury s "indecision." Roxco said that its bid documents envisioned a running bond and that after the contract was awarded, Treasury issued new instructions that required many cuts to the blocks that would not have been needed if the "contract documents" had been followed. Exhibit 350. In February and March 1997, Treasury and Roxco met and discussed the coursing issue. Treasury's position was that the contract documents contained enough information to permit Roxco to install the CSM blocks and that Treasury had not directed Roxco to install the blocks as shown on the reissued drawings. Exhibits 351, 352, 353, 355. On May 22, 1997, Roxco submitted a certified claim to the contracting officer. Roxco said that the reissued drawings required it to change the start and stop point of the coursing at each added control joint, and also required it to make many extra cuts to the CSM blocks. Roxco said that it had prepared a drawing to show that if it had placed the blocks in accordance with the contract's specifications, it would have made 2306 cuts to the blocks. Roxco had also prepared a drawing to show that it actually made 4020 cuts when it placed the blocks in accordance with the subsequently issued drawings. Roxco said that the original contract drawings "glaringly omitted" a "key piece of the design" and that the scope of the work required by the contract could only be determined after the additional drawings were issued. Roxco asked for $97,170.98 for 1146 additional CSM blocks that it had to use due to Treasury's actions, 1714 additional cuts that it had to make to the CSM blocks, four months of additional scaffolding and lift time, and overhead and profit. Roxco also asked for $141,345.12 for 120 days of "extended fixed field expenses" that resulted in part from the coursing problem and in part from the discoloration problem, discussed below.[foot #] 4 Exhibit 355. On September 18, 1997, Treasury s contracting officer denied Roxco s claim. Exhibit 357. Mr. Leech explained that additional blocks and additional cuts were required because Treasury's design of the buildings was "incomplete," meaning that Treasury should have completely revised the drawings when it decided to use CSM blocks so that the dimensions of all of the walls were neatly divisible by two. Transcript at 39-41. Mr. Robbins thought that all of the drawings were "useless" and "terrible." Transcript at 377. In his view, Treasury did not permit Roxco to lay the blocks in the most economical way. Transcript at 327. He said that the reissued drawings implied that Treasury wanted symmetrical coursing, but did not show all of the specific cuts to the CSM blocks that Roxco was required to make in order to achieve symmetrical coursing. Transcript at 377-79. Mr. Gulde testified, "I don't think I've ever seen a drawing that shows every stone." Transcript at 494. Mr. Leech arrived at the number of additional cuts for which Roxco asked to be paid by asking a draftsman to prepare two drawings, A-1 and A-2. Transcript at 41-42. Mr. Leech said that drawing A-1 tried to "develop a way that you could build the ----------- FOOTNOTE BEGINS --------- [foot #] 4 At the hearing, Mr. Leech agreed that it was fair to say that the time taken to resolve the coursing issue did not affect the date that Roxco began installing the CSM blocks. Transcript at 88. ----------- FOOTNOTE ENDS ----------- building with . . . the minimum number of cuts possible." Transcript at 42. For this drawing, Mr. Leech said, the draftsman began one course with a whole block, proceeded with whole blocks along the course, and then at the end of the course, "whatever the leftover dimension was he'd make a cut there and stick a piece of stone in." The next course would come back in the opposite direction, beginning with a whole block and ending with a cut piece of block. Transcript at 117, 119. Drawing A-1 shows 2306 cuts to blocks. Exhibit 355. The second drawing, A- 2, "calculated, as best we could tell, without looking at every single block, essentially how the building was actually constructed with a generally symmetrical coursing pattern." Transcript at 42. Mr. Leech said that drawing A-2, unlike drawing A-1, included cuts made due to the use of symmetrical coursing, as well as cuts made due to control joints. Transcript at 145-46. Drawing A-2 shows 4020 cuts to blocks. Exhibit 355. The difference between the number of cuts shown on the two drawings is 1714, which is the number of cuts Roxco claimed were additional. Transcript at 42; Exhibit 355. Mr. Leech explained that because Treasury compensated Roxco for the costs associated with adding the control joints in modification 10 to the contract, the 1714 cuts for which Roxco claimed compensation should not include any cuts related to adding the control joints. Transcript 134. Drawing A-2, however, clearly shows cuts made as a result of adding the control joints. In fact, drawing A-1 is labeled as showing the "number of stones cut without control joints," and drawing A-2 is labeled as showing the "number of stones cut with control joints." Exhibit 355. Although Mr. Leech could not tell whether some of the cuts were made due to the addition of control joints or due to the use of symmetrical coursing, he believed that only 672 cuts were made due to the addition of the control joints. These cuts, he said, were mistakenly included as part of Roxco's claim. Transcript at 138, 159; Exhibit 355. Taking this mistake into account, Roxco revised its claim to ask for payment for making 1042 additional cuts and 115 days of extended fixed field expenses. Exhibit 527; Transcript at 358-60. Mr. Robbins read drawings A-1 and A-2 differently from the way that Mr. Leech read them. In Mr. Robbins's view, the difference between drawing A-1 and A-2 was, as stated on the drawings, that one showed a building with control joints and the other showed a building without control joints. He thought that drawing A-1 showed the building as it would have been constructed with symmetrical coursing, but without control joints. Transcript at 386. Drawing A-1 does not show how Roxco planned, at the time it submitted its bid, to place the CSM blocks on the walls of the buildings. Transcript at 386-87. Mr. Leech was not involved in the bidding process, and he did not recall ever seeing the estimate for the masonry that was included in Roxco's bid. Transcript at 49. Although Mr. Robbins remembered seeing the estimate, he was not involved in preparing either the estimate or the bid. Transcript at 320. The draftsman who prepared drawing A-1 was not involved in preparing Roxco's bid. Transcript at 89. We do not have Roxco's estimate or bid documents available. Whoever prepared Roxco's masonry estimate did not testify about the preparation of the estimate. Further, our record contains no explanation of how Roxco planned to place the blocks when it prepared its bid, so we do not know how many cuts Roxco planned to make to the blocks when it entered into the contract.[foot #] 5 Transcript at 387, 409. Drawing A-2 does not accurately illustrate either the control joints in the buildings or the cuts that were actually made to the CSM blocks. Transcript at 462; Exhibit 530 at 4 of 9. Shortly before the hearing in this appeal, Mr. Robbins went to the job site in order to count the cuts that were actually made at the control joints. Transcript at 328-30. Based upon his observations, Mr. Robbins developed two new drawings. He decided that there were approximately twice as many cuts attributable to the addition of the control joints as the number of cuts (672) that the parties thought had been made due to the addition of the control joints. Transcript at 328-30; Exhibits 521, 522. Mr. Jones reviewed Mr. Leech's two drawings and determined that they did not accurately depict the buildings as built. Transcript at 462. He also reviewed Mr. Robbins's two new drawings and determined that they were not entirely accurate. Mr. Jones measured the blocks at each control joint in order to determine how many cuts were actually made. Usually, the control joints were located so that some courses contained a full block at the joint and the alternating courses had a cut block, so that there were not as many cut blocks as shown on Mr. Robbins's drawings. Mr. Jones made two drawings of his own to show what he saw on the buildings. He counted approximately 860 cuts that were made necessary by the control joints. Transcript at 462-66; Exhibits 528-29. During the hearing, Mr. Robbins reviewed Mr. Jones's drawings, and he and Mr. Turnage went to see the buildings in order to determine how many cuts were made due to control joints. He decided that the number was not as great as he had previously thought, and that he and Mr. Jones were probably very nearly in agreement as to the number of cut blocks. Transcript at 556-60. ----------- FOOTNOTE BEGINS --------- [foot #] 5 Mr. Robbins testified that Roxco gave Treasury's Inspector General the only hard copy of its estimate, which Roxco used to prepare its bid. Mr. Robbins said that he was not worried about turning over Roxco's only hard copy of the estimate, because Roxco had the estimate on a computer. Transcript at 353-54. Subsequently, Roxco was not able to locate a hard copy of the estimate. Transcript at 355. Our record does not contain a copy of the estimate or any other bid documents, either in hard copy or in a computerized format. ----------- FOOTNOTE ENDS ----------- CSM Block Discoloration Before Treasury issued the solicitation for the classroom buildings contract, Saxelbye provided Treasury with product literature concerning CSM blocks manufactured by a Canadian company, Arriscraft Corporation. Exhibit 6 at 14. Section 04220 of the specifications contained in the classroom buildings solicitation said that CSM blocks manufactured by Arriscraft would be acceptable to Treasury, as would any equivalent blocks. Exhibit 1 at 272-73. Saxelbye did not know of any blocks that were the equivalent of those made by Arriscraft. Exhibit 501. In December 1994, Roxco submitted to Treasury information concerning the CSM blocks manufactured by Arriscraft, and asked Treasury to approve the use of the Arriscraft blocks, which Treasury did later that month. The information submitted by Roxco consisted largely of Arriscraft's product information, which described the characteristics of the blocks and included handling, storage, cleaning, and protection instructions. The handling instructions explained that the blocks should be isolated from contact with the ground and other materials "to prevent staining" and that no blocking material should be used that would "stain or discolour exposed surfaces of units." One of the storage instructions was to "[p]rovide necessary means to prevent staining of units during storage." Another was to "[c]over stored units if exposed to the weather for an extended period of time." The cleaning instructions stated that the blocks were supposed to be washed down and brushed with fiber (not wire) brushes to remove mortar and stains. If this cleaning method was ineffective, other methods could be used only with advice from Arriscraft. If the corners and edges of blocks needed to be protected during construction, they were supposed to be protected in a manner that would not stain the blocks. The product literature said that the CSM blocks should not be used where they would be exposed to de-icing salts or salt-laden snow. Exhibits 6, 7, 8. The contract's Permits and Responsibilities clause provided that Roxco was responsible for all materials delivered until completion and acceptance of the entire work. In addition, the Payments Under Fixed-Price Construction Contracts clause provided that payment by Treasury did not relieve Roxco from its sole responsibility for all material. Treasury supplemented this clause to provide that if Treasury paid for materials stored at the job site, Roxco would remain responsible for adequately securing those materials. The Inspection of Construction clause provided that if Treasury made any inspections, they were for its sole benefit and did not relieve Roxco of responsibility for damage to material before acceptance. The Material and Workmanship clause provided that all material incorporated into the work was to be new and of the most suitable grade for the purpose intended. Exhibit 1 at 711, 715, 742. The contract required Roxco to store all supplies on the project site in a manner that would preclude climatic damage. Exhibit 1 at 732. It also required Roxco to remove any blocks that were stained and to replace them with blocks that matched the adjoining blocks. Exhibit 1 at 275. The solicitation provided bidders with climatological data from the United States Weather Station at Fort Stewart, Georgia, which is approximately fifty-seven miles north of Glynco. According to that data, the average rainfall at Fort Stewart during May, June, July, and August is 28.1 inches. During those four months there are, on average, forty-six days when the temperature is in excess of ninety degrees Fahrenheit. Exhibit 1 at 728. In December 1994, Roxco ordered white Arriscraft blocks from William M. Wood Company (Wood), the local Arriscraft representative in Jacksonville, Florida. Roxco had the majority of the blocks (19,430) delivered to the job site in February and March 1995, and a few more blocks (2304) delivered in early June 1995. Exhibits 146, 362, 369. Treasury designated a "lay down" area and Roxco unloaded the blocks in that area. Transcript at 443. Treasury paid for the blocks soon after they were delivered. Exhibit 350; Transcript at 229. The blocks as they arrived from Arriscraft were covered with shrink wrap and on pallets. Exhibit 112; Transcript at 166, 443. In mid-April 1995, Roxco finished installing the CSM blocks on a sample wall that it constructed. Exhibit 34. Roxco stored the CSM blocks in their original shrink wrap. Transcript at 50, 79, 166. According to Mr. Robbins, Roxco also placed a sheet of plastic on top of the blocks. Transcript at 333-36. Mr. Robbins referred to some photographs, dated June 21 and November 22, 1995, that he said showed the additional plastic covering. The June 21 photograph, he said, showed rolls of the plastic covering sitting on top of the stacks of CSM blocks. Transcript at 338-41; Exhibits 525, 526. Mr. Hulbert believed that Roxco placed a sheet of plastic on top of the blocks. Transcript at 166. Mr. Mahrenholz did not recall that any added plastic was placed on top of the CSM blocks when they were stored at the job site. Transcript at 413-15. Mr. Tatum testified that no added covering was placed on top of the blocks until after October 1995, when Roxco began sorting them according to whether they were stained or unstained. Transcript at 443. He believed that he took the June 21 photograph referred to by Mr. Robbins, and he explained that the photograph showed piping, and not rolls of plastic, on top of the stacks of blocks. Transcript at 444-45. Mr. Jones testified that when he first saw the CSM blocks in the autumn of 1995, the only covering on the blocks was the original shrink wrap. Transcript at 458-60. We conclude that Roxco covered the blocks with an additional sheet of plastic after it began separating the stained blocks from the unstained blocks. If Roxco covered the blocks with a sheet of plastic any earlier than this, it did not do so consistently throughout the time the blocks were stored at the job site. On August 14, 1995, in a letter to Wood, Arriscraft said that one of its representatives had recently visited the job site and that the blocks were quite wet from rain and from condensation that had accumulated under the shrink wrap due to the length of time the blocks had been left in their original packaging. Arriscraft recommended unwrapping the blocks seven to ten days before they were installed, so that they could dry. Wood sent Saxelbye a copy of Arriscraft's letter. Saxelbye forwarded the letter to Treasury, and Treasury sent a copy to Roxco. Exhibit 112. Mr. Hulbert confirmed that there was condensation under the shrink wrap. Transcript at 179. Roxco began installing the CSM blocks in October 1995. Exhibits 123, 125, 350; Transcript at 86. In accordance with Arriscraft's suggestion, it unwrapped the blocks seven to ten days before they were installed. Transcript at 170. Some of the blocks were stained when they were taken off their pallets, and Roxco installed some of the stained blocks in the walls. Transcript at 307, 455-56; Exhibit 407. Roxco did not notify Treasury that the blocks were stained. Transcript at 315. On October 19, Treasury issued a construction compliance notice to Roxco, stating that blocks in various areas of one wall were stained and dirty. Exhibit 129. Also on October 19, Mr. Mahrenholz, Roxco's quality control manager, ordered work with the blocks to stop until a problem with the wall ties could be resolved. Exhibit 131. Roxco immediately requested that Arriscraft visit the project site and explain what was causing the blocks to stain. Exhibit 350. Wood arranged to visit the job site with a chemical engineer from Arriscraft and to inspect the blocks on October 24 or 25, 1995. Exhibits 132, 137, 139. On October 26, representatives from Treasury, Roxco, Wood, Saxelbye, and consulting engineers hired by Saxelbye visited the project and looked at the CSM blocks. Wood reported that according to Arriscraft s engineer, sandblasting the blocks to clean them was acceptable so long as it did not harm the surface of the blocks. Wood and Arriscraft were looking into chemical cleaning options, as well. Also according to Arriscraft s engineer, the blocks were to be cleaned individually before they were placed in the wall, so as to remove any risk of damaging the mortar joints by sand blasting or chemical cleaning. Mr. Smithson directed Roxco to clean the blocks before placing them in the wall. He said that Roxco could sandblast the blocks so long as their surfaces were acceptable. He also told Roxco to provide a complete method for the process Roxco would use to clean the blocks. It is not clear whether Mr. Smithson expected Roxco to obtain Treasury's approval before utilizing whatever method it developed for cleaning the blocks. Exhibit 139. Mr. Smithson's direction to clean the blocks before placing them in the wall was based upon Arriscraft's statements and his concern that the blocks might not come clean after they were placed in the wall. Transcript at 308-09. On October 27, 1995, Roxco wrote to Wood and asked what its position was regarding the discoloration and how Wood thought that Roxco should proceed. Exhibit 508. Roxco told Treasury on October 27 that it was investigating the discoloration of the CSM blocks and that it would advise Treasury of its findings. Exhibit 134. That same day, Mr. Mahrenholz allowed the CSM block work to resume. He directed Roxco's subcontractor to continue setting aside any blocks that were not suitable for use. Exhibit 135. Also on October 27, Treasury asked Roxco what it would do to correct the problem with the stained CSM blocks and whether the stained blocks would affect Roxco's schedule for performing the contract work. Exhibit 137. Roxco responded later that day and explained that it was investigating the problem and that it was segregating the discolored blocks and installing only the blocks that had not discolored. Roxco said that approximately one-third of the blocks were discolored, but that it was proceeding to work "without delay" and that culling the discolored blocks should not affect its schedule. Exhibit 138. On November 6, 1995, Treasury told Roxco that some blocks were showing stains after they dried in place, and said not to place any more blocks until the staining problem was resolved. Exhibit 140. The next day, Treasury told Roxco that it could continue to install blocks that met the contract's requirements, including blocks that Roxco had cleaned. Treasury said that it continued to be concerned about the effect that the stained blocks might have upon Roxco's schedule. Exhibit 142. On November 8, Wood wrote to Roxco and stated that it wanted to find a solution to the discoloration problem and that it had spoken with Arriscraft. As to the origin of the problem, Wood was convinced that the blocks discolored because they were allowed to get very wet and to remain that way for eight months. Wood said that the proper way to proceed would be to dry the blocks thoroughly and remove any stains with a chemical treatment or a light sandblast. Wood said that, as Roxco was aware, Wood's representative had been at the project numerous times and tested various chemicals to see if they would remove the stains. Wood had also asked a representative from a solvent company to visit the job site and make a recommendation for cleaning the blocks. Wood said that it and Arriscraft were ready to work with Roxco to solve the problem. Exhibit 509. On November 14, 1995, Roxco told Treasury that it was working diligently to solve the problem with the stained blocks and that it was proceeding to work by using blocks that had not discolored. Roxco suggested that if the blocks discolored while stored at the job site, they would probably also discolor after they were placed in the walls. Roxco also suggested that the stability of the finish of the blocks was altered by the atmospheric conditions and the environment at the project site. Roxco said that the discoloration was a natural condition of the reaction of the blocks to the elements, and that it had contacted Georgia Tech University to see if there was a "chemical explanation" for the discoloration. Roxco reported that the discoloration could be "easily removed" with 100 grit sandpaper or by sandblasting without harming the surface of the blocks. Roxco suggested that cleaning the blocks be deferred until after they were installed and were thoroughly dry. Roxco also said that the mortar between the blocks would not be harmed if the blocks were cleaned in place. Roxco stated that it was experiencing no schedule impact due to the discolored blocks. Exhibit 144, 145. Also on November 14, Treasury counted the pallets of blocks that Roxco had sorted according to their condition and determined that 40.8% of the blocks were stained. Exhibit 146. Treasury decided to retain contract payments in an amount equal to 40.8% of the amount that it had paid Roxco for the blocks. Exhibit 149. In response to Treasury's decision regarding retention of payments, Roxco explained that it was sorting the blocks according to whether they could be laid immediately. Even though Roxco could not use the discolored blocks immediately, it expected to be able to use them eventually. Roxco stated that Treasury was "responsible for the discoloration" of the blocks because it had specified a "proprietary product" and because only Treasury knew the "unique characteristics" of the blocks. Roxco said that Treasury had a duty to assist Roxco to develop creative solutions to the discoloration problem. Exhibit 150. On December 12, 1995, Mr. Jones reminded Roxco that it needed to propose a method for cleaning the blocks, and that the blocks were to be cleaned before they were laid. Exhibit 159. At a meeting that same day, Roxco said that it was cleaning the discolored blocks by hand sanding the smooth faced blocks with sandpaper before installation. It intended to sandblast the rough faced blocks, and it would use these same methods to clean discolored blocks that had already been installed. Roxco summarized the meeting in a letter dated December 14, and attached a letter from Arriscraft stating that the blocks would not be harmed if Roxco used eighty grit sandpaper to clean them. Exhibit 160. On December 15, Roxco sent Treasury a drawing of a template that it would use to cover the mortar joints between the blocks when it sandblasted blocks in place. Exhibit 163. On December 19, Mr. Smithson wrote to Roxco about its proposed cleaning method. In the letter, Mr. Smithson said he concurred with Roxco's December 14 proposal for cleaning the blocks not yet in place, and for cleaning the smooth faced blocks that had already been installed. He did not concur with Roxco's proposal for cleaning the small number of stained rough faced blocks that had already been installed. Also, he wanted Roxco to demonstrate that its proposed cleaning procedures worked. If they did, Treasury would release one-half of the retained contract payments immediately and the remainder of the retained payments when all of the blocks were cleaned. Mr. Smithson said that Treasury would designate an area of blocks that were in place as a benchmark for accepting the remaining blocks. In addition, he asked Roxco for the results of the examination made by the Georgia Institute of Technology. Exhibit 164. Roxco demonstrated its method for cleaning the blocks on December 20. During the demonstration, Treasury designated an area of blocks that were in place and had been cleaned acceptably. Treasury accepted Roxco's proposed procedures, except for the method of cleaning the rough faced blocks that were in place. Roxco planned to clean the smooth faced blocks that were in place and then wash the walls, and Treasury said that it would work with Roxco to resolve the issue of the rough faced blocks before it washed the walls. Exhibit 169. Roxco cleaned the small number of stained rough faced blocks that it had laid in the walls, although it is not entirely clear in our record what method it used to perform that work. Transcript at 455-56, 566. Roxco was able to remove the stains from all but approximately one percent of the stained blocks that it attempted to clean. Exhibit 350. Mr. Robbins testified that after Roxco developed an acceptable method for cleaning the blocks, he asked Mr. Smithson if Roxco could install the blocks and then clean them, but Mr. Smithson denied that request. Mr. Robbins was not sure when this conversation occurred, but he thought it was in December 1995 or January 1996. Transcript at 555. Mr. Smithson did not recall any such request. Transcript at 309. On December 22, 1995, Arriscraft's chemical engineer wrote to Roxco about the stained blocks. She said that the statements made in Wood's November 8 letter to Roxco were consistent with her laboratory analysis, which showed growths of algae and mold on the CSM blocks. She explained that such growths can occur on masonry and how they thrive in warm, moist conditions. Exhibit 511. On February 5, 1996, Treasury asked Roxco for a copy of any written opinion it had received concerning the stained CSM blocks. Exhibit 189. On February 12, 1996, Arriscraft sent a letter to Wood. Apparently, Arriscraft had visited the job site recently. Based upon its observations, Arriscraft suggested that the project architect verify that the standards for cleaning and selecting blocks were maintained and not allowed to slip as construction progressed. Also, Arriscraft said that "due to the way the stone was allowed to get wet and discolor," it was likely that Roxco would have to order additional blocks and those blocks would not match precisely the color and consistency of the blocks that were originally supplied. Arriscraft suggested that Roxco take steps to determine how many added blocks it would need and plan to place blocks so that any color or consistency differences would not be noticeable. Wood sent Saxelbye a copy of Arriscraft's letter, and Saxelbye forwarded the letter to Treasury. Exhibit 198. On March 7, 1996, Roxco informed Treasury that its current progress schedule update showed that critical path activities were not behind schedule. Roxco also mentioned that it was trying to supplement the masonry subcontractor. Exhibit 206. Treasury disagreed with Roxco's assessment of its progress and asked Roxco to substantiate its statements. Exhibit 210. Roxco hired a new masonry subcontractor and its work force was on site beginning March 18, 1996. Exhibit 211. On March 21, 1996, a chemist provided Roxco with a report regarding the CSM blocks. He explained that CSM blocks are expected to contain iron and aluminum, as did the Arriscraft blocks. As the condensation grew inside the shrink-wrapped pallets of blocks, water was wicked into the blocks. As the water came in contact with the iron and aluminum in the blocks, the metals oxidized, and the result was the stains on the face of the blocks. Salt makes the oxidation process occur more rapidly, and the water that the blocks absorbed was saturated with salt because Brunswick, Georgia, is near the Atlantic Ocean. Exhibit 350. On April 2, Roxco stated that it had initiated a "considerable amount of scientific investigation performed by independent sources" concerning the cause of the discoloration of the CSM blocks. Roxco said that its chemical analysis concluded that the presence of magnesium in the blocks predisposed them to discoloration, due to the moist, salty environment in Glynco. Roxco concluded that Treasury was at fault for specifying the Arriscraft blocks, saying: The Government should have adverted to the proximity of [the project] to the Atlantic Ocean (that it is an area of high humidity and salinity) and the fact that the stone manufacturer, in their literature, highlighted very conspicuously that salt has a deleterious effect on their product. Exhibit 218. Roxco said that the staining problems had delayed its progress and that it had accelerated its work in order to meet the scheduled completion date. Exhibit 218. On April 11, Treasury again asked Roxco for the test data or reports it had concerning the cause of the stains on the blocks. Treasury stated that Roxco had not substantiated that any delay to the completion of the project had occurred due to the problems with the blocks, and also stated that it had not directed Roxco to accelerate its performance. Exhibit 220. On July 19, 1996, Roxco said that it intended to have chemical analysis work performed in the near future in order to establish the reason the blocks discolored. Exhibit 289. On October 18, 1996, Roxco sent a letter to Mr. Smithson asking for a contract modification in an unspecified dollar amount, due to the block discoloration. Roxco stated that its request for a modification was based upon a defect in the contract specifications. Roxco said that Treasury, by specifying a proprietary product, "professed to have expert knowledge of the unique characteristics" of the blocks. Roxco pointed out that Arriscraft's product literature warns against exposing the blocks to de-icing salt or salt-laden snow, but claimed that it first knew about this warning when it started to unpack the blocks at the project site. Roxco pointed out that Brunswick, Georgia, is on the ocean and is generally very humid, and attributed the stains on the blocks to the use of the blocks in a humid, salty environment. Roxco said that the stains were caused by the presence of iron and aluminum in the blocks, and that it experimented for approximately thirty days with different cleaning methods before determining that sandpaper would remove the stains. To its October 18 letter, Roxco appended a copy of the chemist's March 21, 1996 report. Exhibit 350. On October 28, 1996, Mr. Smithson sent a letter to Arriscraft and attached a copy of the March 21, 1996 chemist's report. In his letter, Mr. Smithson stated that according to that report, Treasury's specification of Arriscraft's CSM blocks was defective because the blocks were not suitable for use in the south Georgia climate. Mr. Smithson asked Arriscraft for its comments, so that he could consider them when he prepared a response to Roxco. Exhibit 355. Arriscraft reviewed the chemist's report and responded to Treasury on November 13, 1996. Arriscraft stated that the blocks arrived in Georgia in good condition. It also noted that the sample wall that Roxco constructed did not become stained, although it remained in place in the same environment during the same period that the CSM blocks in the shrink wrap became stained. The blocks in place on the buildings did not stain, although they were in the same environment as were the blocks stored in the shrink wrap. Arriscraft explained that it made a special production run in February 1995 to produce the blocks in order to meet what Roxco said was a tight timetable. Arriscraft stated that Roxco then left the blocks wrapped tightly for months on end in a very damp, humid environment without taking any steps to ensure that the blocks would stay dry and clean. According to Arriscraft, virtually all masonry materials have some degree of absorption, and iron and aluminum would be found in many, if not most, such materials. In Arriscraft's view, most masonry materials would have suffered a fate similar to that of the CSM blocks if they were left in hot, wet conditions for an extended period of time. Arriscraft noted that it had been producing and selling blocks throughout North America for over thirty-five years, and that the blocks had been used on many projects on the Atlantic Ocean coast without any problems due to the environment. The Arriscraft response also pointed out that Arriscraft and other masonry manufacturers warn about exposing masonry to salt because dissolved salts can recrystalize within pores at the surface of masonry and create damaging stresses. The warning about salt had nothing to do with the possibility of staining. Exhibit 355. In February and March 1997, Treasury and Roxco met and discussed the staining issue. Treasury's position was that the stains appeared on the blocks because of the manner in which Roxco stored them and the length of time they were stored. Exhibits 351, 352, 353, 355. Mr. Smithson explained that the contract placed the responsibility on Roxco to store the blocks properly. When Roxco chose to have the CSM blocks delivered to Glynco months before it was ready to begin laying the blocks, it should have conferred with Arriscraft or Wood to become familiar with the product and to learn how to store the blocks for an extended period of time. Transcript at 250-54. On May 22, 1997, Roxco submitted a certified claim to Mr. Smithson. Roxco pointed out that the sample wall of CSM blocks did not discolor, and neither did the blocks after they were in place on the buildings. Roxco attributed this to the fact that the sample and the buildings were not shrink wrapped. Roxco said that Treasury was an expert regarding the CSM blocks and should have told Roxco not to store the blocks in the shrink wrap. Roxco asked for $315,640.21 for the effort it made to determine the appropriate method for cleaning the discolored blocks, separating the blocks into piles for cleaning, cleaning 18,445 blocks before they were installed with sixty grit sandpaper, cleaning the same 18,445 blocks after they were installed, 120 days of "expanded general conditions/accelerative costs," claim preparation costs, and overhead and profit. Roxco also asked for $141,345.12 for 120 days of "extended fixed field expenses" that resulted in part from the discoloration problem and in part from the coursing problem, discussed above. Exhibit 355. On September 18, 1997, Treasury's contracting officer denied Roxco s claim. Exhibit 357. Mr. Leech testified that cleaning the stains from the CSM blocks delayed the contract work significantly, and that this delay was obvious. He also said that Roxco could have finished earlier if Treasury had allowed it to place the stained blocks in the walls and then clean the stains. Transcript at 21-23. At the hearing, Mr. Leech explained that the claim for 120 days of expanded general conditions was for the cost of personnel who were added to the project in its final few months. Transcript at 34, 46. He also explained that the claim for 120 days of extended fixed field expenses is his estimate of the time it took to incorporate the activities required due to the staining and the block cutting into the project schedule. He thought that the CSM block work was a critical path activity. Transcript at 46- 47. He estimated that the project could have finished four months earlier than it did, but for the time it took to resolve the coursing and staining problems with the CSM blocks. Transcript at 46-47. Mr. Leech testified that the contract required Roxco to remove mortar splashes from the blocks, and that Roxco could have sanded the blocks to remove the stains and the mortar splashes at the same time. He acknowledged, however, that the process used to remove the mortar splashes was different from the process used to remove the stains, and it was much more difficult to remove the stains than the mortar splashes. Transcript at 22-23, 71-73. Mr. Tatum reviewed a photograph taken after all of the blocks were laid in one wall, but before they were cleaned, and pointed out that the mortar splashes were not found on every block. Transcript at 453-54; Exhibit 407. Mr. Gulde testified that a good mason will keep the masonry units clean as he works, in order to minimize the amount of cleaning needed at the end of the job. Transcript at 496. Mr. Robbins testified that Roxco hired additional masons in order to accelerate its work, and that Roxco cleaned the blocks as fast as they could be laid. Transcript at 395-96. Mr. Turnage agreed with Mr. Robbins that there was never a shortage of CSM blocks on site that were ready to be installed. Transcript at 407. In the opinion of Mr. Mahrenholz, it was cheaper for Roxco to clean the blocks on the ground than it would have been to clean them after they were placed in the walls. Roxco's workers lined up thirty or forty blocks and stood on the ground while they sanded them. No special equipment had to be rented, and the workers did not have to spend time going up and down the walls. Transcript at 421-22. Mr. Hulbert thought that it would have been more efficient to clean the blocks after they were installed because Roxco would not have to segregate the blocks before installing them. Transcript at 175. Mr. Robbins agreed with Mr. Hulbert. He felt that after Roxco demonstrated that it could clean the blocks, Treasury should have permitted Roxco to clean them after they were installed in the walls of the buildings. This would have eliminated the need to sort and stack the blocks according to whether they were stained or clean. Transcript at 545. Arriscraft CSM blocks were used for the expansion of a hospital in Brunswick, Georgia, approximately two to three years before the hearing in this appeal. The blocks were stored on site for approximately six weeks before being installed. No stains appeared on the blocks. Transcript at 249-50, 297-98. The sample wall that Roxco constructed at the beginning of the project was still standing at the end of the project. The CSM blocks on the wall never stained. Transcript at 442-43. Discussion Before turning to the merits of the appeal, we address two preliminary issues. The first concerns admissions, and the second concerns accord and satisfaction. On the merits, we deny both the claim for CSM block coursing and for CSM block discoloration. Admissions Roxco argues that Treasury's internal memoranda, its correspondence with Saxelbye, and its actions constitute an admission that Roxco needed revised drawings in order to erect the CSM blocks. Appellant's Post-Hearing Brief at 2-3. Roxco also argues that Treasury's payment for the blocks constituted an admission that the blocks were properly stored. Appellant's Post-Hearing Brief at 5; Appellant's Reply Brief at 6. In addition, Roxco argues that contacts between Arriscraft and either Treasury or Saxelbye constitute an admission that Arriscraft was supplying Government-furnished material, which makes Treasury responsible for whatever happened to the blocks. Appellant's Post-Hearing Brief at 9. In support of the proposition that Treasury's documents and actions constituted an admission, Roxco cites to Hedin Construction Co. v. United States, 347 F.2d 235 (Ct. Cl. 1965); Winn-Senter Construction Co. v. United States, 75 F. Supp. 255 (Ct. Cl. 1948); George A. Fuller Co. v. United States, 69 F. Supp. 409 (Ct. Cl. 1947); and Daly Construction, ASBCA 32457, 87- 3 BCA 20,182. Roxco relied upon these same four cases in its opposition to Treasury's motion for summary relief, when it argued that Treasury's approval of Roxco's submittal and payment for the blocks constituted an admission that Roxco's storage method was acceptable. As we explained in our interlocutory decision,[foot #] 6 Winn-Senter stands for the proposition that a statement made during contract performance is entitled to great weight; the case did not involve an admission. Hedin and Fuller state that a contracting officer's extension of time (Hedin) and acknowledgment of delays (Fuller) constitute a rebuttable evidentiary admission. The court of appeals in Wilner v. United States, 24 F.3d 1397 (Fed. Cir. 1994), "expressly overrule[d] Hedin to the extent that it stands for the proposition that a contracting officer's decision constitutes a strong presumption or an evidentiary admission of the extent of the government's liability, albeit subject to rebuttal." 24 F.3d at 1403. Daly did not concern an admission and was also decided before Wilner. None of Roxco's cited cases convinces us that Treasury's memoranda, correspondence, or actions constitute an admission. ----------- FOOTNOTE BEGINS --------- [foot #] 6 Roxco, Ltd. v. Department of Treasury, GSBCA ______________________________________ 14430-TD, 00-1 BCA 30,685 (1999). ----------- FOOTNOTE ENDS ----------- To the extent that the evidence of record supports Roxco's arguments, we have taken that evidence into account in resolving the merits of this appeal. Accord and Satisfaction Roxco argues that modification 10 to the contract does not represent an accord and satisfaction of its claim for costs associated with symmetrical coursing. Appellant's Post-Hearing Brief at 3-4. Treasury agrees. Respondent's Reply Brief at 7. The modification does, however, represent an accord and satisfaction of Roxco's claim for costs associated with the addition of control joints. On May 12, 1995, Treasury asked Roxco for a price proposal to cover the costs of providing control joints and symmetrical coursing. Roxco responded with a price proposal on June 6 that included the cost of additional CSM blocks and cuts to blocks. After some discussion, Roxco submitted price proposals on September 13 and 19, 1995, that included the cost of material and cuts to blocks. In modification 10, which the parties signed in June 1996, Treasury accepted Roxco's September proposals for exterior masonry control joint work and increased the contract price by $13,000, and Roxco agreed to release Treasury from all liability for further adjustments attributable to the facts or circumstances that gave rise to the modification. Modification 10 constitutes an accord and satisfaction of Roxco's claim for costs associated with the addition of the control joints. A party claiming an accord and satisfaction must show "'proper subject matter, competent parties, meeting of the minds of the parties, and consideration.'" Brock & Blevins Co. v. United States, 343 F.2d 951, 955 (Ct. Cl. 1965) (quoting Nevada Half Moon Mining Co. v. Combined Metals Reduction Co., 176 F.2d 73, 76 (10th Cir. 1949), cert. denied, 338 U.S. 943 (1950)). In most cases, an executed bilateral contract modification that does not contain a reservation of rights constitutes an accord and satisfaction. Id. All of the elements needed in order to establish an accord and satisfaction are present in this case. To the extent that any part of Roxco's claim for CSM coursing includes costs incurred as the result of the addition of control joints, that part of the claim is barred by accord and satisfaction. CSM Block Coursing In its claim, Roxco asked for $97,170.98 for 1146 additional CSM blocks, 1714 additional cuts that it had to make to the blocks, and four months of additional scaffolding and lift time, plus overhead and profit. Roxco also asked for $141,345.12 for 120 days of extended fixed field expenses that resulted in part from the coursing problem and in part from the discoloration problem, discussed below. Roxco says that Treasury is responsible for these claimed costs because the specifications and the original drawings did not provide Roxco with the information that it needed to lay the blocks, and because the reissued contract drawings made a change to the work required by the original contract documents. The specifications and the original drawings Roxco says that the specifications and original drawings did not contain a "specific expression" of how Treasury wanted Roxco to place the CSM blocks. Appellant's Post-Hearing Brief at 2-3; Appellant's Reply Brief at 4. Roxco also says that Treasury agreed that Roxco needed information in addition to that provided in the original contract documents, in order to place the blocks. Appellant's Reply Brief at 2, 3. Mr. Leech and Mr. Robbins testified about the deficiencies they found in the original contract documents. Mr. Leech wanted Treasury's drawings to contain a specific expression of wall lengths that were neatly divisible by the size of the blocks. He believed that after Treasury decided to use CSM blocks instead of precast concrete panels, it should have revised its drawings to take into account the dimensions of the blocks. Mr. Leech considered the dimensions stated on the drawings to be a fundamental, glaring error because the blocks were two feet long and the walls of the buildings were not all multiples of two feet. He said that the drawings were incomplete and did not tell Roxco what to do with the left over dimensions that would exist at the end of a course of blocks when walls were not multiples of two feet. He also said that Roxco did not know what to do when the dimensions of the walls were not governed by the dimensions of the blocks. Mr. Robbins wanted Treasury's drawings to contain a specific expression of its desire for bonded corners. He said that it was fairly obvious to him that the details on the drawings were inconsistent and that the details provided him with useless information. The testimony of Mr. Leech and Mr. Robbins is consistent with the questions that Roxco asked Treasury on February 6, 1995. In Mr. Gulde's opinion, the original contract documents provided all of the information that Roxco needed in order to construct the walls because the specifications set out the size of the CSM blocks and the type of bond required, and the drawings set out the dimensions of the buildings. There was no doubt in Mr. Gulde's mind that a mason who was given these three pieces of information would have been able to construct the walls of the classroom buildings. Mr. Gulde did not find it unusual that the dimensions of the buildings were not stated in multiples of two feet. He explained that masons take the actual dimensions of buildings into account and then determine how many cuts to masonry units will be needed. In addition, when the width of a masonry unit is not half its length, as is the case with the CSM blocks, the mason will need to make special cuts to the blocks in order to make them bond around a corner. Mr. Gulde agreed with Mr. Reed that masons build walls by laying the corners first, and they do that by first determining what cuts will be needed. He also explained that bonded corners are used in 99.9% of masonry construction. He did not find the differences between the details shown on the contract drawings to be an error, because he read the more specific detail (8/A3.6), which showed a bonded corner, as elaborating upon the more general detail (B/A3.6), which did not show a bonded corner, but which referred to the more specific detail. Mr. Gulde's explanation of the contents of the original contract documents was cogent and persuasive. We are not convinced that the original contract documents deprived Roxco of information that it needed in order to lay the CSM blocks. Although the lengths of some of the walls of the classroom buildings were not multiples of the length of the CSM blocks, this did not make the drawings defective so as to require Treasury to re-design the classroom buildings. Buildings are not always designed with the size of masonry units in mind, and part of a mason's craft is to determine how to lay masonry units when a wall is not a precise multiple of the size of the units. The original contract documents gave Roxco the size of the masonry units, the type of bond required, and the dimensions of the walls. With these three pieces of information, a mason can determine how to build a wall. Drawings are not defective simply because the dimensions of the walls are not ideally suited to the size of the masonry units. In addition, although one contract drawing detail did not show a bonded corner, it referred to a second, more specific detail that showed a bonded corner and that said to cut the CSM blocks to course as shown. Because the specific detail showed a bonded corner and because very nearly all masonry construction incorporates bonded corners, Roxco should not have been surprised to discover that Treasury wanted bonded corners. Further, because the width of the CSM blocks was not half the length of the blocks, Roxco should have realized that it would need to make special cuts to the blocks so that they would bond around the corners. If the dimensions of the building constituted a fundamental, glaring error, as Mr. Leech testified, or if the drawing details provided useless information about the corners and contained an inconsistency that was fairly obvious, as Mr. Robbins testified, Roxco should have asked about those defects and deficiencies before it submitted a bid for the project. Apparently, whoever prepared Roxco's estimate and its bid did not find any glaring, obvious, fundamental errors in the original contract documents, because Roxco did not ask Treasury any questions about those documents before submitting its bid. Likewise, Mr. Gulde did not find that the original contract documents deprived Roxco of information it needed in order to lay the blocks. Drawings do not typically show how every block will be placed in a building, and if Roxco needed such specific information in order to perform its work, it should not have waited until after award of the contract to inform Treasury of its needs. In support of its argument that Treasury agreed that Roxco needed information in addition to that found in the original contract documents in order to lay the CSM blocks, Roxco relies upon Mr. Reed's March 24, 1995 note to Saxelbye, and upon the fact that Saxelbye provided additional drawings for free. Roxco says that Treasury would not have ordered additional drawings only for the purpose of assisting Roxco, and that Saxelbye would not have provided additional drawings for free except to correct its own errors. Appellant's Reply Brief at 2, 3. Treasury did not agree that Roxco needed help in performing the CSM block work because of omissions in the original contract documents.[foot #] 7 Mr. Reed's note on the March 24 telefacsimile cover sheet says, in part, "Need coursing layout for CS masonry." The note does not say that Roxco needed additional information about coursing because something was missing from the contract documents. Mr. Reed did not understand why Roxco was confused by the contract documents, and Roxco's questions about "which corner gets the 2" piece" and regarding a five feet-two inch long "panel" that was actually the distance between two points on a drawing and unrelated to the masonry work, caused him to question the competence of Roxco's masonry workers. He concluded that Roxco needed some help in order to accomplish the contract work, and that is why he sent his March 24 request to Saxelbye. Even though Roxco finds it unlikely that Treasury would have ordered revised drawings only for the purpose of assisting Roxco, Mr. Reed's testimony was credible, and his request for added drawings was consistent with his position as the person responsible for managing the construction project. Roxco reads too much into Mr. Reed's note when it argues that the note evidences an agreement by Treasury that the original contract documents omitted information that Roxco needed in order to lay the CSM blocks. Saxelbye provided the April 14, 1995 revised drawings for free. Roxco concludes that Saxelbye did so because the original drawings omitted necessary information regarding the placement of the CSM blocks. We have no evidence, however, to establish why Saxelbye did not charge Treasury for the drawings. Perhaps Saxelbye provided the additional drawings for free because, as Mr. Smithson suggested, not much effort was required to prepare the drawings. Perhaps Saxelbye provided the drawings for free because it had a delivery order contract with Treasury, and it wanted to satisfy its customer. Perhaps Saxelbye simply neglected to send Treasury a bill for the revised drawings. No evidence supports the inference, which Roxco asks us to draw, that Saxelbye provided the revised drawings because the original ----------- FOOTNOTE BEGINS --------- [foot #] 7 As discussed earlier, Treasury did agree that the drawings omitted control joints, and the parties signed modification 10 in order to compensate Roxco for the work it performed in order to add the control joints. ----------- FOOTNOTE ENDS ----------- drawings contained omissions that deprived Roxco of information that it needed in order to lay the CSM blocks. The reissued drawings Roxco's second argument in support of its CSM block coursing claim is that the reissued contract drawings made a change to the work required by the original contract drawings. Specifically, Roxco asserts that the first set of reissued drawings required it to use symmetrical coursing, which caused it to need additional blocks and which also caused it to make many extra cuts to the blocks.[foot #] 8 Appellant's Reply Brief at 3; Appellant's Proposed Finding of Fact 26. The original contract documents required Roxco to lay CSM blocks of a specified size, in a running bond, on walls of specified dimensions, in a skillful and workmanlike manner. These requirements remained the same for the duration of the contract. The original contract drawings did not show CSM blocks, except in two details. Treasury provided the first reissued drawings, dated April 7, 1995, in response to Roxco's request for additional details regarding the blocks. The reissued drawings showed CSM blocks on some sections of the elevations of the buildings, and they also contained a new detail, 1/A3.1, which showed a typical corner.[foot #] 9 In the corner detail and the elevations, some blocks at the ends of some courses were cut in order to achieve a running bond in the remainder of those courses. Where the reissued elevation drawings showed CSM blocks, the blocks were laid symmetrically. In order to establish that the reissued drawings changed the contract requirements, Roxco must show that symmetrical coursing was a new requirement and was not a part of the contract until Treasury provided the reissued drawings, and it must also show that the symmetrical coursing depicted in the reissued drawings constituted a contract requirement. Roxco has made neither showing. ----------- FOOTNOTE BEGINS --------- [foot #] 8 Treasury sent Roxco three sets of reissued drawings. The final two sets, dated April 27 and May 5, 1995, dealt with the addition of the control joints. In modification 10, Treasury compensated Roxco for whatever costs it incurred as a result of the addition of the control joint work. Roxco's claim is based, therefore, upon the first set of reissued drawings, dated April 7, 1995, that Saxelbye forwarded to Treasury on April 14, 1995. [foot #] 9 The reissued drawings also contained a second new detail, 1/A3.2, which showed a wing wall, as did original contract drawing detail 8/A3.6. The only difference between the original detail and the new detail is that they show the wing wall from different perspectives. ----------- FOOTNOTE ENDS ----------- Roxco argues that Treasury's May 12, 1995 request for a price proposal proves that Treasury added symmetrical coursing to the contract requirements. The evidence, however, does not support Roxco's argument. After Treasury added the control joints to the buildings, Roxco asked for additional details regarding the coursing at the control joints and cuts to the blocks at the control joints. As Mr. Robbins explained, whenever a feature such as a control joint is added to a building, the coursing will have to be adjusted. In response to Roxco's request, Treasury provided a reissued drawing that showed the cuts at a typical control joint. A few days later, Treasury asked for a price proposal to cover the costs of providing control joints and symmetrical coursing. Mr. Smithson testified that symmetrical coursing had always been required, and that the May 12 request was meant to cover whatever changes to the symmetrical coursing would be needed due to the addition of the control joints. Reading the request's reference to symmetrical coursing in context, we conclude that Treasury asked for a price to change the coursing as needed when the control joints were added, not for a price to change the coursing by imposing a requirement for symmetry. The hearing testimony established that symmetrical coursing in a project such as the classroom buildings is to be expected. According to Mr. Leech, symmetry might not be important in the interior of a warehouse, but it is important to try to achieve symmetrical coursing in a "monumental" project, such as the classroom buildings. Mr. Robbins said that if he were laying blocks in a running bond, he would make the courses symmetrical because it is a better way of performing the work. Mr. Gulde said simply that symmetry is part of building a wall. Mr. Smithson said that symmetry was required from the beginning of the contract. If Roxco is correct that symmetry was not required until Treasury provided Roxco with the April 7 reissued drawings, we would expect to have evidence in our record showing that before bidding, Roxco read the original contract documents as permitting it to lay the blocks using something other than symmetrical coursing, or that at some point Roxco planned to lay the blocks using something other than symmetrical coursing. Our record, however, contains no such evidence. Due to the consistency of the testimony that symmetry was an implied part of laying the blocks on the walls of the classroom buildings in a skillful manner and the lack of evidence to the contrary, we are not convinced that the original contract documents lacked a requirement for symmetry. Even if Roxco had established that the original contract documents permitted it to lay the blocks without symmetrical coursing, Roxco has not shown that Treasury required Roxco to lay the blocks as shown on the April 7 reissued drawings. Although Treasury provided Roxco with the reissued drawings as Roxco requested, Treasury never insisted that Roxco follow the coursing scheme shown in those drawings. Several times, Treasury told Roxco that it could use whatever coursing scheme it intended to use when it entered into the contract. Roxco asks us to ignore Treasury's statements because, Roxco says, Treasury was merely attempting to evade responsibility for defective drawings. Appellant's Reply Brief at 3. As discussed previously, however, we do not find any defects in the drawings. Further, even if the drawings were defective, we do not see how Treasury was trying to evade responsibility when it said that Roxco did not have to use the coursing scheme shown on the revised drawings. Treasury quite logically assumed that when Roxco submitted a bid for the project without asking any questions, it had in mind some workable plan for installing the blocks. We accept at face value Treasury's statements to Roxco that it was free to use whatever coursing scheme it intended to use when it submitted its bid. Apparently, Roxco also accepted Treasury's statements, because it did not always build its courses as shown on the reissued drawings. The evidence establishes that Treasury provided Roxco with the April 7 reissued drawings at Roxco's request, but did not require that Roxco lay the blocks as shown on the drawings. Summary Laying the CSM blocks was difficult for Roxco, and the contractor was surprised at the number of cuts it had to make to the blocks. The evidence does not establish, however, that Roxco's difficulties or surprise can be ascribed to Treasury. Roxco has not shown that the original contract documents omitted any vital information about laying the CSM blocks. The specifications and the drawings provided Roxco with all of the information it should have needed when they spelled out the dimensions of the buildings, the size of the CSM blocks, and the type of bond required. If Roxco found that those documents lacked the information that it needed in order to lay the blocks, it should have asked Treasury for additional information before bidding. Treasury did not conclude that Roxco needed more information due to any deficiencies in the original contract documents. Roxco has not convinced us that Treasury changed the original contract requirements by adding a requirement for symmetry. Treasury provided Roxco with reissued drawings at Roxco's request, but did not require that Roxco lay the blocks as shown on the drawings. We deny the claim for CSM block coursing. CSM Block Discoloration In its claim, Roxco asked for $315,640.21 for the effort it made to determine the appropriate method for cleaning the discolored blocks, separating the blocks into piles for cleaning, cleaning 18,445 blocks before they were installed, cleaning 18,445 blocks after they were installed, 120 days of expanded general conditions/accelerative costs, claim preparation costs, and overhead and profit. Roxco also asked for $141,345.12 for 120 days of extended fixed field expenses that resulted in part from the discoloration problem and in part from the coursing problem, discussed above. A number of contract clauses seem to allocate to Roxco the responsibility for the CSM blocks. For example, the Permits and Responsibilities clause provided that Roxco was responsible for all materials delivered until completion and acceptance of the entire work. In addition, the Payments Under Fixed-Price Construction Contracts clause provided that payment by Treasury did not relieve Roxco from its sole responsibility for all material. Treasury supplemented this clause to provide that if Treasury paid for materials stored at the job site, Roxco would remain responsible for adequately securing those materials. The Inspection of Construction clause provided that if Treasury made any inspections, they were for its sole benefit and did not relieve Roxco of responsibility for damage to material before acceptance. The Material and Workmanship clause provided that all material incorporated into the work was to be new and of the most suitable grade for the purpose intended. The contract also required Roxco to store all supplies on the project site so as to preclude climatic damage, and the solicitation provided bidders with climatological data about the Glynco area. Despite these contract clauses, Roxco asserts that it is not responsible for the stains that appeared on the CSM blocks. Roxco argues that there was no source of the CSM blocks other than Arriscraft, and that we should treat the blocks as Government-furnished material and hold Treasury responsible for the discoloration problem.[foot #] 10 Appellant's Post-Hearing Brief at 8-10. Roxco also argues that Treasury did not cooperate with Roxco's efforts to solve the discoloration problem and imposed a constructive change when it told Roxco to ----------- FOOTNOTE BEGINS --------- [foot #] 10 Roxco says that Treasury and Saxelbye had "consistent direct contacts" with Arriscraft, showing that Arriscraft was not Roxco's supplier "in any normal sense; but was essentially a supplier of Government-furnished material." Appellant's Post-Hearing Brief at 9. Although the facts regarding the contacts among Treasury, Saxelbye, Wood, Arriscraft, and Roxco do not establish that Arriscraft was anything other than Roxco's supplier, we address Roxco's argument below. ----------- FOOTNOTE ENDS ----------- clean the blocks before placing them in the buildings. Appellant's Post-Hearing Brief at 6-8.[foot #] 11 Sole source/Government-furnished material Assuming, without deciding, that the CSM blocks should be treated as Government-furnished material or that they were available from only one source, it does not necessarily follow that Treasury is responsible for the discoloration problem. If the CSM blocks had been Government-furnished material, Treasury would be responsible for the discoloration problem if Roxco could establish that the blocks were not suitable for use. Topkis Brothers Co. v. United States, 297 F.2d 536 (Ct. Cl. 1961). If the blocks were available from only one source, Treasury would be responsible for the discoloration problem if Roxco could establish that Treasury breached its warranty that the requirements of the contract could be met by using the blocks. Cascade Electric Co., ASBCA No. 28,674, 84-1 BCA 17,210. We cannot simply conclude that because the blocks stained, they were unsuitable for use or could not meet contract requirements. Instead, we must determine whether the blocks stained due to some characteristic or quality of the blocks, or whether the stains appeared due to some other factor. During the course of performance, Roxco provided Treasury with several theories as to why the stains appeared on the blocks. In November 1995, Roxco told Treasury that the atmospheric conditions and environment in Glynco were responsible for the stains, which were caused by a reaction of the blocks to the elements. Roxco suggested that the blocks would continue to stain after they were laid in the walls because the stains were a natural condition of the blocks. In April 1996, Roxco told Treasury that the presence of magnesium in the blocks predisposed them to discolor in the moist, salty environment, and Roxco criticized Treasury for not telling bidders that Glynco is near the Atlantic Ocean and is a humid, salty area. In October 1996, Roxco told Treasury that the stains were caused by the presence of iron and aluminum in the blocks, exacerbated by the salty environment. Roxco's theories as to why the stains appeared are not borne out by the facts. The blocks laid in the sample wall that Roxco constructed a few weeks after the blocks arrived at the job site did not discolor. Similarly, the clean, dry blocks that Roxco laid in the buildings did not stain. In addition, blocks that were stored for approximately six weeks before being installed in ----------- FOOTNOTE BEGINS --------- [foot #] 11 In one sentence in its opening brief, Roxco says that Treasury, through its architect, had superior knowledge of potential problems with the CSM blocks. Appellant s Post- Hearing Brief at 6. Roxco does not develop this argument, however, and there is no evidence to show that Treasury knew any more than Roxco did about the blocks. ----------- FOOTNOTE ENDS ----------- a hospital near Glynco did not stain. The blocks in the sample wall, the blocks in the buildings, and the blocks in the hospital did not stain even though they were exposed to the same moist, salty environment as were the shrink-wrapped blocks. Arriscraft explained that iron and aluminum would be found in many, if not most, masonry products and that all masonry has some degree of absorption. Arriscraft has been manufacturing and selling blocks throughout North American for many years, and the blocks have been used successfully on many projects on the Atlantic Ocean coast. Many masonry manufacturers warn about exposing their products to salt, but the warning has nothing to do with the possibility of stains appearing on the masonry units. Clearly, the environment in Glynco did not, by itself, cause the blocks to discolor. The evidence in the record establishes that the blocks discolored because they were stored in their original shrink wrap for many months in a moist, hot environment. When Roxco ordered the blocks in December 1994, Arriscraft made a special production run in order to supply them quickly because Roxco said that it needed them quickly. Most of the blocks arrived at the job site in February and March 1995. Arriscraft's product information made it clear that the CSM blocks could stain, and cautioned Roxco to cover the blocks if they were exposed to the weather for an extended period of time. From the time the blocks arrived until Roxco began laying them in October 1995, they remained in their original shrink wrap. Summers in Glynco are typically hot and humid, as the climatological data that Treasury provided to bidders showed, and condensation built up inside the shrink wrap. Wood, the distributor for the blocks, told Roxco that the blocks discolored because they were allowed to get very wet and to remain that way for eight months. Arriscraft's chemical engineer told Roxco that Wood's statements were consistent with her laboratory analysis, which showed growths of algae and mold on the blocks. She explained how such growths thrive on masonry in warm, moist conditions. Arriscraft concluded that the blocks stained after Roxco left them tightly wrapped for many months in a damp, humid environment without taking any steps to ensure that they would stay dry and clean. Roxco contends that the stains appeared even though it stored the blocks properly, so it should not be held responsible for the problems that resulted from the stains. Appellant's Post-Hearing Brief at 4-6. In support of its contention that it stored the blocks properly, Roxco points out that Treasury did not immediately criticize Roxco's storage methods when the stains came to light. It is true that instead of forming a conclusion in the absence of facts, Treasury repeatedly asked Roxco for information regarding the cause of the stains. Roxco told Treasury several times that it was gathering information about the cause of the stains, and it received an opinion from a chemist on March 21, 1996, regarding the cause of the stains. Roxco did not share that information with Treasury, however, until October 28, 1996. We cannot fault Treasury for wanting to gather information instead of hurriedly blaming Roxco's storage method for the discoloration problem. Treasury's lack of swift criticism about the manner in which the blocks were stored does not prove that Roxco stored them properly. Roxco also points out that Treasury did not show a correlation between the presence of stains and the position of the blocks in the pallets, and says that this shows that the presence of the stains was unrelated to the manner in which the blocks were stored. We have no proof, however, that there would have been such a correlation if the stains were related to storage. Also in support of its position that it stored the blocks properly, Roxco says that Treasury agreed with the manner in which the blocks were stored by approving the information that Roxco submitted about the CSM blocks, by providing an area to place the blocks when they were delivered, and by paying for the blocks soon after they were delivered. Although Treasury approved Roxco's submittal, it did so well before the blocks were delivered to the job site. Thus, approving the submittal did not constitute Treasury's concurrence that Roxco stored the blocks properly. Treasury's payment for the blocks does not show that Treasury monitored and reassessed the storage of the blocks during the following seven or eight months. Likewise, although Treasury selected a spot where Roxco could store the blocks when they were delivered, this does not establish that for the next few months Treasury continued to evaluate Roxco's storage of the blocks. The contract made Roxco responsible for the blocks until completion of the work, and provided that payment by Treasury did not relieve Roxco of that responsibility. The contract did not impose upon Treasury an obligation to participate in decisions regarding when materials would be delivered, whether they would be stored at the job site or in a warehouse, or how they would be stored. The contract did not require Treasury to monitor the blocks so that it could keep Roxco informed of how they were faring while stored. Even if Treasury had inspected the blocks, the contract provided that Roxco would have remained responsible for the blocks. None of Treasury's actions establishes that Roxco stored the blocks properly. Roxco also says that it followed all of Arriscraft's directions, which proves that it stored the blocks properly. Arriscraft's product literature said to cover the blocks if they were exposed to the weather for an extended period of time, and Roxco says that it complied with this direction by placing a sheet of plastic on top of the shrink-wrapped blocks. Although Roxco covered the blocks with plastic after the discoloration problem came to light, we are not convinced that Roxco covered the blocks for the entire time before they were installed. In addition, Roxco has not established that a sheet of plastic would have constituted an appropriate cover for the extended period of time that it stored the blocks. Arriscraft's representative saw that the blocks were wet in August 1995, and recommended that Roxco remove the shrink wrap from the blocks for several days before installing them so that they could dry. Arriscraft's recommendation, which was made before the discoloration problem came to light, was hardly a ringing endorsement of the manner in which Roxco had stored the blocks. Although Roxco took the blocks out of the shrink wrap for several days before installing them, it did not allow all of the blocks to dry before installation. Roxco did not comply with what turned out to be the most important direction contained in Arriscraft's product literature, which was to provide the necessary means to prevent staining during storage. If Roxco did not know what means were necessary to prevent stains when the blocks were stored outdoors for a lengthy period of time, it should have consulted with either Wood or Arriscraft. When Roxco ordered the blocks in December 1994, however, it said that it needed them right away. We do not know what Arriscraft's storage directions would have been had it known that Roxco intended to store the blocks for seven or eight months, through a summer in south Georgia, before installing them. Roxco has not established that the properties of the CSM blocks made them either unsuitable for use or unable to be used to meet the contract's requirements. When the blocks arrived at the job site, they were in good condition. When the blocks were not left in their shrink wrap for an extended period of time, they did not discolor. The blocks did not have any characteristic that made them discolor when they were exposed to the climate in Glynco. The discoloration problem only arose when the blocks were stored in shrink wrap for many months before being used. It was Roxco's decision to have the blocks delivered early and stored outdoors at the project site for an extended period of time, and it was Roxco's decision to leave the blocks in their shrink wrap. The contract provided that Roxco was responsible for the blocks until completion and acceptance of the work, and Roxco has not established that it should be relieved from that responsibility. Constructive change/Failure to cooperate Roxco contends that Treasury constructively changed the contract and failed to cooperate with Roxco's efforts to overcome the effects of the discoloration of the CSM blocks. Specifically, Roxco says that once it demonstrated that the blocks could be cleaned, Treasury should not have refused to permit Roxco to remedy the stains by laying the stained blocks in the walls and then cleaning them, as Roxco wanted to do. Roxco says that Treasury knew that some stains appeared only after blocks were in place, and it allowed Roxco to clean those blocks in the walls. Roxco contends that Treasury's decision not to permit Roxco to install stained blocks and clean them after they were laid was arbitrary and contrary to Roxco's right to choose any method permitted by the contract for performing the work. Treasury's direction in October 1995, to clean the blocks before placing them in the wall was not arbitrary, and it was not a change to the terms of the contract. The Arriscraft product information that Roxco submitted to Treasury said that the blocks were supposed to be cleaned with fiber, not wire, brushes and that other methods could be used only with advice from Arriscraft. In October, after Treasury called the staining problem to Roxco's attention, Arriscraft said that the blocks could be cleaned by sandblasting so long as the surface of the blocks was not harmed, and Wood and Arriscraft were looking into chemical cleaning options. Arriscraft said that the blocks should be cleaned individually before being laid, so as to remove any risk of damaging the mortar joints by sand blasting or chemical cleaning. At that time, Roxco did not propose or demonstrate a method for cleaning the blocks. Treasury's direction to clean the blocks before placing them in the walls was rationally based in part upon Arriscraft's statements and in part upon the contracting officer's concern that some of the blocks might not ever come clean. Treasury's direction was not a change to the terms of the contract, because the contract did not permit Roxco to install stained blocks intentionally. Treasury's decision not to act upon Roxco's November 1995 suggestion that it could clean the blocks in place was not arbitrary. On November 6, Wood told Roxco that the proper way to proceed would be to sandblast or to clean the blocks chemically, and said that it was experimenting with different chemical cleaners. On November 14, Roxco told Treasury that it could easily clean the blocks with 100 grit sandpaper or by sandblasting, and it also said that the mortar joints would not be damaged if the blocks were cleaned in place. Roxco did not provide anything to show that Arriscraft would approve using sandpaper to clean the blocks, and did not explain its basis for saying that cleaning in place would not harm the mortar joints, which was contrary to Arriscraft's earlier statement. In addition, Roxco's suggestion for cleaning the blocks was inconsistent with the conclusion that Wood reached a few days earlier. Roxco also said that it was still investigating the cause of the stains. Treasury did not act arbitrarily when it failed to accept Roxco's bare statement that the blocks could be cleaned after they were laid. Treasury's decision to accept Roxco's December 1995 proposal for cleaning the blocks was rational. On December 12, Treasury's project manager reminded Roxco that it needed to propose a method for cleaning the blocks, and mentioned the contracting officer's earlier direction that the blocks were to be cleaned before being laid. Roxco responded that it was using sandpaper to clean the smooth faced discolored blocks before installing them, that it intended to sandblast the rough faced blocks, and that it would use these same cleaning methods for stained blocks already in place. On December 14, Roxco provided Treasury with a letter stating that Arriscraft agreed that the blocks could be cleaned with eighty grit sandpaper, and Roxco gave Treasury a template that it intended to use to protect the mortar joints when it sandblasted blocks in place. Roxco did not propose to clean blocks after they were laid, except for the stained blocks that were already in place. Roxco demonstrated its proposed cleaning method on December 20, and Treasury accepted that cleaning method. Treasury's decision to accept Roxco's proposed cleaning method was not arbitrary because the proposal was supported by Arriscraft and had been shown to work successfully. The evidence does not establish that after December 1995, Roxco clearly expressed its desire to place stained blocks in the walls and then clean them.[foot #] 12 Although the parties exchanged numerous letters regarding the project, we found no written request by Roxco to lay stained blocks and clean them later. Mr. Robbins testified that after Roxco developed an acceptable method for cleaning the blocks, he asked Mr. Smithson if Roxco could install stained blocks and clean them after they were laid, and Mr. Smithson denied the request. Mr. Robbins did not recall exactly when he spoke with Mr. Smithson, although he thought it was in December 1995, or January 1996. Mr. Smithson did not remember Mr. Robbins's request. If the subject had been discussed in any depth, we would expect that one of the witnesses would have had a better recollection of the event or that there would be some documentation of Roxco's request. If Mr. Robbins's recollection is correct, he made his request before Roxco had acquired much experience using its approved cleaning method, and it would not have been unreasonable for Treasury to take some time to see whether all of the blocks could be cleaned. After the parties learned that approximately one percent of the stained blocks would not come clean, it would not have been arbitrary for Treasury to have rejected a request to place stained blocks in the walls. In any event, Roxco never presented Treasury with a proposed procedure for laying stained blocks and then successfully cleaning them in place, or anything to show that Arriscraft would have agreed to such a procedure. Treasury's decision regarding the stains that appeared after the blocks were in place does not show that its other decisions were arbitrary. In early November 1995, Treasury told Roxco that some blocks had been wet when Roxco laid them and stains appeared as the blocks dried. Roxco's December 14 proposal for cleaning the blocks included a method for cleaning stained blocks that were already in place. Instead of requiring Roxco to remove and replace the stained blocks, as the contract provided, Treasury decided that Roxco could clean those blocks in place, as it proposed to do. This decision does not show that Treasury arbitrarily rejected a later request to place thousands of ----------- FOOTNOTE BEGINS --------- [foot #] 12 Roxco says that it was not required to protest constantly Treasury's direction to clean the blocks before they were laid. Appellant's Reply Brief at 10. Although this is true, neither was Treasury required to intuit that Roxco wanted that direction changed. ----------- FOOTNOTE ENDS ----------- stained blocks in the walls deliberately. Although Roxco was entitled to develop its own method for cleaning the blocks, the contract did not allow Roxco to incorporate stained blocks into the walls of the buildings on purpose. Relaxing the contract's requirements by allowing Roxco to clean a known number of stained blocks already in place, was not inconsistent with adhering to the contract's requirements by not allowing Roxco to place an unknown number of stained blocks in the walls of the buildings intentionally. Summary Like most Government contracts, the one between Roxco and Treasury distributed responsibility between the parties for a variety of events. Responsibility for materials was given to Roxco until its work was completed and accepted. The fact that the blocks stained does not establish that there was something wrong with the blocks. The blocks stained because Roxco stored them in shrink wrap for an extended period of time in a moist, hot environment. Roxco, not Treasury, decided when to have the blocks delivered to the job site and how to store the blocks. Treasury had no obligation to allow Roxco to place thousands of stained blocks in the wall deliberately and hope that they could all be cleaned later. The contract allocates to Roxco the responsibility for the problems associated with the discoloration of the blocks. We deny the claim for CSM block discoloration Decision The appeal is DENIED. ___________________________________ MARTHA H. DeGRAFF Board Judge We concur: ________________________________ _________________________ _________ ANTHONY S. BORWICK CATHERINE B. HYATT Board Judge Board Judge