Georgetown Steel Corporation v. Union Carbide Corporation, Law Engineering Testing Company Pittsburgh Testing Laboratory, Inc., Georgetown Steel Corporation v. Union Carbide Corporation, Law Engineering Testing Company Pittsburgh Testing Laboratory, Inc.

892 F.2d 1041 NOTICE: Fourth Circuit I.O.P. 36.6 states that citation of unpublished dispositions is disfavored except for establishing res judicata, estoppel, or the law of the case and requires service of copies of cited unpublished dispositions of the Fourth Circuit. GEORGETOWN STEEL CORPORATION, Plaintiff-Appellant, v. UNION CARBIDE CORPORATION, Defendant-Appellant, Law Engineering Testing Company; Pittsburgh Testing Laboratory, Inc., Defendants-Appellees. GEORGETOWN STEEL CORPORATION, Plaintiff-Appellant, v. UNION CARBIDE CORPORATION, Defendant-Appellant, Law Engineering Testing Company; Pittsburgh Testing Laboratory, Inc., Defendants-Appellees. Nos. 88-2884, 88-2979. United States Court of Appeals, Fourth Circuit. Argued: June 5, 1989. Decided: Dec. 15, 1989. Rehearing and Rehearing In Banc Denied Jan. 29, 1990. Charles Porter (E. Russell Jeter, Jr., James B. Moore, Jr., McNair Law Firm, P.A., on brief); Thomas S. Tisdale, Jr. (Stephen P. Groves, Young, Clement, Rivers & Tisdale, on brief), for appellants. Robert Bruce Wedge (Stokes, Shapiro, Fussell & Wedge, on brief); Robert O'Neal Fleming, Jr. (Smith & Fleming, on brief), for appellees. Before WIDENER and K.K. HALL, Circuit Judges, and GEORGE ROSS ANDERSON, Jr., United States District Judge for the District of South Carolina, sitting by designation. PER CURIAM: — 1 — This diversity case arises out of Georgetown Steel Corporation's ("GSC") use of steel slag as backfill under a railroad track scale, a truck scale, a refractory warehouse, and an oxygen separation plant, all of which are located on GSC's steel mill site in Georgetown, South Carolina. Shortly after swelling slag destroyed the oxygen separation plant which was owned and operated by Union Carbide Corporation ("Carbide"), GSC filed suit against Pittsburgh Testing Laboratory, Inc. ("PTL") and Law Engineering Testing Company ("Law") alleging negligence and breach of express and implied warranties on the part of PTL and Law in connection with their opinions regarding the suitability of slag as fill material. GSC also filed suit against Carbide for declaratory judgment to determine the respective rights and responsibilities of GSC and Carbide. Carbide cross-claimed against Law and PTL and counterclaimed against GSC. The court allowed GSC to recover for the damage to the track scale as a result of PTL's negligence but denied GSC's remaining claims against both PTL and Law. The court denied all of Carbide's claims. GSC and Carbide appeal. For the reasons set forth below, we affirm in part and reverse in part and remand. I. — 2 — A rather complex set of facts gave rise to this lawsuit. In 1979, Stephen Rishel, a civil engineer, was hired by GSC. His job required him to perform the civil engineering necessary for all of GSC's construction projects. His duties included some engineering design, the writing of specifications, preparation of drawings and the supervision of construction contractors. Prior to going to work at GSC, Rishel worked as a construction engineer for a highway contractor in the Pittsburgh area. He had some experience with steel slag, having encountered it on at least two construction jobs and having seen it used behind retaining walls, as a base to support pavement, as backfill underneath drain pipe, and behind bridge abutments. In addition, he knew something of the expansive properties of certain slag having heard from a carpenter foreman on one of his construction jobs that a swimming pool under which steel slag had been placed was destroyed when the slag swelled. — 3 — Shortly after starting work at GSC, Rishel considered using GSC's steel slag as backfill as a cost saving measure.1 Because of his knowledge that certain slag does swell, Rishel obtained approval from his supervisor, Bill Dobinski, to have the slag tested for its expansive properties. When he decided to have the slag tested, Rishel was not aware of any expansive characteristic of the particular slag produced by GSC nor was he aware that GSC frequently had its slag chemically analyzed and had been doing so since 1977.2 Rishel contacted PTL and requested that certain tests be run on the slag to determine its expansiveness. An employee of PTL, Greg Kuske, conducted several tests on a slag sample including a swell test, under the direction of Dr. Harry Wu, who was totally unfamiliar with slag and its expansive nature. After the ninety-six hours of testing required by the American Society of Testing Materials, the percentage of swell was found to be .09%. Although Rishel had not asked PTL to determine whether the slag would be appropriate for use as backfill under any particular structures, on March 27, 1979, PTL issued a report with the following conclusion: — 4 — Based on our tests results, the subject material is a well-graded, course-grained, greenish gray mill slag. The material has a very little effect on swell due to change in its moisture content. This material may be suitable for using as an engineered fill, provided that the material should be compacted to a minimum of 95% maximum density as obtained by Modified Proctor ASTM D-1557. — 5 — On the basis of this report, Rishel made the decision to use the slag for fill. His receipt of this report was the last contact Rishel or anyone else at GSC had with PTL until after problems with the slag backfill under the oxygen separation plant surfaced in 1982. — 6 — After he received the PTL report, Rishel used slag as backfill around the walls of a railroad track scale. A track scale contains the weighing mechanism for railroad cars, and is a concrete structure the size of a railroad car, sitting in the ground. Piles were placed under the scales, but slag was used to backfill around the walls of the structure. Shortly after its construction, problems developed with the track scale when the walls deflected inward and bound the weighing deck. Rishel testified that he did not suspect expanding slag as the problem at this point. The scale was repaired by its builder, Colt Industries, by placing jacks in the walls for structural support. Heavy truck traffic was also eliminated from the scales at the builder's suggestion. In September 1984, GSC had the slag replaced with gravel at a cost of $3,233. — 7 — In early 1980 slag was used as backfill to support the walls of a truck scale which was being installed on GSC's site. A truck scale resembles a track scale but weighs trucks instead of railroad cars. When problems developed with the pavement around the scale, it was moved to another location on the GSC site and a material other than slag was used as backfill because it was more convenient. — 8 — In the fall of 1980 slag was used as backfill under an extension built on to an existing refractory warehouse. Approximately a year later, GSC became aware of cracks in the floor of the refractory warehouse extension and attempted to repair the cracks by filling them with grout. The cracks reopened within a year. Rishel testified it was at this point, in the fall of 1982, that he first suspected that there might be a problem with the slag fill. — 9 — Three years earlier, in the fall of 1979, Rishel had learned that GSC and Carbide were negotiating for the construction of an oxygen separation plant on the GSC plant site. The plant's purpose was to provide GSC with oxygen and nitrogen in gaseous form at a price less than that at which GSC previously purchased the gases. Under an agreement entered into by Carbide and GSC on January 23, 1981, GSC agreed to provide a site for Carbide to build the plant, have an analysis of the soil conducted, and perform the necessary preparation of the site for construction. Preparation of the site involved two key elements. First, because the site was in close proximity to a river, proper support for the plant had to be determined. Second, because of the plant's structure and the sensitive nature of its equipment, Carbide would be able to tolerate only minimal differential settlement of the underlying support. — 10 — GSC contacted Law about providing professional engineering advice and testing services to assist in preparation of the site. On February 8, 1980, Law submitted a proposal to conduct a geotechnical exploration of the plant site, specifically, to provide "a general subsurface exploration to determine one or two reasonable solutions to major foundation and earthwork design and construction problems." On February 12, 1980, GSC and Law entered into a contract which provided that Law would furnish all labor, equipment, supervision, materials, tools, engineering, insurance, etc., "necessary to perform a subsurface investigation within the proposed construction area for Georgetown's New Oxygen plant," and would "analize [sic] the conditions encountered to determine any solutions to major foundation and earthwork design problems and submit a written report with these findings to Georgetown." — 11 — Pursuant to this contract, and after conducting an exploration of the plant location and performing various soil tests, Law issued a report on March 24, 1980. Stated in the report the purpose of the exploration was "to develop information about the site [for the oxygen separation plant] and subsurface conditions that could be used for assessing foundation alternatives for the new facilities." Law's investigation revealed that the area had been built up to its current elevations using a fill consisting of slag from the furnace operations at GSC. The fill contained soil, metal, concrete, and both organic and inorganic debris. Law made three alternative recommendations for site preparation: (1) dynamic compaction of the existing slag and sand fill to eliminate voids; (2) excavation of the soil/slag mixture and replacement with properly compacted sand; and (3) placement of the plant on properly driven piles. GSC had originally requested that Law evaluate the possible use of piles. The third alternative was offered in response to GSC's request even though Law noted that shallow foundations would be adequate to carry the loads involved, assuming proper site preparation. As to shallow foundations, a section in the report stated: — 12 — In our opinion, mat-type foundations bearing on slag fill or new fill of silty sand (properly improved as outlined in the following section) could be used for the support of the proposed facility. — 13 — * * * — 14 — * * * — 15 — There is a greater degree of risk associated with foundations bearing in the slag fill rather than in virgin sands or sandy fill (used to replace the slag fill) since it is possible that voids and variations could exist in the slag fill, which may not be detected in site preparation procedures. This higher degree of risk may be reduced somewhat by the procedures outlined in the Site Preparation section. The acceptable risk level for this facility must be determined by Georgetown Steel and considered along with the cost of replacing the slag fill with properly compacted sandy fill or the cost of deep foundations. If the risk associated with placing foundations in the slag fill is deemed unacceptable, then replacement of the slag fill with properly compacted sand fill should be less expensive than deep foundations. — 16 — After receiving this report Carbide officials had several conversations with Stephen Blevins, chief soils engineer at Law, and reiterated the fact that the proposed plant could tolerate only minimal differential settlement due to the movement-sensitive nature of its equipment. In light of Carbide's concerns and to deal with the intolerance of differential settlement, Law clarified its recommendations in a second report on August 29, 1980, and recommended removal of the soil/slag/organic material in place and replacement with properly compacted clean sand backfill. On September 5, 1980, after receiving Law's second report, Rishel contacted Blevins and asked about the possibility of replacing the soil/slag/organic material with pure slag instead of clean sand, a move that would result in substantial savings to GSC ($1.40 per cubic yard). Rishel informed Blevins that the slag contained lime and heavy metals and at Blevins' request gave him the results of PTL's grain-size analysis, but did not tell him of his knowledge that certain types of slag swelled and were inappropriate for use as backfill.3 At this point, Blevins had no knowledge that certain types of slag were expansive. He approved the substitution of slag with confirmation of its gradation.4 — 17 — During this same conversation, Rishel asked Blevins to prepare a proposal for Law to oversee the site preparation project. Law subsequently submitted a proposal which GSC accepted. It provided that Law would furnish engineers to observe the excavation and backfill replacement and would periodically conduct field density tests to insure that the material was being properly compacted. The proposal also stated: — 18 — We understand from Mr. Rishel that the metal slag material, a by-product of the plants [sic] steel production, can be processed locally into a well graded artificial aggregate. With verification of this gradation, we anticipate the processed material will be suitable for use in the backfill required within the Oxygen Plant. — 19 — In October 1980, Fred Sharpe, a Law engineer, visited the plant site and obtained samples both of GSC's pure slag and of the soil/slag mixture which was in place at the site. During this visit, Rishel told Sharpe that GSC had previously used backfill with favorable results. He gave Sharpe a copy of the PTL report. Sharpe reviewed the report within two days of receiving it and then contacted Blevins who told Sharpe to confirm the gradation of the slag. Sharpe told Blevins that PTL had performed a swell test on the slag and although both pondered as to why the test was done, neither knew the reason for the test or asked anyone at GSC about it. So that Law could properly oversee the preparation of the plant site, Rishel also forwarded to Sharpe a copy of the site drawing and the site preparation specifications which required that the backfill material be compacted to a minimum of 95% maximum density. — 20 — During construction of the plant, Law performed inspection and testing at the construction site as per the proposal. That included observation of the excavation, placement and compaction of fill, and testing of the in-place density of the fill. On January 5, 1981, Law issued a report briefly summarizing its activities and giving the test results of two sieve analyses and two compaction tests. After the report was issued, and at Carbide's request, Rishel asked Sharpe to include information about the design bearing pressure. Sharpe reissued the report on January 20, 1981, with the following paragraph: — 21 — Based on monitoring and approval of the excavation to suitable materials, and subsequent field density testing of structural fill, we recommend that foundations bearing in the fill material utilize a total maximum design bearing pressure of 3000 pounds per square foot. — 22 — Construction of the oxygen separation plant was completed in September 1981. Problems developed at the plant shortly thereafter. The problems were first noticed in 1982 when expansive slag caused stress on the electric cables and floor heaving caused tripping hazards. The Carbide plant manager initially thought the problems might be due to the foundations sinking, but he subsequently learned that the floor of the plant was rising. — 23 — In November 1982, Fred Sharpe suggested that Law conduct tests to determine if the slag beneath the plant was expansive and Carbide authorized those tests. Law performed chemical analyses and an autoclave test on the slag and on January 14, 1983, issued a report concluding that the slag was indeed expansive: "[T]he laboratory tests performed indicate a definite potential for expansion, both short term and long term, of the soil/slag mixture which supports the slab in question at the Union Carbide Oxygen Plant." Law's specific findings showed that: — 24 — During the autoclave expansion test of samples molded from the soil/slag mixture and portland cement ..., the samples exhibited severe cracking and crumbling. The cracking and crumbling indicate a high potential for expansion of the soil/slag mixture when exposed to moisture changes. — 25 — The chemical analyses of the soil/slag mixture indicate a significant percentage by weight of ferric oxide, calcium oxide and magnesium oxide. The hydration of calcium oxide has been found to cause short term expansion and the magnesium oxide is thought to cause long term expansion during hydration. Also, the very significant amount of ferric oxide (47.5%) could possibly lead to further expansion due to oxidation, carbonation or crystalline hydration. — 26 — In 1985 Carbide was forced to close the plant because of problems arising from the unstable foundation. The parties do not dispute that the problems at the oxygen separation plant were caused by expansion of the slag beneath the plant. — 27 — GSC subsequently filed suit against PTL, Law, and Carbide for damages resulting from the slag swell. Against PTL and Law, GSC alleged breach of express warranty, breach of implied warranty, and negligence. As to Carbide, GSC sought a declaratory judgment pursuant to 28 U.S.C. § 2201 that GSC had no duty to Carbide for the damage to its air separation plant under the January 23, 1981, contract entered into between the parties. Carbide cross-claimed against Law and PTL and counterclaimed against GSC. Prior to trial, Carbide settled its claims against GSC and assigned GSC its claims against PTL and Law, which claims remained separate from GSC's claims. — 28 — Following a bench trial, the court found that PTL was negligent in approving the use of steel slag as backfill but that GSC could recover only for the cost of replacement of the slag at the track scale because of its own contributory negligence in continuing to use slag as backfill after the problems at the track scale. The court also found PTL liable to GSC for breach of implied warranty, but denied damages finding that it was not foreseeable that the slag would be placed under any particular structure. The court found that there was no express warranty from PTL to GSC. The court denied GSC's negligence and express and implied warranty claims against Law on grounds that the expansiveness of the slag did not fall within the scope of Law's duties under the proposals. Finally, the court denied all of Carbide's claims against Law on the theory that GSC was Carbide's agent, and therefore, Carbide's rights were governed by the same result reached on GSC. GSC and Carbide appeal. II. — 29 — GSC's first contention on appeal is that the district court erred in finding GSC contributorily negligent. GSC also contends that the court erred in finding that the costs of repairing the warehouse extension and the oxygen separation plant were not recoverable from PTL for breach of implied warranty. Finally, GSC argues that the court erred in holding that Law was not liable for negligence or for breach of implied or express warranty. Carbide argues that the court erred in finding that Carbide was not a foreseeable victim of PTL's negligence and therefore could not recover for that negligence. Carbide also argues that the court erred in finding that GSC was Carbide's agent for purposes of dealing with Law and that because GSC was not entitled to recover against Law, neither was Carbide. We address these issues seriatim. — 30 — The district court found that PTL was negligent when it rendered an opinion that the slag it tested for GSC "may be suitable for using as an engineered fill," when by its own admission PTL gave this opinion without conducting a full-scale soil investigation and with knowledge of the expansive qualities of electric are slag.5 The finding of negligence on the part of PTL is not disputed on appeal. What is disputed is the court's finding that GSC could only recover for PTL's negligence with regard to the track scale because of its own contributory negligence with regard to the remaining structures. — 31 — The court based the finding of contributory negligence on the fact that GSC failed to have the slag tested further after the problems with the track scale developed and went on to use slag as backfill around three other structures, even though Rishel was aware of the potential for swell in steel slag and GSC knew its slag swelled, frequently had it chemically analyzed, and had done so since 1977. Under the circumstances, the district court found it "incredulous" that no one at GSC suspected slag swell as the cause when the problems at the track scale developed. The court allowed GSC to recover for damage to the track scale, $3,233, but denied recovery for damage to the other structures. — 32 — In South Carolina, contributory negligence of the plaintiff is a complete bar to recovery on a negligence claim. South Carolina Insurance Co. v. James C. Greene and Co., 290 S.C. 171, 348 S.E.2d 617 (1986). Contributory negligence is defined as — 33 — a lack of ordinary care on the part of a person injured by the negligence of another which combines and concurs with that other's negligence and contributes to the injury as a proximate cause without which injury would not have occurred. — 34 — Id. at 177, 348 S.E.2d at 620. — 35 — The question before us is whether, under this standard, the district court's finding that GSC was contributorily negligent was clearly erroneous. Evidence at trial showed that in 1977 GSC's president, Owen Gochenaur, knew that GSC's slag contained chemical properties that caused it to swell. Although Gochenaur left GSC in 1978, Bill King, the director of purchasing, remained an employee for some time and he also knew about the swelling properties of the slag. Further, although Steven Rishel, the civil engineer in charge of GSC's construction projects, testified that he did not suspect that slag swell was the problem with the various structures until 1982, and therefore took no action to determine whether the slag was expansive, he knew in 1979 that certain slag swelled. Under the circumstances, Rishel should have been on notice that the slag may have been swelling. There was testimony that a reasonably prudent civil engineer with Rishel's background and knowledge should have suspected slag swell as the problem with the track scale, and should have, at that point, had the slag tested again to determine if it was causing the problem at the track scale. On the basis of this evidence and that which indicated that GSC itself had knowledge that its slag had expansive qualities, we cannot say that the district court's finding that GSC was contributorily negligent was clearly erroneous. Instead, it is supported by the evidence and must be affirmed. III. — 36 — GSC next contends that the district court erred when it found that PTL had breached its implied warranty to GSC, but denied damages for the cost of repairing the warehouse and the oxygen separation plant. The court found that PTL was not liable for damages because, in March 1979 when the PTL report was issued, neither party contemplated that the refractory warehouse addition or the oxygen separation plant would be built on the slag. Had PTL known that the slag would be placed under a multi-million dollar structure, the court found that it may have conducted further tests, conducted literature searches, and/or contacted other PTL offices to substantiate its conclusion with regard to the steel slag. — 37 — In reaching its decision, the court relied on Georgetown Towing Co. v. National Supply Co., 204 S.C. 445, 29 S.E.2d 765 (1944). There the plaintiff purchased a marine engine from the defendant under a written contract of sale. The plaintiff alleged that the contract required the defendant to furnish a competent mechanic to supervise the installation of the engine in plaintiff's boat. When the engine broke down, the plaintiff sued alleging breach of implied warranty in failing to properly install the engine. The court found that defendant impliedly warranted the proper installation of the engine and that defendant breached that warranty. In addressing the issue of damages, the court said: — 38 — In a case of this kind, the buyer may recover, in addition to general damages, any special or consequential damages which he may have suffered by reason of breach of warranty, and which are the natural and direct or proximate result of the breach, and may reasonably be considered as within the contemplation of the parties at the time the contract was made. — 39 — 29 S.E.2d at 767 (citations omitted). The court allowed the plaintiff to recover the cost of repairing the engine and damages for loss of business during the time the boat was laid up for repairs, finding that the defendant was aware that the engine was purchased to enable plaintiff to conduct its business. The district court in this case relied upon the language in Georgetown Towing in finding that GSC could not recover for damage to the warehouse addition and the oxygen separation plant because neither structure was within the contemplation of the parties when the implied warranty arose. — 40 — Appellants argue that the cost of repairing the warehouse extension and the oxygen separation plant comes within the category of general damages and as such, their recovery does not depend upon whether the structures were within the contemplation of the parties. We disagree. In Sheek v. Lee, 289 S.C. 327, 345 S.E.2d 496 (1986), the South Carolina court distinguished between general and special damages. "General damages are those which must necessarily result from the wrongful act upon which liability is based," 345 S.E.2d at 497. "Damages for losses that are the natural and proximate, but not the necessary, result of the injury may be recovered only when such special damages are sufficiently stated and claimed." Id. Damages for the cost of repair of a multi-million dollar structure can hardly said to be general damages that necessarily resulted from PTL's negligence. It is undisputed that GSC did not tell PTL anything about the proposed construction of either the warehouse extension or the oxygen separation plant when it engaged PTL's services to conduct the slag tests. The evidence on this point is clear. Therefore, we conclude that the district court did not err in finding that GSC was not entitled to recover for the cost of repair to those structures their construction not being within the contemplation of the parties either when PTL was hired or when it issued its report. The district court's finding on this point is affirmed. IV. — 41 — Next, GSC contends that the district court erred in denying its negligence and breach of warranty claims against Law for damages arising from destruction of the oxygen separation plant. The district court found that Law had no duty under its agreements with GSC to determine whether the slag was expansive and therefore was not negligent in failing to conduct a literature search, failing to test the slag for expansiveness, or failing to inform GSC of its knowledge that slag swelled. In addition, the court found that GSC was contributorily negligent in not telling Law of the problems at the track scale and in telling Law that GSC had used slag successfully and without incident in the past. The court found GSC's contributory negligence a proximate cause of the damage to the plant, for had Law been put on notice that the slag had caused problems in the past or that slag had been placed around the walls of the track scale and problems later developed, it could have investigated the properties of the slag more thoroughly.6 — 42 — After thoroughly reviewing the evidence in this case, we conclude that the district court's finding that Law was not negligent in approving the use of slag as backfill under the oxygen separation plant is clearly erroneous. The court found that it was not part of Law's duties to test for expansiveness of the slag. We recognize that there is no express language in any agreement or proposal in the record that specifically required Law to test for expansiveness of the slag. However, Law was asked to give an expert opinion on the type of foundation necessary to support the oxygen separation plant and in doing so it specifically approved the use of GSC's slag to support the foundations. In its March 24, 1980, report Law set out three alternative recommendations for site preparation. One of those recommendations was to place the plant on the existing slag and sand fill after dynamic compaction. The report stated about the recommendations: "The enclosed guideline recommendations for foundation design represent foundation approaches which we feel would be appropriate." — 43 — Subsequently, in response to a question from GSC, Law approved the use of pure slag as backfill. On September 9, 1980, Law wrote that "[w]ith verification of ... gradation, we anticipate the processed material will be suitable for use in the backfill required within the Oxygen Plant." As a professional geotechnical and materials engineering company, hired for its expertise in the field, Law was clearly negligent in approving the use of GSC's slag as appropriate backfill material, especially in view of Stephen Blevins' admission that he was not qualified by experience or training to deal with slag, that he knew nothing about the expansive properties of slag, and that he had no prior experience with the use of slag. Further evidence of Law's negligence is seen from the testimony of Paul Gooding who, as an employee of Law from 1968 to 1986, testified as Law's corporate representative. Gooding testified that he and probably all other geotechnical engineers employed in Law's Birmingham office knew in the early 1970's that slag had expansive qualities and that it could cause problems if used as backfill material. — 44 — The district court found that it was reasonable for Law to rely upon Rishel's statements that slag had been used in the past as backfill at GSC and not to pursue the expansiveness issue because of these statements. We cannot agree. Law's corporate knowledge of the expansive qualities of slag, coupled with Stephen Blevins' lack of familiarity with the material, should have put Law on notice that before it rendered an opinion that slag was suitable for backfill under a multi-million dollar plant, it should have inquired further. It is undisputed that Law knew exactly what structure was going to be built and that the construction of the foundation of that structure was going to be based on its recommendation. An opinion that the slag was suitable as backfill inherently and necessarily included a finding that the slag would not swell and destroy the plant. Under these circumstances, it is clear that Law was negligent in approving the slag as backfill without testing it for expansiveness. — 45 — The district court also found that GSC was contributorily negligent in failing to inform Law that there had been a problem at the track scale and in informing Law that slag had been used successfully in the past. We do not believe these facts support a finding of contributory negligence. Although making and omitting to make such statements was negligent, GSC's conduct was not contributorily negligent so as to bar recovery because it did not contribute "to the injury as a proximate cause without which injury would not have occurred." South Carolina Insurance Co. v. James C. Greene and Co., 290 S.C. 171, 348 S.E.2d 617 (1986). The proximate cause of the destruction of the plant was Law's opinion that the oxygen separation plant could be built on foundations placed in slag. — 46 — The district court denied GSC's warranty claims against Law for the same reason that it denied its negligence claim, i.e., that Law had no duty to determine whether the slag was expansive. The court found that the only warranty running from Law to GSC was defined by the relevant contract documents. In those documents Law warranted that it would perform its tasks with that degree of skill and care exercised by engineers practicing in the locality. The issue is whether Law breached that warranty when it rendered its opinion that slag would be suitable backfill for the foundation of the oxygen separation plant. Having concluded that Law was negligent because its duties included the determination of the expansiveness of the slag and it failed to make that determination, we must also reverse the district court on this point. Law clearly breached its warranty when it approved the use of slag under the oxygen separation plant without first testing its suitability. V. — 47 — Carbide contends that the district court erred in holding that it could not recover for PTL's negligence. The court denied recovery finding that PTL owed no duty to Carbide because it was not foreseeable that Carbide would rely on PTL's report and suffer damages as a result of that reliance. Carbide argues that it was foreseeable to PTL that some structure would be built on the slag fill; the fact that Carbide built the structure instead of GSC does not obviate PTL's duty to perform without negligence. — 48 — In South Carolina, a cause of action for negligence requires: (1) the existence of a duty on the part of the defendant to protect the plaintiff; (2) the failure of the defendant to discharge the duty; and (3) injury to the plaintiff resulting from the defendant's failure to perform. South Carolina State Ports Authority v. Booz-Allen & Hamilton, Inc., 289 S.C. 373, 346 S.E.2d 324 (1986). The absence of any one of these elements renders the cause of action insufficient. — 49 — Foreseeability of injury, in the absence of a duty to prevent the injury, is an insufficient basis upon which to rest liability. Tolar Construction Company v. GAF Corporation, 154 Ga.App. 127, 267 S.E.2d 635, rev'd on other grounds, 246 Ga. 411, 271 S.E.2d 811 (1980). Foreseeability itself does not give rise to a legal duty.7 A tortfeasor's duty arises from his relationship to the injured party. This relationship may arise out of the tortfeasor's contractual relationship with a third party. Barker v. Sauls, 289 S.C. 121, 345 S.E.2d 244 (1986). This is exactly the argument Carbide makes, i.e., that out of its relationship with GSC, who had a contractual relationship with PTL, arose a relationship between Carbide and PTL. — 50 — In South Carolina State Ports Authority v. Booz-Allen & Hamilton, Inc., 289 S.C. 373, 346 S.E.2d 324 (1986), the South Carolina court recognized that liability for negligence in preparing reports has, in certain circumstances, been extended to parties other than those to the contract and that consultants may be found liable in negligence to non-contracting parties who have reasonably relied on their reports in taking some action. In that case the Georgia-Ports Authority contracted with Booz-Allen & Hamilton, a consulting firm, to prepare a report comparing the merits of the Savannah port with the Charleston port for commercial traffic. The report was highly favorable to the Savannah port and contained false facts and figures concerning the Charleston port. Distribution of the report by the Georgia Ports Authority to domestic and foreign customers and potential customers resulted in decreased traffic in the Charleston port. The South Carolina State Ports Authority, the Pilots Association, and two local chapters of the Longshoremen's Association sued Booz-Allen & Hamilton for negligence. — 51 — The South Carolina court held that Booz-Allen & Hamilton owed a duty to the South Carolina State Ports Authority to exercise due care to accurately report objective factual data concerning the Charleston port, if it knew or should have known the report was intended to be used by Georgia Ports Authority as a marketing device. — 52 — We hold a duty to use due care, running from a consultant to the commercial competitor who is being critiqued, arises when the consultant undertakes to objectively analyze and compare the attributes of commercial competitors for the purpose of giving one a market advantage over the other. — 53 — 345 S.E.2d at 326. — 54 — However, the court found no duty owed by Booz-Allen & Hamilton to individuals who relied upon the shipping traffic in the Charleston port for commercial profit. The court found this relationship far too attenuated to rise to the level of a duty flowing between them. "The concept of duty in tort liability must not be extended beyond reasonable limits." 346 S.E.2d at 326 (citation omitted). — 55 — Application of the principles of Booz-Allen & Hamilton to the facts of this case mandates that Carbide's claim be denied. There is absolutely no evidence that PTL knew or should have known that GSC was going to enter into a contract with Carbide and that under that contract Carbide would build a multi-million dollar oxygen preparation plant on top of slag fill, relying on PTL's report that slag was appropriate backfill. Nor is there evidence that Carbide relied on PTL's report or even that anyone at GSC informed Carbide of the existence of the report. To extend PTL's duty to Carbide under these circumstances would in our opinion be to extend it beyond reasonable limits. We therefore affirm the district court's denial of Carbide's negligence claim against PTL. VI. — 56 — Finally, Carbide contends that the district court erred in denying its claims against Law for negligence and breach of warranty on the theory that an agency relationship existed between GSC and Carbide thus binding Carbide, the principal, to the same result as that reached by the court as to GSC, the agent. We agree. — 57 — Agency is a fiduciary relationship that depends initially on intent. 3 AM.JUR.2d Agency § 17 (1986). "Before an agency relationship can exist, the principal must intend that the agent shall act for him, and the agent must intend to accept the authority and act on it, and the intention of the parties must find expression either in words or conduct between them." Courtney v. Remler, 566 F.Supp. 1225, 1230 (D.S.C.1983). However, under South Carolina law an agency relationship does not require an express agreement to that effect between the parties, but may be implied or inferred from the words and conduct of the parties. Bankers Trust of South Carolina v. Bruce, 323 S.E.2d 523 (S.C.Ct.App.1984). The existence of an agency, nonetheless, must be clearly established. McCall v. Finley, 294 S.C. 1, 362 S.E.2d 26 (1987). The cornerstone of an agency relationship is the power of the principal to control the conduct of his agent. Sharpe v. Bradley Lumber Co., 446 F.2d 152, 153 (4th Cir.1971). This power to control the means and methods employed by the agent, and not just the result, distinguishes an agency relationship from that of independent contractor. Fernander v. Thigpen, 278 S.C. 140, 293 S.E.2d 424 (1982). — 58 — Applying these principles to the facts of the case before us, we conclude that there was no agency relationship between GSC and Carbide and that the court's finding to the contrary is clearly erroneous.8 We do not find from the facts of this case any evidence of intent on the part of GSC to act as Carbide's agent with Law, nor do we find evidence of intent on the part of Carbide that GSC act on its behalf and subject to its control. — 59 — The contract entered into by GSC and Carbide was an arms-length transaction. It obligated each party to undertake and complete certain tasks. One of GSC's duties under the contract was to perform the necessary preparation of the site for construction. To assist it in the site preparation, GSC hired Law. There is no evidence that Carbide gave GSC consent to hire Law, controlled GSC's conduct in any manner, or that GSC intended to act in Carbide's behalf in entering into the agreement with Law. GSC was clearly attempting to fulfill its responsibilities under its agreement with Carbide, not to conduct business on Carbide's behalf.9 — 60 — The district court found an agency relationship from the facts that after receiving Law's first report, Carbide officials contacted Law and reiterated that the plant could tolerate only minimal differential settlement and that after Law's January 1981 report was issued, at Carbide's request, Rishel asked Law to clarify its position on design bearing pressure. — 61 — Neither of these instances evidence an agency relationship. Neither shows that Carbide had the right to control the conduct of GSC that is required for an agency to exist. This evidence merely shows that Carbide was concerned with the end result--that the plant site be properly prepared to support the oxygen separation plant as required by its agreement with GSC. — 62 — Under the circumstances, we are of the view that the court erred in finding an agency relationship and denying Carbide's claims against Law on the same basis that it denied GSC's claims. VII. — 63 — Accordingly, the judgment of the district court is affirmed in part, reversed in part, and remanded for proceedings consistent with this opinion including a determination of damages. — 64 — AFFIRMED IN PART; REVERSED IN PART; and REMANDED WIDENER, Circuit Judge, concurring: — 65 — I concur in the result reached by Judge Hall's opinion. — 66 — Joseph F. ANDERSON, Jr., District Judge, concurring in part and dissenting in part: — 67 — I concur with parts I, II, III, and V of the majority's decision. I am of the opinion, however, that the district judge's findings as to Law's negligence and breach of warranty, GSC's contributory negligence, and GSC's status as an agent of Carbide were not clearly erroneous, because they are "plausible in light of the record viewed in its entirety." Anderson v. City of Bessemer City, North Carolina, 470 U.S. 564, 574 (1985). I therefore respectfully dissent from parts IV and VI of the opinion. I would affirm the decision below in all respects. Footnotes [1] GSC manufactures wire rod by melting scrap metal and iron in an electric arc furnace. A by-product of the manufacturing process is a steel slag. GSC has large piles of slag available and constantly creates more [2] In February 1977, GSC entered into a contract with R.B. Pond Construction Company, Inc. for the sale of GSC's entire output of steel slag to Pond for a five-year period. Prior to entering into the contract, Pond had conversations with Owen Gochenaur, then president of GSC, in which Gochenaur told him that GSC's slag was not used in the production of concrete because it swelled. Doug Lane of IMS, the company hired by GSC to process the steel slag, also informed Pond that the slag produced at GSC swelled. GSC periodically provided Pond with chemical analyses of the slag, none of which revealed a calcium oxide content of less than 40 percent. Bill King, the director of purchasing at GSC, received at least one of the analyses in 1977. It revealed that the slag contained 41.6 percent calcium oxide and 5.3 percent magnesium oxide. It is undisputed that calcium oxide causes rapid swelling and that magnesium oxide causes long-term swelling [3] Blevins testified that at some point Rishel told him that slag had been successfully used as backfill at other plant locations [4] Confirmation of gradation involves a grain-size analysis, the purpose of which is to determine if a material will compact below the waterline. Silts will not compact below the waterline, so Blevins asked for the grain-size analysis from Rishel because he was concerned about the compactibility of the slag [5] In 1979, when PTL wrote the report at issue here, its Jacksonville, Florida office had a library that contained a professional article entitled "Building Damage from Expansive Steel Slag Backfill." In addition, Jack Roseman, manager of PTL's geotechnical department, testified that he knew in 1979 that electric arc furnace slag could swell, and included a statement in all of his geotechnical soil reports warning against the use of this type of slag as backfill [6] The evidence showed that Law, like PTL, had a copy of the article "Building Damage from Expansive Steel Slag Backfill." The article was in Law's Atlanta, Georgia office. The report to GSC was issued out of Law's Birmingham, Alabama office [7] A legal duty is that which the law requires to be done or forborne with respect to a particular individual or the public at large. S.C.E. & G. v. Utilities Construction Company, 244 S.C. 79, 135 S.E.2d 613 (1964) [8] As the district court noted, the determination of the existence of an agency relationship is a question of fact and such is reviewed under the clearly erroneous standard. Gathers v. Harris Teeter Supermarket, Inc., 282 S.C. 220, 317 S.E.2d 748 (1984) [9] Law makes much of the fact that the contract between GSC and Carbide was not entered into until after GSC had already hired Law to perform certain soil tests which Carbide required for proper site preparation. All this indicates is that prior to execution of the written contract, GSC was aware from its discussions with Carbide that it was going to have the task of site preparation for the plant and that GSC hired Law to assist in that preparation. This fact does not establish an agency relationship