The initial question this case presents is whether a domestic corporation is a resident of all the districts of a multidistrict state of incorporation or of only the district of incorporation. We confront again the riddle of seaman status under the Jones Act,1 and the liability of a shipowner to a Sieracki seaman.2 In addition, we must consider challenges to the district court's findings on admiralty venue, liability, and damages.

The defendant-appellant, Hill Engineering, Inc. ("Hill"), entered a contract with the Columbia Gulf Transmission Company ("Columbia Gulf") in October 1971, which required Hill to fabricate and install metering facilities, including piping, on a gas gathering station located in the Gulf of Mexico approximately ninety miles off the Louisiana coast in an area commonly referred to as Eugene Island, Block 287. The fabrication phase of the project was completed on shore near Houma, Louisiana. Then the prefabricated structures were loaded onto barges and transported to the site of the gathering station where they were installed. The project was started at the beginning of November, 1971, and completed in mid-1972.

The defendant-appellant, Williams-McWilliams Company, Inc. ("Williams"), was the owner of two barges chartered by Hill in connection with the offshore phase of its work. The derrick barge W-701, a special purpose vessel used to transport men, equipment, and material from one location to another, was chartered complete with its own six-man crew and was at all times under the control of Williams, the shipowner. Located in the stern of the barge was a large steam-powered crane, or derrick, used to load or unload equipment and material from W-701 or from other vessels or offshore structures. The W-701 contained a galley and permanent sleeping quarters for its crew. Tugs moved the vessel. Vernon Rickerson, a Williams employee, was superintendent of the W-701 during the time it was chartered to Hill.

Under a "bareboat" charter Hill also obtained the use of the Williams barge W-504, over which Hill maintained control during the term of the project. The vessel was a "material barge" which had no derrick, living quarters, or crew. With it Hill transported material and equipment from Houma, Louisiana to the gathering station in the Gulf of Mexico. The derrick on the W-701 was used to unload the prefabricated material from the W-504 onto the gathering station platform.

The plaintiff, Jerry Wayne Davis was hired by Hill on November 4, 1971, as a welder's helper to assist in the fabrication of the pipe, the loading and unloading of the fabricated structures onto the barges, and its assembly at the platform. The fabrication was done on the bank of the intercoastal canal near Houma. When the onshore fabrication phase of the work was completed, the plaintiff helped to secure the pipe and equipment to the decks of the anchored barges W-504 and W-701 by welding U-bolts into their surfaces. In addition, he did some minor fabrication work on the pipes as defects in earlier welds were discovered through inspection. The barges were loaded on December 3 and 6, 1971. Temporary sleeping quarters were placed on the deck of the W-701 for the six Hill employees who were to install the fabricated structures at the platform. On the morning of December 7, 1971, the W-701 was placed under the tow of the tug Jonas H. for the twenty-four hour trip to Block 287 at Eugene Island. The material barge W-504 was moved into the area later under separate tow. On the trip into the Gulf the plaintiff, another Hill employee, and several crew members helped wash down the front of the barge which had become muddied as a result of the loading operation. In addition, the plaintiff and the journeyman welder he assisted, Tim Holt, continued to secure the equipment and materials to the barge. Further, at the request of the barge superintendent, they welded two cracks in the deck of the barge.

On the morning of December 8, 1971, the W-701 arrived in the Eugene Island area. After a delay in anchoring, the unloading operations were started; but, because of rough seas, only one lift of equipment to the platform was made. During the brief period unloading the plaintiff and Tim Holt remained on board the barge. They cut the equipment loose from the deck and then assisted Williams riggers who attached the load to the four long cables, or chokers, of the derrick. After the load was placed on the platform, it was cut loose. During that phase of the work, Hill employees on the platform requested that a cutting torch be sent up. At Holt's direction Davis got a torch from the bow of the vessel. The torch was then tied onto a tag line and hoisted up to the platform. After they had finished with it the platform workers sent the torch back down to the deck of the barge where the several hundred feet of hose attached to it had to be retrieved. Holt fed the slack to Davis who started walking backwards toward the bow of the barge while rolling up the hose. Just five to ten feet from the gantry Davis slipped and fell at the port stern of the vessel. He landed on his back on the iron deck. There was oil, water, and grease on the deck in the area where Davis fell. After the fall Davis lay for a while on his bunk in the temporary crew quarters on deck. He told the Hill foreman, Roger Meeks, about the incident but did not report it to the superintendent of the barge.

Within an hour or two after the plaintiff's fall, the barge was placed under tow for a trip to shallow water and out of the rough seas. During the twelve-hour trip no additional work was done by the Hill construction workers. On the morning of the next day, December 9, several Hill employees, including Davis and Holt, said that they were terminating their employment over a wage dispute; and the following day, December 10, a crew boat took them to shore. That was the plaintiff's last day on the job. Although the offshore phase of Hill's work for Columbia Gulf on the gathering system was originally scheduled to last only a few weeks, the project was not completed until June 1972.

On September 5, 1972, the plaintiff filed an action in the United States District Court for the Eastern District of Texas invoking the Jones Act against his employer, Hill, to recover for damages resulting from the injury he sustained on December 6, 1971, while working on the W-701 off the coast of Louisiana. Hill is incorporated to do business in the State of Texas and has its principal place of business in Houston, in the southern district, where the complaint was served. On February 26, 1973, Hill filed a motion to dismiss the complaint for lack of admiralty jurisdiction on the ground that venue was improper in the eastern district of Texas where Hill was doing no business.

On March 1, 1973, the plaintiff amended his complaint to add Williams, owner of the barge on which he was injured, as a defendant under the admiralty jurisdiction of the court. The plaintiff sought damages for breach of the warranty of seaworthiness and for maintenance and cure. Williams, incorporated in Delaware with its principal place of business in Metairie, Louisiana, and licensed to do business in Texas, was served with the complaint in Houston through its registered agent.

Williams joined Hill's motion to dismiss for lack of venue on the ground that it too was doing no business within the eastern district of Texas. In support of its motion Williams filed the affidavit of its president, O. M. Gautreaux. After this Court refused the defendants' request to mandamus the district court to rule on the venue issue before discovery in the case, Gautreaux, in his deposition of January 22, 1974, admitted that Williams was engaged in business in the eastern district of Texas.

On January 30, 1974, the plaintiff filed a second amended complaint against Hill and Williams which was served on Joseph Champagene, deck captain of the dredge Port Arthur, located in the Neches River near Beaumont in the eastern district of Texas and owned by Williams. The dredge had no connection with the events relating to the plaintiff's injury.

The defendants filed no additional motions with the court; and on May 22, 1974, the trial court filed its memorandum opinion and order rejecting the defendants' claims of lack of venue. On May 27, 1974, the defendants filed a motion for entry of an order under 28 U.S.C. § 1292(b) (1970) allowing an immediate appeal on the venue issue to advance the ultimate termination of the litigation. On May 31, 1974, the motion was denied.

After a two-day trial without a jury held on July 29 and 30, 1974, the district court found for the plaintiff and against both defendants, awarding damages of $20,261 for loss of past earnings, $628,991 for loss of future earnings, $33,000 for past and future pain and suffering, and $8 a day for 964 days for maintenance and cure.III.

The Jones Act,3 46 U.S.C. § 688, establishes special venue requirements. It provides, in jurisdictional terms, that venue is proper in the district in which the defendant employer resides or in which his principal office is located. In Pure Oil Co. v. Suarez, 1966, 384 U.S. 202, 86 S. Ct. 1394, 16 L. Ed. 2d 474, the Supreme Court held that the definition of corporate residence established in 28 U.S.C. § 1391(c) (1970)4 of the general venue statute is applicable to the Jones Act. Thus, venue is expanded under the Act to include districts in which the defendant is incorporated, licensed to do business, or doing business.

At issue in appellant Hill's claim that the district court erred in finding venue in the eastern district of Texas is the meaning of the phrase, "any judicial district in which it (may be) incorporated", in section 1391(c). Hill contends that venue lies only in a district in which it is doing business or where its principal place of business or designated agent for service of process is located. At the time of the suit, the defendant Hill was not conducting business in the eastern district of Texas; it was served with process in the southern district at its principal office in Houston. Therefore, Hill claims venue was improper in the eastern district. The issue whether a corporation may be sued under section 1391(c), and consequently the Jones Act, in any district of the state of its incorporation has not been authoritatively determined by the Supreme Court or by a circuit court since the 1948 recodification of the Judicial Code. The district courts which have considered the matter are in disagreement. Under one view, venue in a suit against a corporate defendant is proper in any district of the state of incorporation, regardless of whether the defendant is doing business in the district.5 The opposing view is that, except where the defendant is doing business, venue can be laid only in the district of incorporation, that is, the district in which the principal office is located, the one in which the agent for service of process resides, or the district designated as the place of incorporation in the state charter.6 A parallel split in authority exists among district courts on the question of the proper district or districts for venue in a state in which a defendant corporation is licensed to do business.7

There are three tools to assist us in deciphering the meaning of the term "residence" in section 1391(c) as it relates to the phrase "in any judicial district in which it is incorporated": the law prior to 1948, the effect of the 1948 codification represented by section 1391(c), and the analysis of courts that have considered the question since 1948.

The first predecessor of section 1391(c) was the Act of March 3, 1875, Ch. 137, § 1, 18 Stat. 470, later amended by the Act of March 3, 1887, Ch. 373, § 1, as corrected by the Act of August 13, 1888, Ch. 866.8 In Shaw v. Quincy Mining Co., 1892, 145 U.S. 444, 12 S. Ct. 935, 36 L. Ed. 768, the Supreme Court concluded that the district of a person's residence is equivalent to the "district whereof he is an inhabitant" and that the phrase "district of residence" "restricts the jurisdiction to the district in which one of the parties resides within the State of which he is a citizen". 145 U.S. at 449, 12 S.Ct. at 937. As to corporations, the Court's reasoning was similar:

In the case of a corporation, the reasons are, to say the least, quite as strong for holding that it can sue and be sued only in the State and district in which it has been incorporated, or in the State of which the other party is a citizen.

Id. The principle that for venue purposes corporations are considered residents only of the district of incorporation was reiterated and emphasized by the Court in Galveston, Harrisburg and San Antonio Railway Co. v. Gonzales, 1894, 151 U.S. 496, 14 S. Ct. 401, 38 L. Ed. 248:

(A) domestic corporation is both a citizen and an inhabitant of the state in which it is incorporated; but in none of (these cases) is there any intimation that, where a state is divided into two districts, a corporation shall be treated as an inhabitant of every district of such state, or of every district in which it does business, or, indeed, of any district other than that in which it has its headquarters . . ..

151 U.S. at 503-504, 14 S. Ct. at 404. See also In Re Keasbey and Mattison Co., 1895, 160 U.S. 221, 16 S. Ct. 273, 40 L. Ed. 402.

The immediate predecessor to Section 1391(c) was section 51 of the Judicial Code enacted in 1940.9 In Suttle v. Reich Brothers Construction Co., 1948, 333 U.S. 163, 68 S. Ct. 587, 92 L. Ed. 614, the Supreme Court confirmed that interpretations of corporate venue under the earlier statutes applied equally to the later enactment.10 Thus, on the eve of the 1948 recodification, the rule of venue, that a defendant corporation was a resident only of the district of its incorporation, was settled. Trial courts treated venue as a factual question and determined the district of residence by examining the certificate of incorporation for the designated official residence of the corporation, Gorman v. A. B. Leach & Co., E.D.N.Y.1926, 11 F.2d 454; or by determining the location of its general or principal office, Bacon v. Federal Reserve Bank, E.D.Wash.1923, 289 F. 513; Edgewater Realty Co. v. Tennessee Coal, Iron & Railroad Co., D.Md.1943, 49 F. Supp. 807. See also 1 Moore's Federal Practice, P 0.142(5.1-1-3) at 1400 and n. 9 (1975).

The principal question for resolution in interpreting section 1391(c) is the effect of the 1948 codification. Legislative documents concerning the enactment of the 1948 Code by Congress give the impression that only a few minor changes were made in the venue provisions with the adoption of section 1391(c). See Legislative History of Title 28, United States Code "Judiciary and Judicial Procedure" 1969 (1971). The Reviser's Note published in connection with House deliberations on the Code and later in the annotations to section 1391 does not mention the question of venue for a defendant domestic corporation in a multidistrict state.11 See H.R.Rep. No. 308, 80th Cong., 1st Sess. 6, A127 (1947). As a result of this dearth of comment, some courts have concluded that there was no congressional intent to change the results in the Shaw and Suttle cases regardless of the way in which the new statutory provision is phrased. See, e. g., Jacobson v. Indianapolis Power & Light Co., N.D.Ind.1958, 163 F. Supp. 218. Cf. Ruth v. Eagle-Picher Co., 10 Cir. 1955, 225 F.2d 572.

Several commentators have reached a contrary conclusion.12 In Moore's Judicial Code, published just after the 1948 recodification was enacted, Professor James William Moore discussed the change in the meaning of the term "residence" for corporate venue purposes:

Prior to the Code a corporation was a resident of the state in which it was incorporated; and, more specifically, where the state of incorporation was divided into two or more districts, then in the district where its official residence was designated by the charter or state law and, in the absence of such a designation, then in the district where the corporation maintained its home office. The Code has made no change in this rule relative to the residence of a corporate plaintiff. For venue purposes it has, however, radically altered the rule relative to a corporate defendant.

Moore's Judicial Code 193-194 (1949). Professor Moore continues by quoting verbatim the language of section 1391(c) thereby emphasizing that venue for a corporate defendant is proper "in any judicial district" (emphasis supplied) in which it is incorporated.

Judge Leibell in Cleverley v. Nelson, 122 N.Y.Law J. No. 17, 1 (July 26, 1949), reached the same conclusion and held that because the defendant was incorporated in New York and therefore authorized to do business in any district in the state, it was a resident for venue purpose in every district of the state. The court noted that only the year before, under the old Judicial Code, it had reached the opposite result in Eastman Kodak v. Boyce Motor Lines, S.D.N.Y.1948, 74 F. Supp. 981.

In recent years, as noted above, the district courts have split on the question of how section 1391(c) should be interpreted. According to some district courts, the statute is unambiguous on its face and by its terms allows a corporation to be sued in any district in the state in which it is incorporated.13 Another group has concluded that the language is ambiguous and reached the opposite conclusion. They reason that if Congress had intended to change the meaning of "residence" with the 1948 recodification, it would have used the phrase "such judicial district or districts" in the second clause of the provision instead of the two singular usages in the phrases, "such judicial district" and "the residence of such corporation".14 (Emphasis supplied.) Thus, these courts conclude, although the phrase "any judicial district" would ordinarily be read as plural, the uses of the singular form in the next clause must lead to the interpretation that a single district of incorporation was intended. While the language of section 1391(c) is not free from ambiguity, we do not agree with those courts that have concluded that a domestic corporation is a resident of only one district of a multidistrict state by virtue of its incorporation there.

In adopting the Reviser's proposed language for section 1391(c) Congress must have been aware that a state certificate of incorporation recognizes the corporation's right to conduct business throughout the state, not just in one district. Bertha Building Corp. v. National Theatres Corp., E.D.N.Y.1952, 103 F. Supp. 712; C. Wright, Law of Federal Courts 154 (2d ed. 1970). In addition, the phrase "any judicial district" in the first clause of section 1391(c) applies to all three types of corporate activities within a state for venue purposes: incorporation, license to do business, and doing business. A corporation may obviously do business in more than one district of a state thereby subjecting itself to suit in more than one district, and the language of the statute makes no distinction among the three types of corporate activities. Thus, the parallel phraseology suggests that Congress expected that more than one district of a state would be appropriate for venue purposes whether the state was one in which the corporate defendant was licensed to do business, doing business, or incorporated. Further, appellant Hill's emphasis on the fact that it engaged in no business in the Eastern District of Texas is irrelevant because the statute makes any one of the three activities a sufficient basis for venue.

Some district courts have found only one district of the state of incorporation appropriate for venue purposes on policy grounds. In Energy Resources Group, Inc. v. Energy Resources Corp., S.D.Tex.1969, 297 F. Supp. 232, 234, the court discussed two purposes of the venue statutes:

to lay venue in a place having a logical connection with the parties to the litigation, and to afford the defendant some protection against the hardship of having to litigate in some distant place.

Other courts have also expressed concern over forcing a defendant to answer suit hundreds of miles from where corporate activities are conducted. See, e. g., Westerman v. Grow, S.D.N.Y.1961, 198 F. Supp. 307. The difficulty with this approach as a way of determining whether residence is in all districts of the state of incorporation or only in one is that often a corporation is incorporated in a state with no connection to its activities. In that situation the district of corporate residence derived from the address given in the articles of incorporation does not have any greater connection with the corporation, nor is it necessarily any less distant from corporate activities, than any other district in the state. Thus, distance from corporate activities is not a relevant basis for determining proper venue under section 1391(c). As Congress was aware, if a forum is inconvenient for a corporate defendant because of distance, a motion for transfer of venue to a district with greater connection to the parties is always available under section 1404(a), 28 U.S.C. § 1404(a) (1976).15

We therefore concur with those district courts16 and commentators17 that have concluded that venue is proper under section 1391(c), and consequently the Jones Act, in every district of the state in which a defendant corporation is incorporated. The district court, therefore, did not err in concluding that venue was proper for Hill Engineering, Inc. in the eastern district of Texas.

As to Hill's claim that the district court was in error in not, alternatively, granting a change of venue under section 1404(a), we do not agree. Although in all of its motions to the court, Hill asked in the alternative that venue be changed to either the southern district of Texas or the eastern district of Louisiana, Hill failed to specify in the motions its reasons for this request. In its brief on appeal Hill also did not specify in what ways the district court's decision not to grant the change of venue was in error. Therefore, because this Court cannot determine from the record in what ways the district court may have erred in failing to grant a change of venue, there is no basis on which we can consider the appeal.

VENUE IN ADMIRALTY AND WAIVER OF OBJECTION TO SERVICE OF PROCESS

Appellant Williams also contends that the district court erred in finding venue proper in the eastern district of Texas. The Williams contention as to venue focuses on whether service of the complaint was properly made on a managing agent within the meaning of Rule 4, Fed.R.Civ.P. We do not reach this issue because we find that appellant Williams waived objection to venue on that ground.

Venue in an in personam admiralty action has long been established as proper "wherever a monition can be served upon the libellee, or an attachment made of any personal property or credits of his . . . ." Ex Parte Louisville Underwriters, 1890, 134 U.S. 488, 490, 10 S. Ct. 587, 588, 33 L. Ed. 991. After one false start in which the plaintiff served appellant Williams' registered agent for service of process in the southern district of Texas, he served the second amended complaint in the eastern district upon the deck captain of the dredge Port Arthur located in the Neches River. There is no dispute about the Port Arthur's lack of connection with the facts relating to the plaintiff-appellee's injury; it had none, nor does either party contend that commencement of an action in rem was attempted. Here the plaintiff sought to initiate an action in personam based on personal service of process upon a managing agent of Williams.

The chronology of procedural events is undisputed. On March 6, 1973, the plaintiff first served its complaint on Williams' agent for service of process, C. T. Corporation Systems, in Houston. On June 12, 1973, the defendant Williams filed a motion to dismiss on the grounds that venue was improper in the eastern district of Texas for three reasons: the vessel connected with the injury had not been seized, Williams' principal place of business was not located in the district, and the defendant did not conduct a substantial amount of business there. On January 30, 1974, the plaintiff served its second amended complaint on Joseph Champagene, the deck captain of the dredge Port Arthur, as managing agent for the defendant Williams. Williams filed no additional motions, and on May 22, 1974, the district judge denied both defendants' motions to dismiss. As to Williams, the court found service of process on the deck captain of the dredge Port Arthur proper and accordingly sustained venue in the eastern district.

In response, the defendants moved the court for entry of an order under 28 U.S.C. § 1292(b) (1970) to facilitate the filing of an interlocutory appeal. The defendant Williams sought to appeal the court's ruling on venue this time on the grounds that it was doing minimal business in the eastern district, that the dredge Port Arthur had no connection with the plaintiff's injury, and that the vessel involved in activities relating to the injury had not been seized. The court denied the motion on May 31, 1974.

Under Rule 12(h)(1),18 if a party fails to object to the insufficiency of service of process, he waives any future objection on that ground. Although Williams never moved to quash service of process on the Port Arthur's deck captain nor to object to venue on the grounds of improper service, Williams argues on appeal that there was no waiver of the objection. First, Williams contends that the court's ruling that service of process was proper made any motion to quash service of process a nugatory act. We reject that argument. Service on the deck captain was made on January 30, 1974; and the court did not file its order denying the motion to dismiss until May 22. During the intervening four months Williams had more than sufficient opportunity to file a motion to quash service or to amend the grounds stated in its first motion to dismiss for lack of venue. In addition, even after the court ruled that service was improper, in moving for an interlocutory appeal the defendant did not suggest that the court erred in finding service on the deck captain proper. Williams cannot now successfully accuse the district court of error in a ruling made without benefit of contrary argument when the defendant had more than sufficient time to bring any objections to the court's attention.

Williams presents a second ground for contending there was no waiver of objection to service of process. The defendant argues that because venue in an in personam admiralty action is predicated on proper service of process on the defendant, an objection to venue implicitly includes an objection to service of process. This argument against waiver also fails. Rule 7(b)(1)19 requires that a motion state with particularity the grounds on which it is based. Even if we assume that an objection to venue in an in personam admiralty action includes an objection to service of process, under Rule 7 the stated grounds for the motion to dismiss for lack of venue would have to include insufficient service of process. Here Williams failed to suggest such a ground to the district court. Without the particularity requirement district courts would be subject to reversal for rulings on which they did not have the benefit of contrary argument, and opposing parties would suffer rulings against their interests without notice of opposing contentions. Hence, we apply the requirement here to reject Williams' claim that there was no waiver.

The defendant's citation of two cases, Martin v. Lain Oil & Gas Co., E.D.Ill.1941, 36 F. Supp. 252, and Martens v. Winder, 9 Cir. 1965, 341 F.2d 197, cert. denied, 1965, 382 U.S. 937, 86 S. Ct. 391, 15 L. Ed. 2d 349, does not deter us from finding a waiver of the objection to service of process. In Martin the defendant added an additional ground to its motion to dismiss four days after the original motion was filed and prior to the date briefs on the motion were due. The court found there was no waiver of the additional ground. As to Martens, the Ninth Circuit held that in each consolidated motion filed by the defendant prior to the district court's ruling, he reestablished his contention of lack of personal jurisdiction and therefore did not waive his objection on that ground. Here the defendant did not move to quash service of process in the four months between service and the district court's order nor did he offer improper service as a ground for lack of venue. Therefore, we hold that the defendant Williams waived its objection to service of process on the deck captain of the dredge Port Arthur and uphold the district court's finding that venue was proper as to Williams.

As with the defendant Hill, we also reject Williams's claim that the trial court erred in not granting a change of venue under section 1404(a). Williams never stated to the court below its reasons for contending that venue in the eastern district of Texas was inconvenient and in fact has not made any arguments in that connection here. The bare assertion of error is insufficient.

The Jones Act gives "any seaman who shall suffer a personal injury in the course of his employment" the right to seek damages for the negligence of his employer.20 The Act does not define the term "seaman". In Swanson v. Marra Brothers, Inc., 1946, 328 U.S. 1, 66 S. Ct. 869, 90 L. Ed. 1045, the Supreme Court held that its benefits are restricted to "members of a crew of a vessel". More recently, the traditional meaning of "crew member" under the Jones Act has been expanded to include individuals not formally assigned to the crew of a vessel.21 At the very core of the term "seaman" is the requirement that a claimant have more than a transitory connection with a vessel in order to recover under the Jones Act. See, e.g., Keener v. Transworld Drilling Co., 5 Cir.1972, 468 F.2d 729, 732.

In Offshore Co. v. Robison, 5 Cir.1959, 266 F.2d 769, 779, we established the parameters of seaman status. "There is an evidentiary basis for a Jones Act case to go to the jury:

(1) If there is evidence that the injured workman was assigned permanently to a vessel (including special purpose structures not usually employed as a means of transport by water but designed to float on water) or performed a substantial part of his work on the vessel;

and

(2) if the capacity in which he was employed or the duties which he performed contributed to the function of the vessel or to the accomplishment of its mission, or to the operation or welfare of the vessel in terms of its maintenance during its movement or during anchorage for its future trips."

266 F.2d 779. Seaman status under the Jones Act is a mixed question of law and fact, Holland v. Allied Structural Steel Co.,5 Cir.1976, 539 F.2d 476. We conclude that the district court properly applied the law to the facts of this case. Our examination of the record within the framework of the Robison standards leads us to the conclusion that the plaintiff was "assigned permanently" to the vessel on which he was injured and that the duties which he performed contributed to the accomplishment of the vessel's mission.

The first criterion of the Robison test for seaman status treats the issue of the injured worker's connection with a vessel; it offers alternative grounds for meeting the standard. The plaintiff meets the first alternative, permanent assignment to the vessel.22 In reaching this conclusion we stress that the word "permanent" has never been given a literal interpretation under the Jones Act. As with other elements of the Robison test, the permanency criterion is "more frequently an analytical starting point than a self-executing formula"; the test should not be given a "wooden application". Brown v. ITT Rayonier, Inc., 5 Cir.1974, 497 F.2d 234. An examination of the facts of this case in relation to the reasoning of this Court in other cases involving the permanency requirement explains our conclusion that the plaintiff was "permanently assigned" to the barge on which he was injured.

At the completion of the onshore phase of Hill's pipe fabrication and installation project for Columbia Gulf, the entire Hill crew, including the plaintiff, boarded the W-701 for the trip to the gathering station at Eugene Island. The deck of the barge had been fitted with temporary quarters for the Hill crew; and they ate and slept on the vessel. Tim Holt, the journeyman welder with whom Davis worked, testified that the Hill employees boarded the barge with their clothes prepared for a "long stay". They expected to stay twenty or thirty days until the installation job at the station was completed. Holt stated that it was his understanding that they would be working in conjunction with the derrick barge because of the necessity of using a derrick, or crane, for or unloading barges and for rigging the pipes onto the platform.

Although the derrick had its own crew, Hill employees, including Holt and Davis, assisted in the loading and unloading of the W-701 and the material barge. Holt also testified that the Hill employees were instructed to assist the barge crew in any other way they could. On the trip to the platform, at the superintendent's request, he and Davis welded a davit, or bit, onto the side of the barge and a crack in the surface of the vessel's side. Also Davis assisted Williams employees in washing the deck of the barge. Vernon Rickerson, the W-701's superintendent, testified that another vessel was scheduled to provide the Hill crew with living quarters and rigging assistance after the W-701 had unloaded the equipment and pipes at Eugene Island, but in fact the W-701 remained at the site until the pipe installation project was completed in June of 1972.

The defendant Hill contends that Davis was a "mere passenger" aboard the W-701; yet, the facts lead to the conclusion that the plaintiff and other Hill employees not only ate and slept on the W-701 but that they also worked there. Thus, Davis and his co-workers underwent many of the same hazards of the sea to which members of the W-701's crew were subjected. Kimble v. Noble Drilling Coop., 5 Cir.1969, 416 F.2d 847, cert. denied, 397 U.S. 918, 90 S. Ct. 924, 25 L. Ed. 2d 99; Offshore Co. v. Robison, 266 F.2d at 780; Garrett v. Gutzeit, 4 Cir.1974, 491 F.2d 228.

In contrast, the cases in which this Court has found the permanency requirement a fatal stumbling block have largely involved land-based workers providing shore services to docked vessels. In Thibodeaux v. J. Ray McDermott & Co., 5 Cir.1960, 276 F.2d 42, the Court reasoned that the decedent had a transitory relation to the barge on which he was injured where he was a regular shore worker who ate, slept, and lived at home, was not assigned to any particular vessel, and had worked only four days loading and securing deck cargo on unmanned barges. Similarly, this Court sustained a directed verdict for the defendant on the seaman issue in Burns v. Anchor-Wate Co., 5 Cir.1972, 469 F.2d 730, because the plaintiff, a talleyman at a pipe-coating yard, who never worked more than two or three hours aboard any particular barge and never slept or ate aboard the barges or towing vessels, was not permanently assigned to the barge on which he was injured. See also Thomas v. Peterson Marine Service, Inc., 5 Cir.1969, 411 F.2d 592, cert. denied, 1970, 396 U.S. 1006, 90 S. Ct. 562, 24 L. Ed. 2d 499; Labit v. Carey Salt Co., 5 Cir.1970, 421 F.2d 1333; Cox v. Otis Engineering Corp., 5 Cir.1973, 474 F.2d 613; Dugas v. Pelican Construction Co., 5 Cir.1973, 481 F.2d 773 cert. denied sub nom.; Union Oil Co. of California v. Dugas, 414 U.S. 1093, 94 S. Ct. 723, 38 L. Ed. 2d 550; Ross v. Mobil Oil Corp., 5 Cir.1973, 474 F.2d 989.

Here the plaintiff assisted in welding cracks in the W-701 at the barge superintendent's request; he worked with Williams employees in washing the deck; and he helped load and unload the barge along side Williams crew members operating the derrick. Had he remained in Hill's employ, Davis would also have cooperated with Williams derrick operators in the installation of fabricated pipe at the gathering station. Thus Davis was "permanently assigned" to the W-701 within the meaning of the Robison test.23

The second prong of the Robison test is conjunctive with the first; consequently, the plaintiff must also show either that his duties contributed to the function of the vessel or the accomplishment of its mission or that his work aided the operation or welfare of the vessel. All parties to this suit stipulated as to the functions of the derrick barge W-701. It is a special purpose vessel used to transport men, equipment, and material from one location to another and to load and unload equipment and material from itself, other vessels, and offshore structures. The barge is also used to lift heavy components into place in the course of constructing offshore facilities. There was substantial evidence that all of these functions made up the W-701's mission in connection with Hill's fabrication project. Davis's work included assisting in the barge's loading and unloading operations both onshore and at the offshore structure and in helping to rig the gathering station with fabricated pipes. Thus the plaintiff's duties directly contributed to the functioning of the special purpose W-701 and to the accomplishment of its mission on the Hill project. The testimony of defense witnesses Vernon Rickerson and John Hill substantiates this conclusion.

Owens v. Diamond Drilling Co., 5 Cir.1973, 487 F.2d 74, cited by the defendant Hill, is inapposite because the plaintiff's duties aboard a stationary drilling platform had no relation to the function of the tender, where he lived, in servicing the structure. Rather, a case directly supportive of our conclusion here is Noble Drilling Corp. v. Smith, 5 Cir.1969, 412 F.2d 952, cert. denied, 396 U.S. 906, 90 S. Ct. 221, 24 L. Ed. 2d 182, in which the plaintiff was a mud pumper assigned to operate a mud pump permanently affixed to the deck of the tender. The Court concluded that the plaintiff's duties directly contributed to the fulfillment of one of the primary functions of the tender. See also Kimble v. Noble Drilling Corp., 416 F.2d at 848-849.

The plaintiff has shown that he meets both parts of the Robison test; therefore, the district court's finding that the plaintiff was a seaman under the Jones Act was not in error.VI.

The determination that Davis had seaman status under the Jones Act during his employment with Hill provides the jurisdictional basis for an award of damages to the plaintiff for injuries sustained as a result of his employer's negligence. The Jones Act was adopted as remedial legislation designed to give a seaman the same right to recovery against his employer as the Federal Employers' Liability Act ("FELA") gives to railroad employees.46 U.S.C. § 688, (1970). See Ferguson v. Moore-McCormack Lines, Inc., 1957,352 U.S. 521, 523, 77 S. Ct. 457, 458, 1 L. Ed. 2d 511, 513. Consequently, evidence of "the slightest negligence" is sufficient to sustain a finding of Jones Act liability. Sanford Bros. Boats, Inc. v. Vidrine, 5 Cir.1969, 412 F.2d 958; Perry v. Morgan Guaranty Trust Co. of New York, 5 Cir.1976, 528 F.2d 1378. The district court found the defendant Hill negligent in its failure to: use reasonable care to provide the plaintiff with a safe place in which to work; inspect the vessel for hazards; and take reasonable precautions to protect the plaintiff and others similarly situated from defects.

The district court properly supported its determination of Hill's negligence with reference analogously to Nivens v. St. Louis Southwestern Railway Co., 5 Cir.1970, 425 F.2d 114, cert. denied, 400 U.S. 879, 91 S. Ct. 121, 27 L. Ed. 2d 116 in which a railroad employee sued his employer under the FELA. In discussing the employer's obligations the Nivens Court concluded:

This duty includes a duty to inspect the third party's property for hazards and to take precautions to protect the employee from possible defects. Although an employee may be on another's property for only a short time, and although the employer may have no supervision over the property, there is no de minimus rule. The "brevity of the sojourn" is irrelevant with respect to the duty to provide a safe place to work . . . . The employer may protect itself by simply refusing to permit its employees from going on the property. . . .

425 F.2d at 118-119. See also Vickers v. Tumey, 5 Cir.1961, 290 F.2d 426. That reasoning applies equally well to the facts of this case. The record contains no evidence that Hill carried out its responsibility to inspect the vessel or to take precautions in order to provide its employees with a safe working place, and there is substantial credible evidence that an unsafe condition existed on the deck of the W-701.

Both the plaintiff and Tim Holt testified to the existence of an accumulation of oil, grease, and water on the metal deck. They stated that the mixture caused the deck to be slippery; and, as a result, the plaintiff fell. They suggested that the substance appeared to come from the crane or derrick located approximately five or ten feet from where the plaintiff fell. The derrick was steam powered and its exhaust contained a mixture of condensed steam or water and oil. In addition, Tim Holt testified that the steel deck was not covered by a so-called "riprap boardwalk" as is often the case on such vessels; rather, Davis indicated that because the deck was old and worn, it was "wavy".

From this evidence the district court could have reasonably concluded that Hill failed in its duty to provide its employees a safe place in which to work and in its obligation to inspect the work site.

Under the admiralty doctrine of warranty of seaworthiness the owner of a vessel traditionally is absolutely liable for any injury sustained by a crew member in the course of his employment. The State of Maryland, 4 Cir.1936, 85 F.2d 944. In Seas Shipping Co. v. Sieracki, 1946, 328 U.S. 85, 66 S. Ct. 872, 90 L. Ed. 1099, the Supreme Court expanded the protection of the warranty, giving longshoremen working in the service of the ship a seaman's remedy of damages for personal injury proximately caused by the vessel's unseaworthiness. The "ship's service test" has been applied to include other categories of workers within the protected class of seamen. See, e. g., Pope & Talbot, Inc. v. Hawn, 1953, 346 U.S. 406, 74 S. Ct. 202, 98 L. Ed. 143. It is possible for a claimant to be a Sieracki seaman and not meet the requirements of seaman status under the Jones Act. Burns v. Anchor-Wate Co., 5 Cir.1972, 469 F.2d at 733; Brown v. ITT Rayonier, Inc., 5 Cir.1974, 497 F.2d at 239; Dugas v. Pelican Construction Co., 5 Cir.1973, 481 F.2d at 778. The reverse is not true. Where the Robison requirements are met the claimant a fortiori conforms to the "ship's service" criterion. McWilliams does not contest the district court's finding that Davis is a seaman within the meaning of Sieracki and therefore one to whom the benefits of the warranty of seaworthiness applies. Williams however challenges the court's finding of liability under that doctrine.

The Supreme Court has made clear that the owner's duty to furnish a seaworthy ship is absolute. Mitchell v. Trawler Racer, Inc., 1960, 362 U.S. 539, 80 S. Ct. 926, 4 L. Ed. 2d 941. Whether the owner had notice of an unseaworthy condition, is not an issue in such a claim. Rice v. Atlantic Gulf & Pacific Co., 2 Cir.1973, 484 F.2d 1318; Webb v. Dresser Industries, 5 Cir.1976, 536 F.2d 603. Williams' contention, therefore, that there is no evidence the vessel's superintendent or other personnel knew of the existence of the substance on the deck is irrelevant to unseaworthiness liability.

As discussed above in Section VI, the record contains ample evidence of the accumulation of oil, grease, and water on the "wavy" metal deck of the barge from which the district court could conclude that a slippery condition rendered the deck unreasonably safe for its intended use. Rice v. Atlantic Gulf & Pacific Co., 484 F.2d at 1321. Nevertheless, both Hill and Williams argue that there was insufficient credible evidence of the unsafe condition. They contend that Holt's testimony was suspect because of his relationship to Davis in that his uncle was married to the plaintiff's mother. They also challenge the court's conclusions on liability because of evidence that the deck of the barge was cleaned the day before the accident and that a pump tank was attached to the base of the derrick to catch the exhaust materials of oil and water. From our review of the record however we cannot find fault with the district court's decision to give greater weight to the testimony of Holt and Davis than to the circumstantial evidence adduced by the defendants. The court's findings of fact as to liability are not clearly erroneous.

The defendants contest the court's finding that a fall by the plaintiff on the deck of the barge was the cause of his back condition.

The plaintiff's treating physician, Dr. George W. Tate, testified that Davis was suffering from a congenital back ailment aggravated by degenerative arthritis caused by the fall on the deck of the barge. The defendants contend that other facts do not support Dr. Tate's conclusion. They point out that the plaintiff did not seek medical attention until April 7, 1972, five months after the accident and that Dr. Tate did not diagnose the arthritis until November 29, 1973, twenty-three months after the fall. In addition, on cross-examination Dr. Tate testified that he normally would have expected signs of a trauma to appear in the diagnostic x-rays within four months after the fall; but there were no such signs evident in the April 7, 1972, examination. The x-rays revealed only a muscle spasm. None of these facts support the defendants' claim that Dr. Tate's testimony must be rejected. That signs of a back trauma from such a fall usually will show up within four months, does not refute the conclusion that here the fall caused the plaintiff's arthritis; Dr. Tate testified that sometimes signs of a trauma do not appear until almost six months after an accident. The physician's tardiness in diagnosing the arthritis does not counter his testimony on the cause of the malady.

The burden on the plaintiff to prove proximate cause in actions based on general maritime law and the Jones Act is very light, even "featherweight". Landry v. Two R. Drilling Co., 5 Cir.1975, 511 F.2d 138. The plaintiff has offered credible evidence that he fell as a result of oil and grease on the deck of the vessel and that he sustained an injury from the fall. The treating physician has offered a largely unchallenged opinion that the fall aggravated an existing congenital defect causing degenerative arthritis, an incapacitating ailment. The defense presented no counter medical testimony and insufficient challenges to Dr. Tate's opinion as to the cause of the plaintiff's back condition for us to conclude that the district court's finding of proximate cause must be rejected. Because the court's determinations are not clearly erroneous, we affirm its determination that both of the defendants are liable in damages to the plaintiff for injuries sustained as a result of Davis's fall on the deck of the W-701.

Finally, the doctrine of comparative negligence under which a plaintiff's damage award is reduced in proportion to the amount of his contributory negligence is inapplicable here because there is no evidence that the type of shoes worn by the plaintiff was a contributory cause of his fall. See Stein v. Sea-Land Services, Inc., 5 Cir.1971, 440 F.2d 1181.

The defendants contend that the district court erred in finding Williams alternatively liable on a theory of negligence and in imputing that negligence to Hill. Because we have concluded that the district court was not clearly erroneous in finding Hill negligent under the Jones Act, we do not need to decide whether Hill can also be found liable on the alternative ground of imputed negligence. In addition, we do not reach the question of negligence on the part of Williams because we have affirmed the district court's finding of unseaworthiness liability.

As a seaman with more than incidental attachment to the vessel on which he was injured, see G. Gilmore and C. Black, The Law of Admiralty §§ 6-7 at 282-283, the plaintiff has a right to recover for maintenance and cure at the amount stipulated by the parties, eight dollars a day for 964 days. We affirm the district court's judgment in this regard.

The amount of damages sustained by an injured person is a question of fact, and we will not tamper with the district court's determination unless it is clearly erroneous. Neal v. Saga Shipping Co., 5 Cir.1969, 407 F.2d 481, cert. denied, 1969, 395 U.S. 986, 89 S. Ct. 2143, 23 L. Ed. 2d 775. The defense challenges the damage award on a variety of grounds; it contends that the court's errors stem from reliance on the testimony of the plaintiff's expert, Dr. Sigmund Horvitz. Neither defendant presented the testimony of an opposing expert and Dr. Horvitz's testimony was not seriously challenged by the defendants on cross-examination. "The weight to be accorded unimpeached expert opinion evidence is solely for the judge sitting without a jury." Eason v. Weaver, 5 Cir.1973, 484 F.2d 459. Thus, the district court as the trier of fact is free to rely on an unimpeached expert's opinion unless such reliance would result in a clearly erroneous finding.

The district court made an alternative finding on damages that incorporated a 5.4 percent rate per annum for wage and price inflation. In Freeport Sulphur Co. v. S/S Hermosa, 5 Cir.1976, 526 F.2d 300, 309-310 (Wisdom, J., concurring specially) the author of this opinion advocated that the Court adopt the Alaska Supreme Court's approach in Beaulieu v. Elliott, 1967, 434 P.2d 665, 671, with respect to including an inflation element in damage calculations. On the theory that inflation offsets the interest that could be earned on "safe" investments, under Beaulieu the loss of future earnings component of damages is not discounted to present value.24 This approach yields the claimant an income over the remaining years of his life more accurately reflective of his loss of future earnings than does a discounted figure because it prevents inflation from eroding the damage award. Nevertheless, as the plaintiff concedes, this Court's decisions en banc in Johnson v. Penrod Drilling Co. and Starnes v. Penrod Drilling Co., 5 Cir.1975, 510 F.2d 234, cert. denied, 423 U.S. 839, 96 S. Ct. 68-69, 46 L. Ed. 2d 58, establish that the incorporation of projected future inflation as an element of damages is at present "too speculative a matter for judicial determination". 510 F.2d at 241. Until Penrod is overruled, therefore, an inflation element cannot be included in damage computations, either in the form of calculating loss of future earnings without discounting to present value or of increasing the basic damage award, prior to discounting, with a figure representing the projected inflation rate. The former is the Beaulieu method, the latter the district court's method in this case. Thus, the court's award of $628,991 for the total loss of the plaintiff's future earning capacity must be reduced by the amount attributed to expected inflation. The court set the award at $425,321 absent the inflation element.

Even without the inflation factor, the defense contends that the award for loss of future earning capacity is much too high, given the plaintiff's admissions that he worked only six months a year and applied for unemployment compensation every winter and because of his income record for the preceding five years. Davis testified that he had been working as a welder's helper in the pipelining industry since 1965. His unchallenged testimony established that fair weather is crucial both to pipeline work in general and to the requirements for a proper weld and that consequently pipelining in the Texas and Louisiana area where he worked was largely done in the spring and summer. While on the job, however, a pipeliner ordinarily works seven days a week and from seventy to eighty hours during the week. The plaintiff's tax returns establish his gross earnings and business expenses, including travel, room, and board costs, for the years 1967 through 1971 at the following amounts:

Year  Gross Income  Business Expenses
-------------------------------------
1967   $7,516.14        $2,597.90
1968    5,862.35         2,951.10
1969    5,665.91         1,284.50
1970    4,294.50         1,850.90
1971    4,641.99         2,528.00

The defendant Williams contends that the 21.2 percent figure established by Dr. Horvitz as the percentage of the plaintiff's income spent on costs of employment is too high on the basis of this data. As the plaintiff points out, the defendant has confused the business expense deduction permitted under the personal income tax laws with the "cost of obtaining employment", a factor subtracted from past or future lost income calculations in damage computations because the incapacitated worker no longer pays such costs. Although meal and laundry expenses are permissible components of the business expense deduction, they are constant expenditures incurred by an individual whether or not he is working; consequently, they are not included in the cost of employment for damage calculations. The plaintiff's expenses attributable to travel and lodging, without board, for the five years total $5,958.28, a sum which equals 21.2 percent of Davis' total income of $27,980.00 for the five-year period. The Court's acceptance of this figure for the cost of employment is therefore supported by the record.

The defendant Williams contends that on the basis of the plaintiff's earnings record the base year salary of $8,545 derived by Dr. Horvitz for computing loss of future earning capacity was completely incorrect. The court accepted the base year figure in reliance on Dr. Horvitz's testimony on the relevant assumptions and parameters for computing past and future earnings losses. Dr. Horvitz concluded that the plaintiff worked 1656 hours per year on the average for the five-year period from 1967 to 1971. He made this calculation from the information on the plaintiff's tax returns showing periods of employment and annual earnings during each year. With data from the business agent of Davis' local-union on the hourly wage rates in effect during that period for welder's helpers, Dr. Horvitz calculated the plaintiff's average yearly hours at 1656 with 1091 hours of regular time and 565 hours of overtime. Dr. Horvitz explained that the total number of yearly hours thus derived equalled approximately seventy-five percent of the average number worked in the industry at that time. Thus he concluded that the figures accurately represented the plaintiff's individual work habits and accounted for his absences from work because of injury and illness.25

For the year beginning July 29, 1975, Dr. Horvitz calculated the base year figure of $8,54526 from which future earnings computations were computed, using the union contract wage rate for a welder's helper during the twelve months following trial. The defendant contends that a projected contract wage rate is too speculative a basis for future earnings calculations. We disagree. Cf. Noack v. American Steamship Co., 6 Cir.1974, 491 F.2d 937. The hourly rates used by Dr. Horvitz were established by contract prior to trial. The fact that the plaintiff had not worked more than infrequently as a pipeliner during the three and a half years following his injury and thus had not earned those particular contract rates does not make them any more speculative or less appropriate. The loss of future earning capacity element of damages must be projected from the date of trial and the union contract rates in effect at that time are a reliable source of data on the prevailing wage rate. The use of contract rates in effect at the time of trial does not introduce an inflation element because plaintiff's future earnings capacity is "frozen" for the purposes of these calculations at the base year rate.

The defendants contest the use of union wage rates for deriving the base year figure and for computing loss of past earnings capacity from December 1971 to the date of trial in mid-1975 on another ground. They contend that union rates are inapplicable because the plaintiff worked on union and non-union jobs during the period. In fact, the Hill job on which Davis received his injury was not governed by the union contract. A comparison of the wage rate paid to the plaintiff by the defendant Hill at the end of 1971 (Appendix 522) with union rates for that period (Appendix 540) indicates that the hourly rate Hill paid was within the union contract range. In addition, neither defendant presented evidence during the trial challenging the applicability of union wage scales to the plaintiff's employment activities. Therefore, on the basis of the evidence presented, the trial judge's acceptance of the union rates was not clearly erroneous.

On cross-examination Dr. Horvitz admitted that in calculating the plaintiff's past and future loss of earnings he assumed the plaintiff did not work prior to trial and would not work in the future. As to the future, the evidence of Dr. Tate, the plaintiff's medical expert, established that the arthritis in Davis' lower back was worsening and that Davis was incapable of performing the heavy manual labor for which he was trained by the time of trial. The plaintiff left school in the middle of the seventh grade and started pipelining work when he was 19 or 20 years old. During the years following the injury Davis attempted to return to pipelining and tried housepainting and salvage work, but each time because of the pain generated by the condition of his back the plaintiff was forced to stop working after only a couple of days or to work at an unusually slow pace. From this evidence the trial court could reasonably conclude that the plaintiff was not suited by educational background, experience, or training for work on a non-manual basis and that he could not engage in the work for which he was trained because of his injury. Thus, calculations of lost future earnings on the basis of the plaintiff's complete incapacity for work after trial was proper.

In connection with the plaintiff's loss of wages between the date of injury and the trial, the defense contends that the district court erred in its computations by neglecting to decrease the $20,261 figure suggested by Dr. Horvitz by the amount actually earned by the plaintiff during the period. As we have already noted, Dr. Horvitz testified that past loss of earnings calculations were based on the assumption that Davis did not work after his injury. The plaintiff stated at trial that he tried a variety of jobs after his injury including three or four attempts at pipelining, a two or three-month stint junking cars at ten to twenty dollars a day, and the painting of several houses. From the plaintiff's testimony, the trial court could have concluded that the amount of money the plaintiff earned during the short periods he worked was de minimus. We observed in Neal v. Saga Shipping Co., S.A., 407 F.2d at 489:

Damages in personal injury cases cannot be computed by mathematical formulae nor derived from fixed principles susceptible of being programmed into a computer. Another trier of fact might properly have used a method slightly different from the one the trial court did. He might well have fixed the damages at more or less than the amount that served as a basis for the award here. Neither judge nor jury is expected to be infallible in such matters. The standards by which damages are determined leave room for the play of judgment and discretion of the type properly exercised by the trial court. Here it cannot be said that the result he reached was clearly erroneous. (Footnote omitted.)

In connection with another of its claims of error the defense argues that the trial judge erroneously concluded that the plaintiff in all probability would have qualified to work as a journeyman welder and that loss of future earnings calculations and the base year rate, should be augmented to account for the increased wage rate. Tim Holt and the plaintiff testified that under the union bylaws a welder's helper who works in the apprentice program for five years is eligible for journeyman status. Holt, the journeyman welder whom the plaintiff assisted, stated that Davis would have qualified easily for a journeyman "book". The plaintiff also testified that he met the union's requirements for a journeyman. According to the evidence, the plaintiff worked as a welder's helper in the pipelining industry from 1965 until his injury in December 1971. The defendants contend that these dates prove that the plaintiff could not qualify as a journeyman welder because he had already worked five and a half years as a welder's helper at the time of his injury. The plaintiff counters that he did not work in pipelining for a number of months during 1966 as a result of head injuries. The defendants presented no evidence to challenge the testimony that Davis was qualified to be a journeyman welder nor to suggest that the plaintiff had actually worked more than a total of five years in the apprentice program. The Sixth Circuit established the principles relevant to this issue in United States Steel Corp. v. Lamp, 6 Cir.1970, 436 F.2d 1256, 1270, cert. denied, 402 U.S. 987, 91 S. Ct. 1649, 29 L. Ed. 2d 153:

The claimant must first establish his normal annual earning capacity, which, in the absence of evidence of special circumstances indicating an ability to rise beyond his prior level of employment, would consist of a projection of claimant's earnings history, taking into account all available data relevant to wage adjustment.

With evidence in the record that the plaintiff had the ability to rise to journeyman status and no evidence challenging that conclusion, the trial court's finding that the plaintiff was qualified to work as a journeyman welder was not clearly erroneous.

In urging this Court to affirm the district court's damage award absent the inflation factor, the plaintiff argues that the district court reduced its prediction of the plaintiff's loss of future gross earnings to present value by using a discount rate of six percent per annum, a rate about equal to that paid on short term government securities, according to Dr. Horvitz. We have no objection to the discount rate suggested by the plaintiff's expert. Our quarrel is with the plaintiff's contention that the trial judge applied that rate. The record reveals that the court's findings as to damages were made in complete reliance on the testimony of the plaintiff's expert. Examination of his explanation of the calculations for "Case 2", from which the court's damage award for future earnings was taken, reveals that Dr. Horvitz used a discount rate of only 2.9 percent in computing the figure accepted by the court. (Appendix 209-211, 213.) He explained that he selected that rate in an attempt to reflect expected increases in productivity. (Appendix 213.) Such estimates are as speculative for courts at this time as are projections of the future rate of inflation. See Johnson v. Penrod Drilling Co., 510 F.2d at 241. We remand the case to the district court in order that the damage award for loss of future earnings can be recalculated using a discount rate of six percent.

AFFIRMED in part; REVERSED in part; and REMANDED for proceedings not inconsistent with this opinion.

INGRAHAM, Circuit Judge, concurring specially.

I concur in Judge Wisdom's opinion despite initial reservations about venue as to Hill. Energy Resources Group, Inc. v. Energy Resources Corp., 297 F. Supp. 232 (S.D.Tex.1969), an opinion authored by me, illustrates the problems encountered in large multi-district states. The State of Texas, whose statute of incorporation is the basis for venue in this suit, has recognized these difficulties and has enacted a complex set of rules governing venue. However, I agree that § 1404(a) may be utilized to alleviate such problems in the federal courts.