This Jones Act case comes to us on appeal from the district court’s grant of summary judgment for the appellee on the issue of the decedent’s seaman status. For the reasons set forth below, we affirm.

The decedent, Mantón Bouvier, was an employee of the appellee, Avondale Shipyards. ^[1] He held various jobs during the course of his employment, but the only work alleged to have made him a Jones Act seaman was his five-year stint as a rigger. The plaintiff, Bouvier’s widow, alleges that he died of asbestosis and silicosis caused by the inhalation of asbestos and sand particles in the shipyard. The district court gave summary judgment for Avondale on the ground that Bouvier had not been a Jones Act seaman. This appeal followed.

While summary judgment on seaman status in Jones Act cases is rarely proper, and even marginal cases should go to the jury, it is in some circumstances possible to rule as a matter of law that a worker is not a seaman. See, e.g., Guidry v. Continental Oil Co., 640 F.2d 523 (5th Cir.), cert. denied, 454 U.S. 818, 102 S.Ct. 96, 70 L.Ed.2d 87 (1981); Burns v. Anchor-Wate Co., 469 F.2d 730 (5th Cir.1972). There are three different formulations in this circuit of the test for seaman status, but they are all essentially equivalent. The original standard, set forth in McKie v. Diamond Marine Co., 204 F.2d 132 (5th Cir.1953), required that the worker have (1) a more or less permanent connection with (2) a ship in navigation and (3) that he be aboard primarily to aid in navigation. The McKie test was refined in Offshore Co. v. Robison, 266 F.2d 769 (5th Cir.1959), which held that seaman status was a jury question where (1) the worker was permanently assigned to or did a substantial portion of his work on a vessel, and (2) his work contributed to the function of the vessel, the accomplishment of its mission, or its operation or welfare. The third version of the test is essentially a combination of the first two prongs of McKie and the second prong of Robison. See Watkins v. Pentzien, Inc., 660 F.2d 604 (5th Cir.1981), cert. denied, 456 U.S. 944, 102 S.Ct. 2010, 72 L.Ed.2d 467 (1982); Guidry v. South Louisiana Contractors, Inc., 614 F.2d 447 (5th Cir.1980). All of these formulations express basically the same idea, and are not to be applied mechanically, but rather used as a guide in weighing the total circumstances of an individual’s employment to determine whether they had sufficient nexus with the navigation of vessels and the perils attendant thereon to implicate the concerns of the Jones Act. See Davis v. Hill Engineering, Inc., 549 F.2d 314 (5th Cir.1977); Brown v. ITT Rayonier, Inc., 497 F.2d 234 (5th Cir.1974).

In this case, we do not think that Bouvier’s work as a rigger had that nexus, for it lacked the aspect of permanency or substantiality. ^[2] Avondale’s riggers were shore-based workers who worked their shift at the shipyard and then went home; they never ate or slept on board a vessel, and[*91] they never went to sea. ^[3] The evidence indicates that they spent approximately fifty percent of their time working on ships. They removed machinery from the ships so that it could be taken ashore for repairs and then reinstalled it when the repairs were done. They also worked in new ship construction, apparently doing similar work; it is unclear what portion of their shipboard work was construction and what portion was repair. ^[4] On any given day, they might work on as many as ten different ships. They would not necessarily complete work on one ship before starting on another, nor did they work on any specific or identifiable group of vessels.

Under these circumstances, there simply was no element of permanency or substantiality in Bouvier’s relationship with the vessels. While a worker can be a seaman with respect to a group of vessels where he would not be with respect to any single vessel in the group, Braniff v. Jackson Ave. — Gretna Ferry, Inc., 280 F.2d 523 (5th Cir.1960), the work “must not be spasmodic and the relationship between the individual and the ... ships must be substantial in point of time and work.” Id. at 528. Even if we assume arguendo that a group of vessels may permissibly be defined as the variously owned vessels putting into Avon-dale for repairs, or even as the group of vessels on which Bouvier worked, nevertheless Bouvier’s work as a shore-bound ship repairman working for a shipyard does not make him a Jones Act seaman. While perhaps none of the facts chronicled above with respect to his work would, taken alone, deny him seaman status as a matter of law, their aggregation does.

We observe also that the Longshoremen’s and Harbor Workers’ Compensation Act specifically covers “any harborworker including a ship repairman [or] shipbuilder ... but ... not ... a master or member of a crew of any vessel .... ” 33 U.S.C. §§ 902(3), 903 (1976). The term “master or member of a crew” in the LHWCA has been held to be the equivalent of “seaman” in the Jones Act, with the Robison test the guide to both determinations. McDermott, Inc. v. Boudreaux, 679 F.2d 452 (5th Cir. 1982). The language of the LHWCA thus strongly supports, indeed arguably demands, the conclusion that a harbor-bound ship repairman is as a matter of law not a “member of a crew” and thus not a Jones Act seaman.

The plaintiff relies heavily on two cases: Abshire v. Seacoast Products, Inc., 668 F.2d 832 (5th Cir.1982), and Landry v. Amoco Production Co., 595 F.2d 1070 (5th Cir.1979). Both are distinguishable. In Abshire, a jury had found that the plaintiff was a seaman. There was conflicting evidence as to the plaintiff’s actual work: the plaintiff’s evidence showed that he worked virtually exclusively on ships, would spend days or weeks continuously on a vessel, and occasionally performed the duties of a seaman in navigation. Nonetheless, it was “a close case” whether the defendant was entitled to judgment notwithstanding the verdict. Abshire, 668 F.2d at 835. Bouvier’s case is much weaker. In Landry, the evidence showed that the plaintiff spent seventy percent of her working time on water, was injured at a site which could only be reached by water, worked continuously side-by-side with the same crew members, and at least occasionally operated boats. This shows considerably more vessel-related activity than Bouvier performed, with a[*92] much higher degree of continuity and a much more substantial relationship between the work performed and the mission of the vessel.

We thus find that neither Abshire nor Landry requires a holding in Bouvier’s favor. His case more closely resembles Fazio v. Lykes Bros. Steamship Co., 567 F.2d 301 (5th Cir.1978) (shoregang employee who repaired lifeboats, rigged and changed booms, handled lines; as a matter of law not a seaman) and Rotolo v. Halliburton Co., 317 F.2d 9 (5th Cir.), cert. denied, 375 U.S. 852, 84 S.Ct. 111, 11 L.Ed.2d 79 (1963) (shore-based welder who repaired hulls of boats; as a matter of law not a seaman).

We find that Bouvier’s relationship to the vessels on which he worked was not sufficiently continuous or substantial to present a jury question of Jones Act seaman status. We therefore AFFIRM the judgment of the district court.