The petitioner asks damages for personal injuries he allegedly sustained in a shipboard accident while a crew member aboard the respondent's Great Lakes vessel, the tanker Orion. His complaint alleges respondent's liability both for negligence under the Jones Act, 46 U.S.C. § 688, 46 U.S.C.A. § 688, and for unseaworthiness under the general maritime law;1 a claim for maintenance and cure is also alleged. The parties settled the claim for maintenance and cure at the trial, which was before a jury in the District Court for the Northern District of Ohio. Judgment was entered for the respondent on the unseaworthiness and Jones Act claims upon a verdict directed by the trial judge on the ground of insufficiency of the evidence. The Court of Appeals for the Sixth Circuit affirmed. 271 F.2d 194. We granted certiorari, 362 U.S. 909, 80 S.Ct. 661, 4 L.Ed.2d 618.

Michalic claims that in a shipboard accident on December 28, 1955, a two-and-one-half-pound wrench dropped on his left great toe. Michalic was afflicted with Buerger's disease when he joined the Orion three months earlier as a fireman in the engine room. We are informed by the testimony of one of the medical witnesses that Buerger's disease 'is a disease of unknown origin * * * it produces a narrowing of the blood supply going to the foot through the arteries, and it runs a very foreseeable course; it is slowly progressive in most cases and leads to progressive loss of blood supply to the extremities involving usually the legs'; for one afflicted with the disease to drop 'a hammer on his toe * * * is a very serious thing and frequently leads to amputation. * * * Because the circulation is already impaired and the wound will not heal properly, and any appreciable trauma will frequently lead to gangrene.'

Michalic did not report the accident at the time but continued working until January 6, 1956, a week later, when the vessel was laid up for the winter. Meanwhile he treated the toe every night after work in hot water and Epsom salts. He was at his home from January 6 to March 15 and used hot boric acid soaks 'practically every day.' He was called back to the Orion on March 15. On April 1, 1956, he reported to the Orion's captain that '(m)y leg was so bad, so painful, I couldn't take it no more * * *. I want a hospital ticket.' The captain gave him the ticket after filling out a report in which he stated that Michalic told him that on December 28, 1955, 'While working with pumpman in pumproom man said he dropped a wrench on his foot and his toe has been sore ever since.' This was the first notice respondent had of any accident.

At the hospital in April, a diagnosis was made of 'an infected left great toe nail and gangrene of the left great toe secondary to the Buerger's Disease.' During the spring three amputations were performed on the left leg. first the great left toe, next the left leg below the knee and then part of the leg above the knee. Medical experts, three on behalf of the petitioner and one for the respondent, differed whether, assuming that the wrench dropped on Michalic's left great toe on December 28, there was a causal connection between that trauma and the amputations. This plainly presented a question for the jury's determination. Sentilles v. Inter-Caribbean Corp., 361 U.S. 107, 80 S.Ct. 173, 4 A.L.R.2d 142, and we do not understand that the respondent contends otherwise.

The basic dispute between the parties is as to the sufficiency of the proofs to justify the jury's finding with reason that respondent furnished Michalic with a wrench which was not reasonably fit for its intended use. Here a distinction should be noticed between the unseaworthiness and Jones Act claims in this regard. The vessel's duty to furnish seamen with tools reasonably fit for their intended use is absolute, Mahnich v. Southern S.S. Co., 321 U.S. 96, 64 S.Ct. 455, 88 L.Ed. 561; Seas Shipping Co. v. Sieracki, 328 U.S. 85, 66 S.Ct. 872, 90 L.Ed. 1099; The Osceola, 189 U.S. 158, 23 S.Ct. 483, 47 L.Ed. 760; Cox v. Esso Shipping Co., 5 Cir., 247 F.2d 629; and this duty is completely independent of the owner's duty under the Jones Act to exercise reasonable care. Mitchell v. Trawler Racer, Inc., 362 U.S. 539, 80 S.Ct. 926, 4 L.Ed.2d 941. The differences are stated in Cox v. Esso Shipping Co., supra:

'One is an absolute duty, the other is due care. Where * * * the ultimate issue (is) seaworthiness of the gear * * *. The owner has an absolute duty to furnish reasonably suitable appliances. If he does not, then no amount of due care or prudence excuses him, whether he knew, or could have known, of its deficiency at the outset or after use. In contrast, under the negligence concept, there is only a duty to use due care, i.e., reasonable prudence, to select and keep in order reasonably suitable appliances. Defects which would not have been known to a reasonably prudent person at the outset, or arose after use and which a reasonably prudent person ought not to have discovered would impose no liability.' 247 F.2d, at page 637.

Thus the question under Michalic's unseaworthiness claim is the single one as to the sufficiency of the proofs to raise a jury question whether the wrench furnished Michalic was a reasonably suitable appliance for the task he was assigned. To support the Jones Act claim, however, the evidence must also be sufficient to raise a jury question whether the respondent failed to exercise due care in furnishing a wrench which was not a reasonably suitable appliance.

The wrench dropped on Michalic's foot while he was using it to unscrew nuts from bolts on the casing of a centrifugal pump in the pumproom. He had been assigned this task by the pumpman after the first assistant engineer sent him from the engine room to the pumproom to help ready the pumps for the vessel's winter lay-up. There were about twenty-five 1 5/8 nuts tightly secured to the bolts on the casing. The pumpman gave him a 1 5/8 straight-end wrench weighing two and one-half pounds and ten to eleven inches long, and also a mallet. The pump was located alongside and some inches below a catwalk, and Michalic had to step down from the catwalk to reach the casing. His task required the gripping of each nut in the claw of the wrench and the hammering of the side of the wrench with the mallet to apply pressure to loosen it. Michalic had removed all but a few of the nuts when he 'had hold of a nut' with the wrench and 'I hit it (the wrench) with the mallet and it slipped off the nut and came down the side of the pump and hit my big toe. * * * Yes, she slipped off the nut on the pump and came down the side of the pump and smashed my big toe.'

Michalic contends that the proofs were sufficient to justify the jury in finding with reason that there was play in the claw of the wrench which prevented a tight grip on the nut, thus entitling him to the jury's determination of his unseaworthiness claim, and were also sufficient to justify the jury in finding with reason that the respondent negligently furnished him with a defective wrench, thus entitling him also to the jury's determination of his Jones Act claim. The evidence viewed in a light favorable to him was as follows: The wrench and other pumproom tools were kept in the pumproom toolbox and were used only when the vessel was being prepared for lay-up. The tools were four or five years old. Because of the danger of fire, the tools, including the wrench and mallet which Michalic used, were made of a special spark-proof alloy. The second mate, who had left the Orion on December 19,2 testified that the tools were bronze because 'Bronze tools are for non-striking.' It was the practice to inspect the pumproom tools and replace worn ones before their use at lay-up time, but the first assistant engineer who testified to the practice did not say this inspection was made in 1955; and the pumpman testified that 'It could be' that no one looked at the toolbox for nine months before December 28. The second mate testified that the tools 'had been very beaten and battered, perhaps there for some time.' Michalic testified that he noticed when the pumpman gave him the wrench that it was an 'old beat-up wrench * * * all chewed up on the end.' Michalic said that when he started work 'the wrench was slipping off the nuts; it slipped off every one of them.' He 'had a hard time loosening them off.' He protested to the pumpman that 'This wrench keeps slipping off,' and the pumpman answered 'Never mind about that, do the job as best you can.'

The trial judge found the evidence to be insufficient to present a jury question whether the wrench was a reasonably suitable appliance, because 'on the theory the grip is worn * * * there is never any mention of the grip in the case * * *.' The Court of Appeals took the same view, saying 'There was no evidence that the open or jaw end of the wrench was in any way deficient * * * (t)he fact that the wrench slipped is not evidence that its slipping was the consequence of some condition in the jaw or handle of the wrench.' 271 F.2d at page 199. We think that both lower courts erred. True, there was no direct evidence of play in the jaw of the wrench, as in Jacob v. New York City, 315 U.S. 752, 754, 62 S.Ct. 854, 855, 86 L.Ed. 1166. But direct evidence of a fact is not required. Circumstantial evidence is not only sufficient, but may also be more certain, satisfying and persuasive than direct evidence. Rogers v. Missouri Pacific R. Co., 352 U.S. 500, 508, 77 S.Ct. 443, 449, 1 L.Ed.2d 493, note 17.3 The jury, on this record, with the inferences permissible from the respondent's own testimony that inspections were necessary to replace tools of this special alloy because of wear which impaired their effectiveness, could reasonably have found that the wrench repeatedly slipped from the nuts because the jaw of the wrench did not properly grip them. Plainly the jury, with reason, could infer that the colloquy between Michalic and the pumpman, and Michalic's testimony as to slipping, related to the function of the jaw of the wrench in gripping the nuts and that there was play in it which caused the wrench to slip off. Thus the proofs sufficed to raise questions for the jury's determination of both the unseaworthiness and Jones Act claims. 'It does not matter that, from the evidence, the jury may also with reason, on grounds of probability, attribute the result to other causes * * *.' Rogers v. Missouri Pacific R. Co., supra, 352 U.S. at page 506, 77 S.Ct. at page 448.4

The Jones Act claim is double-barreled. Michalic adds a charge of negligent failure to provide him with a safe place to work to the charge of negligence in furnishing him with a defective wrench. However, the case was not tried, nor is it argued here, on the basis that the charge of negligence in failing to provide a safe place to work rests solely on evidence tending to show a cramped and poorly lighted working space, regardless of the suitability of the wrench. On the contrary, Michalic also makes the allegedly defective wrench the basis of this charge, arguing in effect that the described conditions under which he was required to do the work increased the hazard from the use of the defective wrench. Under that theory, the relevance of the testimony is only to the charge of furnishing a defective wrench and the causal connection between that act and his injury. Phrasing the claim as a failure to provide a safe place to work therefore adds nothing to Michalic's case, and he was not entitled to have that claim submitted to the jury as an additional ground of the respondent's alleged liability.

The judgment of the Court of Appeals is reversed and the cause remanded to the District Court for a new trial. It is so ordered.

Judgment of Court of Appeals reversed and cause remanded to the District Court for new trial.

For the reasons set forth in his opinion in Rogers v. Missouri Pacific R. Co., 352 U.S. 500, 524, 77 S.Ct. 443, 459, 1 L.Ed.2d 493, MR. JUSTICE FRANKFURTER is of the view that the writ of certiorari was improvidently granted.

Mr. Justice HARLAN with whom Mr. Justice WHITTAKER and Mr. Justice STEWART join, dissenting.

At the opening of a Term which finds the Court's docket crowded with more important and difficult litigation than in many years, it is not without irony that we should be witnessing among the first matters to be heard a routine negligence (and unseaworthiness)1 case involving only issues of fact. I continue to believe that such cases, distressing and important as they are for unsuccessful plaintiffs, do not belong in this Court. See dissenting opinions in Rogers v. Missouri Pacific R. Co., 352 U.S. 500, at pages 524, 559, 77 S.Ct. 443, at pages 459, 478.

The District Court, finding that the evidence presented no questions for the jury, directed a verdict for the respondent. The Court of Appeals, in an opinion which manifests a conscientious effort to follow the precepts of the Rogers case, unanimously affirmed, after a painstaking assessment of the record. 6 Cir., 271 F.2d 194. My own examination of the record and of the opinion of the Court of Appeals convinces me that there is no warrant for this Court overriding the views of the two lower courts.

The core of petitioner's case was the condition of the wrench, his unsafe-place-to-work theory having evaporated in thin air, as the Court recognizes. Having had to abandon his original theory that the claw of the wrench had defective teeth (since the wrench was toothless), petitioner testified (1) that the instrument was an 'old beat-up wrench * * * all chewed up on the end' (whether at the claw or handle does not appear); and (2) that the wrench had slipped off nuts at various times during the operation (albeit petitioner had before the accident successfully removed some 15 out of 20 nuts without mishap).

While the Court, in stating that 'there was no direct evidence of play in the jaw of the wrench,' seems to recognize that this testimony did not suffice to show any actionable flaw in the wrench, it nonetheless concludes that the jury should have been permitted to infer one, in light of two other factors. These are (1) the second mate's testimony that as of some 10 days before the accident,2 the tools in the pumproom toolbox 'had been very beaten and battered' (whether at the claw or handle, or anywhere else, does not appear); and (2) other evidence which, as I read its opinion, the Court takes as establishing that the tools were old and infrequently inspected. (Actually the record shows that the tools had been used only four or five times and that the wrench had been inspected just before it was handed to petitioner.3)

Judged by any reasonable standard this evidence, fragmentized or synthesized as one may please, did not in my opinion make a case for the jury. The additional factors on which the Court relies add nothing to the inherent deficiencies of petitioner's testimony which the Court seems to recognize did not of itself make out a case of either negligence or unseaworthiness. If it is permissible for a jury to rationalize 'into being' a defective wrench from this sort of evidence, then wrenches have indeed become dangerous weapons for those operating vessels on the Great Lakes. If the rule of Rogers means that in FELA cases4 trial courts are deprived of all significant control over jury verdicts, and juries are in effect to be allowed to roam at large, I think the lower federal courts should be so told. See Harris v. Pennsylvania R. Co., 361 U.S. 15, 25, 80 S.Ct. 22, 29, 1 L.Ed.2d 1 (dissenting opinion). At least this would be better than continuing to require the lower courts to operate in what must be an atmosphere of increasing bewilderment over what is expected of them in these federal negligence cases.

I would affirm.