WISDOM, Circuit Judge:

Douglas Steamship Company, the defendant-appellant, appeals a district court judgment on a jury verdict that awarded a total of $150,000 damages to Irby J. Robertson, a longshoreman who was injured while working on one of Douglas' vessels. The jury awarded the longshoreman $138,000 in basic damages and $12,000 as compensation for losses due to inflation or the decreased purchasing power of money. We affirm the judgment of the district court with respect to the basic award of $138,000 and we reverse with respect to the $12,000 inflation compensation award.

I.

In November 1970, Irby J. Robertson, a longshoreman employed by Atlantic and Gulf Stevedores, Inc., was injured in two accidents that occurred while he was working in the Port of New Orleans aboard the M/V Inchoma, a vessel owned by the Douglas Steamship Company. The first accident occurred on November 8 when a cargo batten pulled loose while Robertson's gang attempted to stretch a safety net across the hatch where they were working. The cargo batten struck Robertson in the head and left him stunned for several minutes. Dr. Samuel Romano, a physician retained by the stevedore company, examined Robertson and treated him for contusion of the scalp.

On November 10, the day Robertson returned to work, he was injured again. While he was helping to build a landing platform, a large number of sacks, each filled with 100 pounds of chemical fertilizer, fell on him. The first sack that fell knocked him down into a crouched position so that his knee jammed into his stomach and his body doubled into a knot; the other sacks piled on top of him. Robertson again reported to Dr. Romano. He complained of a contusion above his left eye as well as various pains in his neck, ribs, knee, and ankle.

From November 10, 1970, until January 11, 1971, Robertson remained under Dr. Romano's care on an outpatient basis. On January 11, Dr. Romano discharged Robertson and certified him fit to resume work. Thereafter Robertson made numerous visits to Dr. Romano and complained of various pains related to his injuries. Dr. Romano referred Robertson to a neurosurgeon and to an orthopedic consultant for examination. Neither specialist was able to discover the objective cause of Robertson's pain. Accordingly, Robertson was discharged and received no further treatment from Dr. Romano after June 23, 1971.

In October 1971, the longshoremen were preparing to strike against their employers. Robertson anticipated that he would run short of cash. On his own initiative and without advice of counsel, Robertson settled with Douglas and executed a general release for $2,000. About two months later, Robertson began to suffer epigastric distress in the upper abdomen. Robertson entered Flint-Goodridge Hospital on January 19, 1972 where he was treated for an alleged ulcer-type condition. He was discharged on February 11, but his condition later deteriorated and he was readmitted to the hospital on April 14. Exploratory surgery revealed a fibrous attachment between the transverse colon and the right lobe of the liver. Because the surgeon was interested in the upper rather than the lower abdomen, he did not examine the peritoneal area. He testified however, that Robertson's continuing downhill medical progression led him to believe that the patient was suffering from some condition that he was unable to diagnose.

On August 11, 1972, Robertson entered Touro Infirmary. There his condition was diagnosed as a very rare disease, idiopathic retroperitoneal fibrosis, a scarring of the peritoneal lining in the lower abdomen. The surgeons operated to free the ureter from the fibrous tissue of the peritoneal lining. Robertson's condition further deteriorated and he underwent a second operation, this time to have an osphyotomy tube implanted in the right kidney to prevent drainage. Robertson has undergone extensive hospitalization in connection with this condition and will probably require future hospitalization. His medical expenses to the time of trial amounted to $20,000.

Robertson brought this suit in federal district court and it was tried to a jury. As noted, the jury found his damages amounted to $138,000, but awarded him an additional $12,000 to compensate for losses attributable to inflation. Douglas moved for a judgment notwithstanding the verdict or, in the alternative, for a new trial. The Company appeals the district court's denial of these post-trial motions.

On this appeal, Douglas Steamship urges three grounds for reversal: first, it contends that the evidence at trial is insufficient to support a finding that Robertson's condition was caused by either accident aboard the M/V Inchoma; second, that the jury instructions concerning the concept of mutual mistake and the release were incorrect; third, that the instructions concerning the effect of inflation and the decrease in purchasing power of money were incorrect.

II.

At trial, Judge Rubin required the jury to present its verdict in the form of answers to interrogatories. One of these interrogatories asked whether "the M/V Inchoma [was] unseaworthy on November 8 or November 10 or on both dates in a manner that proximately caused Mr. Robinson's (sic) injuries." The jury answered affirmatively. This interrogatory directed the jury's attention to two questions of fact: the unseaworthiness of the vessel and the causal relationship between the vessel's unseaworthiness and the injuries allegedly sustained by the plaintiff. Douglas contends that the evidence adduced at trial is insufficient to support the jury's finding that Robertson's condition, eventually diagnosed as idiopathic retroperitoneal fibrosis, was attributable to either or both of these accidents.

The thrust of Douglas' argument is that the plaintiff's expert medical testimony was not sufficiently certain to support the jury's conclusion. In assessing the defendant's burden in attacking the verdict, we note first that the trial judge concluded that the evidence was sufficient to allow the case to go to the jury. Moreover, he also concluded that the evidence was sufficient to require the denial of Douglas' motions for judgment notwithstanding the verdict and for a new trial.

The trial court has the initial responsibility and authority to correct erroneous jury verdicts. "When the ends of justice require it, a federal trial judge has the power and the duty to set aside a jury's verdict, to grant a new trial, or to grant a judgment notwithstanding the verdict." Stewart v. Gilmore, 5 Cir. 1963, 323 F.2d 389. But the trial judge cannot and here did not substitute his view of the facts for the jury's finding. This Court sitting en banc, in a careful and extensively researched opinion by Judge Ainsworth, established the test to be applied in determining the sufficiency of the evidence to submit a case to the jury when there has been a motion for a judgment notwithstanding the verdict. The Court declared:

On motions for directed verdict and for judgment notwithstanding the verdict the Court should consider all of the evidence — not just that evidence which supports the non-mover's case — but in the light and with all reasonable inferences most favorable to the party opposed to the motion. If the facts and inferences point so strongly and overwhelmingly in favor of one party that the Court believes that reasonable men could not arrive at a contrary verdict, granting of the motions is proper. On the other hand, if there is substantial evidence opposed to the motions, that is, evidence of such quality and weight that reasonable and fair-minded men in the exercise of impartial judgment might reach different conclusions, the motions should be denied, and the case submitted to the jury. . . . However, it is the function of the jury as the traditional finder of the facts, and not the Court, to weigh conflicting evidence and inferences, and determine the credibility of witnesses.

Boeing Company v. Shipman, 5 Cir. 1969, 411 F.2d 365, 374-75.

In the case at bar, the relevant evidence consisted of sophisticated medical testimony concerning an extremely rare disease, the causes of which have not been conclusively established by medical science. Since the causes of this rare disease are not entirely known, Robertson's expert witnesses were unable to state with certainty that the traumatic injuries sustained in the November 1970 accidents caused the plaintiff's present condition. Robertson's experts, however, offered their professional opinion that some causal connection between the accidents and the plaintiff's present condition was likely.

Dr. Marks of Louisiana State University Medical School, testified that the present state of medical knowledge prevented his saying with certainty that the accident definitely caused the disease. He noted, however, that one possible cause of retroperitoneal fibrosis is an abnormal release of material from fatty tissue. Dr. Marks testified that such a release can be caused by a simple tortional strain, such as that commonly suffered from a heavy weight falling on one's back; a direct blow to the abdomen is not a necessary element in the causation of such a release. Dr. Marks said that "when a specific, reasonably severe injury occurs . . . some degree of disruption of tissues was possible, and within a reasonable period of time this type of condition developed, I think it would be very difficult not to relate the two in a circumstantial manner." Finally, he stated that, in his opinion, there was a "logical association" between the accident and the disease, meaning that they were in fact "causally related in some manner."

Dr. Marks was not the only expert who testified as to causation. Another expert, Dr. Schlegel, testified that "trauma in a patient like Mr. Robinson (sic) may be an explanation for his retroperitoneal fibrosis, because he has such an unusual reaction to trauma in his connective tissue."

No expert testimony supports the view that Robertson's accident could not, theoretically, have caused the disease. The only substantial basis for doubt, apart from that inherent in the present limits of medical knowledge, stems from the fact that the disease was not diagnosed at an earlier stage when Robertson was initially treated. Although the lack of an early diagnosis has common sense credibility as proof that the disease did not exist at the time of diagnosis, and thus could not have been caused by an even earlier accident, the medical experts who testified at trial were unanimous in stating their belief that retroperitoneal fibrosis simply cannot be diagnosed at an early stage.

Although Dr. Schlegel now believes that the condition was present at the time of Robertson's operation at Flint-Goodridge, he testified that there was no reason for the surgeons to suspect the proper diagnosis at that time, because there was no obstruction of the ureters, which is the key to diagnosis. He said that, in the absence of such an obstruction, retroperitoneal fibrosis would be one of the last things a diagnostician would consider with respect to a patient's "vague abdominal symptoms". All of the experts concurred in this view. Moreover, despite Robertson's inarticulate expression of his symptoms, it must be conceded that he complained of the same epigastric distress throughout his treatment. This fact suggests that, although diagnosis was not feasible, the same condition existed and deteriorated during this period.

There is always some degree of uncertainty when a litigant seeks to establish causation by expert testimony. One well accepted authority states:

Where opinion evidence is relied on to establish the causal link a question of sufficiency may also arise in deciding whether the opinion has been expressed in terms definite enough to support the needed conclusion. If a qualified witness gives it as his opinion that the causal relation did exist or probably did exist, there is not much question about the legal sufficiency of the testimony. But cautious or hostile witnesses are often unwilling to go so far, and there are serious questions and variant rulings about the sufficiency of such expressions as "may have caused", "possibly caused," and the like.

2 F. Harper F. James, The Law of Torts § 20.2, p. 1117-1118 (1956).

Douglas would have us rule that an expert must state categorically that a given result came from a particular cause. But there are situations in which medical experts cannot testify dogmatically as to cause and effect. Justice may also be served when expert witnesses testify candidly, cautiously, and avoiding dogmatic conclusions when the causal relationship between an accident and an injury is a question of fact for the jury to decide.

Our close review of the record in the light of Boeing v. Shipman convinces us that the trial judge correctly denied the defendant's post-trial motions.

III.

Douglas' second assignment of error attacks the trial court's instructions to the jury concerning the validity of the release and the issue of mutual mistake, on the ground that the trial court's jury instructions should have conformed to Louisiana law, not to general maritime law. Douglas contends that Louisiana law is applicable because the release had "no exclusive maritime purpose" and "was intended only to put to rest the controversy between Robertson and Douglas." The flaw in Douglas' argument rests in its failure to acknowledge that the controversy between Robertson and Douglas is based on rights secured to Robertson as a longshoreman protected by maritime law. It is well-established that questions related to substantive rights created in admiralty, such as the validity of a seaman's release or the effect of mutual mistake, must be determined in accordance with general maritime principles rather than state law. See Garrett v. Moore-McCormack Company, 1942, 317 U.S. 239, 63 S.Ct. 246, 87 L.Ed. 239; Pope Talbot, Inc. v. Hawn, 1953, 346 U.S. 406, 74 S.Ct. 202, 98 L.Ed. 143. Cf. Stevens, Erie R.R. v. Tompkins And The Uniform General Maritime Law, 64 Harv.L.Rev. 246 (1950). As the Supreme Court said of the longshoreman injured in Pope Talbot, supra, "he was injured on navigable waters while working on a ship to enable it to complete its loading for safer transportation of its cargo by water. Consequently, the basis of [his] action is a maritime tort, a type of action which the Constitution has placed under national power to control in `its substantive as well as procedural features'". 346 U.S. 409, 74 S.Ct. 205.

Gulfwide contends that state law should be applicable because a longshoreman's contacts with the sea are less substantial than those of a seaman, and less significant than the longshoreman's contacts with his onshore community. This argument is wide of the mark. For purposes of determining the applicability of general maritime law, a longshoreman injured in the course of his duties stands in the same position as a seaman with respect to "the traditional remedies of the sea."

IV.

The trial court instructed the jury that the release must be set aside if it was executed as a result of mutual mistake concerning a material fact. Douglas contends that the instruction is erroneous, even if general maintenance law is applicable, because it did not accurately state the general maritime law concerning mutual mistake. We hold that the instructions were sufficiently accurate to inform the jury correctly on the relevant law and to form a basis for its verdict.

A mistake with regard to diagnosis has long been recognized as cause for setting aside a seaman's release. "A release induced by the diagnosis given by the employer's doctor should not prevent the award of additional maintenance when the diagnosis is shown to be erroneous, even though it may have been quite honestly made." Bonici v. Standard Oil Co., 2 Cir. 1939, 103 F.2d 437, 439. See also Wooten v. Skibs A/S Samuel Bakke, 4 Cir. 1970, 431 F.2d 821.

Douglas attempts to distinguish the case at bar on the ground that a physician other than Dr. Romano treated Robertson and thus Robertson cannot be said to have relied entirely on Dr. Romano's diagnosis when he signed the release. As authority for its position, Douglas relies on language from another Second Circuit case: "[w]hen a seaman has made a settlement after full investigation and with independent advice, we can see no ground for holding it invalid." Sitchon v. American Export Lines, Inc., 2 Cir. 1940, 113 F.2d 830, 832. In Sitchon, Judge Augustus Hand noted that the plaintiff had in fact been examined by two disinterested physicians relative to his injury and had sought legal advice before settlement. We find the reasoning of Sitchon inapplicable here because the record shows that Robertson's independently retained physician treated him exclusively for hypertension, a condition in no way related to his accident. There is no evidence in the record that Robertson consulted his personal physician regarding his injuries or that he relied on that physician's diagnosis in making a settlement; the mere fact that Robertson could have obtained such an opinion if he desired it is irrelevant in the absence of proof that he did obtain it or rely on it.

Douglas points out that a panel of this Court has recently refused to set aside a longshoreman's release under similar facts in Strange v. Gulf Southern S.S. Co., 5 Cir. 1974, 495 F.2d 1235. We think that Strange must be distinguished. In that case, the Court said:

The record facts and the applicable authorities do not sustain the challenge based on mutual mistake. There would be little security in the settlement of a personal injury claim if the binding effect of such a settlement depended upon the certainty of the extent and the outcome of the injuries involved. It is the very consequences of these uncertainties which the parties seek to foreclose by settlement and to take their chances on their outcome.

495 F.2d 1235, 1237. While it is true that a release should not be set aside for mutual mistake concerning the extent and outcome of injuries, which are necessarily future rather than present facts, it does not follow that a release should not be set aside for mutual mistake concerning the nature of the injuries, which is a present fact. The legal distinction must rest on the medical difference between diagnosis and prognosis.

A longshoreman who signs a release may have to take his chances that a properly diagnosed condition was the subject of an overly optimistic prognosis and that his injuries may be more serious and extensive than originally thought. However, the law does not require him to take his chances when the diagnosis is itself erroneous and he is suffering from a disease entirely different in nature from that diagnosed. In Robertson's case, there is evidence in the record that the accident caused the injury now identified as idiopathic retroperitoneal fibrosis. Because the diagnosis did not uncover this disease, Robertson signed the release with the understanding that the nature of his injury consisted of superficial injuries to his neck, back, and limbs. The injury that was later discovered demonstrated a mistake of fact that went not merely to the extent and outcome of his injury, but indicated that he was the victim of an injury entirely different in nature from that diagnosed. This difference supports a finding of mutual mistake of a material fact.

The trial court instructed the jury that "if the parties were not mistaken at the time — if they both acted on the basis of Mr. Robinson's (sic) condition at the time, there was no mistake, even though his condition later changed." We think that the instructions taken as a whole fairly represent the applicable law relating to mutual mistake. The jury's verdict reflects its finding that the parties entered into the release in reliance on a mutual mistake of fact.

V.

The jury awarded Robertson a total of $150,000. Consistent with the trial court's instructions, the jury divided this award into two parts: the amount of basic damages, which it computed as $138,000, and the amount sustained as a result of loss of future wage increases or inflation or decrease in the purchasing power of money, which it computed as $12,000. Douglas challenges the inflationary compensation award on the ground that such an adjustment, to compensate for the effects of projected inflationary trends, is too speculative to be included as an element of damages. Although some confusion has existed as to the propriety of an award to compensate for inflation, that confusion has now been dissolved by this Court's recent en banc decision in Johnson v. Penrod Drilling Co., 5 Cir. 1975, 510 F.2d 234, and in Starnes v. Penrod Drilling Co., 5 Cir. 1975, 510 F.2d 234. Despite the inflationary trend now existing in this country, the court noted in Penrod that "we still cannot so surely discern the shadow of inflation as a coming event as to warrant requiring its inclusion in a present rule for calculating future damages." For the reasons stated in Penrod, we must reverse the district court's award of $12,000 as compensation for the future effects of inflation.

Affirmed in part and reversed in part.