These cross-appeals arise out of an airplane disaster at about 5:29 A.M. on June 20, 1956. The libellants' decedent, Marshal Noel, was a passenger on the ill-fated aircraft, a Lockheed Constellation, call numbers YV-C-AMS ('AMS'), owned and operated by Venezuelan Airlines Linea Aeropostal Venezuela ('LAV'). The flight originated from Idlewild Airport, New York, at 3:18 A.M., having as its destination Marquetia, Venezuela. At about 4:30 A.M. engine trouble developed and the pilot notified Idlewild that he was returning. At about 5:29 A.M., after beginning to dump fuel in preparation for its landing, the airplane burst into flame and crashed into the ocean approximately 30 miles off the coast of Asbury Park, New Jersey.1 Libellants brought this action against United Aircraft Corporation ('United'), the manufacturer of the airplane's propeller system, alleging the accident was caused by complications (discussed below) in AMS's #2 propeller,2 and that United was negligent in failing in its duty (1) to perfect a safer propeller system, and (2) to warn LAV of certain weaknesses in AMS's propeller system which resulted in its destruction. The accident occurred outside the territorial waters and hence this action was brought pursuant to the Death on the High Seas Act, 46 U.S.C. 761 et seq., properly alleging admiralty jurisdiction. See Weinstein v. Eastern Airlines, Inc., 316 F.2d 758 (3d Cir. 1963).

The case was tried to the District Court without a jury. It found in libellants' favor in the amount of $387,387.00. 219 F. Supp. 556 (D.Del.1963).

Respondent, United, appeals (at No. 14730) urging that (1) libellants failed to show a causal relationship between the airplane's destruction and failure in its propeller system, and (2) it owed no 'continuing duty' to improve the propeller system once the propellers had been sold to LAV, and had been approved by the C.A.A.

Libellants' cross appeal (at No. 14727) contends that the trial court erred in failing to apply proper legal standards in fixing damages and in making 'clearly erroneous' fact-findings bearing thereon.

The record shows that AMS' #2 engine had begun to overspeed approximately 50 minutes prior to the fatal crash, and that the pilot was unable to feather the propeller. Overspeed is a condition in which the propeller rotates at a rate greater than its maximum capacity. Feathering refers to the operation whereby the blades of the propeller are turned on their own axis so that they parallel the airstream in order to limit rotation once the propeller has been turned off. If overspeed is not brought under control either by feathering or some other way it is considered highly dangerous because of the strong likelihood of disintegration of the engine and its component parts, decoupling, i.e., separation of the propeller shaft from the engine shaft, separation of the propeller from the aircraft and fire.

The District Court made these factfindings:-- Engine overspeed developed at about 4:24 A.M.; after signaling Idlewild that the overspeeding propeller could not be feathered, the plane proceeded to return thereto at gradually reduced speed and altitude; at 4:46 A.M. it signaled an emergency due to overspeed, and asked for assistance; a Coast Guard plane with crew of four, under the command of Commander Hancox, was dispatched to intercept AMS and accompany it on its return flight; at about the same time an Eastern Airlines plane under the command of Captain Fisher changed course in order to attempt interception; the Coast Guard plane spotted AMS and took up escort position to the left and slightly astern about 100 feet above, while the Eastern Airlines plane was approaching; at 5:29 A.M. AMS reported that New York was in sight, and proceeded to dump fuel; immediately thereafter, according to the deposition of Captain Fisher, a 'white puff' appeared followed by a steady trail of mist bubbling or rolling from beneath AMS and extending aft; the plane then became a ball of orange flame; it veered sharply to the right, went into a shallow dive, appeared to gain altitude, and then crashed into the ocean; several small objects were seen to fall free and a piece, which may have been part of the wing, was observed on radar to have split from the main body; a double airplane seat cut completely through by some sharp descending object, and the bodies, right legs severed, of a Mrs. DeArmis and her son, who had been assigned to the #4 seat immediately behind the #2 propeller, were found in the ocean some distance from the scene of the accident; there were no survivors.

The libellants' theory as to the plane crash was presented by its expert, Eugene L. Grindle, as follows:-- the propeller decoupled as a result of the unchecked overspeed; the pilot, relieved by an accompanying decrease in engine screech, decided to dump fuel; at that moment some movement of the plane caused the #2 propeller to separate; the propeller moved inboard, slashed through the fuselage and knocked the plane out of control; when the propeller separated it left hot nose section particles exposed, which flamed into fire and ignited the fuel as it was being dumped; the loss of the propeller forced the plane to veer to the right; it also caused the ejection of at least one seat.

The respondent's theory was that the accident was caused by a fire and explosion in the right wing. It must immediately be noted that the trial judge rejected this theory as 'sheer speculation.' 219 F. Supp. 563-566.

An 'Accident Report' of Frank Gates, Jr., a Lockheed investigator, placed in evidence by the respondent, stated in part:

'* * * one seat, the only one recovered, had been cut from top to bottom by a downward, angular, smashing blow that went right through the seat left hand arm rest, seat cushion, and lower frame. Paint marks and spacing of the marks on seat structure were visibly similar to the paint banding which is on propellers adjacent to the tip. The cut was from the left hand side, and at the same angle at which the #2 propeller would hit if it tipped back around the cowling, as it would if it came off while windmilling, and happened to go to the right and into the fuselage. It would hit at about the left hand third-seat-back position, right over the #5 fuel tank. * * * we checked and found that the #5 tank was supposed to have minimum fuel in it at the time of the fire, only about 50 gallons, so we don't quite know whether or not the propeller could have released enough fuel by chopping down into it to cause the general fire.'

The first issue raised by respondent's appeal is that of causation. The trial court found that the 'accident was the result of a prolonged overspeed and inability to feather the propeller, leading to a decoupling, fire, and a separation of the propeller. Simultaneously, the propeller crashed through the fuselage dislodging the #4 seat, ejecting it and its occupants, and gashed into the #5 belly tank which exploded. The flames from the #2 engine were swept back over the wing directly into the raw plumes of gasoline being dumped from the port dump chutes, igniting them. Also, the flames from the explosion added to and increased the holocaust resulting in the total loss of the plane.' 219 F. Supp. 566.

Respondent contends that the evidence does not support the court's findings as to causation and further urges that (1) Grindle was not qualified to testify as an expert; and (2) it was error to credit Captain Fisher's version of events immediately preceding the crash of AMS in view of the testimony of Commander Hancox and the two members of his crew who said they saw an undulating blue flame on the starboard or right side of AMS which appeared above and below the fuselage and continued for about seven seconds.

As to Grindle's qualification as an expert, it is settled that in the absence of a clear abuse of discretion the trial judge's determination of competency will not be set aside. Roberts v. United States, 316 F.2d 489 (3d Cir. 1963); Hickey v. United States, 208 F.2d 269 (3d Cir. 1953), cert. den. 347 U.S. 919, 74 S. Ct. 519, 98 L. Ed. 1074 (1954); Trowbridge v. Abrasive Co. of Philadelphia, 190 F.2d 825 (3d Cir. 1951).

The record shows that Grindle had been assistant chief flight engineer with Pan American Airways for 11 years, and that he had investigated about 30 cases of overspeed. He was also familiar with hundreds of other cases through Air Force reports and similar sources. While at Pan American part of his work was to initiate corrective action and procedures necessary to cope with overspeeds. He also participated in the investigation of between 30 and 40 airplane crashes. Finally he was the inventor of devices having to do with propeller vibration and featherability.

While it is contended that Grindle never had any direct contact with planes of AMS' type, Constellation 1049, and was unfamiliar with the dumping facilities on such aircraft, the record shows that he had some familiarity with the 1049 through his reading and that he had worked with earlier models of the Constellation. Respondent further contends that Grindle was lacking in education, lacked government experience and certification, had little actual flying experience, had never before been used as an expert, and was not up-to-date on C.A.B. accident reports.

On review of the record we cannot say that the trial court abused its discretion in crediting Grindle as an expert.

The respondent's further contention that the trial judge's findings as to causation are not supported by the evidence must be considered in the light of the fact that this case is in admiralty, and the 'clearly erroneous' rule is applicable. McAllister v. United States, 348 U.S. 19, 75 S. Ct. 6, 99 L. Ed. 20 (1954); Pennsylvania Railroad Company v. S.S. Marie Leonhardt, 320 F.2d 262 (3d Cir. 1963); Petition of Oskar Tiedemann & Co., 289 F.2d 237 (3d Cir. 1961).

As to this contention we are of the opinion that the record affords substantial basis for the trial court's factfindings on the score of causation.

We come now to the respondent's contention that it was not negligent because there was no 'continuing duty' on its part, as a manufacturer, 'to develop an improvement' of its propellers, and that its 'duty' with respect to the propeller system ended when the plane was delivered to LAV with CAA approval. The contention is directed to the trial court's finding that the respondent was under a 'continuing duty to improve its propeller system in view of the factor of human safety involved' and that it breached this duty. 219 F. Supp. 574.

The respondent cites these three cases in support of its contention that it was not under a 'continuing duty' here: Pontifex v. Sears, Roebuck & Co., 226 F.2d 909 (4th Cir. 1955); Marker v. Universal Oil Products Co., 250 F.2d 603 (10th Cir. 1957); Mitchell v. Machinery Center, Inc., 297 F.2d 883 (10th Cir. 1962). We are of the opinion that these cases are inapposite. Pontifex deals with gas powered lawn mowers and improvement of their rope starting cord. The difference lies in the type product involved. There the court apparently felt that the rope starting device was of such a nature that the party to whom the duty was allegedly owed would be alerted to its limitations. Obviously that is not the case here. Marker and Mitchell stand for the proposition that a manufacturer need not develop safety devices to protect against every remotely possible danger. Here, as the trial court's findings make clear, the danger of the occurrence of uncontrolled overspeed and its effects was not hypothetical but a generally recognized danger.

Applicable here is the rule of MacPherson v. Buick Motor Co., 217 N.Y. 382, 111 N.E. 1050, L.R.A.1916F, 696 (1916). See North American Aviation, Inc. v. Hughes, 247 F.2d 517 (9th Cir. 1957), cert. den. 355 U.S. 914, 78 S. Ct. 341, 2 L. Ed. 2d 273 (1958); Sieracki v. Seas Shipping Co., Inc., 149 F.2d 98 (3d Cir. 1945), affd. 328 U.S. 85, 66 S. Ct. 872, 90 L. Ed. 1099 (1946), which extends the duty of care owed by the manufacturer of an instrumentality likely to endanger the public to injured persons making proper use of that instrumentality, regardless of contract notions of privity.

The District Court made the following uncontroverted findings of fact at 219 F. Supp. 568:

'* * * Prior to June 20, 1956, there existed a number of instances of overspeeds accompanied by inability to feather resulting in decoupling, fire, or a propeller separation or a combination of the three.3 LX (Libellants' Exhibit) 77 and 78 reveal some 79 actual instances of overspeeds of which there were 73 cases of inability to feather. LX 53 demonstrates 8 common types of overspeeds of which, it was conceded by United, in three types the propeller definitely would not feather and in the remaining five, the ability to feather would be marginal. As early as 1950, United was aware of inabilities to feather due to sticking valves in the governor from oil contamination (LX 65). United admitted that the need for an alternative feathering device was known as early as 1954 but LX 133 indicates that United was aware of this fact as early as 1952 and LX 94 would seem to demonstrate as early as 1950. LX 64, dated March, 1954, over two years prior to the accident, reveals Mr. Pond, a responsible official of United advising Mr. Thorell of the Woodward Governor Co. (a wholly owned subsidiary) that a plane could be lost as a result of governor failure. Mr. Rhines, United's Assistant Chief Engineer, also admitted this knowledge (RP 2023). Significantly, in LX 94, we see Mr. Borger, Chief Engineer for Pan American saying in exasperation to Mr. Hupp, the United official, 'Only after 8 years of overspeeding and inability to feather Hamilton Standard (United) starts to think about changes."

The Court also found that numerous other instances of overspeeds may have occurred which could not be documented because records had been destroyed by United 'for reasons not deemed satisfactory to this Court.' 219 F. Supp. 569.

It made the further fact-findings that three approaches to the problem of overspeed had been made at the time of the accident: the first, Integral Oil Control, was a device which acted as a governor having its own oil supply in order to avoid contamination, since oil contamination is the principal cause of inability to feather, but this device, although used to some degree by the military, had been rejected for the most part by the commercial airlines as being too expensive, too heavy, too complicated and poorly located, and not providing a workable solution; a second device developed was a separate feathering valve which did not solve the problem of oil contamination; and a third device was Pitch Lock, a mechanism which detects when an overspeed has started or is about to start, and prevents the blade angle from going to low pitch by locking oil in front of the piston-- the purpose of Pitch Lock is to limit overspeeds.

The District Court found that while Pitch Lock had been certified by CAA and had been earlier developed by the respondent for use on Lockheed Electras4 it had not developed them for use on Lockheed Constellations at the time of the accident. Its conclusions on the issue of negligence are stated at 219 F. Supp. 572 as follows:

'Assessing United's development program for the correction of overspeeding problems against the background of the standard of care imposed upon it by law, I am forced to the conclusion that United was negligent in that it should and could have produced pitch lock for use on Constellation type planes prior to June 20, 1956, and had it done so, the ability of AMS substantially to control the overspeed would have been materially increased.'

The record establishes that the respondent was aware that absent some type of control mechanism, continued use of its propeller systems on Lockheed Constellations endangered the public, and that despite this awareness it permitted the development of an effective safety device adaptable to these planes to lag behind similar development for other airplanes. Accordingly, we cannot say that the trial court erred in its findings that respondent was negligent.

Libellants in their cross appeal contend in the main that the trial court's fact-finding that the decedent would have had no net income or earnings during the first five years of his 20-year life expectancy was 'clearly erroneous' and based on an improper legal standard. They also urge that the District Court erred in failing to allow was an item of damages interest for the seven-year period between the date of decedent's death on June 20, 1956, and July 9, 1963, when judgment was entered in their favor.

The libellants further complain that the trial court's 'subsidiary' fact-finding as to the decedent's earnings during the five years preceding his death were 'clearly erroneous' because it failed to take into account $70,000 of 'consulting fees' received by the decedent, a mechanical engineer, during the period.5

The trial court made these fact-findings with respect to its allowance of damages at 219 F. Supp. 575-576:

'* * * just prior to his death, he had liquidated his own business for $400,000. He was engaged in a heavily losing business venture in Texas. He was just embarking on a completely new business venture in Venezuela which, at the time of his death, had not begun operations. 'Mr. Noel was very obviously a heavy spender. Between 1951 and the date of his death, his luxurious mode of life coupled with heavy business losses exceeded his entire five previous years' earnings by $100,000. While he had discussed the liquidation of the losing Texas venture, we do not know when or at what further loss this would have been accomplished, but it probably would have been deferred as long as possible because the sole purpose of this business was to provide a livelihood for Mr. Noel's father and brother. We do know he was starting a completely new, untried business venture in Venezuela, a risky sort of enterprise in view of the history of political instability and confiscation of foreign assets common in most South American countries. 'In this connection, also, it is to be emphasized that not one iota of evidence was introduced as to the background, capitalization, contracts or prospects, immediate or future, of the Venezuelan venture. The combination of the relatively heavy and continuous losses on the Texas business and the complete absence of any proof as to the probability of success of the Venezuelan venture strongly suggests the unlikelihood of any net income to Mr. Noel for a number of years during which he might have had difficult times and been forced to live on capital. It is entirely within the realm of possibility, and I so find, that it would have taken Mr. Noel five years 26 to '27. One could reasonably assume that if, by this time, the Venezuelan enterprise was not successful, Mr. Noel, would have abandoned it and devoted his considerable talents to the sale of heavy machinery in this country. establish his Venezuelan business on a profitable basis and at the same time to extricate himself from his financial difficulties in Texas. During this time, he not only would decrease in the value of the dollar but, in fact, been compelled to reduce sharply his relatively luxurious manner of living. It is of course unrealistic to predict any exact, annual net income thereafter. However, I find that, in all, for the balance of the 15 years of his life, the total damages, reduced to present worth at 4%, would amount to $387,387.

'Of this sum neither Pat nor Sharon would be legally entitled to anything because of the finding that Mr. Noel would have had no net earnings for 5 years after his death. However, Marsha would have been entitled to 6 years of maintenance and support and, in addition, a college education. I therefore find that, reduced to present worth at 4%, Marsha would be entitled to $30,000.

'Because of his lavish way of life, I cannot conclude that Mr. Noel would have left any substantial estate even though in law each daughter as she arrived at the age of 21 would cease to be a dependent. If any estate were left, it would have been bequeathed entirely to Mrs. Noel. I decline to speculate on any over future years and I further decline any award to the members of the deceased's family for so-called loss of nurture. I conclude that his wife and daughters would have outlived decedent.

                "AWARD
--------------------------------------
"Total damages (reduced
  to present worth at
  4%)                        $387,387
"Patricia                    ......
"Sharon                      ......
"Marsha (reduced to present
  worth at 4%)               $ 30,000
"Mrs. Noel--Balance          $357,387"

It must immediately be said that the trial court's fact-finding that the decedent, had he lived, 'would probably have had no net income', for the first five years of his life expectancy was without evidential basis and wholly speculative, and was therefore 'clearly erroneous'.

It is well-settled that mere speculation cannot be substituted for proof and the requirement is for probative facts capable of supporting, with reason, the conclusion reached in the verdict. Sharon Herald Co. v. Granger,195 F.2d 890, 895 (3 Cir. 1952); In re Lueders' Estate, 164 F.2d 128, 133 (3 Cir. 1947). The rule was succinctly stated in Galloway v. United States, 319 U.S. 372, 395, 63 S. Ct. 1077, 1089, 87 L. Ed. 1458 (1943), as follows:

'The essential requirement is that mere speculation be not allowed to do duty for probative facts, after making due allowance for all reasonably possible inferences favoring the party whose case is attacked.'

To the same effect see Tennant v. Peoria & P.U.R. Co., 321 U.S. 29, 64 S. Ct. 409, 88 L. Ed. 520 (1944), and Myers v. Reading Co., 331 U.S. 477, 67 S. Ct. 1334, 91 L. Ed. 1615 (1947).

It must further be pointed out that in making this fact-finding of 'no net income' for five years, the trial court, insofar as the Texas venture was concerned, disregarded the undisputed testimony that the decedent during the five years preceding his death had managed to achieve a total of $418,000. of 'earned' income despite his involvement with his losing Texas business, and that the decedent just prior to his death had made known his intention to liquidate it. And, insofar as the Venezuelan venture was concerned, the trial court, in finding that 'it is entirely within the realm of possibility * * * that it would have taken Mr. Noel five years to establish * * * (it) on a profitable basis,' disregarded the testimony that it would have yielded him an annual income of 'eighty-some-thousand dollars.'6 That the trial court simply overlooked this testimony is apparent from its statement at 219 F. Supp. 575 that 'it is to be emphasized that not one iota of evidence was introduced as to the background, capitalization, contracts or prospects, immediate or future, of the Venezuelan venture.'

The stated errors in the trial court's fact-finding will require the granting of a new trial restricted to the issue of damages.

On this appeal libellants contend that the trial court erred in failing to allow as an item of damages interest for the seven-year period between the date of decedent's death on June 20, 1956 and July 9, 1963, when judgment was entered in their favor. However, it is not apparent from the trial court's opinion as to whether it disallowed or took into consideration this item in fixing damages. That it should have been a factor in fixing damages is indicated by the provisions of the Death on the High Seas Act, 46 U.S.C.A. 761 et seq. Moore-McCormack Lines, Inc. v. Richardson, 295 F.2d 583, 96 A.L.R. 2d 1085 (2 Cir. 1961), cert. den. 368 U.S. 989, 82 S. Ct. 606, 7 L. Ed. 2d 526 (1962) and National Airlines, Inc. v. Stiles, 268 F.2d 400 (5 Cir. 1959), cert. den. 361 U.S. 885, 80 S. Ct. 157, 4 L. Ed. 2d 121.

A final comment.

The trial court in its fact-findings spoke of 'net income' and 'net earnings'. The Death on the High Seas Act allows as damages 'fair and just compensation for the pecuniary loss sustained by the persons for whose benefit the suit is brought * * *' 'Pecuniary loss' is the 'reasonable expectation of pecuniary benefits which would have flowed from the continued life of the deceased.' The S.S. Black Gull, 90 F.2d 619 (2 Cir. 1937), cert. den. Faye v. American Diamond Lines, 302 U.S. 728, 58 S. Ct. 50, 82 L. Ed. 562. 'Pecuniary loss' is not measured in terms of, nor restricted to, 'net income' or 'net earnings'.

For the reasons stated the judgment of the District Court in favor of the libellants and against the respondent on the issue of liability will be affirmed. The cause will be remanded with directions to grant a new trial restricted to the issue of damages in accordance with the opinion of this Court.

Before BIGGS, Chief Judge, and McLAUGHLIN, KALODNER, STALEY, HASTIE, GANEY, SMITH and FREEDMAN, Circuit Judges.

KALODNER, Circuit Judge.

Upon consideration of the respondent's Petition for Rehearing and the libellants' Answer thereto, it seems appropriate to supplement our Opinion in the instant case, filed December 9, 1964. There we limited our discussion to the respondent's specific challenges to the trial court's fact-finding that the fatal airplane crash 'was the result of a prolonged overspeed and inability to feather the propeller, leading to a decoupling, fire and separation of the propeller,7 and its holding that the respondent had breached 'a continuing duty to improve its (the) propeller system in view of the factor of human safety involved.'8

In limiting our discussion to the issues as presented by the respondent we regretably did not declare that the undisputed evidence required an explicit finding by the trial judge that the respondent was negligent on July 15, 1955, when it supplied a propeller system of a design which it knew9 had earlier caused 10 serious accidents of decoupling, fire or separation,10 and 57 accidents of lesser degree (engine fires or engine failures.)1112 The finding that the respondent was negligent on July 15, 1955, is implicit in the trial court's expressed finding that the defective design and manufacture caused the accident on June 20, 1956.

We should also have declared that the undisputed evidence required an explicit finding that the respondent was negligent in failing to notify LAV, owner and operator of the doomed plane, of the numerous malfunctionings of its propeller system both before July 15, 1955 and June 20, 1956.13

On the score of these required findings of negligence it must be pointed out that the libellants charged the respondent with (1) a defect in the design of the propeller system in the respect that it was unable to feather overspeeds; (2) knowledge that the system was unable to feather overspeeds, and (3) failure to notify LAV of the frequency of failures to feather overspeeds.

It should be noted parenthetically that the issue of 'continuing duty' was not raised by either the Libel or Answer, filed in 1958, and that it was first injected into this case by the respondent in 1962 in the pre-trial proceedings in its statement of 'Issues of Law to be litigated.'14

The testimony adduced at the trial with respect to the issue of 'continuing duty' developed that there was, in accordance with normal practice, a continuing relationship between the respondent and LAV from the delivery of the propeller system on July 15, 1955 to the plane crash on June 20, 1956. In the course of that relationship the respondent's field service department advised LAV with regard to the maintenance, overhaul and operation of the propeller system and supplied it with service bulletins supplementing manuals of instruction. The respondent's own expert witness, Walter Thomas Grady, testified that a manufacturer 'has to look at the performance of his product as it now stands, examine the statistics on it, and also examine the type of failures. * * * If * * * he sees demonstrated that catastrophies can result from the equipment as it now stands, then obviously he would be obligated to take remedial action.'15

The foregoing affords more than ample support for the trial judge's holding that the respondent was under a 'continuing duty' here. His conclusion that the respondent breached that duty in failing to make available Pitch Lock16 for use on the doomed plane prior to its crashing on June 20, 1956, is supported by the undisputed testimony, that the respondent failed to act despite its knowledge of four serious accidents attended by decoupling, fire or propeller separation, and 23 lesser serious mishaps, all resulting from inability to feather overspeeds which occurred subsequent to July 15, 1955 when it delivered the faulty propeller system.

The sum of what has been said here, and in our Opinion of December 9, 1964, is that the trial judge's determination that the respondent was negligent is soundly premised on any or all of these grounds:17 (1) it delivered a propeller system on July 15, 1955 which was defective in design; (2) it failed to discharge its duty to warn LAV of numerous malfunctionings of its propeller system before July 15, 1955 and before June 20, 1956, and (3) it breached its 'continuing duty' to make available to LAV Pitch Lock prior to June 20, 1956.

For the reasons stated the petition for rehearing will be denied.

HASTIE, Circuit Judge (concurring).

The opinion which evoked this petition for rehearing turned upon its approval of the trial court's concept that, after distributing a type of airplane propeller in commerce a manufacturer has a continuing duty to develop and adapt improvements calculated to make that article that is already in use safer. This doctrine is new and far reaching and, in my view, may well be unsound. If this new concept were essential to the disposition of this appeal I would vote for rehearing before the full court.

However, Judge Kalodner's supplementary opinion now points out that the detailed undisputed facts which are disclosed in the opinion of the trial judge, who decided this case without a jury, required a conclusion that the manufacturer was negligent, at least in failing to warn and inform its vendee that there had occurred some sixty or more episodes of propeller malfunction, all of the type which subsequently caused this accident, most of them even before this propeller was sold. no reasonable man could possibly square such conduct with the standard of due care in the circumstances.

There would be no point, therefore, in sending this case back to the district court to find again, as rationally it must, though on a different theory, that the facts already established constitute actionable negligence. For this reason, and without approving the new concept of a manufacturer's continuing duty to improve a device after he has sold it, I vote to deny rehearing before the court en banc.

FREEDMAN, Circuit Judge (dissenting).

The opinion of Judge Kalodner on the petition for rehearing sufficiently removes from the case an apparent acceptance of the doctrine that a manufacturer turer has a continuing duty to develop improvements in the safety of his product which was non-negligently produced. I would therefore, join the concurring opinion of my brother Hastie but for the fact that the District Judge who sat without a jury expressly planted his decision on that ground and made no finding that the propeller system was defective at the time the defendant sold it or that defendant was negligent in failing to give warning of a defect. Thus the District Judge stated: 'Assessing United's development program for the correction of overspeeding problems against the background of the standard of care imposed upon it by law, I am forced to the conclusion that United was negligent in that it should and could have produced pitch lock for use on Constellation type planes prior to June 20, 1956 (the date of the accident), and had it done so, the ability of AMS substantially to control the overspeed would have been materially increased.' 219 F.Supp. at p. 572. Absent an express finding of fact of such negligence I find myself unable to concur in the majority's conclusion.

It will not do to say, as does the majority on the petition for rehearing, that a finding of negligence on the date of the sale 'is implicit in the trial court's expressed finding that the defective design and manufacture caused the accident on June 20, 1956.' For the trial judge's opinion itself recognizes that airplanes and their parts are still not perfect nor free from all defects. The question is not whether there was a defect in design and manufacture which caused the accident but whether such defect existed at the time of the sale, and was one which fell below the standard of due care under the circumstances and constituted negligence. Any other view would absorb negligence in the doctrine of absolute liability, which does not now require consideration as a preferable alternative.

Again, it seems to me impossible to hurdle the absence of a finding by the trial judge of negligence in manufacture or design at the time of the sale by declaring that we believe that such a finding could have been explicitly made from undisputed evidence.1 We may not make findings which were not made by the trial judge, even if the evidence was uncontroverted.

It may seem an overemphasis on procedural niceties to insist that we may not rest our conclusion on a theory for which there are no supporting findings of fact. But the silence of the trial judge on this subject resulted from his adoption of a theory which is now excluded and we may not assume that on a new focus of attention the findings which may appear appropriate to us will inevitably be made by the trial judge. In any event, it seems to me that this is a matter in which we should not speculate. I therefore would vacate the judgment and direct the court below to make specific findings of fact on the charge of negligence at the time of sale and on alleged failure to warn of defects after the sale. The parties will have their opportunity to argue the inferences and conclusions to be drawn from the evidence, the trial judge who heard the witnesses will be able to make express findings on the facts, and we will not be required to become finders of fact in his stead to fill in the presently existing factual void.

I therefore dissent from the denial of the petition for rehearing.

BIGGS, Chief Judge (dissenting).

I concur in the views expressed in Judge Freedman's dissenting opinion. But I wish to make it clear that point '(3)' of the opinion of this court following the filing of the petition for rehearing is one of unique and unusual importance.

The trial court and this court both have treated June 20, 1956, the date of the accident as critical. Liability is said to be bottomed on a 'continuing duty' of a kind with which the writer is unfamiliar and which seems to be an extension of existing law. In such cases as Mac-Pherson v. Buick Motor Co., 217 N.Y. 382, 111 N.E. 1050, L.R.A.1916F, 696 (1916); North American Aviation, Inc. v. Hughes, 247 F.2d 517 (9 Cir. 1957), cert. denied, 355 U.S. 914, 78 S. Ct. 341, 2 L. Ed. 2d 273 (1958), and Sieracki v. Seas Shipping Co., Inc., 149 F.2d 98 (3 Cir. 1945), aff'd., 328 U.S. 85, 66 S. Ct. 872, 90 L. Ed. 1099 (1946), the negligence was found in the manufacture or design of the dangerous machine or instrumentality prior to or at the time of its delivery to the user. But the extension of the MacPherson v. Buick doctrine, applicable to maritime torts, via Sieracki, in the case at bar consists of the injection of the concept that the manufacturer of a dangerous device or instrumentality, when sold, is liable to the user, if at any time prior to the accident, by the further development of the science or the art of which it is a part, the machine or device may be so developed or perfected that the accident could have been avoided. The obligation imposed by the court below and by this court under point '(3)'1 begins with the delivery of the chattel and seemingly persists until it is no longer in use.

Such an extension of the law of liability requires, in my opinion, prior to its adoption, a full review and discussion and a hearing before this court en banc.

For these reasons I dissent from the denial of the petition for rehearing.

I am authorized to state that Judge Freedman joins in the views expressed in this opinion.