In a jury trial, Clyde 0. Darling was awarded damages in the amount of $100,000 for personal injuries upon his complaint charging defendant with negligence in the manufacture of a bulldozer. Defendant made a motion for new trial, which was ordered granted unless plaintiff agreed to remit $20,000 and ordered denied if plaintiff filed the remission. Plaintiff filed a remission of the judgment in that amount. Defendant appeals.
Plaintiff was an experienced tractor operator, employed by the firm of Morrison-Knudson in the construction of a Marine Corps base at Twenty-Nine Palms. For five or six weeks prior to his accident, he had been operating a bulldozer designed *716and manufactured by defendant, pushing dirt and other materials into a hopper and removing materials from underneath certain bunkers.
Between the hood at the front of the bulldozer and the driver’s seat at the rear extended a flat metal sheet, called the deck plate, which provided plaintiff with a level walking area 4 feet wide and 17 inches deep. A smaller sheet of metal, called the inspection cover, fit into an 8% inch by 10% inch aperture in the deck plate; underneath the aperture was the clutch assembly, which was unguarded; the inspection cover could be raised by turning a latch on the rear edge; when the latch was shut, the cover was flush with the surface of the deck plate. The inspection cover was attached to the deck plate by means of a hinge welded to the underside of both sheets of metal on the side opposite the latch. It received additional support from a ledge or lip of the deck plate beneath the rear edge of the cover which extended one inch into the aperture along 8 inches of the underside of the deck plate for about 4 inches to the left of the latch and four to the right. Both sheets were designed and manufactured by defendant; the hinge was manufactured by another company according to defendant’s specifications and was welded to the sheets in defendant’s factory.
December 16, 1952, plaintiff prepared to remove some dirt from beneath a bunker at the job site. Before doing so, he had to disconnect the exhaust pipe of his bulldozer, which extended vertically above the hood. Darling stopped the tractor and left the motor idling, as was his usual practice. He stepped across the deck plate and removed the exhaust pipe, then turned and stepped back toward the driver’s seat. As he stepped on the inspection cover, which was latched shut, the hinge broke off and both the cover and his right leg slipped downward through the aperture and came in contact with the unenclosed moving parts of the clutch assembly. Plaintiff’s leg was badly mangled, requiring amputation above the knee. It was established that the hinge gave way because of a failure of the weld connecting it to the inspection cover and the deck plate.
One question in the ease was whether, as plaintiff contended, the defective weld was the original one put on by defendant’s employes, or whether, as defendant contended, the weld which failed was a later weld made by employes of Morrison-Knudson. As we shall see, expert testimony was introduced in support of each of these irreconcilable theories. A second question *717which was also the subject of expert testimony was whether defendant was negligent in designing the deck plate. The jury found that defendant was guilty of negligence which was the proximate cause of the accident. The finding may have been predicated upon negligence in the manufacture of the bulldozer or in its design or both.
All the points on the appeal may be summarized under the following headings: (1) Defendant’s negligence, if any, in the manufacture of the bulldozer was not a proximate cause of plaintiff’s injuries; (2) Defendant’s negligence, if any, in the design of the bulldozer was not a proximate cause of plaintiff’s injuries; (3) The court committed error in giving an instruction which permitted the jury to find against defendant under either a theory of negligent manufacture or negligent design; (4) The court erred in its instructions on proximate cause; (5) The court should have instructed the jury as to the possible liability of Morrison-Knudson for additional workmen’s compensation on account of possible serious and wilful misconduct.
There was evidence of the following facts. The bulldozer was manufactured in 1949. May 31, 1949, it was delivered to Morrison-Knudson de Sonora, a Mexican affiliate of plaintiff’s employer. From January 21, 1950 until March 1954 the tractor was in use at various points in the United States. While in Mexico, it was operated and repaired by a crew of relatively unskilled Mexican laborers. After 292 hours of use, report was made to the dealer that the clutch was defective and required repair. In the course of repair and at other times prior to January 1950, the deck plate was removed; this was accomplished by unscrewing the bolts, lifting the plate over the control rods and throwing it onto the ground. Deck plates removed in this manner are sometimes damaged through being run over or struck by falling objects while on the ground.
The deck plate and inspection cover which were on the bulldozer at the time of the accident were received in evidence. They are before us as exhibits. Darling testified that the cover was flush with the surface of the deck plate before the accident occurred. According to Towne, a mechanic who started the bulldozer that morning, the inspection cover was intact, and Douglas, the service foreman, testified that he opened the inspection cover and greased the clutch on the morning of the accident, then closed and locked the cover; if *718anything had been out of order, he would have reported it to the master mechanic.
Both exhibits are now seriously deformed and bent and were in that condition at the time of the trial. As mentioned earlier, the inspection cover was designed to rest on the lip of the deck plate when the latch is closed. However, the segment of lip to the right of the latch is now bent and curves upward at the end. And the wing of the inspection cover which was meant to rest on the segment of lip to the left of the latch bends downward at an angle of 70 degrees with the plane of the remainder of its surface. Two of plaintiff’s fellow workmen testified that the cover was bent in this fashion immediately after the accident; when they first observed it, the cover was inside the aperture housing the clutch assembly and it was resting against the fly wheel. A photograph of the underside of the deck plate, taken by plaintiff six weeks after the accident, was received in evidence as an exhibit. In the opinion of defendant’s photographic expert, one Thomas, the lip of the deck plate cast a curved shadow on the photograph which was caused by a bend in the plate. However, it was plaintiff’s best recollection that there was no bend in the plate when he took the picture.
In a further effort to prove that the inspection cover was bent prior to the accident, defendant called as expert witnesses two qualified engineers, Shields and Snyder; Snyder stated his opinion that the deformation was caused by at least 50 separate blows from a ball peen hammer or similar instrument; the cover was in a vertical position while being hammered. Shields gave his opinion that due to the bend in the deck plate lip, the cover could not have been latched shut unless it, too, were bent out of shape; bending the cover would have permitted the latch to be closed. Both witnesses stated their opinion that some of the energy from the hammer blows would have been transmitted through the inspection cover to the hinge and this would have resulted in damage to the weld connecting the hinge to the cover and deck plate. It was also their opinion that the cover could not have been bent through contact with the clutch mechanism because the rotating clutch fingers would have broken off and the marks on the bent wing of the inspection cover would have been uniform in direction rather than random.
It is undisputed that the fractured weld did not comply with defendant’s factory specifications. Dr. Morrelli, a qualified mechanical engineer, testified on behalf of plaintiff that the *719weld was one-eighth inch shorter than required by the specifications and its side dimensions were also substandard; the fracture occurred at the point of junction of the weld and the parent metal of the hinge. In his opinion, the weld broke because there was not a full11 eighth-inch fillet weld, there is less than a sixteenth. ’ ’ Morrelli’s testimony was corroborated by a Mr. Pellett, an expert metallurgist, who stated that the weld broke because of the poor contour of the weld and an insufficiency of weld material at the point of stress. In Pellett’s opinion, the weld did not fail suddenly, but failed gradually over a considerable period of time. Morrelli testified that if the weld had conformed to defendant’s specifications it would have been 100 per cent more weight-bearing.
The weld was examined both visually and microscopically by Mr. Pellett and Dr. Morrelli. The latter took a cross-section of the weld and the adjacent area of the hinge and subjected it to analysis under the microscope. He took only one specimen of the weld because, as he stated, a number of cross-sections taken from different places across the weld would “eliminate all of the evidence.” In his opinion, the defective weld was the original weld made by defendant.
Edward Pess, a qualified metallurgist, testified on behalf of defendant that he was able to identify two runs of welding in his microscopic analysis of a second cross-section. He observed a portion of a prior weld and a % inch chisel mark made during the removal of a prior weld. The fracture line ran through the weld itself rather than through the fusion zone between the weld and the parent metal. In the opinion of the witness, the weld he studied was not the weld put on at defendant’s factory. Furthermore, while defendant used open arc electric welds, the deck plate showed discoloration characteristic of oxygen-acetylene welding.
However, the cross-section taken by Fess was also examined by plaintiff’s expert, Pellett. Pellett testified that he saw no indications either of additional welding or of the removal of weld material. He stated that if the original weld had been removed and replaced by a new weld, chisel marks made in removing the prior weld would have been eliminated in the course of rewelding.
Defendant also offered in evidence the results of a spectrographic analysis purporting to show that the fractured weld contained .10 per cent tin and .38 per cent silicon. A similar analysis of the materials used in the welding rods utilized by *720defendant from 1946 to 1951 purported to show that the tin content varied from nil to an infinitesimal quantity; between 1947 and 1949 the silicon content varied from .10 per cent to .20 per cent. The welding rods were purchased by defendant from the manufacturer and a number were subjected to chemical tests upon receipt. Plaintiff’s expert, Pellett, stated that the normal limit of tin content of American wire used in the manufacture of welding rods varies from nil to .01 per cent and that he knew of no electrodes made for the general American market containing more than .05 per cent tin. The expert witnesses were in general agreement that tin content as high as .10 per cent would affect welds adversely by making them more brittle.
As to the design of the inspection cover and its appurtenances, Dr. Morrelli stated his opinion that the “accepted procedure” is to eliminate stress on the hinge by extending the deck plate lip either all the way around the cover or along the full length of the side opposite the hinge. Otherwise, if the hinge broke due to a failure of the weld, the latch would not remain locked, but would slide out. We mentioned earlier that the clutch mechanism was unenclosed. Jass, a designer in defendant’s engineering department, testified that the tractors manufactured since 1952 were equipped with an enclosure over the clutch to protect it from dust and dirt and that had the clutch in the bulldozer been similarly guarded, plaintiff’s leg would not have come in contact with the moving parts of the clutch assembly.
The first point to be considered is that, as a matter of law, the accident was not proximately caused by negligence in the manufacture of the bulldozer.
There is no dispute as to the applicable legal principles. The manufacturer of a chattel owes a duty of care toward a user, although there is no privity of contract between them, where the article is inherently dangerous, or where it is reasonably certain, if negligently manufactured, to place life and limb in peril. (35 Cal.Jur.2d 596-598.) However, he is not liable for the user’s injuries where they result from the unforeseeable negligent use of the chattel by a third party. (Waterman v. Liederman, 16 Cal.App.2d 483 [60 P.2d 881, 62 P.2d 142].)
Defendant argues that it was not responsible for the defective weld; that the hinge was probably rewelded by the unskilled employes of Morrison-ICnudson’s Mexican affiliate in order to rectify damage to the original weld resulting from *721mishandling of the deck plate and inspection cover in the field; and that the causal chain between the manufacture of the bulldozer and the injury to plaintiff was broken, as a matter of law, either b;r the unforeseeable modification of the machine through the bending of the metal plates and the re-welding of the hinge or by the passage of time. We are not persuaded by these arguments.
Defendant’s attorneys know full well that it is not our function as a reviewing court to reweigh the evidence and to substitute other inferences for those reasonably drawn by the triers of fact and that we are limited to determining whether there was any substantial evidence to support implied findings of the jury. In our opinion the evidence was ample to support findings that the inspection cover was not bent prior to the accident and that the fractured weld was the original one. The argument that there was no substantial evidence of these facts is unavailing. The jury could reasonably have believed the testimony of Darling and his fellow employes that the inspection cover was flush with the surface of the deck plate on the day of the accident and that neither the cover nor the deck plate was deformed at that time. The jury could also have believed the testimony of the expert witnesses called on behalf of plaintiff that the admittedly defective weld was the weld placed on the hinge at defendant’s factory in the process of manufacturing the bulldozer. Of course, a diametrically opposite conclusion would have found support in the testimony of the experts called by the defense. But the weight to be accorded to conflicting expert testimony is primarily a question for the trier of fact. (19 Cal.Jur.2d 39-40.)
Defendant seeks to transform this question into one of law by emphasizing the evidence that the tin and silicon content of the fractured weld were greatly in excess of the amounts contained in the welding rods used by defendant and in other welding rods manufactured in the United States. It is true that the laboratory technician, Barben, who analyzed the material in the weld, testified that he found that it contained .10 per cent tin and that none of plaintiff’s experts analyzed it for tin content. If the jury had given full credit to the opinion of Barben it could with good reason have concluded that the weld was not one made with American-made rod. But the testimony of the technician did not stand uncontradicted. It was opposed by the positive opinions of plaintiff’s experts, Pellett and Morrelli, that the weld was the *722original one. The jury presumably gave greater credit to the opinions of these experts than to the testimony of Barben. The evidence that all American-made welding rod contains not more than .05 per cent of tin and that defendant uses only American-made rod does not overcome, as a matter of law, the expert testimony that only one weld had been made. If there was only one weld, responsibility for its failure rested upon defendant regardless of its tin content. We cannot reweigh the conflicting testimony of these witnesses and hold that the jury could not reasonably have relied upon the testimony of plaintiff’s experts. Their testimony was that there was nothing which indicated removal of the original weld. It is evident that the jurors were convinced of that fact and that they had sufficient reason for their conclusion.
As we say, the jury could reasonably have determined that defendant was guilty of negligence in the manufacture of the bulldozer. Since the jury was not required to accept defendant’s hypothesis that the deck plate and inspection cover were bent by employes of Morrison-Knudson, there is clearly no merit in the contention that the chain of causation was broken as a matter of law through modification or repair of the machine. Nor is there any merit in the contention that the causal chain was broken by the passage of time between the sale of the vehicle in 1949 and the failure of the hinge in 1952. Defendant relies upon a number of cases from other jurisdictions which hold that after the lapse of a certain interval of time, a chattel cannot be called imminently dangerous and no liability can be imposed upon the manufacturer for negligence. But the imminence of the danger is not determinative of a manufacturer’s liability under the law of this state. (Hale v. Depaoli, 33 Cal.2d 228, 232 [201 P.2d 1, 13 A.L.R.2d 183], and cases cited.) Where machine parts have failed and it is necessary to ascertain the cause of the failure important considerations may be the length of time the machine has been in use and the nature of its use. Earth-moving machines are not built to last indefinitely. The manufacturer is not responsible for failures due to ordinary wear and tear or misuse. There is merit in the argument that the longer a machine has been in use before it fails the more unlikely it is that the failure was due to defective construction or design. Parts that have served through their expected life cannot be regarded as defective because they eventually give out. But these obvious facts do not apply to the present case. The hinge in question was not a moving part that was subject to wear and strain *723while the machine was in operation. It was not a hinge that had been properly and efficiently welded to the deck plate and the inspection cover which gave way as a result of long continued use; it was a part that was defective in construction and which gave way as a result of use which would not have caused it to fail if it had been properly and efficiently constructed.
The second point is that, as a matter of law, the accident was not proximately caused by negligence in the design of the bulldozer.
Defendant says in its brief: “Plaintiff, through his expert witnesses, and in argument, asserted that the weld which failed could have withstood the load precipitating such failure had that weld complied with the specifications of the Caterpillar Tractor Company. Hence the design defects asserted by plaintiff—and denied by defendant—would be immaterial since there would be no proximate cause between such defects, if any there were, and the failure of a weld which would have held had it complied with the very design which plaintiff criticizes.’’ The argument is untenable. It misses the point. It assumes that the weld conformed to defendant’s specifications. It evades the fact that the asserted defect in design relied upon by plaintiff to establish design negligence consisted of the inadequacy of the deck plate lip. Defendant has not set out in its briefs the evidence relating to the design of the bulldozer and no contention is made that the jury could not reasonably have found that in the exercise of ordinary care the lip should have been extended all the way around the cover. If the jury found that the fractured weld was the original factory weld, then it could also have found that the accident resulted from two related acts of negligence on the part of defendant, namely, negligence in designing the deck plate and negligence in the welding of the hinge, for it was a reasonable inference from the testimony of Dr. Morrelli that the inspection cover would not have fallen into the aperture upon the failure of the hinge if the deck plate had been properly designed. As designed, the inspection cover could have fallen and did fall upon failure of the hinge. It would not have fallen if it had rested upon a lip upon the hinge side. Defendant chose to rely solely upon the adequacy of the welding of the hinge to the deck plate. It was a question for the jury whether a simple change in the deck plate was required in reasonable anticipation of the failure of the weld for one cause or another. The fact that it did fail shows that failure was not *724out of reason but whether failure should have been anticipated was a proper factual question. (Mosley v. Arden Farms Co., 26 Cal.2d 213 [157 P.2d 372, 158 A.L.R. 872]; Jones v. City of South San Francisco, 96 Cal.App.2d 427 [216 P.2d 25]; DeMirjian v. Ideal Heating Corp., 129 Cal.App.2d 758 [278 P.2d 114].)
The next point to be considered is that the court committed error in defining the duties of defendant with respect to both the design and manufacture of its product in a single instruction. It is unnecessary to quote the instruction since it is not challenged as a misleading or inaccurate statement of those duties. Defendant argues that since, as a matter of law, a verdict upon either of plaintiff’s two theories of liability would have been without support in the evidence, the instruction should not have been given. A sufficient answer to this contention is that we cannot say, as a matter of law, that there was no substantial basis in the evidence for a finding of negligence in the manufacture of the bulldozer or negligence in its design, or negligence in both respects.
The fourth point is that the court committed error in instructing the jury on the subject of proximate cause. Defendant criticizes four instructions which were given and urges that the court should have given three additional instructions at its request. It is not contended that any of the instructions stated the law incorrectly but only that it would have been stated better in defendant’s instructions that were refused.
We think that the challenged instructions were sufficient and that the proposed instructions were properly refused. At the request of both parties, the court gave a standard instruction defining proximate cause. At plaintiff’s request, the court gave two instructions on concurring negligence. At the request of defendant, the court gave a standard instruction on intervening cause. We have set out the instructions in the margin.1,2,3,4
*725It is earnestly contended that these instructions misled the jury to defendant’s prejudice by failing adequately to present its theory that the sole proximate cause of the accident was the unforeseeable acts of employes of Morrison-Knudson and that three additional instructions, which we have likewise set out in the margin, should also have been given.5,6,7
*726We find no merit in either contention. The instructions that were given adequately stated the law applicable to the admittedly complex question of causation to be resolved by the jury. If there is reason for criticism of the instruction on proximate cause, defendant, having requested the instruction, is in no position to criticize it. As to the proposed instructions, it appears from the record that the instruction respecting the duties of a supplier of chattels was withdrawn by defendant following a conference in chambers. The other two instructions could well have been refused upon the ground that they were argumentative and sufficiently covered by other instructions. But even if the requested instructions would have clarified the issue it is inconceivable that during the trial of many weeks’ duration the jury could have failed to understand that the manufacturer was relying upon the defense that it did not make the defective weld, that the deck plate and inspection cover were deformed prior to the accident. Furthermore, defendant requested and obtained an instruction which informed the jury that if the accident was solely and proximately caused by some other person, it would have *727no responsibility for the other’s acts and would be entitled to a verdict in its favor.
The court instructed the jury that plaintiff had received workmen’s compensation benefits from his employer’s workmen’s compensation insurance carrier. The final point on the appeal is that the court erred in rejecting an additional instruction proposed by defendant that “in the event of certain acts or omissions by his employer,’’ Darling would have had a cause of action under the Labor Code against MorrisonKnudson for serious and wilful misconduct and that his employer could not insure against liability for the same.
Defendant argues that the instruction should have been given because several of plaintiff’s fellow employes testified favorably to him and the jury should have been advised of their interest or bias arising out of possible uninsured demands against their employer. It is a far-fetched theory. Defendant has not made any showing in its briefs that these witnesses were aware of any acts of their employer constituting serious and wilful misconduct, or, if they were, that they knew additional liability for those acts could have been imposed upon Morrison-Knudson. If defendant had thought it worthwhile to question the witnesses for possible bias it could have done so in their cross-examination. It cannot be assumed now that their answers would have shown that they may have been biased because of possible liability of their employer for wilful misconduct. However, we do not imply that any instruction on the employer’s liability for workmen’s compensation was necessary or proper. The failure to give the instruction furnishes no ground for a reversal of the judgment.
The judgment is affirmed.
Wood (Parker), J., and Vallée, J., concurred.
A petition for a rehearing was denied August 5, 1959, and appellant’s petition for a hearing by the Supreme Court was denied August 31, 1959.