Action to recover damages for personal injuries, and this appeal is from the judgment and from an order denying defendant’s motion for a new trial.

The injury occurred March 20, 1890. Plaintiff was a carpenter by trade, and some two years prior to the accident was employed by the defendant in that capacity alone, but, after the completion of the buildings, he was used in making alterations and repairs when necessary, and as a general laborer, roustabout, or handy man. He was not an engineer, or machinist, or familiar with machinery at the time he entered defendant’s service, nor afterward and prior to the accident, except so far as his general employment may have given opportunity for observation. A new still or generator had been put in place, and was first attempted to be used the day of the accident. This generator consisted of a wrought-iron shell about two feet in diameter, and was made in two sections, each about ten feet long, connected at the center by a cast-iron drum, and each end closed with a heavy cast-iron head kept in place by iron rods running from end to end on the outside of the shell and through the heads and secured by nuts on the outside of the heads, by which they could be drawn and secured against the ends of the shell. These heads were rabbeted to the depth of five-eighths of an inch, corresponding to the ends of the shell or cylinder, the groove being nearly filled with lead, so that the ends of the cylinder were drawn partially into the grooves against the lead, the object being to make a tight joint. Within the generator or cylinder was a system of steam pipes, connected with the boiler in the engine room, for the purpose of heating the aqua ammonia with which the generator was filled to about two-thirds of its capacity, the steam not escaping into the generator, but passing through it within the pipes. The ammonia being thus heated produces a gas which fills the remaining space in the generator in the same manner that steam fills the vacant space in a boiler, and the pressure of the gas thus generated increases rapidly as the heat of the am- *248 mania is increased. The steam, being confined within pipes, does not add to the pressure of the gas except as it operates in heating the ammonia. According to Mr. Troutwein, a witness for defendant, the proper working steam pressure is forty-five to fifty pounds, and that produces an ammonia pressure of one hundred and eighty to two hundred and thirty-five pounds to the square inch.

On the day of the accident the generator had been charged with ammonia for the first time.

The steam pressure being used at the time of the accident appears to have been twenty pounds, and the ammonia pressure is variously stated at from one hundred and seventeen to one hundred and twenty pounds, as shown by the gouge. A working pressure had not yet been reached; but at this stage it was discovered that the generator was leaking at the drum in the center of the generator, and the plaintiff and another employee, who had no more knowledge or experience, were called upon by the engineer, according to plaintiff’s testimony, “to get a socket wrench and a bar and set up on those bolts or nuts and stop that leak,” and while they were doing so, or had just ceased work on it, the generator exploded, and the plaintiff was seriously injured.

The jury returned a general verdict for the plaintiff, and also answers to certain special issues submitted to them. It is not claimed, however, that the special issues required a different judgment, but it is strenuously insisted that the evidence does not justify the verdict. It would require more space than can be devoted to this opinion to discuss fully, and in detail, the many points made by counsel for appellant on this branch of the case. The controlling questions only can be noticed.

It is urged that plaintiff’s employment was general; that he worked at anything he was called upon to do, and that this included the work of tightening up the nuts on the generator, and that he was not inexperienced in that particular work. Conceding that, in a sense, his employment in course of time became general, the evi *249 dence is uncontradicted that he was wholly inexperienced in the operation of machinery,- except so far as he may have acquired it during his employment by the defendant; but there is no evidence tending to show that he understood or appreciated the expansive force of steam, or of the gas generated from ammonia, or the principles involved in the construction of generators, or the care required to preserve an equal strain upon all its parts. Whether he had, prior to the time of the accident, assisted in this particular work is a matter about which the evidence is' strongly conflicting; but the jury found upon one of the special issues that “the explosion was caused by allowing inexperienced men to tighten the nuts on the bolts with an increasing ammonia pressure.” Even if plaintiff had before this been required to tighten such bolts under similar circumstances, the operation being conceded to be dangerous unless skillfully done, his skill and experience would not thereby be necessarily established. These qualifications require something more than knowing how to handle a wrench. He must be able to determine where the pressure is least, and when he has secured an equal pressure upon all the bolts, then, if further tightening is required, he may turn each uniformly a little, so as not to disturb to any appreciable extent the equal pressure secured. If he is simply told to turn each a little, and go round the whole number, he may commence where the bolts are tightest, and thus increase the already unequal pressure, while strictly obeying instructions.

Whether the plaintiff had ever tightened the bolts before is disputed. Richards, one of the engineers, says he did not set plaintiff and Richmond at work tightening the bolts, but saw them at work at it. Dempsey, another of the engineers, testified that he did not exactly remember whether he set' the men at work at it or not; that he saw them at work at it; that he did not pay any particular attention to how they proceeded. “ Of course they understood how to do it. They had been instructed before. Every man that had ever *250 tightened had always been instructed to turn the nuts first one some and then others, and keep going round and round.” It is not clear from defendant’s testimony that they received any instructions at that time; and whether the plaintiff had ever received instructions as to the mode of doing the work must depend upon whether he had ever done it before. Plaintiff denied that he had done so while the generator had pressure on. It is clear that he was not an engineer, and was not employed as such. The work was dangerous and required skill, and should have been performed by the engineers. No man, however skilled and experienced, can tell where the bolts are slack or tight unless he has hold of the wrench, and therefore cannot tell whether the work is being properly and safely performed though standing by and directing it; for, though the tension of the rods or bolts may have been equal before heat was applied, it is by no means certain that the expansion of them, when heated, would be equal. Whether the tightening could be done with reasonable safety without first reducing the pressure was strongly controverted in the evidence, but there was no controversy that doing it without reducing the pressure .required care and skill.

Appellant also contends that the accident occurred from an unforeseen and unaccountable cause which cannot with certainty be traced to the negligence of defendant, and that therefore no relief can be granted the plaintiff.

In one aspect of the evidence the explosion seems difficult of satisfactory explanation. The generator was manufactured at Carbondale, Pennsylvania, and was there tested with water pressure, and also with air pressure, up to three hundred and fifty pounds to the square inch, and was calculated to sustain a pressure five times greater than the working pressure required of it. The day before the accident, according to the testimony of engineer Richards, they subjected it to an air pressure of one hundred and sixty pounds, and according to the *251 testimony of Mr. Hendrick one hundred and thirty pounds. Just before the explosion the gouge showed a pressure of one hundred and twenty pounds according to Richards. A given number of pounds pressure, whether produced by gas or condensed air, would make no difference in testing the strength of the cylinder or generator, though possibly, if a break occurred, the expansive qualities of the gas might produce more serious results. If, therefore, the generator were in all its parts in the same condition at the time of the accident that it was in the day before when tested with a greater pressure, it would seem impossible that it should have given way. The solution must therefore be found in what was done to the generator at the time of the explosion. All the expert evidence shows that the danger of doing such work while the generator is under pressure lies in disturbing the equality of the strain upon the parts; that is, in making one part sustain more than its just proportion of the pressure; but as to whether it is prudent or reasonably safe to do the work in question while there is a pressure, there is a serious conflict. It is true, as appellant contends, that the mere happening of an accident, the attending circumstances being unknown, is not evidence of negligence on the part of the owner of the machinery or appliance: but, in this case, the tightening of the nuts and the explosion concurring in point of time, we are not authorized to say that the jury could not properly infer that the procuring cause of the explosion was the tightening of the nuts. The explosion did occur. A cause—the tightening of the nuts—which might have produced it, if unslcillfully done, concurred; while the prior test to which the generator was subjected would seem to exclude all other supposable causes.

It is said, however, that the tightening of the nuts and the explosion did not concur in point of time; that plaintiff and Richmond had completed their work. But this statement is not quite accurate. It is true that, after they had been at work some time in tightening *252 the nuts they stopped and stepped aside to see whether the work they did affected the leak, and that at that moment the explosion occurred. But the fact that they were not at work turning up the nuts at the moment, or within a minute or two or three minutes, of the explosion, does not affect the question. The steam had not been turned off the pipes, nor had the steam pressure reached its working point, so that the heating power of the steam was increasing, and the pressure of the gas generated by the heat was also increasing; but this small increase in the pressure could not nave affected the generator had it been in the condition it was in the day before when it sustained a much greater pressure, and hence this extremely short interval between the suspension of the work and the explosion does not affect the theory adopted by the jury, that the explosion was caused by the tightening of the nuts by inexperienced men.

But counsel for appellant, in their very able brief, further contend that, in order to establish negligence upon the part of the defendant, it must not only appear that the explosion was caused by the agencies and in the manner above indicated, but that such result should reasonably have been foreseen by the defendant; “that a party is not to be held responsible for injuries which could not reasonably have been foreseen or ekpected as a result of his misconduct.”

There was evidence tending to show that it was unsafe for a skilled engineer or machinist to tighten up these nuts while the pressure was on, though it may be conceded that the preponderance of the evidence was that it could be done by skilled men with reasonable safety; but there was little conflict in the evidence tending to show that it was dangerous for unskilled or inexperienced men to do it. In order to fix the liability of the master, it is not necessary that he should know that injury will inevitably follow, for that would exhibit express malice or intent to injure; nor that he knew or believed that it would probably occur, for that would be recklessness scarcely less criminal. This court, in Ingerman v. Moore, *253 90 Cal. 421, 422 , 25 Am. St. Rep. 138 , quoted with approbation the following from Jones v. Florence Min. Co., 66 Wis. 277 , 57 Am. Rep. 269 : “We think it is now clearly-settled that if a master employs a servant to work in a dangerous place, or where the mode of doing the work is dangerous and apparent to a person of capacity and knowledge of the subject, yet if the servant employed to do work of such a dangerous character or in a dangerous place, from youth, inexperience, ignorance, or want of general capacity, may fail to appreciate the dangers, it is a breach of duty on the part of the master to expose a servant of such character, even with his own consent, to such dangers, unless he first gives him such instructions or cautions as will enable him to comprehend them, and do his work safely with proper care on his part."

The instructions given plaintiff by the engineer have already been considered; but, in addition to what has been said, it may be added that it nowhere appears that he was informed that the work was dangerous, or that his safety in any degree depended upon the manner of doing it, or the care with which it was done.

What has been said covers appellant’s first and second points. Its third point is that the verdict is against law; but, if we are right in our conclusions upon the propositions above discussed, this contention cannot be sustained. In support of it, however, appellant calls attention to the special finding No. 8. That question and the answer of the jury thereto are as follows: “ Did plaintiff and Richmond tighten on one nut, and then pass to the next, tightening a little on each nut as they went along; if not, what does the proof show in respect to tightening the nuts? State how they tightened. A. Yes; but was ignorant of the manner in which the same should be tightened.” What has been said heretofore in reference to these instructions, and the manner in which the work should be done, is sufficient to show the meaning of the answer given by the jury.

It is also claimed by appellant that the first and sec- *254 and instructions given by the court of its own motion, and instructions 1 and 3, given at plaintiff’s request, violate that provision of the constitution which declares that “judges shall not charge juries with respect to matters of fact, but may state the testimony and declare the law.” (Const., art. VI, sec. 19.)

The objection to these instructions is not well taken. The court instructed the jury that, if they found, by a preponderance of the evidence, that certain enumerated facts existed, they should find for the plaintiff; but that, if the plaintiff failed to prove any of said facts by a preponderance of the evidence, their verdict should be for defendant. The enumerated facts were within the issues, and as to which evidence had been given. No reference was made to the evidence as sustaining, or tending to sustain, any fact.

It is also insisted that the court erred in refusing to give certain instructions requested by defendant.

These instructions were to the effect that Richards and Dempsey, the engineers, were fellow-servants of the plaintiff; that, if defendant had exercised proper care in selecting them, and the plaintiff was injured through their negligence, they and the plaintiff being engaged in the same general business, the defendant was not liable.

The facts did not justify such an instruction. E. E, Hendrick was superintendent and general manager of the corporation defendant. He testified that Richards ‘‘ was an engineer, and had charge of the refrigerating machine under my direction when I was there, and had sole charge of running the refrigerating machine in my absence, and hired and discharged men under my direction.” Mr. Hendrick had been at the works directing the starting up of the refrigerating machine— the generator of which exploded—and was engaged about it until about half an hour before the explosion, when he left, and went up town for some fittings, and did not return until after the explosion. The testimony of Richards and Dempsey, the engineers, was not clear *255 as to which called the plaintiff to do the work; but it is clear that Richards saw him and Richmond doing the work, and did not object; while the plaintiff testified that it was Richards who called them to do the work. Richards, therefore, had the same power to control the men in Hendrick's absence that Hendrick had when he was present.

The remaining instructions requested by defendant are fully covered by the general discussion, and were properly refused, and the motion for a nonsuit was properly denied.

The judgment and order appealed from should be affirmed.

Belcher, C., and Britt, C., concurred.

For the reasons given in the foregoing opinion the judgment and order appealed from are affirmed.

McFarland, J., Temple, J., Henshaw, J.