Opinion bt
Mr. Justice Potter,The plaintiff in this case seeks to recover damages for the death of her husband, who was killed in an explosion at the defendants’ factory on November 27,1901. Richard Bardsley, the husband, was employed by the defendants as a night engineer, performing, among other duties, that of caring for an oil pump which drew fuel oil from a room in which it was stored, in the cellar of the factory, to the furnace room where it was used as fuel for melting glass.
*59On the night of the explosion the flow of oil from the tank became in some way obstructed. The matter was reported to Bardsley, the engineer, and he went to the oil room to ascertain the source of the trouble. A few moments later the explosion occurred, and the night watchman found Bardsley staggering about in the oil room, where he soon fell dead. The fireman who came in a little later, testified that he found Bardsley’s body lying over a box or barrel, with his head facing away from the oil tank. Iiis clothing was scorched. McCormick, who was a companion of Bardsley, was also injured by the explosion, and could give no clear version of the accident beyond the mere fact that it happened. While he would not state positively that Bardsley was carrying a light when the accident happened, yet he thought so. There was positive evidence that there was no fixed light in the room where the explosion occurred. In the daytime some light was afforded by a small door or shutter, which could be opened; but even this was usually closed. There was evidence that defendants supplied their employees with candles or lamps to use about their work. There were three tanks in the oil room, one of which apparently was never used. The larger of the two remaining tanks had a capacity of about 2,800 gallons of fuel oil. It was sunk in the floor, and in its top there was a manhole through which passed an oil pipe and two steam pipes. The tank contained a steam coil used for heating the oil when it became sluggish in cold weather. There was evidence that the general nature of fuel oil is to give off explosive gases at any temperature, and that heating the oil increases the volume of gas produced.
The negligence charged was, in substance, failure to provide safe conditions for the employees to work in and about the oil room. Specifically, defendants were charged, among other things, with improper construction, arrangement and equipment of the tank and oil room, in these respects: (a) No covering provided for the manhole of the tank, (b) Lack of usual and customary devices to indicate quantity of oil in tank. (e) Lack of usual and customary devices to indicate temperature of oil -as affected by the steam pipes used for heating the oil. (d) Lack of adequate means of ventilation for carrying off gases. In addition, they were charged with failure to sup*60ply Bardsley with a safety lamp, and with violation of duty in not notifying Mm of the dangerous condition of the room. Evidence tending to sustain these charges was offered by plaintiff. The defendants offered no testimony, and asked for binding instructions at the close of the evidence for the plaintiff. The trial judge refused the request and submitted the facts and inferences to be drawn therefrom, to the jury in a very careful and discriminating charge.
The first group of assignments, as arranged by counsel for appellant, complains of the admission of evidence as to what other employees did, and as to what instructions they received, and as to the condition of the plant, at other times. Some latitude must be allowed in such cases as this, where much of the evidence was unattainable, by reason of the explosion. The particular duty of the employer in any given case, is to be ascertained from the nature of the employment, and the location and condition of the premises, and the particular facts of the case. It was proper that the jury should be informed as to the methods pursued by the defendants in handling this dangerous kind of fuel. “ A higher degree of care and vigilance is required in dealing with a dangerous agency than in the ordinary affairs of life or business: ” ICoelsch v. Phila. Co., 152 Pa. 355.
In the second group of assignments complaint is made of the admission of the testimony of the witness Ohesshire. He was offered as an expert on the subject of fuel oil, and his knowledge and long experience with oil would seem to qualify him admirably. He testified, in substance, that fuel oil is a byproduct of a crude oil that is produced in the manufacture of other oils. Fuel oils are black and very oily, having a beaume gravity of 29 to 35, with a fire test from 175 degrees to 300 degrees F. It is the general nature of these oils to give off gases, at all temperatures, but, if heated, the volume is increased. These gases are explosive only when brought in contact with an open flame. In answer to a hypothetical question, he stated that defendants’ method of storing the oil was not the customary method. It is usually stored in tanks in the open air with ventilating shafts, and the amount and temperature of oil is determined by gauges. The witness stated that he knew of some tlfirty or forty industrial institutions using *61fuel oil. He was of the opinion that the room, under the evidence as set forth in the hypothetical question, was not property ventilated ; that the manholes should have had lids with vent pipes leading to the open air; that there should have been at least two windows, always open, withj a flue to the outer air; that there should have been an oil gauge outside of the oil room, and that there should have been a fixed electric light with a vapor-proof globe, and that open lights should have been forbidden.
We think this evidence was entirety relevant, and very much to the point, as tending to show the ordinary usage and proper method in dealing with oil, for fuel purposes, as contrasted with the method used by defendants. This answer applies also to the third group of assignments, in which complaint is made of the admission of evidence as to the methods employed in other establishments. To show those methods, employees of certain glass firms were called. These men knew nothing of general customs, but testified as to the methods employed in their particular establishments. In all of these the tanks were ventilated by vent pipes or hoods connected with the outer air, with tight coverings over the manholes. One testified that at his establishment there was a glass gauge to determine the quantity of oil, without going into the room. The other two witnesses stated that their tanks were apart from the main factory buildings. These methods were in direct contrast to that pursued by defendants.
It is true that no single instance is sufficient to prove a custom, but custom is certainty made up of an aggregation of different instances. It was a question of fact for the jury to say what the general custom was, after the evidence of the methods in use in various concerns was before them. The question was, as to the usual and proper method of storing oil, and rendering it reasonably safe to the employees, when used as fuel; whether in glass factories or in other factories did not matter. It was the storage and care of this class of fuel oil which was under consideration. There was evidence that the explosive gases emitted from fuel oil rendered it a more or less dangerous commodity under any conditions, so dangerous that the proper practice was to store it apart, and in a separate place. But if this was not done, defendants were at least *62bound to take such ordinary and reasonable precautions, as were in customary use in other factories using oil for fuel.
We see no merit in the fourth group of assignments, in which error is alleged in the general charge of the court. The question of ventilation, to carry off dangerous gases, was most important, as bearing upon the measure of duty required of the employer, and was properly put before the jury. So also was the fact that under the evidence, the gases would explode onljr when brought into contact with flame, and it therefore became a question for the jury whether failure to furnish illumination for the oil room other than that from the open flame of a lamp or candle, was negligence. The same thing is true of the failure to provide a tight manhole cover to the tank, to confine the explosive gases, and prevent them from spreading into the room.
Without burdening the opinion with further details, it is sufficient to say that we think the evidence, which was uncontradicted, was sufficient to justify the jury in reaching the conclusion that the defendants were negligent in failing to provide the ordinary and well-known facilities in common use in storing fuel oil, for the protection of the workmen engaged upon the premises, in the performance of their duties. And that the jury were justified in drawing the inference that as a consequence of the neglect, the explosion followed. The evidence was circumstantial, it is true, but that is sufficient in such cases to sustain the burden of proof. The inference of negligence upon the part of the employer is not confined to direct evidence alone: Hughes v. Fayette Mfg. Co., 214 Pa. 282. The facts proven were, that fuel oil generates explosive gases at all times, but especially when heated ; that in this case the fuel oil was heated; that the open manhole in the oil tank permitted the gas to permeate the room; that contact with flame would explode the gas; that there was no permanent light of any kind in the oil room; that lamps and candles wrere used habitually by the employees in entering the room, and were provided for that purpose; that the witness who accompanied the deceased into the room stated that he thought Bardsley had a light, but, owiDg to his own injuries received at the time, he could not be positive. From these proven facts, we think the jury were entitled to draw the in*63ference that Bardsley did what he was expected by his employers in case of need to do; that is, he entered the oil room with an open light, and that this light, coming in contact with the gas, caused the explosion. The inference that the accident happened in this Avay was certainly reasonable and natural.
We do not regard the evidence of alleged contributory negligence upon the part of Bardsley as sufficient to justify a binding instruction to the jury to find for the defendants. This was also a question for the jury, and the rights of the defendants in this respect were carefully and sufficiently guarded by the trial judge in his charge. The jury were instructed that if Bardsley Avas negligent in the slightest degree, in any matter contributing to the accident, there could be no recoArery.
The defendants did not see fit to offer any testimony, but if as is argued, Bardsley took a light with him into the oil room, he Avas only making use of that which had been provided for him, for that very purpose, by his employers. “ Servants may assume that all instrumentalities are fit and suitable for the use to which the master applies them, and that they are properly adjusted to each other: ” Shearman & Redfield on Negligence, sec. 217. As was said in Durst v. Steel Co., 173 Pa. 162: “The master owes-certain duties to the employee, and among these is to furnish a reasonably safe place to Avork, and to notify the employee of any latent danger of which the employer has knowledge, or Avith reasonable care would know. These duties cannot be avoided.” Of course, if Bardsley had known that the oil room was filled with gas or fumes from the oil,'and had then gone in with an open lamp, or lighted a match, he would have assumed the risk. But there is no evidence of that. He had the right to rely upon the intelligence and faithfulness of his employers in providing reasonably safe methods of storing the oil, and he cannot be charged with contributory negligence because he made use of the facilities supplied to him for thak'purpose in his work. We do not see that the defendants have any just cause for complaint as to the admission of evidence, or as to the manner in vvhich this case was submitted to the jury.
The assignments of error are overruled, and the judgment is affirmed.