These are negligence cases based upon the same facts, except as to damage. They were tried together in the District Court of Maine and a verdict for the plaintiff wa,s returned in each.
For convenience the parties will be designated herein as they were below.
The defendant maintained a gasoline filling station at Caribou, in the state of Maine, at which gasoline delivery automobile trucks were loaded. Each of the plaintiffs was the owner of certain personal property stored in a storehouse within a short distance of the defendant’s filling station.
The declaration alleged that on May 31, 1920, as one of the automobile delivery trucks of the defendant was being loaded with gasoline from the tank at this station, the flow of gasoline through the iron pipe by which it entered the tank generated static electricity; that an electric spark ignited the gasoline vapor, and a stream of burning gasoline spread to the storehouse in which the goods of the plaintiffs were stored, setting it o.n fire and destroying or damaging the plaintiffs’ goods; that, previous to this day the defendant had used a safety chain, which extended from the feed pipe on the storage tank into the tank of the automobile delivery truck and conducted to the ground any static electricity which was generated by the flow of gasoline; that upon the day in question the safety chain was not used; and that, because of the negligence of the defendant in failing to use it, the delivery truck which was- being filled became charged with static electricity, and an electric spark ignited the gasoline vapor in the tank of the delivery truck.
The errors assigned are the refusal to direct a verdict for the defendant at the conclusion of all the evidence, incorrect instructions, and the exclusion and admission of certain testimony.
In support of the first assignment of error, the defendant strenuously contends that there was no evidence upon which the jury could base a verdict for the plaintiffs, because there was no evidence as to the-origin of the fire, and they could infer negligence of the defendant from other inferences only, and not from proven facts.
The evidence was undisputed that previous to the day in question the defendant had used a safety chain for three years; that it had *681become broken and was discarded; that upon this day, immediately before the attempt was made to fill the delivery truck which was set on fire, another was filled from the same tank without the use of the chain, and with no disastrous results; and it was claimed that the delivery truck itself was sufficiently grounded through its wet wooden spokes and the chains which were used upon the tires; that the ground was wet and muddy, and that therefore the tank itself, in its then condition, was a good conductor without the safety chain.
The defendant also contended that frictional or static electricity could not be generated by the flow of gasoline through a metallic pipe, and that the filling pipe, resting upon the edge of the opening into the gasoline tank and pressed firmly against its edge by its weight, furnished a good conductor and would offer no opportunity for the creation of a spark. These contentions were supported by the testimony of experts.
The plaintiffs introduced evidence tending to prove that static electricity could be developed in this way, and- that a spark would be caused by the vibration of the pipe upon the edge of the manhole of the delivery truck as the gasoline flowed through it under pressure, and that the safety chain or some other device should have been used for grounding the delivery truck.
Walter L. Wedger, for 13 years state expert on explosives for the state of Massachusetts, testified that he had made a study of the generation of static electricity by gasoline friction, and that it could be developed by the flow of gasoline through a metallic pipe; that if there is a good conductor the electricity so developed will be conducted off as fast as it is generated, but that if it is not it will accumulate; and that, if there is a gap in the connection between the pipe through which the gasoline is flowing and the tank into which it flows, a spark will be created, which will set fire to the gasoline vapor. He testified that he had investigated several hundred fires caused by gasoline and had determined these facts by experiments.
In answer to a hypothetical question he testified that in his opinion static electricity was the cause of the ignition of the gasoline vapor in this case, and that the safety chain would have conducted it away. The chain, when in use, was connected with the filling pipe at some distance back from where it entered the manhole of a delivery truck and extended down through the manhole into the tank, thus furnishing a metallic connection through which the static electricity which might accumulate in the tank could be conducted to the ground.
Mr. Wedger also testified that it was necessary for safety that such a ground connection be made and that it was required by the police and fire departments of Massachusetts when delivery tanks were being filled with gasoline.
Whether the metallic automobile delivery tank itself, because of its metallic body and wooden wheels with rubber tires, constituted a ground, was one of the questions upon which the evidence was contradictory.
The defendant introduced testimony tending to show that the wheels of the automobile tank had chains upon them, that they were *682wet and covered with mud, and that the wheels themselves sat in a rut in which there was mud and that a good connection was thus established by which any static electricity which might have been generated within the tank would be conducted to the ground.
On the other hand, there was testimony that there had been no rain upon the day in question, nor upon the previous day, and that an examination of the automobile truck immediately after the accident showed that there were no chains upon the wheels.
Plaintiffs’ expert admitted that, if the automobile truck was grounded, there would be no accumulation of static electricity. Upon the important question of whether it was grounded or not, there was conflicting testimony.
The filling pipe had upon its outer end a sleeve made of some sheet metal which was one-half inch larger than the pipe, according to the estimate of one witness. This sleeve had a crook in it, which was so adjusted that it could be pulled out -to the desired length to reach the manhole of an automobile truck. This filling pipe was connected by a swivel joint with a pipe which led back to. the storage tank, and while the defendant introduced testimony tending to show that the pipe, when it was pulled down to be used for filling an automobile truck, had such a loose connection at the swivel joint that it fell down of its own weight and rested on the edge of the manhole, there was testimony on the part of the plaintiffs that the joint was stiff, and a picture, introduced at the trial, showed the pipe, stationary within a foot of the top of the manhole of the delivery truck.
While there was evidence that the pipe rested upon one edge of the manhole 'and that the sleeve of galvanized iron, on which was the crook, after it was inserted into the manhole, was pressed against the opposite side, there was evidence that, when the gasoline flowing, under pressure, escaped from the pipe into the enlarged sleeve, it caused the sleeve and pipe to vibrate, thus furnishing a gap which would cause an electric spark, and also that the iron pipe and the sleeve were exposed to the air and were likely to become rusted. In connection with this the expert for the plaintiff testified as follows; it having appeared that the pipe was exposed to the air at least nine times a day when the sleeve was pulled out for the purpose of filling trucks:
“A. I should expect that sleeve and the pipe underneath to be covered with rust, if exposed to the air say nine times a day.'
“Q. If it was not covered with rust, would there be an electrical connection? A. I do not see how it would not be possible to be covered with rust, and yet the vibration of filling that truck would be sufficient to establish a shaking connection, not only between the pipe and sleeve, but between the edge of the manhole and the sleeve, and also where the pipe rested on the rack, and in all three of these places you would get static sparks by vibration.
“Q. Now, assuming that there are three trucks filled from two to three times a day in two different compartments, which would double the times you adjust the sleeve, and the iron pipe, one end supported back at the rack where the swivel is, the other end supported on the edge of the manhole of the truck, would not there be a connection between that sleeve and the iron pipe, the edge of which is necessarily supported in a sleeve?
“A. No; I do not think it would be a good electrical connection. I think there would be a spark jump there, and my basis for that is actual experi*683enee in Brockton in a similar case, where a spark jumped from iron to iron that was supposed to he connected, and we lost two lives on it.”
It appeared in evidence, also, that the tank held 312 gallons^ of gasoline, and that at the time of the accident it was nearly two-thirds full,- and the driver testified that, as the tank was being filled, it settled upon its springs, and in answer to the inquiry if the movement of the machine would make a spark the witness answered:
“The depression of the springs, the natural vibration caused by the flow as gasoline, would cause a depression.
“Q. Which would make these two things separate?
“A. I should anticipate that there would be a spark at that point.”
There was evidence from which the jury could find that frictional or static electricity could be generated by the forcing of gasoline under pressure through an iron pipe; that the automobile tank was not grounded; that the depression of the springs of the automobile tank and the vibration of the filler pipe and the sleeve would produce a break in the electrical connection between the manhole of the tank and the pipe or sleeve, and produce an electric spark.
It was not necessary that the plaintiffs show that a static spark was actually produced; but it was sufficient if conditions and circumstances were shown from which the inference could be reasonably drawn that a static spark was produced. It was in evidence that gasoline gives off a vapor at all temperatures above 4 degrees above zero, and that its volatility increases with the temperature, and that the day of the accident was a very hot one.
Circumstances and conditions having been shown which, if the testimony in regard to them were believed by the jury, would furnish ground for a reasonable inference that the gasoline vapor was ignited by an electric spark generated as claimed by the plaintiffs, it cannot be said that the jury were forced to base their conclusions upon inferences drawn from other inferences^ instead of proven facts.
It was' suggested that the vapor of the gasoline might have been ignited in some other way, and as a suggestion of what might possibly have been the cause of its ignition the defendant proved that the storage tank was situated within a few rods of the railroad yard of .the Bangor & Aroostook Railroad; but there was no proof that any engines were being operated in this part of the yard at the time of the fire. There was also testimony that the occupant of a house nearby had just started a fire with light burning materials, and, gasoline vapor being heavier than air and sinking to the ground, it was suggested that it might have been ignited by a spark from the chimney of this house and a backfire caused, which extended to the gasoline vapor in the tank. The conclusive answer to these suggestions, however, was that the man in charge of the automobile truck, who was upon it, testified that his attention was first attracted by an explosion within the tank, and that the first fire he saw was that bursting from it through the manhole. The jury must have found that all other means of causing the fire, except an electric spark, had been eliminated, and we cannot say as a matter of law that such a finding was clearly wrong.
*684There was substantial evidence to support a finding of negligence because of the failure to use the safety chain or some other device for grounding the automobile truck. This was strengthened by the uncontradicted evidence that in all the stations of the Standard Oil Company such a safety chain had been in use, that it had had'one fire, the cause of which experts traced to the generation of static electricity, and that they were then informed that a safety device was fiecessar-y in order to ground the automobile trucks when they were being filled*
The filling pipe through which the gasoline was flowing into the delivery truck fell to the ground, and it was some time before the flow of gasoline through it was stopped. This spread over the ground and took fire, which rapidly spread. The fire company of the town was summoned, and upon their arrival the firemen directed a stream of water upon the flowing gasoline, which, defendant claimed, diverted its course toward the storehouse where plaintiffs’ goods were stor'ed, although one of the plaintiffs testified he arrived before any line of hose had been laid and water turned upon the gasoline, and at this time the gasoline had spread under the storehouse. It was claimed by the defendant that this action of the firemen constituted an independent intervening cause, which was the proximate cause of the destruction of the plaintiffs’ goods. Upon this question the jury were instructed by the presiding judge as follows:
“As the evidence has developed, I do not think that the fire department intervention has any effect upon any possible liability of the defendant. The fire department may have been over zealous and may have spread the burning gasoline in the direction of the buildings, hut without discussing it any further I will rule to you that so far as the evidence appears here, ■the intervention of the fire department would not change the liability.”
This instruction is assigned as error.
While the question of what is proximate cause is ordinarily for the jury, yet, where the whole evidence offers no substantial dispute upon material points, it becomes a question of law for the court. McDonald v. Snelling, 14 Allen (Mass.) 290, 92 Am. Dec. 768; Eaton v. Boston & Lowell R. R., 11 Allen (Mass.) 500, 505, 87 Am. Dec. 730; Teis v. Smuggler Mining Co., 158 Fed. 260, 85 C. C. A. 478, 15 L. R. A. (N. S.) 893; Milwaukee, etc., R. R. Co. v. Kellogg, 94 U. S. 469, 474, 24 L. Ed. 256; O’Brien v. J. G. White & Co., 105 Me. 308, 314, 74 Atl. 721.
It is no defense to-an action for negligence that the negligence or trespass of a third party contributed to the injury, although such third party acted entirely independently of the wrongdoer. Eaton v. Boston & Lowell R. R. Co., supra; O’Brien v. J. G. White & Co., supra.
Upon the .record in this case there was no error in the instruction of the presiding judge.
Testimony was offered by the defendant to show what the Gulf Refining Company did in the way of precautionary measures to prevent the generation of static electricity, which was excluded, and this is assigned as error, as was also the exclusion of testimony offered to .show the practice of the Texas Oil Company. The witnesses whose *685testimony was excluded were not inquired of in regard to any general custom of other companies in regard to the use or nonuse of devices to prevent the accumulation of static electricity, but were each asked about the practices of the company with which he was connected. In regard to the Texas Company the judge ruled as follows:
“Whether the Texas Company does or does not, I exclude testimony of what the Texas Company does in relation to its business.”
In regard to the other witness the presiding judge stated:
“X have already ruled that I will not permit testimony to be introduced to show what the Gulf Refining Company do.”
One witness who had been connected with the Gulf Refining Company testified that he had visited a great many other stations besides those of his own company throughout New England. He was specifically asked “what system the Gulf Refining Company is using at its filling stations.” When this testimony was excluded, the presiding judge stated that he would permit the defendant “to show by experts thkt there are no safety devices that would be adequate, if you can.”
The common usage of a business, while it may be a test of negligence, is not a conclusive nor controlling test. Bennett v. Long Island R. R. Co., 163 N. Y. 1, 4, 57 N. E. 79; Burke v. Witherbee, 98 N. Y. 562, 566; Boyce v. Manhattan Ry., 118 N. Y. 314, 319, 23 N. E. 304. Thompson on Negligence, § 3770.
In Texas & Pacific Ry. Co. v. Behymer, 189 U. S. 468, at page 470, 23 Sup. Ct. 622, at page 623 (47 L. Ed. 905), Justice Holmes makes this cogent statement of the law:
“Wbat usually is done may be evidence of wbat ought to be done, but wbat ought to be done is fixed by a standard of reasonable prudence, whether it usually is complied with or not.”
As the testimony excluded did not relate to any general custom or practice in the use or nonuse of safety devices at gasoline filling stations, but only to the custom or practice of a single company, there was no error in its exclusion.
After the warning which the defendant had received, it would seem as if the question of whether it had exercised ordinary care, when the warning was disregarded in this case, could not be affected by the evidence of the practices and usages of other companies.
It is also assigned as error, that testimony was excluded as to other fires; but in relation to them it was stated by the presiding judge that such testimony would not be admitted, unless circumstances were shown to be the same. In the case of one fire in which a safety chain was used, the testimony was excluded because, as admitted by counsel for the defendant, “the pipe was not resting on the manhole of the tank in the case of that fire.”
In regard to this the presiding judge ruled:
“That would make it altogether different; that involves other details. 1 cannot admit testimony of other fires.”
Testimony which was offered and excluded did not relate to a fire which occurred when a safety chain was used and the circumstances and conditions were the same as those at the time of the fire in question. Upon this the court ruled:
*686“I think you ought to show the circumstances of the same. Circumstances would make a difference.”
Without offer of proof that the circumstances and conditions surrounding the other fires were substantially the same as those existing at the time the fire in question took place, there was no error in excluding testimony of such other fires.
The judgment of the District Court is affirmed in each case, with costs in this court to the defendant in error.