Opinion by

On January 13, 1919, John W. Miller, Jr., plaintiff’s deceased husband, was a passenger on one of defendant’s local trains, and, as a result of a collision between it and another of defendant’s trains traveling in the same direction on the same track, received “a very small insignificant looking cut [skin deep]......across the bridge of the nose.” The wound was “healing slowly [the slowness probably being due to his age, which was seventy years], but it had never entirely healed,” when, on February 16, 1919, “symptoms of erysipelas” developed, which also “had practically cleared up” when, on March 1st or 2d, 1919, “he commenced to have a trouble in his chest, and......developed a pleuro-pneumonia,” from which he died on March 9, 1919. Plaintiff averred in her statement of claim that “the erysipelas and the pneumonia and his death were all the direct result of the injury and blow received upon his nose” at the time of the accident.

At the trial the case was submitted to the jury, which rendered a verdict for defendant, and from the judgment entered thereon plaintiff appeals, alleging as error only the following extract from the charge to the jury : “The germ [of erysipelas] may be brought into contact with the membranes of the body in a variety of ways. It *332is my view of the law with regard to this case, that the 'burden rests upon the plaintiff of .showing-ymi..l>y_eyi*-dence'fliat the germ of erysipelas was implanted in this gentleman’s body at the time of the accident. If you cannot find that the germ was implanted at that time, then the connection of the accident with the death is not established.” This is not an accurate statement of the law. It is a well known scientific fact that even a slight injury to the structure of the body may render it liable to attack by disease germs, from which it ordinarily would be immune, but which, after the injury and owing thereto, it is not able to resist. Since it is impossible to see a germ, and hence no one can state, with any degree of certainty, that it “was implanted at the time of the accident,” as distinguished from another date reasonably near thereto, the ruling complained of would relieve a defendant from liability though but for its negligence plaintiff would not have been afflicted by the disease. No principle of law, and certainly no public policy, requires us to so hold; and hence we once again decide it is sufficient in such cases “to show with reasonable certainty that the [disease from which death resulted] was caused by the injury received,” and that “there was an unbroken connection in the series of events between the wrongful act” and the death, without “any intermediate cause, not attributable to the original wrong”; and, in considering this question, the period of time between the injury and the death, while an important factor, is not conclusive: McCafferty v. Penna. R. R., 193 Pa. 339; Temme v. Schmidt, 210 Pa. 507.

The error complained of, is, however, a harmless one, since the court below should have given binding instructions for defendant. In addition to the injury to the nose, decedent’s physician, called by plaintiff, testified that, at the time of the accident, and continuously for at least six months prior thereto and until his death, decedent had an enlarged prostate gland from which *333cystitis developed. Assuming that from the evidence the jury might have found the erysipelas resulted from the’ injury to the nose (though this is gravely doubtful), there remains the further question as to whether or not plaintiff carried the burden of establishing with reasonable certainty that the pneumonia resulted from the erysipelas; and here her case failed. The evidence shows that the kind of pneumonia which caused the death is produced by a streptococcus germ which travels through the blood vessels to various parts of the body, in this case to the lungs. This germ might have come from the facial erysipelas; but, as plaintiff’s medical expert testified, it might also have been “produced by the cystitis [and] could be transmitted to the lungs in the same way through the blood vessels”;. decedent “might [have received] his infection from that point ......[from] his bladder, from his prostate or his kidneys” ; there is no way of determining “which of the organisms was the cause of this particular pneumonia” except by a test of the blood, which was not had. True, the family physician said that if decedent “had not had erysipelas he would not have had pneumonia, probably,” but he refused to testify as an expert, and said “I am only saying what I think as the consequence of events,” “it looked as though it was just a sequence of events that took place there”; that is, because erysipelas existed and pneumonia followed, he thought “probably” the former was the cause of the latter. But, as stated, decedent also had cystitis at the same time; in the “sequence of events” pneumonia followed it, and hence by this method of reasoning, for that is all it is, the witness might just as well have concluded that cystitis and not erysipelas was the cause. The same thing is true of the testimony of the medical expert, who, though hedging himself about by the “might have beens” above quoted, drew a like inference from the “close sequence” between the “practically cleared up” erysipelas and the pneumonia, though there was even a closer sequence be*334tween the unabated cystitis and the pneumonia. The jury, therefore, had no real.evidence from which -to-decide between the two possible causes, and, hence, the case is squarely within the long settled rule of law, which was stated again in Bruggeman v. York, 254 Pa. 430, 435: “It is not enough for plaintiffs to show the injury complained of was due to one of two or more possible causes if defendant was responsible for but one of them. They must go further and show the latter cause was the proximate one from which the injury followed as the natural and probable result, and that the events leading up to the injury are linked in succession together as a natural whole without intermediate cause not attributable to defendant’s wrongful act.”

It follows that binding instructions should have been given to the jury, so far as defendant’s liability for the death is concerned; and since, under the Act of April 15, 1851, P. L. 674, plaintiff’s right to personally maintain this action depended on establishing the fact that her husband’s death was “occasioned by unlawful violence or negligence,” the negation thereof necessarily also defeats her right to recover the expenses attendant upon his illness and burial; and this obviates the necessity of considering whether or not, if successful on the main question, she could have recovered them in this form of action.

The judgment of the court below is affirmed.