MARTIN, J.

This action was commenced May 26, 1892. Its purpose was to recover for personal injuries sustained by the plaintiff in consequence of an explosion which occurred in the defendant's office in the city of New York on the fourth day of December, 1891. There is no allegation in the complaint, nor was there any proof upon the trial, which even tended to show that the defendant was in any way responsible for the explosion which was the cause of the plaintiff's injury.

The evidence disclosed that a stranger, whose name was subsequently found to be Norcross, called at the defendant's office December 4, 1891, at about ten minutes past twelve o'clock, said he desired to see the defendant in relation to some railroad bonds and had a letter of introduction from Mr. Rockefeller. When asked to send it to the defendant, he stated that he preferred to present it in person and that he only wanted to say two or three words. Upon receiving this message, the defendant stepped from his private office into the ante room, went to the window and looked into the lobby, where he saw Norcross sitting upon a settee. At that time the defendant met the plaintiff, who said he had a message from Mr. Bloodgood, and the defendant thereupon turned the knob of the door and the plaintiff passed into the ante room of the office. The former then spoke to Norcross, who instantly arose, took his carpet bag in his left hand, and, approaching him, handed him a letter which was supposed to be from Mr. Rockefeller, which he took, opened and read. It was a typewritten communication, the substance of which was: "The bag I hold in my hand contains ten pounds of dynamite. If I drop this bag on the floor, the dynamite will explode and destroy this building in ruins, and kill every human being in the building. I demand $1,200,000, or I will drop the bag. Will you give it? Yes or no?"

The defendant read the letter twice, folded it, handed it back to Norcross, and then commenced parleying with him, stating that he had an engagement with two gentlemen, that he was short of time, and if it was going to take much time he wanted him to come later in the day. Norcross, after a second, said: "Then, do I understand you to refuse my offer?" to which the defendant replied, "Oh, no, I don't refuse your offer. I have an appointment with two gentlemen. I think I can get through with them in about two minutes, and then I will see you." Norcross held the bag at the end of his fingers, walked backwards towards the door through which he came, and when he reached the threshold he stopped and looked at the defendant. The defendant stepped back a little towards the desk that was in the ante room, while Norcross was going the other way. As he reached the threshold he looked at the defendant and said: "I rather infer from your answers that you refuse my offer," to which the defendant answered: "Is there anything in my appearance that would cause you to think that I would not do as I say I would?" and repeated that he had an appointment with two gentlemen, and that he could get through in about two minutes and would then see him. Norcross then gave one look, stepped to one side, when the flash came and it was all over in two seconds. In backing down the room, the defendant came to the desk, and was partially sitting upon the edge of it when the explosion occurred.

After the explosion it was found that everything in the office was wrecked. The partitions, floors, joists, plaster, desks, tables, chairs and other furniture were destroyed, the window sashes and window frames were blown out, even in the private office; Norcross was blown to pieces, and Norton, one of the clerks in the defendant's office, was hurled through the window to the street below, where he met his death. A steel safe which was locked and stood in an adjoining room was blown open, its contents scattered upon the floor with other debris, and every person who was in the room was either killed or seriously injured. Indeed, the explosion was so violent, so general and so destructive in its effect that it seems little less than miraculous that any person who was present should have escaped with his life. This portion of the transaction is undisputed in any essential or material particular.

The plaintiff claims that upon entering the office he passed the defendant and Norcross, who were conversing in the lobby near the door of the ante room; that he entered the ante room, which was about eight by sixteen feet, went to a table or desk near the center of the room, where he stood waiting for the defendant with his back to the door, looking towards Mr. Norton who stood by the ticker at the window looking out on Rector street; that while he stood there he once or twice glanced over his shoulder, saw that the defendant was inside the ante-room door, and that Norcross was just outside; that he heard nothing said, said nothing himself, and saw no paper in the defendant's hand; that he turned and looked towards the window with his back to the defendant, when the latter suddenly came in range of his vision on his left side, came over and placed his hand on his shoulder; that afterwards he dropped his left hand and took the plaintiff's right hand in his and gently moved him over towards the direction in which he stood which was from the plaintiff's right to his left, and that he gently moved him about the width of his body, about fifteen inches, or probably more. He then testified: "I changed my position towards Mr. Sage about fifteen inches. I changed my position in his general direction, but in front of us. I still kept my position as far as Rector street was concerned and the door of the entry. I had my back to the door all the time. I was in a line between Mr. Sage and Mr. Norcross. * * * Mr. Sage rested one thigh on the corner of this table, and then said over my shoulder to this stranger: `If I trust you why can you not trust me?' or, `If you cannot trust me I cannot trust you,' or words in that general line and to that effect, and then the explosion immediately followed."

Upon cross-examination he testified: "I saw him (Mr. Sage) come within the range of my vision to my left. I can safely say that without looking at me he put his hand on my left shoulder, put his left hand on my left arm, and took my left hand in his left hand. It was not quite at that moment that he sat down on the corner of the desk. He did not let go of my left hand with his left hand after taking my left hand. He did not take my left hand in both hands at that moment; he did a moment later, and then he sat down on the corner of the desk with my left hand in both his hands. My hand was not held specially tight; it was covered by both his hands. At that time my position was changed from where I stood when he put his left hand on my shoulder. I was conscious at the time of force being used upon me sufficient to move me; I was conscious of force being used upon me to a certain extent. In a sense it was imperceptible, and in a sense it was not. I spoke of it as being a very gentle movement. I don't think I said it was so gentle as to be imperceptible. I didn't say that. I said it was gentle. It was not violent. I don't think I said it was imperceptible. The whole change of my position was about the width of my body. Should think eighteen inches towards my left."

He also gave evidence to the effect that he had previously testified that the defendant did not use any force upon him, and he never thought of such a thing as that until after the explosion; that he did not think he was conscious that the defendant was pulling him at the time, and he could not say that he was exactly conscious of any force of Mr. Sage's hands in moving him; that he was moved easily and without resistance; that he moved voluntarily because he offered no resistance; that he did not think he was conscious of being pulled at the time, and that that testimony was true.

As to this part of the transaction the defendant testified that when he reached the corner of the table the plaintiff was about four feet from him towards the partition, and that they were in that position when the explosion occurred. He denied that he ever had his hands upon the person of the plaintiff in any manner whatever until after the explosion; testified that at the time the plaintiff was not between him and Norcross for an instant, and that he did not at any time intend or design interposing the body of the plaintiff between himself and Norcross; that he did not put himself behind the plaintiff, and that no portion of his body was behind the plaintiff; that he did not touch him at all, and made no such statement to Norcross as was testified to by the plaintiff; that, after the explosion, they were found thrown together, and that he lifted the plaintiff, which was the first time he had his hand upon him. The evidence of the defendant was corroborated in most of its essential particulars by the testimony of Frank Robertson, who was in the office at the time. There was also other proof which tended to corroborate him, and which was in conflict with the theory and testimony of the plaintiff.

This case has been tried four times, and passed upon three times by the intermediate appellate tribunal. The history of this litigation has shown an evolution in the law held to be applicable to it which is somewhat unusual. Upon the first trial, the complaint was dismissed by the trial court upon the ground that the plaintiff had failed to establish any proper connection between the act of the defendant and the independent act of Norcross which caused the injury. In other words, it held that, under the principles of law applicable to the subject, the acts of the defendant were not shown to be the proximate cause of the plaintiff's injury. That judgment was reversed by the General Term of the Supreme Court, which in effect held that no question of proximate cause was involved; that if the defendant put his hand upon or touched the plaintiff, and caused him to change his position with an intent to shield himself, he was guilty of a wrongful act towards him; that if the plaintiff was injured by the happening of the catastrophe, the burden of proof was upon the defendant to establish the fact that his wrongful act did not in the slightest degree contribute to any part of the injury which the plaintiff sustained by reason of the explosion, and that it was not necessary for the plaintiff to show that he would not have been so severely injured if he had been left standing in his original position.

Upon the second trial the plaintiff had a verdict. The court seems to have charged the jury in accordance with the principles laid down by the General Term upon the first appeal. Upon that trial, however, the defendant's counsel requested the court to charge: "If the jury find from the evidence that the defendant did take the plaintiff and use him as a shield, but that this action was involuntary, or such as would instinctively result from a sudden and irresistible impulse in the presence of a terrible danger, he is not liable to the plaintiff for the consequences of it." That request the court refused, but added: "I will charge it; that the essence of the liability must be a voluntary act." Upon the second appeal that question having been thus sharply presented, the General Term again reversed the judgment, upon the ground that the court erred in refusing to charge that request.

Upon the third trial the jury disagreed.

Upon the last trial, which is now under review, the trial court disregarded the former decisions of the General Term, and charged that the measure of the plaintiff's damages was the difference between those he actually sustained, and such as he would have received if he had not been interfered with, and that the burden of proving such damages was upon the plaintiff. It also charged that, if the defendant involuntarily put his hands upon the plaintiff in a moment of great excitement, confronted with immediate and serious danger, without meaning to interfere with him, he would not be responsible. But it submitted the question whether the act of the defendant was deliberate and intentional to the jury, calling its attention to the fact that the defendant testified that "he was in perfect possession of his senses, recollected everything that was done, that everthing he did there was done intentionally," and then charged that if, under those circumstances, he voluntarily put out his hand and touched the plaintiff, a cause of action was made out, and the plaintiff was entitled to a verdict. In this portion of its charge the court assumed, and stated to the jury, that the defendant testified that everything he did was done intentionally, as proof of his having intentionally interfered with the person of the plaintiff. The propriety of this portion of the charge will be subsequently considered.

On the trial George Baillard was called as a witness, and testified to having seen some one in O'Connell's drugstore whom he believed to be the defendant; that some one stepped up to him and asked if he was injured very much; that he answered that he was not; that he understood him to say something about being protected, or a protection that he had had from the explosion. This evidence was objected to, the objection was overruled and the defendant excepted. At the conclusion of the plaintiff's case, and again when the entire evidence was closed, the defendant moved that this testimony be stricken out. He also asked that the jury be instructed to disregard it. These motions were denied, and the defendant excepted.

When the plaintiff's evidence was closed the defendant moved for a nonsuit. Again, at the close of all the testimony, he moved for a nonsuit; that the plaintiff's complaint be dismissed, and that a verdict be directed in his favor upon sufficient grounds so that the exceptions of the defendant to the denial of those motions fully raise all the questions which are involved or have been discussed upon this appeal. At the conclusion of the charge the defendant excepted to portions of it, requested the court to charge certain propositions which were refused and by exceptions to those rulings again raised the questions involved. Therefore, in the consideration of the legal questions presented, it must be assumed that they were properly raised, not only by a motion to nonsuit and to dismiss the complaint, but also by motions to direct a verdict for the defendant and by exceptions to the charge and the refusal of the court to charge as requested.

Upon an appeal from the judgment entered upon the verdict rendered at the last trial, the Appellate Division obviously intended to follow the previous decisions of the General Term so far as they related to the plaintiff's right of recovery, to the end that the question of the liability of the defendant, under the facts and circumstances proved, might be properly presented to this court, and did not assume the responsibility of passing upon the correctness of the previous decisions in that respect. It, however, discussed many of the questions raised by the defendant's counsel, and among other things attempted to show that there was sufficient evidence to justify a jury in finding that there were two ascertained lines of direction which the explosion followed, and that one of them was in the direction where the plaintiff claims that he and the defendant stood. But when we refer to the evidence bearing upon that question and which is said to justify that conclusion, we are unable to discover its existence in the record. Indeed, in the discussion following, which involved the improbability of the plaintiff's theory that the defendant drew him in front of himself as a shield, when examined in the light of the injuries to the defendant and their location upon his person, the learned judge delivering the opinion advanced as an argument that it was not impossible that "in the titanic whirlwind of the explosion" those injuries might have occurred. He said that there was nothing more extraordinary in that incident than in the circumstance that the force of the explosion blew open a large steel safe and scattered its contents about the room, adding, "No one can account for the eccentricities of such an occurrence. If there were known and provable unvarying incidents of such phenomenal events, some ascertained physical law acting uniformly and equally on all such occasions, we might be able to say what was or what was not impossible within the operation of such law, but we have no such guides or criteria." We think the last suggestions made by the learned judge are entitled to much more weight, and are much more probable than the theory which precedes them, and with which the latter are utterly inconsistent. Indeed, the whole discussion seems to be based upon the idea that the defendant was bound to establish the impossibility of the theory upon which the plaintiff relied, and in its argument that court did not seem to consider that any burden rested upon the plaintiff to prove his theory with any certainty, notwithstanding the lack of substantial proof to show its correctness or existence. That opinion can hardly be read without reaching the conclusion that the learned judge was wrestling with inconsistencies impossible to harmonize, and yet that the purpose of the court was to place the case in a position where the questions of law relating to the right of the plaintiff to a recovery in this action should be presented to, and determined by, the Court of Appeals.

There are certain questions of law involved, which were discussed upon the argument, that we are called upon to decide. We have deemed it necessary to state the facts and history of this protracted litigation somewhat fully, to the end that the legal questions may be plainly presented and clearly understood in their connection with them.

The first question of law presented relates to our jurisdiction to hear and determine this appeal. The respondent contends that, inasmuch as the record does not show affirmatively that the justices of the Appellate Division were divided upon the question as to whether there was evidence supporting, or tending to sustain, the verdict, that question, at least, cannot be considered by this court, and relies upon section 1337 of the Code of Civil Procedure as sustaining his position.

In examining the question of the appealability of this case and the questions which may be determined upon this appeal, it becomes necessary to consider the provisions of the Constitution and statutes, as well as the decisions of this court relating to the subject. Section nine of article six of the Constitution declares: "No unanimous decision of the Appellate Division of the Supreme Court that there is evidence supporting or tending to sustain a finding of fact or a verdict not directed by the court, shall be reviewed by the Court of Appeals." After the adoption of that provision the legislature amended section 190 of the Code of Civil Procedure so as to provide: "From and after the last day of December, eighteen hundred and ninety-five, the jurisdiction of the court of appeals shall, in civil actions and proceedings, be confined to the review upon appeal of the actual determinations made by the appellate division of the supreme court in either of the following cases and no others: 1. Appeals may be taken as of right to said court, from judgments or orders finally determining actions or special proceedings, and from orders granting new trials on exceptions, where the appellants stipulate that upon affirmance, judgment absolute shall be rendered against them." The second subdivision of that section provides for the certification by the Appellate Division of questions of law for determination by the Court of Appeals. Then follows section 191, which contains certain limitations, exceptions and conditions to the provisions of section 190, and in that section is found a provision in which the same language is employed as in the provisions of the Constitution above cited. It also provides that no appeal shall be taken from a judgment of affirmance hereafter rendered in an action to recover damages for a personal injury, when the decision of the Appellate Division is unanimous, unless that court shall certify that in its opinion a question of law is involved which ought to be reviewed by the Court of Appeals, or unless in case of its refusal to so certify, an appeal is allowed by a judge of the Court of Appeals. It likewise provides that the jurisdiction of the court is limited to the review of questions of law.

Here, then, we find that both the Constitution and the statutes provide that no unanimous decision of the Appellate Division, that there is evidence supporting or tending to sustain a finding or verdict not directed, shall be reviewed by this court. But the provisions of section 190 are general and declare that appeals may be taken as of right to this court from judgments or orders finally determining actions or special proceedings. This confers upon a party the absolute right to appeal in those cases, unless it falls within some of the limitations contained in the subsequent section. The question here presented is how or in what manner the fact that the case falls within some limitation, if it does, is to be made to appear. Inasmuch as section 190 in general terms confers the right of appeal, the question is if it is limited by some other provision of the statute, upon whom is the burden of showing that fact. It would seem that inasmuch as the appellant in this case was given the right of appeal in express terms, he might rely upon that general provision, and if it fell within any of the limitations of the statute and the respondent claimed that the appeal was not well taken, the burden of showing that fact rested upon him. In Kaplan v. N. Y. Biscuit Co. ( 151 N.Y. 171) this court held that the burden of showing that a judgment of affirmance in an action for a personal injury was by a unanimous decision of the Appellate Division, rested upon the party asserting it, and that in order to deprive the Court of Appeals of the power to review the case under section 191 of the Code of Civil Procedure, the fact should be established by the party claiming it, either by the judgment or by a certificate of the court appearing in the record. It is true that question arose under the amendment of section 191, made in 1896, and not under subdivision four. Nor were the provisions of section 1337 passed upon in determining that case. Still, we think the principle of that decision is applicable here; that we should hold now, as we held there, that the judgment is reviewable in this court, unless the affirmance was by the unanimous decision of the Appellate Division, and that the burden of showing that fact rests upon the party asserting it, and should appear in the record. We think that section 1337 does not in any way interfere with the principle of the decision in that case, but that the provision that where the justices of the Appellate Division from which the appeal is taken are divided as to whether there is evidence supporting or tending to sustain a finding or verdict not directed by the court, a question for review is presented, is but another way of stating what is contained in subdivision four of section 191 of the Code, and in no way relieves the party who asserts it from the burden of establishing the unanimity of the decision. Hence, we are of the opinion that the contention of the respondent in this respect cannot be sustained, and that the question whether there is evidence supporting or tending to sustain the verdict may be reviewed upon this appeal. We think we should so hold, especially when we consider the fact that the Appellate Division refused to certify that its decision was unanimous, and has so plainly indicated a desire and purpose to have the questions involved decided by this court. Manifestly the amendment to section 191, passed in 1896, limiting appeals in cases of personal injury, has no application, because the judgment appealed from was rendered March 12, 1896, before the passage of that act. ( Croveno v. Atlantic Ave. R.R. Co., 150 N.Y. 225.)

As bearing upon the contention of the respondent as to the appealability of this case, it may be properly added that here every point or question upon which the appellant relies was raised by an exception to the denial of a proper motion to direct a verdict for the defendant, by an exception to the charge of the court, to its refusal to charge as requested, or by some other ruling upon the trial to which a proper exception was taken. Therefore, in its farther examination all the questions of law which are presented by the appellant must be regarded as properly before us for consideration.

The primary question which lies at the foundation of the respondent's right of recovery is whether there was sufficient evidence to justify the court in refusing to direct a verdict for the defendant, or in submitting to the jury the question of the defendant's liability. This general question seems to depend for its solution upon several subordinate ones. These questions are, first, was there sufficient evidence that the defendant performed any act or was guilty of any omission which rendered him even technically liable to the plaintiff; second, if so, was the proof sufficient to justify the court in submitting to the jury the question of substantial damages; and, third, were the alleged acts of the defendant the proximate cause of the plaintiff's injury? A consideration of these questions in the order in which they are stated seems necessary to a proper determination of the original one.

First. That at the time of the occurrence which was the subject of this action, the defendant suddenly and unexpectedly found himself confronted by a terrible and impending danger which would naturally, if not necessarily, terrify and appall the most intrepid, is shown by the undisputed evidence. If with this awful peril before him, he maintained any great degree of self-control, it indicated a strength of nerve and personal bravery quite rare indeed.

That the duties and responsibilities of a person confronted with such a danger are different and unlike those which follow his actions in performing the ordinary duties of life under other conditions, is a well established principle of law. The rule applicable to such a condition is stated in Moak's Underhill on Torts (p. 14), as follows: "The law presumes that an act or omission done or neglected under the influence of pressing danger, was done or neglected involuntarily." It is there said that this rule seems to be founded upon the maxim that self-preservation is the first law of nature, and that where it is a question whether one of two men shall suffer, each is justified in doing the best he can for himself. This principle of pressing danger and an act or omission in its presence was discussed in the squib case ( Scott v. Shepherd, 2 W. Black. 894) and in the wine case ( Vandenburgh v. Truax, 4 Denio, 464). That principle has been many times affirmed by the decisions of the courts of this state as well as others. Indeed, the trial court recognized this doctrine in its charge, but submitted to the jury the question whether the act of the defendant was involuntary and induced by impending danger, adding that the testimony of the defendant that everything he did he did intentionally was sufficient to justify it in finding that he voluntarily moved the plaintiff in the manner claimed by him. But when we examine the defendant's evidence, we find he testified that he never had his hands on the person of the plaintiff in any manner whatever until after the explosion, and that he did not at any time have any intent or design of interposing the body of the plaintiff between himself and the stranger. The testimony of the defendant, to which the court referred in its charge, seems to have been substantially that he was as cool and collected as any man could well be with the intimation made by Norcross; that he exercised his best judgment under the circumstances; that he did nothing unconsciously, spontaneously or without deliberation, but did the best he could and exercised the best judgment he could to avoid any accident; that he did nothing by impulse, and that what he did he did as deliberately as he could under the circumstances. This evidence seems to fall short of justifying the statement of the court that he testified he was in perfect possession of his senses, recollected everything that was done, and that everything he did there was intentional, as it very materially differed from and essentially modified the statement contained in the charge. The statement of the court as to the admission of the defendant can hardly be said to be a fair deduction from his evidence. Nor is the justice of eliminating from its statement to the jury the fact that the admissions he did make were accompanied by evidence that he in no way touched the plaintiff and had no intention of doing so, quite appreciated. If the court desired to use the admission of the defendant as evidence of such a fact, the evidence should have been correctly stated and the attention of the jury called to the entire admission and not to a part alone. Here, as where there is an introduction of any other conversation or admission by a party, the remainder which tends to qualify or explain the portion relied upon should be considered as a part of it, especially where it is a qualification of the other, and rebuts or destroys the inference to be drawn or the use to be made of the portion put in evidence or relied upon. While it is, doubtless, true that a portion of the testimony of a witness may be credited by a jury and a portion discredited, still, when a part of the evidence is modified or qualified by another portion, it is far from clear that one portion may be rejected and the other given credit. But, be that as it may, it is extremely difficult, upon a consideration of all the evidence in the record relating to this subject, to see how a jury was justified in finding that the defendant voluntarily interfered with the person of the plaintiff.

The only witness whose testimony is relied upon to show any interference with the plaintiff by the defendant was the plaintiff himself. He not only had all the interest of a party to the action, but the undisputed proof disclosed that his memory had been very seriously impaired, and to such an extent that he was unable to remember from day to day or hour to hour what he was told to do, and that this condition of his mind continued from the time of the accident until the last trial of this case. Upon the other hand, the defendant clearly and positively denied that he interfered with his person at all, and he is corroborated by at least one unbiased witness upon that subject, and in many of the details of the transaction by the witnesses Osborne, James and Hummel, whose evidence was in direct conflict with that given by the plaintiff. Moreover, when the testimony of the plaintiff is read, it is quite manifest that it does not clearly disclose any movement of his body, if any occurred, that was not his own voluntary act, uncontrolled by any force upon the part of the defendant. There were also certain physical facts established by the proof and uncontradicted, which tend to show that the plaintiff's theory that he was in front of the defendant was impossible. If, as is claimed by the plaintiff, the defendant employed his body as a shield, and it was between him and the place of the explosion, it is quite difficult to comprehend how the missiles, which were found in the defendant's body in front and near the median line, could have reached him, especially if, as is the plaintiff's theory, the explosion followed two straight lines. Nor can we quite understand how, if, when the explosion occurred, the plaintiff's left hand was in both of the defendant's, resting upon the defendant's left thigh, the defendant's hands could have been so seriously injured and wounded by flying substances, as the proof shows they were. With this condition of the proof, it is quite difficult to say that there was any such evidence of the defendant's intentional interference with the plaintiff as would entitle him to recover in this action or have the question submitted to a jury. The evidence which appears to be in conflict with the position of the defendant, to say the most, is nothing more than a mere scintilla and was met not only by the positive testimony of disinterested witnesses, but also by well-known and recognized physical facts about which there is no conflict. Therefore, it would seem that the plaintiff was not entitled to even nominal damages, and that it was the duty of the court to have directed a verdict for the defendant. Second. Was there sufficient evidence to justify the court in submitting to the jury the question of substantial damages? If, for the purpose of this discussion, it be admitted that the plaintiff was moved as testified to by him, and that the act of the defendant was voluntary and intentional, yet we are unable to find any sufficient evidence in the record to justify the court in submitting to the jury the question whether the plaintiff's injuries arose in consequence of the act of the defendant in moving him. The court in effect charged the jury that if the defendant interfered with the body of the plaintiff and his injuries were inflicted because of a change in his position, then it might allow the plaintiff for the injuries which he sustained in consequence of the wounds which it found were caused by his change of position, and the jury was permitted to pass upon the question whether he sustained more or different injuries than he would if he had not been moved. The contention of the respondent is that the evidence disclosed that there were two straight and well-defined lines of explosion, and that it tended to show that the plaintiff was drawn into one of them. But we find no proof either that the lines of explosion claimed did not include the place where the plaintiff originally stood, or that a wave of explosion did not pass over that portion of the room which was as forceful and destructive as that passing in any other direction.

The evidence relied upon by the plaintiff to show that there were these two defined lines was the location of the wounds upon the body of the plaintiff and the testimony of the witness Reeves. When we examine the evidence bearing upon the character and location of the plaintiff's wounds, it falls far short of establishing any well-defined line of explosion. Nor do we think any such inference can be drawn from the situation as described by the plaintiff and other witnesses. And when we examine the evidence of the witness Reeves, we find in it nothing to sustain the contention of the plaintiff. He testified that there were thirty-five to forty joists or beams, which ran north and south, that were injured, and that the breakage ran through the joists in a northeasterly direction. He was then asked whether there were fractures in any other direction in the beams or joists from that hole, to which he answered that he had put in new beams on the Rector street side of the partition leading into Mr. Sage's office; that he put in about twenty-eight or thirty new ones, and that the balance were not so badly fractured, so that they were reinforced without taking the old beams out. We find nothing in this evidence to indicate that there was any distinct line of explosion in any other direction than northeasterly. That the splitting of the joists, if it occurred, would naturally extend lengthwise of them there can be little doubt. But the plaintiff's claim that there is anything in this evidence which shows a second distinct line of explosion within which he was drawn, even if he was moved as he testified, surely cannot be sustained. Consequently, there was nothing but the merest conjecture upon which the jury could base any finding that he was more severely injured by being moved. That he was bound to establish some wrongful act upon the part of the defendant, and that that act was the cause of the injury for which he sought to recover, there can be no doubt. Nor is there any doubt that the burden of proof upon both of those questions rested upon him. The courts below have so held, but, notwithstanding their view of the law, they have submitted those questions to the jury.

In Baulec v. N.Y. H.R.R. Co. ( 59 N.Y. 356; 366) Judge ALLEN said: "It is not enough to authorize the submission of a question, as one of fact, to a jury, that there is some evidence. A scintilla of evidence, or a mere surmise, * * * would not justify the judge in leaving the case to the jury," and in that case it was held that as "at most the jury could only conjecture that the defendant might have been wanting in the care and caution proper to be exercised in such a case, * * * the case was properly withheld from the jury."

In Pollock v. Pollock ( 71 N.Y. 137, 153) FOLGER, J., said: "Insufficient evidence is, in the eye of the law, no evidence," and then cited the language of MAULE, J., in Jewell v. Parr (13 C.B. 916), where he said: "When we say that there is no evidence to go to a jury, we do not mean literally none; but that there is none that ought reasonably to satisfy a jury, that the fact sought to be proved is established."

Again, in Bond v. Smith ( 113 N.Y. 378, 385) EARL, J., stated the duties of the court in considering such a question as follows: "We have no right to guess that he was free from fault; it was incumbent upon the plaintiff to show it by a preponderance of evidence. She furnished the jury with nothing from which they could infer the freedom of the intestate from fault. She simply furnished them food for speculation, and that will not do for the basis of a verdict. The law demands proof, and not mere surmises. The authorities are ample to show in such a case the plaintiff should have been nonsuited," citing, Cordell v. N.Y.C. H.R.R.R. Co. ( 75 N.Y. 330); Dubois v. City of Kingston ( 102 N.Y. 219).

In Pauley v. S.G. L. Co. ( 131 N.Y. 90, 98), which was an action for negligence by which it was claimed that the plaintiff's intestate lost his life, Judge FINCH, in delivering the opinion of the court, said: "But the respondent and the General Term insist that some other theory may be adopted, and in order to do so, enter upon the realm of conjecture, and ask that a jury, in the utter absence of proof, may be allowed to guess that there was some negligence on the part of the defendant which might have tended to cause the death of the intestate. * * * What is claimed is that there is proof that some of the operatives fled to this escape, but could not use it on account of the blinds, and we are asked to permit a jury to guess or conjecture that Pauley was one of these, without any proof of the fact, and in the face of the evidence that no one who did use it or approach it saw him at all," and then adds: "We think the duty imposed by the statute was fairly and fully performed, and even if it was not, that we are not to resort to conjecture and permit a verdict to be based on bare possibilities alone. * * * A mere conjecture built upon a bare possibility will not suffice to transfer the money or property of one man to the possession and profit of another. As we said in Bond v. Smith ( 113 N.Y. 378) food for speculation will not serve as the basis of a verdict."

Again, in discussing this question in Linkhauf v. Lombard ( 137 N.Y. 417, 425) Judge GRAY, writing for this court, declared: "To permit a jury to speculate and surmise upon a question of responsibility, is to withdraw from the litigant a safeguard intended for the protection of his rights. He is entitled to the judgment of the court upon questions, to which the character of the evidence admits of but one answer. No such possibilities of a failure of justice should be countenanced," and then quotes from the opinion of Justice CLIFFORD in Improvement Company v. Munson (14 Wall. 442) the following: "Nor are judges any longer required to submit a question to a jury merely because some evidence has been introduced by the party having the burden of proof, unless the evidence be of such a character that it would warrant the jury in finding a verdict in favor of that party. Formerly it was held that if there was what was called a scintilla of evidence in support of a case, the judge was bound to leave it to the jury; but recent decisions of high authority have established a more reasonable rule; that in every case, before the evidence is left to the jury, there is a preliminary question for the judge, not whether there is literally no evidence, but whether there is any upon which a jury can properly proceed to find a verdict for the party producing it, upon whom the onus of proof rests," and adds: "The rule should be regarded as settled, under all the authorities, as well by the decisions of the courts of this state as by those of England, that, where there is no evidence upon an issue before the jury, or the weight of the evidence is so decidedly preponderating in favor of one side, that a verdict contrary to it would be set aside, it is the duty of the trial judge to nonsuit, or to direct the verdict, as the case may require."

In the case of Hemmens v. Nelson ( 138 N.Y. 517) this court again asserted the doctrine so clearly stated in the Linkhauf case.

In discussing that question in Hudson v. R., W O.R.R. Co. ( 145 N.Y. 408, 412), HAIGHT, J., said: "But where the evidence, which appears to be in conflict, is nothing more than a mere scintilla, or where it is met by well-known and recognized scientific facts, about which there is no conflict, this court will still exercise jurisdiction to review and reverse if justice requires," citing People ex rel. Coyle v. Martin ( 142 N.Y. 352) ; Hemmens v. Nelson and Linkhauf v. Lombard ( supra). (See, also, Cadwell v. Arnheim, 152 N.Y. 182; Hannigan v. L. H.R.R. Co., 157 N.Y. 244.) Thus we see that this court has, in a long line of decisions, uniformly held that, to justify the submission to the jury of any issue, there must be sufficient proof to sustain the claim of the party upon whom the onus rests, and that mere conjecture, surmise, speculation, bare possibility or a mere scintilla of evidence, is not enough. When the principle of these cases is applied, it becomes obvious that the court was not justified in submitting to the jury the question whether the plaintiff suffered any substantial damages by reason of his having been moved in the manner claimed, even though it should find that the defendant moved him. No one can read the evidence in the record as to the nature, power and effect of the explosion, and the results that followed, without reaching the conclusion that it utterly failed to show that the plaintiff was more seriously injured than he would have been if he had remained where he claims to have first stood. Indeed, we can find no proof sufficient to justify the conclusion that there was any place of safety, or comparative safety, in any of the rooms occupied by the defendant. The explosion swept with terrific force over them all, destroying or seriously injuring every person or thing with which it came in contact. Under such circumstances, to permit a jury to guess or conjecture that the plaintiff's injuries were more serious or severe than they would have been if his position had not been changed, is, in the language of Judge GRAY, "to withdraw from the litigant a safeguard intended for the protection of his rights." We think the court erred in not directing a verdict for the defendant, at least so far as substantial damages were concerned, upon the ground that there was no sufficient proof that the plaintiff sustained any injury in consequence of the alleged conduct of the defendant.

Third. We are next brought to the consideration of the question whether the alleged acts of the defendant were the proximate cause of the plaintiff's injury. As has already been suggested, there is no allegation or proof which tends to show that the defendant was in any way responsible for the explosion, or that there was any connection whatever between the defendant's acts and the explosion which followed. Indeed, the counsel for the respondent frankly states in his brief: "It has never been claimed, either in the pleadings or in the argument, that Mr. Sage's liability can be predicated on the fact that any act of his caused or invited a catastrophe, or that he used defiant or injudicious language." Nor does the respondent claim that the defendant was in any way responsible, directly or indirectly, for the explosion itself. Therefore, if the explosion was the proximate cause of the plaintiff's injury, the defendant cannot be held responsible. While the learned General Term stated that there was no question of proximate cause in this case, we are unable to indorse that statement or to understand the basis for it. The doctrine of proximate cause is a fundamental rule of the law of damages, to the effect that damages are to be allowed in general only for the proximate consequences of the wrong, although sometimes the question of proximate cause is applied to consequential damages for the breach of a contract, as well as to damages for negligence or tort. That the principle of proximate cause is applicable in an action for tort seems to be established by all the authorities, and we find none holding a contrary doctrine. While there may be a degree of uncertainty as to the plain signification of the term "proximate cause," or rather in its application to various cases, still in this case there can be no question as to what was the proximate and immediate cause of the plaintiff's injury. Bishop, in his work on Non-Contract Law (§ 42), in discussing this question, after remarking as to the uncertainty with which the term "proximate cause" may have been used and applied, and after defining the terms "proximate" and "remote" cause, says: "If, after the cause in question has been in operation some independent force comes in and produces an injury not its natural or probable effect, the author of the cause is not responsible."

In Shearman Redfield on the Law of Negligence (§ 26) it is said: "The breach of duty upon which an action is brought must be not only the cause but the proximate cause of the damage to the plaintiff. * * * The proximate cause of an event must be understood to be that which, in a natural and continuous sequence, unbroken by any new cause, produces that event, and without which that event would not have occurred."

Wharton thus discusses the question: "Supposing that if it had not been for the intervention of a responsible third party the defendant's negligence would have produced no damage to the plaintiff, is the defendant liable to the plaintiff? This question must be answered in the negative for the general reason that causal connection between negligence and damage is broken by the interposition of independent responsible human action. I am negligent on a particular subject-matter as to which I am not contractually bound. Another person, moving independently, comes in and either negligently or maliciously so acts as to make my negligence injurious to a third person. If so, the person so intervening acts as a nonconductor and insulates my negligence so that I cannot be sued for the mischief which the person so intervening directly produces. He is the one who is liable to the person injured." (Wharton on Negligence, § 134.)

As has been said in an anonymous article in the American Law Review: "A proximate cause is one in which is involved the idea of necessity. It is one the connection between which and the effect is plain and intelligible; it is one which can be used as a term by which a proposition can be demonstrated, that is, one which can be reasoned from conclusively. A remote cause is one which is inconclusive in reasoning, because from it no certain conclusion can be legitimately drawn. In other words, a remote cause is a cause the connection between which and the effect is uncertain, vague or indeterminate. It does not contain in itself the element of necessity between it and its effect. From the remote cause the effect does not necessary flow. * * * This idea of necessity — the necessary connection between the cause and the effect — is the prime distinction between a proximate and a remote cause. The proximate cause being given, the effect must follow. But although the existence of the remote cause is necessary for the existence of the effect (for unless there has been a remote cause there can be no effect) still the existence of the remote cause does not necessarily imply the existence of the effect. The remote cause being given, the effect may or may not follow." (4 Am. Law Review, 201, 205.)

In Marble v. City of Worcester (4 Gray, 396) Chief Justice SHAW said: "On account of the difficulty in unraveling a combination of causes, and of tracing each result, as a matter of fact, to its true, real and efficient cause, the law has adopted the rule before stated, of regarding the proximate, and not the remote cause of the occurrence which is the subject of inquiry."

In Crain v. Petrie (6 Hill, 522, 524) the question of special damages was involved, and, consequently, the question of proximate cause; that is, whether the special damages were the legal and natural consequences of the wrong complained of. NELSON, Ch. J., in that case said: "To maintain a claim for special damages they must appear to be the legal and natural consequences arising from the tort, and not from the wrongful act of a third party remotely induced thereby. In other words, the damages must proceed wholly and exclusively from the injury complained of."

In Milwaukee Saint Paul Railway Co. v. Kellogg ( 94 U.S. 469) it was said: "The inquiry must, therefore, always be whether there was any intermediate cause disconnected from the primary fault, and self-operating, which produced the injury."

In Hofnagle v. N.Y.C. H.R.R.R. Co. ( 55 N.Y. 612) it was held that the act of one person cannot be said to be the proximate cause of an injury when the act of another person has intervened and directly inflicted it.

Another principle of proximate cause which seems to be well established is that an accident or injury cannot be attributed to a cause unless, without its operation, it would not have happened. ( Ring v. City of Cohoes, 77 N.Y. 83; Taylor v. City of Yonkers, 105 N.Y. 202; Ayres v. Village of Hammondsport, 130 N.Y. 665; Grant v. Pa. N.Y. Canal R.R. Co., 133 N.Y. 657.) When damages claimed in an action are occasioned by one of two causes, for one of which the defendant is responsible, and for the other of which he is not responsible, the plaintiff must fail if his evidence does not show that the damage was produced by the former cause, and the jury must not be left to mere conjecture, and a bare possibility that the damage was caused in consequence of the act of the defendant is not sufficient. ( Searles v. Manhattan Railway Co., 101 N.Y. 662.) The plaintiff must fail if the evidence does not show that the injury was the result of some cause for which the defendant is responsible, and where the proof is by circumstances, the circumstances themselves must be shown and not left to rest in conjecture, and when shown it must appear that the inference sought is the only one which can fairly and reasonably be drawn from the facts. ( Ruppert v. Brooklyn Heights R.R. Co., 154 N.Y. 94.)

When we apply to the undisputed facts of this case these rules relating to proximate cause, it becomes quite manifest that the judgment in this action cannot be upheld. All the injuries which the plaintiff sustained were caused directly and immediately by the act of Norcross in exploding the dynamite. That was clearly the proximate, and we think the only, cause of the plaintiff's injury. It was the only efficient cause, as, confessedly, without the explosion the plaintiff would not have been injured, and under no circumstances can it be properly said that the act of the defendant in changing the plaintiff's position a few inches to the left of where he previously stood caused the explosion or occasioned the catastrophe. Surely that was not an act, without which the explosion would not have occurred, nor can it be held to have been the proximate cause of the explosion. The most that can be said is that it produced a situation which existed at the moment it occurred. Obviously, the explosion would have occurred if the defendant had moved the plaintiff in an opposite direction, or had not moved him at all. Nothing which the defendant did could have produced the injury sustained by the plaintiff without another independent intervening cause. There was no evidence in the case of any necessary relation of cause and effect between the act of which the plaintiff complains and the explosion which caused his injury. Under the proof, we think the question whether the defendant's act was the proximate cause of the plaintiff's injury was a question of law, and that the court erred in not directing a verdict for the defendant on that ground.

Another ground upon which the defendant seeks to sustain this appeal is that the court erred in refusing to strike out and in not instructing the jury to disregard the evidence of George Baillard, to which we have already referred. While it is possible that there was sufficient proof to justify the jury in finding that the "elderly stranger" to whom the witness referred was the defendant, still it is obvious that the evidence admitted was so general and inconclusive as to render it only conjectural as to whether it related to the subject at issue at all. It was that the witness understood this "elderly stranger" to say something about being protected — about a protection he had had from the explosion. As to the nature of the protection, its extent, whether by physical matter or an overruling Providence, or a protection of some other character, nothing was attempted to be shown. It seems to us quite clear that no such uncertain statement should have been submitted to a jury or permitted to become the basis of a verdict involving the rights of parties. It was too uncertain, indefinite and vague to be permitted to serve any such purpose.

Our attention is also called to various rulings of the trial court which arose upon the cross-examination of the defendant. His cross-examination was a severe and rigorous one. While perhaps it was not objectionable upon that ground alone, we think the plaintiff's counsel was permitted to go beyond the legitimate bounds of a proper cross-examination in several respects, to which we cannot call attention in detail within the proper limits of this opinion. Counsel was permitted to prove upon the cross-examination of the defendant, or at least to attempt to do so, that the defendant had not shown proper sympathy or paid proper attention to the plaintiff after he was injured. How that evidence could properly bear upon the legal rights of the parties it is extremely difficult to understand. The only purpose it could serve was to prejudice and excite the passions of the jury.

Again, we find rulings which were made on the cross-examination of the defendant in regard to the article in the New York World, which we think cannot be upheld.

It has ever been the theory of our government and a cardinal principle of our jurisprudence that the rich and poor stand alike in courts of justice, and that neither the wealth of the one nor the poverty of the other shall be permitted to affect the administration of the law. Evidence of the wealth of a party is never admissible, directly or otherwise, unless in those exceptional cases where position or wealth is necessarily involved in determining the damages sustained. As to this general proposition there can be no doubt, and no authorities need be cited. Notwithstanding this well-established principle, the plaintiff was permitted to show by the defendant upon his cross-examination substantially, or at least to a great extent, the amount of property possessed by him, its character and his business. He was permitted to show the number of railroads the defendant operated, the banks in which he was a director, that he dealt in stocks, that he loaned money, and other details of his affairs. We think much of this evidence was improperly received, and that the exceptions were valid. But we deem it both unnecessary and unwise to extend this necessarily protracted opinion by either discussing those questions or considering the various other exceptions contained in the record which present questions that are serious, if not fatal, to this judgment. No good purpose could be served by such a course, as the judgment must be reversed upon the more substantial grounds which have been fully discussed.

It is impossible to consider the plaintiff's injuries without a feeling of profound sympathy. His misfortune was a severe one, but sympathy, although one of the noblest sentiments of our nature, which brings its reward to both the subject and actor, has no proper place in the administration of the law. It is properly based upon moral or charitable considerations alone, and neither courts nor juries are justified in yielding to its influence in the discharge of their important and responsible duties. If permitted to make it the basis of transferring the property of one party to another, great injustice would be done, the foundation of the law disturbed and anarchy result. Hence, every proper consideration requires us to disregard our sympathy and decide the questions of law presented according to the well-established rules governing them.

The judgments of the Appellate Division and of the trial court should be reversed and a new trial granted, with costs to abide the event.

All concur, except PARKER, Ch. J., not voting, and GRAY, J., absent.

Judgments reversed, etc.