Opinion,
Me. Justice Steeeett :
As stated in appellant’s history of the case, “ the defence upon which his counsel chiefly relied was insanity.” In view of that fact, several points for charge, bearing more or less directly on the subject, were submitted. One of these is “ that under the plea of not guilty, the defendant has a right to show, by way of defence, the insanity of the defendant at the time of the killing, and that the jury must pass upon the question of the defendant’s sanity or insanity, and if they find him insane at the time of the killing, acquit him by reason of insanity.” This point was rightly affirmed without any qualification, and, of course, it is not assigned for error. Other points *296recited in the thirteenth, fourteenth, fifteenth and seventeenth specifications, respectively, were answered in the negative, and therein it is alleged there was error. In the first of these, the court was requested to charge: “ That the burden of proof never shifts from the commonwealth to the defendant, and the commonwealth must show, beyond a reasonable doubt, that defendant was of sound mind, memory, and discretion at the time of the killing.” The learned judge’s answer was: “ This point is refused. As I understand the point, it is intended to say that the defence of insanity shall be established beyond a reasonable doubt; that, unless it is established beyond a reasonable doubt, it would be your duty to acquit. I do not understand the law to go to that extent, and the matter will be referred to in my general charge, wherein the law, as I understand it, is correctly stated on this subject.” If this answer was intended to be responsive to the point, its meaning is not very clear. The court was not requested to charge “ that the defence of insanity shall be established beyond a reasonable doubt.” On the contrary, the last clause of the point is, in substance, that the burden of proving, affirmatively and beyond a reasonable doubt, the sanity of the defendant at the time of the killing, was on the commonwealth. But, whatever impression this and other answers to the defendant’s points may have made on the minds of the jury, it may be safely assumed that, in considering the evidence bearing on the defence of insanity, they were governed by what the learned judge afterwards» said in that portion of his general charge to which they were specially referred for a correct statement of the law on the subject. After speaking particularly of insanity as a defence, etc., he there said, inter alia: “ It is my duty to say to you, as the law governing the responsibility of men for their acts, that, in all cases, every man is presumed to be sane and to possess a sufficient degree of reason to be responsible for his crimes, until the contrary is proved to the satisfaction of the jury; and, to establish a defence on the ground of insanity, it must be clearly proved that, at the time of committing the act, the party accused was laboring under such a defect of reason from disease of the mind as not to know the nature of the act he was doing, or, if he did know it, that he did not know he was doing what was wrong. This *297looks like, a fair definition of what insanity is. That is what is required to relieve him of responsibility for his acts.”
Without questioning the general correctness of what was said, in that connection, as to the kind of insanity that constitutes a defence to an act, which would otherwise be punishable as criminal, we think the degree of proof necessary to sustain such a defence was too strongly stated in saying “ it must be clearly proved.” This was imposing on the defendant a greater burden than the law requires.
In harmony with the humane principle of the criminal law, that every person accused of crime shall be presumed innocent until his guilt is clearly established, it is incumbent on the commonwealth to prove, not only to the satisfaction of the jury but beyond a reasonable doubt, the presence of every ingredient necessary to constitute the crime charged in the indictment. That burden, as was said in Turner v. Commonwealth, 86 Pa. 54, never shifts, but rests on the prosecution throughout; so that, in all cases, a conviction can be had only after the jury has been convinced, beyond a reasonable doubt, of the defendant’s guilt. It necessarily follows that, if the evidence is such as to leave a reasonable doubt in the minds of the jury as to the existence of any necessary ingredient of the crime charged, they should give the defendant the benefit of that doubt. But presumptions of fact sometimes stand for full and express proof until the contrary is shown. For example, inasmuch as sanity is the normal condition of man, every one is presumed to be sane, and that presumption holds good, and is the full equivalent of express proof, until it is successfully rebutted. When insanity of the defendant is set up as a defence, it is incumbent on him to rebut the ordinary presumption of sanity, and show, not beyond a reasonable doubt, nor eithér clearly or conclusively, but by fairly preponderating evidence, such as is ordinarily required to prove a fact in civil issues, that he was insane at the time of committing the alleged crime. In Lynch v. Commonwealth, 77 Pa. 205, 213, the trial judge refused to charge “ that, if the jury have a reasonable doubt as to the condition of the defendant’s mind at the time the act was done, he is entitled to the benefit of such doubt, and they cannot convict; ” and, for further answer to the point, said: “ The law of this state is that where the killing is admitted, and in*298sanity or want of legal responsibility is alleged as an excuse, it is the duty of the defendant to satisfy the jury that insanity actually existed at the time of the act; a mere doubt as to such insanity will not justify the jury in acquitting on that ground.” That instruction was approved by this court, and substantially the same instruction was afterwards sanctioned in Ortwein v. Commonwealth, 76 Pa. 421, 425, and other cases. In Coyle v. Commonwealth, 100 Pa. 573, the same rule of evidence was again recognized. It was further held to be error, in that case, to instruct the jury that the defence of insanity should be proved by clearly preponderating evidence. The instruction should have been “ fairly preponderating,” instead of “ clearly preponderating evidence.” Speaking of the degree of proof required by the words employed in that case, Mr. Justice MerCUR said: “ It is demanding a higher degree of proof than the authorities require. It may be satisfactorily proved by evidence which fairly preponderates. To require it to clearly preponderate, is practically saying it must be proved beyond all doubt or uncertainty. Nothing less than this will make it clear to the jury.” As applied to the degree of proof required to rebut the presumption of sanity, and sufficiently prove the existence of insanity, there is no appreciable difference between the expressions “ clearly proved ” and “ proved by clearly preponderating evidence.” If there is any difference, the former calls for the higher degree of proof. It is almost equivalent to saying “ proved beyond a reasonable doubt; ” because, if any doubt as to the existence of a particular fact exists, it cannot be said to be “ clearly proved.”
It is true that the learned judge, in another part of his somewhat lengthy charge, said to the jury: “You have to be satisfied of his insanity by the preponderance of the evidence. He has to establish, in other words, his insanity, not by the rule of a reasonable doubt, but by the testimony, what the preponderance of the evidence shows.” But, with two measures of proof before them, one substantially correct and the other erroneous, how is it possible for us to determine which the jury adopted? There should be nothing left to conjecture, especially in a capital ease. It is enough to know that the jury may have been misled by erroneous instructions on a point vital to the defence.
*299The testimony referred to in the sixth, eighth, ninth and tenth specifications appears to have been neither incompetent nor irrelevant, and we think it should have been admitted. Neither of the remaining specifications of error requires special notice. That part of the charge embraced in the eighteenth specification of error contains some expressions of opinion, etc., that might have been profitably omitted, but we are not prepared to say that they are positively erroneous.
Judgment reversed, and a venire facias de novo awarded.