We consider the evidence to be inadmissible. The nature of the transaction in the magistrate’s office was such as to exclude it. “ If you do not tell the truth, I will commit you,” was the language used to her; language sufficient to exclude any confession induced by it. The administering of an oath by the magistrate, under such circumstances, was a gross outrage upon the accused; any information drawn by it, or subsequently given on its basis, is inadmissible. Here there was an interval of only a night and part of a day between the evidence as thus extracted, and that *271now sought to be introduced. Gibson, the constable, was present at both periods; and his presence, as well as the continuance of the same scenery as surrounded the prisoner at the first confessedly inadmissible confessions, were calculated to produce on the mind of the prisoner the same influence at the latter period as existed at the first. No warning was given; but, on the contrary, the only question put was one tending to entrap the prisoner.
A question having arisen as to the duty of the coroner to call in medical assistance, in case of a post mortem examination—
Gibson, C. J.There ought to be post mortem examinations in all cases of death by violence. An action lies against the county at common law by a physician for trouble and labour expended in such examination. A case to this effect was decided by us last spring at Pittsburgh.
Bell, J.In the country, so far as my experience goes, it is always the practice for the coroner in such cases to have an examination by medical men. The court, in fact, always has exacted it.
The prisoner offered no evidence.
Stokes, and Read, Attorney-general, for the Commonwealth, and Daugherty and Lehman, for the defence, having addressed the jury, they were charged as follows:—
Gibson, C. J.You have heard the law■ of homicide so accurately defined by the counsel who assists the attorney-general, that it is unnecessary to discuss it. at length. Murder is the unlawful killing of a human being with malice aforethought, express or implied. If the prisoner at the bar drowned her child, she is guilty, by the law of Pennsylvania, of murder in the first degree. There is no middle ground or questiofi of degrees. If she is guilty of the killing, there is but one verdict which you can give. With this instruction it is our duty to tell you, that what you have to determine, is a question of fact; that it is for your exclusive decision ; and that we intimate no opinion as to the conclusion which you ought to draw.
The evidence is so simple and has been so thoroughly discussed, that a further attempt to analyze it, or collate it, or pass it in review, wmuld render you no assistance ; and I therefore shall confine my remarks to the distinctive character and valué of it. No witness has been produced who saw the act' committed ; and hence it is urged for the prisoner, that the evidence is only circumstantial, and consequently entitled to a very inferior degree of credit, if to any credit at all. But that consequence does not necessarily follow. Circum*272stantial evidence is, in the abstract, nearly, though perhaps not altogether, as strong as positive evidence; in the concrete, it may be infinitely stronger. A fact positively sworn to by a single eye-witness of blemished character, is not so satisfactorily proved, as is a fact which is the necessary consequence of a chain of other facts sworn to by many witnesses of undoubted credibility. Indeed, I scarcely knowr whether there is such a thing as evidence purely positive. You see a man discharge a gun at another, you see the flash, you hear the report, you see the person fall a lifeless corpse ; and you infer from all these circumstances that there was a ball discharged from the gun, which entered his body and caused his death, because such is the usual and natural cause of such an effect. But you did not see the ball leave the gun, pass through the air, and enter the body of the slain; and your testimony to the fact of killing is therefore only inferential—in other words, circumstantial. It is possible that no ball was in the gun; and we infer that there was, only because we cannot account for the death on any other supposition. In cases of death from the concussion of the brain, strong doubts have been raised by physicians, founded on appearances verified by post mortem examination, whether an accommodating apoplexy had not stepped in at the nick of time to prevent the prisoner from killing him, after the skull had been broken into pieces. I remember to have heard it doubted in this court-room, whether the death of a man, whose brains oozed through a hole in his skull, was caused by the wound or a misapplication of the dressings. To some extent, however, the proof of the cause which produced the death rested on circumstantial evidence.
The only difference between positive and circumstantial evidence is, that the former is more immediate, and has fewer links in the chain of connection between the premises and conclusion; but there may be perjury in both. A man may as well swear falsely to an absolute knowledge of a fact, as to a number of facts from which, if true, the fact on which the question of innocence or guilt depends must inevitably follow. No human testimony is superior to doubt. The machinery of criminal justice, like every other production of man, is necessarily imperfect, but you are not therefore to stop its wheels. Because men have been scalded to death or torn to pieces by the bursting of boilers, or mangled by wheels on a railroad, you are not to lay aside the steam-engine. Innocent men have doubtless been convicted and executed on circumstantial evidence; but innocent men have sometimes been convicted and executed on what is called positive proof. What then ? Such convictions are accidents *273which must be encountered; and the innocent victims of them have perished for the common good, as much as soldiers who have perished in battle. All evidence is more or less circumstantial, the difference being only in the degree; and it is sufficient for the purpose when it excludes disbelief; that is, actual, and not technical disbelief; for he who is to pass on the question, is not at liberty to disbelieve as a juror while he believes as a man. It is enough that his conscience is clear. Certain cases of circumstantial proofs to be found in the books, in which innocent persons were convicted, have been pressed on your attention. These, however, are few in number, and they occurred in a period of some hundreds of years, in a country whose criminal code made a great variety of offences capital. The wonder is, that there have not been more. They are constantly resorted to in capital trials to frighten juries into a belief that there should be no conviction on merely circumstantial evidence. But the law exacts a conviction wherever there is legal evidence to show the prisoner’s guilt beyond a reasonable doubt; and circumstantial evidence is legal evidence. If the evidence in this case convinces you that the prisoner killed her child, although there has been no eye-witness of the fact, you are bound to find her guilty. For her sake, I regret the tendency of these remarks, but it has been our duty to make them, and it will be yours' to attend to them.
Two minor points of defence to which I will advert have been made. It has been pressed that there can be no conviction where there is no evidence that the deceased was alive immediately preceding the period of the supposed homicide, and for this, we are referred to the United States v. Hewson, 7 Boston Law Rep. 361, S. C., Whart. Crim. Law, 225, in which Mr. Justice Story held,' that a mother who, in a fit of puerperal fever, had thrown- her child overboard, could not be convicted of murder without proof that it was alive at the time. This is not the case; for there the child may have died a natural death at so tender an age, and the instinctive love of the mother would render it more probable that the fact was so, than that she threw her child into the sea when it was alive. But if Judge Story meant to say that this probability amounted to a legal presumption of antecedent death till it should be rebutted by proof, I dissent from him, notwithstanding the great respect I entertain for every thing that has come from him. The presumption that a person proved to-have been alive at a particular time is still so, holds till it is rebutted by lapse of time, or other satisfactory proof. The prisoner’s child was seen alive in her arms, at half-past six o’clock in the morning, healthy and vigorous; and at eleven at night it was *274found dead, with marks of suffocation on its person. The presumption then is, that it was alive when these marks were impressed.
You have been told that to doubt of the prisoner’s guilt is to acquit her. But a doubt, to work an acquittal, must be serious and substantial—not the mere possibility of a doubt. If the evidence convince you of guilt beyond a reasonable doubt, you are bound to convict. You are the judges of its effect; and if you can reconcile it to any reasonable hypothesis of innocence, you may acquit; if not, you are bound to say so. The issue is in your hands.
After a deliberation of about three hours the jury returned a verdict of guilty of murder in the first degree. Motions for a new trial and in arrest of judgment were made, resting chiefly on the alleged inadequacy of circumstantial evidence, such as was adduced on the trial, but were not pressed by the counsel for the prisoner, and were finally overruled by the court. On Saturday, November 27th, sentence of death was pronounced on the prisoner by Gibson, C. J., Coulter, J., and Bell, J., being with him on the bench.