The defendant was convicted of forging and of uttering a check which purported to be the check of one Frank Dixon. The check as described was drawn in favor of defendant upon the Traders’ and Importers’ Bank of New York. The appeal is from the judgment and an order refusing a new trial. '

The first point made by the appellant is that the court erred in admitting in evidence1 the protest which accompanied the check which it is charged was forged and uttered by the defendant. It is not seriously denied that this ruling was error, but it seems impossible that injury could have resulted therefrom to defendant. The only materiality in the evidence, and the only way in which injury could have resulted, is that it tended to show that the check was not paid upon presentation in New York. But this was never a fact in controversy. It was proven in many ways, and was admitted by the defendant.

The next point relates to the admission of four other checks which there was evidence tending to show defendant negotiated, and upon which he obtained money, and which were not paid by the drawees. They were each objected to as incompetent, irrelevant, and immaterial, and also upon the ground that there was no proof even tending to show that the check set out in the indictment was a forgery.

The checks all purported to be drawn upon banks of the city of New York.

*340 The first was as follows:

It was accompanied by a. protest for nonpayment,, which was also read in evidence.

It was shown that the defendant had procured money upon this check from one Maxwell, who is the same-person upon whom the check mentioned in the indictment had been passed. Also that the check had been presented to the bank in New York and payment was-refused, and that defendant, upon being informed of the-fact, had repaid the money to Maxwell.

The prosecution produced no evidence whatever tending to show that the check was forged, and none to show G. S. Hyman was a real person, and of course none to show that if he was a real person he had not authorized defendant to use his name in that way. Possibly this-statement should be qualified in one respect, for after all the checks had been read in evidence, as well ascertain letters which it was admitted defendant wrote, an expert 'upon the examination of the handwriting testified that in his opinion they were all in the same handwriting. But if this evidence tended to prove that they were all written by the defendant, there was still wanting evidence that the apparent drawer was a real person, or if he was that he had not authorized such use-of hisYrame.

Another of the checks was for two hundred and fifty dollars, drawn on the Bank of Commerce of New York, March 27, 1894. It purported to have been signed by S. W. Williams, and was payable to E. A. Delafield or order. Delafield it was proven was an alias by which defendant was then known. The check was indorsed “ Pay to the .order of A. J. Whiteman. E. A. Delafield,”' *341 and had upon it the indorsement “ A. J. Whiteman.” There was proof that the defendant had passed this at the Bank of California, and had been enabled to do so through the indorsement .of Thomas Day & Co., who ultimately had to pay the check to the Bank of California.

As to this check, also, there was a total lack of evidence as to its being a forgery except as in the case of the last-mentioned check.

Another check dated April- 1, 1894, was drawn upon the Importers’ and Traders’ Bank of New York, and purported to have been signed by Ensign and Carney. Defendant procured money' upon it from W. F. Boeder. It was shown that the check had been presented to the •drawee and that payment was refused. The proof in regard to its being a forgery was exactly as in regard to the first-mentioned check.

Still another of the checks was dated New York, March 24, 1894. It purported to have been drawn by E. A. Delafield. It was indorsed by defendant in that name and by him negotiated. It was shown that payment had been refused. As in the case of the other «checks, there was no proof of the forgery, or the existence of the apparent drawer, or of lack of authority.

The prosecution was also permitted to prove that defendant was registered at the Occidental Hotel under the name of E. A. Delafield, and while there he procured money from the cashier of the hotel upon three checks ■drawn on the Importers’ and Traders’ Bank of New York. The checks purported to be drawn by J. A. Delafield & Co. in favor of E. A. Delafield] and were indorsed by him in that name. There was also some «effort to show that in all of the checks put in evidence the same blank forms had been used. One of the three last-named checks was forwarded to New York and presented to the bank, and payment was • refused. Upon being informed of this the defendant refunded the money and took them all up. They were, therefore, not produced, and, of course, as to them, there was not *342 the slightest attempt to prove that they had been forged. As Delafield was confessedly an assumed name, it is probable that the signature J. A. Delafield & Co, was a fictitious firm. There was, however, no proof upon that subject.

We have- here, then, including the check described in the indictment, eight different checks for small sums purporting to have been drawn by five different drawers upon New York banks. They were all negotiated within a few days in San Francisco, and none were paid on presentation. Letters written by the defendant were also read, in which he seems to admit that he did not think they would be paid. But as to none of them was there evidence that they had been forged.

No evidence short of conclusive proof of guilt could have been more damaging. At .least, this evidence tended very strongly to prove that the defendant was a cheat and reckless swindler.

The evidence was admitted-only for a limited purpose^ and,.in the charge of the court, its effect was expressly restricted to that purpose. When one of the checks was offered and had been objected to, the court said: “I

understand that all these matters are simply to show guilty knowledge in the transaction mentioned here, and for no other purpose.” To which the prosecuting attorney responded, “.Yes, sir.”

In the charge the court said: “Now, in order to show guilty knowledge, the prosecution have been allowed to introduce before -you cases of other checks alleged by them to have been forged, but you will be careful to remember that such evidence is applicable only to the question of guilty knowledge.” The jury was also told, as to the principal check, that the fact that there were no funds to meet it did not tend to prove that the check was forged.

These checks were not, then, offered or recéived for the indorsements which were admitted to be in the handwriting of defendant, for the purpose of comparison, nor as the basis of a theory tending to show that *343 all the checks were forged, but to show guilty knowledge only.

But there was no "proof that the other checks were forged, and, therefore, it was not shown, nor was there evidence tending to show, that about the same time the defendant had passed other forged checks. When the fact of guilty knowledge becomes material it must appear that the check described in the indictment was 'itself a forgery. Proof that defendant passed other checks which were forged cannot be introduced, when unaccompanied with other inculpating facts, to establish the corpus delicti. And, when the body of the offense has been established, and that defendant passed the check, and it is sought to show guilty knowledge by the fact that defendant also passed other forged paper, the prosecution assumes the same burden as to all the other checks introduced. It must show that such checks, were forged.

The evidence did tend very strongly to prove the defendant a swindler, and I have no doubt that it would be much easier to convince a jury that defendant forged-the Dixon check after they had been convinced that he was a swindler. And such an impression is natural and reasonable. But to allow proof of the bad character of a defendant as a step in the proof of guilt has never been the policy of the law. On the contrary, pains is-sometimes taken to prevent knowledge of the bad character of the defendant from reaching the jury.

Thus where a defendant has been previously convicted! of an infamous offense, and such fact is charged in an-indictment, a method is devised by which a defendant may prevent that part of the indictment from being read to the jury.

As to the three checks which w'ere not produced, there was no Evidence of their genuineness, and, as to the others, none except that the expert testified that all were in the same handwriting, as were also certain let-' ters which it is admitted defendant wrote. There was. no evidence that the persons by whom the checks pur *344 ported to be drawn were real persons, and none that, if they were real persons, defendant was not authorized to use their names. It is stated—perhaps through some misapprehension—that the learned judge of the trial court ruled that it was incumbent on the defendant to show that he had such authority, after it has been shown that he has drawn checks in the name of another. The attorney general does not contend for any such doctrine, and it is repudiated in the charge of the court in this case. The jury were told: “When an accused man pleads 1 not guilty/ as he has pleaded here, that immediately there arises a presumption of innocence in his favor which accompanies him all through the trial and never deserts him.” To prove that an accused person signed the name of another to an instrument, and that he passed such instrument as genuine, does not prove the commission of a crime. It must still be shown that it was a false instrument, and this is not proven until it is shown that the person who signed another’s name did so without authority. Until this proof is made it is not shown to be a false instrument, and the defendant is not put to his proof at all. The defendant when on the stand admitted that he drew two of the four checks which were produced at the trial, and that he negotiated them. He claimed, however, that he had authority from the drawers to so use their names. Waiving the necessity of corroborating evidence as to the corpus delicti, would-anyone contend that this proved forgery? Suppose that he had admitted that he drew the Dixon check, and. claimed that he had authority from Dixon to draw it, wotild this admission have put the burden upon him to show such authority ?

It would be a new departure in criminal law. It arises from a difficulty which is inherent in^our system. We have inherited the idea, which we carefully place in all our constitutions, that an accused person must be •confronted by his witnesses in court. State process cannot cross state borders, and hence justice is sometimes defeated. The courts have not created the diffi-, *345 culty, and cannot provide a remedy by depriving a defendant of the presumption of innocence and putting the burden upon him.

I do not think there was reversible error in admitting letters of the defendant which contained certain admissions before there was proof of the corpus delicti. The rule is that there must be other and independent proof that a crime has been committed by someone. Then it can be shown by the extrajudicial confessions that defendant is the guilty party. In some cases it has been held that the order of proof is material. That it is the fairer course that there should be independent proof first, as to the body of the offense, cannot be doubted. But I do not think a case should be reversed merely because of a departure from the natural and just order of evidence.

It is said that the case must be reversed because the prosecution produced no evidence that defendant was not authorized to use Dixon’s name, and also because they did not show that Dixon was a real person. Perhaps there was no such evidence when the prosecution rested, but, after the court had refused an application to instruct the jury to acquit the defendant, he voluntarily took the stand as a witness and supplied the evidence which had been lacking. If there was error, it was thus cured.

For the error above indicated, however, the cause is remanded and a new trial ordered.

Henshaw, J., and McFarland, J., concurred.