At tbe April term, 1889, of tbe Criminal Court of Jefferson, a true bill was found and returned into court, charging five persons — Jim Williams, alias Ben. Elzy, Lawrence Johnson, Henry Joe, Gilbert Lowe and Lindsey Malacbi — with tbe murder of John W.' Meadows. On May 14, 1889, tbe several defendants were arraigned; one of then, Ben. Elzy, pleaded guilty, and all tbe others pleaded severally not guilty. On May 23, 1889, Lawrence Johnson and Gilbert Lowe moved severally for a severance, and that each be tried alone; which motion was granted by tbe court. On tbe same day, Lindsey Malacbi and Henry Joe jointly moved that they be allowed a severance from tbe other defendants, and that they be tried together, but sepa*139rate from Johnson, Elzy and Lowe. This motion was granted.
The trial of the two defendants, Malachi and Joe, was continued to the October term, 1889. On November 4, 1889, during the October term, the Criminal Court, by an order, set November 25,1889, for the trial of Malachi and Joe; but they were not tried at that term. On November 28, 1889, each of the defendants, Malachi and Joe, severally moved the court to be allowed to withdraw their pleas, and to move for a severance in trial as between themselves. This motion was overruled by the court, and they were denied the right to sever further. They were tried together, and Malachi was convicted, and Joe acquitted. This presents the first exception reserved. We hold that the Criminal Court did not err in this ruling. After allowing one severance to these defendants, and after granting their motion as made, any further indulgence in that direction would, at most, be discretionary with the presiding judge — Code of 1886, § 4451; Rule of Practice, 86 Ala. VIII.
The trial court did not err in restricting the defendants each to eleven peremptory challenges of jurors.' — -Act approved February 28, 1889.—Sess. Acts, 77-8; Maxwell v. State, at present term.
Nor was there error in permitting the photographs to be put in evidence. They were proper aids to identification. K. C. M. & B. R. R. Co. v. Smith, at present term.
Ben. Elzy was the only witness on the trial of this case who gave positive testimony of Malachi’s participation in the double crime of robbery and murder of the deceased. Without his testimony there could have been no conviction. He was a confessed accomplice in the crime. It has always been considered dangerous to convict of a heinous crime on the uncorroborated testimony of an accomplice; and emphasizing the sentiment, our statute — Code of 1886, § 4476— declares, that “a conviction of felony can not be had on the testimony of an accomplice, unless corroborated by other evidence tending to connect the defendant with the commission of the offense.”
There was outside testimony on the trial, of the conduct and various conversations of defendant, which were, no doubt, relied on as tending to prove his guilt. These were of not much significance. The main fact relied on as corroboration arose and wás presented as follows: When defendant was arrested in April, 1889, an overcoat was found *140in Ms possession, which had some unusual marks and peculiarities about it. This overcoat was produced before' the jury, and witnesses testified that it was the overcoat of the deceased, which they had seen him wear on many occasions. The testimony produced by the State, as to the identity of the overcoat, was pretty minute and full. Against this, defendant testified that he had owned the coat for three or four years before the homicide was committed; and he produced many witnesses who testified that they had seen him wear it, some fixing the date as far back as three or four years before the time when Meadows was last seen alive. Meadows was last seen alive about the last of December, 1888. His dead body was found about February 1, 1889. Only underclothes were on or about the body when found, and the corpse was considerably mutilated and decayed. It was, however, very fully identified in several ways.
Ben. Elzy, the accomplice, testified minutely as to the killing, he being present, and he testified to facts which clearly prove he was a guilty participant in the crime. If his testimony be true, it was a suggestion of Malachi which led to the murder, after Meadows had been robbed. He did not testify that Meadows’ clothes, or the overcoat was taken. He stated that, when Meadows approached them, he had on a coat, pants, hat and shoes, and had another coat thrown over his arm. He did not testify it was an overcoat, and he did not, in any way, identify the overcoat as having been in the possession of Meadows. As we have before stated, neither he nor any one else made positive proof that any clothes were taken from the person or body of Meadows; and it was thus left to inference how it was that the body had on it only under garments when it was found. It will be observed that Malachi’s arrest, and the discovery of the overcoat in his possession, occurred more than three months after the homicide was committed.
On the question of corroborating Elzy’s testimony by the fact that Meadow’s overcoat was found in Malachi’s possession, the defense assumes two forms. First, that Elzy did not identify the overcoat, and that he gave no material testimony in regard to it. The contention is, to corroborate the testimony of an accomplice, and thereby render it sufficient to found a conviction upon, the corroboration must be of some specific fact or circumstance to which the accomplice has testified. We think this is too narrow a view. Corroborate means to strengthen, to make stronger; to strengthen, not *141the proof of any particular fact to which the witness has testified, but to strengthen the probative, criminating force of his testimony. And under our statute, the corroborating testimony must tend “to connect the defendant with the commission of the offense. And this, at last, was but giving voice to the common-law rule.
This subject has been very largely discussed.—Note to § 4476 of the Code of 1886; 1 Greenl. Ev. §§ 34, 35, and notes; 1 Amer. & Eng. Encyc. of Law, 78 et seq.; Note to Com. v. Price, 71 Amer. Dec. 671 et seq.; Lumpkin v. State, 68 Ala. 56; People v. Haynes, 55 Barb. 450; People v. Clough, 73 Cal. 348. The fullest and ablest discussion of the question which has been called to our attention is Com. v. Holmes, 127 Mass. 424; s. c., 34 Amer. Rep. 391, and note. The opinion was by Gray, C. J., now Associate Justice of the Supreme Court of the United States. Among other things, he said: “Evidence which tends to prove the guilt of the defendant is sufficient by way of corroboration, although it does not directly confirm any particular fact stated by the accomplice; as, for instance, evidence of the possession of stolen goods, by one indicted for stealing or receiving them.” In Ross v. State, 74 Ala. 532, the corroboration was not of' any particular fact testified to by the accomplice. The corroboration relied on was, first, the flight of the defendant, and, second, proximity and opportunity for committing the crime, it having been committed at an unseasonable hour. The trial court left it to the jury to decide whether these two facts sufficiently corroborated the testimony of. the accomplice as to authorize the jury to convict upon it. We decided there was no error in this. The effect of our ruling was, that while the corroboration must tend to connect the defendant with the commission of the offense, it need not refer to any statement or fact testified to by the accomplice. We adhere to that view.
The second form of the contention makes it necessary for us to determine, whether too great time had elapsed between the homicide and the discovery of the overcoat in defendant’s possession, to authorize its submission to the jury as a corroborating factor.
When a larceny, robbery, arson, or even a murder with robbery, has been committed, and soon afterwards goods which were stolen at the time the crime was committed are found in the possession of one not having the rightful custody, there is a presumption, more or less strong, that the *142possession was a guilty possession, and that the custodian committed the crime, of which the acquisition of the goods was a presumed concomitant. The better authorities hold that this is not a presumption of law, but one of fact. All agree, however, that if the discovery follow close upon the crime, the duty is cast on the custodian to., explain how he acquired the possession. Failing to do so to the reasonable satisfaction of the jury, or to generate a reasonable doubt of his guilt, he may be convicted, on this evidence alone, of the crime of which the theft was the motive or concomitant. We use the word may intentionally; for it is a question for the jury. Mr. Greenleaf, vol. 1, § 34, thus expresses the principle: “Possession of the fruits of crime, recently after its commission, is prima facie evidence of guilty possession; and if unexplained, either by direct evidence, or by the attending circumstances, or by the character and habits in life of the possessor, or otherwise, it is taken as conclusive. This rule of presumption is not confined to the case of theft, but is applied to all cases of crime, even the highest and most penal.” — 1 Taylor’s Ev. § 63; Roscoe’s Cr. Ev. 18, 19; Whart. Cr. Ev. § 759. White v. State, 72 Ala. 195, as well as the text of Wharton, and the authorities he cites, show that the presumption is one of fact to be passed on by the j™-7-
What is the precise definition of the adjective recent, as found in the expression of the principle we are considering? In White's Case, following Wharton, it was said it “is incapable of exact or precise definition,” and that it varies, “within a certain range, with the conditions of each particular case.” In England this question has often arisen, when possession of stolen property has been the evidence on which it was sought to convict such custodian of the theft of the property. The ruling there has been, when the larceny was recent, to allow the simple fact of such possession, if unaccounted for, to be considered and weighed by the jury, in determining whether they would convict on that testimony alone. On the other hand, if the larceny was not recent, the trial judges have refused to permit juries to convict, or deliberate on such unaided testimony. In Harris' Case, 8 Cox Cr. Ca. 333, Channell, B. refused to let the jury retire, when the discovery of the stolen property was six months after the larceny, and ordered a verdict of not guilty. In Partridge's Case, 7 C. & P. 554, two months had elapsed between the larceny and the finding; for certain specified reasons, *143the testimony was allowed to go to the jury, and be weighed by them. In Adams' Case, 3 C. & P. 160, the same period of two months had elapsed, and Baron Parke took the case from the jury, and ordered an acquittal.
If the only testimony against the defendant was the finding of the overcoat in his possession more than three months after the homicide, we should think it wholly insufficient to justify his conviction of the murder.
That is not, however, the form in which the question comes before us. It is offered not as sole evidence of guilt, but as corroboration of the testimony of Elzy, the accomplice; Does it tend to connect the defendant with the commission of the offense, and does it so corroborate the testimony of Elzy, that the jury is thereby convinced beyond a reasonable doubt that Meadows was murdered, and that the defendant was present, aiding, abetting, or consenting to the murder? Whether the overcoat was the property of Meadows, was one of the severely contested questions on the trial. Unless the jury were convinced to a moral certainty that such was the case, then it should exert no influence in their deliberations, and Elzy would stand uncorroborated. This was a question for the jury, and they could not convict unless they solved it in favor of the State.
In Williams' Case, 8 Iredell’s L. R. 140, that able jurist, Ruffin, C. J., said: “The possession of a stolen thing is evidence to some extent against the possessor, of a taking by him. Ordinarily, indeed, it is s+ronger or weaker in proportion to the period intervening between the stealing and the finding in the possession of the accused; and after the lapse of a considerable time before a possession is shown in the accused, the law infers not his guilt, but leaves that question to the jury under a consideration of all the circumstances.” We approve this language, with the exception that the guilt of the possessor, however strong the presumption of fact may be engendered thereby, never rises to the dignity of a presumption of law. In the case of State v. Rights, 82 N. C. 675, the principles of Williams' Case were reaffirmed.—1 Taylor Ev. § 140. We feel safe in holding, that the testimony in relation to the overcoat, alleged to have been found in the possession of the accused, was properly submitted to'the jury, on the inquiry of corroboration of the testimony of Elzy, the accomplice. It results, that its sufficiency for that purpose was a question for their decision.
*144Before entering upon the argument to the jury, the defendant requested that the charge of the court be given in writing. This was done. At the close of the' charge, defendant reserved two exceptions to it, one of which was, “to that portion of the charge which undertakes to state the facts and theories as having been proved or claimed on the part of the state, and also on the part of the defendant.” To this, the court replied orally, “I have not attempted to state what the facts are, but simply what is claimed.” The defendant excepted to this statement, because it was not in writing. There was nothing in this. The remark was simply a reply to counsel, and neither asserted, nor attempted to assert, a legal proposition for the guidance of the jury. It was in no sense a charge.
There was nothing in either of the exceptions to the general charge, for two reasons: First, because each failed to point out the particular part or parts of the charge excepted to." — 3 Brick. Dig. 80, §§ 33 et seq. Second, no part of the general charge is obnoxious to the objections urged.
Under the principles declared above, charges asked by defendant from 1 to 8, inclusive, were, each and all, rightly refused.
Charges 9 and 10 raise the question of the proper purpose and scope of corroboration, which will justify a conviction of felony on the testimony of an accomplice. Our statute, as we have seen, requires that it shall tend “to connect the defendant with the commission of the offense.” Whether the overcoat was the property of the deceased, and whether it was about his person at the time he lost his life, were questions for the jury. If that body found each of these inquiries in favor of the State, with that measure of conviction required in criminal cases, then the tendency of that fact was to connect the defendant with thfe commission of the offense. The tendency was more or less convincing, graduated, among other matters of consideration, by the length of time elapsing between the homicide and the discovery of the overcoat. As we have said before, it did not and could not necessarily prove his guilt, — owing to the length of time before the discovery. Hence, if the discovered overcoat had been the only criminating testimony, it was not enough. Such has been the ruling in England, and in States of this Union, whose utterances we respect, and feel it our duty to follow. It only tended to connect the *145accused with the commission of the homicide; only tended to corroborate the testimony of Elzy, the accomplice.
What are the spirit and import of our statute, and of the English rule? What is it that must be corroborated, to meet the requirements of the statute? Not necessarily any specific fact or facts to which the witness has testified. This we have considered in a former part of this opinion. Nor is it required that, of itself, it shall establish the guilt of the accused. To require that, would be to render the testimony of the accomplice unnecessary and redundant. The testimony of the accomplice is not self-supporting, and to authorize conviction upon it, it must be corroborated; that is, its credibility must be strengthened; for, unless it is believed, the charge is not made good, and there can be no conviction. The purpose of the corroboration, then, is, not necessarily to prove the defendant is guilty, but to meet the requirements of the statute — thus convincing the jury that the testimony of the accomplice is true, and authorizing them to found their conviction upon it. If it fails to produce this conviction, then the accomplice’s testimony is not sufficiently strengthened to found a verdict upon, and there should be an acquittal.—Marler's Case, 68 Ala. 580.
Charges 9 and 10 raise the precise question we have been considering. ' They assert that the finding of the overcoat in defendant’s possession, more than three months after the homicide was committed, is not, of itself and unaided, enough to found a verdict of guilty upon; and unless that fact so far corroborates the testimony of Elzy as to convince the jury he testified truly, then they ought to acquit. This is an exact statement of the legal principle involved, and of the mental process by which it is to be worked out. These charges ought to have been given, for the reason, if no other, that without them there may have been danger that the jury would believe the finding of the overcoat in possession of the defendant was sufficient evidence to convict upon.
Reversed and remanded.