CROUCH, J.

On the night of August 6, 1931, two robbers held up a cider stube, so called, on West Forty-fourth street in the city of New York. In the course of the robbery, one of the victims was shot and killed. The robber who fired the shots was pursued and captured. His name was Baumann. The other one escaped. Baumann and his companion, designated as John Doe, were indicted for murder in the first degree. Baumann was tried and convicted as charged. The judgment of conviction was affirmed by this court ( People v. Baumann, 259 N.Y. 600). Subsequently the defendant here was arrested and brought to trial as being Baumann's companion and the person named as John Doe in the indictment. His appeal is from a judgment of conviction for murder in the first degree. That conviction was had upon the testimony of Baumann. Unless Baumann be corroborated by other evidence tending to connect the defendant with the commission of the crime, the conviction must be set aside. (Code Crim. Pro. § 399.)

Baumann testified that he was twenty-one years of age and had known the defendant Reddy, who was commonly called "Howie," for five or six years, although he had seen little of him during the year preceding the murder; that on the afternoon of the murder he had driven with Reddy from Keansburg, New Jersey, where they had both been staying for several days; that after they reached New York, Reddy suggested that they join in a hold-up, to which Baumann agreed; that they separated with an understanding that they would meet at Tenth avenue and Forty-ninth street at seven o'clock in the evening, Reddy in the meantime to get pistols and arrange for the use of a taxicab; that they met at the time and place stated, boarded a taxicab and drove off; that Reddy gave witness a pistol and some extra cartridges; that the taxicab stopped at Forty-fourth street about six doors from Ninth avenue; that Reddy alighted from the taxicab, followed by the witness, the motor being kept running; that the witness took out his pistol, opened the door and went into the cider stube, followed by Reddy; that they ordered the people in the place to put up their hands and go to the rear of the store; that Reddy did most of the searching while the witness kept the victims covered with his pistol; that when the cash register had been rifled and the pockets of the victims searched, Reddy stepped past the witness saying, "Come on, let's go," and proceeded rapidly toward the front door; that as witness turned to follow, Munich, one of the victims, grappled with him; that at that time Reddy was close to the front door; that witness shouted out, "Wait a minute, Howie, this guy has got me," but Reddy proceeded out; that the struggle between witness and Munich ended after a short time with the firing of three shots, after which Munich fell away from the witness, whereupon the witness ran out and saw the taxicab turning into Ninth avenue; that the witness never saw Reddy again until the trial.

It further appears that when Baumann was searched, following his capture, there was found on him a small photograph of Reddy. He repeatedly denied that that was a photograph of the man who was with him; on the contrary, he asserted that his companion was an "Italian fellow." Baumann's conviction was affirmed by this court on June 1, 1932, and his execution was set for July 14, 1932. Within a few hours of the time set for execution, Baumann made a statement which for the first time implicated Reddy. A reprieve of twenty-four hours resulted, during which the district attorney took a statement from Baumann. A second reprieve put off execution until the night of August 18th. The statements which Baumann had made were not wholly true, and it was not until almost the moment of execution on the night of August 18th that Baumann sent for the district attorney and made still another statement which apparently satisfied that official. A further reprieve was thereupon granted to cover the time of Reddy's trial. After that trial Baumann's sentence was commuted to life imprisonment. Baumann on this trial asserted that his testimony on his own trial and all statements made by him relating to the identity of his companion had been intentional and deliberate lies for the purpose of shielding Reddy. He frankly admitted that he finally implicated Reddy and became a witness for the People solely to save his own life, and to that end he would not hesitate to lie. The defendant did not take the stand.

Here, then, is a typical case of conflict between the public need of bringing to justice one against whom suspicion of guilt exists, and the individual right of the suspect to be safeguarded within the law against the effect of tainted evidence. We are to examine the record and say whether or not the testimony of Baumann, the accomplice, was so far corroborated within the rule of faw, that the conviction may safely and justly stand.

The law of corroboration under section 399 of the Code of Criminal Procedure was concisely but comprehensively stated in People v. Dixon ( 231 N.Y. 111, 116). "The corroborative evidence," said the opinion, "need not show the commission of the crime; it need not show that defendant was connected with the commission of the crime. ( People v. Mayhew, 150 N.Y. 346, 353; People v. Cohen, 223 N.Y. 406, 426.) It is enough if it tends to connect the defendant with the commission of the crime in such a way as may reasonably satisfy the jury that the accomplice is telling the truth." Since the web of proof in every case is unique, corroboration "may vary in its nature according to the circumstances of the particular case." (Id.) When the trial judge finds that there is some corroborative evidence, it is his duty to submit it "to the jury for them to say first, whether it was worthy of belief, and secondly, whether if true it tended to connect defendant with the commission of the crime." (Id. 117.)

The learned trial judge found the possibility of corroboration in evidence relating to two matters. One was the alleged flight of defendant following the crime; the other was Baumann's stronger motive for implicating as his companion in crime the man actually guilty, rather than one who was innocent. Accordingly, he submitted to the jury the evidence relating to each of these matters, to say whether it was true, and, if so, whether it tended to connect the defendant with the commission of the crime.

In respect to the matter of the stronger motive, the argument was that Baumann would naturally feel resentment against the man who, instead of responding to his appeal for help, had deserted him and fled in the taxicab with all the loot, leaving him to be hunted through the streets and captured; and hence that his testimony incriminating the defendant was credible in spite of the admitted character and interest of Baumann. So the trial judge charged the jury as follows: "If you find that that contention has merit, you may consider whether that is corroborated by the evidence of the other witnesses. * * * You have heard the testimony of several of the witnesses who were in the place, and of the officers. It is for you to say whether they establish, apart from Baumann, that the other man did escape, did get away in a taxicab, and that Baumann, in consequence of having no taxicab to get into, was captured. If you find that their testimony corroborates Baumann on the question that his accomplice abandoned him, thereby furnishing Baumann with a motive to testify truthfully against the accomplice, you may consider whether that is evidence that tends to connect this defendant with the crime; that is, to show that this defendant is that unknown man that escaped, and that the unknown man was not some other man, some guilty man whom Baumann is shielding."

The testimony of the non-accomplice witnesses to which the charge referred was merely to the effect that the robbery was committed by two men; that one of them escaped, probably in a taxicab, leaving the other struggling with Munich; and that Baumann was captured. None of these witnesses was able to identify the defendant. On the other hand, one of them testified that Baumann's companion was noticeably heavier and stouter than the defendant. None of them, moreover, testified to hearing Baumann's alleged call for help, "Wait a minute, Howie, this guy has got me." Clearly there is nothing in this testimony tending directly or circumstantially to connect the defendant with the commission of the crime. All that it does and all that can be claimed for it, is to give support to some of the facts — and those not the most essential — testified to by Baumann, from which one might infer the existence of resentment, and hence a motive, on the basis of which the truth of Baumann's accusation against the defendant might be conjectured. At the end of that chain, one still has nothing but Baumann's word that his companion was the defendant. There is nothing in the testimony of these witnesses pointing to the defendant's connection with the crime which would not with equal weight point to anybody else whom Baumann might have named. While this testimony was relevant and was corroborative of certain circumstances relating to the crime, it had no tendency, as we think, to connect the defendant with its commission. Its submission to the jury on that theory was, therefore, error.

In respect to the matter of flight, the general rule is well settled. Evidence of flight, concealment or analogous conduct is always admissible. From of old, when unexplained, it has been deemed indicative of a consciousness of guilt, and hence of guilt itself. (Proverbs, chapter XXVIII, 1; People v. Ogle, 104 N.Y. 511, 514; People v. Fiorentino, 197 N.Y. 560, 567; Hickory v. United States, 160 U.S. 408.) But "ordinarily it is of slight value, and none whatever unless there are facts pointing to the motive which prompted it." ( People v. Fiorentino, supra; People v. Stilwell, 244 N.Y. 196, 199.) Its consideration, as pointed out long ago, should not be pressed too far. ( Commonwealth v. Webster, 5 Cush. 295.) To the extent that it is evidence of guilt itself, it must be taken as corroborative of an accomplice's incriminating testimony. (Cf. State v. Taylor, 196 Iowa, 1015; People v. Duffy, 160 App. Div. 385.)

There is testimony by a police detective that on the night of the crime he began efforts to locate defendant and continued such efforts for several months. It is not entirely clear whether he was looking for this particular defendant, knowing him to be Reddy, or merely for some one whose face resembled the face in the photograph found on Baumann. There is also testimony by other detectives and policemen of efforts to find Reddy after Baumann had implicated him. From July 18 to August 20, 1932, when he was arrested, they looked for him at 100 Pilling street, Brooklyn, and at Keansburg, New Jersey; also around Lexington avenue from One Hundred and Tenth to One Hundred and Seventeenth streets, and along the upper west side. That those were known haunts of Reddy is not satisfactorily shown. A flat on West One Hundred and Eightieth street was under observation for two days. On August 20, 1932, Reddy with two other men was seen to come out. He was followed to premises on One Hundred and Sixtieth street, where he was arrested. Reddy had been living there, but for how long and in what manner does not appear. The only testimony which may be said to have significance was given by a parole officer under whose supervision Reddy was. To him Reddy was required to report twice a month. During April, May, June and July, 1931, the reports were duly made. The next report was to be on August 11, 1931. It was not made, nor did Reddy ever thereafter report. The parole officer at once tried unsuccessfully to find Reddy in certain reputed haunts. Since the crime was committed on August 6, 1931, the evidence would warrant a finding that within five days after the crime, Reddy disappeared from his usual haunts and broke his parole. While it was of slight value at best, this evidence of flight or concealment was admissible and relevant upon the main issue of guilt, if there were facts pointing to this crime as the motive which prompted it. Such facts appear in the testimony of Baumann. Whether, being thus in the case, that evidence was corroborative, is another question. Since its only value depended upon the word of Baumann, it is difficult to see how it could be said to corroborate Baumann. But even granting that it could, we are of the opinion that its tendency to connect the defendant with the commission of the crime was so inherently weak and inconclusive as to furnish no reasonable ground for a finding by the jury that the accomplice was telling the truth.

The judgment of conviction should be reversed and a new trial ordered.

POUND, Ch. J., CRANE, LEHMAN, KELLOGG and O'BRIEN, JJ., concur; HUBBS, J., not sitting.

Judgment reversed, etc.