Patrick Ootter, a patrolman on the New York City police force, while on duty in Hughes avenue, between East One Hundred and Eighty-seventh and East One Hundred and Eighty-eighth streets, was killed by a pistol-shot wound at about half-past seven o’clock on the evening of the 4th of August, 1913. The defendant, an Italian 24 years of age, who came to this country in 1904 and is a barber by occupation, was charged with the homicide by an indictment for murder in the first degree. He has been twice tried and convicted: first, in December, 1913, and a second time in May, 1915, on the trial which now comes up for review. The judgment upon the first -conviction was reversed by this court and a new trial ordered in December, 1914, for error in permitting the prosecution to impeach its own witnesses. (See People v. De Martini, 213 N. Y. 203, 32 N. Y. Crim. 265, opinion by Werner, J.) Judge Werner, in Ms opinion, said: “ The one vital, perplexing question in the case is the identity of the slayer of Officer Cotter” (p. 217).
Only one witness for the prosecution claims to have been an -eye-witness of the homicide. This is Salvatore Oiviletti, a lad who was 16 years old at the time'of the trial. He testified that he saw the policeman shot by a man who stood behind a tree, *9920 or 25 feet from the officer who was out in the gutter about 30 feet from the witness. This man “ was just walking and when he got to the tree he stopped.” The witness did not see what became of him. He had on “ a light- gray suit or a dark gray suit, or something,”— the witness could not tell “ because it was just growing dark.” He saw no one running after the shooting of the officer. The witness did not identify the man who he thus says did the shooting.
There being no other witness who testified to having seen the shooting, the identification of the defendant as the murderer depends wholly upon circumstantial evidence. The trial judge, however, refused so to instruct the jury. On the contrary, he said: “ I hold that this is not a case where a conviction depends upon circumstantial evidence.”
The fact that evidently weighed most heavily against the defendant with the jury, was his alleged flight after the homicide. To appreciate the bearing of the evidence on this branch of the case, we must consider some events preceding the killing of Officer Cotter.
There appear to have 'been two shooting affrays on the evening of the murder. The first occurred in Belmont avenue, which is the next avenue east of Hughes avenue, in which the officer was killed. Two pistol shots rang out in Belmont- avenue, at a point about half-way between East One Hundred and Eighty-eighth and East One Hundred and Eighty-ninth streets, whereupon a crowd gathered and began to run or move rapidly toward East One Hundred and Eighty-eighth street, following a man running in the same direction. The pursuit continued, according to the testimony for the prosecution, through East One Hundred and Eighty-eighth street, down Hughes avenue, past the spot where Officer Cotter was shot, one block to East One Hundred and Eighty-seventh street, through the latter street to Hoffman avenue, up Hoffman avenue to East One Hundred and Eighty-eighth street again, along the same to Lorillard place, down *100Lorillard place back again to East One Hundred and Eighty-seventh street, and thence along the same to the intersection of Third avenue, where the defendant who was five or ten feet ahead of the crowd was arrested by a police officer, whose home was in the vicinity and who- had rushed out upon hearing the summons of a police whistle. The route of this hue and cry, which appears to have been set in motion by the sound of the pistol shots in Belmont avenue, is delineated on a map or street-diagram in the record.
The theory of the prosecution was that the defendant was concerned in some manner with a shooting affray in Belmont avenue; that he fled from the scene thereof, taking the course which has been described and pursued by an excited crowd; that on his- way through Hughes avenue he murdered Officer Ootter; and that- he continued on in his flight until he was taken into custody as has been stated. Ho pistol was found upon him; but a loaded cartridge and two empty cartridges were found in an areaway on Lorillard place through which the defendant passed after Officer Ootter was- shot. It is argued that the circumstances connected with this incident justified the jury in finding that the defendant threw away his pistol and cartridges at this point in his flight.
The defendant testified in his own behalf denying all complicity -in the murder. He said that he was walking on Belmont avenue with a young woman named Horma Scarenci and a friend, Charles Dellamonico-, when they were assaulted by two unknown Italians. An affray ensued in which the Italians fired two pistol shots and Dellamonico fell to the sidewalk. Thinking that his friend was killed or wounded, the defendant, although armed only with a razor, pursued the assailants. A crowd of about two hundred persons gathered and joined in the pursuit. As it proceeded the defendant evidently realized that while he was pursuing the Italians the crowd was pursuing him. “ As I turned around the corner, One Hundred and *101Eighty-seventh street and Hughes avenue,” he says, “ a crowd of people were running after me and throwing stones, bricks, and so forth. I was afraid they would kill me, being a stranger up there. I did not know anybody up there, so I kept on running.” He denies that when he was arrested he knew that Officer Cotter or any policeman had been shot.
On the present, as on the prior appeal, the one vital perplexing question is the identity of the slayer of Officer Cotter.
There was no direct proof that the man who did the shooting •was the defendant; nor was there any direct proof that the shooting was done by the person who was seen flying before the pursuing crowd before or after the policeman was shot. The evidence adduced by the People to identify the defendant as the man who shot Officer Cotter was purely circumstantial, consisting largely of testimony relating to his flight. The learned trial judge, nevertheless, instructed the jury otherwise, saying:
“ I hold that this is not a case where a conviction depends upon circumstantial evidence, because there is direct evidence of witnesses who saw the man in front of the crowd, saw the man who was in front of the crowd kill the policeman, and saw the man who was in front of the crowd run into One Hundred and Eighty-seventh street and these other streets and saw the man finally captured. If there were two men who were seen in Hughes avenue and one of the men ran down the street and this defendant was the man that ran down the street, the question would be one of circumstantial evidence, but you have the direct testimony in this case of a witness who says that he saw the officer killed by the man who ran down immediately in front of the crowd and he then ran into One Hundred and Eighty-seventh street and followed the course indicated.”
The learned trial judge erred in his recollection of the testimony and thus inadvertently misled the jury. We think that the exception to this instruction requires a reversal of the judgment- There was no direct evidence that the man who was seen *102pursued by the crowd before and after the homicide was the man who shot the policeman. In telling the jury that there was the court must have conveyed an erroneous idea as to the character of the proof upon which their verdict would be based. The examination of talesmen upon challenges shows how much more willing the ordinary juryman is to convict upon direct than upon circumstantial evidence. Hence, the probable effect of the instruction in the present ease was to make the jury think the proof against the defendant much stronger than it would, have seemed to them if they had been told that their verdict depended solely on circumstantial evidence. In his great treatise Prof. Wigmore points out that the rules of admissibility have nothing to say concerning the weight of evidence when once admitted, although judicial utterances have often dealt with the question whether circumstantial or direct evidence is-relatively the more persuasive. (1 Wigmore on Evidence, § 26.) Mr. William Wills, the author of a work on circumstantial evidence which ranks as a legal classic, says that the best writers, ancient and modern, on the subject of evidence, have concurred in treating circumstantial as inferior in cogency and effect to direct evidence. (Wills on Circ. Evid. [5th ed.] p. 35.) The experience of the writer as a trial judge convincesbim that such is the view which prevails with jurors and that ordinarily they will deem a case stronger when based upon direct evidence than if it rests solely on circumstantial evidence-to establish the guilt of the-accused. It is for this reason that the instruction in the case at bar was so harmful. There is a persistent notion that direct evidence is always of a higher-quality, althoúgh this court has declared otherwise. (People v. Place, 157 N. Y. 584.)
The flight of the defendant- through the streets at and about the time of the homicide is the strongest part of the circumstantial evidence against him. A portion of his so-called flight, however, preceded the shooting of Officer Oottef and, therefore,. *103could not be deemed an attempt to escape from that crime. The further perplexity in the case arises from the inability of the only eye-witness of the murder to identify the defendant as the murderer. The man whom he describes as having shot the officer had been walking, not running, before he stepped behind the tree to shoot, and the witness did not observe what became of him afterward. In a case as close as this, an erroneous statement by the court to the jury to the effect that there was direct evidence of the defendant’s guilt cannot be disregarded as harmless.
The judgment of conviction should be reversed and a new-trial ordered.
Hiscocit, Chase, Collih, Hogah, Cardozo and Sbabury, JJ., concur.
Judgment of conviction reversed, etc.