The plaintiff was convicted of murder in the first degree for the killing of Grover S. Hart at the Indian Harbor Yacht Club in Greenwich in the early morning of Saturday, July 23,1949. Upon appeal that conviction was sustained. State v. Smith, 138 Conn. 196, 82 A.2d 816. Smith made application for a new trial on the ground of newly discovered evidence. The trial court rendered judgment denying the application and from that judgment this appeal has been taken.

*204The evidence produced on the trial of the case in which Smith was convicted left no doubt that Hart, the night watchman at the Indian Harbor Yacht Club, had been lolled in the perpetration of a robbery at the club. The bullet which caused Ms death was from a .22 caliber Smith and Wesson revolver. Another bullet and four cartridge cases found on the premises came from a .22 caliber Colt automatic pistol, and this indicated that at least jrwo persons were involved in the crime. George F. Lowden had pleaded guilty to murder in the second degree in connection with the killing.

Smith’s connection with the crime was proved by circumstantial evidence, wMch may be summarized as follows: At about half past one in the morning of Sunday, July 24,1949, a gray Cadillac automobile, parked outside the Hollywood Cafe in Brewster, New York, was noticed by New York police. One man was seated in the ear and another was standing beside it. Before the police reached the car the two men vanished. Shortly before that, Smith had been seen in the Hollywood Cafe by the proprietor. In the car was found the Colt automatic pistol from which the cartridge cases found at the scene of the crime had been discharged, a hat, the property of the manager of the Indian Harbor Yacht Club, and six neckties, several small pieces of jewelry and two tie clasps, all of which had been taken from the club. All of this indicated that the car was one which had been used in the perpetration of the murder. The car was peculiar in that the cover of the trunk was tied down by a rope. That Smith had been in possession of the car was proved not only by the fact that he had been seen in the Hollywood Cafe while the car was parked there but also by the fact that a sMrt which belonged either to Smith or some member of *205Ms family was found in the car and by the testimony of at least three persons that they had seen Smith driving either that automobile or one that looked like it earlier in the month of July. One of these witnesses, Edith Springer, has since attempted to repudiate her testimony. On the strength of her repudiation a new trial was sought, but the trial court in that case found that her recantation was not credible and denied the new trial. That judgment was .affirmed on appeal. Smith v. State, 139 Conn. 249, 93 A.2d 296.

Other evidence against Smith produced on his trial was his guilty conduct at the time he was arrested on July 28. lie was found concealed in the woods at the Wilton reservoir. He gave a fictitious name, denied that he was Francis Smith, falsely claimed that he was a member of a camping party in the neighborhood, and had in his possession a bottle of hair dye. Smith himself testified at his trial. He denied that he had any connection with the Cadillac car found at Brewster, although he did admit that he had been driving a different model of a Cadillac provided for him by a gambler by whom he was employed. He attempted to avoid the effect of his conduct at the time of his arrest by testifying that he thought the police were after him for some violation of his parole from the Connecticut state prison.

The claimed newly discovered evidence upon which the present application for a new trial is based is that set forth in a written statement and in a deposition made by David Blumetti, now a prisoner in Kilby prison, Montgomery, Alabama. The gist of this statement and deposition is that it was Blumetti and not Smith who was associated with Lowden in the perpetration of the murder. His story is that he and Smith spent most of the day of Friday, July *20622, 1949, riding around in the gray Cadillac, which Smith had, discussing where they could find a place to “rob.” They concluded that they would look for a place after dark. They met again after supper time, but then Smith said he did not want to go through with any robbery. Accordingly, Smith loaned Blumetti the car and got out of it in the center of Stamford. At a little after 5 o’clock Blumetti met Lowden, who said he wanted to pick up some money. They drove to Danbury and into New York state but finally decided on the Indian Harbor Yacht Club. They arrived there at about 2 o’clock in the morning and finally effected an entrance. Blumetti made a noise by throwing an ash tray on the floor. Soon they heard steps and a man came through a door. Blumetti jumped in front of him. “ ... I jumped in front of him told him to put Ms hands up, he jumped, I didn’t know what was happening. I started, I shot and FranMe shot from behind me. The man started to holler. All we did then, we both turned around and run out and run right out through the back this time. ... We got in the car and left.” Blumetti accounted for his reference to “FranMe” in the above quotation by stating that he meant George Francis Lowden rather than Francis C. Smith.

As Blumetti’s story goes on, the two drove back to Stamford. Blumetti let “FranMe” out and then he slept in the car outside the town until 9 or 10 o’clock in the morning. He met “Georgie” about 4 o’clock in the afternoon and they then agreed to go out again that night. He picked Lowden up after supper. They drove around and finally arrived at the Hollywood Cafe. They were inside that place for a while, then went out to the car and were making plans to rob the place when they saw the police. They became alarmed and ran away. Other details *207of Blumetti’s story and other facts connected therewith which the court could reasonably have found will be recited later in the opinion.

The trial court concluded that no injustice was done Smith on the former trial, that due diligence in an effort to obtain the present evidence for use at that time was not proven, that the credible evidence connecting Smith with the murder was strengthened rather than weakened by Blumetti’s deposition, that, in so far as Blumetti attempted to exonerate Smith, his testimony was unworthy of credit, and that there was no reasonable ground for believing that on a new trial, if one were to be granted, a jury would bring in a different or more favorable verdict. Upon argument before us, the state’s attorney stated that he does not now take the position that there had been a lack of due diligence in failing to obtain the evidence of Blumetti at the former trial. It is quite clear from the record that although Blumetti was available as a witness at the former trial, the defendant’s counsel could not then by due diligence have become aware that he would testify as he now has. There was no inkling at that time that Blumetti would testify that it was he and not Smith who was implicated. He could hardly be expected to make such a startling disclosure. The trial court was not warranted in concluding that due diligence on the part of the defendant would have produced such evidence on the former trial.

Whether a new trial on the ground of newly discovered evidence is to be granted rests in the sound legal discretion of the trial court, and upon appeal the ultimate question is whether that discretion has been abused. Smith v. State, 139 Conn. 249, 251, 93 A.2d 296; Link v. State, 114 Conn. 102, 107, 157 A. 867; Burns v. State, 84 Conn. 518, 521, 80 A. 712. *208The test to be applied in the trial court is whether the newly discovered evidence indicates that an injustice was done on the former trial and whether it is probable that on a new trial a different result would be reached. Krooner v. State, 137 Conn. 58, 67, 75 A.2d 51; Gonirenki v. American Steel & Wire Co., 106 Conn. 1, 12, 137 A. 26. Accordingly, the question before us on this appeal is whether the trial court’s conclusions that no injustice had been done on the former trial and that it was improbable that Blumetti’s testimony would change the result on a new trial were so. unreasonable that they constituted an abuse of discretion. The decision of this question turns upon the correctness of the trial court’s other conclusion, that Blumetti’s present story is utterly unworthy of credence. If this conclusion is correct, it follows that no injustice was done on the earlier trial. It also follows, since it should be presumed that no jury will believe an incredible story, that the trial court would be warranted in concluding that a new trial would not produce a different verdict.

In considering the credibility of Blumetti’s present story, the first question which naturally arises is why he should implicate himself in a first degree murder unless in doing so he was telling the truth. The circumstances under which he told the story go a long way toward answering that question. He had been a lifelong friend of Smith. They had served prison terms together. When he first told the story, on February 10, 1953, he was serving a sentence of fifteen years in Kilby prison in Alabama on a conviction of armed robbery. Of that sentence, more than twelve years was yet to run. By reason of his having escaped and by reason of his having been involved in a prison riot, he had been kept in segregation since October, 1951, and he knew that he *209would continue in segregation until the warden saw fit to move him. The conditions of life in segregation in this prison as described by the prison authorities themselves were so horrendous that it is at least conceivable that Blumetti would be so desperate that he would not much care what became of him.

More significant than that, however, is the fact that he not only had twelve years yet to serve in Kilby but also in all probability would have to serve as much as ten more years in Florida on another robbery with violence charge before he could be brought back to Connecticut to face a charge of murder. His experience in life had been such that he was fully aware of that probability. He was also aware that, if he were then brought to trial, or indeed, if he were brought to trial at a much earlier date, he could, by repudiating his present statement, probably escape conviction because there would be no other witness to connect him with the crime. That some such thoughts were in his mind is indicated by the fact that after he made his statement he wrote his mother that he had had nothing to do with the murder and was making the statement to help Smith. He told her not to worry because nothing would happen to him. In short, therefore, no particular verity is to be imputed to Blumetti’s story simply because by telling it he confessed that he was guilty of murder.

Courts in other jurisdictions in essentially similar cases, that is, where one of several accomplices convicted of a crime has made an affidavit exonerating another, have quite uniformly sustained the trial court in its denial of a new trial. Cochrane v. State, 48 Ariz. 124, 135, 59 P.2d 658; People v. Merkle, 89 Cal. 82, 85, 26 P. 642; People v. Sullivan, 3 Cal. App. 502, 513, 86 P. 834; Smith v. People, 39 Colo. 202, 206, 88 P. 1072; Perry v. People, 38 Colo. 23, 31, *21087 P. 796; Rawlins v. State, 126 Ga. 96, 99, 54 S.E. 924; Herndon v. State, 110 Ga. 313, 35 S.E. 154; State v. Anderson, 14 Mont. 541, 551, 37 P. 1. The facts in Smith v. People, snpra, are closely alón to those in the case at bar. A witness at the first trial had testified to an alibi. After Smith’s conviction, the witness made an affidavit that it was he, not Smith, who had committed the crime. A new trial was denied on the ground that the affidavit was not worthy of credence, in part because when it was made the affiant was in Texas, close to the Mexican boundary. In some of these cases it is pointed out that testimony of a convicted felon, “taking the rap” for another when it will not seriously harm him to do so, is not ordinarily entitled to much credence.

The credibility of Blumetti’s statement and deposition is open to attack along four lines. In the first place, Blumetti himself is of such a degraded character that he is not worthy of belief. Secondly, there are inconsistencies and improbabilities within the story itself. Thirdly, the story is contradicted by the testimony of other witnesses. Fourthly, it is contradicted by other statements and conduct of Blumetti himself.

Blumetti is not a savory character. On his own admission armed robbery has been his chief business. He has been twice convicted of that crime and sent to prison, once in Connecticut and now in Alabama. He made no claim that the charge now pending against him in Florida is ill-founded. He broke parole in Connecticut by leaving the state and going to Texas. He escaped from the Alabama prison and then participated in a prison riot there. Obviously, he is a man with no sense of responsibility. He is the sort of man who might well be expected to prevaricate *211whenever prevarication would suit his purposes.

The story which he now tells is in itself unconvincing. In spite of skillful examination by the plaintiff’s counsel, the deposition is lacking in several details as to just how the crime was committed. Blumetti’s knowledge of the incidents which he does recount may be easily accounted for on the basis that he read about the case in the newspapers or even on the hypothesis that he was himself a third participant in the crime.

The most glaring feature of his deposition is that, in recounting the shooting, he said, “I shot and Frankie shot from behind me.” Francis is Smith’s first name. When the use of the name Frankie was called to his attention by counsel, Blumetti said, “I call Georgie Frankie; everybody calls George Lowden Frankie; his name is Francis; everybody calls him Frankie.” After that, in describing his escape from the police officers at the Hollywood Cafe, he said, “[T]he police started looking for us but I got away; Frankie got away.” Then came the following questions and answers: “Q.— You remember where you left Frank Smith? A.— That night? Q. — That night. A. — I left him right there.” At this point, although it is true that Blumetti had previously been claiming that Frank Smith had not been with him at the Hollywood Cafe, it is clear that by the use of the name Frankie he was referring to Francis Smith. The use of the full name Frank Smith in counsel’s question could leave no doubt of that. This incident raises the strong suspicion that to Blumetti the name Frankie really meant Smith and not Lowden and that when he testified that “Frankie shot from behind me” he referred to Francis Smith and the real truth thereby slipped out.

*212In a particular of major importance Blumetti’s story does not ring true. He says that he and Smith spent the whole day Friday, July 22,1949, planning to commit robbery and that it was not until evening that Smith withdrew from the enterprise but allowed Blumetti to use his car. The only consideration which could have motivated Smith, a man with a prison record, in hesitating to go on with the planned crime was fear of detection. If he were deterred by that fear it is hardly probable that he would be willing to subject himself to the peril of being implicated when it was discovered that the automobile used in the commission of the robbery was traceable to him.

Blumetti’s story is vulnerable because it is contradicted in at least three essential particulars by other witnesses. Cathryne Blumetti, the deponent’s mother, testified that just prior to the crime there was some rivalry between her son and Lowden. “He [Blumetti] said it in front of everybody that he detested him [Lowden], he hated him, and there was always jealousy because when they used to go out with different girls, I guess David was the biggest rival of George Lowden.” If this testimony was true, it is improbable that the two young men would venture together on the commission of a serious crime. Furthermore, it was an essential part of Blumetti’s story that he and Lowden were together, riding around in the car, from about 5 or 6 o’clock in the evening of Friday, July 22, 1949, until after the murder was committed at between 2 and 3 o’clock on Saturday morning. Ann Schaffner, an apparently unbiased witness, testified with every appearance of certainty that George Lowden was with her from about half past five until about midnight on that Friday night. Another essential part of Blumetti’s *213story is that, after the shooting, he and his associate immediately ran from the clubhouse without stealing anything. When asked to account for the presence of the stolen property in the Cadillac on Sunday morning, Blumetti claimed that it, together with some brandy, had been stolen on another visit to the clubhouse three or four weeks before. Aside from the improbability of their having made a previous visit without disturbing the watchman and the improbability of their keeping stolen property in the ear for four weeks, the manager of the club testified that there had been no thefts of brandy from the club within the six months prior to July 23, 1949.

Of even greater significance than anything which has already been set forth in this opinion is the fact that Blumetti’s presentstory directly contradicts previous statements made by him concerning the murder. In the summer of 1949, Blumetti broke his parole from the state prison in this state by going to Texas. The supervisor of parole, while returning him, asked him if he had had anything to do with the murder at the Indian Harbor Yacht Club, and Blumetti declared that he had not. Blumetti also told county investigator Bowes that he knew nothing about the crime. Later, while in prison at Wethersfield, Blumetti was visited by a brother of Frank Smith in an attempt to persuade Blumetti to establish an alibi for Smith. While Smith’s trial was in progress, one of his counsel talked to Blumetti at Wethersfield. At first, Blumetti refused to disclose any information; then he said that he and Smith were together in Ridgefield at the time the crime was committed. Later, he said that that was not the truth. Two or three days thereafter, he said he would testify to an alibi, but when he was brought to court on a habeas corpus ad testificandum, he first indicated that he *214and Smith had been together elsewhere than at the scene of the crime and then refused to testify. He gave as his reason for this refusal that he did not wish to expose his mother to a charge of perjury, she having given sworn statements to the state police that at the time in question he was at home with her. She had never made any such statement. At none of these times did Blumetti intimate that he himself was implicated in the crime. Not only are all these statements in direct contradiction of his current story but the fact that he seemed ready to testify to an alibi and was deterred apparently only by fear of the consequences to his mother is indicative of his willingness at that time to commit perjury to serve his friend.

Under all the circumstances, it is clear that the trial court was fully justified in concluding that no credence could be given to Blumetti’s story in so far as it purported to exculpate Smith. It follows that the court’s further conclusions that it did not appear that an injustice had been done on the former trial and that Blumetti’s evidence would not probably produce a different result on another trial were also correct.

There is no error.

In this opinion Baldwin, Wynne and Daly, Js., concurred.

(dissenting). The statute provides that the court may grant a new trial of any cause upon the discovery of new evidence. General Statutes § 8013. We have always held with uniformity that a party is entitled to a new trial (1) if the evidence claimed to be newly discovered is in fact newly discovered, that is, discovered after trial; (2) if the evidence is material to the issues; (3) if *215it is not cumulative; (4) if, in reasonable probability, it is sufficient to produce a different result on a new trial; and (5) if it could not have been discovered, by the exercise of due diligence, for presentation at the former trial. Hamlin v. State, 48 Conn. 92, 93. One of the main grounds upon which the trial court based its denial of a new trial was that the plaintiff had failed to prove due diligence in discovering the new evidence. During oral argument, the state conceded that the trier was in error in this respect because there had, in fact, been ample proof of due diligence on the plaintiff’s part.

This concession is in itself somewhat disquieting since it tends to undermine one’s confidence in the other grounds adopted by the trier in denying Smith a new trial. The concession, disturbing as it is, has one commendable feature. It eliminates from consideration one of the five conditions referred to in the Hamlin case. The question now is whether the other four conditions mentioned in that case have been met. To this question there ought to be, it seems to me, but one answer. The evidence was admittedly discovered months after the former trial; nothing could be more material to the issues and, in the face of a record entirely devoid of any proof of Blumetti’s connection with the crime, nothing could be less cumulative than his confession; and finally, if the jury should accept that confession as true — a contingency lying well within the field of reasonable probability — it would legally be impossible for them to reach any other conclusion than that of not guilty of the crime charged.

One weakness in the position of my colleagues lies in their justification of the trial court’s denial of a new trial because that court did not believe Blumetti’s story. A motion for a new trial on the ground *216of newly discovered evidence should never be denied, at least in a capital case, merely because the trial judge refuses to believe the evidence. The test to be applied in determining whether another trial should be granted is whether the evidence meets the five conditions enumerated in the Hamlin case and whether the evidence is of such a nature that the jury, upon a new trial, could reasonably extend credibility to it. To state the matter affirmatively, the above test requires the trial judge to grant a new trial if he is satisfied that, regardless of his own disbelief in the trustworthiness of the new evidence, 1 j / there is not only reasonable certainty that the evidence will be admitted at the new trial but also a reasonable probability that the jury will accept it. Had the trial court applied this test, it could logically have reached no other conclusion than that Smith ought to have a new trial.

The cold, merciless fact which neither reasoning nor wishful thinking can destroy is that Smith, pursuing the identical course he has consistently followed, is still protesting his innocence of the crime of which a jury, relying exclusively upon circumstantial evidence, found him guilty, while another person not only has confessed to the commission of the murder but has placed Smith far from the scene of the crime.

The philosophical attitude of this court toward the appraisal of newly discovered evidence in a capital ease is stated in Andersen v. State, 43 Conn. 514. In that case Andersen sought a new trial on the ground of newly discovered evidence. He had been convicted of murder in the first degree and had been sentenced to be hanged. His defense had been insanity. The newly discovered evidence upon which he sought another trial was to provide proof of his *217insanity additional to' that submitted at the trial. This proposed new evidence consisted of the testimony of various persons who on different occasions had observed his conduct and had regarded it as indicative of insanity. One of the claims of the state was that the new evidence was merely cumulative, as it obviously was, and that a new trial should be refused for that reason. This court, to which the matter came by way of reservation, ignored the cumulative nature of the evidence and, in advising the Superior Court to order a new trial, expressed itself in this manner (p. 517): “If we were to make a rigid application of the rules which govern the Superior Court in civil causes, we should doubtless advise that a new trial should be denied; but in a case where human life is at stake, justice, as well as humanity, requires us to pause and consider before we apply those rules in all their rigor. . . . [P. 518] The rule we are considering, which is a salutary one in its application to civil causes, becomes harsh and oppressive when it requires the sacrifice of a fellow-being, who may possibly in the sight of God be innocent of the crime with which he is charged. . . . [P. 519] In a case where life is at stake, if the new evidence [it is to be noted that the evidence is not limited to that which is believed by the trial court to which the petition for a new trial has been presented], in addition to that already produced, will have the effect to raise a reasonable doubt whether the prisoner was in a condition of mind to commit the highest crime known to our law, and to incur the severest penalty which human law can inflict, it ought not to be regarded as cumulative so as to prevent another trial.”

If in the Andersen case a new trial was proper in order to permit the accused to have the jury’s con*218sideration of evidence which, was clearly cumulative and, hence, violative of the requirements set forth in the Hamlin case, how much more certain should be Smith’s right to a new trial, when the newly discovered evidence complies fully with the Hamlin requirements.

This is a most unusual case. Upon its outcome depends the life of a human being. To be sure, the man appears to be of a lawless crew, but no individual, however low his status, is to be dismissed as too insignificant or degraded for the law to shield, when justice requires it. Smith is, in my judgment, entitled to a new trial.