Applying the independent source test approved in Commonwealth v. Botelho, 369 Mass. 860 , 866-868 (1976), and Commonwealth v. Venios, 378 Mass. 24 , 26-28, 30 (1979), the judge found that identifications of the defendant by the victim and a bystander who saw the two perpetrators (one of whom the bystander knew by name) fleeing from the scene of the crime were admissible despite suggestiveness which tainted pretrial photographic displays. Both the victim and the bystander had been shown large photographic arrays from which they separately selected several pictures as showing similarities. The bystander, having selected "four or five," one of which was in fact a picture of the defendant, then pointed to one of the other pictures and described it as looking "most like the man I had seen";
whereupon she was told by a police officer that the defendant's picture was of the man they suspected. The bystander replied that the man she had seen was much older and heavier than the person depicted in the defendant's photograph. A similar scenario transpired with the victim, who reduced the photographs to four, then selected one, saying "[t]hat it might be looking like him, but it's not exactly." Again, the police pointed to the defendant's picture (one of the four from which the victim had made his final selection) and identified it as the man they suspected. However reprehensible the police conduct, the judge made no error in finding that the victim's and the bystander's later identifications of the defendant in person were not the product of the suggestive photographic procedures. The particular mode of suggestiveness duplicates that found in such cases as United States v. Eatherton, 519 F.2d 603, 607, 609 (1st Cir.), cert. denied, 423 U.S. 987 (1975), Washington v. Cupp, 586 F.2d 134, 136-137 (9th Cir. 1978), and Cepeda v. Henderson, 493 F.Supp. 186, 190 (S.D.N.Y. 1979), in which the subsequent identifications were held admissible, and in such cases as United States v. Russell, 532 F.2d 1063, 1068 (6th Cir. 1976), and United States v. Field, 625 F.2d 862, 868-870 (9th Cir. 1980), where the subsequent identifications were suppressed. Compare State v. Upham, R.I. , - (1980) (415 A.2d 1029, 1031-1032 [R.I. 1980]); State v. Tyson, 603 S.W.2d 748, 753 (Tenn. Crim. Appeal 1980). See also United States v. Leonardi, 623 F.2d 746, 754-755 (2d Cir.), cert. denied, 447 U.S. 928 (1980); United States v. Moskowitz, 581 F.2d 14, 19-20 (2d Cir.), cert. denied, 439 U.S. 871 (1978); Styers v. Smith, 659 F.2d 293, 295, 297-299 (2d Cir. 1981); State v. Cass, 356 So.2d 936, 942 (La. 1978). The judge here had sound bases for finding that the in-person identifications had an independent source. The victim and the bystander had good opportunity to see the perpetrator; they were both spontaneous and decisive in their in-person identifications; neither had definitely identified someone other than the defendant; and both insisted before and during the trial that the picture of the defendant which had been included in the photographic arrays was simply unlike the perpetrator in too many crucial respects -- principally in size and age. For that there was good reason. The photograph in the arrays was taken in March, 1977, when the defendant was apparently seventeen years old; the break was committed in May, 1981, when the defendant was four years older and substantially heavier. We have the array picture before us, along with pictures of the defendant at the time of his arrest. The differences are striking: it is unlikely an observer would link the two unless it occurred to him to imagine what the defendant might have looked like as a youth. From this perspective the refusal of the victim and the bystander to identify the youthful photo as the perpetrator was properly considered as a factor strengthening the reliability of their in-person identifications. Compare Commonwealth v. Correia, 381 Mass. 65 , 81 (1980). From the point of view of suggestiveness (the defendant's argument being confined to that),
the trial display of the youthful photograph to the bystander prior to the lineup is covered by what has already been said. The lineup was photographed, and the resulting exhibit supports the judge's finding that the lineup was free of suggestion. The accidental confrontation in the courthouse between the defendant and the victim was only marginally suggestive. The defendant was apparently one of a group of several people and, although he was in handcuffs, the victim did not see the handcuffs. The spontaneity of the victim's recognition, when he was not expecting to see a suspect, supports the judge's ruling. Compare Commonwealth v. Harris, 11 Mass. App. Ct. 165 , 175 (1981). The motion to suppress was properly denied. 2. Considering the claim of prosecutorial misconduct in light of the high standard implicit in Commonwealth v. Smith, 387 Mass. 900 (1983), we conclude that the claim is insufficiently made out. It is true that the testimony underlying the prosecutor's reference to the inability of the defendant's mother to control him had been excluded, but the reference was, in context, brief and little emphasized and on the whole insufficient to support a firm inference of deliberate misconduct. 3. The other contentions argued also lack merit.